state-of-the-art paper …imaging.onlinejacc.org/content/jimg/4/3/292.full.pdf ·...

J A C C : C A R D I O V A S C U L A R I M A G I N G V O L . 4 , N O . 3 , 2 0 1 1

© 2 0 1 1 B Y T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N I S S N 1 9 3 6 - 8 7 8 X / $ 3 6 . 0 0

P U B L I S H E D B Y E L S E V I E R I N C . D O I : 1 0 . 1 0 1 6 / j . j c m g . 2 0 1 0 . 1 2 . 0 0 6

S T A T E - O F - T H E - A R T P A P E R

Cardiac PET: A Versatile, QuantitativeMeasurement Tool for Heart FailureManagement

Henry Gewirtz, MD

Boston, Massachusetts

Current American Heart Association/American College of Cardiology practice guidelines classify conges-

tive heart failure (CHF) in 4 stages (A, B, C, and D). This review focuses on state-of-the-art and future

applications of quantitative positron emission tomography (PET) myocardial perfusion and metabolic

imaging in the clinical evaluation and treatment of patients in all CHF stages. Basic physiological and

metabolic principles related to the regulation of myocardial blood flow and metabolism at various stages

of CHF are briefly reviewed. The advantages of quantitative PET image analysis in contrast to simple

qualitative visual analysis of the scans also will be addressed. Finally, potential future clinical applications of

quantitative PET for CHF evaluation and treatment will be discussed. (J Am Coll Cardiol Img 2011;4:

292–302) © 2011 by the American College of Cardiology Foundation

chusfrom

Current American Heart Association/AmericanCollege of Cardiology guidelines recommendconsideration of congestive heart failure (CHF)in 4 stages (A, B, C, and D; Fig. 1) (1). StagesA and B represent preclinical CHF (stage A:patient at high risk but without structuralheart disease; stage B: structural heart diseasepresent but symptoms/signs of CHF absent).Stages C and D reflect overt CHF (stage C:structural heart disease present with prior orcurrent symptoms of CHF; stage D: refrac-tory CHF requiring specialized interven-tion). The incidence of CHF in the UnitedStates has been increasing steadily in recentyears (400,000 new cases every year estimatedby the National Heart, Lung and BloodInstitute in 1996), and mortality, particularlyin stage D, is in excess of 50% per year (2).Accordingly, the syndrome represents a ma-

From the Department of Medicine (Cardiology Division), MassaBoston, Massachusetts. Dr. Gewirtz has received a research grant

Manuscript received November 9, 2010; revised manuscript received D

jor public health problem. Ischemic heartdisease (IHD) and hypertension are the mostcommon causes of CHF, although obesityand diabetes, even in the absence of overtIHD but frequently present together andoften with hypertension, are increasingly rec-ognized as common etiologies (2). Quantita-tive positron emission tomography (PET)imaging plays an important role in bothdiagnosis of etiology and assessment of treat-ment of CHF at various stages of the syn-drome and provides prognostic informationas well. The current state of the art andfuture directions for quantitative PET imag-ing in CHF is the focus of this review, whichwill also address the advantages of the quan-titative approach in contrast to simple qual-itative visual interpretation of PET cardiacimages.

etts General Hospital, Harvard Medical School,FluoroPharma, Inc.

ecember 20, 2010, accepted December 23, 2010.

[

rkftCO

ectftmisCmiosmv[

raphy

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 4 , N O . 3 , 2 0 1 1

M A R C H 2 0 1 1 : 2 9 2 – 3 0 2

Gewirtz

Quantitative PET for Evaluation of Patients With HF

293

Quantitative PET Imaging in Patients With CHF:State of the Art

The role for state-of-the-art quantitative PETmeasurement of myocardial blood flow (MBF)(Fig. 2) and metabolism (glucose and fatty acid)depends on the specific clinical question(s) to beaddressed. However, in more general terms, thereare several points that should be made concerningthe advantages of quantitative versus qualitativesimple visual analysis of cardiac PET images inpatients with CHF. First, simple visual analysis ofPET myocardial perfusion images requires the as-sumption of a “normal” perfusion or metaboliczone—an erroneous assumption and clearly a majorlimitation of that approach. Second, simple visualinterpretation is by definition subjective and henceliable to all associated shortcomings. Third, it hasbeen shown in a modern clinical trial (PARR 2[PET and Recovery Following Revascularization])(3,4) and older observational studies (5–9) thatobjective measurement of parameters indicative ofmyocardial viability or myocardial flow reserve arevaluable in improving patient management or as-sessing prognosis in patients with CHF. Suchparameters (e.g., quantitative, objective determina-tion and extent of MBF/18F-fluorodeoxyglucoseFDG] “mismatch” and 82Rb myocardial tracer

kinetics and measurement of myocardial flow re-serve) necessitate state-of-the-art quantitative PETimaging of MBF and metabolism and cannot beobtained by simple visual analysis of static PETimages. Finally, potential future clinical applica-tions for PET cardiac imaging in CHF (e.g., riskassessment for sudden cardiac death in patientswith ischemic cardiomyopathy; PARAPET [Pre-diction of Arrhythmic Events With Positron Emis-sion Tomography] trial) (10–13) will depend onadoption of current state-of-the-art quantitativePET methodologies used in such trials for assess-ment of MBF, glucose metabolism, and sympa-thetic nervous system function.

Pathophysiology of MBF and Clinical Applicationsof PET in Left Ventricular Dysfunction and CHF

Ischemic cardiomyopathy. Endothelial dysfunctionelated to dyslipidemia and oxidant stress (14,15) isnown to affect the coronary microcirculation be-ore the onset of overt coronary atherosclerosis andherefore places affected individuals in stage AHF (high risk, without structural heart disease).
ngoing dyslipidemia, oxidant stress, and resulting
ndothelial dysfunction commonly progress to overtoronary atherosclerosis (stage B CHF), which inurn may be complicated by acute myocardial in-arction and resulting left ventricular (LV) dysfunc-ion or frank CHF (stage C CHF). Extensiveyocardial damage from either a single or multiple

nfarcts may result in refractory CHF requiringpecialized treatment or intervention (stage DHF). Positron emission tomography measure-ents of absolute myocardial blood flow play an

mportant role in patient management at each stagef CHF related to IHD or predisposing conditionsuch as obesity, hypertension, diabetes, dyslipide-ia, and smoking. PET assessment of myocardial

iability (typically with combined MBF and glucose18F-FDG] metabolism study) is especially impor-

tant in stages C and D CHF.The role of PET measurements of MBF in assess-

ment of coronary microvascular function hasbeen reviewed extensively (14–16). Evenasymptomatic patients with dyslipidemiawithout manifest IHD may have abnormalmyocardial flow reserve owing to coronarymicrovascular dysfunction (17). The same istrue of patients with obesity, diabetes, andhypertension either alone or in combination(18). A common mechanism in each ofthese conditions is endothelial dysfunctionrelated to oxidant stress and subsequentreduction in nitric oxide bioavailability (18–21). Depending on whether or not LVhypertrophy is present in association withhypertension, mechanical compressive forcesand interstitial fibrosis also may play a role(22). Circulating vasoactive peptides (e.g.,endothelin) (23) also may contribute to impairment ofcoronary microvascular dilator capacity. Finally, it hasbeen shown that evidence of increased serum levels ofcirculating biomarkers of inflammation increase therisk of developing CHF (24–26) and may do so atleast in part via the oxidant stress mechanism notedearlier.

Several small observational studies in humanshave suggested that PET measurements of absoluteMBF and MBF reserve may be useful in providingprognostic information concerning progressionfrom stage A/B to C/D CHF in patients withhypertrophic (5,27) or idiopathic dilated cardiomy-opathy (DCM) (7). Although the studies havecertain limitations in addition to small sample size,such as reliance on flow reserve ratio as an endpoint, they nevertheless demonstrated the potential

A B B

A N D

CAV �

vascu

CHF �

DCM

FDG �

FFA �

IHD �

IVUS

LV �

MBF �

PET �

tomog

for quantitative measurements of absolute M

R E V I A T I O N S

A C R O N YM S

coronary artery

lopathy

congestive heart failure

� dilated cardiomyopathy

fluorodeoxyglucose

free fatty acid

ischemic heart disease

� intravascular ultrasound

left ventricular

myocardial blood flow

positron emission

BF to

tenCmbtqobs

smaaen

tion

and dobut


M A R C H 2 0 1 1 : 2 9 2 – 3 0 2

Gewirtz


294

provide important clinical information unavailablefrom simple qualitative visual analysis of the images.Similar data regarding prognosis have been reportedfor patients with IHD (28). More widespreadapplication of clinical insights derived from thesestudies is now possible with generator-derived 82Rbfor PET measurements of absolute MBF and MBFreserve with a commercially available Food andDrug Administration–approved computer program

Figure 1. American Heart Association/American College of Ca

ASCVD � atherosclerotic cardiovascular disease; CHF � congestivefraction; LVH � left ventricular hypertrophy; MI � myocardial infarc

Parametric 13N-Ammonia Myocardial Blood Flow Images

scale (ml/min/g) is shown at the right for rest, adenosine (ADO),

ramine (DBTMN) conditions in a healthy human subject.

(Corridor 4DM, CFR option, INVIA MedicalImaging Solutions, Ann Arbor, Michigan). Theprogram has been validated by comparison with13N-ammonia measurements of MBF in humans(29) and therefore is suitable for use with 13N-ammonia as well. As noted earlier, use of quantita-tive measurements of MBF do not depend on theinvalid assumption of an obligatory “normal” zoneand so are inherently superior to simple qualitativevisual assessment of these images. Further, as 18F-labeled MBF tracers (e.g., 18F-BMS-747158-02[30] and possibly 18F-BFPET [31]) become available,he ability to make these measurements will becomeven more widespread. In addition to providing prog-ostic information, which may be helpful in guidingHF management at an early stage, quantitative PETeasurements of MBF will permit establishment of

aseline pre-treatment levels of maximal MBF andherefore provide an objective parameter for subse-uent follow-up exams designed to assess the effectsf therapy (e.g., statins [32]) in patients judged toe at high risk for progression to more advancedtages (C and D) of CHF.

The use of PET quantitative assessment of ab-olute MBF and myocardial glucose metabolism isore clearly established in the setting of stages C

nd D CHF. When signs and symptoms of CHFre first manifest but the etiology has not beenstablished, it is common practice to perform diag-ostic cardiac catheterization and coronary angiog-

logy Guideline for Clinical Staging of CHF

rt failure; LV � left ventricular; LVEF � left ventricular ejection.

rdio

hea

Figure 2.

The color

aphy. However, PET quantitative assessment of


M A R C H 2 0 1 1 : 2 9 2 – 3 0 2

Gewirtz


295

absolute MBF will be very helpful in a variety ofsettings. Patients with chronic renal insufficiency,not yet on dialysis, are at high risk for severeworsening of renal function with coronary angiog-raphy. PET quantitative assessment of absoluteMBF with 13N-ammonia or 82Rb is an excellentmethod for assessment of the coronary circulationin such cases. Detection of hemodynamically signif-icant stenoses in each of the 3 major coronary vascularterritories is possible with a single MBF measurementobtained during adenosine infusion (33). The samemethodology also should be considered for patientswhose prior probability for IHD is low (�10%) andcertainly for those with coronary stenoses of uncertainphysiological significance (see the following paragraphand Figure 3).

PET quantitative assessment of absolute MBFduring adenosine stress may be thought of as anoninvasive fractional flow reserve test (33) (Fig. 3).In this regard, it should be noted that coronaryintervention(s) performed in patients with abnormalfractional flow reserve were associated in the FAME(Fractional Flow Reserve Versus Angiography forMultivessel Evaluation) trial (34) with reduced rates ofthe primary end point at 1 year (death, nonfatalmyocardial infarction, and repeat revascularization)compared with that of similar patients in whompercutaneous coronary interventions were performedbased on visual assessment of the lesions alone. Thisstudy thus provided another example of the superiorityof objective physiological assessment over simple qual-

Figure 3. Schematic Representation of the Physiological Basis f

The pressure drop (dP) across a coronary stenosis is given by the estenosis geometry and blood viscosity (80–82). Under conditions inapproaches 0, and so there is no dP along the epicardial vessel duris approximately 1 (mean aortic pressure [Pao] � mean pressure dia and b increase and, more importantly, flow accelerates across the na
(note the term bQ2), with resulting decline in Pd. Thus, FFR becomes muc
itative visual inspection of images in the evaluationand treatment of patients with IHD. Moreover, ac-quiring an electrocardiogram-gated image of the myo-cardium following completion of the dynamic acqui-sition for MBF measurements makes it possible toassess LV function, including LV volumes, LV ejec-tion fraction, and stroke work and power (35–37).These measurements of LV contractile function notonly are useful in following the response to therapy butalso are essential to the proper interpretation of abso-lute measurements of MBF in the case of stroke workand power.

PET measurements of absolute MBF also havebeen employed in the past to help elucidate thepathophysiology of myocardial hibernation andstunning (38–44), which are frequently present instages C and D CHF. Although it is clear that achronic state characterized by matched reduction ofMBF and contractile function (hibernation) existsin humans (41), chronic stunning (relatively pre-served MBF with impaired contraction) also isknown to occur (44), sometimes in the same pa-tients (41), and may progress to chronic hibernation(42). These states are of considerable importance inthe clinical evaluation of patients with IHD withstages C and especially D CHF. In particular,efforts to detect myocardial viability in one or morecoronary vascular territories frequently play a keyrole in clinical decision making regarding revascu-larization.

The established PET method for assessment ofmyocardial viability compares the uptake of 18F-

FR Measurement

ion dP � aQ � bQ2, for which a and b are constants related toich there is no stenosis (A), the value of each constantadenosine infusion. Accordingly, the fractional flow reserve (FFR)to the stenosis [Pd]). With increasing stenosis severity, the values ofed vessel lumen (B) such that there is a geometric pressure loss

or F

quatwhingstalrrow

h less than 1 because Pao �� Pd. Q � myocardial blood flow.

R(siormsa

ur

tlr(ocas

(

ftFrp2omp


M A R C H 2 0 1 1 : 2 9 2 – 3 0 2

Gewirtz


296

FDG with that of rest MBF (9), as represented bystatic PET 13N-ammonia or 82Rb uptake images.

egions with 18F-FDG uptake in excess of MBF“mismatch”) generally are viable and have beenhown to improve contraction following revascular-zation (6,9,41). It is important to note that theriginal description of the observation (9) andecent clinical trials (4,45) used quantitative nor-alized measurement of tracer concentration and

tandardized definitions to determine the presencend extent of 18F-FDG MBF “mismatch.” Simple

qualitative visual assessment was not used. More-over, the extent of FDG/blood flow “mismatch,”objectively measured, has been shown both to haveprognostic information and to be helpful in decisionmaking regarding coronary revascularization inischemic cardiomyopathy (3,4,6). Such informationis not available from simple visual analysis of theimages. Accordingly, to obtain optimal results,especially to define the extent of “mismatch” terri-tory, state-of-the-art clinical practice should applymethods shown to be of value in prior clinical trials.Although, in theory, Patlak analysis to determinethe metabolic rate for glucose usage combined withabsolute measurement of MBF should be superiorto the quantitative normalized method for deter-mining viability precisely because it avoids theassumption of an obligatory “normal” zone (typi-cally that with the best 13N-ammonia or 82Rbptake), this approach has not been reported andemains a direction for future research.

PET measurements of absolute MBF alone inhis setting are predictive of myocardial scar at veryow flows, �0.3 ml/min/g, which commonly cor-espond to regions of poor 18F-FDG uptake38,46,47). However, at higher levels of rest MBF,verlap between viable and nonviable regions pre-ludes exclusive use of absolute MBF for viabilityssessment (Fig. 4). Nevertheless, quantitative mea-urements of PET 82Rb washout (8) or 13N-

ammonia tracer kinetic parameters (e.g., K1/k2[46] and k3 [47]), which cannot be obtained fromsimple visual analysis of uptake images, have beenshown to enhance detection of viable myocardiumin patients with ischemic cardiomyopathy and inconjunction with absolute measurement of MBFsuggest a potential future alternative to 18F-FDGMBF “mismatch” for viability assessment.

Another PET method, measurement of myocar-dial oxygen consumption with 11C-acetate, hasbeen used to assess myocardial viability and wasshown to be superior to that of 18F-FDG in one lab
48). The method is limited, however, by the need
or an on-site cyclotron to produce the tracer andherefore has not gained widespread acceptance.inally, it should be noted that cardiac magnetic

esonance (CMR), with gadolinium contrast andulse sequence designed to optimize delayed (5 to0 min post–contrast injection) hyperenhancementf scar tissue (49), has been found to be an accurateethod for assessment of myocardial viability, com-

ares quite favorably with PET 18F-FDG method-ology (50), and is becoming increasingly used forthis purpose.

Other potential applications for PET measure-ments of absolute MBF that have not yet achievedwide clinical application but are pertinent, particu-larly to stage D CHF, often but not always ofischemic origin, include baseline and follow-upmeasurements in patients after cardiac transplant(51–54) and those with LV assist device support(55) or cardiac resynchronization therapy withbiventricular pacemaker (56). Follow-up of patientsafter cardiac transplantation represents perhaps themost useful of these potential applications, giventhe importance of post-transplant coronary arteryvasculopathy (CAV) in determining the clinicalcourse of these patients (57). Coronary angiogra-

Figure 4. Group Mean (� SD) Data Depicting the RelationshipBetween Rest MBF and Myocardial Viability as Defined byPET, ECHO, and Contractile Response to CABG Criteria

By all 3 methods, regions of chronically reduced myocardialblood flow (MBF) corresponded to viable myocardial segments.In the MBF range of � 0.3 ml/min/g, there is considerable over-lap between viable and nonviable segments. Reprinted, withpermission, from Tawakol et al. (41). CABG � coronary arterybypass graft; ECHO � echocardiography; NZ � normal zone;PET � positron emission tomography.

phy, coronary intravascular ultrasound (IVUS), or

mamwmqwaioao

h


M A R C H 2 0 1 1 : 2 9 2 – 3 0 2

Gewirtz


297

both are routinely employed annually to detectanatomic evidence of epicardial coronary stenoses ordiffuse intimal thickening of varying degree relatedto CAV (57–59). Both methods are invasive andprovide anatomic as opposed to functional informa-tion concerning the status of coronary circulation.Moreover, because of diffuse intimal thickening, theepicardial coronary vessels may appear angiographi-cally normal (58,59), whereas disease primarilyinvolving very small intramyocardial vessels includ-ing the microcirculation (51,60) cannot be visual-ized by routine angiography or IVUS.

Because the vasodilator capacity of the coronarymicrocirculation changes as a function of timepost-transplant even in the absence of epicardialstenoses (51), it is important in interpreting PETmeasurements of absolute MBF (rest and withadenosine or other direct coronary vasodilator) to becognizant of the duration post-transplant at thetime of measurement (Fig. 5). In the first yearpost-transplant, rest MBF may be elevated relativeto demand, although dilator capacity is normal (vs.healthy controls) (51). Between 1 and 3 yearspost-transplant, both rest MBF and maximal dila-tor capacity typically are normal. However, after 3years, microvascular dilator capacity declines versusnormal (51). A rough inverse correlation betweenextent of intimal thickening by IVUS (plaque vol-ume) and myocardial flow reserve assessed by PET13N-ammonia recently has been reported (52). Theloose inverse relationship clearly illustrates the well-known problem of comparing anatomic with phys-iological measurements and, more importantly,raises the question of which is better for assessmentof prognosis and evaluation of therapy aimed atpreventing or at least ameliorating CAV. Althoughthe answer is not known, it appears likely that thefunctional measurements (i.e., absolute MBF) mayprove superior because in the case of coronarymicrocirculation, abnormal vasodilator capacity re-lated to abnormal endothelial function precedesrecognizable structural abnormalities.

Stem cell and gene therapy to treat stages C andD CHF, particularly that related to IHD, is anactive research area. Whereas studies with positron-emitting reporter probes (61) have sought to docu-ment the presence, location, and fate of engraftedcells or gene expression, functional studies of thecoronary circulation in such patients have beenlimited and involved standard single-photon emis-sion computed tomography myocardial perfusion
imaging (62). However, a recent PET study of a
MBF reserve in patients treated with vascular en-dothelial growth factor has been reported (63).Given that the placebo effect is known to compli-cate interpretation of these studies, objective meth-ods to document improvement are essential notonly in the research setting but also for subsequentclinical management. Indeed, a recent pilot studyemployed both quantitative PET measurements ofMBF (13N-ammonia) and 18F-FDG for glucose

etabolism to define viability by objective criteriand obtain gated 18F-FDG images for objectiveeasurement of LV function in patients treatedith bone marrow–derived stem cells followingyocardial infarction (64). Although preliminary,

uantitative PET data identified a subset of patients inhom stem cell therapy appeared to reduce scar size

nd improve rest MBF and LV function. The study isllustrative of the role and future direction for state-f-the-art quantitative PET measurements of MBFnd metabolism in assessment of the efficacy of state-f-the-art therapy for stages C and D CHF.

Nonischemic cardiomyopathy. Obesity, diabetes, andypertension commonly coexist in a given patient

Figure 5. Group Mean (� SD) Data Depicting the TimeCourse of Changes in Rest and Minimal (Adenosine) CVRFollowing Cardiac Transplantation in Humans WithNormal Coronary Angiograms

Note that in the early post-transplant period, rest coronary vas-cular resistance (CVR) is reduced versus both normal controls(WNL) and other transplant groups (�1 year, 1–3 years, and �3years). More than 3 years following cardiac transplantation,there is evidence of progressive microvascular dysfunction witha significant increase in minimal CVR. *p � 0.05 versus othergroups. Reprinted, with permission, from Kushwaha et al. (51).Ado � adenosine; WNL � within normal limits.

nd either alone or in combination may result in

ucdiprCtihmhcoMptib


M A R C H 2 0 1 1 : 2 9 2 – 3 0 2

Gewirtz


298

DCM absent atherosclerotic coronary artery dis-ease. In addition to important coronary microvas-cular dysfunction in these patients (14,15), abnor-malities in both glucose and fatty acid metabolismare known to occur and have been reviewed in detailrecently (65–69). Although beta oxidation of freefatty acids (FFA) is the principal source of ATPproduction by cardiomyocyte mitochondria underresting conditions in the normal heart, it is impor-tant to recall that glucose also is used and that thesubstrate utilization will vary depending on themetabolic milieu and physiological conditions(65,67). Thus, with exercise, glucose usage increasesrelative to FFA beta oxidation. Moreover, it hasbeen pointed out from a metabolic view that theheart does best when using an appropriate mix ofcarbohydrate and FFA as fuels (Fig. 6) and thatexcessive shift to either glucose or FFA metabolismmay result in cardiotoxicity (67,68).

Diabetes has been noted to be an important exam-ple in which insulin resistance leads to excessive use ofFFA by the myocardium, with resulting lipotoxicity(68,69). Under such conditions, the heart’s ability tooxidize FFA (and glucose) is impaired, and conse-quently ATP production declines, with the result thatthe heart behaves as “if starved in the midst of plenty”(66–69). Lipotoxicity is associated with a number ofother maladaptive processes including accumulation oftriglycerides, reactive oxidant species, and activators ofvarious genetic transcription factors (68,69). In addi-tion, excessive FFA metabolism inhibits glucose oxi-dation and results in accumulation of intermediaryproducts of glycolysis. These products are potentiallycardiotoxic and may active genetic pathways, particu-larly fetal programs, that may become maladaptive(70), and together result in what has been termedglucolipotoxicity (69,71). Finally, it has been reportedthat more advanced stages of CHF (C and D) may

Figure 6. Schematic Representation of Myocardial Use of FA an

Under physiological conditions, the myocardium employs a mixturetion. Excessive reliance on one or the other is maladaptive and lead
as glucolipotoxicity (69).
themselves cause myocardial insulin resistance and there-fore result in a vicious cycle in which CHF impairsintermediate energy metabolism, which in turn exacer-bates CHF (70,72).

Currently, there is great research potential forPET imaging of cardiac metabolic pathways andgene programs involved in the basic mechanismsunderlying CHF of ischemic and nonischemic eti-ologies at all stages (66,73). Clinical applications,unfortunately, are limited by a number of factors,including requirement for an on-site cyclotron andradiochemist to manufacture 11C-labeled metabolicintermediates and complexity of the studies thatrange from patient dietary preparation to advancedtracer kinetic models needed for quantitative re-sults. It may be possible in the future to minimizemany of these issues with a previously reported18F-labeled fatty acid tracer (74).

One quantitative study with 18F-FDG in humansnder conditions of a hyperinsulinemic, euglycemiclamp demonstrated that young patients with insulin-ependent diabetes of �5 years’ duration had similar

nsulin-stimulated myocardial glucose uptake com-ared with healthy controls (75). Further, given cur-ent interest in cardiac metabolism as a target forHF therapy at various stages (67,70,76), the ability

o quantitatively measure glucose use with 18F-FDGn humans (Fig. 7) under well-defined metabolic andemodynamic conditions may prove helpful in deter-ining if a particular therapy (drug or device) en-

ances myocardial glucose usage, which is more effi-ient in comparison with FFA beta oxidation in termsf oxygen consumption for production of ATP (70).oreover, the ability to gate the FDG acquisition

ermits near simultaneous assessment of LV contrac-ile function (64,77) and therefore provides directnformation on the therapeutic efficacy of the meta-olic intervention. Accordingly, quantitative PET

O in the Human Heart

fatty acids (FA) and carbohydrates (CHO) to fuel energy produc-lipotoxicity (FA), glucotoxicity (CHO), or a combined state known

d CH

ofs to


M A R C H 2 0 1 1 : 2 9 2 – 3 0 2

Gewirtz


299

measurements of cardiac glucose metabolism andcontractile function are readily available in the clinicalsetting and are likely to play a more important role inpatient management as metabolic interventions be-come more widely applied and the need to demon-strate efficacy increases in a socioeconomic environ-ment that demands the optimal use of scarce healthcare resources.

Further, it may be anticipated that the advent of18F-labeled fatty acid and sympathetic nerve (78)tracers will substantially expand the field of cardiacmetabolic and autonomic nervous system imaging forclinical use in patients with CHF, especially whencombined with quantitative measurements of glucosemetabolism, LV function. and MBF. Lastly, quanti-tative PET imaging of 18F-FDG in humans has beenreported useful in distinguishing idiopathic DCMfrom that owing to isolated cardiac sarcoidosis and forassessing both disease activity and response to therapy(79). The quantitative methodology employed re-quired determination of standard uptake values of18F-FDG in a standard 17-segment model, alongwith means, SDs, and coefficients of variation, all ofwhich are not possible with simple qualitative visualanalysis. The last parameter (coefficients of variation)was essential to objectively defining “heterogeneous”tracer uptake and best separated patients with cardiacsarcoidosis from those with DCM, as well as controlsand patients with sarcoidosis but without cardiacinvolvement.

Conclusions

PET quantitative measurements of MBF and metab-olism provide state-of-the-art methodology for eval-uation and management of patients at all stages ofCHF. In stages A and B, quantitative measurementsof absolute MBF permits assessment of coronarymicrovascular function, which are impossible to obtainwith simple visual analysis of uptake images and whichhave been shown to provide prognostic information(5,7) and information on response to therapy (32). Inthe more advanced stages of CHF (C and D), quan-titative PET measurements of MBF and glucosemetabolism are of proven value in assessment ofmyocardial viability (4,9,45), prognosis (6), and selec-tion of patients for coronary revascularization (3).Applications that are immediately and readily avail-able but not yet widely employed include absolutemeasurement of MBF for detection of CAV (51), akey prognostic factor, in patients with cardiac trans-
plants; quantitative measurement of the metabolic rate
for glucose usage and LV function for assessment ofresponse to metabolic therapy of stages C and DCHF; objective treatment responses to device, gene,or stem cell therapy (64), as defined by absolute MBF,quantitative glucose metabolism, and LV function;and quantitative 18F-FDG uptake to improve thediagnosis of at least one treatable cause of cardiomy-opathy, sarcoidosis (79). Finally, it is important toemphasize that: 1) quantitative measurements ofMBF in the stenosis setting are linked to basic fluiddynamic principles (80–82) which are the basis forthe invasive fractional flow reserve measurement;and 2) quantitative measurements of the metabolicrate for glucose usage depend on the value assignedto the “lumped constant” which in fact is notconstant and requires correction depending onphysiological condition (83).

Future applications will depend on availability of

Figure 7. Typical LV Blood Pool and Myocardial TAC From PETDynamic Data Acquisition

Shown here are typical LV blood pool and myocardial time activity(TAC) (upper panel) from a PET 18F-fluorodeoxyglucose (FDG) dynaacquisition. The lower panel shows the resulting Patlak plot for theThe slope of the line, Ki, is the uptake rate constant for the tracer amultiplied by measured plasma glucose concentration divided by tlumped constant (LC) gives the metabolic rate for glucose usage (m(�mol/min/100 g). Note that the plot can be implemented with anspreadsheet, and the LC is not constant over all physiological condmethod for correcting it based on the myocardial FDG TAC has beeposed (83). Other abbreviations as in Figures 1 and 4.

18F-FDG

curvesmic datadata.nd whenheGLU)ordinaryitions; an pro-

18F-labeled tracers for widespread use and include

Cp


M A R C H 2 0 1 1 : 2 9 2 – 3 0 2

Gewirtz


300

assessment of myocardial fatty acid metabolism andautonomic nervous system function, both of which areimportant clinically with respect to risk stratificationand evaluation of innovative, emerging therapies forCHF. As illustrated earlier, optimal use of PET forthese applications will require state-of-the-art quanti-tative methods to provide objective, measureable pa-

myocardial cell membrane integrity

1

1

1

1

1

1

1

1

1

2008;118:968–76.

mated, and followed over time, and are independentof and inherently superior to even a skilled reader’ssubjective impression of a static PET image.

Reprint requests and correspondence: Dr. Henry Gewirtz,ardiac Unit/Yawkey 5E, Massachusetts General Hos-ital, Boston, Massachusetts 02114. E-mail: hgewirtz@

rameters that can be accurately reproduced, auto- partners.org.

1

2

2

2

2

2

2

2

2

2

R E F E R E N C E S

1. Jessup M, Abraham WT, Casey DE,et al. 2009 focused update: ACCF/AHA guidelines for the diagnosis andmanagement of heart failure in adults:a report of the American College ofCardiology Foundation/AmericanHeart Association Task Force onPractice Guidelines: developed in col-laboration with the International So-ciety for Heart and Lung Transplan-tation. J Am Coll Cardiol 2009;53:1343–82.

2. Kannel WB. Incidence and epidemi-ology of heart failure. Heart Fail Rev2000;5:167–73.

3. Abraham A, Nichol G, Williams KA,et al. 18F-FDG PET imaging of myo-cardial viability in an experienced cen-ter with access to 18F-FDG and inte-gration with clinical managementteams: the Ottawa-FIVE substudy ofthe PARR 2 trial. J Nucl Med 2010;51:567–74.

4. Beanlands RS, Nichol G, Huszti E, etal. F-18-fluorodeoxyglucose positronemission tomography imaging-assistedmanagement of patients with severe leftventricular dysfunction and suspectedcoronary disease: a randomized, con-trolled trial (PARR-2). J Am Coll Car-diol 2007;50:2002–12.

5. Cecchi F, Olivotto I, Gistri R, et al.Coronary microvascular dysfunctionand prognosis in hypertrophic cardio-myopathy. N Engl J Med 2003;349:1027–35.

6. Maddahi J, Schelbert H, Brunken R,Di Carli M. Role of thallium-201 andPET imaging in evaluation of myo-cardial viability and management ofpatients with coronary artery diseaseand left ventricular dysfunction.J Nucl Med 1994;35:707–15.

7. Neglia D, Michelassi C, Trivieri MG,et al. Prognostic role of myocardialblood flow impairment in idiopathicleft ventricular dysfunction. Circula-tion 2002;105:186–93.

8. vom Dahl J, Muzik O, Wolfe ER Jr.,et al. Myocardial rubidium-82 tissuekinetics assessed by dynamic positronemission tomography as a marker of

and viability. Circulation 1996;93:238–45.

9. Tillisch J, Brunken R, Marshall R, etal. Reversibility of cardiac wall-motion abnormalities predicted bypositron tomography. N Engl J Med1986;314:884–8.

0. Fallavollita JA, Banas MD, Suzuki G,et al. 11C-meta-hydroxyephedrine de-fects persist despite functional im-provement in hibernating myocar-dium. J Nucl Cardiol 2010;17:85–96.

1. Fallavollita JA, Canty JM Jr. Dysin-nervated but viable myocardium inischemic heart disease. J Nucl Cardiol2010;17:1107–15.

2. Fallavollita JA, Luisi AJ Jr., MichalekSM, et al. Prediction of arrhythmicevents with positron emission tomog-raphy: PAREPET study design andmethods. Contemp Clin Trials 2006;27:374–88.

3. Fallavollita JA, Luisi AJ Jr., Yun E,deKemp RA, Canty JM Jr. An abbre-viated hyperinsulinemic-euglycemicclamp results in similar myocardialglucose utilization in both diabeticand non-diabetic patients with isch-emic cardiomyopathy. J Nucl Cardiol2010;17:637–45.

4. Camici PG, Crea F. Coronary micro-vascular dysfunction. N Engl J Med2007;356:830–40.

5. Schindler TH, Schelbert HR, Quer-cioli A, Dilsizian V. Cardiac PETimaging for the detection and moni-toring of coronary artery disease andmicrovascular health. J Am Coll Car-diol Img 2010;3:623–40.

6. Gould KL. Does coronary flow trumpcoronary anatomy? J Am Coll CardiolImg 2009;2:1009–23.

7. Dayanikli F, Grambow D, Muzik O,et al. Early detection of abnormalcoronary flow reserve in asymptom-atic men at high risk for coronaryartery disease using positron emis-sion tomography. Circulation 1994;90:808 –17.

8. Levy BI, Schiffrin EL, Mourad JJ, etal. Impaired tissue perfusion: a pathol-ogy common to hypertension, obesity,and diabetes mellitus. Circulation

9. Loscalzo J. Oxidant stress: a key determi-nant of atherothrombosis. Biochem SocTrans 2003;31(Pt 5):1059–61.

0. Quyyumi AA, Dakak N, Andrews NP, etal. Contribution of nitric oxide to meta-bolic coronary vasodilation in the humanheart. Circulation 1995;92:320–6.

1. Treasure C, Klein L, Vita J, et al.Hypertension and left ventricular hy-pertrophy are associated with im-paired endothelium—mediated relax-ation in human coronary resistancevessels. Circulation 1993;87:86–93.

2. Hamasaki S, Al Suwaidi J, HiganoST, et al. Attenuated coronary flowreserve and vascular remodeling inpatients with hypertension and leftventricular hypertrophy. J Am CollCardiol 2000;35:1654–60.

3. Parris R, Webb D. The endothelinsystem in cardiovascular physiologyand pathophysiology. Vasc Med 1997;2:31–43.

4. Suzuki T, Katz R, Jenny NS, et al.Metabolic syndrome, inflammation,and incident heart failure in the el-derly: the cardiovascular health study.Circ Heart Fail 2008;1:242–8.

5. Suzuki T, Hirata K, Elkind MS, et al.Metabolic syndrome, endothelial dys-function, and risk of cardiovascularevents: the Northern ManhattanStudy (NOMAS). Am Heart J 2008;156:405–10.

6. Bahrami H, Bluemke DA, Kronmal R, etal. Novel metabolic risk factors for inci-dent heart failure and their relationshipwith obesity: the MESA (Multi-EthnicStudy of Atherosclerosis) study. J AmColl Cardiol 2008;51:1775–83.

7. Olivotto I, Cecchi F, Gistri R, et al.Relevance of coronary microvascularflow impairment to long-term remod-eling and systolic dysfunction in hy-pertrophic cardiomyopathy. J AmColl Cardiol 2006;47:1043–8.

8. Herzog BA, Husmann L, Valenta I,et al. Long-term prognostic value of13N-ammonia myocardial perfusionpositron emission tomography addedvalue of coronary flow reserve. J Am
Coll Cardiol 2009;54:150–6.
mailto:[email protected]

mailto:[email protected]

3

3

3

3

3

4

4

4

4

4

4

4

4

4

4


M A R C H 2 0 1 1 : 2 9 2 – 3 0 2

Gewirtz


301

29. El Fakhri G, Kardan A, Sitek A, et al.Reproducibility and accuracy of quan-titative myocardial blood flow assess-ment with 82Rb PET: comparisonwith 13N-ammonia PET. J Nucl Med2009;50:1062–71.

0. Nekolla SG, Reder S, Saraste A, et al.Evaluation of the novel myocardial per-fusion positron-emission tomographytracer 18F-BMS-747158-02: compari-son to 13N-ammonia and validationwith microspheres in a pig model. Cir-culation 2009;119:2333–42.

1. Shoup TM, Elmaleh DR, Brownell AL,et al. Evaluation of (4-[18F]fluoro-phenyl)triphenylphosphonium ion. A po-tential myocardial blood flow agent forPET. Mol Imaging Biol 2010 Jun 19[Epub ahead of print].

32. Huggins GS, Pasternak RC, AlpertNM, Fischman AJ, Gewirtz H. Ef-fects of short-term treatment of hy-perlipidemia on coronary vasodilatorfunction and myocardial perfusion inregions having substantial impairmentof baseline dilator reverse. Circulation1998;98:1291–6.

33. Hajjiri MM, Leavitt MB, Zheng H,et al. Comparison of positron emis-sion tomography measurement ofadenosine-stimulated absolute myo-cardial blood flow versus relative myo-cardial tracer content for physiologicalassessment of coronary artery stenosisseverity and location. J Am Coll Car-diol Img 2009;2:751–8.

34. Tonino PA, De Bruyne B, Pijls NH,et al. Fractional flow reserve versusangiography for guiding percutaneouscoronary intervention. N Engl J Med2009;360:213–24.

35. Porenta G, Cherry S, Czernin J, et al.Noninvasive determination of myo-cardial blood flow, oxygen consump-tion and efficiency in normal humansby carbon-11 acetate positron emis-sion tomography imaging. Eur J NuclMed 1999;26:1465–74.

36. Khorsand A, Graf S, Eidherr H, et al.Gated cardiac 13N-NH3 PET for as-sessment of left ventricular volumes,mass, and ejection fraction: compari-son with electrocardiography-gated18F-FDG PET. J Nucl Med 2005;46:2009–13.

7. Gregory SA, MacRae CA, Aziz K, etal. Myocardial blood flow and oxygenconsumption in patients with Fried-reich’s ataxia prior to the onset ofcardiomyopathy. Coron Artery Dis2007;18:15–22.

8. Gewirtz H, Fischman AJ, AbrahamSA, et al. Positron emission tomo-graphic measurements of absolute re-gional myocardial blood flow permitsidentification of nonviable myocar-dium in patients with chronic myocar-dial infarction. J Am Coll Cardiol
1994;23:851–9.
9. Berman M, Fischman AJ, Southern J,et al. Myocardial adaptation duringand after sustained, demand-inducedischemia. Observations in closed-chest, domestic swine. Circulation1996;94:755–62.

0. Holmvang G, Fry S, Skopicki HA, etal. Relation between coronary “steal”and contractile function at rest incollateral-dependent myocardium ofhumans with ischemic heart disease.Circulation 1999;99:2510–6.

1. Tawakol A, Skopicki HA, AbrahamSA, et al. Evidence of reduced restingblood flow in viable myocardial re-gions with chronic asynergy. J AmColl Cardiol 2000;36:2146–53.

2. Heusch G, Schulz R, Rahimtoola SH.Myocardial hibernation: a delicatebalance. Am J Physiol Heart CircPhysiol 2005;288:H984–99.

3. Fallavollita J, Canty JJ. Differential18-F-2-deoxyglucose uptake in viabledysfunctional myocardium with nor-mal resting perfusion: evidence forchronic stunning in pigs. Circulation1999;99:2798–805.

4. Gerber BL, Vanoverschelde JJ, Bol A,et al. Myocardial blood flow, glucoseuptake, and recruitment of inotropicreserve in chronic left ventricular dys-function. Implications for the patho-physiology of chronic myocardial hi-bernation. Circulation 1996;94:651–9.

5. Beanlands RS, Ruddy TD, deKempRA, et al. Positron emission tomog-raphy and recovery following revascu-larization (PARR-1): the importanceof scar and the development of aprediction rule for the degree of recov-ery of left ventricular function. J AmColl Cardiol 2002;40:1735–43.

6. Beanlands RS, deKemp R, Scheffel A,et al. Can nitrogen-13 ammonia ki-netic modeling define myocardial via-bility independent of fluorine-18 fluo-rodeoxyglucose? J Am Coll Cardiol1997;29:537–43.

7. Kitsiou AN, Bacharach SL, BartlettML, et al. 13N-ammonia myocardialblood flow and uptake: relation tofunctional outcome of asynergic re-gions after revascularization. J AmColl Cardiol 1999;33:678–86.

8. Gropler RJ, Geltman EM, Sampath-kumaran K, et al. Comparison ofcarbon-11-acetate with fluorine-18-fluorodeoxyglucose for delineating vi-able myocardium by positron emissiontomography. J Am Coll Cardiol 1993;22:1587–97.

9. Simonetti OP, Kim RJ, Fieno DS, etal. An improved MR imaging tech-nique for the visualization of myocar-dial infarction. Radiology 2001;218:
215–23.
50. Kuhl HP, Beek AM, van der WeerdtAP, et al. Myocardial viability inchronic ischemic heart disease: com-parison of contrast-enhanced mag-netic resonance imaging with 18F-fluorodeoxyglucose positron emissiontomography. J Am Coll Cardiol 2003;41:1341–8.

51. Kushwaha SS, Narula J, Narula N,et al. Pattern of changes over time inmyocardial blood flow and microvas-cular dilator capacity in patientswith normally functioning cardiacallografts. Am J Cardiol 1998;82:1377– 81.

52. Wu YW, Chen YH, Wang SS, et al.PET assessment of myocardial perfu-sion reserve inversely correlates withintravascular ultrasound findings inangiographically normal cardiac trans-plant recipients. J Nucl Med 2010;51:906–12.

53. Preumont N, Berkenboom G, VachieryJ, et al. Early alterations of myocardialblood flow reserve in heart transplantrecipients with angiographically normalcoronary arteries. J Heart Lung Trans-plant 2000;19:538–45.

54. Preumont N, Lenaers A, Goldman S,et al. Coronary vasomotility and myo-cardial blood flow early after hearttransplantation. Am J Cardiol 1996;78:550–4.

55. Xydas S, Rosen RS, Pinney S, et al.Reduced myocardial blood flow dur-ing left ventricular assist device sup-port: a possible cause of prematurebypass graft closure. J Heart LungTransplant 2005;24:1976–9.

56. Knaapen P, van Campen LM, deCock CC, et al. Effects of cardiacresynchronization therapy on myocar-dial perfusion reserve. Circulation2004;110:646–51.

57. Konig A, Kilian E, Rieber J, et al.Assessment of early atherosclerosisin de novo heart transplant recipi-ents: analysis with intravascularultrasound-derived radiofrequencyanalysis. J Heart Lung Transplant2008;27:26 –30.

58. St. Goar FG, Pinto FJ, Alderman EL,et al. Intracoronary ultrasound in car-diac transplant recipients. In vivo evi-dence of “angiographically silent” in-timal thickening. Circulation 1992;85:979–87.

59. Johnson DE, Alderman EL, Schroe-der JS, et al. Transplant coronary ar-tery disease: histopathologic correla-tions with angiographic morphology.J Am Coll Cardiol 1991;17:449–57.

60. Treasure CB, Vita JA, Ganz P, et al.Loss of the coronary microvascularresponse to acetylcholine in cardiactransplant patients. Circulation 1992;
86:1156–64.

7

7

7

7

7

7

7

7

7

7

8

8

8

8


M A R C H 2 0 1 1 : 2 9 2 – 3 0 2

Gewirtz


302

61. Inubushi M, Tamaki N. Positronemission tomography reporter geneimaging in the myocardium: for mon-itoring of angiogenic gene therapy inischemic heart disease. J Card Surg2005;20:S20–4.

62. Hendel RC, Henry TD, Rocha-SinghK, et al. Effect of intracoronary re-combinant human vascular endothe-lial growth factor on myocardial per-fusion: evidence for a dose-dependenteffect. Circulation 2000;101:118–21.

63. Tio RA, Tan ES, Jessurun GA, et al.PET for evaluation of differentialmyocardial perfusion dynamics afterVEGF gene therapy and laser therapyin end-stage coronary artery disease.J Nucl Med 2004;45:1437–43.

64. Castellani M, Colombo A, GiordanoR, et al. The role of PET with 13N-ammonia and 18F-FDG in the assess-ment of myocardial perfusion and me-tabolism in patients with recent AMIand intracoronary stem cell injection.J Nucl Med 2010;51:1908–16.

65. Taegtmeyer H. Switching metabolicgenes to build a better heart. Circula-tion 2002;106:2043–5.

66. Taegtmeyer H. Tracing cardiac me-tabolism in vivo: one substrate at atime. J Nucl Med 2010;51 Suppl1:80S–7S.

67. Taegtmeyer H, Ballal K. No low-fatdiet for the failing heart? Circulation2006;114:2092–3.

68. Taegtmeyer H, McNulty P, YoungME. Adaptation and maladaptationof the heart in diabetes: part I:general concepts. Circulation 2002;105:1727–33.

69. Young ME, McNulty P, TaegtmeyerH. Adaptation and maladaptation of

the heart in diabetes: part II: potentialmechanisms. Circulation 2002;105:1861–70.

0. Taegtmeyer H. Cardiac metabolism asa target for the treatment of heartfailure. Circulation 2004;110:894–6.

1. Taegtmeyer H. Glucose for the heart:too much of a good thing? J Am CollCardiol 2005;46:49–50.

2. Nikolaidis LA, Sturzu A, Stolarski C,et al. The development of myocardialinsulin resistance in conscious dogswith advanced dilated cardiomyopa-thy. Cardiovasc Res 2004;61:297–306.

3. Gropler RJ, Beanlands RS, DilsizianV, et al. Imaging myocardial meta-bolic remodeling. J Nucl Med 2010;51Suppl 1:88S–101S.

4. Shoup TM, Elmaleh DR, Bonab AA,Fischman AJ. Evaluation of trans-9-18F-fluoro-3,4-methyleneheptadecanoicacid as a PET tracer for myocardial fattyacid imaging. J Nucl Med 2005;46:297–304.

5. vom Dahl J, Herman WH, Hicks RJ,et al. Myocardial glucose uptake inpatients with insulin-dependent dia-betes mellitus assessed quantitativelyby dynamic positron emission tomog-raphy. Circulation 1993;88:395–404.

6. Nikolaidis LA, Elahi D, Hentosz T,et al. Recombinant glucagon-likepeptide-1 increases myocardial glu-cose uptake and improves left ven-tricular performance in consciousdogs with pacing-induced dilatedcardiomyopathy. Circulation 2004;110:955– 61.

7. Slart RH, Bax JJ, de Jong RM, et al.
Comparison of gated PET with MRIfor evaluation of left ventricular func-
tion in patients with coronary arterydisease. J Nucl Med 2004;45:176–82.

8. Mistry M, Kagan M, Lazewatsky J, etal. Dosimetry in nonhuman primatesof [18F]LMI1195, a novel PET tracerfor imaging the cardiac sympatheticnervous system. J Nucl Med 2010;51Suppl 2:1447 (abstr).

9. Tahara N, Tahara A, Nitta Y, et al.Heterogeneous myocardial FDG up-take and the disease activity in cardiacsarcoidosis. J Am Coll Cardiol Img2010;3:1219–28.

0. Gould KL. Pressure-flow character-istics of coronary stenoses in unse-dated dogs at rest and during coro-nary vasodilation. Circ Res 1978;43:242–53.

1. Sun Y, Most AS, Ohley W, GewirtzH. Estimation of instantaneous bloodflow through a rigid coronary arterystenosis in anesthetized domesticswine. Cardiovasc Res 1983;17:499–504.

2. Young DF, Cholvin NR, Roth AC.Pressure drop across artificially in-duced stenosis in the femoral artery ofdogs. Circ Res 1975;36:735–43.

3. Botker HE, Goodwin GW, HoldenJE, et al. Myocardial glucose uptakemeasured with fluorodeoxyglucose: aproposed method to account for vari-able lumped constants. J Nucl Med1999;40:1186–96.

Key Words: heart failure ymyocardial blood flow y PET
imaging.

state-of-the-art paper …imaging.onlinejacc.org/content/jimg/4/3/292.full.pdf ·...

Documents