Abstract
The present study included 14,550 postmenopausal female participants in the UK Biobank who completed cardiac magnetic resonance imaging. Earlier age at menopause was significantly and independently associated with smaller left ventricular end-diastolic volume and smaller stroke volume, a pattern suggesting acceleration of previously described age-related left ventricular remodeling. These findings may have implications for understanding mechanisms of heart failure, specifically heart failure with preserved ejection fraction, among women with early menopause.
Premature menopause1–3 has been associated with elevated risk of heart failure (HF), but the mechanisms underlying these associations remain unclear. Although aging is associated with progressively smaller left ventricular end-diastolic volume (LVEDV),4 whether premature age at menopause is associated with altered LV structural indices is unknown. Analysis of differences in LV morphology may yield insights into HF phenotype and mechanisms among women with early menopause.
Methods
We included women in the observational UK Biobank who completed cardiac magnetic resonance (CMR) imaging5 and who were 50–75 years old and postmenopausal at the time of CMR. Age at menopause, age at menarche, parity, and menopausal hormone therapy (MHT) use were ascertained by participant self-report. Women with both natural (ie, spontaneous) menopause and surgical menopause were included. Women who were premenopausal, had missing or unknown age at menopause, or had congenital heart disease were excluded. Alcohol consumption and tobacco use were captured from self-report. Hypertension, type 2 diabetes mellitus, and coronary artery disease were ascertained by self-report or by the presence of a qualifying International Classification of Diseases code in the subject’s medical record (Supplementary Table I). Clinical covariates were updated with follow-up data between study enrollment and CMR.
CMR was performed on 1.5-T scanners (MAGNETOM Aera, Syngo Platform vD13A, Siemens Healthcare) with electrocardiographic gating for cardiac synchronization at 3 UK centers between 2014 and 2019.5 Postprocessing was performed using cvi42 Version 5.1.1.6 A validated machine-learning algorithm analyzed CMR images to extract LVEDV and LV end-systolic volume (LVESV),7 from which stroke volume (SV) and LV ejection fraction (LVEF) were derived. LV mass was available in a subset of participants, derived from manually traced endocardial and epicardial borders with papillary muscles excluded from LV mass as previously described.6 The LV mass-to-volume ratio was calculated as LV mass divided by LVEDV.
Multivariable linear regression tested associations of menopausal age with LVEDV, LDESV, SV (both crude and indexed for body-surface area using the Mosteller formula), and with LVEF; multivariable-adjusted change in these structural indices per 5-year difference in age at menopause constituted the co-primary study outcomes. Models were adjusted for CMR scanner, age, race, current or former smoking, alcohol consumption, age at menarche, parity, ever-use of MHT, hypertension, type 2 diabetes, and coronary artery disease. Secondary models tested multivariable-adjusted change in LV mass and mass-to-volume ratio; others compared women with age at menopause <45 years and ≥55 years to women with age at menopause 45–54 years. Sensitivity analyses tested associations among women who never used MHT and those with age at menopause ≥50 years. Two-sided P < .05 was considered statistically significant.
This research was supported by grants from the US National Heart, Lung, and Blood Institute and by a Harvard Medical School John S. LaDue Fellowship. The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the paper, and its final contents.
Results
Among 14,550 women included, mean (SD) age at CMR was 63.9 (6.1) years. Mean (SD) time from study enrollment to CMR was 8.8 (1.7) years. Most women (97.3%) were white. Mean age at menopause was 50.4 (4.5) years. A total of 1,304 women (9.0% of the overall cohort) experienced menopause before age 45 years (378 [2.6%] with surgical menopause and 926 [6.4%] with natural menopause). As expected, women with menopause before age 45 were more likely to have ever used MHT (Table I).
Table I.
Clinical and cardiac MRI characteristics of the study cohort
| Age at menopause, y | ||||
|---|---|---|---|---|
| <45 (n = 1304) |
45–54 (n = 10,980) |
≥55 (n = 2266) |
P value | |
| Age at CMR imaging, y | 64.3 (6.5) | 63.4 (6.1) | 65.7 (5.1) | <.001 |
| White | 1275 (97.8%) | 10,670 (97.2%) | 2217 (97.8%) | .14 |
| Age at menopause, y | 40.2 (3.9) | 50.5 (2.4) | 56.1 (1.4) | <.001 |
| Age at menarche, y | 13.0 (1.7) | 13.0 (1.5) | 13.0 (1.6) | .90 |
| Reproductive duration (difference between age at menopause and menarche), y | 27.2 (4.3) | 37.5 (2.9) | 43.2 (2.1) | <.001 |
| Final parity (median [IQR]) | 2 (1–2) | 2 (1–2) | 2 (1–3) | <.001 |
| History of gestational hypertension/preeclampsia | 12 (0.9%) | 109 (1.0%) | 16 (0.7%) | .44 |
| History of hysterectomy | 189 (17.5%) | 280 (2.6%) | 32 (1.4%) | <.001 |
| History of bilateral oophorectomy | 127 (9.9%) | 316 (2.9%) | 29 (1.3%) | <.001 |
| Current or former smoking | 565 (43.3%) | 3,905 (35.6%) | 790 (34.9%) | <.001 |
| Alcohol consumption (median [IQR]) | 1–2 times weekly (1–3 times monthly to 3–4 times weekly) | 1–2 times weekly (1–3 times monthly to 3–4 times weekly) | 1–2 times weekly (1–3 times monthly to 3–4 times weekly) | .58 |
| Exercise ≥2 times weekly | 384 (29.4%) | 3599 (32.8%) | 780 (34.4%) | <.001 |
| Ever-use of menopausal hormone therapy | 806 (61.9%) | 3408 (27.8%) | 667 (29.5%) | <.001 |
| Duration of hormone therapy use among prior users | 7.7 (5.8) | 4.6 (4.0) | 5.2 (4.6) | <.001 |
| Body mass index, kg/m2 | 26.4 (4.6) | 25.8 (4.4) | 26.3 (4.5) | <.001 |
| Systolic blood pressure, mm Hg | 131.6 (18.2) | 130.8 (17.7) | 134.5 (18.1) | <.001 |
| Diastolic blood pressure, mm Hg | 79.6 (9.7) | 79.3 (9.7) | 80.3 (9.7) | <.001 |
| Type 2 diabetes mellitus | 30 (2.3%) | 143 (1.3%) | 37 (1.6%) | .01 |
| Hypertension | 297 (22.8%) | 2058 (18.7%) | 503 (22.2%) | <.001 |
| Coronary artery disease | 12 (0.9%) | 76 (0.7%) | 17 (0.8%) | .65 |
| Heart failure | 3 (0.2%) | 17 (0.2%) | 4 (0.2%) | .64 |
| Chronic kidney disease | 6 (0.5%) | 35 (0.3%) | 10 (0.4%) | .43 |
| Antihypertensive medication use | 158 (12.1%) | 990 (9.0%) | 259 (11.4%) | <.001 |
| Cholesterol-lowering medication use | 115 (8.8%) | 647 (5.9%) | 156 (6.9%) | <.001 |
| Total cholesterol, mg/dL | 230.8 (42.3) | 227.1 (40.8) | 229.1 (40.4) | .003 |
| High-density lipoprotein cholesterol, mg/dL | 63.3 (14.7) | 63.7 (14.3) | 63.5 (13.8) | .52 |
| Low-density lipoprotein cholesterol, mg/dL | 142.5 (32.2) | 139.4 (31.5) | 140.1 (31.8) | .002 |
| Triglycerides, mg/dL (median [IQR]) | 114.2 (84.5–162.4) | 105.6 (78.6–148.3) | 110.2 (81.4–154.3) | <.001 |
| High-sensitivity C-reactive protein, mg/L (median [IQR]) | 1.4 (0.7–2.8) | 1.0 (0.5–2.1) | 1.1 (0.6–2.3) | <.001 |
| LVEDV, mL | 121.1 (20.0) | 122.6 (19.9) | 122.0 (19.8) | .01 |
| LVEDVi, mL/m2 | 68.4 (10.5) | 69.8 (10.4) | 69.0 (10.3) | <.001 |
| LVESV, mL | 41.1 (11.4) | 41.5 (11.3) | 41.1 (11.2) | .24 |
| LVESVi, mL/m2 | 23.2 (6.1) | 23.6 (6.1) | 23.2 (6.0) | .007 |
| SV, mL | 79.9 (12.1) | 81.2 (12.1) | 80.9 (12.1) | .002 |
| SVi, mL/m2 | 45.2 (6.5) | 46.2 (6.5) | 45.8 (6.4) | <.001 |
| LVEF, % | 66.4 (5.3) | 66.5 (5.2) | 66.7 (5.2) | .30 |
| LV mass,* g | 71.5 (14.9) | 73.6 (15.0) | 75.2 (14.5) | .04 |
| LV mass-to-volume ratio,* g/mL | 0.61 (0.13) | 0.60 (0.11) | 0.62 (0.11) | .03 |
Numbers are displayed for continuous variables as mean (SD), unless otherwise specified, and for binary variables as count (proportion). Participant characteristics were compared using analysis of variance or the Kruskal-Wallis test, as appropriate, for continuous variables and using the Pearson χ2 test or Fisher exact test, as appropriate, for categorical variables.
LV mass was available in 1,707 women (11.7% of total cohort).
Overall, the mean (SD) LVEDV was 122.4 (19.9) mL, LVESV was 41.4 (11.3) mL, and SV was 81.0 (12.1) mL. Unadjusted LVEDV, LVEDV index (LVEDVi), SV, SV index (SVi), and LV mass were smaller in women with menopause before age 45 years compared with other women (Table I). After multivariable adjustment, each 5 years of earlier menopausal age was independently associated with reduced LVEDV by 0.44 mL (95% CI 0.08–0.80 mL, P = .02). Although the association with LVEDVi did not reach significance (Table II), a post hoc model for LVEDV additionally adjusting for body mass index retained nominal significance (β=−0.41 mL per 5 years of earlier menopause [95% CI −0.76 to −0.06 mL], P = .02). There was no association with LVESV or LVESV index (LVESVi) (Table II). Per 5 years of earlier menopausal age, SV and SVi were significantly smaller by 0.44 mL (95% CI 0.22−0.65 mL, P < .001) and 0.16 mL/m2 (95% CI 0.04−0.27 mL/m2, P = .008), respectively. Women with menopause <45 years had, on average, an SVi similar to women 3.3 years older for whom menopause occurred at age 45–54 years (Figure 1). A small, nominally significant association with LVEF was present (−0.10% per 5 years of earlier menopause [95% CI −0.20% to −0.01%], P = .04).
Table II.
Associations of clinical predictors with left ventricular (LV) structural indices measured by cardiac magnetic resonance imaging among 14,550 postmenopausal women in the UK Biobank Values are displayed as β (95% CI).
| Predictor | LV end-diastolic volume, mL | LV end-diastolic volume index, mL/m2 | LV end-systolic volume, mL | LV end-systolic volume index, mL/m2 | Stroke volume, mL | Stroke volume index, mL/m2 | LV ejection fraction, % | LV mass,+ g | LV mass-to-volume ratio,+ g/mL |
|---|---|---|---|---|---|---|---|---|---|
| Menopausal age, per 5 years of earlier menopausal age | −0.44 (−0.80 to −0.08) | 0.12 (−0.06 to 0.31) | 0.01 (−0.20 to 0.21) | −0.03 (−0.15 to 0.08) | −0.44 (−0.65 to −0.22) | −0.16 (−0.27 to −0.04) | 0.10 (0.02 to 0.20) | −1.32 (−2.06 to −0.57) | −0.006 (−0.01 to − 0.001) |
| Participant age at imaging, per year | −0.82 (−0.88 to −0.76) | −0.41 (−0.44 to −0.38) | −0.31 (−0.34 to −0.27) | −0.16 (−0.17 to − 0.14) | −0.51 (−0.55 to −0.47) | −0.25 (−0.27 to −0.23) | 0.03 (0.02 to 0.05) | −0.15 (−0.28 to − 0.01) | 0.002 (0.001 to 0.003) |
| Non-White race | −9.69 (−11.79 to − 7.60) | −3.21 (−4.30 to −2.13) | −4.99 (−6.20 to −3.78) | −2.12 (−2.77 to −1.47) | −4.67 (−5.97 to −3.43) | −1.09 (−1.77 to −0.42) | 1.55 (0.99 to 2.12) | −0.84 (−5.60 to 3.92) | 0.05 (0.01 to 0.09) |
| Ever-smoking | 0.58 (−0.09 to 1.25) | −0.52 (−0.86 to − 0.17) | 0.31 (−0.08 to 0.69) | −0.11 (−0.32 to 0.10) | 0.27 (−0.13 to 0.68) | −0.41 (−0.62 to − 0.19) | −0.07 (−0.25 to 0.11) | 2.06 (0.57 to 3.56) | 0.02 (0.01 to 0.03) |
| Use of alcohol at enrollment, per incremental category of greater alcohol intake* | 0.50 (0.27 to 0.73) | 0.68 (0.56 to 0.80) | 0.09 (−0.04 to 0.22) | 0.19 (0.12 to 0.26) | 0.41 (0.27 to 0.55) | 0.49 (0.42 to 0.56) | 0.07 (0 to 0.12) | −0.17 (−0.68 to 0.33) | −0.005 (−0.008 to −0.001) |
| Type 2 diabetes | −2.39 (−5.09 to 0.31) | −6.37 (−7.77 to −4.97) | −0.58 (−2.14 to 0.98) | −2.05 (−2.86 to − 1.21) | −1.82 (−3.45 to − 0.18) | −4.32 (−5.20 to −3.45) | −0.13 (−0.86 to 0.59) | 0.99 (−3.92 to 5.90) | 0.03 (−0.01 to 0.06) |
| Hypertension | 0.99 (0.17 to 1.80) | −1.88 (−2.30 to −1.45) | −0.50 (−0.98 to −0.03) | −1.11 (−1.36 to −0.86) | 1.49 (0.99 to 1.99) | −0.77 (−1.03 to −0.50) | 0.73 (0.51 to 0.95) | 6.10 (4.43 to 7.77) | 0.05 (0.04 to 0.06) |
| Coronary artery disease | 4.98 (1.25 to 8.72) | 2.99 (1.06 to 4.92) | 4.51 (2.35 to 6.67) | 2.49 (1.33 to 3.65) | 0.47 (−1.75 to 2.73) | 0.50 (−0.70 to 1.70) | −2.11 (−3.11 to − 1.11) | −0.27 (−6.72 to 6.18) | −0.03 (−0.07 to 0.02) |
| Parity, per live birth | 0.43 (0.16 to 0.71) | 0.27 (0.12 to 0.41) | 0.28 (0.12 to 0.44) | 0.16 (0.08 to 0.25) | 0.15 (−0.01 to 0.32) | 0.10 (0.01 to 0.19) | −0.10 (−0.18 to −0.03) | 0.16 (−0.45 to 0.78) | 0 (−0.004 to 0.005) |
| Age at menarche, per year | −0.08 (0.28 to 0.13) | 0.25 (0.15 to 0.36) | −0.01 (−0.12 to 0.11) | 0.10 (0.04 to 0.16) | −0.07 (−0.19 to 0.05) | 0.15 (0.08 to 0.22) | −0.02 (−0.07 to 0.04) | −0.37 (−0.83 to 0.08) | −0.002 (−0.005 to 0.001) |
| Ever-use of menopausal hormone therapy | −1.20 (−1.98 to −0.44) | −0.62 (−1.02 to −0.22) | −0.28 (−0.72 to 0.17) | −0.13 (−0.37 to 0.11) | −0.93 (−1.40 to −0.47) | −0.49 (−0.74 to −0.24) | −0.11 (−0.32 to 0.10) | 0.13 (−1.50 to 1.77) | 0.02 (0.004 to 0.03) |
Effect estimates, 95% confidence intervals, and P-values are derived from a multivariable linear regression model adjusted for all variables listed in the Table, plus cardiac magnetic resonance imaging scanner.
Alcohol use incorporated as an ordinal variable according to the following categories: Never, special occasions only, 1–3 times monthly, 1–2 times weekly, 3–4 times weekly, daily or almost daily
LV mass was available in 1,707 women (11.7% of total cohort)
Figure 1.
Loess regression-smoothed LV SVi versus age at CMR imaging, stratified by age at menopause.
Among 1,707 women with available LV mass, as expected, older chronological age was associated with a decline in LV mass (−0.15 g/y, P = .03) but an increase in LV mass-to-volume ratio (+0.002 g/mL*y, P < .001). After multivariable adjustment, per 5 years of earlier menopause, LV mass was smaller by 1.32 g (95% CI 0.57−2.06 g, P < .001), and LV mass-to-volume ratio was smaller by 0.006 g/mL (95% CI 0.001−0.01 g/mL, P = .03).
In models comparing women with age at menopause <45 years and ≥55 years to women with menopause between age 45 and 54 years, consistent associations with LVEDVi, SVi, and LV mass were observed (Supplementary Table II). In sensitivity analyses, associations with menopausal age were similar or larger after excluding 4,521 women with ever-use of MHT: Per 5 years of earlier menopause, women who never used MHT had smaller LVEDV by 0.67 mL (95% CI 0.17–1.19 mL, P = .009) and smaller SVi by 0.25 mL/m2 (95% CI 0.09–0.41 mL/m2, P = .003).
In multivariable models, hypertension was associated with smaller LVEDVi, LVESVi, and SVi; higher LVEF; and greater LV mass and mass-to-volume ratio (Table II). In addition, ever-use of MHT was independently associated with smaller LVEDVi and SVi and greater LV mass-to-volume ratio. Associations with MHT use were similar in models restricted to women with menopausal age ≥50 years (LVEDVi: −0.76 mL/m2 [95% CI −1.26 to −0.26 mL/m2], P = .003; SVi: −0.62 mL/m2 [95% CI −0.93 to −0.31 mL/m2], P < .001).
Discussion
In the largest study to date of LV volumes versus age at menopause to our knowledge, we find evidence of accelerated LV remodeling in women with earlier menopause as evidenced by small but significant and independent reductions in LVEDV and SV, analogous to previously described age-related alterations.4 In a sub-study of the Multi-Ethnic Study of Atherosclerosis, 2,123 postmenopausal women underwent CMR; among Chinese American women, women with menopause before age 45 years had significantly smaller LVEDV and SV and higher LV mass-to-volume ratio, although similar findings were not detected in other racial/ethnic groups.8 In the present study, we observe an association between earlier menopause and smaller LVEDV and SV among 14,550 women, of whom >97% had white European ancestry. However, we also observed independent reductions in LV mass beyond the expected age-related decline and, in contrast with the Multi-Ethnic Study of Atherosclerosis study,8 lower LV mass-to-volume ratio, suggestive of relative cardiac sarcopenia in women with earlier menopause versus women of comparable chronologic age.
These findings may have implications for understanding mechanisms of HF among postmenopausal women. Compared to men with HF, women with HF are disproportionately likely to have HF with preserved ejection fraction (HFpEF).9 Postmenopausal estrogen deficiency has been proposed as a key driver of HFpEF in women, with associated activation of the reninangiotensin-aldosterone system, endothelial dysfunction, elevated arterial stiffness, and oxidative stress proposed as potential mediating pathways.10 Recent work in the current cohort indicates that premature menopause may be associated with HF risk, but analyses may have been limited because of limited available clinical HF phenotyping.1 The present analysis in a CMR subcohort indicates key LV structural alterations among women with premature menopause; the prognostic significance requires analyses in larger cohorts. Our findings lend further support to the hypothesis that early menopause may predispose to development of HFpEF, although the association between earlier menopause and lower LV mass-to-volume ratio suggests concentric hypertrophy may not drive this predisposition.
Limited data also suggest postmenopausal estrogen replacement may improve parameters of diastolic function.10 However, similar to a study of 1,604 women in the UK Biobank, we also observed that MHT use was associated with reduced LVEDV and SV in a 7-fold larger cohort.11 Furthermore, we found no evidence of confounding by an indication of premature menopause, suggesting that observed patterns of remodeling are not merely from decreased cumulative estrogen exposure in women with earlier menopause. Postmenopausal changes in other sex hormones (eg, androgens) may play a role.12
Our study has several limitations. LV mass data were available in a minority of the overall cohort. Data on MHT doses and preparations used or sex hormone levels at the time of CMR were unavailable. Potential survival and volunteer bias in participants undergoing CMR, suggested by low prevalence of comorbidities, may have minimized observed associations and biased results toward the null. Lack of clinical follow-up beyond the imaging study visit precludes assessment of the impact of differences in LV structural indices on incident HF. Finally, whether findings generalize to nonwhite women requires further study.
Overall, these data suggest that menopausal age may influence long-term cardiac morphology, with evidence of accelerated reduction in LV size, SV, and LV mass in women with early menopause. Further research is needed to clarify mechanisms linking early menopause to LV remodeling and HF.
Supplementary Material
Acknowledgements
The authors thank Seyedeh M. Zakavat, BS, and Derek Klarin, MD, for critical review of this manuscript. This research was conducted using the UK Biobank Resource under Application Number 7089.
Footnotes
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.org/10.1016/j.ahj.2020.08.005.
Contributor Information
Michael C. Honigberg, Cardiology Division, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA; Program in Medical and Population Genetics, Broad Institute of Harvard and MIT, Cambridge, MA; Cardiovascular Research Center and Center for Genomic Medicine, Massachusetts General Hospital, Boston, MA; Corrigan Women’s Heart Health Program, Massachusetts General Hospital, Boston, MA.
James P. Pirruccello, Cardiology Division, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA; Program in Medical and Population Genetics, Broad Institute of Harvard and MIT, Cambridge, MA; Cardiovascular Research Center and Center for Genomic Medicine, Massachusetts General Hospital, Boston, MA.
Krishna Aragam, Cardiology Division, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA; Program in Medical and Population Genetics, Broad Institute of Harvard and MIT, Cambridge, MA; Cardiovascular Research Center and Center for Genomic Medicine, Massachusetts General Hospital, Boston, MA.
Amy A. Sarma, Cardiology Division, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA; Corrigan Women’s Heart Health Program, Massachusetts General Hospital, Boston, MA.
Nandita S. Scott, Cardiology Division, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA; Corrigan Women’s Heart Health Program, Massachusetts General Hospital, Boston, MA.
Malissa J. Wood, Cardiology Division, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA; Corrigan Women’s Heart Health Program, Massachusetts General Hospital, Boston, MA.
Pradeep Natarajan, Cardiology Division, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA; Program in Medical and Population Genetics, Broad Institute of Harvard and MIT, Cambridge, MA; Cardiovascular Research Center and Center for Genomic Medicine, Massachusetts General Hospital, Boston, MA.
References
- 1.Honigberg MC, Zekavat SM, Aragam K, et al. Association of premature natural and surgical menopause with incident cardiovascular disease. JAMA 2019;322(24):2411–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Appiah D, Schreiner PJ, Demerath EW, et al. association of age at menopause with incident heart failure: a prospective cohort study and meta-analysis. J Am Heart Assoc 2016;5(8). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hall PS, Nah G, Howard BV, et al. Reproductive factors and incidence of heart failure hospitalization in the Women’s Health Initiative. J Am Coll Cardiol 2017;69(20):2517–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Cheng S, Fernandes VR, Bluemke DA, et al. Age-related left ventricular remodeling and associated risk for cardiovascular outcomes: the Multi-Ethnic Study of Atherosclerosis. Circ Cardiovasc Imaging 2009;2(3):191–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Petersen SE, Matthews PM, Francis JM, et al. UK Biobank’s cardiovascular magnetic resonance protocol. J Cardiovasc Magn Reson 2016;18:8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Petersen SE, Aung N, Sanghvi MM, et al. Reference ranges for cardiac structure and function using cardiovascular magnetic resonance (CMR) in Caucasians from the UK Biobank population cohort. J Cardiovasc Magn Reson 2017;19(1), 18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Sanghvi MM, Feuchter P, Zemrak F, et al. Automatic left ventricular analysis with inline VF performs well compared to manual analysis: results from Barts Cardiovascular Registry. J Cardiovasc Magn Reson 2016;18(Suppl 1):P31. [Google Scholar]
- 8.Ebong IA, Watson KE, Goff DC, et al. Age at menopause and incident heart failure: the Multi-Ethnic Study of Atherosclerosis. Menopause 2014;21(6):585–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Honigberg MC, Lau ES, Jones AD, et al. Sex differences in exercise capacity and quality of life in heart failure with preserved ejection fraction: a secondary analysis of the RELAX and NEAT-HFpEF trials. J Card Fail 2020;26(3):276–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sabbatini AR, Kararigas G. Menopause-related estrogen decrease and the pathogenesis of hfpEF. JACC review topic of the week J Am Coll Cardiol 2020;75(9):1074–82. [DOI] [PubMed] [Google Scholar]
- 11.Sanghvi MM, Aung N, Cooper JA, et al. The impact of menopausal hormone therapy (MHT) on cardiac structure and function: insights from the UK Biobank imaging enhancement study. PLoS One 2018;13(3), e0194015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Subramanya V, Zhao D, Ouyang P, et al. Sex hormone levels and change in left ventricular structure among men and post-menopausal women: the Multi-Ethnic Study of Atherosclerosis (MESA). Maturitas 2018;108:37–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.