hi per tensy

Indian Heart J 2003; 55: 21–26 Ramakrishnan et al. Hypertensive Heart Failure 21

Hypertensive Heart Failure

S Ramakrishnan, SS Kothari, VK BahlDepartment of Cardiology, All India Institute of Medical Sciences, New Delhi

Heart failure is being increasingly recognized the worldover. Hypertension and ischemic heart disease are the

two cardinal causes of heart failure.1,2 Over the years, avoluminous amount of literature has accumulatedregarding various facets of hypertensive heart failure.Despite this, the risk and mechanisms of heart failure inpatients with hypertension is not completely understood.Further, due to the common coexistence of coronary arterydisease (CAD) and hypertension in the population, therelative contributions of CAD and hypertension to heartfailure have been difficult to disentangle. This articleprovides an overview of the salient aspects of hypertensiveheart failure (HHF) from a clinical standpoint.

Epidemiology

The data on the prevalence of hypertension in patients withheart failure are rather discordant. In the FraminghamHeart Study cohort,3 hypertension antedated thedevelopment of congestive heart failure (CHF) in 91% ofcases and was associated with a two- to three-fold risk ofdevelopment of CHF after adjusting for age and other riskfactors. Hypertension also had a high population-attributable risk (the percentage of heart failure cases thatcan be attributed to hypertension) for CHF, viz. 39% in menand 59% in women in the Framingham study. In contrast,hypertension was found to be the primary factor in only17% of hospitalized heart failure patients.4 Moreover,hypertension was reported to be the primary etiologicalfactor in only 4% of heart failure patients in an overviewof 31 studies.5

Recent data from two population-based studies indicatethat hypertension is responsible for CHF in 4%–20% ofpatients. In a Swedish study of 7500 patients followed upfor 27 years,6 the identified etiological factors for CHF werehypertension in 20.3%, and CAD either alone or incombination with hypertension in 58.8%. Fox et al.7 havereported hypertension to be the primary attributable factorfor CHF in only 4.4% of incident cases of CHF examinedprospectively. This study emphasized the fact that associated

CAD might be clinically underestimated if coronaryangiography is not routinely performed. However, thecontribution of hypertension to heart failure in patientswith significant CAD was not analyzed in the study.Myocardial infarction (MI) is associated with a five- to six-fold increase in the risk of heart failure in hypertensivepatients.3 Antecedent hypertension interacts withneurohumoral activation, and thereby adversely modifiesearly ventricular remodeling to increase the risk of heartfailure after MI.8 Taken together, these data suggest thatwhile hypertension clearly causes or contributes to heartfailure, the absolute risk of heart failure in an individualremains low in the absence of other factors.

Individual Risk Assessment

In addition to CAD, several other patient-related risk factorsinfluence the risk of heart failure. These include age, gender,race, diabetes, valvular heart disease, obesity, severity ofhypertension, left ventricular hypertrophy (LVH), andalcoholism. An Italian study9 suggests that in a 60-year-old asymptomatic man with a systolic blood pressure (BP)of 160 mmHg, the risk of developing heart failure is 0.37%per year in the absence of LVH, which increases to 0.9%per year in the presence of hypertrophy. If ischemic heartdisease, valvular heart disease, and diabetes coexist in thesame subject, then the risk of heart failure rises to 5.1%and 9.5% in the absence and presence of LVH, respectively.

In humans, what degree and duration of hypertensionwould cause LVH or heart failure has not been determined.No threshold of BP has been observed for a substantivechange in risk for heart failure,3,10 yet 70%–80% of patientswith heart failure and hypertension have a BP >160/100mmHg.3 The incidence of CHF is clearly greater atincreasing levels of blood pressure and lower with lessersystolic BP.3,10 The etiology of hypertension could alsoinfluence the risk of CHF. Renovascular hypertension isassociated with more target organ damage,11 and possiblymore CHF.

The incidence of heart failure increases as a function ofage.3 In the young and middle-aged, the incidence of heartfailure correlates with diastolic and mean pressures.12

However, in the elderly, both these measures correlate

Editorial

Correspondence: Professor VK Bahl, Department of Cardiology,All India Institute of Medical Sciences, New Delhi 110029.e-mail: [email protected]

EDITORIAL.p65 5/6/2003, 9:57 AM21

22 Ramakrishnan et al. Hypertensive Heart Failure Indian Heart J 2003; 55: 21–26

poorly with heart failure. Instead, pulse pressure, a measureof pulsatile load, is a powerful and independent predictorof heart failure.13 A wide pulse pressure is either anindicator or a consequence of aortic stiffness, whichincreases the systolic BP and lowers the diastolic BP by avariety of mechanisms. An increase in the systolic BPincreases the load, and lower diastolic BP reduces thecoronary perfusion pressure, thereby increasing thevulnerability of the heart to failure.13 Both hypertensionand LVH are stronger risk factors for CHF in women ascompared to men. However, women have a lowerprevalence of LVH than men for any given level of bloodpressure.14 In obese hypertensive patients, varyingcombinations of pressure and volume overload are seen,resulting in a mixed eccentric–concentric form of LVH.15

Moreover, obesity is an important independent predictorof left ventricular (LV) mass, and the effects of hypertensionand obesity are additive.16

Diabetes is emerging as an important precursor of heartfailure.3 In the United Kingdom Prospective Diabetic Study(UKPDS),10 the incidence of heart failure in 4801 whitemen with type 2 diabetes was 2.4 per 1000 person-years’follow-up in patients with a systolic BP in the range of 120–129 mmHg, and rose to 7.0 in patients with a systolic BP>160 mmHg. However, no threshold of BP was seen forthe occurrence of heart failure.

Left Ventricular Hypertrophy

Left ventricular hypertrophy represents the major biologicaladaptation to increased pressure load and its limitations.Thus, understanding ventricular remodeling influences theentire issue of heart failure in hypertension and its therapy.Although cardiac failure would possibly occur earlier in theabsence of LVH, LVH is clearly a double-edged sword.Hypertensive LVH has repeatedly been shown to be anindependent marker of cardiac failure, acceleratedatherosclerosis of the coronary arteries, lethal cardiacarrhythmias, and sudden death.17 The increased risk ofsudden death can be explained at least in part by therepolarization abnormalities described in patients withhypertensive LVH. Graded prolongation of the QT intervaloccurs with increasing LV mass index, and measures of QTdispersion are also related to the LV mass.18 Abnormalitiesof the QT interval occur irrespective of the geometry ofLVH.18 Each 50 g/m2 increase in LV mass is associated witha 1.49 increase in the relative risk of cardiovascular diseasefor men and a 1.57 increase for women.16 However,individuals differ in their response of LVH to a givenpressure load. Only 15%–20% of hypertensive patients

show echocardiographic evidence of LVH16 and, inhypertensive patients, it is estimated that approximately40% of the variance in LV mass is accounted for by the totalLV load.19 Further, LVH occurs in the absence ofhypertension and, in some cases, precedes its development(cardiogenic hypertension).20 Old age, male sex, obesity,diabetes in women, black race, and certain geneticinfluences such as angiotensin-converting enzyme (ACE)polymorphisms lead to a greater LVH for a given BP.21 Thegeometric pattern of LVH in hypertension also influencesthe risk. The 10-year incidence of cardiovascular events isreported to be 30% in those with concentric LVH, 25% inthose with eccentric LVH, 15% in those with concentricremodeling (increased relative wall thickness with a normalLV mass), and 9% in those with a normal LV mass.22

Cellular Mechanisms of Left VentricularHypertrophy and Heart Failure

Mechanical forces are thought to be the principaldeterminants of cardiac hypertrophy. The principalmechanical contributor that promotes LVH in earlyhypertension is sustained increase in systemic vascularresistance.19 The molecular mechanisms that translateincreased wall stress to cellular hypertrophy are beginningto be elucidated.23 Increased wall stress activates a stretchreceptor that releases intracellular calcium and activatescalcineurin, which, acting through a series of subcellularevents, activates fetal cardiac and growth genes, such as c-myc and c-jun, to upregulate protein synthesis.24 Themechanical forces work in tandem with potentneurohumoral mediators such as angiotensin II,norepinephrine, and insulin.23

The hypertrophy that occurs in hypertensive heartdisease (HHD) is not homogeneous. In contrast to thehypertrophy seen in athletes, in HHD it is disproportionate,and predominantly involves noncardiomyocytes. Thus, itis not the quantity but the quality of the myocardium thatdistinguishes HHD from the adaptive hypertrophy of theathlete.25 In certain disease states such as chronic anemia,small arteriovenous fistula, atrial septal defect, orhyperthyroidism, cardiac hypertrophy occurs withpreserved homogenecity. These conditions are notassociated with activation of the circulating renin–angiotensin system.26 Several attempts have been made todifferentiate physiologic and pathologic hypertrophy bynoninvasive means. It has been shown that the long-axismean annular velocities measured by tissue Doppler aresignificantly decreased in HHD as compared to athletes.27

Fibrosis, an integral feature of the adverse structural

EDITORIAL.p65 5/6/2003, 9:57 AM22


remodeling of the heart seen in HHD, appears in twomorphologically distinct forms: a reactive form expressedas a perivascular/interstitial fibrosis, and a reparativefibrosis represented by microscopic scars that replacenecrotic myocytes.26 Adverse accumulation of extracellularmatrix initially increases myocardial stiffness, and itscontinued accumulation impairs contractile behavior. Earlyevidence suggests that the measurement of serumconcentrations of procollagen type I C-terminal propeptide(a peptide that is cleaved from procollagen type I duringthe synthesis of fibril-forming collagen type I) may provideindirect information on the extent of myocardial fibrosis.28

Gadolinium-DTPA delayed-enhancement magneticresonance imaging (de-MRI) is accurate in assessingregional fibrosis noninvasively.29

Imaging studies using 111In-labeled monoclonalantimyosin antibodies30 have shown that cell damageoccurs early in HHD, and abnormally stimulated apoptosisof cardiomyocytes and noncardiomyocytes could beresponsible for such cell damage.31 Local factors that maytrigger apoptosis include mechanical forces, oxidative stress,hypoxia, and an unbalanced presence of growth factors andcytokines (e.g. angiotensin II) or neurotransmitters (e.g.norepinephrine).32 Since signals that potentially mediatehypertrophy are largely the ones proposed to mediateapoptosis also, it has been proposed that when growthsignals persist chronically in terminally differentiated cells,they produce a contradictory genetic demand and triggerthe apoptosis.32 The absence of a clear association betweencardiomyocyte apoptosis and absolute BP values indicatethat nonhemodynamic factors, especially angiotensin, maybe more important.32

Clinical Transition to Heart Failure

In humans, HHD leads to three recognizable stages inevolution of heart failure: LVH, diastolic dysfunction, andsystolic dysfunction, not necessarily in the same order.

Abnormalities of LV diastolic filling are observed invarious forms of hypertension in adults as well aschildren,21 and the prevalence may be as high as 22% inasymptomatic hypertensive patients with a BP of >140/90 mmHg.33 Diastolic filling abnormalities have beenshown to generally correlate with LV mass,34 and BP.33

However, diastolic dysfunction could precede the onset ofhypertension in the young male offspring of hypertensiveparents.33

In hypertensive patients, diastolic heart failure isincreasingly being recognized as a cause of CHF. However,the incidence is not clear, as establishing a definite diagnosis

is difficult. In previous published studies on unselectedheart failure populations, the reported prevalence ofpatients having preserved systolic function varied widely,from 13% to 74%.35 Each 1 mmHg increase in pulmonarycapillary wedge pressure, a measure of diastolicdysfunction, is shown to be associated with a 23% increasein the risk of all-cause mortality, and a 13% increase in therisk of cardiovascular events in the 174 patients ofuncomplicated hypertension followed up for 10 years.36

In the absence of MI, hypertensive patients with LVHcommonly have supernormal ejection phase indices onechocardiography, and a significant decrease in LV ejectionfraction occurs very late in the course of HHD.37 Yet,normalcy of these ejection indices may not imply normalsystolic function in hypertension. When indices such asmid-wall fractional shortening are used, up to 16% ofhypertensive patients were found to have depressed LVsystolic function.38 Hypertensive patients also have adepressed end-systolic stress relationship. Furthermore,hypertension may lead to exercise-induced systolic LVdysfunction even though the resting LV function may benormal.39

Acute heart failure (pulmonary edema) in ahypertensive patient could be due to transient systolicdysfunction, diastolic dysfunction, ischemia, or mitralregurgitation. However, nearly 40% of hypertensivepatients have a normal LV ejection fraction indicatingdiastolic dysfunction as the cause.40 Normally, the leftventricle compensates for an increase in systolic load byincreasing end-diastolic volume (i.e., using preload reserve).In a patient with diastolic dysfunction, this small increasein left ventricular end-diastolic volume may be associatedwith a marked elevation in diastolic pressure, because ofthe reduced distensibility of the left ventricle, therebyprecipitating pulmonary edema.41 Common precipitants ofovert heart failure in patients with diastolic dysfunctioninclude old age, tachycardia, sudden severe increase inoverload such as a hypertensive crisis, and loss of atrialkick.42 The relative contribution of significant ischemia dueto epicardial obstructive CAD in the causation ofpulmonary edema is difficult to quantify. More than 60%of hypertensive patients presenting with flash pulmonaryedema had epicardial obstructive CAD in one study.43

Interestingly, however, the flash pulmonary edemarecurred in half the patients even after revascularization.43

The diagnosis of CAD in patients suspected of havingHHF is dif ficult. The sensitivity and specificity ofnoninvasive stress tests are altered by both heart failure andhypertension. In heart failure patients, noninvasive stresstests can be less sensitive due to the low level of stress

EDITORIAL.p65 5/6/2003, 9:57 AM23


obtainable, and less specific due to doubtfulechocardiographic pictures and perfusion images of dilated,hypokinetic ventricles. The baseline electrocardiogram alsooften shows intraventricular conduction delays and alteredrepolarization at rest. Hence, it has been suggested that inheart failure patients the coronary angiogram may be thesole determining test for excluding obstructive epicardialCAD.44 However, to assign ischemia as the etiology of heartfailure in the presence of hypertension, the anatomicdisease should be associated with regional wall motionabnormalities, perfusion abnormalities, or ischemicvalvular dysfunction. It has been suggested that in HHFpatients with a preserved LV ejection fraction, two groupscould be identified based on LV mass. If there was a highdegree of reactive hypertrophy, the patients had a lowerchance of a positive stress test, whereas patients with onlymoderate hypertrophy have a high rate of epicardialobstructive CAD.45

Recently, tissue Doppler measurement of the cyclicvariation index of the backscatter signal at the septum levelhave shown significant alterations in hypertensive patients,and it correlates with the LV mass and geometry of LVH.The alterations could be reversed with antihypertensivetreatment.46 Hence, ultrasonic tissue characterization hasthe potential to identify early those hypertensive patientsat risk of adverse remodeling.

Impact of Treatment

The reported median survival following the diagnosis ofheart failure in hypertensive subjects is 1.37 years in menand 2.48 years in women. The 5-year survival rates arereported to be 24% in men and 31% in women.3 Theprognosis may be better for patients with diastolicdysfunction. For example, in heart failure patients withpreserved systolic function, the annual mortality wasreported to be 8.7% v. 3.0% for matched controls, and inpatients with LV systolic dysfunction, the annual mortalitywas 18.9% v. 4.1% for matched controls in theFramingham cohort.47 Hence, treatment is required beforethe onset of clinical heart failure.

The efficacy of antihypertensive medications in reducingthe incidence of heart failure in diastolic48,49 as well asisolated systolic hypertension50 has been well documented.Meta-analysis of long-term hypertension has also shownthat sustained lowering of blood pressure is effective inpreventing LVH and CHF, regardless of the agent used,51 afinding not corroborated by a recent meta-analysis.52

Currently, there is no clear evidence that one class ofantihypertensive agent is more effective than the other in

retarding the progression of HHD; however, there, is someevidence that calcium-channel blockers may be lesseffective. In the recently reported ALLHAT study,53 theamlodipine group had a 38% higher risk of heart failure,and a 35% higher risk of hospitalized/fatal heart failure ascompared to chlorthalidone. The recent meta-analysis hasalso shown that the risk of heart failure was 15% higherwith calcium-channel blockers as compared to beta-blockers, and 18% higher as compared to ACE inhibitors.52

The relative superiority of ACE inhibitors and diuretics isnot clear, as two recent large trials53,54 have yieldedconflicting results. In the ALLHAT study,53 the lisinoprilgroup had a 19% increased risk of heart failure ascompared to patients treated with chlorthalidone, whereasin the second Australian national blood pressure study,54

the incidence of heart failure was not significantly differentbetween ACE inhibitor- and diuretic-treated patients.

Management of hypertension should not focus merelyon a reduction in BP, but must also target the adversestructural remodeling that begets HHD. Such approachestarget diastolic dysfunction, adverse remodeling, apoptosis,fibrosis, and neurohumoral mediators. Drugs that areshown to be reparative include ACE inhibitors, angiotensin-1 receptor antagonists, endothelin antagonists, andaldosterone receptor antagonists.25 Two approaches thatinfluence apoptosis include inhibition of apoptotic signalsthat trigger the process, and direct blockade of theintracellular apoptotic mechanisms. Drugs shown to beeffective antiapoptotics include ACE inhibitors, losartan,alfa-blockers, and calcium-channel blockers. Hydralazineand diuretics have been shown to have no effect onapoptosis.32 Gene therapy, targeting abnormalities of ionhandling, cellular signaling, neurohormonal control, andapoptosis, hold promise in retarding the progression toheart failure.55 ACE inhibitors have been shown to causeregression in fibrosis, leading to improvement in diastolicdysfunction as compared to diuretics.56 Further progresson specific therapy is awaited.

Conclusions

The data on the prevalence of hypertension in heart failureare rather discordant. The duration and severity ofhypertension critical to cause heart failure is not known.For the same degree of hypertension, several patient-relatedfactors modify the occurrence of heart failure. In patientswith HHF, CAD needs to be carefully evaluated.Mechanisms of hypertrophy and remodeling thatcontribute to systolic and diastolic dysfunction arebeginning to be understood. Antihypertensive treatment

EDITORIAL.p65 5/6/2003, 9:57 AM24


reduces the risk of HHF (possibly with some differencesamong various antihypertensive drugs). Specific therapiestargeting adverse cardiac remodeling are underdevelopment.

References

1. Gillum RF. Epidemiology of heart failure in the United States. Am HeartJ 1993; 126: 1042–1047

2. Mckee PA, Castelli WP, McNamara PM, Kannel WB. The naturalhistory of congestive heart failure: The Framingham study. N Engl JMed 1971; 285: 1441–1446

3. Levy D, Larson MG, Vasan RS, Kannel WB, Ho KK. The progressionfrom hypertension to congestive heart failure. JAMA 1996; 275:1557–1562

4. Andersson B, Waagstein F. Spectrum and outcome of congestive heartfailure in a hospitalized population. Am Heart J 1993; 126: 632–640

5. Teerlink JR, Goldhaber SZ, Pfeffer MA. An overview of contemporaryetiologies of congestive heart failure. Am Heart J 1991; 121: 1852–1853

6. Wilhelmsen L, Rosengren A, Eriksson H, Lappas G. Heart failure inthe general population of men—morbidity, risk factors and prognosis.J Intern Med 2001; 249: 253–261

7. Fox KF, Cowie MR, Wood DA, Coats AJS, Gibbs JS, Underwood SR, etal. Coronary artery disease as the cause of incident heart failure inthe population. Eur Heart J 2001; 22: 228–236

8. Richards AM, Nicholls MG, Troughton RW, Lainchbury JG, Elliott J,Frampton C, et al. Antecedent hypertension and heart failure aftermyocardial infarction. J Am Coll Cardiol 2002; 39: 1182–1188

9. Verdecchia P. [Cardiac failure in hypertensive cardiopathy] [Articlein Italian]. Ital Heart J 2000; 1 (Suppl): 72–77

10. Adler AI, Stratton IM, Neil HA, Yudkin JS, Matthews DR, Cull CA, etal. Association of systolic blood pressure with macrovascular andmicrovascular complications of type 2 diabetes (UKPDS 36):prospective observational study. BMJ 2000; 321: 412–419

11. Losito A, Fagugli RM, Zampi I, Parente B, de Rango P, Giordano G, etal. Comparison of target organ damage in renovascular and essentialhypertension. Am J Hypertens 1996; 9: 1062–1067

12. Kannel WB, Gordon T, Schwartz MJ. Systolic versus diastolic bloodpressure and risk of coronary heart disease. The Framingham study.Am J Cardiol 1971; 27: 335–346

13. Vaccarino V, Holford TR, Krumholz HM. Pulse pressure and risk formyocardial infarction and heart failure in the elderly. J Am Coll Cardiol2000; 36: 130–138

14. Agabiti-Rosei E, Muiesan ML. Left ventricular hypertrophy and heartfailure in women. J Hypertens 2002; 20 (Suppl): S34–S38

15. Thakur V, Richards R, Reisin E. Obesity, hypertension, and the heart.Am J Med Sci 2001; 321: 242–248

16. Levy D, Anderson KM, Savage DD, Kannel WB, Christiansen JC,Castelli WP. Echocardiographically detected left ventricularhypertrophy: prevalence and risk factors. The Framingham HeartStudy. Ann Intern Med 1988; 108: 7–13

17. Frohlich ED. Fibrosis and ischemia: the real risks in hypertensive heartdisease. Am J Hypertens 2001; 14: 194S–199S

18. Oikarinen L, Nieminen MS, Viitasalo M, Toivonen L, Wachtell K,Papademetriou V, et al. Relation of QT interval and QT dispersion toechocardiographic left ventricular hypertrophy and geometric patternin hypertensive patients. The LIFE study. The Losartan InterventionFor Endpoint Reduction. J Hypertens 2001; 19: 1883–1891

19. Ganau A, Devereux RB, Pickering TG, Roman MJ, Schnall PL, SantucciS, et al. Relation of left ventricular hemodynamic load and contractileperformance of left ventricular mass in hypertension. Circulation1990; 81: 25–36

20. Post WS, Larson MG, Levy D. Impact of left ventricular structure onthe incidence of hypertension. The Framingham Heart Study.Circulation 1994; 90: 179–185

21. Vasan RS, Levy D. The role of hypertension in the pathogenesis ofheart failure—a clinical mechanistic overview. Arch Intern Med 1996;156: 1789–1796

22. Koren MJ, Devereux RB, Casale PN, Savage DD, Laragh JH. Relationof left ventricular mass and geometry to morbidity and mortality inuncomplicated essential hypertension. Ann Intern Med 1991; 114:345–352

23. Morgan HE, Baker KM. Cardiac hypertrophy. Mechanical, neural, andendocrine dependence. Circulation 1991; 83: 13–25

24. Molkentin JD, Lu JR, Antos CL, Markham B, Richardson J, Robbins J,et al. A calcineurin-dependent transcriptional pathway for cardiachypertrophy. Cell 1998; 93: 215–228

25. Weber KT. Cardioreparation in hypertensive heart disease.Hypertension 2001; 38: 588–591

26. Weber KT, Brilla CG. Pathological hypertrophy and cardiacinterstitium. Fibrosis and renin–angiotensin–aldosterone system.Circulation 1991; 83: 1849–1865

27. Vinereanu D, Florescu N, Sculthorpe N, Tweddel AC, Stephens MR,Fraser AG. Differentiation between pathologic and physiologic leftventricular hypertrophy by tissue Doppler assessment of long-axisfunction in patients with hypertrophic cardiomyopathy or systemichypertension and in athletes. Am J Cardiol 2001; 88: 53–58

28. Lopez B, Gonzalez A, Varo N, Laviades C, Querejeta R, Diez J.Biochemical assessment of myocardial fibrosis in hypertensive heartdisease. Hypertension 2001; 38: 1222–1226

29. Wilson JM, Villareal RP, Hariharan R, Massumi A, Muthupillai R,Flamm SD. Magnetic resonance imaging of myocardial fibrosis inhypertrophic cardiomyopathy. Tex Heart Inst J 2002; 29: 176–180

30. Pons-Llado G, Ballester M, Borras X, Carreras F, Carrio I, Lopez-Contreras J, et al. Myocardial cell damage in human hypertension. JAm Coll Cardiol 2000; 36: 2198–2203

31. González A, López B, Ravassa S, Querejeta R, Larman M, Díez J, et al.Stimulation of cardiac apoptosis in essential hypertension: potentialrole of angiotensin II. Hypertension 2002; 39: 75–80

32. Fortuño MA, Ravassa S, Fortuño A, Zalba G, Díez J. Cardiomyocyteapoptotic cell death in arterial hypertension: mechanisms andpotential management. Hypertension 2001; 38: 1406-1412

33. Phillips RA, Goldman ME, Ardeljan M, Arora R, Eison HB, Yu BY, etal. Determinants of abnormal left ventricular filling in earlyhypertension. J Am Coll Cardiol 1989; 14: 979–985

34. Ren JF, Pancholy SB, Iskandrian AS, Lighty GW Jr, Mallavarapu C,Segal BL. Doppler echocardiagraphic evaluation of the spectrum ofleft ventricular diastolic dysfunction in essential hypertension. AmHeart J 1994; 127: 906–913

35. Vasan RS, Benjamin EJ, Levy D. Prevalence, clinical features andprognosis of diastolic heart failure: an epidemiologic perspective. JAm Coll Cardiol 1995; 26: 1565–1574

36. Fagard R, Pardaens K. Left ventricular diastolic function predictsoutcome in uncomplicated hypertension. Am J Hypertens 2001; 14:504–508

37. Frohlich ED, Tarazi RC, Dustan HP. Clinical–physiological correlationsin the development of hypertensive heart disease. Circulation 1971;44: 446–455

38. de Simone G, Devereux RB, Roman MJ, Ganau A, Saba PS, AldermanMH, et al. Assessment of left ventricular function by midwall fractionalshortening/end-systolic stress relation in human hypertension. J AmColl Cardiol 1994; 23: 1444–1451

39. Cuocolo A, Sax FL, Brush JE, Maron BJ, Bacharach SL, Bonow RO.Left ventricular hypertrophy and impaired diastolic filling in essentialhypertension. Diastolic mechanisms for systolic dysfunction duringexercise. Circulation 1990; 81: 978–986

40. Gandhi SK, Powers JC, Nomeir AM, Fowle K, Kitzman DW, Rankin

EDITORIAL.p65 5/6/2003, 9:57 AM25


KM, et al. The pathogenesis of acute pulmonary edema associatedwith hypertension. N Engl J Med 2001; 344: 17–22

41. Vasan RS, Benjamin EJ. Diastolic heart failure—no time to relax. NEngl J Med 2001; 344: 56–59

42. Shah PM, Pai RG. Diastolic heart failure. Curr Probl Cardiol 1992; 17:781–868

43. Kramer K, Kirkman P, Kitzman D, Little WC. Flash pulmonary edema:association with hypertension and re-ocurrence despite coronaryrevascularization. Am Heart J 2000; 140: 451–455

44. Tavazzi L. Towards a more precise definition of heart failure aetiology.Eur Heart J 2001; 22: 192–195

45. Iriarte M, Murga N, Sagastagoitia D, Molinero E, Morillas M, SalcedoA, et al. Congestive heart failure from left ventricular diastolicdysfunction in systemic hypertension. Am J Cardiol 1993; 71: 308–312

46. Di Bello V, Giorgi D, Talini E, Dell’ Omo G, Palagi C, Romano MF, et al.Incremental value of ultrasonic tissue characterization (backscatter)in the evaluation of left ventricular myocardial structure andmechanics in essential arterial hypertension. Circulation 2003; 107:74–80

47. Vasan RS, Larson MG, Benjamin EJ, Evans JC, Reiss CK, Levy D.Congestive heart failure in subjects with normal versus reduced leftventricular ejection fraction: prevalence and mortality in apopulation-based cohort. J Am Coll Cardiol 1999; 33: 1948–1955

48. Veterans Administration Cooperative Study Group onAntihypertensive Agents. Effects of treatment on morbidity inhypertension. Results in patients with diastolic blood pressuresaveraging 115 through 129 mm Hg. JAMA 1967; 202: 1028–1034

49. Veterans Administration Cooperative Study Group onAntihypertensive Agents. Effects of treatment on morbidity in

hypertension. II. Results in patients with diastolic blood pressureaveraging 90 through 115 mm Hg. JAMA 1970; 213: 1143–1152

50. SHEP Cooperative Research Group. Prevention of stroke byantihypertensive drug treatment in older persons with isolated systolichypertension. Final results of the Systolic Hypertension in the ElderlyProgram (SHEP). JAMA 1991; 265: 3255–3264

51. Moser M, Hebert PR. Prevention of disease progression, left ventricularhypertrophy and congestive heart failure in hypertension treatmenttrials. J Am Coll Cardiol 1996; 27: 1214–1218

52. Neal B, MacMohan S, Chapman N. Effects of ACE inhibitors, calciumantagonists, and other blood-pressure-lowering drugs: results ofprospectively designed overviews of randomised trials. Blood PressureLowering Treatment Trialists’ Collaboration. Lancet 2000; 356:1955–1964

53. The ALLHAT officers and coordinators for the ALLHAT collaborativeresearch group. Major outcomes in high-risk hypertensive patientsrandomized to angiotensin-converting enzyme inhibitor or calciumchannel blocker vs diuretic: The Antihypertensive and Lipid-LoweringTreatment to Prevent Heart Attack Trial (ALLHAT). JAMA 2002; 288:2981–2997

54. Wing LM, Reid CM, Ryan P, Beilin LJ, Brown MA, Jennings GL, et al. Acomparison of outcomes with angiotensin-converting-enzymeinhibitors and diuretics for hypertension in the elderly. N Engl J Med2003; 348: 583–592

55. MacNeill BD, Hayase M, Hajjar RJ. Targeting signaling pathways inheart failure by gene transfer. Curr Atheroscler Rep 2003; 5: 178–185

56. Brilla CG, Funck RC, Rupp H. Lisinopril-mediated regression ofmyocardial fibrosis in patients with hypertensive heart disease.Circulation 2000; 102: 1388–1393

EDITORIAL.p65 5/6/2003, 9:57 AM26

Indian Heart J 2003; 55: 27–34 Laroia et al. Infections in Atherogenesis 27

Role of Infections in Atherogenesis

Sandeep T Laroia, Apar Kishor Ganti, Anil PottiDepartment of Medicine, University of North Dakota School of

Medicine and Health Sciences, USA

In 1921, Ophuls1 proposed that infections could lead toatherosclerosis. This hypothesis was based on pathologic

specimens of blood vessels, which showed macrophageinfiltrates and foam cells. Five decades later, in 1978,Fabricant et al.2 once again showed arterial lesions similarto atherosclerosis in chicken infected with the avianherpesvirus. Recent advances in diagnostic techniques havefacilitated the re-emergence of this hypothesis. It is now wellaccepted that atherosclerosis is an inflammatory process, anda natural corollary of this concept is that microorganismscould be the prime initiators of this process.3 Data fromseveral studies indicate an increased prevalence of chronicinfections in atherogenesis. The agents that have beenimplicated in this process are Chlamydia pneumoniae,Helicobacter pylori, herpes simplex virus, and cytomegalovirus(CMV).4–6 In this article, we evaluate the evidence availableboth for and against the different etiological agents, and theircurrent status in the pathogenesis of atherosclerosis.

Basic Scheme of Atherogenesis

Simplified, atherogenesis is the passage of low-densitylipoprotein (LDL)-cholesterol through dysfunctionalendothelium at points of low shear stress. In addition,mechanical stress can lead to the aggravation of thisphenomenon.7 LDL-cholesterol penetrates thedysfunctional endothelium and undergoes oxidation. Theoxidized LDL causes further endothelial dysfunction.Monocytes penetrate the endothelium, differentiate intomacrophages, attract more macrophages and the resultantfoam cells, which are lipid-laden macrophages, accumulatein the region. These cells may ultimately rupture, causingthe release of toxic inflammatory mediators that trigger afibroproliferative response from smooth muscles.8, 9

Potential Role of Infections in Atherogenesis

Endothelial dysfunction, the trigger for atherogenesis, canbe induced by systemic or local infection. Multiple

mechanisms have been proposed which include thefollowing:

1. Bacterial endotoxins and tumor necrosis factor (TNF-alpha) can inhibit vasodilator nitric oxide generated byendothelial-dependent processes;10

2. Endothelial stunning, a mechanism which hypothesizesthat periods of endothelial inactivity can be induced bya brief exposure to endotoxin;11

3. Direct infection of the endothelium by infectious agents,especially the herpesviruses including CMV;12

4. Altered expression of growth-controlling proteins byvascular smooth muscles after infection with certainviruses, which leads to these cells obtaining a growthadvantage and thus may contribute to atherosclerosisand restenosis.13

Furthermore, C-reactive protein (CRP) and fibrinogen(acute phase reactants), which are strong independentpredictors of subsequent cardiovascular events, are foundto be elevated in infectious states as well, and elevated levelsof cytokines are found in both infections and acutecoronary syndromes. These circulating cytokines maycause abnormal endothelial function, increasedthrombosis, and toxic free radical generation, leading toaccelerated atherogenesis.14,15

Evidence for the Role of Infectionsin Atherosclerosis

Herpesviridae, especially CMV, H. pylori, and C. pneumoniae,have been extensively studied for their ef fect onatherosclerosis. Tables 1, 2, and 3 list the various studiesthat showed a positive correlation between atherogenesisand infection by herpesviruses,16–30 H. pylori,31–41 andC. pneumoniae,42–63 respectively. These studies included bothhuman and experimental research. The experimentalworks listed include those involving both experimentalanimals and tissue studies.

Evidence Against the Role of Infectionsin Atherosclerosis

Though there are a number of reports favoring the role of

Correspondence: Dr Sandeep T Laroia, Department of Internal Medicine,University of North Dakota School of Medicine, 1919 Elm Street N, Fargo,ND 58102, USA. e-mail: [email protected]

Review Article

IHJ-328-02.p65 5/6/2003, 9:58 AM27

28 Laroia et al. Infections in Atherogenesis Indian Heart J 2003; 55: 27–34

Table 1. Evidence favoring the role of CMV and other herpesviridae in atherogenesis

Reference Year Number of Findingssubjects (n)

Grattan et al.16 1989 301 Increased risk of developing vascular lesions in post-transplant patients with clinicalCMV infection

Zhu et al.17 2000 238 CMV seropositivity associated with increased CRP levels, a significant determinantof CAD

Muhlestein et al.18 2000 985 CMV seropositivity and elevated CRP, in combination, are independent predictorsof mortality in patients with CAD

Siscovick et al.19 2000 213 Presence of IgG antibodies to HSV-1 is associated with a 2-fold increase in the riskof incident MI and CAD-related death

Horvath et al.20 2000 244 Genomic DNA of CMV is found in arterial walls at a significantly higher rate in IHDpatients than the nonischemic control group

Sorlie et al.21 2000 515 Population with the highest antibody levels of CMV (approximately the upper 20%)showed an increased relative risk of CHD

Kaftan et al.22 1999 310 Titer of anti-CMV antibody and the levels of CRP can predict patients with a highrisk of CAD

Neumann et al.23 2000 551 Previous CMV infection increased the risk of coronary thrombotic events after stentplacement

Biocina et al.24 1999 284 Patients undergoing orthotopic heart transplantation who experienced CMV diseasehad significantly worse long-term survival compared to those with the infectiononly or without the infection

Blum et al.25 1998 65 Patients with high anti-CMV titer had a higher prevalence of CAD and a higherrestenosis rate than those with a lower antibody titer

Zhou et al.26 1996 75 Patients who were seropositive for CMV had a greater than 5-fold increased rate ofrestenosis following coronary atherotomy

Alber et al.27 2000 NA Atheroma formation was accelerated in apo E-/- mice infected with murine gamma-herpesvirus-68 compared with control uninfected apo E-/- mice

Lemstrom et al.28 1997 NA Treatment with ganciclovir significantly reduced intimal thickening in the presenceof CMV infection

Zhou et al.29 1999 60 CMV infection of immunocompetent adult rats increased the neointimal responseto vascular injury

Lin et al.30 2000 40 Bovine herpesvirus-4 can accelerate the atherosclerotic process in rabbits

CMV: cytomegalovirus; HSV: herpes simplex virus; CRP: C-reactive protein; CAD: coronary artery disease; CHD: coronary heart disease; apo E: apolipoproteinE; IHD: ischemic heart disease

infections in atherosclerosis, the number of reports arguingagainst such a role is equally impressive. In the discussionthat follows, a few reports which showed no effect ofinfection in the pathogenesis of atherogenesis have beenlisted.

Evidence against herpesviridae: Siscovick et al.19 did notfind any correlation between the presence of IgG antibodiesto CMV and the risk of acute myocardial infarction (AMI)or coronary artery disease (CAD) in elderly patients.Choussat et al.63 studied systemic markers of inflammationin patients with unstable angina or non-Q wave MI, andthe relationship between these markers, seropositivity tochronic infections (CMV, H. pylori, and C. pneumoniae), andprognosis. They found no association between the levels ofeach inflammatory marker and the serologic status.Furthermore, levels of inflammatory proteins in patients

seronegative to all 3 agents were comparable to those ofpatients seropositive to 2 or 3 infectious agents. Thecomposite end-points of death, MI, recurrent angina, orrevascularization at 1-year follow-up did not dif feraccording to the serologic status.63 To determine if CMVinfection is a risk factor for primary CAD and the associationbetween CMV infection and CAD (>50% blockage in anycoronary artery), Adler et al.64 investigated nearly 900successive nontransplant patients undergoing coronaryangiography. By the use of logistic regression, they foundthat CMV seropositivity (p=0.462), the level of IgGantibodies to CMV whole-cell antigen (p=0.98), or the levelsof IgG antibodies to CMV glycoprotein B (p=0.67) were notsignificantly associated with CAD. These data suggest thatCMV infection is not a major risk factor for the developmentof primary CAD in adults.64

IHJ-328-02.p65 5/6/2003, 9:58 AM28


Table 2. Evidence favoring the role of Helicobacter pylori in atherogenesis


Mendall et al.31 1994 200 Seropositivity for H. pylori confers a 2-fold risk of CADPatel et al.32 1995 388 76.6% of men with ischemia or infarction were seropositive compared with 45.5%

of men with normal electrocardiogramsGunn et al.33 2000 556 1.80-fold increase in AMI risk, which increased further to 2.25-fold in subjects <55

years of ageKahan et al.34 2000 200 Seropositivity for H. pylori is associated with previous AMIFarsak et al.35 2000 85 H. pylori DNA was found in 17 of 46 (37%) endarterectomy specimens with

atherosclerotic plaques and in none of the controlsHoffmeister et al.36 2001 405 Significant association is observed between H. pylori infection and decreased

HDL-cholesterol levelsPieniazek et al.37 1999 157 Infection significantly increases the risk of CADAmeriso et al.38 2001 38 H. pylori is present in a substantial number of carotid atherosclerotic lesionsLaurila et al.39 1999 880 The serum triglyceride and total cholesterol concentrations were significantly higher

in males with positive antibody titers than in those with no infection (p<0.001)Markus et al.40 1998 357 H. pylori seropositivity was associated with large-vessel disease and lacunar strokeBirnie et al.41 1998 136 Anti-heat shock protein 65 titers correlated with both the severity and extent of

coronary atherosclerosis100 Successful eradication led to a significant fall in anti-heat shock protein 65 titers

H. pylori: Helicobacter pylori; CAD: coronary artery disease; AMI: acute myocardial infarction; HDL: high-density lipoprotein

Evidence against Helicobacter pylori: A Finnish groupfailed to show a statistically significant relationship betweenpatients infected with H. pylori and CAD; interestingly, thisstudy showed higher levels of serum triglycerides in patientsseropositive for H. pylori.65 In their study, Tsai and Huang66

showed that H. pylori seropositivity was not associated withseveral coronary risk factors in either cases or controls. Theproportion of H. pylori-positive patients was higher amongcases with triple-vessel disease (77.5%) than in those withdouble-vessel (67.3%) and single-vessel (65.7%) disease;however, the differences were not statistically significant.In this study, no increase was found in H. pyloriseropositivity in subjects with CAD. In a recent review,Menge et al.67 found that the present data were inconclusiveregarding the association between H. pylori infection andCAD. They concluded that proposed links between H. pyloriinfection and coronary heart disease (CHD), such ashyperhomocysteinemia or autoimmune mechanisms dueto cross-reacting antibodies to H. pylori heat-shock proteinwith human endothelium-derived heat-shock protein, needfurther confirmation. Quinn et al.68 studied the relationshipbetween angiographically defined CAD and serologicevidence of H. pylori infection in 488 patients undergoingelective coronary angiography. There was no associationbetween H. pylori infection and CAD.68 Basili et al.69

retrospectively analyzed 149 subjects who underwent anesophagogastroduodenoscopy, in whom the search forH. pylori was histologically performed, and found that the

prevalence of CAD was not significantly different from thatobserved in H. pylori-free patients (26% v. 21%; p=0.527).Lastly, a meta-analysis of 18 epidemiological studiesinvolving over 10 000 patients failed to demonstrate anysignificant association between H. pylori infection andCAD.70 Khurshid et al.71 prospectively studied 179 patientsundergoing coronary angiography for suspected CAD andfound that H. pylori infection rates were similar in patientswith normal and abnormal coronary arteries, and infectionwith H. pylori was not an independent risk factor for CAD.In patients with CAD, H. pylori infection was not a risk factorfor more severe disease.71

Evidence against Chlamydia pneumoniae

Patient data: The Physicians Health Study72 prospectivelymeasured IgG antibodies against C. pneumoniae in 343participants with first MI. A similar number of age- andsmoking-matched controls were also followed up for aperiod of 12 years. The prevalence of seropositivity was thesame in both the groups.72 Markus et al.73 obtainedultrasonic images of the carotid artery to determine theintima–media thickness (IMT) and the thickness of anyatheroma plaques, and found no evidence that serologicalevidence of C. pneumoniae infection is associated with earlyatherosclerosis. They also found no evidence thatC. pneumoniae results in a chronic systemic inflammatorystate.73 Hoffmeister et al.74 investigated the associationbetween seropositivity to chlamydial lipopolysaccharide

IHJ-328-02.p65 5/6/2003, 9:58 AM29


Table 3. Evidence favoring the role of Chlamydia pneumoniae in atherogenesis


Hu et al.42 1999 40 Infection exacerbated the development of atherosclerosis in miceMuhlestein et al.43 1998 30 Intranasal infection accelerated intimal thickening in rabbits given a cholesterol-

enhanced diet. Also, treatment with azithromycin after exposure preventsaccelerated intimal thickening

Liu et al.44 2000 NA Infection for 6 months produced a significantly greater exacerbation of aorticatherosclerosis in LDLR-/- mice in the presence of a high-cholesterol diet

Kuo et al.45 1993 36 C. pneumoniae was detected in coronary artery atheromas by immunocytochemistryand PCR in 20 of 36 autopsy cases with coronary artery atheromas

Muhlestein et al.46 1996 114 Atherectomy specimens of patients with symptomatic CAD showed positivechlamydial antigen by direct immunofluorescence in 73% of patients

Ericson et al.47 2000 60 Direct immunofluorescence for chlamydial antigen was reactive in 86% of caseswith severe atherosclerosis but in only 6% of cases with mild atherosclerosis(p<0.01)

Maass et al.48 1998 70 Coronary artery endarterectomy specimens showed viable chlamydial organismsin 16% and chlamydial DNA in 30%

Bartels et al.49 1999 58 Viable C. pneumoniae was recovered from 16% of occluded vein grafts while exceptfor 1 native saphenous vein, all control vessels were negative for C. pneumoniae

Farsak et al.35 2000 85 C. pneumoniae DNA was found in 26% of 46 endarterectomy specimens and noneof the healthy vascular wall specimens (p<0.001)

Ouchi et al.50 2000 177 62% of atherosclerotic plaques from symptomatic patients were infected with C.pneumoniae compared with just 2% of nonatherosclerotic tissues

Saikku et al.51 1998 213 In a series of 213 patients with AMI, 38% showed strong seropositivity, 35% showedintermediate seropositivity, and 28% showed no detectable antibodies

Gabriel et al.52 1998 282 Patients with CAD had a higher prevalence of C. pneumoniae PCR from pharyngealspecimens (36% v. 22%)

Toss et al.53 1998 256 Among 256 patients with unstable angina, increased antibody titers to C. pneumoniaewere more common than in controls (36% v. 19%). Even asymptomaticatherosclerosis specimens showed increased antibody titers

Melnick et al.54 1993 300 In 300 patients, cardiac wall thickening seen on ultrasound (a marker forasymptomatic atherosclerosis) was correlated with positive antibody titers (73% v.63%)

Burian et al.55 2001 405 Presence of high-level antibodies to C. pneumoniae is an independent risk factor forthe development of coronary atherosclerosis

Leowattana et al.56 2000 243 A significantly higher percentage of patients with CAD had positive IgG and IgAantibodies to C. pneumoniae as compared to healthy controls

Maass et al.57 2000 188 C. pneumoniae was detected in 52 (28%) of 188 persons with unstable angina andin 13 (26%) of 50 persons with MI

Kaftan et al.58 2000 160 Levels of triglyceride, LDL-cholesterol, CRP, fibrinogen, and total cholesterol/HDL-cholesterol ratio had a direct relationship, but the level of HDL-cholesterol had anegative relationship with seropositivity

Sessa et al.59 1999 228 Significant correlation exists between chronic C. pneumoniae infection anddyslipidemias in the AMI and CAD groups

Wong et al.60 1999 804 Circulating C. pneumoniae DNA is a predictor of CAD in menDechend et al.61 1999 NA Tissue factor, PAI-1, and interleukin-6 expression was increased in infected

cells, and NF-κ B was activated in human vascular endothelial and smooth musclecells

Fong62 2000 NA Early institution of antimicrobials with antichlamydial activity within 5 days ofinfection largely prevented aortic lesions in New Zealand white rabbits

C. pneumoniae: Chlamydia pneumomiae; CAD: coronary artery disease; AMI: acute myocardial infarction; LDL: low-density lipoprotein; CRP: C-reactiveprotein; HDL: high-density lipoprotein; PCR: polymerase chain reaction; PAI: plasminogen activator inhibitor

IHJ-328-02.p65 5/6/2003, 9:58 AM30


(cLPS) or C. pneumoniae and angiographically documentedCAD. They also examined the relationship betweenserostatus and markers of systemic inflammation. Theirresults indicated no strong association betweenC. pneumoniae and CAD, and they concluded that theincreased systemic inflammation in patients with CAD didnot seem to be due to seropositivity to cLPS orC. pneumoniae.74 In their meta-analysis, Danesh et al.75

reviewed 15 studies to examine the association betweenCAD and serum markers of chronic C. pneumoniae infection.They did not find any strong association betweenC. pneumoniae IgG titers and incident CAD. Romeo et al.76

tried to correlate the severity of CAD with seropositivity toC. pneumoniae prospectively. They found no significantdifference in IgG and IgA seropositivity among patients withstable CAD, unstable CAD, and controls. They concludedthat only a small percentage of patients with CADdemonstrate seropositivity against C. pneumoniae. A recentcase–control study investigated the relationship betweenthe presence of C. pneumoniae IgG and IgA andangiographically diagnosed CAD. When cases werecompared with controls whose angiographic results werenormal, after adjusting for established risk factors(cholesterol, smoking, hypertension, diabetes, age, gender,and family history), the estimated risk of CAD was 0.79 forthe presence of IgG and 0.94 for IgA. These results do notsupport an association between C. pneumoniae infection andCAD.77 In a recent review, Wong et al.78 concluded thatmore evidence is required before C. pneumoniae can beaccepted as playing a role in atherosclerosis.

Animal data: Aalto-Setala et al.79 infected apolipoproteinE (apo E)-deficient mice with C. pneumoniae and placed themon either a high- or low-fat diet. They found thatC. pneumoniae infection did not influence the lesion size ineither mouse strain. They also could not demonstrateC. pneumoniae by polymerase chain reaction in any of theatherosclerotic lesions of the infected animals. They did notfind any inflammatory signs in the myocardium ofC. pneumoniae-infected mice. They concluded thatC. pneumoniae infection did not accelerate atherogenicchanges in the aortic root of apo E-deficient mice.79 Blessinget al.80 inoculated C57BL/6J mice with C. pneumoniae. Theyobserved inflammatory changes in the heart or aorta in asmall number of chronically infected mice but no evidenceof atherosclerotic lesions in any of them. Their findingssuggested that chronic C. pneumoniae infection could induceinflammatory changes in the heart and aorta of C57BL/6Jmice but did not initiate definitive atherosclerosis.80 In arecent review, Haberbosch and Jantos81 concluded from the

present data that chronic infection with the pathogen isnot an independent risk factor for atherosclerosis.

Intervention Studies

The ACADEMIC trial82 reported 302 patients withseropositivity to C. pneumoniae. Subjects receivedazithromycin and placebo for 3 months. At 6 months, theazithromycin-treated group showed a reduced global indexof inflammation, which comprised CRP, TNF-alpha; andinterleukins 1 and 6 compared with a placebo. However,cardiovascular events were similar in the two groups at6 months. At 2-year follow-up, there was a 20%–30% riskreduction.82 Parchure et al.83 carried out a randomized,prospective, double-blind, placebo-controlled trial in 40male patients with documented CAD and positiveC. pneumoniae IgG antibody titers. They showed thatpatients who received azithromycin had a significantimprovement in flow-mediated dilatation of the brachialartery. They then concluded that treatment withazithromycin had a favorable effect on endothelial functionin patients with documented CAD and evidence ofC. pneumoniae infection, irrespective of antibody titerlevels.83

A pilot study of 60 survivors of AMI with persistentelevated anti-chlamydial antibody titers was designed sothat subjects were randomized to receive placebo orazithromycin. Azithromycin-treated patients showed anapparent reduction in cardiovascular events from 28% to8%. There was no significant difference between a single-and double-dose course of azithromycin.84 The ROXIStrial85 randomized 202 patients with unstable angina ornon-Q wave MI to roxithromycin or a placebo. At the endof the treatment period, the rates of recurrent ischemiawere 1% v. 5.4%, MI 0% v. 2.2%, and ischemic events 0% v.2.2% in the roxithromycin v. placebo group, respectively.85

At 6 months, the individual and composite event ratesremained lower in the roxithromycin group, but thedifference was not statistically significant.86 In the AZACStrial,87 patients with acute coronary syndromes (unstableangina or MI) were randomized in a double-blind, placebo-controlled fashion to either azithromycin 500 mg/dayfollowed by 250 mg/day for 4 days or a matching placebo.They found that in patients with AMI or unstable angina,short-term treatment with azithromycin did not have aneffect on the recurrence of ischemic events during a6-month follow-up period. There was no difference betweenpatients who tested positive for the presence ofC. pneumoniae antibodies and those who did not.87 Similarly,in the WIZARD trials,88 which sought to assess the use of

IHJ-328-02.p65 5/6/2003, 9:58 AM31


antibiotics to prevent recurrent CHD, the antibiotic regimencomprising weekly azithromycin in adult patients >6 weekspost-MI with elevated C. pneumoniae IgG titers, achieved a7% nonsignificant reduction in the incidence of recurrentcardiovascular disease at 11 weeks. The baseline titer ofIgG antibodies against C. pneumoniae had no effect on theoutcome.88

Conclusions

Atherosclerosis is an inflammatory process. Hence,microorganisms could be the prime initiators of thisprocess. Endothelial dysfunction, the trigger foratherogenesis, can be induced by systemic or local infection.Circulating cytokines, released by any stimulus, may causeabnormal endothelial function, increased thrombosis andtoxic free-radical generation, leading to acceleratedatherogenesis. As in leprosy, Whipple’s disease, syphilis andehrlichiosis, Koch’s postulates cannot be used to determinethe role of infection in atherosclerotic disease. Regardingthe role of infection in CAD, important indirect evidencecomes from prevention of the disease by means of specificinterventions.

The data for and against the role of infection inatherosclerotic vascular disease are equally impressive.However, the few prospective clinical trials evaluating therole of antibiotics in the secondary prevention of CAD havenot shown a significant decrease in clinically majorcardiovascular events. Larger studies with a longer durationof follow-up may be more useful in assessing the exactpathogenetic mechanism of infection in atheroscleroticheart disease, and also the role of antibiotic therapy in thetreatment of CAD.

Based on the current data, however, it is not possible tobe certain one way or the other about the role of infectionin the pathogenesis of atherogenesis, and the subsequentcomplications of atherosclerosis. Further studies arerequired to solve this complex problem.

References

1. Ophuls W. Arteriosclerosis and cardiovascular disease. JAMA 1921;76: 700–701

2. Fabricant CG, Fabricant J, Litrenta MM, Minick CR. Virus-inducedatherosclerosis. J Exp Med 1978; 148: 335–340

3. Lorber B. Are all diseases infectious? Another look. Ann Intern Med1999; 131: 989–990

4. Danesh J, Collins R, Peto R. Chronic infections and coronary heartdisease: is there a link? Lancet1997; 350: 430–436

5. Ridker PM. Inflammation, infection, and cardiovascular risk: howgood is the clinical evidence? Circulation 1998; 97: 1671–1674

6. Fuster V. Human lesion studies. In: Numano F, Ross R (eds).Atherosclerosis IV: recent advances in atherosclerosis research. New York

City: The New York Academy of Sciences;1997. pp. 207–2257. Drexler H. Endothelial dysfunction: clinical implications. Prog

Cardiovasc Dis 1997; 39: 287–3248. Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathogenesis of

coronary artery disease and the acute coronary syndromes (1). NEngl J Med 1992; 326: 242–250

9. Stary HC. Evolution and progression of atherosclerotic lesions incoronary arteries of children and young adults. Arteriosclerosis 1989;9: (Suppl) I19–I32

10. Wang P, Ba ZF, Chaudry IH. Administration of tumor necrosis factor-alpha in vivo depresses endothelium-dependent relaxation. Am JPhysiol 1994; 266: H2535–H2541

11. Bhagat K, Moss R, Collier J, Vallance P. Endothelial “stunning”following a brief exposure to endotoxin: a mechanism to link infectionand infarction? Cardiovasc Res 1996; 32: 822–829

12. Hajjar DP, Pomerantz KB, Falcone DJ, Weksler BB, Grant AJ. Herpessimplex virus infection in human arterial cells. Implications inarteriosclerosis. J Clin Invest 1987; 80: 1317–1321

13. Speir E, Modali R, Huang ES, Leon MB, Shawl F, Finkel T, et al. Potentialrole of human cytomegalovirus and p53 interaction in coronaryrestenosis. Science 1994; 265: 391–394

14. Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH.Inflammation, aspirin, and the risk of cardiovascular disease inapparently healthy men. N Engl J Med 1997; 336: 973–979

15. Woodhouse PR, Khaw KT, Plummer M, Foley A, Meade TW. Seasonalvariations of plasma fibrinogen and factor VII activity in the elderly:winter infections and death from cardiovascular disease. Lancet 1994;343: 435–439

16. Grattan MT, Moreno-Cabral CE, Starnes VA, Oyer PE, Stinson EB,Shumway NE. Cytomegalovirus infection is associated with cardiacallograft rejection and atherosclerosis. JAMA 1989; 261: 3561–3566

17. Zhu J, Shearer GM, Norman JE, Pinto LA, Marincola FM, Prasad A, etal. Host response to cytomegalovirus infection as a determinant ofsusceptibility to coronary artery disease: sex-based differences ininflammation and type of immune response. Circulation 2000; 102:2491–2496

18. Muhlestein JB, Horne BD, Carlquist JF, Madsen TE, Bair TL, PearsonRR, et al. Cytomegalovirus seropositivity and C-reactive protein haveindependent and combined predictive value for mortality in patientswith angiographically demonstrated coronary artery disease.Circulation 2000; 102: 1917–1923

19. Siscovick DS, Schwartz SM, Corey L, Grayston JT, Ashley R, Wang SP,et al. Chlamydia pneumoniae, herpes simplex virus type 1, andcytomegalovirus and incident myocardial infarction and coronaryheart disease death in older adults: the Cardiovascular Health Study.Circulation 2000; 102: 2335–2340

20. Horvath R, Cerny J, Benedik J Jr, Hokl J, Jelinkova I, Benedik J. Thepossible role of human cytomegalovirus (HCMV) in the origin ofatherosclerosis. J Clin Virol 2000; 16: 17–24

21. Sorlie PD, Nieto FJ, Adam E, Folsom AR, Shahar E, Massing M. Aprospective study of cytomegalovirus, herpes simplex virus 1, andcoronary heart disease: the atherosclerosis risk in communities(ARIC) study. Arch Intern Med 2000; 160: 2027–2032

22. Kaftan HA, Kaftan O, Kilic M. Markers of chronic infection andinflammation. Are they important in cases with chronic coronaryheart disease. Jpn Heart J 1999; 40: 275–280

23. Neumann FJ, Kastrati A, Miethke T, Pogatsa-Murray G, Seyfarth M,Schomig A. Previous cytomegalovirus infection and risk of coronarythrombotic events after stent placement. Circulation 2000; 101: 11–13

24. Biocina B, Husedzinovic I, Sutlic Z, Presecki V, Wallwork J.Cytomegalovirus disease as a possible etiologic factor for earlyatherosclerosis. Coll Antropol 1999; 23: 673–681

25. Blum A, Giladi M, Weinberg M, Kaplan G, Pasternack H, Laniado S,et al. High anti-cytomegalovirus (CMV) IgG antibody titer is associated

IHJ-328-02.p65 5/6/2003, 9:58 AM32


with coronary artery disease and may predict post-coronary balloonangioplasty restenosis. Am J Cardiol 1998; 81: 866–868

26. Zhou YF, Leon MB, Waclawiw MA, Popma JJ, Yu Zx, Finkel T, et al.Association between prior cytomegalovirus infection and the risk ofrestenosis after coronary atherectomy. N Engl J Med 1996; 335; 624–630

27. Alber DG, Powell KL, Vallance P, Goodwin DA, Grahame-Clarke C.Herpesvirus infection accelerates atherosclerosis in the apolipoproteinE-deficient mouse. Circulation 2000; 102: 779–785

28. Lemstrom K, Sihvola R, Bruggeman C, Hayry P, Koskinen P.Cytomegalovirus infection-enhanced cardiac allograft vasculopathyis abolished by DHPG prophylaxis in the rat. Circulation 1997; 95:2614–2616

29. Zhou YF, Shou M, Guetta E, Guzman R, Unger EF, Yu ZX, et al.Cytomegalovirus infection of rats increases the neointimal responseto vascular injury without consistent evidence of direct infection ofthe vascular wall. Circulation 1999; 100: 1569–1575

30. Lin TM, Jiang MJ, Eng HL, Shi GY, Lai LC, Huang BJ, et al. Experimentalinfection with bovine herpesvirus-4 enhances atherosclerotic processin rabbits. Lab Invest 2000; 80: 3–11

31. Mendall MA, Goggin PM, Molineaux N, Levy J, Toosy T, Strachan D,et al. Relation of Helicobacter pylori infection and coronary heartdisease. Br Heart J 1994; 71: 437–439

32. Patel P, Mendall MA, Carrington D, Strachan DP, Leatham E,Molineaux N, et al. Association of Helicobacter pylori and Chlamydiapneumoniae infections with coronary heart disease and cardiovascularrisk factors. BMJ 1995; 311: 711–714

33. Gunn M, Stephens JC, Thompson JR, Rathbone BJ, Samani NJ.Significant association of cagA positive Helicobacter pylori strains withrisk of premature myocardial infarction. Heart 2000; 84: 267–271

34. Kahan T, Lundman P, Olsson G, Wendt M. Greater than normalprevalence of seropositivity for Helicobacter pylori among patients whohave suffered myocardial infarction. Coron Artery Dis 2000; 11: 523–526

35. Farsak B, Yildirir A, Akyon Y, Pinar A, Oc M, Boke E, et al. Detectionof Chlamydia pneumoniae and Helicobacter pylori DNA in humanatherosclerotic plaques by PCR. J Clin Microbiol 2000; 38: 4408–4411

36. Hoffmeister A, Rothenbacher D, Bode G, Persson K, Marz W, NauckMA, et al. Current infection with Helicobacter pylori, but notseropositivity to Chlamydia pneumoniae or cytomegalovirus, isassociated with an atherogenic, modified lipid profile. ArteriosclerThromb Vasc Biol 2001; 21: 427–432

37. Pieniazek P, Karczewska E, Duda A, Tracz W, Pasowicz M, KonturekSJ. Association of Helicobacter pylori infection with coronary heartdisease. J Physiol Pharmacol 1999; 50: 743–751

38. Ameriso SF, Fridman EA, Leiguarda RC, Sevlever GE. Detection ofHelicobacter pylori in human carotid atherosclerotic plaques. Stroke2001; 32: 385–391

39. Laurila A, Bloigu A, Nayha S, Hassi J, Leinonen M, Saikku P.Association of Helicobacter pylori infection with elevated serum lipids.Atherosclerosis 1999; 142: 207–210

40. Markus HS, Mendall MA. Helicobacter pylori infection: a risk factor forischaemic cerebrovascular disease and carotid atheroma. J NeurolNeurosurg Psychiatry 1998; 64: 104–107

41. Birnie DH, Holme ER, McKay IC, Hood S, McColl KE, Hillis WS.Association between antibodies to heat shock protein 65 and coronaryatherosclerosis. Possible mechanism of action of Helicobacter pyloriand other bacterial infections in increasing cardiovascular risk. EurHeart J 1998; 19: 387–394

42. Hu H, Pierce GN, Zhong G. The atherogenic effects of chlamydia aredependent on serum cholesterol and specific to Chlamydia pneumoniae.J Clin Invest 1999; 103: 747–753

43. Muhlestein JB, Anderson JL, Hammond EH, Zhao L, Trehan S,Schwobe EP, et al. Infection with Chlamydia pneumoniae acceleratesthe development of atherosclerosis and treatment with azithromycin

prevents it in a rabbit model. Circulation 1998; 97: 633–63644. Liu L, Hu H, Ji H, Murdin AD, Pierce GN, Zhong G. Chlamydia

pneumoniae infection significantly exacerbates aortic atherosclerosisin an LDLR -/- mouse model within six months. Mol Cell Biochem2000; 215: 123–128

45. Kuo CC, Shor A, Campbell LA, Fukushi H, Patton DL, Grayson JT.Demonstration of Chlamydia pneumoniae in atherosclerotic lesions ofcoronary arteries. J Infect Dis 1993; 167: 841–849

46. Muhlestein JB, Hammond EH, Carlquist JF, Radicke E, Thomson MJ,Karagounis LA, et al. Increased incidence of Chlamydia species withinthe coronary arteries of patients with symptomatic atheroscleroticversus other forms of cardiovascular disease. J Am Coll Cardiol 1996;27: 1555–1561

47. Ericson K, Saldeen TG, Lindquist O, Pahlson C, Mehta JL. Relationshipof Chlamydia pneumoniae infection to severity of human coronaryatherosclerosis. Circulation 2000; 101: 2568–2571

48. Maass M, Bartels C, Kruger S, Krause E, Engel PM, Dalhoff K.Endovascular presence of Chlamydia pneumoniae DNA is a generalizedphenomenon in atherosclerotic vascular disease. Atherosclerosis 1998;140 (Suppl): S25–S30

49. Bartels C, Maass M, Bein G, Malisius R, Brill N, Bechtel JF, et al.Detection of Chlamydia pneumoniae but not cytomegalovirus inoccluded saphenous vein coronary artery bypass grafts. Circulation1999; 99: 879–882

50. Ouchi K, Fujii B, Kudo S, Shirai M, Yamashita K, Gondo T, et al.Chlamydia pneumoniae in atherosclerotic and nonatherosclerotictissue. J Infect Dis 2000; 181 (Suppl): S441–S443

51. Saikku P, Leinonen M, Mattila K, Ekman MR, Nieminen MS, MakelaPH, et al. Serological evidence of an association of a novel Chlamydia,TWAR, with chronic coronary heart disease and acute myocardialinfarction. Lancet 1988; 2: 983–986

52. Gabriel AS, Gnarpe H, Gnarpe J, Hallander H, Nyquist O, MartinssonA. The prevalence of chronic Chlamydia pneumoniae infection asdetected by polymerase chain reaction in pharyngeal samples frompatients with ischaemic heart disease. Eur Heart J 1998; 19: 1321–1327

53. Toss H, Gnarpe J, Gnarpe H, Siegbahn A, Lindahl B, Wallentin L.Increased fibrinogen levels are associated with persistent Chlamydiapneumoniae infection in unstable coronary artery disease. Eur Heart J1998; 19: 570–577

54. Melnick SL, Shahar E, Folsom AR, Grayston JT, Sorlie PD, Wang SP, etal. Past infection by Chlamydia pneumoniae strain TWAR andasymptomatic carotid atherosclerosis. Atherosclerosis Risk inCommunities (ARIC) Study Investigators. Am J Med 1993; 95: 499–504

55. Burian K, Kis Z, Virok D, Endresz V, Prohaszka Z, Duba J, et al.Independent and joint effects of antibodies to human heat-shockprotein 60 and Chlamydia pneumoniae infection in the developmentof coronary atherosclerosis. Circulation 2001; 103: 1503–1508

56. Leowattana W, Mahanonda N, Bhuripanyo K, Pokium S, Kiartivich S.Chlamydia pneumoniae antibodies and angiographically demonstratedcoronary artery disease in Thailand. J Med Assoc Thai 2000; 83: 1054–1058

57. Maass M, Jahn J, Gieffers J, Dalhoff K, Katus HA, Solbach W. Detectionof Chlamydia pneumoniae within peripheral blood monocytes ofpatients with unstable angina or myocardial infarction. J Infect Dis2000; 181 (Suppl): S449–S451

58. Kaftan AH, Kaftan O. Coronary artery disease and infection withchlamydia pneumonia. Jpn Heart J 2000; 41: 165–172

59. Sessa R, Di Pietro M, Santino I, del Piano M, Varveri A, Dagianti A, etal. Chlamydia pneumoniae infection and atherosclerotic coronarydisease. Am Heart J 1999; 137: 1116–1119

60. Wong YK, Dawkins KD, Ward ME. Circulating Chlamydia pneumoniaeDNA as a predictor of coronary artery disease. J Am Coll Cardiol 1999;34: 1435–1439

IHJ-328-02.p65 5/6/2003, 9:58 AM33


61. Dechend R, Maass M, Gieffers J, Dietz R, Scheidereit C, Leutz A, et al.Chlamydia pneumoniae infection of vascular smooth muscle andendothelial cells activates NF-kappaB and induces tissue factor andPAI-1 expression: a potential link to accelerated arteriosclerosis.Circulation 1999; 100: 1369–1373

62. Fong IW. Antibiotics effects in a rabbit model of Chlamydia pneumoniae-induced atherosclerosis. J Infect Dis 2000; 181 (Suppl): S514–S518

63. Choussat R, Montalescot G, Collet J, Jardel C, Ankri A, Fillet A, et al.Effect of prior exposure to Chlamydia pneumoniae, Helicobacter pylori,or cytomegalovirus on the degree of inflammation and one-yearprognosis of patients with unstable angina pectoris or non-Q-waveacute myocardial infarction. Am J Cardiol 2000; 86: 379–384

64. Adler SP, Hur JK, Wang JB, Vetrovec GW. Prior infection withcytomegalovirus is not a major risk factor for angiographicallydemonstrated coronary artery atherosclerosis. J Infect Dis 1998; 177:209–212

65. Niemela S, Karttunen T, Korhonen T, Laara E, Karttunen R, IkaheimoM, et al. Could Helicobacter pylori infection increase the risk ofcoronary heart disease by modifying serum lipid concentrations?Heart1996; 75: 573–575

66. Tsai CJ, Huang TY. Relation of Helicobacter pylori infection andangiographically demonstrated coronary artery disease. Dig Dis Sci2000; 45: 1227–1232

67. Menge H, Lang A, Brosius B, Hopert R, Lollgen H. Helicobacter pyloriand coronary heart diseases—hypotheses and facts. Z Gastroenterol2000; 38: 315–323

68. Quinn MJ, Foley JB, Mulvihill NT, Lee J, Crean PA, Walsh MJ, et al.Helicobacter pylori serology in patients with angiographicallydocumented coronary artery disease. Am J Cardiol 1999; 83: 1664–1666

69. Basili S, Vieri M, Di Lecce VN, Maccioni D, Marmifero M, Paradiso M,et al. Association between histological diagnosis of Helicobacter pyloriand coronary heart disease: results of a retrospective study. Clin Ter1998; 149: 413–417

70. Danesh J, Peto R. Risk factors for coronary heart disease and infectionwith Helicobacter pylori: meta-analysis of 18 studies. BMJ 1998; 316:1130–1132

71. Khurshid A, Fenske T, Bajwa T, Bourgeois K, Vakil N. A prospective,controlled study of Helicobacter pylori seroprevalence in coronaryartery disease. Am J Gastroenterol 1998; 93: 717–720

72. Ridker PM, Kundsin RB, Stampfer MJ, Poulin S, Hennekens CH.Prospective study of Chlamydia pneumoniae IgG seropositivity and risksof future myocardial infarction. Circulation 1999; 99: 1161–1164

73. Markus HS, Sitzer M, Carrington D, Mendall MA, Steinmetz H.Chlamydia pneumoniae infection and early asymptomatic carotidatherosclerosis. Circulation 1999; 100: 832–837

74. Hoffmeister A, Rothenbacher D, Wanner P, Bode G, Persson K, BrennerH, et al. Seropositivity to chlamydial lipopolysaccharide and Chlamydiapneumoniae, systemic inflammation and stable coronary arterydisease: negative results of a case–control study. J Am Coll Cardiol2000; 35: 112–128

75. Danesh J, Whincup P, Walker M, Lennon L, Thomson A, Appleby P, et

al. Chlamydia pneumoniae IgG titers and coronary heart disease:prospective study and meta-analysis. BMJ 2000; 321: 208–213

76. Romeo F, Martuscelli E, Chirieolo G, Cerabino LM, Ericson K, SaldeenTG, et al. Seropositivity against Chlamydia pneumoniae in patients withcoronary atherosclerosis. Clin Cardiol 2000; 23: 327–330

77. Cellesi C, Sansoni A, Casini S, Migliorini L, Zacchini F, Gasparini R, etal. Chlamydia pneumoniae antibodies and angiographicallydemonstrated coronary artery disease in a sample population fromItaly. Atherosclerosis 1999; 145: 81–85

78. Wong YK, Gallagher PJ, Ward ME. Chlamydia pneumoniae andatherosclerosis. Heart 1999; 81: 232–238

79. Aalto-Setala K, Laitinen K, Erkkila L, Leinonen M, Jauhiainen M,Ehnholm C, et al. Chlamydia pneumoniae does not increaseatherosclerosis in the aortic root of apolipoprotein E-deficient mice.Arterioscler Thromb Vasc Biol 2001; 21: 578–584

80. Blessing E, Lin TM, Campbell LA, Rosenfeld ME, Lloyd D, Kuo C.Chlamydia pneumoniae induces inflammatory changes in the heart andaorta of normocholesterolemic C57BL/6J mice. Infect Immun 2000;68: 4765–4768

81. Haberbosch W, Jantos C. Chlamydia pneumoniae infection is not anindependent risk factor for arterial disease. Herz 2000; 25: 79–83

82. Anderson JL, Muhlestein JB, Carlquist J, Allen A, Trehan S, Nielson C,et al. Randomized secondary prevention trial with azithromycin inpatients with coronary artery disease and serological evidence forChlamydia pneumoniae infection: The Azithromycin in CoronaryArtery Disease: Elimination of Myocardial infection with Chlamydia(ACADEMIC) study. Circulation 1999; 99: 1540–1547

83. Parchure N, Zouridakis EG, Kaski JC. Effect of azithromycin treatmenton endothelial function in patients with coronary artery disease andevidence of Chlamydia pneumoniae infection. Circulation 2002; 105:1298–1303

84. Gupta S, Leatham EW, Carrington D, Mendall MA, Kaski JC, CammAJ. Elevated Chlamydia pneumoniae antibodies, cardiovascular events,and azithromycin in male survivors of myocardial infarction.Circulation 1997; 96: 404–407

85. Gurfinkel E, Bozovich G, Daroca A, Beck E, Mautner B. Randomisedtrial of roxithromycin in non-Q-wave coronary syndromes: ROXISPilot Study. ROXIS Study Group. Lancet 1997; 350: 404–407

86. Gurfinkel E, Bozovich G, Beck E, Testa E, Livellara B, Mautner B.Treatment with the antibiotic roxithromycin in patients with acutenon-Q-wave coronary syndromes. The final report of the ROXIS Study.Eur Heart J 1999; 20: 121–127

87. Cercek B. Azithromycin in Acute Coronary Syndrome Investigators.The effect of short-term treatment with azithromycin on recurrentischemic events in patients with acute coronary syndrome. Programand abstracts of the American College of Cardiology 51st AnnualMeeting; 2002 March 17–20; Atlanta, Georgia. [Abstr] p. 503

88. Dunne M, O’Connor C, Pfeffer M, Muhlestein B, Gupta S, Yao L. WeeklyIntervention with Zithromax for Atherosclerosis and Its RelatedDisorders (The WIZARD Study). Program and abstracts of theAmerican College of Cardiology 51st Annual Meeting; 2002. March17–20; Atlanta, Georgia [Abstr] p. 504

IHJ-328-02.p65 5/6/2003, 9:58 AM34

Indian Heart J 2003; 55: 35–39 Jose et al. Utility of NT-proBNP for the Diagnosis of Heart Failure 35

Utility of N-Terminal Pro-Brain Natriuretic Peptidefor the Diagnosis of Heart Failure

Jacob V Jose, Satya N Gupta, Dhayakani SelvakumarChristian Medical College and Hospital, Vellore

Heart failure (HF) is a common disorder, and itsprevalence increases with age. In the Framingham

Heart Study, the prevalence of HF increased from 0.8% inthe 50–59 years age group to 9.1% in the population 70years of age and older.1,2 The prevalence of HF maycontinue to rise as the proportion of elderly people withinthe population increases. Heart failure is not only animportant reason for hospitalization but is also second onlyto hypertension as a cause of outpatient physician visits.3

Most patients in the emergency department present withacute shortness of breath. It is extremely important todifferentiate between HF and other causes of shortness ofbreath.

Symptoms

and physical examination alone are not

sufficient to always make an accurate diagnosis. Currently,

echocardiography is considered the gold standard

diagnostic test to evaluate patients with suspectedventricular dysfunction but it is expensive and not alwayseasily available.4

Natriuretic peptides (cardioneurohormones)—atrialnatriuretic peptide (ANP) and brain natriuretic peptide(BNP)—are secreted from the heart in response to increasedintracardiac volume or pressure.5,6 Brain natriureticpeptides are specifically secreted from the ventricles inresponse to volume expansion or pressure overload. Levelsof BNP have been shown to be elevated in patients with leftventricular dysfunction and correlate with the New YorkHeart Association (NYHA) classification, as well asprognosis.7–10

Recently, it has been reported that N-terminal pro-brainnatriuretic peptide (NT-proBNP) levels are elevated inpatients with cardiac impairment.11,12 NT-proBNP is derivedfrom proBNP when it splits into BNP. In normal subjects,NT-proBNP levels are similar to those of BNP, whereas in

Original Article

Correspondence: Dr Jacob V Jose, Department of Cardiology, CMC Hospital,Vellore 632004

Background: The goal of this study was to evaluate the utility of plasma N-terminal pro-brain natriureticpeptide for the diagnosis of heart failure in patients presenting with shortness of breath.Methods and Results: We measured plasma levels of N-terminal pro-brain natriuretic peptide in 119 patientspresenting with shortness of breath. The patients were divided into two groups based on the Framingham criteriaand echocardiographic results—those with heart failure and those not in heart failure. Plasma levels ofN-terminal pro-brain natriuretic peptide were compared in the two groups. The mean N-terminal pro-brainnatriuretic peptide concentration in patients with heart failure (n=73) was higher than that in those not inheart failure (389±148 fmol/ml v. 142±54 fmol/ml, p<0.001). N-terminal pro-brain natriuretic peptide valuesincreased significantly as the functional severity of heart failure increased (p<0.001). The mean N-terminalpro-brain natriuretic peptide levels were 261±34 fmol/ml for patients in New York Heart Association functionalclass I, 300±161 fmol/ml for patients in New York Heart Association functional class II, 427±103 fmol/ml forpatients in New York Heart Association functional class III and 528±170 fmol/ml for patients in New YorkHeart Association functional class IV. Using a cut-off value of 200 fmol/ml, the sensitivity of N-terminalpro-brain natriuretic peptide was 97%, specificity was 89% and accuracy for differentiating heart failure fromother causes of shortness of breath was 93%.Conclusions: Our results suggest that N-terminal pro-brain natriuretic peptide can be reliably used for thediagnosis of heart failure in an outpatient setting, and this will improve the ability of clinicians to differentiatepatients with shortness of breath due to heart failure from those with other causes of shortness of breath.(Indian Heart J 2003; 55: 35–39)

Key Words: Heart failure, Natriuretic peptide, Echocardiography

IHJ-378-02.p65 5/6/2003, 9:59 AM35

36 Jose et al. Utility of NT-proBNP for the Diagnosis of Heart Failure Indian Heart J 2003; 55: 35–39

those with cardiac impairment, the proportional andabsolute increment of NT-proBNP above the normal levelexceeds that of BNP, which suggests that it may be a morediscerning marker.13,14 Though there are a number ofstudies regarding the role of BNP in HF, NT-proBNP hasnot been studied extensively. Thus, this study wasundertaken to evaluate the utility of NT-proBNP for thediagnosis of HF.

Methods

Patient characteristics: One hundred nineteen patientspresenting with shortness of breath (either acute orchronic) over a 6-month period to our emergency andoutpatient departments were enrolled in the study.Associated symptoms were edema, weight gain, cough, orwheezing.

Patients with acute coronary syndromes were

excluded.All the subjects underwent physical examination, chest

X-ray, supine 12-lead electrocardiogram (ECG) and bloodpressure (BP) measurement. Other data recorded includedhistory, risk factors, and drug treatment.

Echocardiography

was performed in all the patients using a Sonos 5500machine.

Left ventricular end-systolic and diastolic

volumes were measured, and ejection fraction determinedbased on the American Society of Echocardiographyrecommendations.15

NT-proBNP measurement: Two ml of blood was collectedin a chilled clotted tube and centrifuged immediately. Theplasma was stored at –80 °C until the NT-proBNP assay wascarried out. Plasma NT-proBNP concentration wasmeasured using a Biomedica kit.

This kit was a competitive

enzyme immunoassay (EIA) designed to measure theimmunoreactive NT-proBNP in diluted human serum,plasma or urine samples.

To achieve high specificity, the kit

incorporates an immunoaffinity purified sheep antibodyspecific for NT-proBNP (8-29) immobilized to the surfaceof a microtiter plate well.

The assay is based on competitive reaction of theunlabeled peptide in the standards or samples, andhorseradish peroxidase-labeled peptide (tracer) forthe limiting binding sites of the NT-proBNP (8-29)specific antibody.

The assay had a detection limit (±2 SD)

of 5 fmol/ml. The intra-assay and interassay coefficients of

variation were 6.5% and 4.4%, respectively. The assay was

performed within 3 months of sampling. The author, who

was involved in NT-proBNP measurement, was blinded toclinical information regarding HF.

Diagnosis of heart failure: We divided our patients into

two groups: patients with shortness of breath due to HFand patients with shortness of breath not due to HF. Thediagnosis of HF was based on the Framingham criteria andechocardiography.16 For patients with a diagnosis otherthan HF, confirmation was attempted using the followingvariables: normal chest X-ray and normal left ventricularfunction by echocardiogram.

In the group of patients with

HF, there were patients with systolic and diastolicdysfunction: of the 73 patients with HF, 14 had diastolicdysfunction.

Statistical analysis: To evaluate the utility of NT-proBNPmeasurement for the diagnosis of HF, we computedsensitivity, specificity, and accuracy. All the values areexpressed as mean±2 SD. Comparison among valuesobtained for the HF group and for those not in HF was madeby multivariate analysis. A value of p<0.05 was consideredstatistically significant. The plasma level of NT-proBNP wasalso compared in each stage of NYHA functional class inpatients with HF. A receiver–operating characteristics curvewas obtained to determine various cut-off values ofNT-proBNP peptide for the diagnosis of HF.

Baseline characteristics: The baseline characteristics ofthe overall study group of 119 patients are shown inTable 1. The mean age of the study population was

Table 1. Baseline characteristics of patients with andwithout heart failure

Heart failure No heart failureCharacteristics n=73 (61.3%) n=46 (38.7%)

Number (%) or mean±SD

DemographicsAge (years) 53.35±11.62 54.26±13.37Male sex 51 (69.8) 27 (58.6)

HistoryHypertension 33 (45.2) 16 (34.7)Diabetes mellitus 27 (36.9) 14 (30.7)History of MI 28 (38.3) 6 (13.0)Smoking 31 (42.4) 26 (56.5)

SymptomsDyspnea 73 (100) 46 (100)Orthopnea 42 (57.5) 3 (6.5)Nocturnal cough 16 (21.9) 11 (23.9)

SignsRales 49 (67.1) 17 (36.9)Increased JVP 50 (68.4) 11 (23.9)Ankle edema 52 (71.2) 5 (10.8)S

3 gallop 36 (49.3) 2 (4.3)

Resting tachycardia 26 (35.6) 15 (32.6)Cardiomegaly on CXR 44 (60.2) 9 (19.5)

MI: myocardial infarction; JVP: jugular venous pressure; CXR: chest X-ray

IHJ-378-02.p65 5/6/2003, 9:59 AM36


53.7±12.4 years. There were 78 men (65.5%) and 41women (34.5%). Patients were grouped according towhether they had HF or not, based on the Framinghamcriteria. The final assessment revealed that 73 patients(61.3%) had HF as a cause of their shortness of breath,while 46 (38.7%) had a cause other than HF; these includedbronchial asthma, chronic obstructive pulmonary disease,pneumonia, carcinoma lung, psychogenic dyspnea. Therewas no significant difference in the age and sex ratio in boththe groups of patients. Patients with HF were more likelyto have had a previous history of myocardial infarction andclinical signs of HF.

Results

NT-proBNP and final diagnosis: Patients with a finaldiagnosis of HF had a mean (±SD) NT-proBNP level of389±148 fmol/ml compared with 142±54 fmol/ml inpatients not in HF. The difference between the groups wasstatistically significant (p<0.001).

Figure 1 is a box plot of

NT-proBNP values for patients with HF and those not inHF.

NT-proBNP and clinical severity of heart failure: Themean NT-proBNP level was 261±34 fmol/ml for NYHAfunctional class I, 300±161 fmol/ml for NYHA functionalclass II, 427±103 fmol/ml for NYHA functional class IIIand 528±170 fmol/ml for NYHA functional class IVpatients. NT-proBNP values increase significantly as thefunctional severity of HF increases (p<0.001). Figure 2shows NT-proBNP values in relation to the NYHAfunctional class.

NT-proBNP and left ventricular ejection fraction:There was a significant negative linear correlation betweenplasma NT-proBNP and the left ventricular ejection fraction(r=–0.69;

p<0.001) (Fig. 3). The level of NT-proBNP in

patients with a left ventricular ejection fraction less than40% was significantly higher in comparison to its level inpatients with left ventricular ejection fraction more than40% (406±152 fmol/ml v. 171±83 fmol/ml).

Receiver–operating characteristics curve: The capacityof NT-proBNP to differentiate between shortness of breathdue to HF and that due to other causes was assessed with areceiver–operating characteristics curve analysis (Fig. 4).The area under the receiver–operating characteristics curvewhen NT-proBNP was used to differentiate HF from othercauses of shortness of breath was 0.94 (95% confidenceinterval: 0.90–0.98; p<0.001). An NT-proBNP cut-offvalue of 200 fmol/ml had a sensitivity of 97%, a specificityof 89% and an accuracy of 93% for differentiating betweenHF and other causes of shortness of breath.

Multivariate analysis: In multiple logistic regressionanalysis, we found that NT-proBNP was a strongindependent predictor for HF with an odds ratio of 8.94

Fig. 1. N-terminal pro-brain natriuretic peptide levels in patients with andwithout heart failure. The boxes show interquartile ranges, and the bars representthe highest and lowest values.

Fig. 2. Plots showing the median level of N-terminal pro-brain natriureticpeptide levels among patients in each New York Heart Association functionalclass. Boxes show interquartile ranges, and bars represent the highest and lowestvalues.

Fig. 3. Scatter plot showing the correlation of the left ventricular ejectionfraction (LVEF) with plasma levels of N-terminal pro-brain natriuretic peptide(NT-proBNP). There was a significant negative linear correlation between LVEFand NT-proBNP.

IHJ-378-02.p65 5/6/2003, 9:59 AM37

38 Jose et al. Utility of NT-proBNP for the Diagnosis of Heart Failure Indian Heart J 2003; 55: 35–39

(95% confidence interval: 3.9–20.48). The other variable,

which was better than NT-proBNP, was the presence of S3

gallop (Table 2).

enough during an echocardiographic study for a goodimage to be obtained. In addition, technical difficulty dueto obesity and lung disease may also interfere in getting agood echocardiographic image. Therefore, even

if

echocardiography is available in the emergencydepartment, an accurate, sensitive, and specific blood testfor HF would be a useful addition to the clinicalarmamentarium.

Brain natriuretic peptide measurement

is a useful tool for evaluating possible left ventriculardysfunction and ventricular failure.10 Insufficient data areavailable on NT-proBNP, but BNP and NT-proBNP appearto be equivalent markers.17 In our study, the NT-proBNPlevel was significantly higher in patients with HF incomparison with those not in HF. Our results concur withthose reported previously from the West.18–22

Our results suggest that the levels of NT-proBNP increasewith increasing severity of HF.

NT-proBNP levels were

higher in patients in a higher NYHA class. Our results are

in accordance with the study by Hunt et al.12 Patients withleft ventricular ejection fraction <40% had NT-proBNPlevels of 406±152 fmol/ml, whereas patients with leftventricular ejection fraction >40% had values of 171±83fmol/ml.

In the study by Campbell et al.,21 NT-proBNP had

a sensitivity of 100% for differentiating patients with leftventricular ejection fraction <45%.

In our study, NT-

proBNP levels showed a negative correlation with the leftventricular ejection fraction (Fig. 3). When we used a cut-off value of 200 fmol/ml for NT-proBNP, the sensitivity was97% for the diagnosis of HF. Our findings suggest that NT-proBNP could be a reliable marker for the diagnosis of HF.

Our results suggest that the determination of cardiacventricular peptides substantially improves the diagnosticaccuracy of the assessment of left ventricular dysfunctionin a patient population with suspected HF. The resultsdemonstrate that measurement of NT-proBNP levels in theblood improves the ability of the clinician to differentiatebetween patients with shortness of breath due to HF andshortness of breath due to other causes in an acute caresetting. This should be especially true among patients inwhom the diagnosis of HF is not clinically obvious. TheBNP/NT-proBNP test is now available in a rapid form, thusmaking diagnostic information immediately available to theacute care physician. Use of this test in conjunction withother clinical information should lead to a more accurateinitial diagnosis of HF.

Limitations of the study: As this study was conductedin a tertiary care center, it needs to be confirmed whetherthese results can be generalized to the primary care level.A larger study, including the patient’s clinical findings,

Fig. 4. Receiver–operating characteristics curve for various cut-off levels of N-terminal pro-brain natriuretic peptide for the diagnosis of heart failure. Thenumber in parenthesis is the 95% confidence interval. AUC: area under thereceiver–operating characteristics curve

Table 2. Multivariate logistic regression analysis ofsignificant variables for differentiating patients with andwithout heart failure

Variable Odds ratio 95% CI p value

Rales 1.8 1.20–2.730 0.001Increased JVP 2.86 1.67–4.900 <0.001Cardiomegaly 3.08 1.66–5.690 <0.001Ankle edema 6.55 2.82–15.18 <0.001Orthopnea 8.8 2.90–26.81 <0.001NT-proBNP 8.94 3.9–20.48 <0.001S

3 gallop 11.34 2.86–44.87 <0.001

CI: confidence interval; JVP: jugular venous pressure; NT-proBNP: N-terminal pro-brain natriuretic peptide

Discussion

Our study was designed to assess the diagnostic value ofthe plasma level of circulating NT-proBNP as a noninvasiveindicator of HF. Our results suggest that NT-proBNP canbe used as a reliable marker for identification of leftventricular dysfunction in a group of patients presentingwith shortness of breath. We found that a cut-off value of200 fmol/ml had an accuracy of 93%.

Echocardiography, although currently the gold standardfor the diagnosis of left ventricular dysfunction, is costlyand has limited availability in an urgent care setting.Dyspneic patients may be unable to hold their breath long

IHJ-378-02.p65 5/6/2003, 9:59 AM38


chest X-ray, and NT-proBNP level may be worthwhile. Also,

we did not look at the group of patients who could not bediagnosed to have HF by routine clinical methods.

Such a

study will give the incremental value of NT-proBNP overroutine clinical methods.

Conclusions: Plasma NT-proBNP is a sensitive indicatorof cardiac dysfunction. Our results suggest that NT-proBNPcan be used reliably for the diagnosis of HF in an outpatientsetting. The results demonstrate that measurement of NT-proBNP levels will improve the ability of clinicians todifferentiate between patients with shortness of breath dueto HF and that due to other causes.

Acknowledgment

We would like to express our thanks to Mr Pramod,statistician, for helping us with the statistical work.

References

1. Cheitlin MD, Zipes DP. Cardiovascular disease in special population.

In: Braunwald E, Zipes DP, Libby P (eds). Heart disease: a text book ofcardiovascular medicine. 6th ed. Philadelphia: WB Saunders Co; 2001.p. 2019

2. Kannel WB, Belanger AJ. Epidemiology of heart failure. Am Heart J

1991; 121: 951–9573. O’Connell JB, Bristow MR. Economic impact of heart failure in the

United States: time for a different approach. J Heart Lung Transplant1994; 13 (Suppl): S107–S112

4. Dao Q, Krishnaswamy P, Kazanegra R, Harrison A, Amirnovin R,Lenert L, et al. Utility of B-type natriuretic peptide in the diagnosis ofcongestive heart failure in an urgent-care setting. J Am Coll Cardiol2001; 37: 379–385

5. Nagagawa O, Ogawa Y, Itoh H, Suga S, Komatsu Y, Kishimoto I, et al.Rapid transcriptional activation and early mRNA turnover of brainnatriuretic peptide in cardiocyte hypertrophy: evidence for brainnatriuretic peptide as an “emergency” cardiac hormone againstventricular overload.

J Clin Invest 1995; 96: 1280–1287

6. Maeda K, Tsutamoto T, Wada A, Hisanaga T, Kinoshita M. Plasma

brain natriuretic peptide as a biochemical marker of high leftventricular end-diastolic pressure in patients with symptomatic leftventricular dysfunction.

Am Heart J 1998; 135: 825–832

7. Clerico A, Iervasi G, Del Chicca MG, Emdin M, Maffei S, Nannipieri M,et al.

Circulating levels of cardiac natriuretic peptides (ANP and BNP)

measured by highly sensitive and specific immunoradiometric assaysin normal subjects and in patients with different degrees of heartfailure.

J Endocrinol Invest 1998; 21: 170–179

8. Omland T, Aakvaag A, Bonarjee VV, Caidahl K, Lie RT, Nilsen DW, etal.

Plasma brain natriuretic peptide as an indicator of left ventricular

systolic function and long-term survival after acute myocardialinfarction. Comparison with plasma atrial natriuretic peptide and

N-terminal proatrial natriuretic peptide Circulation 1996; 93: 1963–

19699. Krishnaswamy P, Lubien E, Clopton P, Koon J, Kazanegra R, Wanner

E, et al. Utility of B-natriuretic peptide levels in identifying patients

with left ventricular systolic or diastolic dysfunction. Am J Med 2001;

111: 274–27910. Maisel AS, Krishnaswamy P, Nowak RM, McCord J,

Hollander JE, Duc

P, et al. Rapid measurement of B-type natriuretic peptide in theemergency diagnosis of heart failure. N Engl

J Med 2002; 347: 161–

16711. Hunt PJ, Yandle TG, Nicholls MG, Richards AM, Espiner EA.

The

amino-terminal portion of pro-brain natriuretic peptide (Pro-BNP)circulates in human plasma.

Biochem Biophys Res Commun 1995; 214:

1175–118312. Hunt PJ, Richards AM, Nicholls MG, Yandle TG, Doughty RN, Espiner

EA. Immunoreactive amino-terminal pro-brain natriuretic peptide(NT-ProBNP): a new marker of cardiac impairment.

Clin Endocrinol

1997; 47: 287–29613. Richards AM, Nicholls MG, Yandle TG, Frampton C, Espiner EA,

Turner JG, et al. Plasma N-terminal pro-brain natriuretic peptide and

adrenomedullin: new neurohormonal predictors of left ventricularfunction and prognosis after myocardial infarction.

Circulation 1998;

97: 1921–192914. Hughes D, Talwar S, Squire IB, Davies JE, Ng LL.

An

immunoluminometric assay for N-terminal pro-brain natriureticpeptide: development of a test for left ventricular dysfunction.

Clin Sci

(Lond) 1999; 96: 373–38015. Schiller NB, Shah PM, Crawford M, DeMaria A, Devereux R,

Feigenbaum H, et al. Recommendations for quantification of the leftventricle by two-dimensional echocardiography. American Society ofEchocardiography Committee on Standards, Subcommittee onQuantitation of Two-Dimensional Echocardiograms. J Am SocEchocardiogr 1989; 2: 358–367

16. Ho KK, Pinsky JL, Kannel WB, Levy D. The epidemiology of heart

failure: the Framingham Study. J Am Coll Cardiol 1993; 22: 6A–13A

17. Mair J, Hammerer-Lercher A, Puschendorf B. The impact of cardiacnatriuretic peptide determination on the diagnosis and managementof heart failure. Clin Chem Lab Med 2001; 39: 571–588

18. Schulz H, Langvik TA, Lund Sagen E, Smith J, Ahmadi N, Hall C.Radioimmunoassay for N-terminal probrain natriuretic peptide inhuman plasma.

Scand

J Clin Lab Invest 2001; 61: 33–42

19. Hammerer-Lercher A, Neubauer E, Muller S, Pachinger O,Puschendorf B, Mair J.

Head-to-head comparison of N-terminal pro-

brain natriuretic peptide, brain natriuretic peptide and N-terminalpro-atrial natriuretic peptide in diagnosing left ventriculardysfunction. Clin Chim Acta 2001; 310: 193–197

20. Pemberton CJ, Yandle TG, Rademaker MT, Charles CJ, Aitken GD,Espiner EA.

Amino-terminal proBNP in ovine plasma: evidence for

enhanced secretion in response to cardiac overload. Am J Physiol1998; 275: H1200–H1208

21. Campbell DJ, Mitchelhill KI, Schlicht SM, Booth RJ. Plasma amino-

terminal pro-brain natriuretic peptide: a novel approach to thediagnosis of cardiac dysfunction.

J Card Fail 2000; 6: 130–139

22. Talwar S, Squire IB, Downie PF, Mccullough AM, Campton MC, DaviesJE, et al.

Profile of plasma N-terminal proBNP following acute

myocardial infarction; correlation with left ventricular systolicdysfunction. Eur Heart J 2000; 21: 1514–1521

IHJ-378-02.p65 5/6/2003, 9:59 AM39

40 Baran et al. IV Metoprolol in Patients with Positive ITTT Indian Heart J 2003; 55: 40–43

Usefulness of Intravenous Metoprolol During PositiveIsoproterenol Tilt-Table Test in the Choice of Treatment for

Neurocardiogenic Syncope

Ibrahim Baran, Kani Gemici, Bülent Özdemir, Murat Saraç, Sümeyye Güllülü, Ali Aydinlar, Jale CordanDepartment of Cardiology, Uludag University, School of Medicine, Bursa, Turkey

Hypotension and bradycardia associated withneurocardiogenic syncope are thought to be due to a

reflex triggered by enhanced left ventricularmechanoreceptor activity in response to diminishedventricular volume and accentuated ventricularcontractility.1 The medications generally considered to beeffective in the treatment of neurocardiogenic syncopeinclude beta-blockers, selective serotonin re-uptake

inhibitors (SSRIs), and midodrine.2 Increased sympatheticactivity preceding this reflex

may intensify the response.3

Beta-receptor blockers such as metoprolol have been usedin the treatment of neurocardiogenic syncope due tonegative inotropic and chronotropic effects as well as thepresumed effects on central mechanoreceptors.4

We carried out this study to evaluate whether the useof intravenous metoprolol in preventing symptoms duringthe positive isoproterenol tilt-table test (ITTT) would behelpful in selecting patients for long-term therapy withbeta-blockers or SSRIs (sertraline or paroxetine) andmidodrine.

Original Article

Correspondence: Dr Kani Gemici, Uludag University, School of Medicine,Department of Cardiology, 16059 Görükle, Bursa, Turkey.e-mail: [email protected]

Background: Isoproterenol tilt-table testing provides a diagnosis of neurocardiogenic syncope in patients withsyncope or near-syncope. Although acute beta-blockade may prevent the development of syncope duringisoproterenol tilt-table testing, the use of beta-blockers for chronic prophylaxis may not be effective for somepatients who show a positive response to isoproterenol tilt-table testing. We evaluated whether the efficacy ofintravenous metoprolol in preventing symptoms during repeated tests would be helpful in selecting patientssuitable for long-term therapy.Methods and Results: We studied 55 patients (35 females, 20 males; mean age 36±11 years) who had beenchosen from a group referred to our institute with a history of unexplained syncope (>2 syncopal episodes) anda positive response to isoproterenol tilt-table testing. After a positive response to isoproterenol tilt-table testing,5 mg metoprolol was infused intravenously as a bolus and the test repeated. Thirty-five patients (group 1) showeda positive response again and 20 (group II) showed a negative response. We started 50 mg metoprolol once aday for patients in group 1 while group 2 was divided into 2 subgroups: the first subgroup (group 2a, 12 patients)was started on 50 mg sertraline or 20 mg paroxetine once a day and the second subgroup (group 2b, 8 patients)was started on 5 mg midodrine orally once a day. Two months later, isoproterenol tilt-table testing was repeated.In group 1, 13 of 35 patients (37%) were positive on isoproterenol tilt-table testing while in group 2, 8 of 20patients (40%) were positive on isoproterenol tilt-table testing (p not statistically significant). The therapies ofthe two groups were then interchanged. Two months later (4 months from the beginning of the study), theisoproterenol tilt-table test was repeated. Eleven patients in group 1 (31%) and 6 in group 2 (30%, p notstatistically significant) showed a positive response again.Conclusions: We conclude that acute beta-blockade response to positive isoproterenol tilt-table testing is not auseful predictor for the assessment of chronic prophylaxis for neurocardiogenic syncope. (Indian Heart J 2003;55: 40–43)

Key Words: Neurocardiogenic syncope, Tilt-table test, Beta-blockers

.

IHJ-368-02.p65 5/6/2003, 10:00 AM40

Indian Heart J 2003; 55: 40–43 Baran et al. IV Metoprolol in Patients with Positive ITTT 41

Methods

Patients: We studied 89 patients (53 females and 36 males;mean age 38±16 years)

, who had been referred to our

institute with a history of unexplained syncope (>2syncopal episodes). Cardiac and neurologic examination,electrocardiogram, chest X-ray and echocardiogram werenormal in all the patients. Fifty-five patients (35 females,20 males; mean age 36±11 years) with a positive responseto ITTT (Table 1) formed the study group.

Around 1 week later, the ITTTs were repeated with

standard protocol and completed after an infusion of 5 mgmetoprolol (Table 2). Thirty-five patients (group 1) againshowed a positive response

while 20 (group 2) showed a

negative response. We started patients in group 1 on 50mg metoprolol once a day. Group 2 was divided into twosubgroups; the first subgroup: group 2a (12 patients) wasstarted on 50 mg sertraline or 20 mg paroxetine once aday

and the second subgroup: group 2b (8 patients) on 5

mg midodrine orally once a day. Two months later, theITTTs were repeated. During the ITTT, 3 patterns ofpositive response were observed. A vasodepressor response

was associated with a marked decrease in systolic bloodpressure (a decrease of at least 20 mmHg in the systolicblood pressure). A cardioinhibitory response wascharacterized by a marked decrease in heart rate (≥20%)or asystole ≥5 s at the onset of symptoms. A mixed patternof response was characterized by a decrease in both bloodpressure and heart rate, frequently associated with thedevelopment of junctional rhythm.

Tilt-table test protocol: A modified protocol was speciallydesigned for our clinic (Table 2). Patients were informedbefore the test, and the procedure performed in the morningbetween 9 and 10 a.m. after all cardioactive and vasoactivemedications had been stopped for at least 5 half-lives. Thepatients had only

a light breakfast 2 hours before the test.

The study room was cool, semilighted, quiet, and free ofdisturbing factors such as noise. The patients lay on ahydraulic table for 15 min in the supine position. Heart rate,electrocardiographic, and noninvasive blood pressuremonitoring were performed continuously. Baselinemeasurements were obtained every minute in the supineposition.

Patients were then tilted to an 80° upright position

using a handboard on the hydraulic table.. Heart rate, blood

pressure, electrocardiographic changes, and symptomswere recorded every 1–2 min.

The end-point was either

development of symptoms of syncope or presyncopeaccompanied by substantial hypotension with or withoutbradycardia, or at the end of a 25-min head-up tilt-tabletest.

Definitions: A positive head-up tilt-table test was definedas the development of hypotension or bradycardia (orasystole), or both, to a degree sufficient to cause syncopeor presyncope. Syncope was defined as a transient state ofunconsciousness characterized by recovery in the supine

Table 2. The ITTT protocol form prepared for our clinic

Name and surname:Age:Symptoms:

Stages Minutes Tilt Systolic BP Heart rate ECG Symptoms Isoproterenol Metoprolol

Step I 0, 3, 5, 10 and 15 0° – –Step II 20, 25, 30, 35, 40 and 45 80° – –Step III 47 and 50 0° 3 mcg/min –Step III 53 80° 3 mcg/min –Step III 55,57 and 60 80° 5 mcg/min –

Step IV 61 and 62 0° 5 mcg/min 5 mg i.v.Step IV 64, 66, 68 and 70 80° 5 mcg/min –

ITTT: isoproterenol tilt-table test; BP: blood pressure; ECG: electrocardiogram

Table 1. Clinical characteristics of the study groups

Group 1 Group 2a Group 2b

Number of patients 35 12 8Mean age (years) 37±12 34±10 35±11Women/men ratio 22/13 7/5 5/3Syncopal episodes/year 5 4 6Time to positive ITTT (seconds) 39 47 44Baseline heart rate (bpm) 110 106 103Baseline systolic BP (mmHg) 118 121 114

ITTT: isoproterenol tilt-table test; BP: blood pressure; bpm: beats perminute; p value not statistically significant

IHJ-368-02.p65 5/6/2003, 10:00 AM41

42 Baran et al. IV Metoprolol in Patients with Positive ITTT Indian Heart J 2003; 55: 40–43

position. Presyncope was defined as a state oflightheadedness associated with symptoms of decreasedvision, nausea, vomiting, partial loss of postural tone orslow response times to verbal stimuli, which substantiallyreproduced the clinical presyncope of the patients.

Follow-up: All the patients were followed up to detectrecurrent syncopal or presyncopal attacks, and alsocomplaints related to adverse effects of the drugs. Twomonths later, the ITTTs were repeated, and the previoustreatment changed to an alternative one.

Statistical analysis: Data are expressed as mean±SEM.The paired and unpaired Student’s t test was used tocompare all numerical variables among the groups. Ap value <0.05 was considered statistically significant.

Results

The study population consisted of 35 females and 20 maleswith a positive response to the ITTT. They had had a medianof 4.8 syncopal attacks per year. During the second testafter a positive response to the ITTT, 5 mg metoprolol wasinfused as a bolus

in the resting supine position and the

patient was tilted again to 80°. Thirty-five patients(group 1) showed a

positive response again while 20

(group 2) showed a negative response with metoprolol.Group 1 was started on 50 mg metoprolol orally once aday and group 2 subdivided into two subgroups—the groupgiven SSRIs and that given midodrine. The SSRIs group(group 2a) was started on 50 mg sertraline or 20 mgparoxetine once a day

, and the midodrine group (group 2b)

was started on 5 mg midodrine once a day orally. No patientsuffered from side-effects necessitating stoppage of therapy.Two months later, the ITTTs were repeated. In group 1, 13of 35 ITTTs (37%) were positive while 8 of 20 ITTTs (40%)were positive in group 2 (p not statistically significant).The therapies of the two groups were interchanged.

The

patients in group 1 were given sertraline/paroxetine ormidodrine randomly. Two months later (4 months fromthe beginning of the study), the ITTTs were repeated.Eleven patients in group 1 (31%) and 6 patients in group2 (30%, p not statistically significant) again showed apositive response to the ITTTs.

Discussion

The treatment of patients with neurocardiogenic syncope

is controversial. Neither the underlying mechanisms northe best modality of treatment have been clarified. Manydrugs, including disopyramide, beta-blockers, scopolamine,

fludrocortisone, midodrine, SSRIs, and etilefrine have beenused for treatment.5–9

The tilt-table test is an important diagnostic tool for theevaluation of patients with unexplained syncope.10,11

ITTTprovides a diagnosis of neurocardiogenic syncope in manypatients who faint.12,13

Beta-blockers have been widely used in the treatmentof patients with neurocardiogenic syncope and arefrequently chosen as first-line drug therapy. Metoprolol andatenolol have been the most frequently studied beta-blockers in neurocardiogenic syncope. Following a positivetilt-table test, treatment with beta-blockers may not have asubstantial effect in preventing recurrence of syncope.14

Mahanonda et al.15 performed the only randomized, double-blind, placebo-controlled trial to evaluate the efficacy oforal beta-blockers in the treatment of neurocardiogenicsyncope. They compared the efficacy of a beta-blocker witha placebo in patients who had had at least 1 episode ofsyncope or 2 episodes of presyncope within one monthbefore presentation.

. The response rate after one month of

treatment was 62% versus 5% in the atenolol and controlgroups, respectively. However, the efficacy of beta-blockerswas also examined in other studies, yielding controversialresults. Sheldon et al.,14 for example, concluded thattreatment with beta-blockers had no major effect onrecurrence of syncope.

Cox et al.16 prospectively evaluated

the efficacy of propranolol in preventing neurocardiogenicsyncope. Oral beta-blockers were found to be effective bythe tilt-test criteria in the majority (94%) of patients. Whiletaking beta-blockers, 10% of patients had recurrent clinicalsymptoms. In conclusion, they stated that intravenouspropranolol was very effective in blocking the neuro-cardiogenic reflex during tilt-table testing, and that it couldpredict a good response to oral beta-blockers.

In our study, we evaluated the efficacy of beta-blockerscompared with midodrine and SSRIs in preventingrecurrences in patients with clinically diagnosedneurocardiogenic syncope. We followed up the patients forat least 4 months.

Due to the acute ef ficacy of an

intravenous beta-blocker during a positive ITTT, we hopedthat it would also be useful for long-term therapy. However,in the long term, beta-blockers did not show as good anefficacy as expected. Also, the positive response tointravenous metoprolol did not reliably predict a goodresponse to long-term oral administration.

Conclusions: The study is ongoing. We conclude that

acute beta-blockade response to a positive ITTT is not auseful predictor of the success of beta-blockers in long-termprophylaxis of neurocardiogenic syncope.

IHJ-368-02.p65 5/6/2003, 10:00 AM42

Indian Heart J 2003; 55: 40–43 Baran et al. IV Metoprolol in Patients with Positive ITTT 43

References

1. Brignole M, Alboni P, Benditt D, Bergfeldt L, Blanc JJ, Bloch ThomsenPE, et al. Guidelines on management (diagnosis and treatment) ofsyncope. Eur Heart J 2001; 22: 1256–1306

2. Boudoulas H, Nelson SD, Schaal SF, Lewis RP. Diagnosis andmanagement of syncope. In: Hurst’s the heart. 9th ed.

McGraw-Hill;

1998. pp. 1059–10803. Perry JC, Garson A Jr. The child with recurrent syncope: autonomic

function testing and beta-adrenergic hypersensitivity. J Am Coll Cardiol1991; 17: 1168–1171

4. O’Marcaigh AS, MacLellan-Tobert SG, Porter CJ. Tilt-table testing andoral metoprolol therapy in young patients with unexplained syncope.Pediatrics 1994; 93: 278–283

5. Raviele A, Brignole M, Sutton R, Alboni P, Giani P, Menozzi C, et al.Effect of etilefrine in preventing syncopal recurrence in patients withvasovagal syncope: a double-blind, randomized, placebo-controlledtrial. The Vasovagal Syncope International Study. Circulation 1999;99: 1452–1457

6. Morillo CA, Leitch JW, Yee R, Klein GJ. A placebo-controlled trial ofintravenous and oral disopyramide for prevention of neurallymediated syncope induced by head-up tilt. J Am Coll Cardiol 1993;22: 1843–1848

7. Madrid AH, Ortega J, Rebollo JG, Manzano JG, Segovia JG, Sanchez A,et al. Lack of efficacy of atenolol for the prevention of neurallymediated syncope in a highly symptomatic population: a prospective,double-blind, randomized and placebo-controlled study. J Am CollCardiol 2001; 37: 554–559

8. Lee TM, Su SF, Chen MF, Liau CS, Lee YT. Usefulness of transdermalscopolamine for vasovagal syncope. Am J Cardiol 1996; 78: 480–482

9. Ward CR, Gray JC, Gilroy JJ, Kenny RA. Midodrine: a role in themanagement of neurocardiogenic syncope. Heart

1998; 79: 45–49

10. Fogoros RN. Electrophysiological testing, 3rd ed. Blackwell Science;1999

11. Braunwald E, Zipes DP, Libby PT (eds). Heart disease. A text book ofcardiovascular medicine. Philadelphia: WB Saunders Company; 2001.pp. 935–936

12. Kapoor WN, Brant N. Evaluation of syncope by upright tilt testingwith isoproterenol. A nonspecific test. Ann Intern Med 1992; 116:358–363

13. Zeng C, Zhu Z, Hu W, Liu G, Zhu S, Zhou Y, et al. Value of sublingualisosorbide dinitrate before isoproterenol tilt test for diagnosis ofneurally mediated syncope. Am J Cardiol 1999; 83: 1059–1063

14. Sheldon R, Rose S, Flanagan P, Koshman ML, Killam S. Effect of beta-blockers on the time to first syncope recurrence in patients after apositive isoproterenol tilt table test. Am J Cardiol 1996; 78: 536–539

15. Mahanonda N, Bhuripanyo K, Kangkagate C, Wansanit K, KulchotB, Nademanee K, et al. Randomized double-blind, placebo-controlledtrial of oral atenolol in patients with unexplained syncope and positiveupright tilt table test results. Am Heart J 1995; 130: 1250–1253

16. Cox MM, Perlman BA, Mayor MR, Silberstein TA, Levin E, Pringle L,et al. Acute and long-term beta-adrenergic blockade for patients withneurocardiogenic syncope. J Am Coll Cardiol 1995; 26: 1293–1298

IHJ-368-02.p65 5/6/2003, 10:00 AM43

44 Jadhav et al. Endothelial Dysfunction and Prediction of CAD Indian Heart J 2003; 55: 44–48

Noninvasive Assessment of Endothelial Dysfunction byBrachial Artery Flow-Mediated Dilatation in Prediction of

Coronary Artery Disease in Indian Subjects

Uday M Jadhav, Anand Sivaramakrishnan, NN KadamDepartment of Non-Invasive Cardiology, MGM New Bombay Hospital, Mumbai

Arterial endothelial dysfunction (ED) is one of the keyearly events in atherogenesis, preceding structural

atherosclerotic changes. It is also important in the latestages of obstructive atherosclerosis, predisposing toconstriction and/or thrombosis.1 Testing the function of theendothelium in vivo has proved challenging. Although theendothelium releases a number of products, no single bloodtest has yet proved useful in determining normal endothelialfunction or early abnormalities. The most common test usedfor the evaluation of endothelial function in vivo to datedepends on measuring endothelium-dependent dilatationin response to physiologic or pharmacologic stimuli. This

depends mostly on the ability of the endothelium to releasenitric oxide (NO), which is not only a potent vasodilatorbut also an inhibitor of platelet aggregation, monocyteadhesion, and smooth muscle proliferation.2

Endothelial function can be measured in the coronaryarteries and peripheral vasculature by measuringvasomotor function after intra-arterial infusion ofpharmacologic substances, which enhance the release ofendothelial NO. The disadvantage of this method is itsinvasive nature, which generally makes it unsuitable forstudies involving asymptomatic subjects. For this reason,noninvasive tests of endothelial function have beendeveloped. The most widely used of these tests, anultrasound-based method, involves measurement ofarterial diameter in response to shear stress, which causes

Original Article

Background: A noninvasive technique for testing endothelial function by ultrasound measurement of flow-mediated dilatation has recently generated considerable interest as a marker of atherosclerosis, and in theprediction of clinical coronary events and coronary artery disease.Methods and Results: We measured the flow-mediated dilatation of the brachial artery (endothelium-dependent vasodilatation) in 136 subjects, with or without evidence of coronary artery disease. Endothelialdysfunction was diagnosed if flow-mediated dilatation was less than 4.5%. Of the 136 subjects (age group40–70 years) recruited for the study, 94 were males and 42 females. Sixty-eight subjects had evidence of coronaryartery disease as diagnosed by documented hospitalization due to myocardial infarction or acute coronarysyndrome, proved by coronary angiography when feasible or noninvasive cardiac evaluation. Endothelialdysfunction was detected in 90 subjects (66.2%). Prevalence of coronary artery disease was higher amongsubjects with endothelial dysfunction compared to those without (57.5% v. 34.7%, p=0.013). Prevalence ofendothelial dysfunction was significantly higher among subjects with coronary artery disease as compared tothose without coronary artery disease (76.4% v. 55.8%, p=0.012). The present study showed a sensitivity of76%, specificity of 44%, positive predictive value of 58% and negative predictive value of 65% for endothelialdysfunction in the prediction of coronary artery disease. Multiple regression analysis using coronary arterydisease as a dependent variable revealed a statistically significant association with endothelial dysfunction(p=0.033) even after the inclusion of traditional risk factors into the model.Conclusions: We conclude that endothelial dysfunction shows a strong association with coronary artery diseaseand can be a useful noninvasive tool for the evaluation of coronary artery disease. (Indian Heart J 2003; 55:44–48)

Key Words: Coronary artery disease, Endothelial dysfunction, Doppler ultrasound

Correspondence: Dr Uday M Jadhav, MGM New Bombay Hospital, Sector3, Vashi, Navi Mumbai 400703. e-mail: [email protected]

IHJ-363-02.p65 5/6/2003, 10:01 AM44

Indian Heart J 2003; 55: 44–48 Jadhav et al. Endothelial Dysfunction and Prediction of CAD 45

endothelium-dependent dilatation.1 Endothelium-independent dilatation can be assessed by measuring thearterial diameter after administration of sublingualnitroglycerin.

We undertook this study to assess the role of a simpletechnique of flow-mediated dilatation (FMD) measurementin predicting coronary artery disease (CAD). This studyassesses the endothelium-dependent dilatation of conduitvessels, and there are other methods for resistance vessels.Dif ferences may exist with regard to endothelium-dependent dilatation of the conduit versus resistancearteries, and these differences have potential implications.

Methods

A total of 136 patients, both hospitalized as well asoutpatient, were included in the study. Informed consentwas obtained from the study subjects who were dividedinto two groups: with CAD (68 subjects) and without CAD(68 subjects). Coronary artery disease was diagnosedalong with clinical presentation by electrocardiography(ECG), symptom-limited exercise stress test, Dopplerechocardiography, and coronary angiography, whenfeasible. Of the 68 subjects with CAD, 42 had evidenceof myocardial infarction or acute coronary syndrome,13 had evidence on coronary angiography, and 13 hadECG, stress test, and echocardiographic manifestationssuggestive of CAD. Of the 136 subjects, 20 had type 2diabetes mellitus, 51 had hypertension, 37 had diabeteswith concomitant hypertension, and 28 had no diabetesor hypertension. The mean age was 52 years with a rangeof 40–74 years. The subjects included 94 males and 42females.

The ultrasound method for measuring endothelium-dependent and endothelium-independent arterial dilatationhas been described previously.3,4 The brachial arterydiameter was measured on B-mode ultrasound images,with the use of a 7.0 MHz linear-array transducer withImagepoint Hx ultrasound equipment (Agilent Technology,India). The right brachial artery was studied in all thesubjects. Brachial artery endothelial function was studiedafter the subject had abstained from alcohol, caffeine, andsmoking for 8 h. Scans were obtained with the subject atrest, during reactive hyperemia and again with the subjectat rest. The subjects were asked to lie quietly for at least 10min before the first scan. The brachial artery was scannedin longitudinal section 2–15 cm above the elbow, and thecenter of the artery was identified when the clearest pictureof the anterior and posterior intimal layers was obtained.The transmit (focus) zone was set to the depth of the near

wall, because of the greater difficulty in evaluating the “m”line (the interface between the media and adventitia) of thenear wall as compared with that of the far wall.5 Depth andgain settings were set to optimize images of the interfacebetween the lumen and the arterial wall, and the imageswere magnified. Settings for operating the machine werenot changed during the study.

When a satisfactory transducer position was found, theskin was marked and the arm was kept in the same positionthroughout the study. A resting scan was obtained, and thevelocity of arterial flow measured with a pulsed Dopplersignal at a 70° angle to the vessel, with the range gate (1.5mm) in the center of the artery. Increased flow was theninduced by the inflation of a sphygmomanometer cuffplaced around the forearm (distal to the scanned part ofthe artery) to a pressure of 200 mmHg for 4.5 min, followedby release. A second scan was performed continuously for30 s before and 90 s after deflation of the cuff, including arepeated recording of flow velocity for the first 15 s afterthe cuff was released.

Flow-mediated dilatation was calculated, and theaverage results of the two observations recorded. Thesystem software was designed for online recording, whichwas assessed visually to ensure that the best possiblerecording was obtained and artefacts minimized.Automatic tracking was not performed. ECG was monitoredthroughout the scans and the artery diameter measuredat end-diastole. Interobserver and intraobserver variabilitywas studied in 20 healthy subjects with a mean age of 42.4years. Interobserver variability was 1.6% and intraobservervariability was 2.2%. This method is accurate andreproducible for measuring small changes in arterialdiameter6 with low rates of interobserver error inmeasuring FMD.3,4 Flow-mediated dilatation was presentedas the percent change from baseline to hyperemia.Endothelial dysfunction was defined as FMD <4.5%, as hasbeen previously described.7

Clinical examination included blood pressuremeasurement, cardiovascular examination, anthro-pometrical measurements, and body-mass index (BMI).Biochemical assessment included fasting blood sugar (FBS)and post-prandial blood sugar levels, glycated hemoglobin,urine for microalbumin, and comprehensive lipid profile.Plasma glucose and lipid estimation were done after anovernight fast of 12 h. Biochemical analysis was done onTechnicon RA-1000 Auto Analyser. Plasma glucose, serumcholesterol, serum triglycerides, and high-densitylipoprotein (HDL)-cholesterol (HDL-c) were estimated withkits supplied by AUTOPAK-Bayer Diagnostics, India. Plasmaglucose estimation was done by the GOD/POD method,

IHJ-363-02.p65 5/6/2003, 10:01 AM45


serum cholesterol by the enzymatic method andtriglycerides by the enzymatic calorimetric method. HDL-cwas estimated after precipitating low-density lipoprotein(LDL) and chylomicron fractions by the addition ofphosphotungstic acid in the presence of magnesium ionsand very low-density lipoprotein (VLDL). LDL-cholesterol(LDL-c) was calculated using the Friedewald formula,namely,

LDL-c = Total cholesterol – (Triglycerides

+ HDL)5

The baseline characteristics of the study group and therelevant brief results of the lipid profile are shown in Tables1 and 2.

Statistical analysis: Data collected were managed on anExcel spreadsheet. One-way ANOVA or Student’s t test was

used as appropriate to compare the mean of the continuousvariables. Chi-square test was used to compare proportion.Pearson’s correlation coefficient was used to determine ifan association existed among the risk factors. All the groupswere combined for Pearson’s correlation analysis. Multipleregression analysis was carried out using CAD as thedependent variable and factors such as BMI, smoking,hypertension, FBS, serum cholesterol, triglycerides, HDL-c,LDL-c as independent variables. The subjects werecompared for the age groups less than and greater than 50years, smokers and non-smokers, BMI <23 kg/m2 and >23kg/m2, LDL-c <130 mg/dl and >130 mg/dl, HDL-c <40mg/dl and >40 mg/dl, total cholesterol to HDL-c ratio <5.5and >5.5, triglycerides <200 mg/dl and >200 mg/dl, andthe subgroup of hypertension±diabetes mellitus. Allanalyses were performed with the SPSS version 10 and pvalues less than 0.05 were considered significant.

Results

The percentage distribution of FMD in subjects with variousrisk factors is shown in Table 3. There was a 66.2%prevalence of ED in the present study population. Insubjects with evidence of CAD, 76.4% had endothelialdysfunction as indicated by FMD <4.5% as compared to55.8% in those without obvious evidence of CAD (Table4). This was highly significant with p<0.001. Thecorrelation of FMD with CAD is illustrated in Fig.1. The oddsratio for FMD <4.5% was 3.90 in those with CAD as againstthose without obvious CAD, making it an important markerfor preclinical and clinical CAD.

The prevalence of CAD among subjects with FMD<4.5% was calculated by the following method:

Number of subjects with CADand FMD <4.5%

Prevalence (%) = ——————————–––––× 100Number of subjects withFMD <4.5%

The prevalence of CAD among subjects with FMD>4.5% and prevalence of FMD <4.5% in subjects with andwithout CAD were calculated in the same way. Theprevalence of CAD was significantly higher among subjectswith FMD <4.5% compared to those with FMD >4.5%(57.5% v. 34.7%, p=0.013). Prevalence of FMD <4.5% wassignificantly higher among subjects with CAD comparedto those without CAD (76.4% v. 55.8%, p=0.012). Multipleregression analysis using CAD as a dependent variablerevealed a statistically significant association withendothelial dysfunction (p=0.033) even after the inclusionof traditional risk factors into the model, as shown inTable 5.

Table 1. Clinical characteristics of the study group

Parameters CAD Non-CAD p value(n=68) (n=68)

Age (years) 56.53±11.47 48.74±11 0.0001Male (%) 50 (53.1) 44 (46.8) 0.265Smoking (%) 51 (75) 17 (25) 0.012Hypertension (%) 21 (30.9) 30 (44.1) 0.374Diabetes mellitus (%) 10 (14.7) 10 (14.7) 0.374Body-mass index (kg/m2) 25.68±4.14 26.03±3.66 0.602Fasting plasma glucose (mg/dl) 114.61±41.92 114.11±50.39 0.951Total cholesterol (mg/dl) 194.93±43.71 184.73±38.66 0.270LDL-cholesterol (mg/dl) 122.26±41.29 111.18±33.89 0.110HDL-cholesterol (mg/dl) 40.93±5.80 41.73 ±6.10 0.457Triglycerides (mg/dl) 155.02±66.96 160.44±107.70 0.754Cholesterol–HDL ratio 4.79±1.11 4.52±0.92 0.157

Values in parentheses are percentagesLDL: low-density lipoprotein; HDL: high-density lipoprotein

Table 2. Clinical characteristics of the study group in thecontext of endothelial dysfunction

Parameters FMD FMD p value<4.5% >4.5%

Age (years) 53.73±11.67 50.48±12.06 0.130Male (%) 67 (74.4) 27 (58.7) 0.060Smoking (%) 71 (78.9) 42 (91.3) 0.068Hypertension (%) 29 (32.2) 22 (47.8) 0.048Diabetes mellitus (%) 15 (16.7) 5 (10.9) 0.072Body-mass index (kg/m2) 25.48±3.89 26.61±3.85 0.109Fasting plasma glucose (mg/dl) 118.7±49.1 106.0±39.1 0.140Total cholesterol (mg/dl) 192.7±77.9 141.7±66.8 0.270LDL-cholesterol (mg/dl) 120.0±39.7 110.9±34.5 0.206HDL-cholesterol (mg/dl) 41.1±6.3 41.7±5.4 0.634Triglycerides (mg/dl) 161.9±96.7 150.0±74.8 0.511Cholesterol–HDL ratio 4.8±1.1 4.4±0.8 0.096

Values in parentheses are percentagesLDL: low-density lipoprotein; HDL: high-density lipoprotein

IHJ-363-02.p65 5/6/2003, 10:01 AM46

Indian Heart J 2003; 55: 44–48 Jadhav et al. Endothelial Dysfunction and Prediction of CAD 47

The present study showed ED to have a sensitivity of76%, specificity of 44%, positive predictive value of 58%and negative predictive value of 65% in the prediction ofCAD. Changes in the flow velocity during the hyperemicphase were inconsistent and not statistically significant insubjects with and without CAD (group without CAD–34.25% to 83.10%, mean 11.03%, SD 28.64; versusgroup with CAD –22.06% to 51.91%, mean 14.29%, SD21.47, p>0.05).

Discussion

Ultrasonography is a reliable and accurate technique todetermine FMD in the superficial arteries. Reproduciblityof FMD determination is best in the brachial artery inhealthy subjects and in patients with atherosclerosis.3,6

B-mode ultrasound scan including brachial artery FMDmay be of clinical value in the screening of patients withCAD.6

In a study by Schroeder et al.,7 patients with CAD had asignificantly lower FMD% than patients without CAD. Theyreported a sensitivity of 71%, a specificity of 81% with apositive predictive value of 0.95 and a negative predictivevalue of 0.41 for FMD in the prediction of CAD. They founda better specificity and higher sensitivity for FMD ascompared to angina pectoris (sensitivity 95%, specificity47.6%), exercise ECG (sensitivity 82.4%, specificity 57.1%),and myocardial perfusion imaging (sensitivity 100%). Thepresent study showed a sensitivity of 76%, specificity of44%, positive predictive value of 58%, and negativepredictive value of 65% for ED in the prediction of CAD.The specificity and positive predictive value can be improvedby the selection of a healthy control group in the true sense,inclusion of only coronary angiography-documented cases,and utilizing automated tracking for the measurement ofarterial diameter.

In a study by Neunteufl et al.,8 impaired FMD in CADpatients was related to the extent of CAD, to the maximumpercent diameter stenosis in one of the major coronaryvessels, brachial artery diameter, and plasma cholesterollevel on univariate analysis. On multiple regressionanalysis, the extent of CAD (1-, 2- or 3-vessel disease) andthe baseline brachial artery diameter were independentlyassociated with FMD in CAD patients.8

Table 4. Correlation of flow-mediated dilatation of thebrachial artery with CAD on univariate analysis

Without CAD With CAD Total

FMD <4.5% 38 52 90(55.9) (76.5) (66.2)

FMD >4.5% 30 16 46(44.1) (23.5) (33.8)

Total 68 68 136

Values in parentheses indicate percentagesFMD: % change in flow-mediated dilatation; CAD: coronary artery diseasePearson Chi-Square=6.439, p=0.011Odds ratio: FMD <4.5%/FMD >4.5% for CAD=3.90(95% CI lower: 0.187, CI upper: 0.814)

Table 5. Multiple regression analysis using CAD as adependent variable

p value Odds ratio

Sex 0.934 1.047Age 0.019 1.053Smoking 0.198 2.411BMI >23 kg/m2 1.000 1.000Cholesterol >180 mg/dl 0.555 1.473LDL-c >130 mg/dl 0.624 1.395HDL-c >40 mg/dl 0.539 0.717Triglycerides >200 mg/dl 0.560 1.390FMD (%) <4.5 0.033 0.321Risk groups 0.173 0.723

BMI: body-mass index; LDL-c: low-density lipoprotein cholesterol;HDL-c: high-density lipoprotein cholesterol; FMD: flow-mediateddilatation

Fig. 1. Percentage distribution of subjects with endothelial dysfunction inrelation to CAD.

Table 3. Percentage distribution of flow-mediated dilatationin subjects with risk factors

Normal Type 2 Hyper- Diabetes withdiabetes tension hypertension

FMD <4.5% 17 15 29 29within subgroup (18.9) (16.7) (32.2) (32.2)

FMD >4.5% 11 5 22 8within subgroup (23.9) (10.9) (47.8) (17.4)

Total 28 20 51 37within subgroup (20.6) (14.7) (37.5) (27.2)

Values in parentheses indicate percentagesFMD: flow-mediated dilatationPearson Chi-square= 5.506, p<0.138

IHJ-363-02.p65 5/6/2003, 10:01 AM47


Endothelial-dependent and -independent functions areimpaired in patients with CAD, and the extent of FMD hasbeen shown to significantly correlate with serum HDL-clevels on univariate and multivariate analysis.9 Patientswith unstable angina have also been shown to have EDcompared to normal individuals.10 Long-term follow-up of28 months in subjects with severe ED in the absence ofobstructive CAD was shown to be associated with increasedcardiac events. Coronary ED may play a role in theprogression of coronary atherosclerosis.11

The obvious limitation of the present study is the lackof angiographic documentation in both the groups, exceptfor undisputed evidence of CAD in 41 subjects in the CADgroup. This was not performed for ethical, logistic, andsocioeconomic reasons. It was not known whetherparticipants in the group without CAD with FMD >4.5%already had significant subclinical atherosclerosis at thetime of enrollment to the study, which can negate theargument for the above limitation. The present study is nota prospective cohort study, and a long-term follow-up ofthe subjects without CAD is required.

Conclusions: In conclusion, the present study shows thatED is associated with and is also an independent risk factorfor CAD. Assessment of ED should be included along withother conventional risk variables such as lipoproteins toidentify individuals at increased risk for CAD at an earlystage, given the paucity of facilities for invasive evaluationin India. Longitudinal studies are required to demonstratethe utility of ED as a predictor of CAD in the Indianpopulation.

Acknowledgment

We would like to thank Dr DP Singh for his assistance withthe statistical analysis.

References

1. Raitakari OT, Celermajer DS. Flow-mediated dilatation. Br J ClinPharmacol 2000; 50: 397–404

2. Celermajer DS. Testing endothelial function using ultrasound. JCardiovasc Pharmacol 1998; 32: 29–32

3. Celermajer DS, Sorensen KE, Gooch VM, Spiegelhalter DJ, Miller OI,Sullivan ID, et al. Non-invasive detection of endothelial dysfunctionin children and adults at risk of atherosclerosis. Lancet 1992; 340:1111–1115

4. Celermajer DS, Sorensen KE, Georgakopoulos D, Bull C, Thomas O,Robinson J, et al. Cigarette smoking is associated with dose-relatedand potentially reversible impairment of endothelium-dependentdilation in healthy young adults. Circulation 1993; 88: 2149–2155

5 Wendelhag I, Gustavsson T, Suurkula M, Berglund G, Wikstrand J.Ultrasound measurement of wall thickness in the carotid artery:fundamental principles and description of a computerized analysingsystem. Clin Physiol 1991; 11: 565–577

6. Sorensen KE, Celermajer DS, Spiegelhalter DJ, Georgakopoulus D,Robinson J, Thomas O, et al. Non-invasive measurement of humanendothelium dependent arterial responses: accuracy andreproducibility. Br Heart J 1995; 74: 247–253

7. Schroeder S, Enderle MD, Ossen R, Meisner C, Baumbach A, Pfohl M,et al. Noninvasive determination of endothelium-mediatedvasodilation as a screening test for coronary artery disease: pilot studyto assess the predictive value in comparison with angina pectoris,exercise electrocardiography, and myocardial perfusion imaging. AmHeart J 1999; 138: 731–739

8. Neunteufl T, Katzenschlager R, Hassan A, Klaar U, Schwarzacher S,Glogar D, et al. Systemic endothelial dysfunction is related to the extentand severity of coronary artery disease. Atherosclerosis 1997; 129:111–118

9. Zhang X, Zhao SP, Li XP, Gao M, Zhou QC. Endothelium-dependentand -independent functions are impaired in patients with coronaryheart disease. Atherosclerosis 2000; 149: 19–24

10. Esper RJ, Vilarino J, Cacharron JL, Machado R, Ingino CA, GarciaGuinazu CA, et al. Impaired endothelial function in patients withrapidly stabilized unstable angina: assessment by noninvasive brachialartery ultrasonography. Clin Cardiol 1999; 22: 699–703

11. Suwaidi JA, Hamasaki S, Higano ST, Nishimura RA, Holmes DR Jr,Lerman A. Long-term follow-up of patients with mild coronary arterydisease and endothelial dysfunction. Circulation 2000; 101: 948–954

IHJ-363-02.p65 5/6/2003, 10:01 AM48

Indian Heart J 2003; 55: 49–54 Harikrishnan et al. Supravalvar Aortic Stenosis 49

Supravalvar Aortic Stenosis: Clinical and HemodynamicProfile, and Surgical Outcome

S Harikrishnan, SR Krishna Manohar, Krishna Kumar Nair, JaganmohanTharakan, Thomas Titus,VK Ajith Kumar, Anil Bhat, S Sivasankaran, Francis Bimal, KM Krishna Moorthy, R Padma Kumar

Departments of Cardiology and Cardiac Surgery, Sree Chitra Tirunal Institute forMedical Sciences and Technology, Thiruvananthapuram

Supravalvar aortic stenosis (SVAS) is the least commonform of congenital left ventricular outflow tract (LVOT)

obstruction. Congenital narrowing of the ascending aortamay be localized or diffuse, originating at the superiormargin of the sinus of Valsalva just above the level of theorigin of the coronary artery.1 The features of SVAS whichdistinguish it from other forms of LVOT obstruction areinvolvement of the coronary arteries and aortic leaflets, andan association with characteristic facies and mentalretardation.2,3

Reports regarding this entity are few, especially in theIndian context. This study is a retrospective one, whichdescribes our 24-year experience of the clinical, angio-

graphic, and hemodynamic profile of this entity, and oursurgical experience and its outcome.

Methods

We reviewed the medical records of all patients with adiagnosis of SVAS from our patient database startingfrom 1976 till date. Only those patients who hadundergone cardiac catheterization and angiography toconfirm the diagnosis were enrolled in the study. Datacollected included anthropometric particulars, clinicalhistory and findings, electrocardiograms (ECGs), chest X-rays, and echocardiographic and cardiac catheterizationdata. All cine films were retrospectively reviewed by twoexperienced angiographers. Surgical data, including thetechnique of surgery, peroperative findings, and

Original Article

Background: Supravalvar aortic stenosis is the rarest of left ventricular outflow obstructions. Data on this rareentity from India are scarce.Methods and Results: We retrospectively analyzed the data of 15 patients (13 males, mean age 15.5±10.18years) with a diagnosis of supravalvar aortic stenosis confirmed by cardiac catheterization. Five patients hadmorphological features of Williams’ syndrome. One patient had diffuse while the rest had discrete type ofsupravalvar aortic stenosis. Five patients did not have any associated lesions. A 9-year-old male had an ascendingaortic aneurysm, and 3 patients had associated peripheral pulmonary artery stenosis. One child had a subaorticventricular septal defect, and another had severe mitral regurgitation. Twelve patients had electrocardiographicevidence of left ventricular hypertrophy. Three patients had mild aortic valvar stenosis while 2 had aorticregurgitation. Six patients had dilated coronary arteries. Two patients with supravalvar aortic gradients of 20and 40 mmHg were kept on close follow-up. One patient was not willing to undergo surgery while the other isawaiting surgery. Eleven patients underwent surgical correction. Dacron or pericardial patch aortoplasty wasdone in all the patients. In addition, one patient each underwent pulmonary artery plasty, ventricular septaldefect closure, repair of ascending aortic aneurysm, and mitral valve replacement. The patient with diffuse typeof supravalvar aortic stenosis underwent augmentation aortoplasty. Two patients died perioperatively. One waslost to follow-up. Two had moderate residual gradients. The rest of the patients were in New York Heart Associationfunctional class I on follow-up of 6.3±4.7 years.Conclusions: Repair of supravalvar aortic stenosis by single sinus aortoplasty is safe and produces good results.(Indian Heart J 2003; 55: 49–54)

Key Words: Supravalvar aortic stenosis, Williams’ syndrome, Aortoplasty

Correspondence: Dr S Harikrishnan, SCTIMST, Thiruvananthapuram695011. e-mail: [email protected]

IHJ-408-02.p65 5/6/2003, 10:01 AM49

50 Harikrishnan et al. Supravalvar Aortic Stenosis Indian Heart J 2003; 55: 49–54

complications were also collected. Follow-up datawere collected and those patients who did not have recentfollow-up data were called for a review and their datacollected.

Results

Among the 12 400 patients who underwent cardiaccatheterization for evaluation of congenital heart disease,there were 15 patients (0.12%, 13 males) who had thediagnosis of SVAS confirmed by catheterization andangiography. The average age was 15.5±10.18 years. Theyoungest was a 15-month-old male child and the oldestwas a 34-year-old male. Five patients had morphologicalfeatures of Williams’ syndrome; of them, only 1 hadhypercalcemia (Tables 1 and 2).

All the patients had discrete type of SVAS (Figs 1 and 2),except 1 (6.6%) who had the diffuse variety (Fig. 3). Themean peak gradient across the supra-aortic region was101.63±59.69 mmHg.

The mode of presentation was congestive heart failure(CHF) in 3 patients. One patient, a 14-year-old male,presented with severe anemia and CHF (hemoglobin 4 g%),the other 2 patients presented with CHF in infancy. A 27-year-old male patient presented with recurrent presyncope.He had a very high LVOT gradient of 215 mmHg. Threepatients were in the New York Heart Association (NYHA)

functional class I at presentation. The rest of the patientswere in NYHA class II.

Two patients underwent repeat cardiac catheterizationon follow-up. One 10-year-old boy with class II symptoms,who had an initial gradient of 24 mmHg, which increasedto 80 mmHg after 6 years, was advised surgery. Another10-year-old boy whose initial gradient of 31 mmHgincreased to 76 mmHg after 5 years also underwent surgery.One patient with a gradient of 40 mmHg had the sameechocardiographic gradient 9 years post-catheterizationand has been kept on follow-up.

Associated lesions: Five patients did not have anyassociated lesions. One patient, a 9-year-old male, had anascending aortic aneurysm necessitating repair. Threepatients had associated peripheral pulmonary arterystenosis. Of them, 2 patients had features of Williams’syndrome. One patient had moderate aortic regurgitation(AR) while another had a small perforation in thenoncoronary aortic cusp with mild AR, requiring repair. A10-year-old child had a subaortic ventricular septal defect(VSD), and mild valvar AS (AS gradient of 20 mmHg). A15-month-old child had mitral valve prolapse (MVP), andmoderate-to-severe mitral regurgitation (MR), requiringmitral valve repair.

Electrocardiograms: Two patients with supravalvargradients of 20 and 40 mmHg had normal ECGs. A 6-year-

Table 1. Clinical profile and morphologic data

S. Age Sex Dysmorphism Aortic valve Associated cardiac CoronariesNo. (years) lesions

1 14 M Nil Normal Nil Normal2 27 F Nil Thickened, stenotic Valvular AS Normal3 1.25 M Nil Normal MVP MR NA4 6 M Nil Valve commissures fused Mild AR

to ridge, NCC perforation Dilated5 16 M WS Normal Nil Normal6 18 M WS Thickened, stenotic VSD+valvular AS Normal7 10 M WS Normal RPA stenosis Normal8 27 M Nil Normal Innominate artery Dilated

narrowing9 4.5 M WS Adherent LCC Nil Dilated, tortuous10 10 M Nil Normal Nil Normal11 9 M Nil Normal Ascending aortic Dilated

aneurysm12 15 F Nil Normal LPA stenosis Normal13 26 M Nil Adherent LCC Nil Dilated14 10 M WS Normal LPA, RPA stenosis Dilated, tortuous15 34 M Nil Adherent cusps, AR Nil Normal

WS: Williams’ syndrome; LCC: left coronary cusp; NCC: noncoronary cusp; AS: aortic stenosis; VSD: ventricular septal defect; AR: aortic regurgitation;MVP: mitral valve prolapse; MR: mitral regurgitation; LPA: left pulmonary artery; RPA: right pulmonary artery; NA: data not available

IHJ-408-02.p65 5/6/2003, 10:01 AM50


old patient with a gradient of 70 mmHg also did not haveleft ventricular hypertrophy (LVH). All the other patientshad ECG evidence of LVH. The presence of LV strain did notcorrelate with the severity of stenosis.

Aortic valve morphology: Three patients had gradientsacross the aortic valve in addition to the gradients at thesupravalvar region. These gradients were minimal (10, 20,and 14 mmHg) against supravalvar gradients of 120, 170and 202 mmHg, respectively. All the patients whounderwent surgery had their valves examinedperoperatively and the findings were noted. Nine patientshad normal valves. Details of valve abnormalities are givenin Table 1. No patient needed any intervention for valvularanomalies except one who had a small noncoronary cuspperforation which had to be repaired during patchaortoplasty. One patient with moderate AR is awaitingsurgery.

Coronary morphology: Coronary anatomy was availablefor analysis in 13 of the 15 patients. Seven of the 13 patientshad dilated coronary arteries, especially the proximalcoronary arteries (Figs 2 and 3). Two patients had tortuous,dilated coronary arteries.

Two patients with supravalvar aortic gradients of 20and 40 mmHg were kept on close follow-up. The relativesof one 10-year-old male patient were not willing for surgery.One patient is awaiting surgery. The remaining patients(n=11) underwent surgical correction.

Surgical technique: All the patients were operated(n=11) under standard cardiopulmonary bypass (CPB)with moderate hypothermic and cold cardioplegic

myocardial protection. Ten patients, who had a classicalhourglass type of narrowing, underwent repair by thetechnique of single sinus aortoplasty, using a diamond-shaped dacron patch in 6 patients and gluteraldehyde-tanned autologous pericardial patch in 4. The child withdiffuse narrowing of the ascending aorta underwentwidening of the ascending aorta with a tannedpericardial patch from the noncoronary sinus to theorigin of the innominate artery. Associated proceduresin our patients included mitral valve repair, closure ofVSD, repair of saccular aneurysm of the ascending aorta,and pericardial patch widening of the origin of the rightpulmonary artery (one case each).

Surgical results and follow-up data: There were 2 deaths,both due to uncontrollable bleeding from the aortic sutureline, which occurred in the immediate postoperative period.One patient was lost to follow-up. Follow-up was for6.33±4.71 years. The patient with diffuse type of SVASshowed a gradient of 66 mmHg on follow-upechocardiogram. All the other patients were in NYHAfunctional class I on follow-up and echocardiographicassessment showed good results, except in 2 patientswho showed mild valvar gradient with mild valvarregurgitation.

Discussion

Supravalvar aortic stenosis is relatively rare and the leastcommon of the congenital LVOT obstructions.1 The patientscan be classified into three groups according to the mode ofpresentation.

Table 2. Electrocardiographic and hemodynamic data, and follow-up

S. AgeNo. (years) Sex Type ECG LV pressure SV gradient Course/surgery

1 14 M Diffuse LVH strain 230 150 PA2 27 F Localized No LVH 120 20 Medical follow-up3 1.2 M Localized LVH 154 90 PA+MVR4 6 M Localized No LVH 190 70 PA5 16 M Localized LVH 120 56 PA6 18 M Localized LVH 250 70 PA+VSD closure7 10 M Localized LVH 170 76 PA+pulmonary arterioplasty8 27 M Localized LVH 340 215 PA9 4.5 M Localized LVH 200 122 PA10 10 M Localized No LVH 150 40 Medical follow-up11 9 M Localized LVH strain 302 200 PA+aneurysmorrhaphy12 15 F Localized LVH strain 300 180 PA13 26 M Localized LVH 200 80 PA14 10 M Localized LVH 205 100 Not willing for surgery15 34 M Localized LVH strain 170 50 Awaiting surgery

PA: patch aortoplasty; LVH: left ventricular strain; MVR: mitral valve replacement; VSD: ventricular septal defect; SV: supravalvar

IHJ-408-02.p65 5/6/2003, 10:01 AM51


Morphologically, SVAS can be categorized into three types.

(1) Hourglass type: the most common type, characterizedby a constricting annular ridge at the superior marginof the sinus of Valsalva

(2) Membranous type: produced by an intra-aortic fibrous/fibromuscular circular/semicircular diaphragm with acentral opening

(3) Hypoplastic type: with uniform hypoplasia of theascending aorta which may extend into the archvessels.

Supravalvar aortic stenosis is unique among LVOTobstructions because, among systemic vessels, the coronaryarteries alone are exposed to high systolic pressures,5 whichlead to dilatation, tortuosity and premature atherosclerosisof these arteries, thus making the patient susceptible tosudden cardiac death before and after repair. The coronaryostia may be obstructed by the overhanging thick sinusrims as well as bound-down aortic cusps. This morecommonly occurs in the left sinus, but can also occur in theright. Also, the free edges of the aortic cusps (which arethickened in one-third of patients) may be adherent to theintraluminal supravalvar aortic ridge. This contributes toLVOT obstruction, interferes with coronary blood flow, andis also responsible for the associated aortic regurgitation.

Five of our 15 patients had associated aortic valve

Fig. 1. Echocardiogram (high parasternal view) showing discrete supravalvarnarrowing of the aorta (arrows).

(1) Nonfamilial sporadic cases with normal facies andintelligence

(2) Autosomal dominant familial cases with normal faciesand intelligence3

(3) Williams–Beuren's syndrome2–4 with characteristic elfinfacies and mental retardation—nonfamilial.

Fig. 3. Diffuse variety of supravalvar aortic stenosis. Long-segment aorticnarrowing is present. A dilated left main coronary artery is also seen.

Fig. 2. Discrete variety of supravalvar aortic stenosis. A dilated left maincoronary artery is seen.

IHJ-408-02.p65 5/6/2003, 10:01 AM52


anomalies. In the series by Sharma et al.,6 25% of the patientshad a bicuspid aortic valve,12% had aortic regurgitation,11% had valvar aortic stenosis, and 10% had subvalvarstenosis. In the series of 13 patients by Braunstein et al.7

one patient had subaortic obstruction.Williams–Beuren’s syndrome2–4 is associated with a

microdeletion in the chromosomal region 7q11.23encompassing, among others, the elastin gene. Thesyndrome is routinely confirmed by detecting elastinhemizygosity by fluorescence in situ hybridization (FISH).Previous reports suggest that hemizygosity of the elastingene is responsible for the typical vasculopathy of thissyndrome, namely, SVAS and pulmonary arterial stenosis.4,5

Williams–Beuren’s syndrome is associated with thenonfamilial form of SVAS, in which there is a peculiar facies(small chin, large mouth, patulous lips, blunt and upturnednose, widely set eyes with internal strabismus, epicanthicfolds, broad forehead, baggy cheeks, lacy iris pattern, andmalformed teeth with malocclusion), short stature, andmental retardation. Adult patients are short-statured, mayhave kyphoscoliosis, and may develop progressive jointlimitation and hypertonia. These patients may haveidiopathic infantile hypercalcemia, pulmonary arterystenosis (especially peripheral pulmonary stenosis), renalartery anomalies, tortuous retinal arteries, systemichypertension, and gastrointestinal and urinary tractanomalies.

In Williams’ syndrome, SVAS seems to progress rapidlywhile peripheral pulmonary stenosis improves with time.8

Associated cardiac lesions described are pulmonary stenosis(both valvar and peripheral), mitral valve anomalies withMR, and aneurysm of the ascending aorta and subaorticobstruction. Etiologic factors proposed are idiopathichypercalcemia, disturbance of vitamin D metabolism andcalcium homeostasis,9 and rubella.

In our series, there were no familial cases of SVAS, butthere were 5 cases of Williams’ syndrome. Of them, 2 hadperipheral pulmonary artery stenosis and one had aorticvalvar stenosis. Only one of these patients hadhypercalcemia; however, vitamin D challenge or calciumloading tests were not done in these patients to unmaskcalcium homeostatic defects. Seven of the 13 patients inthe series by Braunstein et al.7 and 14 of the 101 patientsin the series by Brown et al.10 had Williams' syndrome.

The treatment of SVAS is surgical if the patient hassignificant gradients. SVAS is less amenable to operativetreatment than either valvar or discrete subvalvar stenosis.Successful surgery for SVAS using patch graft enlargementof the noncoronary sinus of Valsalva was first reportedfrom the Mayo clinic.11 After that, many series on surgicalresults in patients with SVAS have been reported.6,10–18

The discrete type of SVAS is more amenable to surgery

than the diffuse type. In the discrete type, aortotomy is doneabove the valve, and the incision carried into the right ornoncoronary sinus of Valsalva. The intimal shelf is thenresected and a diamond-shaped pericardial/dacron patchincorporated into the incision to enlarge the aortic diameterto normal.

In the single-center experience of SVAS from the TexasHeart Institute, 32% of the patients had discrete type ofSVAS, and were treated with simple patch aortoplasty(placing a diamond-shaped patch across the sinus rim inthe noncoronary sinus of Valsalva) with good results andonly 4% mortality. Though this operation was effective inrelieving the pressure gradients, it did nothing to rebuild/remodel the diseased aortic root. Doty et al.19 have proposedbisinus repair by extending a bilobed patch into both thenoncoronary sinus and right coronary sinus for a widerand anatomically more symmetrical repair, even in caseswith discrete stenosis.

Surgical treatment options for the diffuse form of SVASare less well defined. It may become necessary to replace orwiden the entire hypoplastic aorta with an appropriateprosthesis.16,17 Extended patch aortoplasty and apicoaorticconduits are the procedures that have been tried. Sharmaet al.6 have treated two patients with the diffuse variety ofSVAS with another technique: extensive endarterectomyof the ascending aorta and arch, with patch aortoplastyextended into the aortic arch. Since the morphologic defectin SVAS involves the media and intima, it seems logical toremove these by endarterectomy.6

Only 1 of our patients had the diffuse variety of SVAS(6.66%), compared with 15% of patients in the series bySharma et al.6 In the series by Brown et al.,10 28/101patients had the diffuse variety of SVAS, and they weretreated with either an apical aortic conduit or extensiveendarterectomy with patch aortoplasty, with encouragingresults. In the series by Braunstein et al.,7 2/13 patientswith the diffuse variety of SVAS had excellent results withbisinus patching and extending the patch to the descendingaorta.

Thirty-three patients in the surgical series by Sharma etal.6 had other associated obstructive lesions of the LVOTrequiring surgical correction. Though 2 patients in our serieshad associated valvar stenosis, this did not warrant surgicalcorrection. Of patients with SVAS, 30% are reported to havethickened aortic valve cusps. Two patients (18%) developedmild AR postoperatively in our series. The exact incidenceof AR in the series by Sharma et al.6 is not clear. However,the incidence of 44% of new AR in the series by Braunsteinet al.7 is much higher.

Four patients in our series had associated cardiac lesions,which required surgical correction. PA plasty was done in1 of our patients. There were no cases which required PA

IHJ-408-02.p65 5/6/2003, 10:01 AM53


plasty in the Braunstein series,7 while 1 patient required PAplasty in the series by Sharma et al.6 One patient in theseries by Sharma et al.6 had pseudoaneurysm of theascending aorta, while 1 of our patients had an aneurysmof the ascending aorta requiring aneurysmorrhaphy.

We had 2 early deaths due to bleeding (18.18%) but nolate deaths at a mean follow-up of more than 6 years.

None of our 11 patients required reoperation at a meanfollow-up of more than 6 years. In the series by Sharma etal.,6 22% of the patients required reoperation at some timeduring the follow-up period, and there were 5 late deaths inthose with complex forms of the disease. We had only 1patient with the diffuse form of the disease who underwentextended patch aortoplasty but had moderate residualgradient on follow-up. In a large experience of 101 patientswith SVAS,10 the overall survival including operativemortality was 98% at 10 years, and 97% at 20 and 30years.

This study has several weaknesses. It was a retrospectivestudy, and the numbers were too small to draw conclusions.The patients did not undergo hemodynamic orangiographic study postoperatively.

We conclude that repair of SVAS by single sinusaortoplasty is safe, and produces good results.

References

1. Sissman NJ, Neill CA, Spencer FC, Taussig HB. Congenital aorticstenosis. Circulation 1959; 19: 458–459

2. Williams JCP, Barrat-Boyes BG, Lowe JB. Supravalvar aortic stenosis.Circulation 1961; 24: 1311–1314

3. Beuren AJ, Schulze C, Eberle P, Harmjanz E, Apitz J. The syndrome ofsupravalvar aortic stenosis, peripheral pulmonary stenosis, mentalretardation and similar facial appearance. Am J Cardiol 1964; 13:471–473

4. Eronen M, Peippo M, Hiippala A, Raatikka M, Arvio M, JohanssonR, et al. Cardiovascular manifestations in 75 patients with Williams’

syndrome. J Med Genet 2002; 39: 554–5585. Stamm C, Friehs I, Ho SY, Moran AM, Jonas RA, del Nido PJ.

Congenital supravalvar aortic stenosis: a simple lesion? Eur JCardiothorac Surg 2001; 19: 195–202

6. Sharma BK, Fujiwara H, Hallman GL, Ott DA, Reul GJ, Cooley DA.Supravalvar aortic stenosis: a 29-year review of surgical experience.Ann Thorac Surg 1991; 51: 1031–1039

7. Braunstein PW Jr, Sade RM, Crawford FA Jr, Oslizlok PC. Repair ofsupravalvar aortic stenosis: cardiovascular morphometric andhemodynamic results. Ann Thorac Surg 1990; 50: 700–707

8. Giddins NG, Finley JP, Nanton MA, Roy DL. The natural course ofsupravalvar aortic stenosis and peripheral pulmonary arterystenosis in Williams’ syndrome. Br Heart J 1989; 62: 315–319

9. Kruse K, Pankau R, Gosch A, Wohlfahrt K. Calcium metabolism inWilliams–Beuren syndrome. J Paediatr 1992; 121; 902–907

10. Brown JW, Ruzmetov M, Vijay P, Turrentine MW. Surgical repair ofcongenital supravalvular aortic stenosis in children. Eur JCardiothorac Surg 2002; 21: 50–56

11. McGoon DC, Manhin HT, Vlad P, Kirklin JW. The surgical treatmentof supravalvar aortic stenosis. J Thorac Cardiovasc Surg 1961; 41:125–129

12. Rastelli GC, McGoon DC, Ongley PA, Mankin HT, Kirklin JW. Surgicaltreatment of supravalvar aortic stenosis. Report of 16 cases andreview of literature. J Thorac Cardiovasc Surg 1966; 51: 873–878

13. Thistlethwaite PA, Madani MM, Kriett JM, Milhoan K, Jamieson SW.Surgical management of congenital obstruction of the left maincoronary artery with supravalvular aortic stenosis. J ThoracCardiovasc Surg 2000; 120: 1040–1046

14. Flaker G, Teske D, Kilman J. Supravalvar aortic stenosis: a 20–yearclinical perspective and experience with patch aortoplasty. Am JCardiol 1983; 15: 256–261

15. McElhinney DB, Petrossian E, Tworetzky W, Silverman NH, HanleyFL. Issues and outcomes in the management of supravalvar aorticstenosis. Ann Thorac Surg 2000; 69: 562–567

16. Kirklin JW, Barratt-Boyes BG (eds). Cardiac surgery. 2nd ed. New York:Churchill Livingstone;1993. pp. 1196–1232

17. Friedman WF. Aortic stenosis. In: Friedman WF (ed). Heart diseasein infants, children and adolescents. 5th ed. Hagerstorm: William andWilkins; 1995. pp. 1087–1111

18. Doty DB. Supravalvar aortic stenosis. Ann Thorac Surg 1991; 51:886–887

19. Doty DB, Polansky DB, Jenson CB. Supravalvular aortic stenosis.Repair by extended aortoplasty. J Thorac Cardiovasc Surg 1977; 74:362–371

IHJ-408-02.p65 5/6/2003, 10:01 AM54

Indian Heart J 2003; 55: 55–59 Bharani et al. Sildenafil in PAH 55

Background: Pulmonary arterial hypertension is a life-threatening disease for which continuous intravenousinfusion of prostacyclin has proved effective. However, it carries the risk of serious complications arising fromthe complex delivery system. Prostacyclin analogs, endothelin antagonists, and the phosphodiesterase-5 inhibitorsildenafil are emerging promising therapies. This study was aimed at evaluating the utility of oral sildenafil inpatients with pulmonary hypertension of varied etiology, poorly controlled on conventional treatment.Methods and Results: Ten consecutive patients with pulmonary hypertension, either primary or related toprevious left-to-right shunts, thromboembolism, or interstitial lung disease, poorly controlled on conventionaltherapy such as warfarin, calcium antagonists, digitalis, and diuretics, were included. A thorough clinical,laboratory, and comprehensive echo Doppler evaluation was performed before enrollment in the trial to establishthe diagnosis and obtain baseline data. Subjects received sildenafil 25 mg 8 hourly, or a matching placebo fortwo weeks each, in a randomized, double-blind, crossover design. A run-in period of two weeks was permittedbetween the two therapies during which patients continued to receive the conventional therapy without anyvasodilator. At the end of each therapy period, the patients were evaluated for symptoms, New York HeartAssociation class, distance covered during the 6 min walk test, rating of modified Borg dyspnea score, andsystolic pulmonary artery pressure using echo Doppler. The differences in the above variables at the end ofsildenafil and placebo therapies were compared. Nine patients completed the study protocol. Sildenafil, comparedto placebo, was associated with improved exercise tolerance as determined by the 6 min walk test(266.67±131.45 m v. 170±105 m; p<0.005), decrease in modified Borg dyspnea score (3.56±1.01 v.5.11±1.45; p<0.01), decrease in Doppler-estimated pulmonary artery systolic pressures (55.33±16.52 mmHgv. 75.33±19.75 mmHg; p<0.005), improvement in New York Heart Association class (2 patients), andimprovement in symptoms. Sildenafil was well tolerated with no untoward effects; further, no significant changesin heart rate or blood pressure occurred during the study period.Conclusions: Sildenafil improves exercise capacity and symptoms, and decreases pulmonary artery pressuresin patients with primary or secondary pulmonary hypertension of varied etiology. (Indian Heart J 2003; 55:55–59)

Key Words: Pulmonary hypertension, Sildenafil, Echocardiography

The Efficacy and Tolerability of Sildenafil in Patients WithModerate-to-Severe Pulmonary Hypertension

Anil Bharani, Vivek Mathew, Ashutosh Sahu, Basant LuniaDivision of Cardiology, Department of Medicine, M.G.M. Medical College and M.Y. Hospital, Indore

Pulmonary arterial hypertension (PAH) is characterizedby a progressive increase of pulmonary vascular

resistance leading to right ventricular failure and death(WHO 1998).1,2 Pulmonary arterial hypertension includesprimary pulmonary hypertension (PPH) with no apparentcause, PAH secondary to congenital shunt lesions,connective tissue disorders, and portal hypertension, drug-induced PAH, and HIV-related PAH.1,2 The currenttherapeutic options available for PAH include

anticoagulants, calcium-channel blockers, epoprostenol(prostacyclin), iloprost, beraprost, the endothelin receptorantagonist bosentan, and phosphodiesterase type-5 (PDE5)inhibitor sildenafil.3 However, in many parts of the world,therapeutic options for patients with PAH are limited in theabsence of availability of prostacyclins3 and thus sildenafilcould be a useful alternative.

Sildenafil-induced inhibition of PDE5 increasesthe cellular levels of cGMP, potentiating vascularsmooth muscle relaxation, particularly in the lungs wherePDE5 is found in high concentrations.4 Sildenafil has beenreported to inhibit hypoxia-induced pulmonary

Original Article

Correspondence: Dr Anil Bharani,119, Ravindra Nagar, Indore 452018.e-mail: [email protected]

IHJ-395-02.p65 5/6/2003, 10:02 AM55

56 Bharani et al. Sildenafil in PAH Indian Heart J 2003; 55: 55–59

hypertension,5 cause sustained reduction in pulmonaryartery pressures and pulmonary vascular resistancein PPH,6 blunt the rebound pulmonary hypertensionseen following withdrawal of inhaled nitric oxide,7 leadto impressive clinical benefits in childhood PPH,8 andcause selective pulmonary vasodilatation in a lambmodel of PAH.9

We report the results of a randomized, double-blind,placebo-controlled, crossover study designed to determinethe efficacy of short-term oral administration of sildenafilon exercise capacity, symptoms, and Doppler-determinedpulmonary artery pressures in NYHA class II–IV patientswith PAH.

Methods

Selection of patients: Ten consecutive, symptomatic(NYHA class >II) patients with Doppler-estimatedpulmonary systolic pressure >35 mmHg with normal leftventricular function and no reversible cause for PAH weretaken up for the study. They gave well-informed writtenconsent for participation in the study and all except onecompleted the study protocol. The institutional ethicscommittee approved the protocol. Patients were excludedif they had any contraindications to sildenafil therapy or ifthey had any reversible cause for pulmonary hypertensionsuch as valvular heart disease.

Study design: The study was carried out at a large referralcenter of central India as a prospective, randomized,double-blind, crossover trial. To establish the diagnosis,patients were subjected to a thorough physicalexamination, routine laboratory evaluation, chest X-ray,electrocardiogram (ECG), and comprehensive echocardio-graphic evaluation at baseline. Before randomization, theywere allowed a run-in period of one week after stoppage oftheir previous vasodilator therapy; they then received eithersildenafil 25 mg 8 hourly or matching placebo for twoweeks in a crossover design with a washout period of atleast two weeks between the two therapies.

Outcome measures: The patients were evaluated forexercise capacity as indicated by the 6 min walkingdistance. The secondary measures of efficacy were changein symptoms and NHYA functional class, change inmodified Borg dyspnea index10 (a measure of perceivedbreathlessness on a scale of 0 to 10, with higher valuesindicating more severe dyspnea), and change in restingpulmonary artery systolic pressures as estimated fromDoppler tricuspid regurgitant jet velocities or from thegradient across the ventricular septal defect or patent

ductus, wherever applicable.11 Echo-Doppler studies (GELogic 500, MD, USA) were done by a single experiencedobserver who was blind to the therapy patients werereceiving.

Statistical analysis: Changes in variables duringtreatment with sildenafil and placebo were analyzed withthe paired Student’s t test. The null hypothesis was thatthere should be no significant difference in the exercisecapacities and the pulmonary systolic pressures before andafter 2 weeks of sildenafil therapy. A probability value ofp<0.05 was considered significant.

Results

Baseline study group characteristics: There were 5females and 4 males, 19 to 60 years of age, in NYHAfunctional classes II–IV, having pulmonary hypertensionof varied etiologies. They were receiving conventionaltherapy including warfarin and nifedipine, and two patientswere on diuretics and digoxin for treatment of heart failure(Table 1).

Table 1. Study group characteristics

Characteristic Patients (n)

Age 18–60 yearsMean 32.11 years, SD 15.06 years

Sex Male 4Female 5

NYHA classBaseline II 3

III 5IV 1

Post-sildenafiltherapy

II 4III 5IV 0

DiagnosisPrimary pulmonary hypertension 3Pulmonary hypertension associated with:

Interstitial lung disease 2Thromboembolism 1Eisenmenger syndrome

Ventricular septal defect 2Patent ductus arteriosus 1

Treatment at inclusionWarfarin 9Nifedipine 4Frusemide 2Spironolactone 2Digoxin 2

IHJ-395-02.p65 5/6/2003, 10:02 AM56


lead to improved symptoms, hemodynamics, and reducedmortality.12 The prostacyclin analogs iloprost13 andberaprost14 produce marked clinical benefits in patientswith PAH. The orally administered endothelin receptorantagonist bosentan15 is reported to be effective and welltolerated in patients with PAH. While mortality benefitswith these newer agents are yet to be ascertained, eachtherapy has certain limitations. Epoprostenol, the currentgold standard of therapy for PPH, is associated with seriouscomplications arising from the complex delivery system.Sildenafil, which is currently approved for the treatmentof erectile dysfunction, is emerging as a new promisingtherapeutic agent for the treatment of secondary PAH andPPH.16,17 Through selective inhibition of PDE5, which isabundant in the lungs, sildenafil increases cellular levelsof cGMP causing vascular smooth muscle relaxationleading to sustained reduction in pulmonary vascular

Table 2. Change in 6 min walking distance, modified Borgdyspnea score and Doppler pulmonary artery pressure

Characteristic Baseline Placebo Sildenafil t p

6 min walkingdistance (m)

Mean 163.89 170 266.67 5.98 <0.005SD 110.73 105 131.45

Borg dyspnea scoreMean 5.22 5.11 3.56 3.56 <0.01SD 1.64 1.45 1.01

Pulmonary arterypressure (mmHg)

Mean 80.78 75.33 55.33 4.33 <0.005SD 21.30 19.75 16.52

Exercise capacity: The distance walked in 6 min atbaseline and at the end of sildenafil or placebo therapies isgiven in Table 2. The patients showed a significantimprovement in exercise capacity while on sildenafiltherapy compared to placebo (p<0.005, Fig. 1A).

Symptoms and Borg dyspnea index: All the patientsshowed improvement in symptoms after 2 weeks ofsildenafil therapy compared to placebo. The Borg dyspneaindex of perceived exertion after the 6 min walk showedstatistically significant improvement at the end of sildenafiltherapy (p<0.01, Table 2, Fig. 1B). There was improvementin the NYHA class in only two patients. There was nomortality.

Pulmonary artery pressures: Table 2 gives thepulmonary artery systolic pressures as estimated from theDoppler studies. Sildenafil therapy was associated withsignificant reduction in the pulmonary artery pressurescompared to placebo (p<0.005, Fig. 1C).

Side-effects and tolerance: Sildenafil therapy wastolerated well with no adverse effects reported during thisshort-term study. No substantial changes in heart rate orsystemic arterial pressure were observed during sildenafiltreatment. Two sexually active males included in the studyexperienced no change in overall sexual performance orlibido during sildenafil therapy.

Discussion

Primary pulmonary hypertension is a rapidly progressivedisease which is uniformly fatal if untreated. Until 1981,when heart–lung transplantation was introduced, notreatment was available. Challenged by the limited numberof suitable donors and the complications associated withorgan transplantation, the search for effective medicaltherapies is ongoing.1 Continuous epoprostenol infusions

Fig. 1. Mean change from baseline at week 8 in results of the (A) 6 minutewalk test, (B) modified Borg dyspnea index, and (C) pulmonary artery systolicpressure in patients while on sildenafil and placebo. Numbers indicate the meanvalues. Standard error of the mean is given in parentheses.

IHJ-395-02.p65 5/6/2003, 10:02 AM57

58 Bharani et al. Sildenafil in PAH Indian Heart J 2003; 55: 55–59

resistance and pulmonary artery pressures.4,8,18 Thesalutary effects of sildenafil on symptoms and exerciseperformance in patients with PAH have been reported insmall studies16,17 and in anecdotal case reports,8,18 whichare comparable to those of the prostacyclins. However, largecontrolled trials with sildenafil to address the issue ofimproving survival in patients with PPH and PAH are notyet available. This short-term, double-blind, placebo-controlled, crossover study has demonstrated that oralsildenafil in a dosage of 25 mg 8 hourly improves exercisecapacity, symptoms, and causes significant decrease inDoppler-estimated pulmonary artery pressures.

As in other trials performed on patients with PAH, theprimary end-point of this study was the 6 min walkingdistance that has been shown to be an independentpredictor of mortality.19.20 Following treatment, the distanceobserved in our patients was 96.67 m, which matches theeffect observed with intravenous epoprostenol in NYHAclass III–IV patients,20 and with beraprost21 in NYHA classII–III patients with PPH. Improvement in the 6 min walkingdistance was observed in all patients (NYHA class II–IV) inour study. However, improvement in the 6 min walk testwas seen only in patients with PPH receiving beraprost butnot in those with secondary PAH.21

Sildenafil alone16 or in combination with inhalediloprost17 has been shown to improve exercise performancein patients with PPH and secondary PAH. The improvementin the 6 min walking distance was associated with aconcomitant significant improvement in the perception ofdyspnea, as assessed by a reduction of the Borg dyspneascore. Perceptual estimates, obtained by psychophysicalratio-scaling method such as the Borg dyspnea scale, arevery useful in judging the perceptual variation and arecomplementary to physiological measurement of workcapacity and physical performance.10 Despite improvementin symptoms and exercise capacity in all the patients,improvement in NYHA class was seen in only two patients.This was perhaps related to the shorter duration of ourstudy. Similar observations were reported in patients withPAH who were treated with beraprost for 12 weeks.21

Sildenafil therapy was associated with a significantdecrease in Doppler-estimated tricuspid regurgitationvelocities and pulmonary systolic pressures in our study.The available studies on sildenafil are few. In a young malewith severe PAH, sildenafil 100 mg five times per day for 3months led to a decrease in the Doppler-estimatedpulmonary artery pressure from 120 mmHg to 90 mmHgwith improvement in the myocardial oxygen consumptionduring exercise, and dramatic improvement in symptomsand aerobic capacity.18 Similarly, in the case of a 4-year-

old girl with severe PPH with poor response to prostacyclin,oral sildenafil (2 mg/kg 4 hourly) led to a significantdecrease in pulmonary artery pressures, rise in oxygensaturation, improved exercise capacity, and impressiveclinical recovery, permitting discontinuation of prostacyclinand oxygen therapy.8

A recent study reported long-lasting reduction in themean pulmonary artery pressure and pulmonary vascularresistance with sildenafil and additional improvement afteriloprost inhalation in patients with PPH. No significantchanges in heart rate or blood pressure were observedduring the treatment.6 Another study reported impressiveclinical benefits associated with combination therapy usingoral sildenafil and inhaled iloprost in patients with severePAH.17

In a randomized, double-blind, placebo-controlled study,sildenafil 100 mg given orally to ten healthy volunteersinhibited hypoxia-induced pulmonary hypertension, aresponse mediated through the eNOS–NO–cGMP pathway.5

Sildenafil has been reported to blunt the reboundpulmonary hypertension seen following withdrawal ofinhaled nitric oxide.7,22 In lambs with acute pulmonaryhypertension, sildenafil decreased the pulmonary arterypressure and pulmonary vascular resistance resulting fromselective pulmonary vasodilatation.9 In a recent study, asingle oral dose of sildenafil (75 mg) was found to be aseffective and as selective a pulmonary vasodilator as inhalednitric oxide in patients with severe PAH referred for heart–lung transplantation. Sildenafil was superior to nitric oxidein that it led to a greater increase in cardiac output and didnot increase wedge pressure.23 Sildenafil was assessed forchildhood and neonatal pulmonary hypertension(16 patients), given either acutely (0.25–0.5 mg/kg) priorto hemodynamic measurements in the cath lab, orchronically after gradual withdrawal of nitric oxide inrefractory suprasystemic PAH, or for chronic treatment ofPPH or PAH. During cardiac catheterization, the meanpulmonary artery pressure decreased (50±8 mmHg to38±12 mmHg, p<0.05), and pulmonary vascularresistance decreased with no change in the mean systemicpressure or systemic vascular resistance. During chronicuse, sildenafil attenuated the rise in pulmonary arterypressure, permitted discontinuation of nitric oxide and ledto 200% improvement in the 6 min walk distance in theremaining patients.24

Sildenafil has been well tolerated without major side-effects.3 We did not come across hypotension, syncope, orany other untoward effect attributable to sildenafil in ourstudy. Our results concur with those of other studies tosupport the safety and impressive clinical benefits

IHJ-395-02.p65 5/6/2003, 10:02 AM58


associated with sildenafil therapy in patients suffering fromPPH or secondary PAH.6,8,17,18 Sildenafil therapy was notfound to have any important hemodynamic effect inpatients with stable coronary artery disease25 but cautionmust be exercised in patients receiving diuretics and/ornitrates.

To conclude, our study results show the beneficial effectsof sildenafil in patients with PPH and secondary PAH inadults. These include significant improvement in exercisecapacity as indicated by improvement in the 6 min walkdistance, decrease in modified Borg dyspnea index ofperceived exertion, and decrease in pulmonary arterysystolic pressures. The drug is well tolerated with nosubstantial change in heart rate or blood pressure. Thesmall sample size and short duration of the study areacceptable limitations. Large, long-term, randomized,multicenter trials are needed to answer questions regardingthe long-term utility and survival benefits with the PDE5inhibitor sildenafil in patients with PPH and secondaryPAH.

Acknowledgments

We acknowledge the help received from Professor LKMathur, Department of Biostatistics, M.G.M. MedicalCollege, Indore, for statistical analysis and Mrs PrasannaNair for secretarial assistance.

References

1. Rich S. Executive summary from the World Symposium on PrimaryPulmonary Hypertension 1998. Available from http/www.who.int/ncd/cvd/pph.html

2. British Cardiac Society Guidelines and Medical Practice Committee.Recommendations on the management of pulmonary hypertensionin clinical practice. Heart 2001; 86 (Suppl): I1–I13

3. Kothari SS, Bahl VK. Primary pulmonary hypertension—an update.Indian Heart J 2002; 54: 255–260

4. Sanchez LS, de la Monte SM, Filippov G, Jones RC, Zapol WM, BlochKD. Cyclic-GMP-binding, cyclic-GMP-specific phosphodiesterase(PDE5) gene expression is regulated during rat pulmonarydevelopment. Pediatr Res 1998; 43: 163–168

5. Zhao L, Mason NA, Morrell NW, Kojonazarov B, Sadykov A, MaripovA, et al. Sildenafil inhibits hypoxia-induced pulmonary hypertension.Circulation 2001; 104: 424–428

6. Wilkens H, Guth A, Konig J, Forestier N, Cremers B, Hennen B, et al.Effect of inhaled iloprost plus oral sildenafil in patients with primarypulmonary hypertension. Circulation 2001; 104: 1218–1222

7. Mychaskiw G, Sachdev V, Heath BJ. Sildenafil (viagra) facilitatesweaning of inhaled nitric oxide following placement of a biventricular-

assist device. J Clin Anesth 2001; 13: 218–2208. Abrams D, Schulze-Neick I, Magee AG. Sildenafil as a selective

pulmonary vasodilator in childhood primary pulmonaryhypertension. Heart 2000; 84: E4

9. Weimann J, Ullrich R, Hromi J, Fujino Y, Clark MW, Bloch KD, et al.Sildenafil is a pulmonary vasodilator in awake lambs with acutepulmonary hypertension. Anesthesiology 2000; 92: 1702 –1712

10. Borg GA. Psychophysical bases of perceived exertion. Med Sci SportsExerc 1982; 14: 377–381

11. Oh JK, Seward JB, Tajik AJ. The echo manual. 2nd ed. Philadelphia:Lippincott Williams & Wilkins; 1999. p. 215

12. Galie N, Manes A, Branzi A. Medical therapy of pulmonaryhypertension. The prostacyclins. Clin Chest Med 2001; 22: 529–537

13. Hoeper MM, Galie N, Simonnean G, Rubin LJ. New treatments forpulmonary arterial hypertension. Am J Respir Crit Care Med 2002;165: 1209–1216

14. Nagaya N, Shimizu Y, Satoh T, Oya H, Uematsu M, Kyotani S, et al.Oral beraprost sodium improves exercise capacity and ventilatoryefficiency in patients with primary or thromboembolic pulmonaryhypertension. Heart 2002; 87: 340–345

15. Rubin LJ, Badesch DB, Barst RJ, Galie N, Black CM, Keogh A, et al.Bosentan therapy for pulmonary artery hypertension. N Engl J Med2002; 346: 896–903

16. Watanabe H, Ohashi K, Takeuchi K, Yamashita K, Yokoyama T, TranQK, et al. Sildenafil for primary and secondary pulmonaryhypertension. Clin Pharmacol Ther 2002; 71: 398–402

17. Ghobrani HA, Weidmann R, Rose F, Olscherski H, Schermuly RT,Weissmann N. Combination therapy with oral sildenafil and inhalediloprost for severe pulmonary hypertension. Ann Intern Med 2002;136: 515–522

18. Prasad S, Wilkinson J, Gatzoulis MA. Sildenafil in primary pulmonaryhypertension. N Engl J Med 2000; 343: 1342

19. Miyamoto S, Nagaya N, Satoh T, Kyotani S, Sakamaki F, Fujita M, etal. Clinical correlates and prognostic significance of six-minute walktest in patients with primary pulmonary hypertension. Comparisonwith cardiopulmonary exercise testing. Am J Respir Crit Care Med2000; 161: 487–492

20. Barst RJ, Rubin LJ, Long WA, McGoon MD, Rich S, Badesch DB, et al.A comparison of continuous intravenous epoprostenol (prostacyclin)with conventional therapy for primary pulmonary hypertension. ThePrimary Pulmonary Hypertension Study Group. N Engl J Med 1996;334: 296–302

21. Galie N, Humbert M, Vachiery JL, Vizza CD, Kneussl M, Manes A, etal. Effects of beraprost sodium, an oral prostacyclin analogue, inpatients with pulmonary arterial hypertension: a randomized, double-blind, placebo-controlled trial. J Am Coll Cardiol 2002; 39: 1496–1520

22. Atz AM, Wessel DL. Slidenafil ameliorates effects of inhaled nitric oxidewithdrawal. Anesthesiology 1999; 91: 307–310

23. Michelakis ED, Tymchak W, Lien D, Dasilva L, Hashimoto K, ArcherSL. Oral sildenafil is an effective and specific pulmonary vasodilatorin patients with pulmonary arterial hypertension [Abstr]. J Am CollCardiol 2002; 39 (Suppl): 1081

24. Erickson S, Reyes J, Bohn D, Adatia I. Sildenafil (Viagra) in childhoodand neonatal pulmonary hypertension [Abstr]. J Am Coll Cardiol 2002;39 (Suppl): 1097–1098

25. Arruda-Olson AM, Mahoney DW, Nehra A, Leckel M, Pellikka PA.Cardiovascular effects of sildenafil during exercise in men with knownor probable coronary artery disease: a randomized crossover trial.JAMA 2002; 287: 719–725

IHJ-395-02.p65 5/6/2003, 10:02 AM59

60 Puri et al. Polymorphism in the Apo B-100 Gene Indian Heart J 2003; 55: 60–64

Background: The aim of this study was to investigate the association of apolipoprotein B gene polymorphismswith coronary artery disease and lipid levels in Indians.Methods and Results: One hundred patients of angiographically proven atherosclerotic coronary artery diseaseand one hundred age- and sex-matched control subjects (treadmill negative) were included in the study. Serumlipids including cholesterol, triglycerides, high-density lipoprotein, low-density lipoprotein, very low-densitylipoprotein, and apolipoprotein B were analyzed. Genomic DNA was extracted and the apolipoprotein B 3'hypervariable region amplified by polymerase chain reaction. Regions carrying Xba1, EcoR1, and Msp1restriction sites present in the apolipoprotein B gene were amplified and digested separately by the respectiveenzymes. Restriction fragment length polymorphism analysis showed that EcoR1 with the R+/R+ genotype wassignificantly more common in patients with coronary artery disease. Overall, the genotypes EcoR1+/+, Msp1+/+,Xba1+/+ and Eco R1+/+ Msp1+/–, Xba1–/– were significantly more common in patients as compared to controls(p<0.05). When gene polymorphisms were compared with lipid abnormalities, the genotypesEcoR1+/+, Xba1–/–, and Msp1+/+ were more frequent in patients with elevated apolipoprotein B and verylow-density lipoprotein levels. On the other hand, these genotypes were less common in patients with increasedtotal cholesterol and low-density lipoprotein levels. When we studied the individual alleles of the variable numberof tandem repeats region, we observed that allele 34 was significantly increased in patients with coronaryartery disease as compared to controls. Allele 36 was present with a frequency of 1% in controls while it wastotally absent in patients.Conclusions: This study identifies the apolipoprotein B gene polymorphism associated with coronary arterydisease. An association between apolipoprotein B gene polymorphisms and elevated apolipoprotein B and verylow-density lipoprotein levels was observed. However, there was no positive association with other elevated lipidlevels in North Indians from Uttar Pradesh. (Indian Heart J 2003; 55: 60–64)

Key Words: Apolipoprotein B, Polymorphism, Coronary artery disease

Polymorphisms in the Apolipoprotein B-100 Gene:Association With Plasma Lipid Concentration

and Coronary Artery Disease

Ratna Dua Puri, Satyendra Tewari, Nakul Sinha, V Ramesh, Faisal Khan,Vivek P Singh, Suraksha Agrawal

Departments of Medical Genetics, Cardiology and Pathology, Sanjay Gandhi PostgraduateInstitute of Medical Sciences, Lucknow

Elevated serum low-density lipoprotein (LDL)concentration is an important risk factor for developing

atherosclerotic coronary artery disease (CAD) in humans.1

Apolipoprotein (apo) B-100 is the principal proteincomponent of LDL. The interaction of apo B-100 with LDLreceptors mediates the uptake of LDL from the liver andperipheral cells; hence, apo B-100 plays an important role

in cholesterol homeostasis.2 The apo B gene is localized onchromosome 2, and the complete structure of the humanapo B-100 gene has been elucidated.3

Cloning and sequencing of the apo B gene has made itpossible to study the variation in the apo B gene at the DNAlevel. A 15 bp AT-rich hypervariable region (HVR) is locatedadjacent to the 3' end of the apo B gene. It consists of avariable number of tandem repeat sequences (VNTR).Previous studies have reported that some of the “3' VNTR”alleles4–6 and restriction fragment length polymorphisms(RFLP)4,7,8 of apo B-100 are directly associated with CAD,

Original Article

Correspondence: Professor Suraksha Agrawal, Department of MedicalGenetics, Sanjay Gandhi Postgraduate Institute of Medical Sciences, RaebareliRoad, Lucknow 226014. e-mail : [email protected]

IHJ-417-02.p65 5/6/2003, 10:03 AM60

Indian Heart J 2003; 55: 60–64 Puri et al. Polymorphism in the Apo B-100 Gene 61

or with variations in plasma lipids. Interestingly, theassociation of apo B-100 VNTR and RFLPs with plasmalipid concentration or CAD varies in different ethnic groups9

and has not always been found to be associated with CADor hyperlipidemia.10,11

The incidence of CAD is increasing in India, especiallyin the younger population. Hence, it is important todelineate risk factors for CAD in Indians. There are veryfew studies from India which report the effect of apo Bpolymorphisms on CAD or lipid levels. The present study isan attempt to analyze the allele frequency of apo B-100 3'VNTR and apo B-100 RFLPs in the Indian population andto determine their association with plasma lipidconcentration and CAD.

Methods

Subjects: One hundred patients of angiographicallyproven CAD evaluated at the Cardiology Department of theSanjay Gandhi Postgraduate Institute of Medical Sciences,Lucknow, Uttar Pradesh, India were included in the study.Patients less than 6 weeks post-myocardial infarction (MI)were excluded from the study. One hundred age- and sex-matched controls were also selected for the study. Thecontrols were subjected to a treadmill test to ensure thatthey were not suffering from any CAD. Further, all controlswith hypertension, diabetes, and endocrine or metabolicdisorders were excluded. Informed consent was taken fromboth the patients and controls.

Lipid analysis: Blood samples were taken after a fastingperiod of 12 hours from all the patients and controls. Serumlipids including cholesterol, triglycerides, high-densitylipoprotein (HDL), LDL, and very low-density lipoprotein

(VLDL) were analyzed according to methods previouslydescribed.12,13 Apolipoprotein B levels were assessed by theimmunoturbidimetric immunoassay using a commercialkit (Randox Laboratories Ltd, UK).

DNA preparation: Genomic DNA was extracted by thehigh salting-out procedure followed by phenol chloroformextraction and ethanol precipitation.14

Analysis of VNTR: Apolipoprotein B 3' HVR amplificationwas carried out by PCR using a forward and reverseoligonucleotide primer encompassing the entireapo B 3' VNTR sequence. The sequence of the primerused was 5' ATGCAAACGGAGAAATTATG 3' and 5'CCTTCTCACTTGCCAAATAC 3'. The polymerase chainreaction (PCR) was performed in an M.J. Research Inc.Thermocycler, with 26 cycles of denaturation at 94 oC for1 min, and annealing and extension at 58 oC for 6 min. Theamplified product was electrophoresced in 5%polyacrylamide gel, and allele sizing was done using theapo B 3' VNTR allelic ladder and a commercial ladder 50 bpin size (Fig. 1).

Analysis of RFLP: Regions of the apo B gene carryingEcoR1, Msp1, and Xba1 restriction sites were amplifiedseparately using their respective primers: 5' CTG, GCT, TGC,TAA CCT, CTC, TG and 3' GAG, AAG, CTT, CCT, GAA, GCT,CG for EcoR1, 5' TCT, CGG, GAA, TAT, TCA, GGA, ACT, ATT,G and 3' CTA, AGG, ATC, CTG, CAA, TGT, CAA, GGT forMsp1 and 5' GGA,GAC, TAT, TCA, GAA, GCT, AA and3' GAA, GAG, CCT, GAA, GAC, TGA, CT for Xba1. Theamplified product was separately digested with therespective enzymes as previously described.15 Alleles of eachpolymorphic site were classified as (+) or (–) according to

C D

Fig. 1. Apo B-gene polymorphism. A: Apo B 3' VNTR analysis; B: Msp1 RFLP analysis; C: Xba1 RFLP analysis; D: EcoR1 RFLP analysis

Lane 1: 25 bp ladderLane 2 and 4: Undigested samplesLane 3 and 5: Digested sample (+/+)

1 2 3 4 5

349 bp239 bp110 bp

Lane 1: Undigested sampleLane 2: Digested sample (+/–)Lane 3: 25 bp ladder

1 2 3

Lane 1: Undigested sampleLane 2: Digested sample (+/–)Lane 3: 50 bp ladder

1 2 3

675 bp

415 bp

260 bp

Lane 1–3: CAD patients’ samplesLane 4: 50 bp ladder

1 2 3 4

35 bp33 bp31 bp

BA

IHJ-417-02.p65 5/6/2003, 10:03 AM61


the presence or absence at the cutting site of eachrestriction enzyme, respectively.

Statistical analysis: Allele and genotypic frequencyanalysis for apo B 3' VNTR and RFLP was done using thePOPGENE-16 version. The Student’s t test was used forcomparisons using the SPSS software.

Results

One hundred patients with angiographically proven CAD(90 males, 10 females) and 100 normal controls (90 malesand 10 females) were evaluated. The mean age of thepatients was 50.74±9.7 years, while that of the controlswas 50.41±12.23 years.

Lipid levels: The mean lipid levels of patients withCAD and controls were not significantly dif ferent.However, lipid levels were higher in patients as comparedto controls.

Allele frequency and genotyping for EcoR1, Xba1,and Msp1: The comparison of the genotypic and allelefrequencies of the EcoR1, Xba1, and Msp1 RFLPs of theapo B gene in patients with CAD and controls is shown inTable 1. The only significant RFLP pattern was with EcoR1,with the R+/R+ genotype expressed more frequently inpatients with CAD (p=0.001), while the R+/R– genotype was

Table 1. Allele and genotype frequencies of apo B geneRFLPs in patients with CAD and controls

CAD (n=100) Controls (n=100)

Allele frequencyEcoR1 (+) 0.950 0.820EcoR1 (–) 0.050 0.180Xba1 (+) 0.285 0.225Xba1 (–) 0.715 0.775Msp1 (+) 0.825 0.850Msp1 (–) 0.175 0.150

Genotypic frequency (%)*R+/R– 8 34*R+/R+ 91 65*

Genotypes *E+ E+ M+ M+ X+ X+ 10 2**E+ E+ M+ M– X– X– 20 8**E+ E– M+ M– X– X– 1 9**

+, – refers to the presence or absence of a cutting site for restrictionendonuclease*only significant genotype frequencies have been shown (p<0.005);**only significant genotype frequencies have been shown (p<0.001)Apo B: apolipoprotein B; RFLP: restriction fragment length polymorphism;CAD: coronary artery disease; R: EcoR1; M: Msp1; X: Xba1

Fig. 2. Apo B 3' VNTR allele frequency distribution among patients with CADand controls.

more common in controls (p=0.001). The genotypes E+ E+

M+ M+ X+ X+ and E+ E+ M+ M– X– X– were significantly moreprevalent in patients with CAD (p<0.05). The genotype E+

E– M+ M– X– X–, on the other hand, was more common incontrols (p<0.05).

VNTR analysis: Relative frequencies of VNTR in patientswith CAD and controls are shown in Fig. 2. The VNTR allele34 was significantly increased in patients with CAD ascompared to that in controls, while allele 36 wassignificantly increased in controls. Interestingly, VNTRallele 36 was totally absent in patients.

Correlation of VNTR, RFLP, and lipid parameterswith the age of the patients with CAD: When youngpatients with CAD (age <45 years, n=46) were analyzed,there was no significant increase in any allelic andgenotypic frequencies of different apo B polymorphisms ascompared to older patients (age ≥45 years, n=54). However,when patients with CAD were compared to controls, theolder patients with CAD had a higher frequency of EcoR1(+) allele, while the EcoR1 (–) allele was less common inboth the age subgroups.

Correlation of RFLP and HVR allele with lipidparameters: In patients with CAD, the genotypes R+/R+,X–/X– and M+/M+ were frequent in those with elevated apoB and VLDL levels. On the other hand, these genotypes wereless frequent in patients with increased total cholesterol andLDL levels. The allele Msp1 (–) was significantly lesscommon in patients with elevated LDL levels.

Among controls, the trend of association was the sameas in the patient group but the results were not significantin the controls, even in those with elevated apo B and VLDLlevels.

IHJ-417-02.p65 5/6/2003, 10:03 AM62

Indian Heart J 2003; 55: 60–64 Puri et al. Polymorphism in the Apo B-100 Gene 63

Discussion

On the basis of evidence obtained over many years fromepidemiological and trial data, LDL- and HDL-cholesterollevels have been the recommended lipid variables ininternational guidelines for treatment. However, newinformation shows the importance of apo B and apo A-I asrisk predictors for CAD.16 A reason why apo B may be astronger predictor of risk than LDL-cholesterol is that apo Bis present not only in LDL but also in VLDL, intermediatedensity lipoprotein, and lipoprotein (a). Therefore, the sumof apo B concentrations in all atherogenic particles mightbe a better risk marker than total cholesterol and LDL-cholesterol levels only.16 In our study, the entire lipid levels,including those of apo B, were higher in patients ascompared to controls; however, these did not reachstatistical significance. One possible explanation for thiscould be that lipid-lowering drugs were not withheld priorto lipid testing for this study as this would not have beenethically justifiable in patients who had angiographicallyproven CAD with dyslipidemia and were already on lipid-lowering drugs.

It has been suggested that in addition to quantitativevariation in apo B levels in the plasma, genetic variation atthe apo B locus may be a new and independent risk factorfor CAD.17

It has been reported that different VNTR alleles may beassociated with CAD and hyperlipidemia.4–6 We found anassociation of VNTR 34 in patients with CAD similar tothat reported by Moreel et al.4 The allelic variation of apo Bgene polymorphisms may have some association withvarious ethnic groups. Deka et al.9 in a study of allelicfrequency distribution at the hypervariable locus 3' to theapo B gene in 5 human populations found 12 segregatingalleles in 319 individuals. They found that the two mostfrequent alleles, 37 and 39, were present in all thepopulations. When we studied VNTR we found that allele34 was significantly increased in patients as compared tocontrols, while allele 36 was completely absent in patientsbut present in controls. This clearly demonstrates thepresence of allelic frequency variation in dif ferentpopulations. These association studies may be of some usewhen genetic factors are considered as one of thepredisposing causes.

There are few studies on Indians that show theassociation of the apo B gene 3' HVR alleles with CAD andplasma lipid levels. Renges et al.18 found an associationbetween Xba1 and ins/del polymorphisms of the apo B genewith total cholesterol and HDL-cholesterol levels in SouthAsians in the UK. Saha et al.19 reported that DNApolymorphisms of the apo B gene were associated with

obesity and serum lipids in healthy Indians in Singapore.Misra et al.,20 on the other hand, have reported that apo B(Xba1 and EcoR1) polymorphisms did not appear toinfluence serum lipid levels. In our study, we found thatthe genotypes R+/R+, X–/X– and M+/M+ were frequent inpatients with elevated apo B and VLDL levels. However,these genotypes were decreased in patients with elevatedcholesterol and LDL levels.

Pan et al.6 found that apo B 3' VNTR genotypic variationhad little impact on the risk of dyslipidemia in a Taiwanesepopulation but despite this, the long apo B 3' VNTR allelesoccurred more frequently in patients with CAD. Hegele etal.17 also found a significant correlation between apo B genepolymorphisms and CAD, without any significantassociation with either LDL or VLDL. Similar results werereported in a porcine model of atherosclerosis in which anapo B genetic variant was associated with atherosclerosis,despite normal lipid levels.21 Our results show that the apoB gene variations possibly do not affect the apo B–LDL-receptor binding region, and thus do not affect the lipidlevels. Other studies have also shown similar trends.17 Wespeculated that a mutation in the protein coding region ofapo B could affect the interaction of LDL with monocytemacrophages, endothelial cells, ground substance, clottingfactors, or the immune system in a manner that wouldpromote atherogenesis.22

The clinical relevance of apo B polymorphisms stillremains unclear. All studies in the past reflect the geneticheterogeneity of the apo B gene. As CAD is a multifactorialdisease, the apo B gene alone may not have a direct effecton the lipid profile or severity and prematurity of CAD.However, it does emphasize the importance of such studies,which may help in future to delineate the high-risk groupfor CAD, and may be of use in the genetic screening ofpatients with CAD belonging to different populations.

References

1. Kannel WB, Castelli WP, Gordon T. Cholesterol in the prediction ofatherosclerotic disease. New perspectives based on the FraminghamStudy. Ann Intern Med. 1979; 90: 85–91

2. Brown MS, Goldstein JL. How LDL receptors influence cholesterol andatherosclerosis. Sci Am 1984; 251: 58–66

3. Knott TJ, Rall SC Jr, Innerarity TL, Jacobson SF, Urdea MS, Levy-WilsonB, et al. Human apolipoprotein B: structure of carboxyl-terminaldomains, sites of gene expression, and chromosomal localization.Science 1985; 230: 37–43

4. Moreel JF, Roizes G, Evans AE, Arveiler D, Cambou JP, Souriau C, et al.The polymorphism ApoB/4311 in patients with myocardial infarctionand controls: the ECTIM Study. Hum Genet 1992; 89: 169–175

5. Cavalli SA, Hirata MH, Salazar LA, Diament J, Forti N, Giannini SD, etal. Apolipoprotein B gene polymorphisms: prevalence and impact onserum lipid concentrations in hypercholesterolemic individuals fromBrazil. Clin Chim Acta 2000; 302: 189–203

IHJ-417-02.p65 5/6/2003, 10:03 AM63


6. Pan JP, Chiang AN, Chou CY, Chan WL, Tai JJ. Polymorphisms of theapolipoprotein B 3' variable number of tandem repeats regionassociated with coronary artery disease in Taiwanese. J Formos MedAssoc 1998; 97: 233–238

7. de Padua Mansur A, Annicchino-Bizzacchi J, Favarato D, Avakian SD,Machado Cesar LA, Franchini Ramires JA. Angiotensin-convertingenzyme and apolipoprotein B polymorphisms in coronary arterydisease. Am J Cardiol 2000; 85: 1089–1093

8. Machado MO, Hirata MH, Bertolami MC, Hirata RD. Apo B genehaplotype is associated with lipid profile of higher risk for coronaryheart disease in Caucasian Brazilian men. J Clin Lab Anal 2001; 15:19–24

9. Deka R, Chakraborty R, DeCroo S, Rothhammer F, Barton SA, FerrellRE. Characteristics of polymorphism at a VNTR locus 3' to theapolipoprotein B gene in five human populations. Am J Hum Genet1992; 51: 1325–1333

10. Helio T, Palotie A, Totterman KJ, Ott J, Kauppinen-Makelin R, TikkanenMJ. Lack of association between the apolipoprotein B gene 3'hypervariable region alleles and coronary artery disease in Finnishpatients with angiographically documented coronary artery disease.J Intern Med 1992; 231: 49–57

11. Turner PR, Talmud PJ, Visvikis S, Ehnholm C, Tiret L. DNApolymorphisms of the apoprotein B gene are associated with alteredplasma lipoprotein concentration but not with perceived risk ofcardiovascular disease. European Atherosclerosis Research Study.Atherosclerosis 1995; 116: 221–234

12. Allain CC, Poon LS, Chan CS, Richmond W, Fu PC. Enzymaticdetermination of total serum cholesterol. Clin Chem 1974; 20: 470–475

13. Fossati P, Prencipe L. Serum triglycerides determined colorometricallywith an enzyme that produces hydrogen peroxide. Clin Chem 1982;28: 2077–2080

14. Olerup O, Zetterquist H. HLA-DR typing by PCR amplification withsequence-specific primers (PCR-SSP) in 2 hours: an alternative toserological DR typing in clinical practice including donor–recipientmatching in cadaveric transplantation. Tissue Antigens 1992; 39:225–235

15. Pan JP, Chiang AN, Tai JJ, Wang SP, Chang MS. Restriction fragmentlength polymorphisms of apolipoprotein B gene in Chinese populationwith coronary heart disease. Clin Chem 1995; 41: 424–429

16. Walldius G, Jungner I, Holme I, Aastveit AH, Kolar W, Steiner E. Highapolipoprotein B, low apolipoprotein A-I, and improvement in theprediction of fatal myocardial infarction (AMORIS study): aprospective study. Lancet 2001; 358: 2026–2033

17. Hegele RA, Huang LS, Herbert PN, Blum CB, Buring JE, HennekensCH, et al. Apolipoprotein B-gene DNA polymorphisms associated withmyocardial infarction. N Engl J Med 1986; 315: 1509–1515

18. Renges HH, Wile DB, McKeigue PM, Marmot MG, Humphries SE.Apolipoprotein B gene polymorphisms are associated with lipid levelsin men of South Asian descent. Atherosclerosis 1991; 91: 267–275

19. Saha N, Tay JS, Heng CK, Humphries SE. DNA polymorphisms of theapolipoprotein B gene are associated with obesity and serum lipids inhealthy Indians in Singapore. Clin Genet 1993; 44: 113–120

20. Misra A, Nishanth S, Pasha ST, Pandey RM, Sethi P, Rawat DS.Relationship of Xba1 and EcoR1 polymorphisms of apolipoprotein-Bgene to dyslipidemia and obesity in Asian Indians in North India.Indian Heart J 2001; 53: 177–183

21. Rapacz J, Hasler-Rapacz J. Investigations on the relationship betweenimmunogenetic polymorphisms of betalipoproteins and thebetalipoprotein and cholesterol in swine. In: Lenzi S, Descovich GC(eds). Atherosclerosis and cardiovascular disease. Bologna: EditriceCompositori; 1984. pp. 99–108

22. Ross R. The pathogenesis of atherosclerosis—an update. N Engl J Med1986; 314: 488–500

IHJ-417-02.p65 5/6/2003, 10:03 AM64

Indian Heart J 2003; 55: 65–67 Roy et al. Isolation of the Left Subclavian Artery 65

Isolation of the Left Subclavian Artery

Ambuj Roy, Shyam S Kothari, Harminder Singh, Sanjiv SharmaCardiothoracic Centre, All India Institute of Medical Sciences, New Delhi

Isolation of the left subclavian artery (LSA) from the rightaortic arch is a rare vascular anomaly,1 in which the left

subclavian artery does not communicate with the aorticarch but instead is connected to the left pulmonary arteryby the ductus arteriosus. We report two cases of isolatedLSA, one of which was diagnosed clinically.

Case Report

Case 1: A two-year-old child presented to us with a historyof mild cyanosis since 6 months of age and a history ofone episode of a cyanotic spell one month ago. Onexamination he had mild cyanosis and clubbing. The leftarm pulse was weaker compared to the right arm. The bloodpressure was 70/40 mmHg in the left arm and 104/60mmHg in the right arm. Cardiovascular examinationrevealed a quiet precordium, single second heart sound, anda grade III/VI ejection systolic murmur in the leftparasternal area. Chest X-ray showed a right aortic arch,no cardiomegaly and oligemic lung fields. Anelectrocardiogram (ECG) showed right axis deviation withright ventricular hypertrophy. Typical anatomy of thetetralogy of Fallot (TOF) was seen on echocardiogram. Inview of TOF, right aortic arch and weak left arm pulse,isolation of the LSA was suspected. Subsequentcatheterization confirmed the presence of TOF withconfluent and good-sized pulmonary arteries. Aortogramrevealed a right aortic arch. The innominate, right and leftcommon carotid arteries were normal. On aortography, theLSA did not show any opacification (Fig. 1). Late framesshowed faint opacification of the LSA by retrograde filling

from the left vertebral artery, suggesting the presence ofisolation of the LSA (Fig. 2). The child underwent correctivesurgery for TOF and reimplantation of the LSA to the leftcommon carotid artery. He is doing well at follow-up after6 months.

Brief Report

Correspondence: Dr Shyam S Kothari, Department of Cardiology,Cardiothoracic Centre, All India Institute of Medical Sciences, New Delhi110029. e-mail: [email protected]

Two cases of isolation of the left subclavian artery from the aortic arch are reported for the rarity of this lesion.One patient was diagnosed clinically, the other after angiography. The isolated left subclavian artery wasreimplanted in one patient. This rare anomaly has clinical and surgical relevance and should be diagnosed bydiligent clinical and angiographic evaluation. (Indian Heart J 2003; 55: 65–67)

Key Words: Isolated left subclavian artery, Subclavian steal, Arch anomalies

Fig. 1. Arch aortogram showing the branches of the aortic arch with failure ofopacification of the left subclavian artery.

Case 2: A 3.5-month-old male child was brought with ahistory of cyanosis and failure to thrive, and recurrentepisodes of lower respiratory tract infection. There was nohistory of stridor, wheezing or loss of consciousness. Hewas born at term and the antenatal history wasunremarkable. His present weight was 2.8 kg and the birthweight was 2.5 kg. His pulse rate was 110/min, regular,and all peripheral pulses were palpable and equal. He hadmild cyanosis on room air. His blood pressure was 80/56

IHJ-386-02.p65 5/6/2003, 10:03 AM65

66 Roy et al. Isolation of the Left Subclavian Artery Indian Heart J 2003; 55: 65–67

mmHg in the right arm and 82/60 mmHg in the left arm.The arterial saturation by oximetry was 85% in room airin all four limbs. Cardiovascular examination revealedcardiomegaly, single second heart sound, left ventricularthird heart sound, and a pansystolic murmur at the apex.ECG showed left axis deviation with a counterclockwise loopand right ventricular hypertrophy. Chest X-ray revealedcardiomegaly, a right aortic arch, and increased pulmonaryblood flow. Echocardiography showed atrioventricularseptal defect (AVSD) and a double outlet right ventricle(DORV) with dilated left ventricle and atrium. The greatvessels were malposed with an anterior aorta. There wasno pulmonic stenosis. Cardiac catheterization andaortography were performed. The aortogram revealed aright aortic arch with absence of filling of the LSA fromthe aorta (Fig. 3) and delayed retrograde filling from theleft vertebral artery. A selective left common carotidangiogram showed retrograde filling of the LSA throughthe left vertebral artery (Fig. 4). Right ventricularangiogram confirmed the presence of a DORV. Hence, adiagnosis of cyanotic congenital heart disease with DORV,AVSD, and isolated LSA was made. The child subsequentlydeveloped lower respiratory tract infection and succumbedto it.

Discussion

Isolation of the LSA is defined as a loss of continuitybetween the LSA and the aorta with a persistent connectionto the homolateral pulmonary artery through the ductus

arteriosus, which may be patent or closed. There isretrograde filling of the LSA through the left vertebralartery, which is fed via the circle of Willis. This leads to thecongenital subclavian steal phenomenon. However, if theductus arteriosus is patent, a shunt between the pulmonaryand the systemic circulation is established. Therefore, theinverted left vertebral blood flow supplies only the proximalpart of the LSA, before entering the pulmonary artery,leading to pulmonary artery steal.2 The distal LSA in thesecases is usually supplied by anastomoses between the leftexternal carotid artery and the LSA.

Fig. 2. Late frames of the same injection as in Fig.1 show retrograde filling ofthe left vertebral artery (white arrow) which supplies the left subclavian artery(black arrow).

Fig. 3. Arch aortogram showing opacification of the branches of the aortic arch.Note the absence of opacification of the left subclavian artery.

Fig. 4. Late frames from selective left common carotid angiogram showingretrograde filling of the left vertebral artery (arrowhead) and the left subclavianartery (black arrow).

IHJ-386-02.p65 5/6/2003, 10:03 AM66

Indian Heart J 2003; 55: 65–67 Roy et al. Isolation of the Left Subclavian Artery 67

Isolated LSA has been diagnosed from infancy to as lateas 52 years of age. The condition is diagnosed usually dueto the presence of associated cardiac anomalies. However,symptoms of subclavian steal and limb ischemia are presentin some patients. Luetmer et al.3 in their review of theliterature reported the presence of symptoms ofvertebrobasilar insufficiency in 5 of 30 patients (17%) andlimb ischemia in 5 patients. These symptomatic patientsare usually older and develop symptoms in the second andthird decades of life.

Isolation of the LSA should be suspected if the left armpulse or blood pressure is found to be lower than that ofthe right arm in a patient with a right aortic arch, as in thefirst case reported here. These features may be absent dueto the associated lesions, which result in equalization ofpressure in the aorta and pulmonary arteries. Bedsidediagnosis of an isolated LSA is dif ficult under suchcircumstances, as in the second case. The diagnosis of anisolated LSA can be reliably established by aortography,wherein delayed films show late opacification of the LSAfrom the left vertebral artery.

An isolated LSA is a rare congenital anomaly.1 It is almostalways associated with a right aortic arch.3 However, only0.8% of patients with a right aortic arch have an isolatedLSA. A solitary case of a left arch with an isolated LSA hasbeen reported in the literature.4 Nearly 60% of the patientshave associated intracardiac anomalies of which TOF is themost common.3 Other associations include atrial andventricular septal defects, bilateral patent ductus arteriosus,truncus arteriosus, transposition of the great arteries,hypoplastic left heart syndrome, and DORV.3,5,6

The diagnosis of a right aortic arch with an isolated LSAis imperative for the future management of patients withintracardiac lesions such as TOF,7 as a Blalock–Taussigshunt in these patients is done on the side opposite the arch.However, because of decreased pressure and flow in anisolated LSA, it cannot be used for a left Blalock–Taussigshunt in these patients.

The embryogenesis of this entity can be explained byusing the hypothetical double arch system proposed byStewart et al.8 and Edwards.9 There is interruption of theleft aortic arch at two levels, one between the left commoncarotid artery and the LSA, and the second between theleft ductus arteriosus and left dorsal aortic root. This resultsin a right aortic arch with typical branching of this entity.

The branches of the right aortic arch in this entity, in order,are the left common carotid, the right common carotid, andthe right subclavian arteries. The LSA becomes detachedfrom the aorta and is connected to the pulmonary arteryby the ductus arteriosus. Similarly, interruption of the leftarch at corresponding points leads to a left aortic arch withisolation of the right subclavian artery.7,10

Surgical treatment of this condition is required ifsymptoms of subclavian steal occur or if surgery is donefor concomitant heart disease. Corrective surgery involvesanastomosis of the LSA with the arch or the left commoncarotid artery. Recent catheter-based abolition ofpulmonary steal has been described by closure of the patentductus arteriosus.11

References

1. Stewart JR, Kincaid OW, Titus JL. Right aortic arch: plain filmdiagnosis and significance. Am J Roentgenol Radium Ther Nucl Med1966; 97: 377–389

2. Carano N, Piazza P, Agnetti A, Squarcia U. Congenital pulmonary stealphenomenon associated with tetralogy of Fallot, right aortic arch,and isolation of the left subclavian artery. Pediatr Cardiol 1997; 18:57–60

3. Leutmer PH, Miller GM. Right aortic arch with isolation of the leftsubclavian artery: case report and review of the literature. Mayo ClinProc 1990; 65: 407–413

4. Engelman DA, Mortazavi A. Congenital subclavian steal syndromein an adult with a left aortic arch. Tex Heart Inst J 1998; 25: 216–217

5. McMahon CJ, Thompson KS, Kearney DL, Nihill MR. Subclavian stealsyndrome in anomalous connection of the left subclavian artery tothe pulmonary artery in d-transposition of the great arteries. PediatrCardiol 2001; 22: 60–62

6. Patel CR, Spector ML, Zahka KG. Hypoplastic left heart syndrome withright aortic arch, bilateral arterial ducts and origin of the leftsubclavian artery from the left pulmonary artery. Cardiol Young 1999;9: 331–334

7. Rodriguez L, Izukawa T, Moes CA, Trusler GA, Williams WG. Surgicalimplications of right aortic arch with isolation of left subclavianartery. Br Heart J 1975; 37: 931–936

8. Stewart JR, Kincaid OW, Edwards JE. An atlas of vascular rings andrelated malformations of the aortic arch system. Springfield: Illinois.Charles C Thomas, 1964

9. Edwards JE. Anomalies of the derivatives of the aortic arch system.Med Clin North Am 1948; 32: 925–949

10. Garti IJ, Aygen MM. Left aortic arch with congenital isolation of theright subclavian artery. Pediatr Radiol 1980; 9: 241–243

11. Jones TK, Garabedian H, Grifka RG. Right aortic arch with isolationof the left subclavian artery, moderate patent ductus arteriosus, andsubclavian steal syndrome: a rare aortic arch anomaly treated withthe Gianturco–Grifka vascular occlusion device. Catheter CardiovascInterv 1999; 47: 320–322

IHJ-386-02.p65 5/6/2003, 10:03 AM67

68 Pandurangi et al. Endoepicardial AV Pacing Indian Heart J 2003; 55: 68–70

Temporary Endoepicardial Atrioventricular SequentialPacing for Complete Heart Block Following Complex

Surgery for Congenital Heart Disease

Ulhas M Pandurangi, PJ Ruth, Satish C Toal, Snehal Kulkarni, KS Murthy, KM CherianDepartment of Cardiology, Madras Medical Mission, Institute of Cardiovascular Diseases, Chennai

Intracardiac surgical repair for complex congenital heartdisease is associated with complete heart block (CHB) in

up to 36% of cases.1 In the presence of ventriculardysfunction, atrioventricular (AV) sequential pacing ispreferred to ventricular pacing alone.2,3 We report the caseof a 12-year-old girl who developed CHB following mitralvalve replacement for residual mitral regurgitation andmoderate biventricular dysfunction. She had undergone anintracardiac repair in the past for complex congenital heartdisease. The epicardial ventricular pacing lead provided VVIpacing. However, the poor hemodynamics resulted in thepatient being dependent on the ventilator. A transvenousendocardial pacing lead was placed in the right atriumthrough the jugular vein and AV sequential pacingachieved using an epicardial ventricular pacing lead, withgratifying results. This report highlights the feasibility ofendoepicardial AV sequential pacing and the need for AVsynchrony in cases of postoperative CHB.

Case Report

A 12-year-old girl had undergone surgical closure for atrialand ventricular septal defects, with subpulmonic resectionat 6 years of age for situs inversus, dextrocardia, atrial andventricular septal defects, corrected transposition of thegreat arteries (c-TGA), subpulmonic stenosis andinterrupted inferior vena cava. The patient underwent a

redo surgery for new-onset progressive mitral valveregurgitation. An Omnicarbon prosthetic valve (29 mm)was placed at the mitral annulus. A bipolar epicardialventricular pacing lead was placed as there was CHBfollowing the surgery (Fig. 1). The patient was paced in theVVI mode (Fig. 2) using a temporary pacemaker (Medtronic5346 DDD temporary pulse generator, Minneapolis, MN,USA). The patient came out of bypass easily. However,despite good surgical results, the patient could not beweaned away from the ventilator even 48 hours post-procedure, as there were signs of reduced cardiac output

Brief Report

Correspondence: Dr Ulhas M Pandurangi, Department of Cardiology,Institute of Cardiovascular Diseases, Madras Medical Mission, 4-A,Dr J Jayalalitha Nagar, Mogappair, Chennai 600050.e-mail: [email protected]

Complete heart block following intracardiac surgical repair for complex congenital heart disease is notuncommon. In the presence of ventricular dysfunction, ventricular pacing alone may not improve the cardiacoutput. We report the feasibility and efficacy of endoepicardial atrioventricular sequential pacing in a case ofpostoperative complete heart block. (Indian Heart J 2003; 55: 68–70)

Key Words: Complete heart block, Atrioventricular sequential pacing, Cardiac surgery

Fig. 1. Complete heart block following surgery.

Fig. 2. Pacing in the VVI mode.

IHJ-392-02.p65 5/6/2003, 10:04 AM68

Indian Heart J 2003; 55: 68–70 Pandurangi et al. Endoepicardial AV Pacing 69

and systemic venous congestion. Despite inotropic support,the systolic blood pressure remained less than 90 mmHgand central venous pressure between 16 and 18 mmHg.The arterial blood gas analyses were suggestive of metabolicacidosis. The loss of atrial kick was considered acontributory factor for the reduced cardiac output and AVsequential pacing was considered to improve thehemodynamics.4 A temporary bipolar 5 F pacing lead wasplaced in the left-sided right atrium through the left internaljugular vein. The tip of this lead was positioned under thefluoroscope to achieve stable and satisfactory pacing andsensing parameters (Fig. 3). This endocardial lead and theepicardial lead were connected to the AV sequentialpacemaker (Medtronic 5388 Inc., Minneapolis, MN, USA).The AV delay was adjusted to 100 ms with the aid of venouspressure monitoring5 and bedside Doppler study6 of themitral inflow and mitral regurgitation (Figs 4 and 5). Therewas marked improvement in the hemodynamics soon afterAV sequential pacing was initiated. The systolic bloodpressure could be maintained between 120 and 130 mmHgwithout inotropic support, and there was promptnormalization of the central venous pressure and arterialblood gas parameters. The patient could be successfullyweaned away from the ventilator over 24 hours andunderwent dual-chamber permanent pacemakerimplantation before discharge from the hospital.

Discussion

Complete heart block following cardiac surgery for complexcongenital heart disease, like the one our patient had, isnot uncommon. Surgical repair for defects associated with

c-TGA results in CHB in up to 36% of cases.1,7 PostoperativeCHB may be managed satisfactorily in most cases byventricular pacing in the VVI mode alone throughepicardial leads.8,9 However, in certain situations in whichthe atrial contribution to ventricular output is critical, asin our case where there was moderate biventriculardysfunction, it has been shown that AV sequential pacingresults in better hemodynamics as compared to ventricularpacing alone. It has been common practice in some surgicalcenters to place atrial and ventricular epicardial pacingleads for post-surgical CHB to achieve AV sequentialpacing.10,11 However, if only one epicardial lead is availablefor ventricular pacing, AV sequential pacing can beobtained by an esophageal lead, which can sense atrialpotentials.12 The disadvantages of using esophageal leadsare unstable lead position and intermittent sensing failure.In addition, an esophageal lead is not practical in aconscious patient. Placement of another epicardial lead forsequential pacing would add to the perioperative morbidity.Using epicardial leads for ventricular pacing and a Swan–Ganz flow-directed pacing catheter for atrial pacing/sensingto obtain AV sequential pacing has met with variableresults.13 Endocardial atrial lead placement is anotheroption for obtaining AV sequential pacing. Since our patienthad interruption of the inferior vena cava, the jugular veinwas used for right atrial access. The left jugular vein was

Fig. 3. Fluoroscopy (PA view) showing endocardial atrial lead (single arrow)and epicardial ventricular lead (double arrow).

Fig. 4. Pacing in the VDD mode.

Fig. 5. Pacing in the DDD mode.

IHJ-392-02.p65 5/6/2003, 10:04 AM69

70 Pandurangi et al. Endoepicardial AV Pacing Indian Heart J 2003; 55: 68–70

chosen as the patient had situs inversus. This case reporthighlights the occasional need for AV sequential pacing ina case of postoperative CHB and the feasibility of achievingsequential pacing by endoepicardial leads.

References

1. de Leval MR, Bastos P, Stark J, Taylor JF, Macartney FJ, Anderson RH.Surgical technique to reduce the risks of heart block following closureof ventricular septal defect in atrioventricular discordance. J ThoracCardiovasc Surg 1979; 78: 515–526

2. Reiter MJ, Hindman MC. Hemodynamic ef fects of acuteatrioventricular sequential pacing in patients with left ventriculardysfunction. Am J Cardiol 1982; 49: 687–692

3. Kargul W, Gasior Z, Zajac T, Pruski M, Grzegorzewski B, Krauze J. Effectof ventricular and sequential stimulation on the left-ventricularsystolic function. Kardiol Pol 1992; 37: 8–12

4. Leonard JJ, Shaver J, Thompson M. Left atrial transport function. TransAm Clin Climatol Assoc 1980; 92: 133–141

5. Kern MJ, Deligonul U. Interpretation of cardiac pathophysiology frompressure waveform analysis: pacemaker hemodynamics. CathetCardiovasc Diagn 1991; 24: 22–27

6. Romero LR, Haffajee CI, Levin W, Doherty PW, Berkovits BV, AlpertJS. Non-invasive evaluation of ventricular function and volumesduring atrioventricular sequential and ventricular pacing. Pacing ClinElectrophysiol 1984; 7: 10–17

7. Doty DB, Truesdell SC, Marvin WJ Jr. Techniques to avoid injury ofthe conduction tissue during the surgical treatment of correctedtransposition. Circulation 1983; 68: II63–69

8. Fitzpatrick A, Sutton R. A guide to temporary pacing. BMJ 1992; 304:365–369

9. Waldo AL, MacLean WAH. Diagnosis and treatment of cardiacarrhythmias following open heart surgery. 2nd ed. Mt Kisco, New York:Futura; 1980. p. 64

10. Moungey SJ. Temporary A-V sequential pacemakers. Programmingand troubleshooting. Prog Cardiovasc Nurs 1989; 4: 49–60

11. Haywood DL. Temporary A-V sequential pacing using an epicardiallead system. Crit Care Nurse 1985; 5: 21–24

12. Roth JV, Huertas R. Atrioventricular sequential pacing usingtransesophageal atrial pacing in combination with a temporary DDDpacemaker for atrial tracking and ventricular pacing. J CardiothoracVasc Anesth 1995; 9: 255–258

13. Roth JV. Temporary transmyocardial pacing using epicardial pacingwires and pacing pulmonary artery catheters. J Cardiothorac VascAnesth 1992; 6: 663–667

IHJ-392-02.p65 5/6/2003, 10:04 AM70

Indian Heart J 2003; 55: 71–74 Baruah et al. Concurrent Coronary, Iliac and Renal Direct Stenting 71

Concurrent Coronary, Bilateral Iliac and Left RenalArtery Direct Stenting

DK Baruah, NK Panigrahi, AN SrinivasDepartment of Cardiology, Apollo Hospitals, Vishakhapatnam

Atherosclerotic vascular disease can involve both thecoronary and peripheral arterial systems.1 Patients

with peripheral vascular disease can present as acutecoronary syndrome due to coexisting coronary arterydisease and require proper therapeutic attention.Percutaneous intervention to treat both coronary andperipheral vascular disease may offer an alternative tosurgery in such situations.2 Rapid progress in the design ofcoronary devices and, to a degree, advances in theequipment for peripheral vascular interventions makes itincreasingly possible to combine this technique in clinicallyand anatomically challenging patients.3,4 This combinedstrategy reduces the need for multiple interventions,repeated hospitalizations, and is also cost-effective. Wereport the case of an elderly woman who presented withacute anterior wall myocardial infarction and recurrentpostinfarct angina with a background history of severeclaudication of the lower limbs. She was managed by directstenting of the left renal and both common iliac arteries inaddition to coronary angioplasty during the same sitting.

Case Report

A 74-year-old woman presented to us with recurrentpostinfarct angina following acute anterior wall myocardialinfarction. She was a known hypertensive and diabetic formore than 10 years. She had history of severe claudicationof both legs (Fontaine class IIb), more on the right side. She

had been on drug therapy for peripheral vascular diseasetill date. She also gave a history of accelerated hypertensionwith an episode of pulmonary edema and was alsodiagnosed to have an early stage of nephropathy.Physical examination showed very feeble right lower limbpulses and feeble left lower limb pulses with a supine bloodpressure in the right upper limb of 200/110 mmHg.Electrocardiography revealed Q waves in leads V

1–V

3 with

T wave inversion in leads V4–V

6. Echocardiogram showed

regional wall motion abnormality in the left anteriordescending coronary artery territory with a left ventricularejection fraction of 45%. After informed consent,angiographic study was performed which revealed 90%complex stenosis of the left anterior descending coronaryartery (Fig. 1a) with minimal disease of the other coronaryarteries and mild LV systolic dysfunction. A peripheralangiographic study revealed 90% stenosis of the rightcommon iliac artery (Fig. 2a), 70% eccentric stenosis ofthe left common iliac artery (Fig. 2a) and 80% eccentricstenosis of the left renal artery (Fig. 3a).

The patient was pretreated with aspirin 325 mg dailyand clopidogrel 75 mg twice a day 48 hours prior to theprocedure and, in anticipation of a large contrast load (inthe presence of borderline nephropathy), the patient washydrated with normal saline, 75 ml/hour for 12 hours. Tominimize the amount of radiation, it was decided to usethe direct stenting technique and all attempts were madeto reduce exposure time.

Coronary artery stenting: A 7 F arterial sheath wasplaced percutaneously in the left femoral artery and the

Brief Report

Correspondence: Dr DK Baruah, Director of Cath. Lab., Apollo Hospitals,Vishakhapatnam

We describe a patient who underwent percutaneous coronary intervention combined with bilateral iliac andleft renal artery angioplasty during the same sitting. Stenting of the coronary and peripheral arteries wasperformed employing the “direct stenting” technique. No complications occurred. The patient was discharged2 days after the intervention and remains asymptomatic, leading a fully active life during 1 year of follow-up. Toour knowledge, unstaged coronary stenting combined with direct stenting of the renal and both common iliacarteries has not been reported previously in India. (Indian Heart J 2003; 55: 71–74)

Key Words: Coronary artery disease, Peripheral vascular disease, Stenting

IHJ-348-02.p65 5/6/2003, 10:04 AM71

72 Baruah et al. Concurrent Coronary, Iliac and Renal Direct Stenting Indian Heart J 2003; 55: 71–74

left coronary artery was engaged using a 6 F Vista BriteTip® left Judkin’s catheter (Cordis). The lesion in the leftanterior descending coronary artery was crossed with a0.014" balance middle weightTM guidewire (Guidant). Afterquantitative coronary analysis, a Bx velocity stent (Cordis)2.5 mm in diameter and 23 mm in length was deployed at14 atm pressure. The final angiogram (Fig. 1b) showed goodangiographic result with TIMI III flow distally. A total of10 000 units of heparin was administered intravenouslyduring the procedure.

Right common iliac artery stenting: We used theipsilateral approach for iliac angioplasty in view of the

ostial location of the lesion and the acute angle of the aorticbifurcation. After cannulating the right femoral vein witha 0.38" guidewire, the right femoral artery was puncturedunder fluoroscopic guidance. This was done because theright femoral artery pulsation was very feeble and difficultto localize. A 7 F sheath was inserted into the right femoralartery and a retrograde angiogram through the sheathdelineated the tight stenosis of the right common iliacartery. A peak pressure gradient of 80 mmHg on pull-backwas recorded across the lesion. The lesion was crossed witha 0.014" balance middle weight coronary guidewire andexchanged for a 0.035" Amplatz super stiffTM (BostonScientific) guidewire. A CorinthianTM stent (Cordis) 17 mm

Fig. 1. Left anterior descending coronary artery stenosis before (a) and after(b) stenting.

Fig. 2. Right and left common iliac artery stenosis before (a) and after (b)stenting.

b

aa

b

IHJ-348-02.p65 5/6/2003, 10:04 AM72

Indian Heart J 2003; 55: 71–74 Baruah et al. Concurrent Coronary, Iliac and Renal Direct Stenting 73

in length and 6 mm in diameter was positioned accordingto the bony landmarks with the help of retrograde injectionthrough the sheath and deployed at 13 atm pressure. Apost-stenting angiogram showed a good result (Fig. 2b).There was no significant gradient across the stentedsegment on pull-back. The activated clotting time at thispoint of time was 250 s and 5000 units of heparin wereadministered.

Left common iliac artery stenting: Selective left iliacangiogram was performed through the left femoral sheathto delineate the lesion in the left common iliac artery.Simultaneous pressure recording revealed a 47 mmHg peakpressure gradient across the stenosis. A 0.035" Radifocus®

guidewire (Terumo Corporation) was used to cross thelesion and exchanged to a 0.035" Amplatz super stiffguidewire. A CorinthianTM stent (Cordis) 17 mm in lengthand 6 mm in diameter was deployed to cover the entirelesion at 13 atm pressure. In view of the bilateral ostialdisease of the iliac arteries, final dilatation of both stentswas carried out by the “kissing balloon” technique. Thistechnique helps in avoiding plaque shift and in properdeployment of both stents. The final injection showed goodangiographic result (Fig. 2b) and no significant pressuregradient was recorded across the stented segments.

Renal artery stenting: The left renal artery was engagedusing a 6 F right coronary guiding catheter (JR 3.5 VistaBrite Tip, Cordis) and the lesion was crossed with 0.014"balance middle weight guidewire. After selective injectionand QCA analysis, a GFX stent (AVE, Medtronic) 3.5 mmin diameter and 18 mm in length was positioned carefullyand deployed at 14 atm pressure. The post-procedureangiogram (Fig. 3b) revealed a good result with rapid distalflow. Usually, a 3.5 mm diameter stent is undersized for anormal renal artery. In this case, QCA analysis revealed thesize of the left renal artery to be 3.6 mm and the left kidneyto be smaller than the right. Thus, we decided to use acoronary stent.

The patient was kept under observation for 48 hoursand discharged without complications. At 1-year follow-up, she is asymptomatic with good effort tolerance and goodlower limb pulses. Her blood pressure is well under controlwith only one antihypertensive drug.

Discussion

Patients with coexistent coronary and peripheral vasculardisease constitute a challenging group in clinical practice,one that requires special therapeutic attention. Coronaryartery bypass surgery in the presence of acute myocardial

infarction and peripheral vascular disease carries asubstantial risk of morbidity and mortality.5,6 On the otherhand, it is well established that in patients who undergovascular surgery, there is a decrease in survival rate ifcoronary artery disease is present.7 In such a clinicalsituation, percutaneous treatment offers a promisingalternative to surgery. There has been rapid growth ininterventional cardiology in the past few decades and theintroduction of stents has improved the long-term outcomeof angioplasty.8,9 With constantly improving stent designs,delivery systems and techniques, multivessel stenting hasbecome routine in clinical practice. Introduction of directstenting has made this procedure more rapid and cost-effective with less exposure to radiation and contrast load.10

Various studies have been carried out to assess the efficacy

Fig. 3. Left renal artery stenosis before (a) and after (b) stenting.

a

b

IHJ-348-02.p65 5/6/2003, 10:04 AM73

74 Baruah et al. Concurrent Coronary, Iliac and Renal Direct Stenting Indian Heart J 2003; 55: 71–74

of direct stenting and it has been found to be equivalent interms of immediate and long-term results to that of stentingwith predilatation.11–13

There are few reports on combined multivessel directstenting involving both the coronary and peripheralvessels.14,15

Our patient was an elderly woman with multiplecoronary risk factors and borderline nephropathy; thiscondition not only carries a higher surgical risk but also ahigher risk of complications during percutaneousintervention. We used the direct stenting technique in thispatient to reduce the contrast load as well as exposure time.The patient tolerated the procedure well without any majorcomplications with a satisfactory 1-year follow-up record.Direct stenting is an evolving technique in interventionalcardiology which, at present, is applied in selected casesduring percutaneous angioplasty. This technique can alsobe useful when multivessel angioplasty is contemplated.

References

1. Hertzer NR, Beven EG, Young JR, O’Hara PJ, Ruschhaupt WF 3rd,Graor RA, et al. Coronary artery disease in peripheral vascularpatients. A classification of 1000 coronary angiograms and resultsof surgical management. Ann Surg 1984; 199: 223–233

2. Joseph G, Baruah DK, Zacharias TU, Krishnaswami S. Combinedpercutaneous coronary and carotid artery stenting. Indian Heart J1997; 49: 425–426

3. Mathur A, Roubin GS, Yadav JS, Iyer SS, Vitek J. Combined coronaryand bilateral carotid stenting: a case report. Cathet Cardiovasc Diagn1997; 40: 202–206

4. Klez RS, Rozek MM, Bouknight D. Bilateral carotid stenting combinedwith three-vessel percutaneous coronary intervention in single sitting.Cathet Cardiovasc Diagn 2001; 52: 100–104

5. Hertzer NR, Young JR, Beven EG, O’Hara PJ, Graor RA, RuschhauptWF, et al. Late results of coronary bypass in patients with peripheralvascular disease. I. Five-year survival according to age and clinicalcardiac status. Cleve Clin Q 1986; 53: 133–143

6. Hertzer NR, Young JR, Beven EG, O’Hara PJ, Graor RA, RuschhauptWF, et al. Late results of coronary bypass in patients with peripheralvascular disease. II. Five-year survival according to sex, hypertension,and diabetes. Cleve Clin J Med 1987; 54: 15–23

7. Crawford ES, Bomberger RA, Glaeser DH, Saleh SA, Russell WL.Aortoiliac occlusive disease: factors influencing survival and functionfollowing reconstructive operation over a twenty-five-year period.Surgery 1981; 90: 1055–1067

8. Sigwart U, Puel J, Mirkovitch V, et al. Intravascular stents to preventocclusion and restenosis after transluminal angioplasty. N Engl J Med1991; 316: 13–17

9. Serruys PW, de Jaegere P, Kiemeneij F, Macaya C, Rutsch W,Heyndrickx G, et al. A comparison of balloon-expandable-stentimplantation with balloon angioplasty in patients with coronaryartery disease. Benestent Study Group. N Engl J Med 1994; 331: 489–495

10. Figulla HR, Mudra H, Reifart N, Werner GS. Direct coronary stentingwithout predilatation: a new therapeutic approach with a specialballoon catheter design. Cathet Cardiovasc Diagn 1998; 43: 245–252

11. Carrie D, Khalife K, Citron B, Izaaz K, Hamon M, Juiliard JM, et al.Comparison of direct coronary stenting with and without balloonpredilatation in patients with stable angina pectoris. BET (BenefitEvaluation of Direct Coronary Stenting) Study Group. Am J Cardiol2001; 87: 693–698

12. Wilson SH, Berger PB, Mathew V, Bell MR, Garratt KN, Rihal CS, et al.Immediate and late outcomes after direct stent implantation withoutballoon predilation. J Am Coll Cardiol 2000; 35: 937–943

13. Veselka J, Mates M, Tesar DB, Aschermann M, Urbanova T, Honek T.Direct stenting without predilatation: a new approach to coronaryintervention. Coron Artery Dis 2000; 11: 503–507

14. Dorros G, Mathiak L. Direct deployment of the iliofemoral balloonexpandable (Palmaz) stent utilizing a small (7.5 French) arterialpuncture. Cathet Cardiovasc Diagn 1993; 28: 80–82

15. Mehan VK, Meier B. Multiple renal artery stenting with half(disarticulated) Palmaz–Schatz coronary stents and simultaneouscoronary angioplasty. Indian Heart J 1995; 47: 259–260

IHJ-348-02.p65 5/6/2003, 10:04 AM74

Indian Heart J 2003; 55: 75–77 Gupta et al. Catheter Ablation Using a New Mapping System 75

Catheter Ablation of Atrial Tachycardia Using a Real-TimePosition Management Mapping System

Anoop K Gupta, Alok Maheshwari, Ranjan K Thakur, Yash Y LokhandwalaThoracic and Cardiovascular Institute, Sparrow Health System, Michigan State University, Lansing, MI, USA

Focal atrial tachycardias are not randomly distributedbut rather tend to cluster in certain anatomic zones,1–3 the

commoner site being along the crista terminalis in the rightatrium and from within a pulmonary vein. There are avariety of approaches to mapping and ablation of focalatrial tachycardias.

The Carto system has been used to develop anelectroanatomic map of the earliest activation; however,this system requires that the map be developed point bypoint.4 Multielectrode basket catheters have also beendeveloped, as have noncontact mapping arrays.5 Anothersystem, Cardiac Pathways, enabling three-dimensional (3-D)real-time mapping, includes real-time position managementin guiding radiofrequency (RF) catheter ablation. We presentour initial experience with the Cardiac Pathways mappingsystem for ablation of focal atrial tachycardia.

Case Report

A 53-year-old woman presented to the arrhythmia clinicwith palpitation. Twenty-four hour Holter monitoringshowed a narrow QRS complex tachycardia with a rate of128 beats/min with warm-up phenomenon. P-wavemorphology during tachycardia was similar to the sinusrhythm, suggestive of a focus in the high right atrium(Fig. 1). Clinical examination, chest X-ray andechocardiography showed no abnormality.

The patient was taken up for electrophysiologic study(EPS) after withdrawing all antiarrhythmic drugs for atleast five half-lives. EPS was performed using the Cardiac

Pathways system. 3-D positional mapping was done withthe ablation catheter using both point and isochronal maps.Identification of major structures and the deformation mapwere performed before mapping the atrium.

Atrial tachycardia was easily inducible with atrialextrastimuli and continuous pacing. Right atrial mappingwas performed; the earliest activation of 36 ms from thesurface P wave was obtained high at the interatrial septum.A trans-septal puncture was performed using the standardtechnique to map the left atrium. The earliest activation of42 ms from the surface P wave was obtained in the roof ofthe left atrium between the right and left superiorpulmonary veins. Isochronal and point mapping of the leftatrium also suggested earliest activation in the roof of theleft atrium (Figs 2a and b).

Successful RF ablation was performed using a 7 F, 4 mmtip bidirectional steerable cooled ablation catheter (CardiacPathways, Sunnyvale, CA, USA), keeping the reference

Brief Report

Correspondence: Dr Anoop K Gupta, Consultant Cardiologist andElectrophysiologist, Krishna Heart Institute, Ghuma, Ahmedabad 380058.e-mail: [email protected]

Catheter ablation for atrial tachycardia is limited by its low success rate and prolonged procedure time becauseof difficulties in mapping the site of the tachycardia. A new three-dimensional mapping system, the CardiacPathways mapping system, using an ultrasound transducer, has recently become available. We report a case offocal atrial tachycardia ablation with this system. (Indian Heart J 2003; 55: 75–77)

Key Words: Ablation, Tachyarrhythmia, Cardiac mapping

Fig. 1. Panel A: 12-lead ECG showing atrial tachycardia (128 beats/min).The P wave morphology in the inferior leads is similar to sinus rhythm ECG.Panel B: 12-lead ECG showing baseline sinus rhythm.

IHJ-364-02.p65 5/6/2003, 10:05 AM75

76 Gupta et al. Catheter Ablation Using a New Mapping System Indian Heart J 2003; 55: 75–77

ghost catheter at the earliest site of activation during theablation. No tachycardia could be induced despite avigorous atrial stimulation protocol with and withoutisoproterenol after radiofrequency ablation (RFA). Theprocedure time was 135 min and fluoroscopic time was 18min. There was no complication during the procedure.Heparin was given at the rate of 100 unit/kg intravenouslyafter the trans-septal puncture.

Cardiac Pathways mapping systemReference and ablation catheters: Two reference cathetersand one mapping/ablation catheter were introduced

percutaneously using the subclavian and/or femoralapproach. One reference catheter was positioned in thecoronary sinus (CS) and the other in the right ventricular(RV) apex. For ablation purposes, a 7 F, 4 mm tipbidirectional steerable cooled ablation catheter (CardiacPathways, Sunnyvale, CA, USA) was used. The referencecatheters (Cardiac Pathways) have a 6 F fixed curve distalshaft. The shaft of the CS reference catheter contains nine1 mm ring electrodes and one 2 mm tip electrode(interelectrode distance 1 mm), whereas the RV referenceand ablation catheters contain three 1 mm ring electrodesand one 4 mm tip electrode (interelectrode distance 1 mm).The reference catheters are equipped with four ultrasoundtransducers; the ablation catheter contains threeultrasound transducers. The ultrasound transmitter andreceiver device sends a continuous cycle of ultrasoundpulses (558.5 kHz) to the transducers of the reference andablation catheters. By measuring the time delay from thedeparture of a transmitted ultrasound pulse and thereception of this pulse at the other transducers, assuminga speed of sound in blood of 1550 m/s, the distance betweenthe individual transducers can be calculated. These dataare subsequently transferred to the computer and used todefine the location of the catheter(s) within the referenceframe. Once the 3-D reference frame is established,triangulation can be used to track the position of additionaltransducers.

Because dimensional and structural characteristics ofthe catheter are known, it is possible to construct a real-time 3-D graphic representation of the catheters, includingthe position of the electrodes and the transducer. As one ofthe transducers is positioned distal to the deflection pointof the shaft of the catheter, it is possible to display the curveof the catheter as well. Furthermore, the real-time positionmanagement system graphically displays the beat-to-beatmovement of the tip of the catheters.

Real-time position management mapping system: The 3-D real-time position management and mapping system (CardiacPathways) uses ultrasound-ranging techniques todetermine the position of a mapping/ablation catheterrelative to two reference catheters. The mapping systemconsists of an acquisition module and an ultrasoundtransmitter and receiver unit, both connected to a SPARC20 computer (Sun Microsystems, Palo Alto, CA, USA). Thesystem is capable of simultaneously processing: (i) sevenposition management catheters; (ii) 24 bipolar/48 unipolarelectrogram signals; (iii) 12-lead ECG; and (iv) two pressuresignals. Signals are sampled at 3 kHz per channel with aresolution of 14 bits. The high-pass filters are set at 30 Hzand low-pass filters at 500 Hz. Electrograms and catheter

Fig. 2. Isochrone mapping of the right and left atrium on the Cardiac Pathwayssystem showing the earliest activation site in the roof of the left atrium.LAO 40o (a) and RAO 30o (b). The right panel of the pictures shows theearliest activation signal in the ablation catheter (68 ms early from thereference catheter). CS: coronary sinus; RV: right ventricle; ABL: ablationcatheter

a

b

IHJ-364-02.p65 5/6/2003, 10:05 AM76

Indian Heart J 2003; 55: 75–77 Gupta et al. Catheter Ablation Using a New Mapping System 77

positions are stored on an optical disk. The original positionof the reference catheters can be displayed on the real-timewindow, thereby allowing repositioning of the cathetersafter displacement.

The frame rate of this real-time imaging is: (i) Cathetercartoon: once per cardiac cycle (ii) Luttle ball of ablationcatheter tip: 15 frames/second.

Discussion

Targeting the site of earliest activation preceding the P waveduring tachycardia best ablates focal atrial tachycardia,regardless of whether it is due to abnormal automaticity,triggering, or micro re-entry. Pace mapping to match thesurface P wave during tachycardia can be usedadjunctively,3 but because of the difficulty in clearlydiscerning surface P wave morphology, activation mappingis more accurate.

A mapping system provides additional help in mappingatrial tachycardia, especially defining the crista terminalis,which is the commonest site of atrial tachycardia. Inaddition, it helps in understanding the breakthrough leftatrial tachycardia by providing the 3-D anatomy of theatrium. Isochrone mapping usually draws the sequence ofactivation within the atrium with the red color showingthe earliest and violet the late activation.

In our case, although the morphology of tachycardiawas suggestive of right atrial origin, successful ablation wasperformed in the left atrium. Left atrial tachycardias usuallyarise from near the pulmonary veins but, for severalreasons, initial localization can at times be confusing. First,activation times during tachycardia in the CS are usuallyrelatively late with respect to many sites in the right atriumbecause the CS is inferior and anterior to the pulmonaryveins, which are inserted in the posterior aspect of the leftatrium. Secondly, rapid conduction from left to right overthe Bachmann’s bundle may cause a wrong diagnosis ofright atrial origin. Finally, the right superior pulmonaryvein enters the left atrium just behind the superior posteriorright atrium.

Ultimately, one must identify a site of earliest activationwhich precedes P wave onset by more than 30 ms forsuccessful ablation.1 In our case, the earliest activation was

42 ms from the surface P wave. It is mandatory to map theother atrium if the earliest activation is less than 30 ms.However, in our case, we did left atrial mapping despite agood signal in the right atrium to rule out any possibilityof breakthrough left atrial tachycardia.

Although the isochrones and point map depend uponthe earliest activation signal from the surface ECG, makingthe activation sequence map does help in localizing thediscrete point. The advantage of the Cardiac Pathwayssystem over the Carto and other 3-D mapping systems isthat it also indicates the catheter position during mapping;therefore, it guides catheter manipulation during mappingand, secondly, availability of the ghost catheter duringablation, which makes a very appropriate site for focalablation and reduces the chances of edema formationaround the site of origin by avoiding burns in thesurrounding area. In addition, the catheters can be re-usedto save the cost of the procedure in developing countriessuch as India. It also significantly reduces the fluoroscopictime as with other electroanatomic mapping systems.

Conclusions: The Cardiac Pathways mapping systemhelps in the precise location of ablation of atrialtachycardia. Our initial experience with this new mappingsystem is good. However, its efficacy needs to be evaluatedin more patients.

References

1. Walsh EP, Saul JP, Hulse JE, Rhodes LA, Hordof AJ, Mayer JE, et al.Transcatheter ablation of ectopic atrial tachycardia in young patientsusing radiofrequency current. Circulation 1992; 86: 1138–1146

2. Kay GN, Chong F, Epstein AE, Dailey SM, Plumb VJ, et al.Radiofrequency ablation for treatment of primary atrial tachycardias.J Am Coll Cardiol 1993; 21: 901–909

3. Tracy CM, Swartz JF, Fletcher RD, Hoops HG, Solomon AJ, KarasikPE, et al. Radiofrequency catheter ablation of ectopic atrialtachycardia using paced activation sequence mapping. J Am CollCardiol 1993; 21: 910–917

4. Kottkamp H, Hindricks G, Breithardt G, Borggrefe M. Three-dimensional electromagnetic catheter technology: electroanatomicalmapping of the right atrium and ablation of ectopic atrial tachycardia.J Cardiovasc Electrophysiol 1997; 8: 1332–1337

5. Eldar M, Fitzpatrick AP, Ohad D, Smith MF, Hsu S, Whayne JG, et al.Percutaneous multielectrode endocardial mapping during ventriculartachycardia in the swine model. Circulation 1996; 94: 1125–1130

IHJ-364-02.p65 5/6/2003, 10:05 AM77

78 Kothari et al. An Infant with “Dying Spells” Indian Heart J 2003; 55: 78–80

An Infant with “Dying Spells”

Shyam S Kothari, Ambuj Roy, Sanjiv Sharma, Anil BhanCardiothoracic Centre, All India Institute of Medical Sciences, New Delhi

Vascular rings and slings in infants usually present asstridor, feeding difficulties, and recurrent aspiration.

Vascular compression of the trachea by the innominateartery is rare. It may present with one of these symptomsor, more dramatically, as reflex apnea. We report such acase to highlight a rare but curable cause for recurrentapneic spells.

Case Report

A 45-day-old infant was brought with a history ofrecurrent episodes of unconsciousness and apnea triggeredusually (but not exclusively) by crying. He was born at fullterm, after an unremarkable pregnancy in a primigravida,and had an Apgar score of 8/10 at birth. There was nohistory of fever, jitteriness, feeding difficulties, seizures,cyanosis, or stridor, and the baby was otherwise healthy.The infant had had several episodes of unconsciousness andapnea since he was 15 days old and was feared dead duringthese spells, but would recover spontaneously within 1–2minutes.

On examination, his weight was 4.5 kg, and heart rate120 per minute, regular. His mental and motor milestoneswere normal. There were no dysmorphic features. He hadsitus inversus, and a pansystolic murmur. The chest X-rayshowed mild cardiac enlargement, situs inversus, andnormal lung fields. Detailed echocardiographic assessmentrevealed situs inversus, corrected transposition of the greatvessels (i.e. atrioventricular and ventriculoarterialdiscordance), a small ventricular septal defect, and mildpulmonic stenosis. The electrocardiogram (ECG)

demonstrated situs inversus but was otherwiseunremarkable. The electroencephalogram (EEG) and CTscan of the head were normal.

A seizure disorder, complete heart block, or cyanoticspells were suspected in view of the history and cardiacdisorder. Breath-holding spells were also considered.Routine blood chemistry was within normal limits. In thehospital, the infant was observed to have episodes of cryingand distress followed by apnea and progressive bradycardia,and even asystole. Two of the episodes required briefassisted ventilation and cardiac massage. Cardiaccatheterization confirmed the cardiac lesions but could notexplain the symptoms. An aortic angiogram revealed anL-posed aorta with a slightly medial origin of theinnominate artery (Fig. 1). In the course of illness, a repeatchest X-ray demonstrated hyperinflated lung fields. In viewof this finding, a chest CT scan was done which revealed aright aortic arch with anterolateral tracheal compressionby the left-sided innominate artery 1 cm above the carina(Fig. 2). Cinefluoroscopy confirmed compression of thetrachea at the level of the innominate artery. The infantunderwent aortopexy and left innominatepexy for the reliefof tracheal compression. The left innominate artery andthe aortic arch were plicated to the second costal cartilage.Subsequently, he has had no recurrence of symptoms andwas doing well at six months’ follow-up.

Discussion

The absence of stridor in our patient delayed the correctinterpretation of reflex apnea due to tracheal compressionby the innominate artery. Innominate artery compressionis a rare cause of airway obstruction in infants. Theanomalous innominate artery compression syndrome wasfirst described by Gross and Neuhauser1 in 1948 as

Brief Report

Correspondence: Dr Shyam S Kothari, Department of Cardiology,Cardiothoracic Centre, All India Institute of Medical Sciences, NewDelhi 110029. e-mail: [email protected]

A 45-day-old infant presented with the unusual and intriguing symptom of episodic crying and loss ofconsciousness. The infant was discovered to have a vascular compression of the trachea by the innominateartery, almost serendipitously. He was cured of his symptoms by anterior suspension of the innominate artery.(Indian Heart J 2003; 55: 78–80)

Key Words: Vascular ring, Apnea, Aortopexy

IHJ-441-03.p65 5/6/2003, 10:05 AM78

Indian Heart J 2003; 55: 78–80 Kothari et al. An Infant with “Dying Spells” 79

consisting of cough, stridor, and occasionally apnea. Otherauthors2–4 have subsequently reported abnormal origin andcourse of the innominate artery as a cause of airwaycompression, with relief of symptoms followingdecompressive surgery. However, some authors5 havedisputed the role of the innominate artery in causingsymptoms in these children, and believe that intrinsicabnormality of the trachea rather than compression by theinnominate artery is responsible for the symptoms. Strifeet al.6 reported that the innominate artery crosses anteriorto the trachea, and produces mild to moderate anteriorindentation of trachea in 30% of normal children withoutcausing symptoms. Mustard et al.7 found that only 39 outof 285 children with innominate artery compression had

symptoms marked enough to warrant surgery. Thus, theinnominate artery probably crosses the trachea in almostall infants, but few children actually have trachealcompression, and even fewer have symptoms requiringintervention. Why only some children have symptoms ofcompression is not known. A tight or crowded mediastinumand esophageal dilatation due to any cause are some of thereasons postulated. The syndrome has been associated withthe presence of congenital heart disease as well asesophageal atresia. The usual symptoms reported due toinnominate artery compression are stridor, respiratorydistress, recurrent lower respiratory tract infections, andreflex apnea. Apnea is thought to be due to reflex respiratoryarrest initiated by compression of the trachea. This mayoccur as a result of a large bolus of food passing throughthe esophagus or from accumulation of secretions in thetracheobronchial tree. The symptoms are maximal duringinfancy with gradual improvement by 2 years of age. Theimprovement is probably due to strengthening of theairways by increased cartilaginous support and cephaladdisplacement of the aorta.

The diagnostic modalities are a lateral X-ray that mayshow an anterior indentation of the tracheal air column.In suspected cases MRI imaging confirms the diagnosis, andis the imaging modality of choice, though a CT scan alsoprovides adequate information, as in our case. Recently, cineMRI has been reported to be useful for dynamic imaging oftracheal compression by the innominate artery.8 Bariumswallow is not helpful because the compression is locatedanteriorly. The diagnosis can be further confirmed bybronchoscopy, which shows anterior tracheal compression1–2 cm proximal to the carina. The area of compression ispulsatile and, when levered anteriorly by the tip of theendoscope, leads to diminution of the corresponding radialpulse.

Surgical correction of compression is usually bysuspension of the innominate artery to the undersurfaceof the sternum by a row of braided Dacron sutures(innominatepexy), though some surgeons prefer itsreimplantation lower down the aortic arch. Surgicaloutcomes have been excellent in judiciously selected casesof innominate artery compression.2

References

1. Gross RE, Neuhauser EBD. Compression of the trachea by ananomalous innominate artery: an operation for its relief. Am J DisChild 1948; 75: 570–574

2. Hawkins JA, Bailey WW, Clark SM. Innominate artery compressionof the trachea. Treatment by reimplantation of the innominateartery. J Thorac Cardiovasc Surg 1992; 103: 678–682

3. Fearon B, Shortreed R. Tracheobronchial compression by congenitalcardiovascular anomalies in children. Ann Otol 1963; 73: 949–969

4. Schuster T, Hecker WC, Ring-Mrozik E, Mantel K, Vogl T. Tracheal

Fig. 2. Contrast-enhanced CT scan of the chest 1 cm above the carina showingtracheal compression by the innominate artery. Note the flattened, trifoliateappearance of the trachea (arrow).

Fig. 1. Aortic angiogram shows L-posed aorta but is otherwise unremarkable.In retrospect, the innominate artery is seen to arise more medially.

IHJ-441-03.p65 5/6/2003, 10:05 AM79

80 Kothari et al. An Infant with “Dying Spells” Indian Heart J 2003; 55: 78–80

stenosis by innominate artery compression in infants: surgicaltreatment in 35 cases. Prog Pediatr Surg 1991; 27: 231–243

5. Fletcher BD, Cohn RC. Tracheal compression and the innominateartery: MR evaluation in infants. Radiology 1989; 170:103–107

6. Strife JL, Baumel AS, Dunbar JS. Tracheal compression by theinnominate artery in infancy and childhood. Radiology 1981; 139:73–75

7. Mustard WT, Bayliss CE, Fearon B, Pelton D, Trusler GA. Trachealcompression by the innominate artery in children. Ann Thorac Surg1969; 8: 312–319

8. Faust RA, Rimell FL, Remley KB. Cine magnetic resonance imagingfor evaluation of focal tracheomalacia: innominate arterycompression syndrome. Int J Pediatr Otorhinolaryngol 2002; 65:27–33

IHJ-441-03.p65 5/6/2003, 10:05 AM80

Indian Heart J 2003; 55: 81–87 Kar et al. LMWH in Coronary Interventions 81

Low-Molecular-Weight Heparins in PercutaneousCoronary Interventions

AK Kar, I DuttaDepartment of Cardiology, Institute of Postgraduate Medical Education and Research, Kolkata

Low-molecular-weight heparins (LMWHs) represent amajor advancement in the management of patients

with acute coronary syndrome (ACS) or as part of thestrategy in candidates undergoing coronary interventions.LMWH are derived from unfractionated heparins (UFH) bychemical or enzymatic depolymerization. They have a meanmolecular weight of 4000–5000, as compared to UFHwhich have a mean molecular weight of 15 000 (5000–30 000). UFH inhibit Factors Xa and IIa in a ratio of 1:1,whereas LMWH inhibit Factors Xa and IIa in ratios between4:1 and 2:1, depending on their molecular weight. LMWHpotentiate antithrombin (AT) III activity, but are relativelyless efficient in binding thrombin, and thus lead to a higherratio of anti-Factor Xa activity to anti-Factor IIa(antithrombin) activity. The anti-Factor Xa activity isparticularly important, as it can prevent thrombingeneration and interrupt feedback amplification ofthrombin production. The thrombin-binding activity ofheparin lies primarily in the higher weight glyco-saminoglycan chains; therefore, the anti-Factor Xa/anti-Factor IIa ratio of various LMWH varies according to themolecular size of the subfractions they contain. The lesserantithrombin effects of LMWH lead to less prolongation ofthe activated partial thromboplastin time (aPTT) than UFH;thus, there is no need for aPTT monitoring duringtherapeutic use of LMWH. LMWH are more resistant toFactor IV-mediated inhibition of heparin activity and donot produce the same degree of platelet activation as UFH.The important pharmacologic characteristics thatdistinguish LMWH from UFH include: greaterbioavailability; minimal plasma protein and vessel wallbinding; more predictable and durable anticoagulantresponse (T½ 2–4 times that of UFH); ease of subcutaneousor intravenous administration; and inhibition of acute-phase release of von Willebrand factor in ACS.

LMWH have earned their place and established theirsuperiority over UFH in the management of post-surgicaldeep vein thrombosis1,2 and ACS, including unstable angina(UA) and non-Q wave myocardial infarction (NQMI).3,4 TheEfficacy and Safety of Subcutaneous Enoxaparin in Non-

q-wave Coronary Events (ESSENCE) study5 showedenoxaparin plus aspirin to be superior to UFH plus aspirinin patients with ACS. The need for urgent revascularizationwas significantly lower in the enoxaparin group. TheHeparin and Aspirin Reperfusion Therapy (HART II) study6

showed that enoxaparin was as effective as UFH as anadjunct to thrombolysis with tissue plasminogen activator(TPA), with a trend towards higher recanalization rates andless reocclusion at 5 or 7 days. Thus, several trials haveshown that the LMWH enoxaparin offers the practical andpotential pharmacologic advantages over UFH in multipleapplications, and logically should also provide a similarbenefit during percutaneous coronary intervention (PCI).

However, during coronary and noncoronaryinterventions, UFH is conventionally used more frequentlythan LMWH. This is largely due to the fact that interventionalcardiologists are sceptical about the efficacy of acuteanticoagulation with LMWH. Furthermore, reversibility ofthe action of heparin is far more prompt and accurate withprotamine in case of UFH than with LMWH. Clinicians arealso concerned about the safety of a combination of LMWHwith glycoprotein (Gp) IIb/IIIa inhibitors, and the safety oftransition to the catheterization laboratory in patients whohave already received LMWH on the floor.

From 1994 to 1998, four clinical trials using LMWHduring or after PTCA were published7–10 (their results aresummarized in Table 1), the most recent being the NationalInvestigators Collaborating on Enoxaparin (NICE) trials 1,3and 4 (Table 2). The use of LMWH eliminates the need forcontinuous IV infusion, anticoagulation monitoring, anddose adjustment associated with UFH.

Data from randomized, controlled clinical trials11,12

support the administration of LMWH and/or platelet GpIIb/IIIa blockade in patients who present with non-ST-elevation ACS. Despite evidence-based support foradministering LMWH and/or Gp IIb/IIIa receptor blockersto patients undergoing PCI and those presenting with ACS,algorithms for integrating these agents into clinical practicehave not been determined. The NICE trials have evaluatedthis issue in detail, and the results of these trials are likelyto have a major impact on the choice of adjunctive therapyduring PCI.13–16

Current Perspective

Correspondence: Professor AK Kar, 82A, Chowringhee Road,Kolkata 700020. e-mail: [email protected]

IHJ-365-02.p65 5/6/2003, 10:05 AM81

82 Kar et al. LMWH in Coronary Interventions Indian Heart J 2003; 55: 81–87

The NICE Story

In the NICE pilot study, 60 patients undergoing PCI wererandomly assigned to receive either UFH 10 000 U as an IVbolus with supplemental doses to achieve a target activatedclotting time (ACT) greater than 300 s or enoxaparin 1 mg/kg IV bolus immediately before PCI. There were nodifferences in procedural outcomes or bleeding eventsobserved between the treatment groups in this smallrandomized trial. Levels of Factor IIa activity weresignificantly higher in the UFH group, though Factor Xaactivity was similar in the two treatment groups.

The pilot trial experience was expanded in the NICE 1study in which 828 patients undergoing PCI withoutplanned use of abciximab were administered enoxaparin1 mg/kg within 15 min of arterial sheath insertion. All thepatients received aspirin in addition, as an antiplateletagent. The incidence of major noncoronary artery bypassgrafting (CABG) related bleeding was 10.5%, and thecomposite clinical end-point [death, myocardial infarction(MI), need for urgent revascularization] at 30 days was7.9%13,14 (Figs 1 and 2).

Table 1. Clinical trials of LMWH after PTCA

Study Year Drug No. of Protocol End-point Relativepatients benefits

ERA7 1994 Enoxaparin 458 R, C, B Clinical and angiographic restenosis –2% (p=ns)REDUCE8 1996 Reviparin 612 R, PC, B 6 months death and MI plus revascularization –4% (p=ns)FACT9 1997 Nadroparin 354 R, PC, B 3 months angiographic restenosis –6% (p=ns)ENTICES10 1998 Enoxaparin 123 R, C 30-day death, MI, revascularization, and stent thrombosis –25%

LMWH: low-molecular-weight heparin; PTCA: percutaneous transluminal coronary angioplasty; R: randomized; B: blinded; C: controlled; PC: placebo-controlled; ns: statistically not significant

Table 2. Overview of National Investigators Collaborating on Enoxaparin (NICE) 4 Study

Inclusion criteria (patients) PCI (excluding planned rotational atherectomy) for acute coronary syndromesStudy design Multicenter, open-labelTreatment regimen Enoxaparin 0.75 mg/kg intravenous bolus, followed by abciximab 0.25 mg/kg intravenous bolus and

12-hour infusion (0.125 µg/kg/min)*Primary end-point Major and minor bleeding post-procedure and need for transfusionSecondary end-points Composite of death, MI, urgent revascularization at 30 days post-PCI, creatine kinase (CK) and CK-MB

elevation at 30 days post-PCIResults (%): first 310 patients Hemorrhage

Major (0)Minor (2.9)Transfusion (0.6)Death/MI/urgent revascularization (2.3)

*No further heparin of any type was administered after the procedure and vascular access sheaths were removed 4 hours after the enoxaparin bolusPCI: percutaneous coronary intervention; MI: myocardial infarction

Fig. 1. Major bleeding incidence with or without events related to CABG andtransfusion requirements up to 30 days post-PCI in the NICE 1 and NICE 4trials.

This large, multicenter experience suggests thatintravenous enoxaparin in a dose of 1 mg/kg providesadequate, safe, and effective anticoagulation during PCI,the antithrombin efficacy (anti-Factor Xa activity) beingsimilar to UFH administered as an IV bolus of 10 000 U. A

IHJ-365-02.p65 5/6/2003, 10:05 AM82


comparison of the NICE 1 trial with the stent plus placebosubgroup (UFH 100 U/kg) of the EPISTENT trial (Figs 3and 4) provides valuable insights, as both trials arecomparable in size, and the definition of major and minorhemorrhage used in both are similar. Major non-CABGbleeding events, ischemic adverse outcomes, death, andneed for urgent revascularization at 30 days were similarin both. Hence, it can be concluded that enoxaparin 1 mg/kg IV during PCI is as safe and effective as the contemporaryexperience utilizing a weight-adjusted heparin (UFH).

The NICE 4 trial: This trial was the first large-scaleexperience with the combination of an LMWH and aplatelet Gp IIb/IIIa antagonist during PCI. Patientsundergoing PCI with a Food and Drug Administration(FDA),USA-approved device (except rotationalatherectomy) were administered enoxaparin 0.75 mg/kgas an IV bolus followed by a standard dose of abciximab(0.25 mg/kg IV bolus followed by 0.125 µg/kg/min to amaximum of 10 µg/kg). The reduced dose of enoxaparinwas chosen to simulate the weight-adjusted reduction inheparin dose used in both the EPISTENT17 and EPILOG18

trials. Approximately 800 patients with an average age of63 years were enrolled in 11 centers; multivessel PCI wasperformed in 52%, and stent deployment in 86%. Vascularsheaths were removed 4 hours following the enoxaparinbolus, and periprocedural ACT was not monitored. Theprimary end-point of NICE 4 was the incidence of major orminor bleeding and need for transfusion, secondaryend-points being (i) clinical (death, MI, urgentrevascularization), and (ii) biochemical changes (CK, CK-MB elevation) in hospital and up to 30 days’ post-PCI.(Table 2).

Remarkably, there was no major non-CABG bleeding inthe first 557 patients and the final analysis revealed ameager 0.2% incidence. The need for transfusion was alsonegligible. Comparing NICE 4 with the EPILOG trial (low-dose weight-adjusted heparin 70 U/kg plus abciximab) andthe EPISTENT trial (stent plus abciximab) cohorts (Fig. 5),it was observed that a combination of low-dose enoxaparin0.75 mg/kg and abciximab was safe and associated with alow incidence of non-CABG related bleeding. Addition ofabciximab to 0.75 mg/kg of enoxaparin bolus was

Fig. 2. Ischemic events up to 30 days post-PCI in the NICE 1 and NICE 4trials.

Fig. 3. Major and minor bleeding with and without CABG and requirement forblood product transfusion in the NICE 1 trial versus the EPISTENT stent plusplacebo cohort; events up to 30 days.

Fig. 4. Comparison of incidence of death and need for urgent revascularizationup to 30 days in the NICE 1 trial versus the EPISTENT stent plus placebo cohort.

IHJ-365-02.p65 5/6/2003, 10:05 AM83


associated with a reduction of periprocedural ischemicevents in the NICE 4 trial compared to the NICE 1 trial.Interestingly, abciximab-associated thrombocytopenia waslower in NICE 4 than in studies using abciximab plus UFHregimens. This might be related to the fact that enoxaparinis less likely than UFH to cause platelet activationor aggregation, or that thrombocytopenia may resultfrom an interaction between abciximab and certainanticoagulants.17

The NICE 1 and NICE 4 trials demonstrated the safetyand efficacy of enoxaparin alone and in combination withabciximab for procedural anticoagulation. The NICE 3study16 investigated the use of enoxaparin with one of thethree Gp IIb/IIIa antagonists (abciximab, tirofiban,eptifibatide) in patients with ACS (including PCI). It aimedto assess the safety of this combination in thecatheterization laboratory with respect to bleeding andpracticality of usage.

A total of 660 patients were enrolled with rest anginaof less than 24 hours’ duration at 46 clinical centers inCanada and the United States. All the patients received IVenoxaparin 1 mg/kg b.d., aspirin 162–325 mg/day, and aGp IIb/IIIa antagonist assigned by the institution at astandard dose. Clinical outcomes in NICE 3 werecomparable to those of prior studies—the combination ofenoxaparin and Gp IIb/IIIa antagonist did not result inexcess major non-CABG bleeding events (Table 3). Inaddition, the study showed that additional UFH was notrequired for the study population treated with enoxaparinand a Gp IIb/IIIa antagonist during coronary intervention,and patients on combination therapy could safely undergoPCI.

Enoxaparin and Ticlopidine after Elective Stenting(ENTICES)Trial10 (Table 4)

Enoxaparin has been shown to be beneficial in patientsrequiring coronary stent implantation, and its use incombination with other antiplatelet regimens in stentinginterventions was evaluated in this trial. One hundredtwenty-three patients scheduled for elective intracoronarystenting were randomized (in a 2:1 ratio) to a combinationof enoxaparin, ticlopidine, and aspirin; or to a conventionalregimen of warfarin, UFH, dextran, dipyridamole, andaspirin. The composite clinical end-point of death, MI, stentthrombosis, CABG, repeat PTCA, and stroke at 30 days wassignificantly reduced in enoxaparin-treated patients thanin the conventionally treated group. The incidence ofvascular complications and major hemorrhage requiringtransfusion was much less in enoxaparin recipients. Thistrial suggested that a modified antithrombotic regimenincorporating both antiplatelet and anticoagulant drugsmay offer improved outcomes compared to anticoagulanttherapy alone.

Intracoronary LMWH

In the POLONIA study,19 100 patients from 4 centers withsingle lesions (single-vessel coronary artery disease) wererandomized to enoxaparin 10 mg administeredintramurally prior to stent implantation, or systemicheparinization with 100 U/kg of UFH. Follow-upangiography at 6 months was performed in all but onepatient. The results showed that local delivery ofenoxaparin resulted in significant reduction in the

Fig. 5. Incidence of death or urgent revascularization (UR) up to 30 days inthe NICE 4 trial versus the EPISTENT (stent plus abciximab) cohort.

Table 3. NICE 3 study findings: enoxaparin and Gp IIb/IIIaantagonist for patients with ACS including PCI (n=616)

In-hospital clinical outcomes (%)

Death 0.3MI 3.4Urgent revascularization 2.1Death/MI/urgent revascularization 5.7Bleeding 27.9

Major 4.5Minor 25

Transfusion 10.5Non-CABG 1.9

Gp: glycoprotein; ACS: acute coronary syndrome; PCI: percutaneouscoronary intervention; MI: myocardial infarction; CABG: coronary arterybypass grafting

IHJ-365-02.p65 5/6/2003, 10:05 AM84


incidence of restenosis and the revascularization ratecompared with systemic heparinization. A few other trialsare also currently evaluating the comparison of localdelivery of enoxaparin before stenting with stenting alone.

Safety Considerations

The most common adverse ef fects reported arehemorrhagic complications. In most cases, these are minorbleeds, e.g. injection site ecchymosis, as would be expectedwith subcutaneous injections. In the ESSENCE study,3 forexample, minor bleeding was significantly more commonin patients receiving subcutaneous enoxaparin 1 mg/kgtwice daily, than those receiving IV UFH (18.4% v. 14.2%at 30 days; p<0.001). There were no significant differencesbetween the 2 treatment groups regarding the incidenceof major hemorrhage. Similarly, when safety results fromother large clinical trials are considered, major bleeding wasreported in <6.5% of patients with CAD treated withenoxaparin 1 mg/kg twice daily or 40 mg once daily.3 Thus,there does not appear to be an increased risk of major

bleeding with enoxaparin therapy in patients with ACS.The other important concern is regarding reversibility

of anticoagulant action. Unlike heparin, the action ofLMWH can only be partially reversed with protamine(50%). The use of fresh frozen plasma to replenishcoagulation factors has been advocated for LMWH.

Timing of PCI

The validated, weight-adjusted dose of enoxaparin to treatunstable angina patients also seems to be suitable for PCIwithin 8 hours of the last injection. In the study of“Percutaneous coronary intervention after subcutaneousenoxaparin pre-treatment in patients with unstable anginapectoris”,20 carried out in the department of cardiology andthe haemostasis laboratory of the University Hospital, Paris,France a total of 451 consecutive patients with UA/NQMIwere treated for at least 48 hours with subcutaneousenoxaparin (1 mg/kg every 12 hours), cycled at 6 a.m. and6 p.m. Two hundred ninety-three patients underwent acoronary angiography within 8 hours of the morning

Table 4. Overview of Enoxaparin and Ticlopidine after Elective Stenting (ENTICES) study

Inclusion criteria of patients Multicenter, randomized

Treatment regimen Control EnoxaparinPre-procedure UFH/warfarin, aspirin 325 mg/day, Aspirin 325 mg/day,

dipyridamole 75 mg 3 times daily, ticlopidine 250 mg twicedextran 100 mg/hour daily, dipyridamole

Procedure Dextran and UFH to maintain ACT UFH to maintain ACTat 300–350 s of 300–350 s

Post-procedure UFH 2500 U IV bolus 6 hours after Enoxaparin 30–60 mgsheath removal, then infusion until subcutaneously twiceprothrombin time 50–70 s, dextran daily for 10 days,until 2 hours after restarting UFH. *ticlopidine 250 mg twiceAspirin, dipyridamole, and warfarin daily for 1 month(INR 2.0–3.0) for 30 days

Clinical end-points Death, MI, stent thrombosis, bypass surgery (CABG), and repeat PTCA at 30 days, individuallyand the composite of all clinical events, in-hospital hemorrhagic complications

Results (%) Control (n=44) Enoxaparin (n=79) p value

Death 5 0 0.13MI 11 4 0.13Stent thrombosis 7 0 0.04Bypass surgery 9 0 0.02Repeat PTCA 7 1 0.13Composite 20 5 0.01Hemorrhagic/vascular complications 16 5 0.05

* The enoxaparin dose was 30 mg, 40 mg or 60 mg in patients weighing <70 kg, 70–100 kg and >100 kg, respectively.UFH: unfractionated heparin; IV: intravenous; ACT: activated clotting time; INR: international normalized ratio; PTCA: percutaneous transluminal coronaryangioplasty

IHJ-365-02.p65 5/6/2003, 10:05 AM85


LMWH injection, followed by immediate PCI in 132patients. PCI was performed without monitoring ofcoagulation and without any additional bolus of UF/LMWH heparin. There were no in-hospital abrupt closuresor urgent revascularization needed in the PCI group. Theincidence of death/MI rates in the PCI group was muchlower than in those not undergoing catheterization. The30-day major bleeding rate in the PCI group was also lessthan in those not undergoing catheterization. Sheathremoval was done with manual or pneumatic compressionmore than 10 hours after the morning injection ofenoxaparin. Importantly, enoxaparin was not restartedafter PCI and the patients left the hospital the next morning.

This study convincingly shows that PCI can beperformed safely within 8 hours in patients treated withLMWH, particularly enoxaparin, without additionalanticoagulation. Whether the time window can beextended to 12 hours is a matter of debate and is beingcurrently evaluated in the PEPCI trial (PharmacokineticStudy of Enoxaparin in Percutaneous CoronaryInterventions). Presumably, an additional bolus ofanticoagulant may be beneficial in procedures performedbetween 8 and 12 hours of the last dose of the LMWH.

Role of Monitoring

The role of coagulation monitoring with the use of LMWHhas also been evaluated in the study from France mentionedabove.20 Anti-Factor Xa activity was measured fromsamples collected immediately before catheterization in allpatients undergoing PCI. However, anti-Factor Xameasurements, obtained retrospectively, were not used fordecision-making during PCI. During the period of medicalstabilization before catheterization, in elderly patients andthose with renal failure, anti-Factor Xa activity wasmeasured 4 hours after the third injection, and the dose ofenoxaparin adjusted, aiming at a target value of 0.5–1.0 Uof anti-Factor Xa. The anti-Factor Xa activity at the timeof catheterization was >0.5 IU/ml in 97.6% of patients,and it was stable over the 8-hour period after the injectionof enoxaparin. The aPTT was also measured prior tocatheterization and there was a weak but significantcorrelation with anti-Factor Xa activity.

From the above study, we can conclude that anti-FactorXa activity need not be measured routinely before PCI donewithin 8 hours of the last dose of enoxaparin. However, inelderly patients and those with renal failure receiving areduced dose of enoxaparin, anti-Factor Xa measurementsmay be beneficial.

Pilot studies are on for monitoring of LMWH in the

setting of UA. A panel of assays to measure the tissue factorclotting time (TiFaCT) is being developed. The tests providea functional measure of tissue factor, including itsinteraction with platelets and plasma components. Theresults can be obtained in 15 min–2 hours.

The issue of sheath removal following PCI has beenaddressed in the NICE 4 trial, in which the vascular sheathwas removed 4 hours after the enoxaparin bolus dosewithout any increased incidence of puncture sitecomplications. In the study reported from Francementioned above,20 the vascular sheath was pulled out >10hours after the last enoxaparin injection.

Conclusions

In present day practice, when interventional cardiology ismaking rapid progress every day, it is worthwhileconsidering safer and easier adjunctive medications duringthe procedure. While UFH has been a trusted companionof interventionists for years, LMWH can offer an easier andassuring alternative of anticoagulation withoutcompromising ef ficacy. The role of LMWH as anintracoronary injection for the prevention of stentcomplications is under evaluation and, if established, wouldcertainly be a powerful weapon for the prevention of stentthrombosis. It is debatable whether the benefitsdemonstrated with enoxaparin can be extended to otherLMWH. The US FDA classifies each LMWH as a distinct drugthat cannot be interchanged with another. In the ESSENCEand TIMI-IIB studies, a combination of enoxaparin plusaspirin was more effective than that of UFH and aspirin.Inversely, in the FRIC study, dalteparin plus aspirin did notdemonstrate such superiority. Evidence is alsoaccumulating regarding the role of enoxaparin as anantithrombotic in the treatment of acute MI. Selection ofan individual LMWH for a particular indication shouldreflect the level of evidence for a the agent in that condition.Thus, at present, evidence-based considerations favor theuse of enoxaparin in the antithrombotic regimen ofpatients undergoing PCI.

References

1. Leizorovicz A. Comparison of the efficacy and safety of low molecularweight heparins and unfractionated heparin in the initial treatmentof deep venous thrombosis. An updated meta-analysis. Drugs 1996;52 (Suppl): 30–37

2. Dunn CJ, Goa KL. Enoxaparin. A pharmacoeconomic appraisal of itsuse in thromboembolic prophylaxis after total hip arthroplasty.Pharmacoeconomics 1996; 10: 179–190

3. Cohen M, Demers C, Gurfinkel EP, Turpie AG, Fromell GJ, Goodman S,et al. A comparison of low-molecular-weight heparin with

IHJ-365-02.p65 5/6/2003, 10:05 AM86


unfractionated heparin for unstable coronary artery disease. Efficacyand Safety of Subcutaneous Enoxaparin in Non-Q-Wave CoronaryEvents Study Group. N Engl J Med 1997; 337: 447–452

4. Antman EM, McCabe CH, Gurfinkel EP, Turpie AG, Bernink PJ, SaleinD, et al. Enoxaparin prevents death and cardiac ischemic events inunstable angina/non-Q-wave myocardial infarction. Results of thethrombolysis in myocardial infarction (TIMI) IIB trial. Circulation1999; 100: 1593–1601

5. Deutsch E, Cohen M, Radley DR. Safety and efficacy of percutaneousprocedures in patients receiving subcutaneous enoxaparin forunstable angina: results of ESSENCE trial [Abstr]. Circulation 1998;98 (Suppl): 1563

6. Ross AM, Malhoek P, Lundergen C, Knudtson M, Draoui Y, RegaladoL, et al. Randomized comparison of enoxaparin, a low-molecularweight heparin with unfractionated heparin, adjunctive torecombinant tissue plasminogen activator thrombolysis and aspirin.Second trial of Heparin and Aspirin Thrombolysis. Circulation 2001;104: 648–652

7. Faxon DP, Spiro TE, Minor S, Cote G, Douglas J, Gottlieb R, et al. Lowmolecular weight heparin in prevention of restenosis after angioplasty.Results of Enoxaparin Restenosis (ERA) Trial. Circulation 1994; 90:908–914

8. Karsch KR, Preisak MB, Bonan R, on behalf of the REDUCE StudyGroup. Low molecular weight heparin, reviparin, in prevention ofrestenosis after PTCA: results of the REDUCE trial. J Am Coll Cardiol1996; 27: 113A

9. Lablanche JM, McFadden EP, Meneveau N, Lusson JR, Bertrand B,Metzger JP, et al. Effect of nadroparin, a low-molecular-weightheparin, on clinical and angiographic restenosis after coronaryballoon angioplasty: the FACT Study. Fraxiparine AngioplastieCoronaire Transluminale. Circulation 1997; 96: 3396–3402

10. Zidar JP. Enoxaparin and ticlopidine after elective stenting, theENTICES trial. Am J Cardiol 1998; 82: 29L–32L

11. Deutsch E. The emerging role of low-molecular-weight heparin andantiplatelet therapies in the cardiac catheterization laboratory. AmHeart J 1999; 138: S577–S585

12. Kereiakes DJ, Grines C, Fry E, Esente A, Barr L, Matthai W, et al.Combination enoxaparin and abciximab during percutaneouscoronary intervention: a new standard of care? Curr Interv Rep 2000;2: 157–164

13. Kereiakes DJ, Grines C, Fry E, Barr L, Matthai W, Broderick M, et al.Abciximab–enoxaparin interaction during percutaneous coronaryintervention. Results of NICE-1 and 4 trials [Abstr]. J Am Coll Cardiol2000; 35: 92A

14. Young JJ, Kereiakes DJ, Grines CL. Low-molecular-weight heparintherapy in percutaneous coronary intervention: the NICE 1 and NICE4 trials. National Investigators Collaborating on EnoxaparinInvestigators. J Invasive Cardiol 2000; 12 (Suppl): E14–E28

15. Kereiakes DJ, Fry E, Matthai W, Niederman A, Barr L, Brodie B, et al.Combination enoxaparin and abciximab therapy duringpercutaneous coronary intervention: “NICE guys finish first”. JInvasive Cardiol 2000; 12 (Suppl): 1A–5A

16. Fergusson JJ, Elliott MA, Bates ER, Cohen M, Every N, Harrington RA,et al. The use of enoxaparin and IIb/IIIa antagonists in acute coronarysyndromes including PCI. Final results of the NICE-3 study. [Abstr]. JAm Coll Cardiol 2001; 37: 648A

17. The EPISTENT investigators. Randomised placebo-controlled andballoon-angioplasty-controlled trial to assess safety of coronarystenting with use of platelet glycoprotein IIb/IIIa blockade. Lancet1998; 352: 87–92

18. The EPILOG investigators. Platelet glycoprotein IIb/IIIa receptorblockade and low-dose heparin during percutaneous coronaryrevascularization. N Engl J Med 1997; 336 : 1689–1696

19. Kiesz RS, Buszman P, Rozek MM,Martin JL, Deutsch E, Gaszewska E,et al. Local delivery of enoxaparin decreases restenosis rate aftercoronary stenting (POLONIA Study) [Abstr]. J Am Coll Cardiol 2000;35 (Suppl): 9A

20. Collet JP, Montalescot G, Lison L, Choussat R, Ankir A, Drobinski G,et al. Percutaneous coronary intervention after subcutaneousenoxaparin pre-treatment in patients with unstable angina pectoris.[Abstr]. Circulation 1999; 100: 1–188

IHJ-365-02.p65 5/6/2003, 10:05 AM87

88 Shrivastava et al. Echocardiographic Anatomy of Atrial Septal Defect Indian Heart J 2003; 55: 88–89

Echocardiographic Anatomy of Atrial Septal Defect:“Nomenclature of the Rims”

Savitri Shrivastava, S RadhakrishnanDepartment of Congenital and Pediatric Heart Disease, Escorts Heart Institute and Research Centre, New Delhi

The accuracy of diagnosis of atrial septal defect (ASD)by two-dimensional (2-D) echocardiography is well

established, and has been utilized in clinical practice foralmost two decades. However, following the easy availabilityof various devices for the closure of ASD, a detailedanatomy of the ASD and the rims around the defect hasassumed great importance. Two-dimensional echocardio-

graphy, both transthoracic (TTE) and transesophageal(TEE), can clearly show the anatomy and is used forselecting cases for device closure of the ASD.1–3 During theprocedure of device closure, TEE serves as the mostimportant investigation for proper placement of the device.However, there is some confusion regarding thenomenclature of the various rims of ASDs, e.g. the rim inrelation to the aorta has been variously called anterior,posterior, and inferior. We propose that for clarity,uniformity and mutual communication the rims should bedesignated according to the structure they are related to,e.g. superior vena caval (SVC), inferior vena caval (IVC),atrioventricular (AV) septal (rim at the crux), aortic andatrial rims (rim of the superior wall of the atrium near theright upper pulmonary vein). The gross anatomic details

Point of View

Correspondence: Dr Savitri Shrivastava, Department of Congenital andPediatric Heart Disease, Escorts Heart Institute and Research Centre, OkhlaRoad, New Delhi 110 025. e-mail: [email protected]

Fig. 1. Gross anatomic details of various atrial septal defects.

Fig. 2. Superior vena caval (SVC) and inferior vena caval (IVC) rims on (a)transthoracic echocardiography in the subcostal sagittal view, and (b) ontransesophageal echocardiography in the basal long-axis view.

b

a b

Fig. 3. Atrioventricular (AV) and atrial rims on (a) transthoracic echocardio-graphy in the subcostal coronal view, and (b) on transesophageal echocardio-graphy in the 4-chamber view.

Fig. 4. Aortic rim seen on (a) transthoracic echocardiography in the parasternalshort-axis view, and (b) on transesophageal echocardiography in the basal short-axis view.

a b

a

IHJ-366-02.p65 5/6/2003, 10:06 AM88

Indian Heart J 2003; 55: 88–89 Shrivastava et al. Echocardiographic Anatomy of Atrial Septal Defect 89

of an ASD with all the above-mentioned rims are depictedin Fig. 1.

The SVC and IVC rims are best visualized on TTE in thesubcostal sagittal view (Fig. 2a), and on TEE in the basallong-axis view (Fig. 2b). The AV septal and atrial rims arebest visualized on TTE in the subcostal coronal view (Fig.3a), and on TEE in the 4-chamber view (Fig. 3b). The aorticrim is best seen on TTE in the parasternal short-axis view(Fig. 4a) and on TEE in the basal short-axis view (Fig. 4b).The rim between the IVC and SVC, located to the right andposterior aspect of the fossa ovalis, is not important indetermining the success of transcatheter closure, andhence has not been considered.

References

1. Reddy SCB, Rao PS, Evenko J, Koscik R, Wilson AD. Echocardio-graphic predictors of success of catheter closure of atrial septal defectwith the buttoned device. Am Heart J 1995; 129: 76–86

2. Rosenfeld HM, Velde MEV, Sanders SP, Colan SD, Parness IA, LockJE, et al. Echocardiographic predictors of candidacy for successfultranscatheter atrial septal defect closure. Cathet Cardiovasc Diagn1995; 34: 29–34

3. Magni G, Hijazi ZM, Pandian NG, Delabays A, Sugeng L, Laskari C,et al. Two-dimensional and three-dimensional transesophagealechocardiography in patient selection and assessment of atrial septaldefect closure by the new DAS-Angel Wings Device: initial clinicalexperience. Circulation 1997; 96: 1722–1728

IHJ-366-02.p65 5/6/2003, 10:06 AM89

90 Anandaraja et al. Migration of Proximally Detached Pacemaker Lead Indian Heart J 2003; 55: 90–91

Unusual Migration of Proximally Detached PacemakerLead into the Coronary Sinus

S Anandaraja, N Naik, R Yadav, R Juneja, KK TalwarDepartment of Cardiology, All India Institute of Medical Sciences, New Delhi

A 72-year-old man had been implanted in 1976 with aunipolar VVI pulse generator in the right pectoral

pocket for symptomatic complete heart block. The patientsubsequently received a new pulse generator and pacinglead (both bipolar) in 1991 when the previous generatorhad exhausted its battery. This time, the pacemaker wasplaced in the left pectoral pocket. The previous generatorwas removed although the pacemaker lead was capped andretained. In 1996, the patient required a new pulsegenerator, which was connected to the bipolar pacing lead.In 1998, there was a fracture of this bipolar pacing leadresulting in pacing failure. Attempts to remove this lead bymanual traction were unsuccessful, and it was severed atits entrance into the venous system and sutured to thepacemaker pocket. Hence, all the pacemaker leadsimplanted so far had been retained.

In December 2001, the unipolar lead retained in theright pectoral pocket spontaneously extruded through theskin. A local physician tried to extract this lead by traction.

This, however, resulted in fracture of the lead, and a partof its distal end was retained, anchored to the rightventricular wall. Six months later, the patient hadpacemaker extrusion from his left pectoral pocket after localtrauma to the pacemaker pocket. The pulse generator wasremoved and the infected lead detached, proximally cutnear its entrance into the venous system and retained inthe pacemaker pocket. After a week of antibiotic therapythe patient was taken up for pacemaker implantation fromthe right pectoral side. However, the superior vena cava(SVC) was blocked and collateral channels connected theinnominate vein to the SVC. It was noticed that the bipolarlead that had been detached and retained in the pocket aweek back had slipped into the SVC while the previouslyimplanted permanent pacemaker lead had also gotdetached from the pocket and was lying freely in the rightatrial cavity. A left subclavian angiogram (by a catheterinserted from the femoral vein) demonstrated a collateralchannel draining into the SVC–right atrial junction.Pacemaker lead implantation was attempted from this siteat a second sitting, which was successful and without anyuntoward effect. Fluoroscopy repeated during this period

Cardiovascular Images

Correspondence: Dr KK Talwar, Department of Cardiology, All IndiaInstitute of Medical Sciences, New Delhi 110029.

Fig. 1. Right anterior oblique view (a) and left anterior oblique view (b) showing migrated pacemaker lead in the coronary sinus (arrow). Also visible is a proximallysevered unipolar lead (white arrowhead), and another detached pacemaker lead lying freely within the right atrium (bold arrow). The white arrow indicates the newpacemaker lead.

a b

IHJ-Images.p65 5/6/2003, 10:07 AM90

Indian Heart J 2003; 55: 90–91 Anandaraja et al. Migration of Proximally Detached Pacemaker Lead 91

on different days showed a marked migration of thesedetached pacemaker leads. Sites of migration included theinferior vena cava (IVC) and the coronary sinus (Fig. 1aand b). The patient was advised surgical extraction of thepacemaker leads, which he refused.

Proximal detachment of the pacemaker lead from itspectoral pocket is a rare complication that produces leadmigration. This usually occurs if there is concomitantinfection or if the leads are cut short and allowed toretract.1,2 Although such leads are usually secure, it is notunusual for sutures to cut through in the presence of localinfection. This allows the lead to slip into the vascular cavityfrom where it is prone to migration. Such leads usuallymigrate into the IVC or the hepatic veins. This preferentiallocalization of pacing leads to the lower half of the body isprobably related to their higher specific gravity, due to theirmetallic content. However, body position also plays a rolein lead migration and these leads can migrate to unusualsites. Lead migration into the main and peripheralpulmonary arteries is reported in the literature.1,2,3 In onereport, the free end of the lead perforated the atrial septumand caused cerebral embolism.4

In our patient, the lead migrated to various sites andradiological examination showed the leads to lie coiled inthe right atrial cavity, at the SVC–right atrial junction,within the IVC and hepatic vein, at different points in time.Migration was also noted deep within the coronary sinusat one juncture. To the best of our knowledge, migrationinto the coronary sinus has not been described previously.Such extensive migration of the lead is likely to producecomplications. There have even been case reports of suddendeath when patients with proximally detached pacing leadswere treated only medically. Although autopsy findings arenot available in these reports, it can be presumed that deathwas related to lead migration. Leads can conceivablyperforate thin-walled structures, such as the right atriumand coronary sinus, thereby producing lethalhemopericardium.

Our patient had other lead-related complications. Theseincluded spontaneous extrusion of one of the pacemaker

leads 26 years after its implantation. The patient alsodeveloped thrombosis of the SVC and, as collateral channelswere adequate, he did not have any evidence of the SVCsyndrome. The retained leads also interfered withappropriate sensing due to their electrical activity, and thepacemaker was programmed to VOO mode to ensureregular pacing.

Extraction of functionless pacemaker leads can beaccomplished by transvenous techniques or surgically. Asuccess rate of 94%–98% has been reported withtransvenous techniques.5,6 Recently, the North AmericanSociety for Pacing and Electrophysiology (NASPE) has comeup with guidelines for the extraction of functionlesstransvenous pacing and defibrillator leads.7 According tothe recommendations, a retained lead that poses animmediate/imminent threat to the patient is a class Iindication for lead extraction. As lead extraction devicesare not available in our country, the patient was advisedsurgical extraction of these leads, which he refused.

References

1. Myers MR, Parsonnet V, Bernstein AD. Extraction of implantedtransvenous pacing leads: a review of a persistent clinical problem.Am Heart J 1991; 121: 881–887

2. Rettig G, Doenecke P, Sen S, Volkmer I, Bette L. Complications withretained transvenous pacemaker electrodes. Am Heart J 1979; 98:587–594

3. Dalvi BV, Rajani RM, Lokhandwala YY, Sathe SV, Kulkarni HL, KalePA. Unusual case of pacemaker lead migration. Cathet Cardiovasc Diagn1990; 21: 95–96

4. Bohm A, Banyai F, Komaromy K, Pinter A, Preda I. Cerebral embolismdue to a retained pacemaker lead: a case report. Pacing ClinElectrophysiol 1998; 21: 629–630

5. Kantharia BK, Kutalek SP. Extraction of pacemaker and implantablecardioverter defibrillator leads. Curr Opin Cardiol 1999; 14: 44–51

6. Wilkoff BL, Bryd CL, Love CJ, Hayes DL, Sellers TD, Schaerf R, et al.Pacemaker lead extraction with the laser sheath: results of the PacingLead Extraction with the Excimer Sheath (PLEXES) trial. J Am CollCardiol 1999; 33: 1671–1676

7. Love CJ, Wilkoff BL, Bryd CL, Belott PH, Brinker JA, Fearnot NE, et al.Recommendation for extraction of chronically implanted transvenouspacing and defibrillator leads: indications, facilities, training. NorthAmerican Society of Pacing and Electrophysiology lead extractionconference faculty. Pacing Clin Electrophysiol 2000; 23: 544–551

IHJ-Images.p65 5/6/2003, 10:07 AM91

92 Letters to the Editor Indian Heart J 2003; 55: 92–96

I read with great interest the article “Tissue DopplerEchocardiography: Principles and Applications” by

Sengupta et al.1 I am concerned about the increasingnumber of papers on tissue Doppler

echocardiography

(TDE) appearing in leading medical journals. Tissue Dopplerechocardiography is a clear case of misuse of technology.It

could turn out to be the greatest hoax in cardiology.

Doppler is best suited to study the direction of free-movingobjects. In the case of echocardiography it is mainly usedto study free-moving blood cells (blood flow). It is thedirection of motion that is important; velocitymeasurements are secondary. A comparison of flowDoppler [color flow mapping (CFM)] and TDE will makethings clear.

Doppler studies are primarily for the study of thedirection of motion. The “red shift” in astronomy, forexample, shows the stars moving away from us, indicatingan expanding universe. Calculation of velocities issupplementary. The key point to note is that Doppler studiesrequire

a priori knowledge of the direction of motion. In

blood flow there is a definite direction of flow and Dopplercan detect any aberration in this. For example, we have priorknowledge of the direction of blood flow in the circuit—pulmonary veins→left atrium→left ventricle→aorta.Suitably directed Doppler interrogation can study anddetect any deviation in this circuit. Myocardial motion iscomplex and not amenable to Doppler studies. In cardiacmotion there are translational, rotational, anddeformational movements. Besides, many tissues near theheart move—due to transmitted cardiac motion, vesselpulsation, respiratory motion, and involuntary musclemovements.2 Doppler interrogation at one point willrepresent the resultant of all these movements. At aparticular point we can never predict the resultant vector.Even if known, the resultant is accurately recorded only ifit is in the line of the Doppler beam. This is due to theinherent problem of directional bias. It is like measuringthe length of a twisted rod with a straight ruler. Cardiacmotion becomes more complicated in the presence of wallmotion abnormalities. Blood flow is simple and suitable forDoppler study. Here the projectile motion of free-movingblood cells in one direction at an instant is studied

In f low Doppler there are definite “points ofinterrogation”, which are the normal and abnormalorifices. In TDE there are no such definite points. In flow

Tissue Doppler Echocardiography: ANeed for Review

Doppler there is a unique unidirectional flow in one part ofthe cardiac cycle at the current interrogation points. Forexample, in mitral valve Doppler interrogation, the uniquedirectional signal is obtained only in diastole. If there is asignal in systole it becomes abnormal. In TDE the to-and-fro motion (systole and diastole) of a tetheredinterconnected syncytium of myocytes is imaged. Suchinformation is useless. This can even otherwise be seen andanalyzed by B-mode imaging. While in flow Doppler highervelocities are studied, TDE is used to study lower velocities,i.e. to study hypofunctioning myocardial segments. Lowervelocities are difficult to appreciate. Higher velocities areeasier to appreciate with the help of aliasing and variance.Thus, hypofunction is difficult to analyze. The derivationsfrom flow Doppler allow us to get orifice size, amount offlow, and pressure gradients, which are clinically of greatimportance.

Thus, Doppler studies are best suited to study

flow.Color flow mapping allows us to “see” what we cannot

see with ultrasonic “eyes”, hence it is of great value. In TDEwe see more or less what we already see by B-mode, henceits value is marginal. In CFM the anatomic landmarks areintact as the color is superimposed on the B-mode image.In TDE the B-mode is eliminated and the entire “picture” isDoppler information resulting in a fuzzy image, anddifferent anatomic regions are difficult to determine. TissueDoppler has been a disappointing imaging modality inclinical practice. This is due to the basic flaw in theapplication of the principles of Doppler. Doppler is bestsuited for flow studies and applying it to tissue motion isunreasonable. In f low Doppler the tissue “noise” issuppressed and

flow is displayed.2 In TDE it is the other way

round. It is like suppressing the PQRST waves in anelectrocardiogram and displaying the “noise”. Besides, TDEis ultrasensitive and so the information gathered is almostuseless (too much false-positive information). In fact,excellent cardiac waveforms can be obtained by placing thesample volume even outside the cardiac region! Once thefoundation of a modality is wrong, all derivations tend tobe wrong.

It is time to look at TDE more realistically. As a new

modality of imaging it appears exciting. However, its realclinical utility is doubtful. TDE does not give any additionalinformation over conventional modalities. Due to the above-mentioned deficiencies it could even give misleadinginformation. Making diagnostic decisions based on thisfaulty application of technology should be unacceptableto the scientific cardiologist.

Letters to the Editor

Letter-to-Editors.p65 5/6/2003, 10:07 AM92

Indian Heart J 2003; 55: 92–96 Letters to the Editor 93

References

1. Sengupta PP, Mohan JC, Pandian NG. Tissue Dopplerechocardiography: principles and applications. Indian Heart J 2002;54: 368–378

2. Maslak SH, Freund JG. Color Doppler instrumentation. In: Lanzer P(ed). Vascular imaging by color Doppler and magnetic resonance. Berlin,Heidelberg: Springer-Verlag; 1991. pp.87–123

George ThomasDepartment of Cardiology

Indira Gandhi Co-operative Hospital, Kochi

Reply

We read with great interest the views and concernsexpressed in the letter regarding the use of tissue

Doppler imaging (TDI). While we greatly admire theauthor’s intelligent analysis of Doppler and its principles,we disagree on the technical contents cited in the letter thatargue against the clinical use of TDI. Ever since the start ofbasic unidimensional echocardiography, it was known thatcardiac motion was a complex combination of rotation,translation, and deformation in a three-dimensional (3-D)format, and its resultant vector was difficult to estimate.However, measures of cardiac motion and function on M-mode or two-dimensional (2-D) echocardiography werestandardized and became invaluable tools for providingdiagnostic and prognostic information. It was not theabsolute 3-D cardiac motion or displacement that wasquantified by these techniques but how these parameterschanged predictably with disease. The use of Doppler withregard to tissue motion or strain, like earlier techniques,till date has centered not on quantifying the absolutedisplacement or true vector of cardiac motion but how analtered velocity profile predicts

the presence of disease or

provides reliable prognostic information.1,2

The limitation of Doppler in not providing the truevelocity vector holds true for both blood flow and tissuemotion. Blood flow in the heart and blood vessels has acomplex profile. Red blood cells have a steady flowcomponent and a series of oscillatory flow components. Themean flow has a parabolic profile while the pulsatingcomponent causes an oscillatory profile, which variesbetween parabolic and blunt. At curves the velocity profiles

are disturbed giving rise to secondary flows, which persistfor a finite distance.3 Doppler, when used for flow imaging,does not show all these components. The color data andspectral velocities merely represent a part of the truevelocity vector in the direction of the transducer. Thus, inthis regard, application of Doppler for blood flow has thesame limitation as its use in tissue motion. On the contrary,TDI has a superior signal-to-noise ratio than thecorresponding blood flow technique, thereby requiringfewer echoes for the estimation of mean velocities at eachpixel location.4

Studies with sonomicrometry (which uses the principleof speed of sound through the cardiac tissue) have alreadybeen carried out for providing veracity of data, and this is adirect proof that information obtained by tissue Doppler iscorrect and scientifically valid.5 The reasons for sustainedclinical and research interest in recent years resulted fromthe unique ability of TDI to provide rapid, incrementalinformation not available currently by any other technique.Some of the areas where the use of TDI has been a clearadvantage and which are incontrovertible are listed below.

1. A-mode, M-mode and real-time B-mode detect themotion of tissue boundaries, which is simply displayedover the screen. TDI detects and measures the motionof weaker and more complex echoes from within thelayers of the myocardial wall.1,6 Velocity gradients areknown to exist in different layers of the myocardium,which are altered in diseased states, and can be detectedby TDI.1

2. Use of TDI and strain-rate imaging have already beenshown to be superior to conventional real-time B-modeimaging for the interpretation of regional wall motionabnormalities.6,7 Distinguishing hypokinesis andakinesis or dyskinesis by conventional echocardiographyhas wide interobserver variability. Routine use of TDIand strain-rate imaging makes this job simple andquantitative.2,7

3. Longitudinal motion of the mitral valve annulus by M-mode has been used for assessing systolic and diastolicLV function. This information is now readily availableby pulsed Doppler TDI and strain-rate imaging.1,2

4. Very high temporal resolution (<4 ms) have providedan insight into a number of physiologic and pathologicshort-lived events during the cardiac cycle, particularlyin relation to the understanding of myocardial ischemiaand its detection. Assessment of segmental viability bypost-systolic thickening or strain is now an invaluabletool in clinical practice.2,8

Similarly, estimation ofregional and global isovolumic phases, and ejection and



filling periods is possible within a single cardiac cyclefrom a single point of sample volume, even without theuse of an ECG and phonocardiogram. This is the onlytechnique that provides temporal data of both phasesof the cardiac cycle simultaneously.

5. Tissue velocity imaging is an invaluable tool for the

assessment of diastolic dysfunction, and can easilydifferentiate normal from pseudonormal patterns ofdiastolic dysfunction. Compared to hemodynamic flowDoppler parameters, it is relatively load-independent anddoes not

fuse even at high heart rates.1

6. Strain and strain-rate imaging may cause problems inclinical use because of noise, angle-dependence andsegments being syncytium, but this seems

to be the right

direction in which to proceed.2 Strain-rate imaging dataobtained from Doppler ultrasound has good agreementwith that obtained from magnetic resonance tagging.8

7. TDI has a distinct advantage with regard to theassessment of right ventricular function, which isdifficult otherwise because of the complex geometry.9

8. Several well-conducted studies have shown that therequirement for resynchronization therapy in heartfailure, optimizing benefits, and follow-up is much betterwith tissue velocity imaging compared to conventionalparameters.10

The clinical applications of TDI are expanding and,despite a learning curve, more and more people are willingto use it. Some of the benefits listed above are obvious, themost important being its role in making echocardiographictechniques and measurements quantitative and lessoperator-dependent. It certainly is not a hoax, nor confinedto research laboratories

alone. Regular use will change

skeptics into converts.

References

1. Waggoner AD, Bierig SM. Tissue Doppler imaging: a usefulechocardiographic method for the cardiac sonographer to assesssystolic and diastolic ventricular function. J Am Soc Echocardiogr 2001;14: 1143–1152

2. Pislaru C, Abraham TP, Belohlavek M. Strain and strain rateechocardiography. Curr Opin Cardiol. 2002; 17: 443–454

3. Evans DH, McDicken WN. Doppler ultrasound physics, instrumentationand signal processing, 2nd ed. West Sussex: John Wiley and Sons; 2000.pp. 5–26

4. Evans DH, McDicken WN. Doppler ultrasound physics, instrumentationand signal processing. 2nd ed. West Sussex: John Wiley and Sons; 2000.pp. 345–347

5. Gorcsan J 3rd, Strum DP, Mandarino WA, Gulati VK, Pinsky MR.Quantitative assessment of alterations in regional left ventricularcontractility with color-coded tissue Doppler echocardiography.Comparison with sonomicrometry and pressure–volume relations.Circulation 1997; 95: 2423–2433

6. Anderson T, McDicken WN. Measurement of tissue motion. Proc InstMech Eng [H] 1999; 213: 181–191

7. Fathi R, Cain P, Nakatani S, Yu HC, Marwick TH. Effect of tissueDoppler on the accuracy of novice and expert interpreters ofdobutamine echocardiography. Am J Cardiol 2001; 88: 400–405

8. Edvardsen T, Gerber BL, Garot J Bluemke DA, Lima JA, Smiseth OA.Quantitative assessment of intrinsic regional myocardial deformationby Doppler strain rate echocardiography in humans: validationagainst three-dimensional tagged magnetic resonance imaging.Circulation 2002; 106: 50–56

9. Frommelt PC, Ballweg JA, Whitstone BN, Frommelt MA. Usefulnessof Doppler tissue imaging analysis of tricuspid annular motion fordetermination of right ventricular function in normal infants andchildren. Am J Cardiol 2002; 89: 610–613

10. Sogaard P, Egeblad H, Pedersen AK, Kim WY, Kristensen BO, HansenPS, et al. Sequential versus simultaneous biventricularresynchronization for severe heart failure: evaluation by tissueDoppler imaging. Circulation 2002; 106: 2078–2084

Partho P Sengupta, Jagdish C Mohan andNatesa G Pandian

GB Pant Hospital,Delhi,India andTufts University, Boston,USA


Indian Heart J 2003; 55: 92–96 Letters to the Editor 95

The potential association between hypertension andcancer mortality has been extensively studied over the

past several years but many aspects of this issue remainobscure.1 A recently published meticulous meta-analysisof previously published data concluded that hypertensionis associated with an increased mortality from cancer,particularly renal cell carcinoma.2 Even though the clinicalimplications have not been clearly established, it isreasonable to assume that elucidation of the underlyingpathophysiologic mechanisms could lead to thedevelopment of effective therapeutic strategies.

Notably, oxidative stress seems to play an important rolein the pathophysiology of various disease states such ashypertension3,4 and cancer.4,5 Several parameters, whichpossibly contribute to the evolution of malignancy inhypertensives, have been associated with oxidative stress.In particular, the apoptotic process, which controls cellulargrowth, can be triggered by intracellular redox imbalancecaused by oxidant stimuli. Furthermore, it is well knownthat oxidative stress leads to deregulation of calciummetabolism. It has been proposed that the accumulationof this ion leads to activation of several mitogens andoncogens. In addition, various vasoactive neurohormonesimplicated in the pathogenesis of hypertension, such asangiotensin II, catecholamines, and insulin, apart fromgrowth-like ef fects, have been shown to deteriorateantioxidant status. Presumably, the observation thatdiuretics increase cancer risk might be explained by the factthat the resulting rise in angiotensin II due to volumedepletion exerts significant mitogenic and oxidative effects.Finally, factors such as aging, smoking, obesity, anddiabetes, which are often related with hypertension, havealso been found to induce oxidative stress.

In a recent “hypothesis” paper, Gago-Dominguez et al.6

proposed that lipid peroxidation, which is increased in obeseand hypertensive subjects, is the mechanism responsible,at least in part, for the increased risk of renal carcinoma insuch individuals. To the best of our knowledge, thehypothesis that oxidative stress contributes to increasedcancer mortality in hypertensives has not yet been tested.In several observational studies, plasma concentrations ofantioxidant vitamins were inversely associated withcardiovascular and cancer mortality.7,8 On the other hand,the evidence in randomized interventional studies that

Hypertension and Cancer Mortality:Is There a Place for AntioxidantInterventions?

antioxidant vitamin supplementation can reducecardiovascular and cancer mortality is scanty or non-existent.9,10 Despite these unpromising results, directevidence is lacking regarding the impact of antioxidantstatus on creased cancer risk observed in individuals withelevated blood pressure.

It is worth noting that besides pure antioxidants, severalcardiovascular and antihypertensive drugs have beenproposed to have intrinsic antioxidant properties.11,12 Theseinclude carvedilol,13 propranolol,14 nebivolol,12 captopril,15

losartan,16 calcium-channel blockers,17 statins,18,19

trimetazidine,20 aspirin,21 and others. Moreover, it has beensuggested that the anti-inflammatory action of some ofthese drugs correlates with their antioxidant capacity.22 Itis prudent to assume that most of these drugs act oncomponents of neurohormonal activation or havemetabolic and anti-inflammatory action. As previouslymentioned, neurohormonal processes are able to induceoxidative stress by themselves. Thus, the study of their effecton oxidative stress in conjunction with cancer mortalityappears to be difficult because their indirect “antioxidanteffect” may confound their intrinsic free-radical scavengingcapacity. Despite these difficulties, we believe that suchmedications may serve as valuable tools in the research ofantioxidant interventions in this setting. Well-designed,prospective studies in the near future may further elucidatethe association between these two major diseases, anddelineate the role of antioxidant interventions.

References

1. Stumpe KO. Hypertension and the risk of cancer: is there newevidence? J Hypertens 2002; 20: 565–567

2. Grossman E, Messerli FH, Boyko V, Goldbourt U. Is there an associationbetween hypertension and cancer mortality? Am J Med 2002; 112:479–486

3. Berry C, Brosnan MJ, Fennell J, Hamilton CA, Dominiczak AF.Oxidative stress and vascular damage in hypertension. Curr OpinNephrol Hypertens 2001; 10: 247–255

4. Halliwell B, Gutteridge JMC. Free radicals in biology and medicine. 3rded. Oxford: Oxford University Press; 1999

5. Kovacic P, Jacintho JD. Mechanisms of carcinogenesis: focus onoxidative stress and electron transfer. Curr Med Chem 2001; 8: 773–796

6. Gago-Dominguez M, Castelao JE, Yuan JM, Ross RK, Yu MC. Lipidperoxidation: a novel and unifying concept of the etiology of renalcell carcinoma. Cancer Causes Control 2002; 13: 287–293

7. Jha P, Flather M, Lonn E, Farkouh M, Yusuf S. The antioxidantvitamins and cardiovascular disease. A critical review ofepidemiologic and clinical trial data. Ann Intern Med 1995; 123: 860–872

8. Khaw KT, Bingham S, Welch A, Luben R, Wareham N, Oakes S, et al.Relation between plasma ascorbic acid and mortality in men andwomen in EPIC-Norfolk prospective study: a prospective populationstudy. Euroean Prospective Investigation into Cancer and Nutrition.Lancet 2001; 357: 657–663



9. The Alpha-Tocopherol, Beta Carotene Cancer Prevention StudyGroup. The effect of vitamin E and beta carotene on the incidence oflung cancer and other cancers in male smokers. N Engl J Med 1994;330: 1029–1035

10. Heart Protection Study Collaborative Group. MRC/BHF HeartProtection Study of antioxidant vitamin supplementation in 20,536high-risk individuals: a randomised placebo-controlled trial. Lancet2002; 360: 23–33

11. Landmesser U, Harrison DG. Oxidant stress as a marker forcardiovascular events: ox marks the spot. Circulation 2001; 104:2638–2640

12. Luscher TF, Spieker LE, Noll G, Cosentino F. Vascular effects of newercardiovascular drugs: focus on nebivolol and ACE-inhibitors. JCardiovasc Pharmacol 2001; 38 (Suppl. 3): S3–S11

13. Nakamura K, Kusano K, Nakamura Y, Kakishita M, Ohta K, NagaseS, et al. Carvedilol decreases elevated oxidative stress in human failingmyocardium. Circulation 2002; 105: 2867–2871

14. Khaper N, Rigatto C, Seneviratne C, Li T, Singal PK. Chronic treatmentwith propranolol induces antioxidant changes and protects againstischemia–reperfusion injury. J Mol Cell Cardiol 1997; 29: 3335–3344

15. Hayek T, Attias J, Smith J, Breslow JL, Keidar S. Antiatheroscleroticand antioxidative effects of captopril in apolipoprotein E-deficientmice. J Cardiovasc Pharmacol 1998; 31: 540–544

16. Khaper N, Singal PK. Modulation of oxidative stress by a selectiveinhibition of angiotensin II type 1 receptors in MI rats. J Am Coll Cardiol2001; 37: 1461–1466

17. Sevanian A, Shen L, Ursini F. Inhibition of LDL oxidation and oxidizedLDL-induced cytotoxicity by dihydropyridine calcium antagonists.Pharm Res 2000; 17: 999–1006

18. Delbosc S, Morena M, Djouad F, Ledoucen C, Descomps B, Cristol JP.Statins, 3-hydroxy-3-methylglutaryl coenzyme A reductase

inhibitors, are able to reduce superoxide anion production by NADPHoxidase in THP-1-derived monocytes. J Cardiovasc Pharmacol 2002;40: 611–617

19. Delbosc S, Cristol JP, Descomps B, Mimran A, Jover B. Simvastatinprevents angiotensin II-induced cardiac alteration and oxidativestress. Hypertension 2002; 40: 142–147

20. Tselepis A, Doulias P, Lourida E, Glantzounis G, Tsimoyiannis E, GalarisD. Trimetazidine protects low-density lipoproteins from oxidation andcultured cells exposed to H2O2 from DNA damage. Free Radic Biol Med2001; 30: 1357–1364

21. Lopez-Farre A, Riesco A, Digiuni E, Mosquera JR, Caramelo C, deMiguel LS, et al. Aspirin-stimulated nitric oxide production byneutrophils after acute myocardial ischemia in rabbits. Circulation1996; 94: 83–87

22. Korantzopoulos P, Galaris D, Papaioannides D, Kokkoris S. Does C-reactive protein represent an oxidative stress marker in cardiovasculardiseases? Int J Clin Pract 2003 (in press)

Panagiotis KorantzopoulosLaboratory of Biological Chemistry, University of

Ioannina Medical School, Ioannina, Greece

Dimitrios PapaioannidesDepartment of Medicine

Arta General District HospitalArta, Greece


Indian Heart J 2003; 55: 97–100 Selected Summaries 97Selected Summaries

In-Stent Restenosis Treated With Stent-Based Delivery of PaclitaxelIncorporated in a Slow-Release Polymer Formulation

K Tanabe et al. TAXUS III Trial Investigators. Circulation 2003; 107: 559–564

Summary

The TAXUS III trial was a single-arm, 2-center studyevaluating the feasibility and safety of a paclitaxel-elutingstent for the treatment of in-stent restenosis (ISR). Itenrolled 28 patients with ISR with a lesion length of <30mm, 50%–99% diameter stenosis in vessels with a diameterbetween 3 and 3.5 mm. Patients with acute myocardialinfarction, poor ejection fraction (<30%), recent stroke,renal dysfunction, or contraindication to antithromboticswere excluded from the present study. The stent used wasTAXUS NIRx paclitaxel-eluting stent (Boston ScientificCorporation, USA), with a total load of 1.0 µg/mm2 ofpaclitaxel incorporated into a slow-release copolymercarrier system that gives biphasic release an early burst overthe first 48 hours followed by slow release over the next 10days. Predilatation was performed in all cases, and post-dilatation was performed when necessary. All the stentswere 15 mm long and 3–3.5 mm in diameter. During theprocedure, ACT was kept at > 250 s, and standardantiplatelet regimen was followed post-procedure.Angiographic and intravascular ultrasonographic (IVUS)follow-up was performed at 6 months, and clinical follow-up at 6 months and 1 year. There were clinical,angiographic, and IVUS-based end-points. Angiographicfollow-up was available in 25 patients, and IVUS follow-upin 17. The major adverse cardiac event (MACE) rate was29% (8 patients; 1 non-Q-wave myocardial infarction, 1coronary artery bypass grafting, and 6 target lesionrevascularization [TLR]). Two patients underwentrevascularization for restenosis in the gap between 2paclitaxel-eluting stents, 1 for restenosis for a bare stentimplanted at edge dissection, 1 for anginal symptoms butno restenosis (diameter stenosis of 32.5%), and in 2 patientsnot for restenosis but for incomplete stent apposition/expansion as detected on IVUS study. One patient had targetvessel occlusion but required no intervention as there wereno symptoms. Binary angiographic restenosis occurred in4 patients (16%). One patient had restenosis of a bare stentimplanted to cover edge dissection due to implantation ofa paclitaxel-coated stent, and 2 patients had restenosis inthe gap between 2 paclitaxel-eluting stents. Thus, of the 4restenoses, only 1 occurred in the area of paclitaxel delivery.The diameter stenosis at follow-up was 30.8%, with anaverage in-stent late loss of 0.54 mm. Late loss at theproximal and distal edges was 0.20 and 0.11 mm,respectively, Of 17 patients undergoing IVUS, the volumeof neointimal hyperplasia was 20.3±23.1 mm3. There wasno evidence of positive or negative remodeling in the 6patients undergoing TLR. Thus, paclitaxel-eluting stentimplantation appears to be safe and potentially effective inthe treatment of ISR.

Comments

In the current era, use of stents has improved the safetyand success of percutaneous coronary interventions (PCI)and even reduced the restenosis rate (BENESTENT andSTRESS trials). However, their use has also brought aboutthe complex and difficult-to-manage problem of ISR andrepeat TLR. Several pharmacologic and mechanicalapproaches have been tried to overcome the problem of ISRbut they have limitations. Plain balloon angioplasty for ISRleads to a re-restenosis rate of 30%–85%, depending onwhether it is a focal restenosis or diffuse ISR. Similarly,directional coronary atherectomy (re-restenosis rate 60%),and excimer laser (TLR 41%) have not been of muchbenefit. Rotablation and the use of cutting balloon mayhave some theoretical advantage but large studies to provetheir efficacy are not available. Only radiation therapy hasbeen found to be consistently useful till date. A number ofdouble-blind clinical trials using gamma sources (SCRIPPS,WRIST, GAMMA-1) and a beta source (START, INHIBIT)have reported a striking efficacy in preventing re-restenosisin patients with established ISR (17%–34%). A majorproblem with brachytherapy, however, has been late stentthrombosis and late restenosis. Implantation of a new stentin the irradiated segment and withdrawal of clopidogrel/ticlopidine have been correlated with late thrombosis.Preliminary evidence from trials using prolongedantiplatelet regimens (>3 months) have suggested a markedreduction in this complication. Another major problemwith brachytherapy is the edge effect. Careful analysis hassuggested that these edge ef fects may represent a“geographic miss” of radiation and this problem can besubstantially ameliorated with careful attention toregistering adequate doses of radiation throughout theinstrumented segment of the coronary artery.Furthermore, brachytherapy is less accessible, and itsdelivery requires special handling. In this context, stent-based local drug delivery have revolutionized the field ofPCI. Smith et al. reported use of sirolimus-eluting stents in15 patients with ISR; at 4 months, MACE occurred in 6.7%(1/15) cases. One patient died suddenly, and late occlusionoccurred in another. In-stent late loss was 0.02 mm, andon IVUS, neointimal hyperplasia was 7.8±0.03 mm2. Inthe present study using paclitaxel-coated stents, althoughthe technical re-restenosis rate is 16%, actual restenosisoccurred in only 1 patient. Similarly, the occurrence of lateloss (0.54 mm) and neointimal hyperplasia (20.3 mm2) waslow. However, these studies are too small and do not have aplacebo arm. Nevertheless, it does seem that drug-elutingstents may be of use in the therapy of ISR.

IHJ-Selected Summary.p65 5/6/2003, 10:08 AM97

98 Selected Summaries Indian Heart J 2003; 55: 97–100

Summary

The Antihypertensive and Lipid-Lowering Treatment toPrevent Heart Attack Trial (ALLHAT) was a randomized,double-blind, active controlled, clinical trial conducted witha view to determine which antihypertensive treatmentdecreased the risk of coronary artery disease (CAD) or othercardiovascular disease (CVD) events. It enrolled 33 357hypertensive subjects, both men and women, >55 years ofage, with at least one other risk factor for CAD from 623North American centers. The patients were randomized toreceive chlorthalidone 12.5–25 mg/day (n=15 255),amlodipine 2.5–10 mg/day (n=9048), or lisinopril 10–40mg/day (n=9054), and followed up for 8 years. The primaryend-point was a combination of CAD deaths plus nonfatalmyocardial infarction (MI). The secondary end-points wereall-cause mortality, stroke, combined CAD end-point (CADdeath, nonfatal MI, coronary revascularization, or anginarequiring hospitalization), and combined CVD (combinedCAD end-point, stroke, treated angina withouthospitalization, CHF and peripheral arterial disease). At amean follow-up of 4.9 years, the primary outcome of CADdeath plus nonfatal MI was not different in any of thetreatment groups. Compared with chlorthalidone (6-yearrate, 11.5%), the relative risks (RRs) were 0.98 (95% CI:0.90–1.07) for amlodipine (6-year rate, 11.3%) and 0.99(95% CI: 0.91–1.08) for lisinopril (6-year rate, 11.4%).Similarly, all-cause mortality was also not dif ferent.Chlorthalidone was more effective in controlling systolicBP as compared to amlodipine and lisinopril (p<0.05 forboth). On the other hand, amlodipine was more effectivein lowering the diastolic BP (p<0.01). Among thesecondary end-points, occurrence of CHF was commonerin the amlodipine group as compared to the chlorthalidonegroup (10.2% v. 7.7%, RR 1.38; 95% CI: 1.25–1.52).Similarly, combined CAD end-points (33.3% v. 30.9%, RR1.10; 95% CI: 1.05–1.16), stroke (6.3% v. 5.6%, RR 1.15;95% CI: 1.02–1.30), and CHF (8.7% v. 7.5%, RR 1.19; 95%CI: 1.07–1.31) were all higher in the lisinopril groupcompared to chlorthalidone group. The authors concludedthat diuretics are superior to amlodipine or lisinopril inreducing the cardiovascular morbidity and are much lessexpensive.

Comments

For decades, the search for an ideal antihypetensive hasbeen on. An ideal antihypertensive is one which not only

Major Outcomes in High-Risk Hypertensive Patients Randomized to Angiotensin-Converting Enzyme Inhibitor or Calcium-Channel Blocker v. Diuretics

The ALLHAT Collaborative Research Group. JAMA 2002; 288: 2981–2997

controls the blood pressure but also reduces cardiovascularmorbidity and, even more importantly, mortality, withoutcausing side-effects. The cost of drug therapy is alsoimportant, especially because it has to be continued lifelong.Classically, diuretics and beta-blockers have beenconsidered drugs of first choice, primarily because of theireffect in reducing mortality. However, these drugs are notwithout side-effects, and are besieged with the problems ofdiuresis, impotence, and depression, as well as dyslipidemiaand metabolic abnormalities. Angiotensin-convertingenzyme inhibitors (ACE-I) have emerged as a crediblealternative because of their efficacy and ability to preventcomplications such as the progression of renal disease andstroke. However, till date, no large study has shownreduction in overall mortality with ACE-I. Calcium-channelblockers (CCB), on the other hand, though efficacious, haveactually shown increased mortality with the use of short-acting ones. Unfortunately, very few large studies haveundertaken head-to-head comparison of variousantihypertensive agents. In this context, the ALLHAT studyis a landmark one. The major finding of this study was astriking and unequivocal null result in reducing theprimary end-point, i.e. CAD death plus nonfatal MI (exceptdoxazocin which was discontinued earlier on in the study).Because of the large number of subjects involved in thestudy, the mortality benefit with diuretics, if at all, is likelyto be very small. However, in the context of reduction ofcardiovascular morbidity, therapy with chlorthalidone doesseem superior to that with amlodipine or lisinopril. In fact,chlorthalidone was found to be superior to even lisinoprilin reducing 6-year CHF rates (8.7% v. 7.7%; CI: 1.07–1.31),a risk reduction of 1.19. However, despite being path-breaking, the study is still fraught with certain limitations.The blood pressure control achieved was not similar in allthe groups (SBP was better controlled with chlorthalidone,while DBP was better controlled with amlodipine). Newerantihypertensive agents such as ramipril, angiotensinreceptor blockers, selective aldosterone antagonists, andbeta-blockers have not been studied. The most importantmessage from this study is that, as of now, thiazide-typediuretics should still be considered first-line therapy becauseof their excellent efficacy, low long-term morbidity andmortality rates, and low cost.


Indian Heart J 2003; 55: 97–100 Selected Summaries 99

Summary

Electrical Stimulation versus Coronary Artery BypassSurgery in Severe Angina Pectoris (ESBY) was arandomized, prospective, open-comparison study ofcoronary artery bypass grafting (CABG) and spinal cordstimulation (SCS) in patients with severe angina despiteoptimal pharmacologic treatment. Prior to ESBY study, SCSwas used in patients for whom revascularization was notpossible. In ESBY study patients were enrolled if angina wasnot controlled with anti-anginal agents, and the patientswere eligible for CABG with an increased but acceptablerisk (as per the ACC/AHA, 1991 indications). Patientsenrolled did not have acute myocardial infarction in thepast 6 months, and were intelligent enough to manage theSCS device. The mean Higgins score (scoring system forestimation of preoperative risk) was 4.2. The SCS devicewas implanted under local anesthesia. The long-termoutcome regarding quality of life and survival of those 104patients included in ESBY study (1992–1995) wasinvestigated. The quality of life was assessed by theNottingham health profile (NHP) and quality of lifequestionnaire (angina pectoris). The quality of life at 6months had shown significant improvement compared tothe run-in (p<0.001). This result persisted even up to 4.8years of follow-up. In the CABG group, however, there wasa deterioration during the postoperative observationalperiod. Within 6 months of randomization, 1 patient in theSCS group and 7 in the CABG group died. At 3 and 5 years,there was no significant difference in mortality betweenthe two groups. Three years after randomization, 84.9% inthe SCS group and 76.5% in the CABG group were alive.At 5 years, 75.5% in the SCS group and 68.6% in the CABGgroup were alive.Thus, spinal cord stimulation as well as

CABG offered long-lasting improvement in quality of lifeand survival up to 5 years was comparable between thegroups.

Comments

SCS or epidural spinal electrical stimulation (ESES) has beenused for treatment of refractory angina. The ESBY studywas published in 1998 with 6 months of follow-up, andhad shown no difference between the CABG and SCS arms.The present study is a long-term follow-up of the ESBYpatients. The mortality rate in this study was similar to thatseen in the angina subpopulation of the Swedish Councilon Technology Assessment in Health Care patients. Ascompared with 0.8% mortality seen in CASS trial or 16%observed in the ECSS study, the mortality is higher in thepresent study. However, the patient population of the ESBYstudy does not match that of CASS/ECSS studies. Thesurvival data in the study showed equivalent survival inboth the arms with equal improvement in the quality oflife. As it is a long-term follow-up study, the placebo effectwill not be seen as it decreases with time. However, the exactmechanism of benefit is unknown. The antianginal effectis secondary to an anti-ischemic effect, which in turn seemsto be due to a reduction in myocardial oxygen consumption;however, a redistribution of coronary blood flow cannotbe excluded. Thus, it seems that SCS, being noninvasive andless traumatic, may be a therapeutic alternative for patientswith severe angina and high risk for bypass surgery. Atpresent, SCS treatment is costly and technically moredemanding, thereby limiting its use. Further prospectivestudies are needed to compare SCS and revascularizationto establish this as an alternative treatment in patients withrefractory angina.

Long-Term Effects of Spinal Cord Stimulation and Coronary Artery BypassGrafting on Quality of Life and Survival in the ESBY Study

O Ekre et al. Electrical Stimulation versus Coronary Artery Bypass Surgery in Severe Angina Pectoris.Eur Heart J 2002; 23: 1938–1945


100 Selected Summaries Indian Heart J 2003; 55: 97–100

Summary

Eplerenone is an aldosterone antagonist that selectivelyinhibits the mineralocorticoid receptors, leaving theglucocorticoid, progesterone or androgen receptorsunaffected. Its selective action is beneficial in preventingside-effects such as sexual dysfunction and gynecomastiathat are encountered with the nonselective blockadeprovided by spironolactone. This study is a multicenter,international, randomized, double-blind, placebo-controlled trial that included patients with acutemyocardial infarction (time to randomization 3–14 days)and left ventricular dysfunction (LVEF <40%) with clinicalheart failure, except in diabetics in whom the symptoms ofheart failure did not have to be demonstrated. Exclusioncriteria were the use of potassium-sparing diuretics, aserum creatinine level of 2.5 mg/dl or above, and a serumpotassium concentration of >5 mmol/L. The potassiumlevel was monitored on subsequent follow-up visits. Sixthousand six hundred forty-two patients were randomizedto eplerenone (25 mg/day initially, titrated to a maximumof 50 mg/day; 3319 patients), or placebo (3313 patients).The study continued until 1012 deaths occurred. Atbaseline, the 2 groups were not significantly different andwere receiving optimal medical therapy including ACEinhibitors or angiotensin-receptor blockers (87%), beta-blockers (75%), aspirin (88%), and diuretics (60%).

The primary end-points were death from any cause anddeath from cardiovascular causes, or first hospitalizationfor a cardiovascular event including heart failure, recurrentacute myocardial infarction (AMI), stroke, or ventriculararrhythmia.

In the mean follow-up period of 16 months, 478 deathsoccurred in the eplerenone group (14.4%) and 554 deathsin the placebo group (16.7%) (relative risk [RR] 0.85;p=0.008). Of these deaths, 407 in the eplerenone groupand 483 in the placebo group were attributed tocardiovascular causes (RR: 0.83; p=0.005), whichincluded sudden cardiac death, acute myocardialinfarction, and heart failure. The reduction in risk of suddendeath from cardiac causes was statistically significant (RR0.79; p=0.03). There was a relative reduction of 15% inthe risk of hospitalization for heart failure and there were23% fewer episodes of hospitalization in those receivingeplerenone compared to placebo. The rate of death fromany cause or any hospitalization was 8% lower in theeplerenone than in the placebo group (RR: 0.92; p=0.02).During the study period, 493 patients in the placebo groupand 528 patients in the eplerenone group discontinued thedrug. There was a significantly greater increase in theserum creatinine concentration in the eplerenone groupthan in the placebo group, but the difference was clinicallysmall. Serious hyperkalemia occurred in 5.5% of patientsin the eplerenone group, as compared with 3.9% of patients

Eplerenone, a Selective Aldosterone Blocker, in Patients With Left VentricularDysfunction After Myocardial Infarction

B Pitt et al. The Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy AndSurvival Study (EPHESUS) Trial Investigators N Engl J Med 2003; 348: 1309–1321

in the placebo group.The incidence of hyperkalemia wasmore in patients with a baseline creatinine clearance of<50 ml/min. Though there were 12 hospitalizations forserious hyperkalemia in the eplerenone group, there wereno deaths. Three patients in the placebo group wereadmitted because of hyperkalemia and there was 1 death.In conclusion, the study showed that eplerenonesignificantly reduces morbidity and mortality in patientswith AMI and LV dysfunction.

Comments

The Randomized Aldactone Evaluation Study (RALES)proved the importance of aldosterone antagonists in themanagement of patients with heart failure. In addition toreducing mortality by 30%, small doses of spironolactoneresulted in an improvement in the ejection fraction andenhanced treatment tolerance. It has been shown thatpatients with heart failure have high levels of plasmaaldosterone because of increased production and decreasedhepatic clearance. Sustained aldosterone levels promoteendothelial dysfunction and oxidative stress in thevasculature and also organ fibrosis. The beneficial effecton ventricular remodeling is evidenced by this study, inwhich the relative risk of death was reduced by 15% andthe risk of hospitalization was also reduced by 15%. Thereduction in cardiovascular mortality is mainly due to 21%reduction in the rate of sudden death due to cardiac causes.An important point in the EPHESUS trial is that patientswere already receiving optimal therapy (ACE inhibitors in87% and beta-blockers in 75%). In RALES, beta-blockerswere used in only about 11%, although ACE inhibitors weregiven to 94%. One-year mortality among the placebo groupwas higher in RALES (25%) as compared with EPHESUS(13.6%). This difference in mortality may reflect the moreseverely depressed systolic function in patients enrolled inRALES (LVEF averaged 25% in RALES as compared to 33%in EPHESUS). Taken together, these 2 trials of aldosteroneblockade have enrolled more than 8000 patients withsystolic dysfunction and symptoms of heart failure, andhave shown that the addition of aldosterone antagoniststo these patients can substantially reduce overall mortalityand rate of sudden death. Currently, the guidelines of theAmerican College of Cardiology and American HeartAssociation specify that the use of spironolactone forpatients with mild-to-moderate heart failure has not beentested. After the RALES and EPHESUS, their beneficial rolein the management of heart failure is quite evident, withjust a little extra effort to monitor the potential side-effects,especially hyperkalemia. Further studies would determinewhether this class of drugs will prove as efficacious inpatients with less severe symptoms or in those with heartfailure due to diastolic dysfunction.


May 1–3, 2003, 18th International Meeting onClinical Cardiology, Athens, GreeceContact: Dr P. Toutouzas

Chairman, Organizing CommitteeDepartment of CardiologyUniversity of Athens, GreeceTel: 0106401477Fax: 0106401478

June 21–24, 2003, Heart Failure Update 2003,Strasbourg, FranceContact: European Society of Cardiology

The European Heart House2035 Route des Colles, Les Templiers – BP 179Sophia Antipolis Cedex 06903, FranceTel: 33 4 9294 7600Fax: 33 4 9294 7601e-mail: [email protected]

June 25–28, 2003, 14th Asian Pacific Congress ofCardiology (APCC), Singapore, SingaporeContact: Dr Michael Lim, Chairman, 14th APCC

The Secretariat, 302, Orchard Road# 16-04, Tong Building, Singapore 238862Republic of SingaporeFax: 65 836 0436e-mail: [email protected]

September 16–21, 2003, Transcatheter Cardio-vascular Therapeutics 2003, Washington, D.C., USAContact: The Course Directors

55 East 59th Street, 6th FloorNew York NY 10022-1112, USATel: 1 212 434 6300Fax: 1 212 434 6386e-mail: [email protected]

Indian Heart J 2003; 55: 101 Calendar of Conferences

October 26–30, 2003, 69th Annual ScientificAssembly, American College of Chest Physicians,Orlando, Florida, USAContact: American College of Chest Physicians

3300 Dundee Road, Northbrook IL 60062, USATel: 1 847 498 1400Fax: 1 847 498 5460

November 2–5, 2003, 76th Scientific Session,American Heart Association (AHA), Orlando, Florida,USAContact: American Heart Association

7320 Greenville Avenue, Dallas TX 75231, USATel: 1 214 373 6300Fax: 1 214 373 3406

December 4–7, 2003, 55th Annual Conference ofCardiological Society of India, Kolkata, IndiaContact: Dr Asok Kumar Kar, Organizing Secretary

Indian Heart HouseP-60, CIT Road, Scheme VIIM,Kankurgachi, Kolkata 700 054, IndiaFax: 033 355 6308e-mail: [email protected]

November 7–10, 2004, 77th Scientific Session,American Heart Association (AHA), New Orleans,Louisiana, USAContact: American Heart Association

7320 Greenville Avenue, Dallas TX 75231, USATel: 1 214 373 6300Fax: 1 214 373 3406

Cal of Conference.p65 5/6/2003, 10:08 AM101

hi per tensy

Documents