Journal of Cardiac Failure Vol. 19 No. 5 2013

Distinctive Clinical Characteristics According to Ageand Gender in Apical Ballooning Syndrome (Takotsubo/Stress

Cardiomyopathy): An Analysis Focusing on Menand Young Women

SANDEEP M. PATEL, MD,1 RAMESH G. CHOKKA, MD,2 KAVITA PRASAD, MD,3 AND ABHIRAM PRASAD, MD, FRCP, FACC2

Pittsburgh, Pennsylvania; and Rochester, Minnesota

From the 1HeMedical Center,Diseases, Mayo CMedicine, Mayo C

Manuscript recMarch 15, 2013; r

Reprint requesMedicine, Mayo55905. Tel: 507-5mayo.eduSee page 310 fo1071-9164/$ - s� 2013 Elseviehttp://dx.doi.org

ABSTRACT

Background: Apical ballooning syndrome (ABS) predominantly affects postmenopausal women. Thereis a paucity of data regarding ABS in men and young women. The aim of this study was to compare theclinical characteristics and outcomes of men and young women (!50 y) to older women ($50 y).Methods & Results: We retrospectively reviewed the records of 224 patients and divided them into men(n5 12), young women (n5 12), and older women (n5 200). Older women were further subdivided intothose who were and were not on hormone replacement therapy (HRT) at the time of presentation. Menwere more likely to present after a physical trigger (100% vs 46%; P 5 .009), have lower ejection frac-tions (30.1 6 8.0% vs 40 6 13.9%; P 5 .04), and have greater need for mechanical ventilation (67% vs17%; P ! .0001) compared with older women. Younger women were more likely to have a history ofpsychiatric disorders (75% vs 24%; P 5 .0001) at presentation and a higher rate of recurrence (16% vs3%; P 5 .017) compared with older women. Of the older women, 15 developed ABS while on chronicHRT. Those without HRT were more likely to require mechanical hemodynamic (7.7% and 0%; P 5 .002)and ventilatory (18.1% and 0%; P 5 .017) support compared with older women who were on HRT.Conclusions: Men appeared to develop ABS as a consequence of a physical trigger, whereas youngwomen had a higher rate of psychiatric comorbidities and a greater propensity for recurrence. Treatmentwith HRT in older women does not preclude the development of ABS. (J Cardiac Fail 2013;19:306e310)Key Words: Apical ballooning syndrome, takotsubo cardiomyopathy, stress cardiomyopathy.

Apical ballooning syndrome (ABS) is a unique cardiomy-opathy primarily characterized by depressed left ventricularsystolic function, typically involving the midventricularand apical segments, and it frequently occurs followinga physical or emotional stressor.1,2 Commonly referred toas stress or takotsubo cardiomyopathy, the condition gener-ally occurs in patients without obstructive epicardial

art and Vascular Institute, University of PittsburghPittsburgh, Pennsylvania; 2Division of Cardiovascularlinic, Rochester, Minnesota and 3Department of Internallinic, Rochester, Minnesota.eived January 9, 2013; revised manuscript receivedevised manuscript accepted March 19, 2013.ts: Abhiram Prasad, MD, FRCP, FACC Professor ofClinic, 200 First Street SW , Rochester, Minnesota38-6325; Fax: 507-255 2550. E-mail: prasad.abhiram@

r disclosure information.ee front matterr Inc. All rights reserved./10.1016/j.cardfail.2013.03.007

306

coronary atherosclerosis, and by and large outcomes arecharacterized by improvement in left ventricular systolicfunction over time. As ABS has gained recognition asa clinical entity, it has become evident that, although it pre-dominantly occurs in postmenopausal women, it may occa-sionally also be diagnosed in men and young women.3

There is a paucity of data regarding these infrequent but im-portant subsets of patients. Therefore, the aim of the presentstudywas to describe the clinical characteristics andoutcomesof men and young women and compare them with the typicalolder (‘‘postmenopausal’’) women who are at risk for ABS.

Methods

We conducted a retrospective analysis of the electronic medicalrecords of patients who were prospectively identified with ABSduring the time period from January 2002 to January 2012 atthe Mayo Clinic. The diagnosis of ABS was based on the MayoClinic criteria2: 1) transient hypokinesis, akinesis, or dyskinesisof the left ventricular midsegments with or without apical

Age and Gender in Apical Ballooning Syndrome � Patel et al 307

involvement, the regional wall motion abnormalities extending be-yond a single epicardial vascular distribution, and a stressful trig-ger often, but not always, present; 2) absence of obstructivecoronary disease or angiographic evidence of acute plaque rup-ture; 3) new electrocardiographic abnormalities (either ST-segment elevation and/or T-wave inversion) or modest elevationin cardiac troponin; and 4) clinical absence of pheochromocytomaor myocarditis. Two-hundred twenty-four cases who met these cri-teria were divided into 3 groups: men (n5 12), women!50 yearsold (n 5 12), and women $50 years old (n 5 200). The women$50 years were further subcategorized into patients who were onsystemic hormone replacement therapy (HRT) for O1 year at thetime of the diagnosis (n5 15). The clinical, laboratory, electrocar-diographic (ECG), and imaging data were obtained from the med-ical records. All subjects consented to the use of their medicalrecords for research purposes, and the Mayo Clinic InstitutionalReview Board approved the study.All continuous variables which demonstrated near-normal dis-

tribution are summarized as mean 6 SD and were analyzedwith the use of a 1-way analysis of variance. The between-groupcomparisons were tested with the use of a Student 2-samplet test. For continuous variables that did not demonstrate symmetricdistribution, data are expressed as median and interquartile rangeand compared with the use of the Wilcoxon rank sum test. Cate-goric variables are represented as a frequency and tested withthe use of the Pearson chi-square test. Kaplan-Maier analyseswere done to compare recurrence-free survivals among the threecohorts. A P value of !.05 was considered to be statistically sig-nificant for comparison of men versus women. In Table 1, owing tomultiple comparisons, P ! .025 was considered to be statisticallysignificant. Statistical analyses were performed with the use ofJMP 9.0.

Results

Men Compared With Women $50 Years

Table 1 summarizes the baseline clinical characteristics,ECG findings, biomarker levels, echocardiographic fea-tures, left ventriculogram data, and clinical outcomes.There was no difference in the age and frequency of cardio-vascular risk factors between the groups. Men were morelikely to present after a physical trigger (100%) comparedwith women (46.5%; P 5 .009; Fig. 1). Physical triggersin males included acute exacerbation of chronic obstructivepulmonary disease (n5 5), invasive procedures (liver trans-plant, exploratory laparatomy, hepatic artery embolization,and bronchoscopy), dobutamine stress echocardiography,alcoholic pancreatitis with withdrawal symptoms and shov-eling snow. There was no difference regarding history ofpsychiatric diseases, connective tissue/autoimmune disor-ders, irritable bowel syndrome, or fibromyalgia. ECG andcardiac biomarker profiles were similar between the groupsexcept for peak B-type natriuretic peptide levels, whichwere significantly elevated in older women comparedwith men (P 5 .0136). Echocardiographically determinedejection fraction at admission was lower in men (P 5.02). Linear regression analysis of initial ejection fractionby echocardiography correlated positively with age (r 50.75; P 5 .009). A similar trend was observed with the

ejection fraction derived by left ventriculography whichwas associated with significantly elevated end-systolicand -diastolic volumes compared with women. Men weremore likely to require mechanical ventilation followingthe diagnosis of ABS compared with women (P ! .0001).

Young Versus Older Women

Older females were more likely to have hypertensionand coronary artery disease at presentation (P values .015and .004, respectively; Table 1). Younger women weremore likely to have a history of psychiatric disorders(75% vs 24%; P 5 .0001), such as anxiety (17% vs 8.0%;P 5 .30) and depression (75% vs 17%; P !.0001), and mi-graines (25% vs 4%; P 5 .0006) at presentation, but therewas no difference in the frequency of Raynaud phenome-non, connective tissue/autoimmune diseases, irritable bowelsyndrome, or fibromyalgia between the groups. There wereno significant differences regarding findings on the ECG,biomarkers, echocardiography, and left ventriculography.

Older Women and Hormone Replacement Therapy

Diabetes mellitus was more prevalent in patients withoutHRT compared with those with HRT. The ECG, biomarker,echocardiographic, and left ventriculographic data weresimilar between these groups. Patients without HRT hada statistically non-significant trend toward lower ejectionfractions (39.7% and 45.4%; P 5 .19), and they weremore likely to require mechanical hemodynamic (7.7%and 0%; P 5 .002) and ventilatory (18.1% and 0%; P 5.017) support.

Clinical Outcomes, Mortality, and Recurrence

The ejection fraction at follow-up (2.2 6 3.7 mo) in-creased to a normal level in each of the 3 groups (Fig. 2).There were no deaths in women !50 years (meanfollow-up 3.2 y). The in-hospital mortality in men was17% (n 5 2), and 3 additional deaths occurred duringfollow-up. All deaths were due to noncardiac causes.The in-hospital mortality in women $50 years was 6%(n 5 11), and an additional 40 patients died duringfollow-up. Of those 51 deaths, 11 were due to cardiaccauses. There was no significant difference betweenmen and women $50 years regarding cause of death(P 5 .36) and mortality rates (P 5 .35). None of the menhad a recurrence of ABS during a mean follow-up durationof 1.3 years (Fig. 3). Of the women !50 years, 2 (16%)had a recurrence over a mean follow-up of 3.5 years. Inthe women $50 years, 5 (3%) developed a recurrence ofABS over a mean follow-up duration of 3.3 years.

Discussion

Our study provides the most comprehensive evaluationof clinical phenotype of men and young women diagnosedwith ABS to date, compared with women $50 years old(‘‘the typical patient’’). The major findings are: 1) Male

Table 1. Baseline Clinical Characteristics

Variable

ABS Cohort

P ValueMen (n 5 12) Women #50 y (n 5 12) Women $50 y (n 5 200)

Patient characteristicsAge, y 65.5 6 11.2 44.9 6 3.8 71.7 6 10.4 dBody surface area, m2y 1.8 6 0.3 1.9 6 0.2* 1.7 6 0.2 .018Hypertension 8 (66.7%) 5 (41.7%)* 146 (73.7%) .05Diabetes mellitus 1 (8.3%) 1 (8.3%) 29 (14.7%) .7Hyperlipidemia 6 (50%) 4 (33.3%) 87 (43.9%) .69Smoking 6 (50%) 5 (41.7%) 50 (26.3%) .12Coronary artery disease 9 (75%) 5 (41.7%)* 145 (78.4%) .02Coronary artery disease O50% 3 (25%) 2 (16.7%) 41 (22.2%) .88Atrial fibrillation 1 (8.3%) 0 (0%) 35 (17.7%) .2In-hospital presentation 5 (41.7%) 5 (41.7%) 81 (40.5%) .99

ECG findingsST-Segment elevation (STE) 5 (41.7%) 5 (41.7%) 91 (46.4%) .91Deep T-wave inversion (TWI) 3 (25%) 6 (50%) 94 (48.2%) .62STE or TWI 8 (66.7%) 10 (83.3%) 146 (73%) .64Maximum corrected QT interval, 476.5 (458.8e541.3) 555.5 (473.8e591.8) 494 (462e533) .18

BiomarkersPeak troponin T (normal !0.01), ng/mL 0.23 (0.11e0.50) 0.29 (0.12e1) 0.49 (0.16e0.94) .14Peak CK-MB, ng/mL 8.1 (7.9e21.6) 19.3 (8.3e36.4) 11.65 (7.3e20.1) .49Peak BNP, pg/mL 73 (19, 233)* 241 (62, 869.5) 705 (359, 1258) .01

Initial echocardiographic findingsEjection fraction, % 30 (25e35)* 31 (27e40) 37 (30e50) .030Right ventricular systolic pressure, mm Hg 45.9 6 11.8 42.4 6 8.2 43.2 6 12.8 .84Mitral regurgitation present 6 (60%) 7 (70%) 109 (63.7%) .89Wall motion score index 2.0 6 0.26 2.0 6 0.44 1.9 6 0.43 .42

Left ventriculogramEjection fraction, % 38 6 7.8 42.6 6 7.5 43.3 6 13.4 .57End-systolic volume, mL 110.1 6 27.1* 82.4 6 26 70.9 6 30.5 .003End-systolic volume/BSA, mL/m2 64.8 6 10.0* 39.9 6 14.2 47.6 6 18.4 .001End-diastolic volume, mL 174.5 6 38.9* 149.7 6 31.3 126 6 39.1 .002End-diastolic volume/BSA, mL/m2 101.3 6 13.2* 85.2 6 21.0 69.8 6 18.9 !.001End-diastolic pressure, mm Hg 26.5 6 8.6 21 6 9.3 24.3 6 6.8 .24

Clinical courseLength of hospitalization (d) 6 (2e10.8) 4 (3e8) 5 (3e8) .72Length of ICU course (d) 3 (0.25e6) 1 (1e1) 1 (1e3) .35Acute heart failure 4 (33.3%) 3 (25.0%) 71 (36.0%) .73Intra-aortic balloon pump 2 (16.7%) 1 (8.3%) 15 (7.6%) .54Mechanical ventilation 8 (66.7%)* 4 (33.3%) 34 (17.2%) !.0001

ABS, apical ballooning syndrome; BSA, body surface area; CK-MB, creatinine kinase; MB subunit; ECG, electrocardiography; ICU, intensive care unit.Values are presented as n (%), mean 6 SD, or median (interquartile range).

*P ! .025 was considered to be statistically significant for comparison with women $50 years.yBody surface area was available in 9 patients (75%) of the men, 9 (75%) of the women !50 y, and 93 (47%) of the women $50 y.

308 Journal of Cardiac Failure Vol. 19 No. 5 May 2013

patients almost always present following a physical triggerand manifest a greater reduction in left ventricular systolicdysfunction; (2) women !50 years old have a higher prev-alence of premorbid psychiatric disorders and a greater like-lihood of recurrence; and 3) HRT does not exclude the riskfor developing ABS in females $50 years old.

ABS in Men

Men accounted for a small minority (5%) of patientswith ABS in our cohort, a finding consistent with mostlarge case series.3e5 The profile of cardiovascular risk fac-tors and comorbid conditions of these patients were similarto that of women $50 years who accounted for the greatmajority (90%) of patients. However, the male patients dif-fered regarding the triggers leading to ABS (Fig. 1). Virtu-ally all cases followed a major medical illness or aninvasive/surgical procedure, and an emotional trigger wasnot identified in any of the men. Because men in generalare not predisposed to developing ABS, we speculate that

physical stressors may be associated with a more sustainedsurge in catecholamines and thus a greater cardiac stresscompared with emotional triggers, and that this may bean essential element in the pathophysiology in men. Thisis in keeping with our observation that men had lower leftventricular ejection fraction at presentation and anincreased need for mechanical ventilation for acute respira-tory failure that was secondary to ABS (66.7%; P! .0001).Our study, conducted at a tertiary center, validates the find-ings of an earlier smaller series from a community hospitalcohort.6 Another study from Japan has reported that menpresent more frequently in the hospital setting,4 but thiswas not so in our cohort. The left ventricular ejectionfraction correlated positively with age among the men(r 5 0.75; P 5 .009), a relationship that was not presentin women and is of unclear clinical significance. In-hospital clinical outcomes and mortality were similaramong men and older women. All 5 deaths occurred dueto noncardiac causes, suggesting that ABS in males is

Fig. 1. Type and distribution of stress triggers in men, youngwomen (!50 y), and older women ($50 y).

Fig. 3. Kaplan-Meier curves for survival free of recurrence in the3 groups.

Age and Gender in Apical Ballooning Syndrome � Patel et al 309

unlikely to be amanifestation of a primary cardiac pathology,but rather a secondary phenomenon related to major comor-bid illnesses.

ABS in Young Women

Women !50 years old accounted for 5% of our cohort,a proportion similar to the men. A notable observation wasthe high prevalence of psychiatric disorders among theyoung women. This findings is consistent with our recentreport from a cohort of local residents that patients withABS are more likely to have chronic anxiety disordersand depression compared with control subjects.7 Thus,given the growing body of evidence linking chronic depres-sion and traits such as anger and hostility with cardiovascu-lar diseases, we speculate that chronic psychologic stress

Fig. 2. Ejection fraction at presentation and during follow-up inthe 3 groups. The ejection fraction increased significantly ineach of the 3 groups.

may be an important predisposing risk factor for ABS inyounger women. One of the largest studies on this subjectwas conducted among 72,359 women without a history ofcardiovascular diseases participating in the Nurses’ HealthStudy. Over 12 years of follow-up, high levels of phobicanxiety was associated with an increased risk of suddencardiac death and fatal coronary heart disease. Only partof the association could be accounted for by confoundingfactors, such as a higher frequency of cardiovascular riskfactors, in the individuals with high anxiety.8 It remains tobe established whether the association is causal or a markerof risk. Potential mechanisms related to chronic psycho-logic stress that may promote atherogenesis include activa-tion of the hypothalamic-pituitary-adrenal and sympatheticnervous systems, serotonergic dysfunction, secretion ofproinflammatory cytokines, and platelet activation.9

Women !50 years old had less hypertension and coro-nary artery disease, but the frequency of other cardiovascu-lar risk factors were similar between the two groups ofwomen. The prevalence of hypertension (42%), smoking(42%), hyperlipidemia (33%), and coronary artery disease(42%) was higher than one might expect in healthy youngwomen. The high prevalence of atherosclerosis and its riskfactors raises the possibility that vascular dysfunction maybe another predisposing factor in the pathophysiology.10

Moreover, a predisposition for developing ABS in thesepatients is suggested by the 5-fold higher rate of recurrenceduring follow-up among younger women (Fig. 3). Therewere no deaths among the young women, indicating a rela-tively benign prognosis, in contrast to the in-hospital mor-tality of 6% in older women.

Hormone Replacement Therapy in Older Women

Estrogen withdrawal is thought to contribute to the path-ophysiology of ABS because the vast majority of cases oc-cur in postmenopausal women. In an animal model of stresscardiomyopathy, estrogen replacement has been shown toattenuate stress-induced sympathoadrenal outflow fromthe brain to the heart, and up-regulate cardioprotectivepathways involving natriuretic peptides and heat shock

310 Journal of Cardiac Failure Vol. 19 No. 5 May 2013

proteins.11 Yet, as far as we know, there has been no clinicaldata published on the impact of HRT in ABS. The presentstudy provides some preliminary insight on this issue: 7.5%of older women were on chronic HRT at the time of devel-oping ABS, suggesting that estrogen replacement may notmitigate the risk of developing the transient cardiomyopa-thy. Of note, women who were not taking HRT had a lowerejection fraction (statistically nonsignificant) on presenta-tion, and a significantly greater need for mechanical hemo-dynamic and ventilator support.

Study Limitations

This was a retrospective investigation subject to the biasesof such analyses. The number of men, women !50 yearsold, and women $50 years old taking HRT were relativelysmall compared with the women $50 years old not takingHRT, and therefore, our observations must be consideredto be preliminary, and they require validation in other co-horts. Finally, we could not verify menopausal status or thecompliance with HRT in our patients because of the retro-spective study design.

Conclusion and Implications

This study indicates that men and young women withABS have certain unique characteristics. ABS does notoccur as a primary disorder in men, appearing to be a sec-ondary complication of major noncardiac illnesses or pro-cedures. Young women with ABS have a greater burdenof chronic psychologic disorders and a predisposition to re-currence. Therefore, a careful psychologic evaluation inthese patients may be important and optimal treatment ofpsychiatric disorders is desirable. Moreover, interventionsaimed at preventing recurrence, such as beta-blocker ther-apy, may yield greatest benefit in this subgroup. The roleof HRT in preventing or modulating the severity of ABSneeds further investigations.

Disclosures

None.

References

1. Bybee KA, Kara T, Prasad A, Lerman A, Barsness GW, Wright RS,

et al. Systematic review: transient left ventricular apical ballooning:

a syndrome that mimics ST-segment elevation myocardial infarction.

Ann Intern Med 2004;141:858e65.

2. Prasad A, Lerman A, Rihal CS. Apical ballooning syndrome (tako-

tsubo or stress cardiomyopathy): a mimic of acute myocardial infarc-

tion. Am Heart J 2008;155:408e17.

3. Sharkey SW, Windenburg DC, Lesser JR, Maron MS, Hauser RG,

Lesser JN, et al. Natural history and expansive clinical profile of

stress (tako-tsubo) cardiomyopathy. J Am Coll Cardiol 2010;55:

333e41.

4. Kurisu S, Inoue I, Kawagoe T, Ishihara M, Shimatani Y, Nakama Y,

et al. Presentation of tako-tsubo cardiomyopathy in men and women.

Clin Cardiol 2010;33:42e5.

5. Citro R, Rigo F, Previtali M, Ciampi Q, Canterin FA, Provenza G,

et al. Differences in clinical features and in-hospital outcomes of older

adults with tako-tsubo cardiomyopathy. J Am Geriatr Soc 2012;60:

93e8.

6. Buja P, Zuin G, Cutolo A, Grassi G, Madalosso M, Millosevich P,

et al. Left ventricular apical ballooning syndrome in men: a case se-

ries. J Cardiovasc Med 2012;13:790e4.7. Summers MR, Lennon RJ, Prasad A. Pre-morbid psychiatric and car-

diovascular diseases in apical ballooning syndrome (tako-tsubo/stress-

induced cardiomyopathy): potential pre-disposing factors? J Am Coll

Cardiol 2010;55:700e1.8. Albert CM, Chae CU, Rexrode KM, Manson JE, Kawachi I. Phobic

anxiety and risk of coronary heart disease and sudden cardiac death

among women. Circulation 2005;111:480e7.9. Everson-Rose SA, Lewis TT. Psychosocial factors and cardiovascular

diseases. Annu Rev Public Health 2005;26:469e500.

10. Martin EA, Prasad A, Rihal CS, Lerman LO, Lerman A. Endothelial

function and vascular response to mental stress are impaired in pa-

tients with apical ballooning syndrome. J Am Coll Cardiol 2010;56:

1840e6.

11. Ueyama T, Kasamatsu K, Hano T, Tsuruo Y, Ishikura F. Catechol-

amines and estrogen are involved in the pathogenesis of emotional

stress-induced acute heart attack. Ann N Y Acad Sci 2008;1148:

479e85.