- Email: [email protected]
Reproductive History of Women With Takotsubo Cardiomyopathy
Reproductive History of Women With Takotsubo Cardiomyopathy Elena Salmoirago-Blotcher, MD, PhDa,b,*, Shira Dunsiger, PhDb,c, Heather H. Swales, MDd, Gerard P. Aurigemma, MDe, Ira Ockene, MDe, Lindsey Rosman, PhDf, and Ilan S. Wittstein, MDg Takotsubo cardiomyopathy (TC) occurs predominantly in postmenopausal women, suggesting a possible role of reproductive and hormonal factors in the pathophysiology of this condition. Yet reproductive characteristics of women with TC have received limited attention. This prospective case-control study sought to explore reproductive characteristics associated with TC. Incident TC cases and myocardial infarction (MI) controls were recruited among consecutive women presenting at the emergency departments of 2 large medical centers in Massachusetts and Connecticut. Female healthy controls were recruited from a registry of research volunteers. Information about reproductive history was collected 1 month after discharge using standardized questionnaires completed during phone interviews. Linear and logistic regression models were used to estimate associations with reproductive factors. From March 2013 to October 2015, 209 women were screened for eligibility and 107 (45 TC, 32 MI, and 30 healthy controls) were enrolled. Conditions uniquely associated with TC were a history of irregular menses (adjusted OR, TC vs MI 8.30; 95% CI 1.01 to 69.18), number of pregnancies (adjusted b coefficient 0.69; SE 0.35, p [ 0.05), and use of post-menopausal hormone replacement therapy (OR 5.79; CI 1.20 to 28.02). We did not find associations with history of infertility, breastfeeding, hysterectomy or oophorectomy, oral contraceptive use, and age at menopause. In conclusion, our findings suggest that premenopausal reproductive factors may play an important role in the onset of TC at a later age. These results need to be confirmed in future studies with larger populations. Ó 2016 Elsevier Inc. All rights reserved. (Am J Cardiol 2016;-:-e-) Takotsubo cardiomyopathy (TC) is a syndrome characterized by acute left ventricular dysfunction accompanied by electrocardiographic changes and cardiac enzyme elevation in the absence of significant coronary artery disease.1,2 A physical or emotional trigger precedes the occurrence of TC in 40% to 70% of cases2e4 and 90% of cases are diagnosed in post-menopausal women.3,4 This association suggests that low estrogen levels may play a role in the pathogenesis of this syndrome. Although there is a general consensus that an exaggerated sympathetic stimulation plays an important role in the development of TC,5,6 the role of hormonal factors is still unclear. Studies conducted in animal models of TC have shown that estrogen supplementation reduces a Department of Medicine, Medical School, and cDepartment of Behavioral and Social Sciences, School of Public Health, Brown University, Providence, Rhode Island, USA; bCenters for Behavioral and Preventive Medicine, The Miriam Hospital, Providence, Rhode Island, USA; d Department of Medicine, Hartford Hospital, Hartford, Connecticut, USA; e Department of Medicine, University of Massachusetts Medical School, Worcester, Massachusetts, USA; fDepartment of Psychology, East Carolina University, Greenville, North Carolina, USA; and gDepartment of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA. Manuscript received May 13, 2016; revised manuscript received and accepted August 19, 2016. This work was supported by the American Heart Association (Dallas, Texas) award 13CRP1334001 to Dr. Salmoirago-Blotcher. See page 6 for disclosure information. *Corresponding author: Tel: (401) 793-8325; fax: (401) 793-8059. E-mail address: [email protected] (E. Salmoirago-Blotcher).
0002-9149/16/$ - see front matter Ó 2016 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2016.08.083
responses to emotional stress.7e9 The few clinical studies examining reproductive characteristics of women with TC have yielded conflicting results.10e13 The purpose of this study was to investigate whether the occurrence of TC is associated with specific reproductive characteristics. We expected that, consistent with laboratory evidence in animal models, conditions associated with low estrogen levels (i.e., a history of oophorectomy) would be more prevalent in TC women. In addition, we were interested in assessing whether conditions associated with a lifelong exposure to lower estrogen levels may play a role in the onset of this syndrome. Methods To study the reproductive history of women with TC, we conducted a prospective (i.e., enrolling incident cases and controls) case-control study. Inclusion criteria for TC cases were age 21 years, a first diagnosis of TC fulfilling Mayo Clinic diagnostic criteria,14 English fluency, and access to a telephone. Exclusion criteria were inability or unwillingness to provide informed consent; a history of pheochromocytoma, myocarditis, or hypertrophic cardiomyopathy; dementia or severe cognitive impairment; and being clinically unstable. Women with TC associated with intracranial bleeding or head trauma were excluded because of cognitive and neurologic deficits impairing the ability to complete study interviews. Controls were women admitted with a confirmed diagnosis of acute nonfatal myocardial infarction (MI) and a group of healthy controls (HC). Inclusion criteria for MI controls were age 21 years, a diagnosis of MI meeting www.ajconline.org
2
The American Journal of Cardiology (www.ajconline.org)
current diagnostic criteria,15 English fluency, and access to a telephone. Women were excluded if they had a previous diagnosis of TC, were unable or unwilling to give informed consent, and if they were clinically unstable. Eligibility criteria for HC were age 21 years, English fluency, and access to a telephone. HC were excluded if they had a previous diagnosis of TC, a chronic condition (any cancer other than non-melanoma skin cancer, cardiovascular disease, liver failure, or renal failure), and a history of dementia or severe cognitive impairment. A trained physician abstractor blinded to the study outcomes confirmed eligibility for the study for all conditions. TC and MI women were recruited among incident consecutive cases presenting at emergency departments of 3 large medical centers in the New England region from March 2013 to October 2015. Participants were identified through physician referral and daily reviews of computerized listings from cardiac catheterization laboratories, echo laboratories, and emergency departments. HC were recruited from a registry of research volunteers at the University of Massachusetts Medical School. All participants received a letter describing the study and inviting them to participate. To avoid selection bias (i.e., excluding women who were too ill or refused a coronary angiogram), TC cases that did not undergo an angiogram but otherwise met all other diagnostic criteria at discharge were also enrolled. A phone interview was scheduled for women expressing interest in study participation. After obtaining verbal informed consent and authorization to access medical records, research personnel conducted the study interview w1 month after discharge using a standardized script to elicit information consistently across participants. Both the interviewer and the study participants were blinded to the study outcomes and to participants’ case or control status. The study was approved by the institutional review board at each participating institution. Research Electronic Data capture technology was used to conduct computer-assisted interviews with direct data entry into electronic study surveys. Information about reproductive characteristics was collected using a survey modeled after the reproductive history questionnaire used for the Women’s Health Initiative Observational Study.16 The information collected included: age at first (“How old were you when you had your first menstrual period?”) and last menstrual period (How old were you when you last had any menstrual bleeding?), history of menstrual irregularities (“during most of your life, were your periods regular, i.e., did they occur about once a month, not including any time when your were pregnant or taking birth control pills?”), history of ovarian dysfunction (“Between your first and your last period, did you go without any period for at least a year, not counting any time when your were pregnant or taking birth control pills?”), infertility (“Have you ever tried to become pregnant for more than one year without becoming pregnant?”), number of pregnancies, breastfeeding history (“Did you breastfeed any children for at least one month?”), history of menopausal symptoms (“Have you ever had menopausal symptoms such as hot flashes or night sweats?”), history of oophorectomy or hysterectomy and age at surgery, oral contraceptive use and duration of use (“Did you ever take birth control pills, diethylstilbestrol, or
shots called Depo-Provera for birth control or any other reasons?”), use of hormone replacement therapy (HRT) (“Did you take hormone replacement therapy, that is, hormones that are taken around the time of menopause or after menopause, not including hormones used for birth control”), and duration of use. Information about age, race/ethnicity, marital status, income, and education was collected directly from the participants using standardized questionnaires. Information about medical history (including coronary risk factors) was abstracted from the medical record with the exception of physical activity (frequency of walking outside the home for at least 10 minutes without stopping) and family history of coronary heart disease (CHD) (self-reported questionnaire). Body mass index was calculated from height and weight measurements at admission reported in the medical record. Demographic and reproductive characteristics were compared using ANOVA (or nonparametric tests where applicable) for continuous measures and chi-square statistics for categorical variables. Generalized linear models and logistic regression models (adjusted for variables that were associated with the outcomes of interest with p 0.1) were used to estimate associations with reproductive characteristics, with HC as the reference group. For all estimates, p values or 95% confidence intervals were calculated. All study analyses were performed using SAS, version 9.3, statistical software. Results From March 2013 to October 2015, 107 women (45 TC cases, 32 MI controls [STEMI, n ¼ 24; non-STEMI, n ¼ 8]) and 30 HC were enrolled in the study (Figure 1). HC had higher income and education and were slightly older compared with MI and TC, whereas other demographic characteristics were similar across groups (Table 1). Although diabetes and dyslipidemia, as well as a higher body mass index, were more common among MI controls, a history of smoking, a family history of CHD, and hypertension were also highly prevalent among TC cases. Clinical characteristics at admission were fairly similar between TC cases and MI controls with the exception of peak troponin levels, ejection fraction, and systolic blood pressure, which were lower in TC women. The diagnosis of TC was not angiographically confirmed in 15 women. These women, however, did fulfill all other diagnostic criteria for TC and received a final diagnosis of TC at discharge. Angiographically confirmed TC cases did not significantly differ from nonconfirmed cases in terms of baseline demographics and medical history (data not shown). Unadjusted reproductive characteristic by case-control status are listed in Table 2, whereas findings from multivariable adjusted models are presented in Tables 3 (logistic regression) and 4 (linear regression). Both TC cases and MI controls were younger than HC at their first menstrual period, with no significant differences between TC and MI. Women with TC were more likely to report irregular menses compared with HC and MI: in adjusted models, TC women were 8 times more likely than MI controls to have a history of irregular menses. The prevalence of a history of infertility was similar across
Cardiomyopathy/Takotsubo Cardiomyopathy and Reproductive History
3
Screened = 209 Ineligible = 94 12 clinically unstable 7 intracranial bleeding 36 diagnosis not confirmed 4 HC with chronic condiƟons 6 deceased 12 unreachable 5 not English speaking 12 myocardiƟs/CABG
Eligible = 115 Refused = 4 2 not interested 2 no Ɵme
Consented = 111
Diagnosis not confirmed =4
Enrolled = 107 45 TC 32 MI 30 HC
Figure 1. STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) diagram. CABG ¼ coronary artery bypass grafting.
groups. A history of amenorrhea of at least 1 year’s duration was more prevalent among both TC cases and MI controls compared with the HC group. TC women had a significantly higher number of pregnancies compared with both HC and MI. We did not find associations between case-control status and breastfeeding history. Regarding menopausal characteristics, TC women were, on average, 3 years younger than HC at menopause onset; these differences, however, were not statistically significant. TC cases were more likely than HC to report menopausal symptoms, such as night sweats and hot flashes. Although these associations did not reach statistical significance, we found a large effect size in that TC cases were 3 times more likely than HC and 50% more likely than MI controls to report menopausal symptoms. There were no differences in the prevalence of a history of oophorectomy or hysterectomy across groups.
There were no significant differences in the use of oral contraceptives during fertile years. Among women who did report using oral contraceptives, TC cases reported to have used it for fewer years compared with HC (3.2 years; p ¼ 0.04) and MI (1.9 years, p ¼ 0.20). Women with TC were almost 6 times more likely to report HRT use compared with MI controls. Discussion This study sought to explore reproductive factors related to female endogenous hormone levels in women admitted with a recent episode of TC. Overall, we found that certain reproductive characteristics, namely a history of irregular menses, number of pregnancies, and HRT use after menopause, appeared to be uniquely associated with the onset of TC. We did not find significant associations between TC
4
The American Journal of Cardiology (www.ajconline.org)
Table 1 Baseline characteristics of cases and controls* TC ¼ 45
MI ¼ 32
Table 2 Unadjusted reproductive characteristics by case/control status HC ¼ 30
Age (years) (Mean 62.4 10.8 64.6 15.3 55.9 12.4 SD) American Indian or 1 (2%) 0 0 Alaskan Asian or Pacific 0 0 1 (3%) Islander Black or African0 1 (3%) 0 American Hispanic/Latino 3 (7%) 6 (19%) 3 (10%) White (not Hispanic) 40 (91%) 25 (78%) 26 (87%) Education High School 20 (46%) 19 (59%) 3 (10%) diploma College or some 23 (52%) 13 (41%) 24 (80%) college Post-graduate 1 (2%) 0 3 (10%) Income Less than $ 35,000 23 (52%) 17 (53%) 4 (13%) $35,000 to $74,999 9 (20%) 6 (19%) 12 (40%) $75,000 or more 9 (21%) 6 (19%) 14 (47%) Don’t know 3 (7%) 3 (9%) 0 Coronary risk factors Family history of 18 (40%) 16 (50%) 0 coronary heart disease Diabetes mellitus 8 (18%) 14 (44%) 3 (10%) Hyperlipidemia 17 (38%) 19 (59%) 8 (27%) Hypertension 26 (58%) 22 (69%) 5 (17%) Body mass index 22.3 (9.7) 29.2 (7.8) (kg/m2) Ever a smoker 31 (71%) 22 (69%) 13 (45%) Physically activez 21 (48%) 17 (53%) 20 (69%) Medical history Coronary bypass 0 2 (6%) 0 Percutaneous 0 3 (9%) 0 coronary intervention Stroke 0 2 (6%) 0 Anxiety disorder 11 (24%) 3 (9%) 0 Mood disorder 9 (20%) 7 (22%) 5 (17%) Clinical characteristics at admission 119 23 130 23 n/a Systolic blood pressure (mmHg) (Mean SD) Diastolic blood 70 17 75 19 n/a pressure (mmHg) (Mean SD) Ejection fraction (%) 40 (10) 48 (12) n/a Peak CPK (IU/L) 398.0 515.8 566.0 367.5 n/a (Mean SD) Peak troponin (ng/ml) 2.9 3.2 6.8 7.8 n/a (Mean SD) Heart rate (bpm) 91 26 86 18 n/a (Mean SD)
p†
TC ¼ 45
0.08
Age at first menses 12.7 1.6 12.7 1.3 13.3 1.8 (mean SD) History of irregular 9 (21%) 1 (0%) 0 menstrual periods Ever pregnant 39 (91%) 28 (90%) 23 (82%) Number of pregnancies 3.4 1.6 2.8 1.3 2.7 1.5 (mean SD) Ever breastfeeding for at 17 (40%) 11 (36%) 14 (52%) least 1 month Ever taken birth control 26 (58%) 13 (41%) 21 (70%) pill Years on birth control pill 3.7 3.4 5.2 5.5 5.9 4.5 (mean SD) History of infertility 1 (2%) 0 1 (4%) History of ovarian 3 (7%) 3 (10%) 0 dysfunction† Previous oophorectomy 15 (35%) 7 (23%) 10 (36%) Age at oophorectomy 41.0 10 44.7 11.6 45.1 9.7 (mean SD) Previous hysterectomy 14 (33%) 8 (27%) 10 (39%) Age at last menstrual 45.6 8.9 45.6 7.4 45.7 8.3 bleeding (mean SD) Menopausal symptoms 30 (70%) 17 (55%) 12 (43%) (Night sweats, hot flashes) Hormone replacement 13 (30%) 2 (6%) 4 (15%) therapy
0.49 0.28 0.31 0.26 0.28 <.001 0.006 0.09 0.001 0.10 0.02 0.25 <.001
0.004 0.03 <.001 0.009 0.06 0.20 0.09 0.03
0.09 0.007 0.87 0.04
0.24
0.02 0.63 0.01 0.37
* Values are n (%) unless otherwise indicated. † ANOVA or chi-square. z Walking outdoors for at least 10 minutes once a week.
and a history of infertility, breastfeeding, hysterectomy or oophorectomy, birth control pill use, and age at menopause. Taken together, these findings indicate that although studies
MI ¼ 32
HC ¼ 30
p* 0.21 .009 0.50 0.14 0.42 0.06 0.23 0.58 0.26 0.45 0.58 0.64 0.99 0.07 0.12
* ANOVA or chi-square. † Defined as history of amenorrhea of at least one year during fertile years.
conducted in animal models suggest that low estrogen levels may play a role in the physiopathology of TC,7e9 in humans, the picture is likely more complex and that, rather than estrogen levels per se, premenopausal factors may play an important role in the onset of TC at a later age. The epidemiologic evidence shows that few cardiovascular events occur before the onset of menopause, and in fact, the loss of cyclical ovarian function occurring with menopause is associated with the development of atherosclerosis and CHD in women.17 During fertile years, estrogens protect from atherosclerosis through favorable effects on lipoprotein levels, vascular reactivity, fat distribution, and hemostasis.18,19 Although this is the first study reporting an association between history of menstrual irregularity and the occurrence of TC, there is a growing body of evidence linking premenopausal abnormalities of ovarian function to the onset of CHD in older age. A prospective cohort study of 82,439 female nurses found that women with irregular cycles had an increased risk for nonfatal or fatal CHD compared with women with a history of regular menstrual cycles after adjustment for potential confounders.20 Similarly, a large prospective cohort study conducted among >23,000 women in The Netherlands found that women with irregular cycles had a 27% higher risk of CHD compared with those with a history of regular cycles, independent of common coronary risk factors.21 Another large prospective study using administrative data has linked irregular cycles and cardiovascular mortality, although risk differences were no longer
Cardiomyopathy/Takotsubo Cardiomyopathy and Reproductive History Table 3 Results of logistic regression analyses* TC vs HC Irregular menstrual periods Unadjusted n/a† Adjustedz Ovarian dysfunctionx Unadjusted n/a† Adjustedz Menopausal symptoms Unadjusted 3.08 (1.14-8.29) Adjusted{ 3.00 (.84-10.71) Ever pregnant Unadjusted 2.12 (.52-8.7) Adjustedk 1.80 (.42-7.7) Ever breastfeeding for at least 1 month Unadjusted .61 (.23-1.60) Adjustedk .74 (.27-2.03) Previous oophorectomy Unadjusted .96 (.36-2.61) Adjusted** .57 (.16-2.10) Previous hysterectomy Unadjusted .77 (.28-2.13) Adjusted†† .78 (.20-3.05) Birth control pill Unadjusted .59 (.22-1.56) Adjusted** .71 (.20-2.50) Hormone replacement therapy Unadjusted 2.64 (.77-9.07) Adjusted 2.96 (.82-10.66)
5
Table 4 Results of linear regression analyses* TC vs MI 8.86 (1.07-73.18) 8.30 (1.01-69.18) .70 (.13-3.72) .73 (.14-3.94) 1.90 (.73-4.97) 1.55 (.51-4.43) 1.05 (.21-5.04) 1.15 (.23-5.66) 1.19 (.46-3.09) 1.07 (.40-2.84) 1.84 (.64-5.25) 1.98 (.57-6.89) 1.33 (.47-3.72) 1.19 (.37-3.89) 2.00 (.80-5.02) 1.99 (.71-5.60) 6.09 (1.27-29.29) 5.79 (1.20-28.02)
* Data are OR (95% CI). † OR could not be computed when comparing to cells with 0 cell count or very low cell count (n ¼ 1). z Adjusted for anxiety, education. x Defined as history of amenorrhea of at least one year during fertile years. { Adjusted for age, anxiety. k Adjusted for education. ** Adjusted for age. †† Adjusted for age, income.
significant after adjustment for body mass index.22 The finding of an association between the occurrence of TC and menstrual irregularities suggests that a lifelong exposure to low estrogen levels (like that occurring with even mild disruptions of ovarian function) may play an important role in the onset of TC.23,24 Cardiovascular changes associated with pregnancy are so vast that they led some investigators to define pregnancy as a “9-month-long cardiovascular stress test.”24 A secondary analysis from 3 case-control studies has shown that parity was positively associated with an increased risk of nonfatal MI,25 whereas a prospective analysis from the Multi-Ethnic Study of Atherosclerosis cohort found that the number of live births was associated with increases in left ventricular mass and end-diastolic and end-systolic volume and decreases in ejection fraction in middle to older age, independent of demographic and CHD risk factors.26 Although ours is the first study showing a significant association between number of pregnancies and TC, this finding deserves further study, in particular as to how repeated pregnancies may predispose women to develop TC later in life.
TC vs HC Age at first menses Unadjusted -0.60 (.36) 0.70 (.37) Adjusted† Age at last menstrual bleeding Unadjusted -0.01 (1.99) Adjustedz -3.04 (2.17) Pregnancies Unadjusted 0.71 (0.38) Adjustedz 1.08 (0.44) Age at oophorectomy Unadjusted -4.11 (4.09) Adjusted -7.24 (4.82) Years on birth control pill Unadjusted -2.20 (1.25) Adjustedz -3.20 (1.54) Years on hormone replacement therapy Unadjusted -7.66 (3.87) Adjustedz 1.11 (4.23)
p
TC vs MI
p
.10 .05
0.02 (0.36) 0.08 (0.36)
.96 .83
0.9 .16
0.17 (1.93) 1.33 (1.84)
0.93 0.47
.06 .01
0.54 (0.36) 0.69 (0.35)
.14 .05
.31 .13
-3.67 (4.69) -3.01 (5.14)
.43 .56
.07 .04
-1.47 (1.43) -1.87 (1.53)
.30 .22
.05 .79
2.59 (5.14) 3.99 (4.21)
.61 .34
* Data are beta coefficients (SE). † Adjusted for education. z Adjusted for age.
We also found that TC women were almost 6 times more likely than MI controls to report use of post-menopausal HRT (differences with HC were in the same direction but were not statistically significant). Information on the role of HRT in the onset of TC is scarce and has largely focused on HRT use at the time of the hospital admission. A recent retrospective study found a higher prevalence of current HRT in both TC and MI women compared with HC.11 A small case series and literature review found that no TC woman was receiving HRT at hospital admission.12 Although our study differed from these studies in that we focused on history of menopausal HRT use, instead of current use, our findings are consistent with those from randomized clinical trials showing that HRT may actually increase the risk of cardiovascular events in postmenopausal women.27 A possible explanation for the higher prevalence of HRT use among TC cases versus MI controls is confounding by indication, for example, TC cases were more likely to be on HRT because they had fewer contraindications to HRT. The important message from this finding, however, is that HRT in menopause was not protective, thus suggesting that estrogen levels per se may not play an important role in the onset of this condition. Although there is a relative abundance of information linking female reproductive history and CHD, there is very limited information about reproductive characteristics of TC women. A small study that compared estradiol, progesterone, luteinizing hormone, and follicle-stimulating hormone levels in women with SC (n ¼ 17), age-matched women with MI (n ¼ 16), and women with normal coronary arteries (n ¼ 15) found higher concentrations of estradiol in TC compared with women with MI or women with normal coronary arteries, but, as the investigators note, estrogen levels were measured at hospital admission and could have increased in response to stress.10 A retrospective casecontrol study conducted among 25 consecutive female
6
The American Journal of Cardiology (www.ajconline.org)
cases of TC, age-matched and event dateematched control women with ST-segment elevation MI (n ¼ 25), and women with neither diagnosis (n ¼ 50) found no differences in (crude) prevalence of menopausal status, frequency of oophorectomy, and average years since the onset of menopause.11 This study has several important strengths. First, this is the first extensive study of reproductive characteristics of TC women. Although findings need to be confirmed in a larger study, our data suggest that reproductive history could be an essential component of the assessment of cardiovascular risk among women. Other strengths include a rigorous case-control design with enrollment of incident cases and controls, standardized procedures for data collection, and blinding of study staff conducting study interviews. Some important limitations need to be acknowledged. First, this study had a relatively small sample size, which limited the power to detect associations. Although we have been cautious in dismissing nonsignificant associations with large magnitude, results from this study need to be reproduced in larger samples. We would like to note that associations with menstrual irregularities and number of pregnancies appear to be both biologically plausible and consistent with epidemiologic evidence in women with CHD. Second, as for all case-control studies, recall bias is an important limitation. We note that in the present study interviews for TC and AMI women were conducted at least 1 month after discharge to minimize the potential effects of hospitalization-related distress on patient-reported reproductive history. It is also unlikely that TC cases would recall their reproductive history differently than MI controls. Third, information on reproductive factors was selfreported. Although women are quite reliable when reporting details about their reproductive history,28,29 nonetheless we were not able to collect detailed information about the characteristics and duration of menstrual irregularities or to document a definite diagnosis, such a polycystic ovary syndrome or functional hypothalamic dysfunction. A final limitation concerns our selection of HC, which were recruited from a registry of research volunteers. With the exception of higher socioeconomic status and age (which were accounted for in the analyses), however, no other group differences were associated with the outcomes of interest. Acknowledgment: The authors wish to thank Sandhya Reddy, MD (data abstractor) and Nancy Mecone, BSN, RN (research coordinator) for their collaboration on this study. Disclosures The American Heart Association did not have any role in the design and conduct of the study, in the collection, analysis, and interpretation of the data, and in the preparation, review, or approval of the manuscript. The authors have no conflicts of interest to disclose. 1. Dote K, Sato H, Tateishi H, Uchida T, Ishihara M. Myocardial stunning due to simultaneous multivessel coronary spasms: a review of 5 cases. J Cardiol 1991;21:203e214.
2. Eitel I, von Knobelsdorff-Brenkenhoff F, Bernhardt P, Carbone I, Muellerleile K, Aldrovandi A, Francone M, Desch S, Gutberlet M, Strohm O, Schuler G, Schulz-Menger J, Thiele H, Friedrich M. Clinical characteristics and cardiovascular magnetic resonance findings in stress (takotsubo) cardiomyopathy. JAMA 2011;306:277e286. 3. Pilgrim TM, Wyss TR. Takotsubo cardiomyopathy or transient left ventricular apical ballooning syndrome: a systematic review. Int J Cardiol 2008;124:283e292. 4. Templin C, Ghadri JR, Diekmann J, Napp LC, Bataiosu DR, Jaguszewski M, Cammann VL, Sarcon A, Geyer V, Neumann CA, Seifert B, Hellermann J, Schwyzer M, Eisenhardt K, Jenewein J, Franke J, Katus HA, Burgdorf C, Schunkert H, Moeller C, Thiele H, Bauersachs J, Tschöpe C, Schultheiss HP, Laney CA, Rajan L, Michels G, Pfister R, Ukena C, Böhm M, Erbel R, Cuneo A, Kuck KH, Jacobshagen C, Hasenfuss G, Karakas M, Koenig W, Rottbauer W, Said SM, Braun-Dullaeus RC, Cuculi F, Banning A, Fischer TA, Vasankari T, Airaksinen KE, Fijalkowski M, Rynkiewicz A, Pawlak M, Opolski G, Dworakowski R, MacCarthy P, Kaiser C, Osswald S, Galiuto L, Crea F, Dichtl W, Franz WM, Empen K, Felix SB, Delmas C, Lairez O, Erne P, Bax JJ, Ford I, Ruschitzka F, Prasad A, Lüscher TF. Clinical features and outcomes of takotsubo (stress) cardiomyopathy. N Engl J Med 2015;373:929e938. 5. Wittstein IS, Thiemann DR, Lima JA, Baughman KL, Schulman SP, Gerstenblith G, Wu KC, Rade JJ, Bivalacqua TJ, Champion HC. Neurohumoral features of myocardial stunning due to sudden emotional stress. N Engl J Med 2005;352:539e548. 6. Abraham J, Mudd JO, Kapur NK, Klein K, Champion HC, Wittstein IS. Stress cardiomyopathy after intravenous administration of catecholamines and beta-receptor agonists. J Am Coll Cardiol 2009;53: 1320e1325. 7. Ueyama T, Ishikura F, Matsuda A, Asanuma T, Ueda K, Ichinose M, Kasamatsu K, Hano T, Akasaka T, Tsuruo Y, Morimoto K, Beppu S. Chronic estrogen supplementation following ovariectomy improves the emotional stress-induced cardiovascular responses by indirect action on the nervous system and by direct action on the heart. Circ J 2007;71: 565e573. 8. Ueyama T, Hano T, Kasamatsu K, Yamamoto K, Tsuruo Y, Nishio I. Estrogen attenuates the emotional stress-induced cardiac responses in the animal model of Tako-tsubo (Ampulla) cardiomyopathy. J Cardiovasc Pharmacol 2003;42 Suppl 1:S117eS119. 9. Cao X, Zhou C, Chong J, Fu L, Zhang L, Sun D, Hou H, Zhang Y, Li D, Sun H. Estrogen resisted stress-induced cardiomyopathy through increasing the activity of beta(2)AR-Galphas signal pathway in female rats. Int J Cardiol 2015;187:377e386. 10. Brenner R, Weilenmann D, Maeder MT, Jorg L, Bluzaite I, Rickli H, De Pasquale G, Ammann P. Clinical characteristics, sex hormones, and long-term follow-up in Swiss postmenopausal women presenting with Takotsubo cardiomyopathy. Clin Cardiol 2012;35:340e347. 11. Scantlebury DC, Prasad A, Rabinstein AA, Best PJ. Prevalence of migraine and Raynaud phenomenon in women with apical ballooning syndrome (Takotsubo or stress cardiomyopathy). Am J Cardiol 2013;111:1284e1288. 12. Kuo BT, Choubey R, Novaro GM. Reduced estrogen in menopause may predispose women to takotsubo cardiomyopathy. Gend Med 2010;7:71e77. 13. Sato A, Yagihara N, Kodama M, Mitsuma W, Tachikawa H, Ito M, Hanawa H, Aizawa Y. Takotsubo cardiomyopathy after delivery in an oestrogen-deficient patient. Int J Cardiol 2011;149:e78ee79. 14. Prasad A, Lerman A, Rihal CS. Apical ballooning syndrome (Takotsubo or stress cardiomyopathy): a mimic of acute myocardial infarction. Am Heart J 2008;155:408e417. 15. Thygesen K, Alpert JS, Jaffe AS, Simoons ML, Chaitman BR, White HD, Writing Group on the Joint ESC/ACCF/AHA/WHF Task Force for the Universal Definition of Myocardial Infarction, Thygesen K, Alpert JS, White HD, Jaffe AS, Katus HA, Apple FS, Lindahl B, Morrow DA, Chaitman BA, Clemmensen PM, Johanson P, Hod H, Underwood R, Bax JJ, Bonow RO, Pinto F, Gibbons RJ, Fox KA, Atar D, Newby LK, Galvani M, Hamm CW, Uretsky BF, Steg PG, Wijns W, Bassand JP, Menasché P, Ravkilde J, Ohman EM, Antman EM, Wallentin LC, Armstrong PW, Simoons ML, Januzzi JL, Nieminen MS, Gheorghiade M, Filippatos G, Luepker RV, Fortmann SP, Rosamond WD, Levy D, Wood D, Smith SC, Hu D, Lopez-Sendon JL, Robertson RM, Weaver D, Tendera M, Bove AA, Parkhomenko AN, Vasilieva EJ, Mendis S; ESC Committee for
Cardiomyopathy/Takotsubo Cardiomyopathy and Reproductive History
16. 17. 18. 19. 20.
21.
22.
Practice Guidelines (CPG). Third universal definition of myocardial infarction. Eur Heart J 2012;33:2551e2567. Design of the Women’s Health Initiative clinical trial and observational study. The Women’s Health Initiative Study Group. Control Clin Trials 1998;19:61e109. Kannel WB, Hjortland MC, McNamara PM, Gordon T. Menopause and risk of cardiovascular disease: the Framingham study. Ann Intern Med 1976;85:447e452. Schwertz DW, Penckofer S. Sex differences and the effects of sex hormones on hemostasis and vascular reactivity. Heart Lung 2001;30:401e426. Vitale C, Mendelsohn ME, Rosano GM. Gender differences in the cardiovascular effect of sex hormones. Nat Rev Cardiol 2009;6:532e542. Solomon CG, Hu FB, Dunaif A, Rich-Edwards JE, Stampfer MJ, Willett WC, Speizer FE, Manson JE. Menstrual cycle irregularity and risk for future cardiovascular disease. J Clin Endocrinol Metab 2002;87:2013e2017. Gast GC, Grobbee DE, Smit HA, Bueno-de-Mesquita HB, Samsioe GN, van der Schouw YT. Menstrual cycle characteristics and risk of coronary heart disease and type 2 diabetes. Fertil Steril 2010;94: 2379e2381. Wang ET, Cirillo PM, Vittinghoff E, Bibbins-Domingo K, Cohn BA, Cedars MI. Menstrual irregularity and cardiovascular mortality. J Clin Endocrinol Metab 2011;96:E114eE118.
7
23. Kaplan JR, Manuck SB. Ovarian dysfunction and the premenopausal origins of coronary heart disease. Menopause 2008;15:768e776. 24. Shufelt C, Waldman T, Wang E, Merz CN. Female-specific factors for IHD: across the reproductive lifespan. Curr Atheroscler Rep 2015;17: 481e486. 25. Bertuccio P, Tavani A, Gallus S, Negri E, La Vecchia C. Menstrual and reproductive factors and risk of non-fatal acute myocardial infarction in Italy. Eur J Obstet Gynecol Reprod Biol 2007;134:67e72. 26. Parikh NI, Lloyd-Jones DM, Ning H, Ouyang P, Polak JF, Lima JA, Bluemke D, Mittleman MA. Association of number of live births with left ventricular structure and function. The Multi-Ethnic Study of Atherosclerosis (MESA). Am Heart J 2012;163:470e476. 27. Boardman HM, Hartley L, Eisinga A, Main C, Roque i Figuls M, Bonfill Cosp X, Gabriel Sanchez R, Knight B. Hormone therapy for preventing cardiovascular disease in post-menopausal women. Cochrane Database Syst Rev 2015;3:CD002229. 28. Bosetti C, Tavani A, Negri E, Trichopoulos D, La Vecchia C. Reliability of data on medical conditions, menstrual and reproductive history provided by hospital controls. J Clin Epidemiol 2001;54:902e906. 29. Hunter DJ, Manson JE, Colditz GA, Chasan-Taber L, Troy L, Stampfer MJ, Speizer FE, Willett WC. Reproducibility of oral contraceptive histories and validity of hormone composition reported in a cohort of US women. Contraception 1997;56:373e378.
0002-9149/16/$ - see front matter Ó 2016 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2016.08.083
responses to emotional stress.7e9 The few clinical studies examining reproductive characteristics of women with TC have yielded conflicting results.10e13 The purpose of this study was to investigate whether the occurrence of TC is associated with specific reproductive characteristics. We expected that, consistent with laboratory evidence in animal models, conditions associated with low estrogen levels (i.e., a history of oophorectomy) would be more prevalent in TC women. In addition, we were interested in assessing whether conditions associated with a lifelong exposure to lower estrogen levels may play a role in the onset of this syndrome. Methods To study the reproductive history of women with TC, we conducted a prospective (i.e., enrolling incident cases and controls) case-control study. Inclusion criteria for TC cases were age 21 years, a first diagnosis of TC fulfilling Mayo Clinic diagnostic criteria,14 English fluency, and access to a telephone. Exclusion criteria were inability or unwillingness to provide informed consent; a history of pheochromocytoma, myocarditis, or hypertrophic cardiomyopathy; dementia or severe cognitive impairment; and being clinically unstable. Women with TC associated with intracranial bleeding or head trauma were excluded because of cognitive and neurologic deficits impairing the ability to complete study interviews. Controls were women admitted with a confirmed diagnosis of acute nonfatal myocardial infarction (MI) and a group of healthy controls (HC). Inclusion criteria for MI controls were age 21 years, a diagnosis of MI meeting www.ajconline.org
2
The American Journal of Cardiology (www.ajconline.org)
current diagnostic criteria,15 English fluency, and access to a telephone. Women were excluded if they had a previous diagnosis of TC, were unable or unwilling to give informed consent, and if they were clinically unstable. Eligibility criteria for HC were age 21 years, English fluency, and access to a telephone. HC were excluded if they had a previous diagnosis of TC, a chronic condition (any cancer other than non-melanoma skin cancer, cardiovascular disease, liver failure, or renal failure), and a history of dementia or severe cognitive impairment. A trained physician abstractor blinded to the study outcomes confirmed eligibility for the study for all conditions. TC and MI women were recruited among incident consecutive cases presenting at emergency departments of 3 large medical centers in the New England region from March 2013 to October 2015. Participants were identified through physician referral and daily reviews of computerized listings from cardiac catheterization laboratories, echo laboratories, and emergency departments. HC were recruited from a registry of research volunteers at the University of Massachusetts Medical School. All participants received a letter describing the study and inviting them to participate. To avoid selection bias (i.e., excluding women who were too ill or refused a coronary angiogram), TC cases that did not undergo an angiogram but otherwise met all other diagnostic criteria at discharge were also enrolled. A phone interview was scheduled for women expressing interest in study participation. After obtaining verbal informed consent and authorization to access medical records, research personnel conducted the study interview w1 month after discharge using a standardized script to elicit information consistently across participants. Both the interviewer and the study participants were blinded to the study outcomes and to participants’ case or control status. The study was approved by the institutional review board at each participating institution. Research Electronic Data capture technology was used to conduct computer-assisted interviews with direct data entry into electronic study surveys. Information about reproductive characteristics was collected using a survey modeled after the reproductive history questionnaire used for the Women’s Health Initiative Observational Study.16 The information collected included: age at first (“How old were you when you had your first menstrual period?”) and last menstrual period (How old were you when you last had any menstrual bleeding?), history of menstrual irregularities (“during most of your life, were your periods regular, i.e., did they occur about once a month, not including any time when your were pregnant or taking birth control pills?”), history of ovarian dysfunction (“Between your first and your last period, did you go without any period for at least a year, not counting any time when your were pregnant or taking birth control pills?”), infertility (“Have you ever tried to become pregnant for more than one year without becoming pregnant?”), number of pregnancies, breastfeeding history (“Did you breastfeed any children for at least one month?”), history of menopausal symptoms (“Have you ever had menopausal symptoms such as hot flashes or night sweats?”), history of oophorectomy or hysterectomy and age at surgery, oral contraceptive use and duration of use (“Did you ever take birth control pills, diethylstilbestrol, or
shots called Depo-Provera for birth control or any other reasons?”), use of hormone replacement therapy (HRT) (“Did you take hormone replacement therapy, that is, hormones that are taken around the time of menopause or after menopause, not including hormones used for birth control”), and duration of use. Information about age, race/ethnicity, marital status, income, and education was collected directly from the participants using standardized questionnaires. Information about medical history (including coronary risk factors) was abstracted from the medical record with the exception of physical activity (frequency of walking outside the home for at least 10 minutes without stopping) and family history of coronary heart disease (CHD) (self-reported questionnaire). Body mass index was calculated from height and weight measurements at admission reported in the medical record. Demographic and reproductive characteristics were compared using ANOVA (or nonparametric tests where applicable) for continuous measures and chi-square statistics for categorical variables. Generalized linear models and logistic regression models (adjusted for variables that were associated with the outcomes of interest with p 0.1) were used to estimate associations with reproductive characteristics, with HC as the reference group. For all estimates, p values or 95% confidence intervals were calculated. All study analyses were performed using SAS, version 9.3, statistical software. Results From March 2013 to October 2015, 107 women (45 TC cases, 32 MI controls [STEMI, n ¼ 24; non-STEMI, n ¼ 8]) and 30 HC were enrolled in the study (Figure 1). HC had higher income and education and were slightly older compared with MI and TC, whereas other demographic characteristics were similar across groups (Table 1). Although diabetes and dyslipidemia, as well as a higher body mass index, were more common among MI controls, a history of smoking, a family history of CHD, and hypertension were also highly prevalent among TC cases. Clinical characteristics at admission were fairly similar between TC cases and MI controls with the exception of peak troponin levels, ejection fraction, and systolic blood pressure, which were lower in TC women. The diagnosis of TC was not angiographically confirmed in 15 women. These women, however, did fulfill all other diagnostic criteria for TC and received a final diagnosis of TC at discharge. Angiographically confirmed TC cases did not significantly differ from nonconfirmed cases in terms of baseline demographics and medical history (data not shown). Unadjusted reproductive characteristic by case-control status are listed in Table 2, whereas findings from multivariable adjusted models are presented in Tables 3 (logistic regression) and 4 (linear regression). Both TC cases and MI controls were younger than HC at their first menstrual period, with no significant differences between TC and MI. Women with TC were more likely to report irregular menses compared with HC and MI: in adjusted models, TC women were 8 times more likely than MI controls to have a history of irregular menses. The prevalence of a history of infertility was similar across
Cardiomyopathy/Takotsubo Cardiomyopathy and Reproductive History
3
Screened = 209 Ineligible = 94 12 clinically unstable 7 intracranial bleeding 36 diagnosis not confirmed 4 HC with chronic condiƟons 6 deceased 12 unreachable 5 not English speaking 12 myocardiƟs/CABG
Eligible = 115 Refused = 4 2 not interested 2 no Ɵme
Consented = 111
Diagnosis not confirmed =4
Enrolled = 107 45 TC 32 MI 30 HC
Figure 1. STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) diagram. CABG ¼ coronary artery bypass grafting.
groups. A history of amenorrhea of at least 1 year’s duration was more prevalent among both TC cases and MI controls compared with the HC group. TC women had a significantly higher number of pregnancies compared with both HC and MI. We did not find associations between case-control status and breastfeeding history. Regarding menopausal characteristics, TC women were, on average, 3 years younger than HC at menopause onset; these differences, however, were not statistically significant. TC cases were more likely than HC to report menopausal symptoms, such as night sweats and hot flashes. Although these associations did not reach statistical significance, we found a large effect size in that TC cases were 3 times more likely than HC and 50% more likely than MI controls to report menopausal symptoms. There were no differences in the prevalence of a history of oophorectomy or hysterectomy across groups.
There were no significant differences in the use of oral contraceptives during fertile years. Among women who did report using oral contraceptives, TC cases reported to have used it for fewer years compared with HC (3.2 years; p ¼ 0.04) and MI (1.9 years, p ¼ 0.20). Women with TC were almost 6 times more likely to report HRT use compared with MI controls. Discussion This study sought to explore reproductive factors related to female endogenous hormone levels in women admitted with a recent episode of TC. Overall, we found that certain reproductive characteristics, namely a history of irregular menses, number of pregnancies, and HRT use after menopause, appeared to be uniquely associated with the onset of TC. We did not find significant associations between TC
4
The American Journal of Cardiology (www.ajconline.org)
Table 1 Baseline characteristics of cases and controls* TC ¼ 45
MI ¼ 32
Table 2 Unadjusted reproductive characteristics by case/control status HC ¼ 30
Age (years) (Mean 62.4 10.8 64.6 15.3 55.9 12.4 SD) American Indian or 1 (2%) 0 0 Alaskan Asian or Pacific 0 0 1 (3%) Islander Black or African0 1 (3%) 0 American Hispanic/Latino 3 (7%) 6 (19%) 3 (10%) White (not Hispanic) 40 (91%) 25 (78%) 26 (87%) Education High School 20 (46%) 19 (59%) 3 (10%) diploma College or some 23 (52%) 13 (41%) 24 (80%) college Post-graduate 1 (2%) 0 3 (10%) Income Less than $ 35,000 23 (52%) 17 (53%) 4 (13%) $35,000 to $74,999 9 (20%) 6 (19%) 12 (40%) $75,000 or more 9 (21%) 6 (19%) 14 (47%) Don’t know 3 (7%) 3 (9%) 0 Coronary risk factors Family history of 18 (40%) 16 (50%) 0 coronary heart disease Diabetes mellitus 8 (18%) 14 (44%) 3 (10%) Hyperlipidemia 17 (38%) 19 (59%) 8 (27%) Hypertension 26 (58%) 22 (69%) 5 (17%) Body mass index 22.3 (9.7) 29.2 (7.8) (kg/m2) Ever a smoker 31 (71%) 22 (69%) 13 (45%) Physically activez 21 (48%) 17 (53%) 20 (69%) Medical history Coronary bypass 0 2 (6%) 0 Percutaneous 0 3 (9%) 0 coronary intervention Stroke 0 2 (6%) 0 Anxiety disorder 11 (24%) 3 (9%) 0 Mood disorder 9 (20%) 7 (22%) 5 (17%) Clinical characteristics at admission 119 23 130 23 n/a Systolic blood pressure (mmHg) (Mean SD) Diastolic blood 70 17 75 19 n/a pressure (mmHg) (Mean SD) Ejection fraction (%) 40 (10) 48 (12) n/a Peak CPK (IU/L) 398.0 515.8 566.0 367.5 n/a (Mean SD) Peak troponin (ng/ml) 2.9 3.2 6.8 7.8 n/a (Mean SD) Heart rate (bpm) 91 26 86 18 n/a (Mean SD)
p†
TC ¼ 45
0.08
Age at first menses 12.7 1.6 12.7 1.3 13.3 1.8 (mean SD) History of irregular 9 (21%) 1 (0%) 0 menstrual periods Ever pregnant 39 (91%) 28 (90%) 23 (82%) Number of pregnancies 3.4 1.6 2.8 1.3 2.7 1.5 (mean SD) Ever breastfeeding for at 17 (40%) 11 (36%) 14 (52%) least 1 month Ever taken birth control 26 (58%) 13 (41%) 21 (70%) pill Years on birth control pill 3.7 3.4 5.2 5.5 5.9 4.5 (mean SD) History of infertility 1 (2%) 0 1 (4%) History of ovarian 3 (7%) 3 (10%) 0 dysfunction† Previous oophorectomy 15 (35%) 7 (23%) 10 (36%) Age at oophorectomy 41.0 10 44.7 11.6 45.1 9.7 (mean SD) Previous hysterectomy 14 (33%) 8 (27%) 10 (39%) Age at last menstrual 45.6 8.9 45.6 7.4 45.7 8.3 bleeding (mean SD) Menopausal symptoms 30 (70%) 17 (55%) 12 (43%) (Night sweats, hot flashes) Hormone replacement 13 (30%) 2 (6%) 4 (15%) therapy
0.49 0.28 0.31 0.26 0.28 <.001 0.006 0.09 0.001 0.10 0.02 0.25 <.001
0.004 0.03 <.001 0.009 0.06 0.20 0.09 0.03
0.09 0.007 0.87 0.04
0.24
0.02 0.63 0.01 0.37
* Values are n (%) unless otherwise indicated. † ANOVA or chi-square. z Walking outdoors for at least 10 minutes once a week.
and a history of infertility, breastfeeding, hysterectomy or oophorectomy, birth control pill use, and age at menopause. Taken together, these findings indicate that although studies
MI ¼ 32
HC ¼ 30
p* 0.21 .009 0.50 0.14 0.42 0.06 0.23 0.58 0.26 0.45 0.58 0.64 0.99 0.07 0.12
* ANOVA or chi-square. † Defined as history of amenorrhea of at least one year during fertile years.
conducted in animal models suggest that low estrogen levels may play a role in the physiopathology of TC,7e9 in humans, the picture is likely more complex and that, rather than estrogen levels per se, premenopausal factors may play an important role in the onset of TC at a later age. The epidemiologic evidence shows that few cardiovascular events occur before the onset of menopause, and in fact, the loss of cyclical ovarian function occurring with menopause is associated with the development of atherosclerosis and CHD in women.17 During fertile years, estrogens protect from atherosclerosis through favorable effects on lipoprotein levels, vascular reactivity, fat distribution, and hemostasis.18,19 Although this is the first study reporting an association between history of menstrual irregularity and the occurrence of TC, there is a growing body of evidence linking premenopausal abnormalities of ovarian function to the onset of CHD in older age. A prospective cohort study of 82,439 female nurses found that women with irregular cycles had an increased risk for nonfatal or fatal CHD compared with women with a history of regular menstrual cycles after adjustment for potential confounders.20 Similarly, a large prospective cohort study conducted among >23,000 women in The Netherlands found that women with irregular cycles had a 27% higher risk of CHD compared with those with a history of regular cycles, independent of common coronary risk factors.21 Another large prospective study using administrative data has linked irregular cycles and cardiovascular mortality, although risk differences were no longer
Cardiomyopathy/Takotsubo Cardiomyopathy and Reproductive History Table 3 Results of logistic regression analyses* TC vs HC Irregular menstrual periods Unadjusted n/a† Adjustedz Ovarian dysfunctionx Unadjusted n/a† Adjustedz Menopausal symptoms Unadjusted 3.08 (1.14-8.29) Adjusted{ 3.00 (.84-10.71) Ever pregnant Unadjusted 2.12 (.52-8.7) Adjustedk 1.80 (.42-7.7) Ever breastfeeding for at least 1 month Unadjusted .61 (.23-1.60) Adjustedk .74 (.27-2.03) Previous oophorectomy Unadjusted .96 (.36-2.61) Adjusted** .57 (.16-2.10) Previous hysterectomy Unadjusted .77 (.28-2.13) Adjusted†† .78 (.20-3.05) Birth control pill Unadjusted .59 (.22-1.56) Adjusted** .71 (.20-2.50) Hormone replacement therapy Unadjusted 2.64 (.77-9.07) Adjusted 2.96 (.82-10.66)
5
Table 4 Results of linear regression analyses* TC vs MI 8.86 (1.07-73.18) 8.30 (1.01-69.18) .70 (.13-3.72) .73 (.14-3.94) 1.90 (.73-4.97) 1.55 (.51-4.43) 1.05 (.21-5.04) 1.15 (.23-5.66) 1.19 (.46-3.09) 1.07 (.40-2.84) 1.84 (.64-5.25) 1.98 (.57-6.89) 1.33 (.47-3.72) 1.19 (.37-3.89) 2.00 (.80-5.02) 1.99 (.71-5.60) 6.09 (1.27-29.29) 5.79 (1.20-28.02)
* Data are OR (95% CI). † OR could not be computed when comparing to cells with 0 cell count or very low cell count (n ¼ 1). z Adjusted for anxiety, education. x Defined as history of amenorrhea of at least one year during fertile years. { Adjusted for age, anxiety. k Adjusted for education. ** Adjusted for age. †† Adjusted for age, income.
significant after adjustment for body mass index.22 The finding of an association between the occurrence of TC and menstrual irregularities suggests that a lifelong exposure to low estrogen levels (like that occurring with even mild disruptions of ovarian function) may play an important role in the onset of TC.23,24 Cardiovascular changes associated with pregnancy are so vast that they led some investigators to define pregnancy as a “9-month-long cardiovascular stress test.”24 A secondary analysis from 3 case-control studies has shown that parity was positively associated with an increased risk of nonfatal MI,25 whereas a prospective analysis from the Multi-Ethnic Study of Atherosclerosis cohort found that the number of live births was associated with increases in left ventricular mass and end-diastolic and end-systolic volume and decreases in ejection fraction in middle to older age, independent of demographic and CHD risk factors.26 Although ours is the first study showing a significant association between number of pregnancies and TC, this finding deserves further study, in particular as to how repeated pregnancies may predispose women to develop TC later in life.
TC vs HC Age at first menses Unadjusted -0.60 (.36) 0.70 (.37) Adjusted† Age at last menstrual bleeding Unadjusted -0.01 (1.99) Adjustedz -3.04 (2.17) Pregnancies Unadjusted 0.71 (0.38) Adjustedz 1.08 (0.44) Age at oophorectomy Unadjusted -4.11 (4.09) Adjusted -7.24 (4.82) Years on birth control pill Unadjusted -2.20 (1.25) Adjustedz -3.20 (1.54) Years on hormone replacement therapy Unadjusted -7.66 (3.87) Adjustedz 1.11 (4.23)
p
TC vs MI
p
.10 .05
0.02 (0.36) 0.08 (0.36)
.96 .83
0.9 .16
0.17 (1.93) 1.33 (1.84)
0.93 0.47
.06 .01
0.54 (0.36) 0.69 (0.35)
.14 .05
.31 .13
-3.67 (4.69) -3.01 (5.14)
.43 .56
.07 .04
-1.47 (1.43) -1.87 (1.53)
.30 .22
.05 .79
2.59 (5.14) 3.99 (4.21)
.61 .34
* Data are beta coefficients (SE). † Adjusted for education. z Adjusted for age.
We also found that TC women were almost 6 times more likely than MI controls to report use of post-menopausal HRT (differences with HC were in the same direction but were not statistically significant). Information on the role of HRT in the onset of TC is scarce and has largely focused on HRT use at the time of the hospital admission. A recent retrospective study found a higher prevalence of current HRT in both TC and MI women compared with HC.11 A small case series and literature review found that no TC woman was receiving HRT at hospital admission.12 Although our study differed from these studies in that we focused on history of menopausal HRT use, instead of current use, our findings are consistent with those from randomized clinical trials showing that HRT may actually increase the risk of cardiovascular events in postmenopausal women.27 A possible explanation for the higher prevalence of HRT use among TC cases versus MI controls is confounding by indication, for example, TC cases were more likely to be on HRT because they had fewer contraindications to HRT. The important message from this finding, however, is that HRT in menopause was not protective, thus suggesting that estrogen levels per se may not play an important role in the onset of this condition. Although there is a relative abundance of information linking female reproductive history and CHD, there is very limited information about reproductive characteristics of TC women. A small study that compared estradiol, progesterone, luteinizing hormone, and follicle-stimulating hormone levels in women with SC (n ¼ 17), age-matched women with MI (n ¼ 16), and women with normal coronary arteries (n ¼ 15) found higher concentrations of estradiol in TC compared with women with MI or women with normal coronary arteries, but, as the investigators note, estrogen levels were measured at hospital admission and could have increased in response to stress.10 A retrospective casecontrol study conducted among 25 consecutive female
6
The American Journal of Cardiology (www.ajconline.org)
cases of TC, age-matched and event dateematched control women with ST-segment elevation MI (n ¼ 25), and women with neither diagnosis (n ¼ 50) found no differences in (crude) prevalence of menopausal status, frequency of oophorectomy, and average years since the onset of menopause.11 This study has several important strengths. First, this is the first extensive study of reproductive characteristics of TC women. Although findings need to be confirmed in a larger study, our data suggest that reproductive history could be an essential component of the assessment of cardiovascular risk among women. Other strengths include a rigorous case-control design with enrollment of incident cases and controls, standardized procedures for data collection, and blinding of study staff conducting study interviews. Some important limitations need to be acknowledged. First, this study had a relatively small sample size, which limited the power to detect associations. Although we have been cautious in dismissing nonsignificant associations with large magnitude, results from this study need to be reproduced in larger samples. We would like to note that associations with menstrual irregularities and number of pregnancies appear to be both biologically plausible and consistent with epidemiologic evidence in women with CHD. Second, as for all case-control studies, recall bias is an important limitation. We note that in the present study interviews for TC and AMI women were conducted at least 1 month after discharge to minimize the potential effects of hospitalization-related distress on patient-reported reproductive history. It is also unlikely that TC cases would recall their reproductive history differently than MI controls. Third, information on reproductive factors was selfreported. Although women are quite reliable when reporting details about their reproductive history,28,29 nonetheless we were not able to collect detailed information about the characteristics and duration of menstrual irregularities or to document a definite diagnosis, such a polycystic ovary syndrome or functional hypothalamic dysfunction. A final limitation concerns our selection of HC, which were recruited from a registry of research volunteers. With the exception of higher socioeconomic status and age (which were accounted for in the analyses), however, no other group differences were associated with the outcomes of interest. Acknowledgment: The authors wish to thank Sandhya Reddy, MD (data abstractor) and Nancy Mecone, BSN, RN (research coordinator) for their collaboration on this study. Disclosures The American Heart Association did not have any role in the design and conduct of the study, in the collection, analysis, and interpretation of the data, and in the preparation, review, or approval of the manuscript. The authors have no conflicts of interest to disclose. 1. Dote K, Sato H, Tateishi H, Uchida T, Ishihara M. Myocardial stunning due to simultaneous multivessel coronary spasms: a review of 5 cases. J Cardiol 1991;21:203e214.
2. Eitel I, von Knobelsdorff-Brenkenhoff F, Bernhardt P, Carbone I, Muellerleile K, Aldrovandi A, Francone M, Desch S, Gutberlet M, Strohm O, Schuler G, Schulz-Menger J, Thiele H, Friedrich M. Clinical characteristics and cardiovascular magnetic resonance findings in stress (takotsubo) cardiomyopathy. JAMA 2011;306:277e286. 3. Pilgrim TM, Wyss TR. Takotsubo cardiomyopathy or transient left ventricular apical ballooning syndrome: a systematic review. Int J Cardiol 2008;124:283e292. 4. Templin C, Ghadri JR, Diekmann J, Napp LC, Bataiosu DR, Jaguszewski M, Cammann VL, Sarcon A, Geyer V, Neumann CA, Seifert B, Hellermann J, Schwyzer M, Eisenhardt K, Jenewein J, Franke J, Katus HA, Burgdorf C, Schunkert H, Moeller C, Thiele H, Bauersachs J, Tschöpe C, Schultheiss HP, Laney CA, Rajan L, Michels G, Pfister R, Ukena C, Böhm M, Erbel R, Cuneo A, Kuck KH, Jacobshagen C, Hasenfuss G, Karakas M, Koenig W, Rottbauer W, Said SM, Braun-Dullaeus RC, Cuculi F, Banning A, Fischer TA, Vasankari T, Airaksinen KE, Fijalkowski M, Rynkiewicz A, Pawlak M, Opolski G, Dworakowski R, MacCarthy P, Kaiser C, Osswald S, Galiuto L, Crea F, Dichtl W, Franz WM, Empen K, Felix SB, Delmas C, Lairez O, Erne P, Bax JJ, Ford I, Ruschitzka F, Prasad A, Lüscher TF. Clinical features and outcomes of takotsubo (stress) cardiomyopathy. N Engl J Med 2015;373:929e938. 5. Wittstein IS, Thiemann DR, Lima JA, Baughman KL, Schulman SP, Gerstenblith G, Wu KC, Rade JJ, Bivalacqua TJ, Champion HC. Neurohumoral features of myocardial stunning due to sudden emotional stress. N Engl J Med 2005;352:539e548. 6. Abraham J, Mudd JO, Kapur NK, Klein K, Champion HC, Wittstein IS. Stress cardiomyopathy after intravenous administration of catecholamines and beta-receptor agonists. J Am Coll Cardiol 2009;53: 1320e1325. 7. Ueyama T, Ishikura F, Matsuda A, Asanuma T, Ueda K, Ichinose M, Kasamatsu K, Hano T, Akasaka T, Tsuruo Y, Morimoto K, Beppu S. Chronic estrogen supplementation following ovariectomy improves the emotional stress-induced cardiovascular responses by indirect action on the nervous system and by direct action on the heart. Circ J 2007;71: 565e573. 8. Ueyama T, Hano T, Kasamatsu K, Yamamoto K, Tsuruo Y, Nishio I. Estrogen attenuates the emotional stress-induced cardiac responses in the animal model of Tako-tsubo (Ampulla) cardiomyopathy. J Cardiovasc Pharmacol 2003;42 Suppl 1:S117eS119. 9. Cao X, Zhou C, Chong J, Fu L, Zhang L, Sun D, Hou H, Zhang Y, Li D, Sun H. Estrogen resisted stress-induced cardiomyopathy through increasing the activity of beta(2)AR-Galphas signal pathway in female rats. Int J Cardiol 2015;187:377e386. 10. Brenner R, Weilenmann D, Maeder MT, Jorg L, Bluzaite I, Rickli H, De Pasquale G, Ammann P. Clinical characteristics, sex hormones, and long-term follow-up in Swiss postmenopausal women presenting with Takotsubo cardiomyopathy. Clin Cardiol 2012;35:340e347. 11. Scantlebury DC, Prasad A, Rabinstein AA, Best PJ. Prevalence of migraine and Raynaud phenomenon in women with apical ballooning syndrome (Takotsubo or stress cardiomyopathy). Am J Cardiol 2013;111:1284e1288. 12. Kuo BT, Choubey R, Novaro GM. Reduced estrogen in menopause may predispose women to takotsubo cardiomyopathy. Gend Med 2010;7:71e77. 13. Sato A, Yagihara N, Kodama M, Mitsuma W, Tachikawa H, Ito M, Hanawa H, Aizawa Y. Takotsubo cardiomyopathy after delivery in an oestrogen-deficient patient. Int J Cardiol 2011;149:e78ee79. 14. Prasad A, Lerman A, Rihal CS. Apical ballooning syndrome (Takotsubo or stress cardiomyopathy): a mimic of acute myocardial infarction. Am Heart J 2008;155:408e417. 15. Thygesen K, Alpert JS, Jaffe AS, Simoons ML, Chaitman BR, White HD, Writing Group on the Joint ESC/ACCF/AHA/WHF Task Force for the Universal Definition of Myocardial Infarction, Thygesen K, Alpert JS, White HD, Jaffe AS, Katus HA, Apple FS, Lindahl B, Morrow DA, Chaitman BA, Clemmensen PM, Johanson P, Hod H, Underwood R, Bax JJ, Bonow RO, Pinto F, Gibbons RJ, Fox KA, Atar D, Newby LK, Galvani M, Hamm CW, Uretsky BF, Steg PG, Wijns W, Bassand JP, Menasché P, Ravkilde J, Ohman EM, Antman EM, Wallentin LC, Armstrong PW, Simoons ML, Januzzi JL, Nieminen MS, Gheorghiade M, Filippatos G, Luepker RV, Fortmann SP, Rosamond WD, Levy D, Wood D, Smith SC, Hu D, Lopez-Sendon JL, Robertson RM, Weaver D, Tendera M, Bove AA, Parkhomenko AN, Vasilieva EJ, Mendis S; ESC Committee for
Cardiomyopathy/Takotsubo Cardiomyopathy and Reproductive History
16. 17. 18. 19. 20.
21.
22.
Practice Guidelines (CPG). Third universal definition of myocardial infarction. Eur Heart J 2012;33:2551e2567. Design of the Women’s Health Initiative clinical trial and observational study. The Women’s Health Initiative Study Group. Control Clin Trials 1998;19:61e109. Kannel WB, Hjortland MC, McNamara PM, Gordon T. Menopause and risk of cardiovascular disease: the Framingham study. Ann Intern Med 1976;85:447e452. Schwertz DW, Penckofer S. Sex differences and the effects of sex hormones on hemostasis and vascular reactivity. Heart Lung 2001;30:401e426. Vitale C, Mendelsohn ME, Rosano GM. Gender differences in the cardiovascular effect of sex hormones. Nat Rev Cardiol 2009;6:532e542. Solomon CG, Hu FB, Dunaif A, Rich-Edwards JE, Stampfer MJ, Willett WC, Speizer FE, Manson JE. Menstrual cycle irregularity and risk for future cardiovascular disease. J Clin Endocrinol Metab 2002;87:2013e2017. Gast GC, Grobbee DE, Smit HA, Bueno-de-Mesquita HB, Samsioe GN, van der Schouw YT. Menstrual cycle characteristics and risk of coronary heart disease and type 2 diabetes. Fertil Steril 2010;94: 2379e2381. Wang ET, Cirillo PM, Vittinghoff E, Bibbins-Domingo K, Cohn BA, Cedars MI. Menstrual irregularity and cardiovascular mortality. J Clin Endocrinol Metab 2011;96:E114eE118.
7
23. Kaplan JR, Manuck SB. Ovarian dysfunction and the premenopausal origins of coronary heart disease. Menopause 2008;15:768e776. 24. Shufelt C, Waldman T, Wang E, Merz CN. Female-specific factors for IHD: across the reproductive lifespan. Curr Atheroscler Rep 2015;17: 481e486. 25. Bertuccio P, Tavani A, Gallus S, Negri E, La Vecchia C. Menstrual and reproductive factors and risk of non-fatal acute myocardial infarction in Italy. Eur J Obstet Gynecol Reprod Biol 2007;134:67e72. 26. Parikh NI, Lloyd-Jones DM, Ning H, Ouyang P, Polak JF, Lima JA, Bluemke D, Mittleman MA. Association of number of live births with left ventricular structure and function. The Multi-Ethnic Study of Atherosclerosis (MESA). Am Heart J 2012;163:470e476. 27. Boardman HM, Hartley L, Eisinga A, Main C, Roque i Figuls M, Bonfill Cosp X, Gabriel Sanchez R, Knight B. Hormone therapy for preventing cardiovascular disease in post-menopausal women. Cochrane Database Syst Rev 2015;3:CD002229. 28. Bosetti C, Tavani A, Negri E, Trichopoulos D, La Vecchia C. Reliability of data on medical conditions, menstrual and reproductive history provided by hospital controls. J Clin Epidemiol 2001;54:902e906. 29. Hunter DJ, Manson JE, Colditz GA, Chasan-Taber L, Troy L, Stampfer MJ, Speizer FE, Willett WC. Reproducibility of oral contraceptive histories and validity of hormone composition reported in a cohort of US women. Contraception 1997;56:373e378.