The Severity of Individual Menopausal Symptoms, Cardiovascular Disease, and All-cause Mortality in the Women's Health Initiative Observational Cohort

Matthew Nudy, MD; Aaron K. Aragaki, MS; Xuezhi Jiang, MD, PhD, NCMP; JoAnn E. Manson, MD, DrPH, NCMP; Matthew A. Allison, MD, MPH; Aladdin H. Shadyab, PhD, MS, MPH; Howard N. Hodis, MD; Robert A. Wild, MD, MPH, PhD, NCMP; John A. Robbins, MD; Simin Liu, MD, ScD; Michelle J. Naughton, PhD, MPH; Sarah Dreibelbis, DO; Margery Gass, MD; Marcia L. Stefanick, PhD; Carolina Valdiviezo, MD; Peter F. Schnatz, DO, NCMP

Disclosures

Menopause. 2022;29(12):1365-1374.

Abstract and Introduction

Abstract

Objective: The aim of this study was to examine the association between common menopausal symptoms (MS) and long-term cardiovascular disease (CVD) and all-cause mortality.

Methods: In an observational cohort of 80,278 postmenopausal women with no known CVD at baseline from the Women's Health Initiative, we assessed individual MS severity (mild vs none; moderate/severe vs none) for night sweats, hot flashes, waking up several times at night, joint pain or stiffness, headaches or migraines, vaginal or genital dryness, heart racing or skipping beats, breast tenderness, dizziness, tremors (shakes), feeling tired, forgetfulness, mood swings, restless or fidgety, and difficulty concentrating. Outcomes included total CVD events (primary) and all-cause mortality (secondary). Associations between specific MS, their severity, and outcomes were assessed during a median of 8.2 years of follow-up. All results were multivariable adjusted, and individual associations were Bonferroni corrected to adjust for multiple comparisons. A machine learning approach (least absolute shrinkage and selection operator) was used to select the most parsimonious set of MS most predictive of CVD and all-cause mortality.

Results: The severity of night sweats, waking up several times at night, joint pain or stiffness, heart racing or skipping beats, dizziness, feeling tired, forgetfulness, mood swings, restless or fidgety, and difficulty concentrating were each significantly associated with total CVD. The largest hazard ratio (HR) for total CVD was found for moderate or severe heart racing or skipping beats (HR, 1.55; 95% confidence interval [CI], 1.29–1.86). The individual severities of heart racing or skipping beats, dizziness, tremors (shakes), feeling tired, forgetfulness, mood swings, restless or fidgety, and difficulty concentrating were associated with increased all-cause mortality. Moderate or severe dizziness had the largest HR (1.58; 95% CI, 1.24–2.01). Multiple symptom modeling via least absolute shrinkage and selection operator selected dizziness, heart racing, feeling tired, and joint pain as most predictive of CVD, whereas dizziness, tremors, and feeling tired were most predictive of all-cause mortality.

Conclusion: Among postmenopausal women with no known CVD at baseline, the severity of specific individual MS was significantly associated with incident CVD and mortality. Consideration of severe MS may enhance sex-specific CVD risk predication in future cohorts, but caution should be applied as severe MS could also indicate other health conditions.

Introduction

Several Women's Health Initiative (WHI) studies have investigated whether vasomotor symptoms (VMS),^[1] depressive symptom,^[2] sleep disturbances,^[3] migraine,^[4] panic attacks,^[5] joint pain,^[6] and cognitive impairment^[7] are associated with development of cardiovascular disease (CVD) risk factors and CVD outcomes (see table, Supplemental Digital Content 1, https://links.lww.com/MENO/B34). It is unclear whether the severity of other menopausal symptoms (MS), in addition to VMS, is associated with an increased risk of all-cause mortality and cardiovascular health outcomes including coronary heart disease (CHD), stroke, and total CVD. In particular, cognitive-affective symptoms, which are known to be associated with menopause and included on validated menopause symptom scales,^[8] were largely overlooked (see table, Supplemental Digital Content 1, https://links.lww.com/MENO/B34). A previous analysis of postmenopausal women in the WHI observational cohort indicated that those who self-reported VMS only at the time of enrollment and not at menopause onset had an increased risk of CHD and all-cause mortality. In contrast, women who only self-reported VMS during menopause onset and not later and at WHI enrollment had decreased risk of stroke, CVD, and all-cause mortality.^[1]

Among 20,050 women in the WHI–calcium and vitamin D (CaD) clinical trial, we found no association between the severity of VMS and hip fracture, colorectal cancer, invasive breast cancer, all-cause mortality, CHD, stroke, cardiovascular death, and total CVD.^[9] However, there were significant associations with stroke and total CVD for those women experiencing more moderate or severe MS. An exploratory analysis assessed individual MS severity and total CVD. This analysis found that those with moderate or severe dizziness, heart racing or skipping beats, restlessness or fidgety, feeling tired, difficulty concentrating, and forgetfulness were more likely to experience a CVD event.^[9] Given these findings, the primary objective of this study was to expand upon prior WHI reports (see table, Supplemental Digital Content 1, https://links.lww.com/MENO/B34) and analyze individual MS among women with no CVD at baseline in the WHI observational cohort. The severity (none, mild, and moderate/severe) of various cognitive-affective and somatic symptoms, beyond VMS, will be assessed for associations with incident CVD and all-cause mortality. In addition, a machine learning model was used to select the parsimonious set of symptoms most predictive of CVD and mortality.

1 2 3 4 5

Next Section

Abstract and Introduction
Methods
Results
Discussion
Conclusion
References

Table 1. Participant characteristics ^a by number of moderate or severe menopausal symptoms at baseline (N = 80,278)
Characteristic	No. moderate or severe symptoms						Total
	2+		1		None		Total
	(n = 29,140)		(n = 22,721)		(n = 28,417)		(N = 80,278)
	n	%	n	%	n	%	n	%
Age, y, mean ± SD	63.0 ± 7.5		63.7 ± 7.3		63.0 ± 7.2		63.2 ± 7.3
Age, 10 y groups
50–59	10,364	35.6	6,962	30.6	9,705	34.2	27,031	33.7
60–69	12,338	42.3	10,237	45.1	12,767	44.9	35,342	44.0
70+	6,438	22.1	5,522	24.3	5,945	20.9	17,905	22.3
Hispanic/Latino^b	1,423	4.9	746	3.3	1,068	3.8	3,237	4.1
Race^b, %
American Indian/Alaska Native	112	0.4	54	0.2	80	0.3	246	0.3
Asian/Pacific Islander	540	1.9	689	3.0	1,130	4.0	2,359	2.9
Black/African American	2,392	8.2	1,326	5.8	1911	6.7	5,629	7.0
White	25,011	85.8	20,107	88.5	24,517	86.3	69,635	86.7
Multiracial/other/unknown or not reported	1,085	3.7	545	2.4	779	2.7	2,409	3.0
Education > high school degree or GED	22,112	76.5	18,266	81.0	23,224	82.3	63,602	79.8
Smoking status
Never	14,259	49.6	11,414	50.8	15,114	53.7	40,787	51.4
Past	12,593	43.8	9,850	43.8	11,345	40.3	33,788	42.6
Current	1,908	6.6	1,202	5.4	1,662	5.9	4,772	6.0
Sleep duration, h
≤6	12,248	42.1	7,567	33.3	7,632	26.9	27,447	34.2
7–8	15,396	52.9	14,158	62.4	19,597	69.0	49,151	61.3
≥9	1,467	5.0	981	4.3	1,170	4.1	3,618	4.5
Hysterectomy	13,342	45.8	8,859	39.0	10,300	36.3	32,501	40.5
Bilateral oophorectomy	6,403	22.5	4,335	19.4	5,090	18.2	15,828	20.1
HT use status
Never used	10,953	37.6	9,085	40.0	11,854	41.7	31,892	39.8
Past user	4,871	16.7	3,151	13.9	3,576	12.6	11,598	14.5
Current user	13,288	45.6	10,465	46.1	12,963	45.7	36,716	45.8
Treated diabetes (pills or shots)	1,201	4.1	621	2.7	633	2.2	2,455	3.1
Hypertension^c	12,829	44.6	9,011	40.1	10,273	36.5	32,113	40.4
History of high cholesterol requiring pills^d	4,042	13.9	2,931	12.9	3,298	11.6	10,271	12.8
Statin use	2,039	7.0	1,539	6.8	1,735	6.1	5,313	6.6
Antihypertensive medication	9,053	31.1	5,863	25.8	6,236	21.9	21,152	26.3
Aspirin use	5,299	18.2	4,614	20.3	5,103	18.0	15,016	18.7
Age at menopause, y, mean ± SD	47.7 ± 6.6		48.6 ± 6.1		48.8 ± 5.9		48.3 ± 6.2
Body-mass index, kg/m², median [IQR]	26.9 [7.5]		25.6 [6.4]		25.3 [5.9]		25.9 [6.7]
Waist circumference, cm, mean ± SD	86.7 ± 14.4		83.6 ± 13.1		83.6 ± 13.1		84.2 ± 13.4
Systolic BP (mm Hg), mean ± SD	126.9 ± 17.8		126.4 ± 17.8		125.6 ± 17.7		126.3 ± 17.8
Diastolic BP (mm Hg), mean ± SD	75.1 ± 9.3		75.1 ± 9.3		74.6 ± 9.2		74.8 ± 9.2
Recreational physical activity (MET-h/wk), median [IQR]	7.8 [15.4]		10.8 [17.4]		12.0 [18.4]		10.3 [16.8]

Summary statistics correspond to number (percentage) of participants, mean ± SD, or median [IQR] for skewed variables (skew >1); summaries were unadjusted.
BP, blood pressure; GED, general equivalency diploma; HT, hormone therapy; IQR, interquartile range; MET, metabolic equivalent task.
^aExcept for aspirin use (P = 0.49), all pairwise associations between participant characteristic and number of moderate or severe symptoms were statistically significant (P < 0.001); computed from a cumulative logit model adjusted for age, race, ethnicity, and HT use.
^bEthnic group and race were self-reported by participants. Multiracial participants self-identified with more than one race. Participants of other races self-identified with that category.
^cSelf-report of ever taking medication for high blood pressure, or clinic-measured systolic/diastolic blood pressure ≥140/90 mm Hg.
^dSelf-report or use of antihyperlipidemic medications at enrollment.

Table 2. Prevalence of menopausal symptom severity at baseline (N = 80,278) and stratified by 10-year age group ^a
Symptom Severity	10-year age group						Total
	50–59 y		60–69 y		70–79 y		Total
	(n = 27,031)		(n = 35,342)		(n = 17,905)		(N = 80,278)
	n	(%)	n	(%)	n	(%)	n	(%)
Vasomotor symptoms
Night sweats
None	18,434	68.2	28,041	79.3	15,123	84.5	61,598	76.7
Mild	6,136	22.7	5,611	15.9	2,205	12.3	13,952	17.4
Moderate/severe	2,461	9.1	1,690	4.8	577	3.2	4,728	5.9
Hot flashes
None	17,885	66.2	28,742	81.3	15,815	88.3	62,442	77.8
Mild	6,488	24.0	5,023	14.2	1,619	9.0	13,130	16.4
Moderate/severe	2,658	9.8	1,577	4.5	471	2.6	4,706	5.9
Somatic symptoms
Did you wake up several times
None	6,775	25.1	7,861	22.2	3,619	20.2	18,255	22.7
Mild	10,790	39.9	13,583	38.4	6,569	36.7	30,942	38.5
Moderate/severe	9,466	35.0	13,898	39.3	7,717	43.1	31,081	38.7
Joint pain or stiffness
None	9,194	34.0	10,040	28.4	4,854	27.1	24,088	30.0
Mild	12,114	44.8	16,735	47.4	8,325	46.5	37,174	46.3
Moderate/severe	5,723	21.2	8,567	24.2	4,726	26.4	19,016	23.7
Headaches or migraines
None	12,332	45.6	20,218	57.2	12,246	68.4	44,796	55.8
Mild	10,818	40.0	11,788	33.4	4,517	25.2	27,123	33.8
Moderate/severe	3,881	14.4	3,336	9.4	1,142	6.4	8,359	10.4
Vaginal or genital dryness
None	19,366	71.6	25,711	72.7	13,621	76.1	58,698	73.1
Mild	5,517	20.4	7,038	19.9	3,155	17.6	15,710	19.6
Moderate/severe	2,148	7.9	2,593	7.3	1,129	6.3	5,870	7.3
Heart racing or skipping beats
None	21,289	78.8	28,514	80.7	14,630	81.7	64,433	80.3
Mild	4,808	17.8	5,733	16.2	2,659	14.9	13,200	16.4
Moderate/severe	934	3.5	1,095	3.1	616	3.4	2,645	3.3
Breast tenderness
None	21,093	78.0	28,543	80.8	15,060	84.1	64,696	80.6
Mild	4,970	18.4	5,667	16.0	2,346	13.1	12,983	16.2
Moderate/severe	968	3.6	1,132	3.2	499	2.8	2,599	3.2
Dizziness
None	22,731	84.1	29,594	83.7	14,574	81.4	66,899	83.3
Mild	3,632	13.4	4,913	13.9	2,810	15.7	11,355	14.1
Moderate/Severe	668	2.5	835	2.4	521	2.9	2,024	2.5
Tremors (shakes)
None	25,633	94.8	33,475	94.7	16,627	92.9	75,735	94.3
Mild	1,131	4.2	1,500	4.2	1,014	5.7	3,645	4.5
Moderate/Severe	267	1.0	367	1.0	264	1.5	898	1.1
Cognitive-affective symptoms
Feeling tired
None	3,259	12.1	5,089	14.4	2,655	14.8	11,003	13.7
Mild	16,914	62.6	22,750	64.4	11,334	63.3	50,998	63.5
Moderate/severe	6,858	25.4	7,503	21.2	3,916	21.9	18,277	22.8
Forgetfulness
None	12,395	45.9	15,026	42.5	6,259	35.0	33,680	42.0
Mild	12,282	45.4	17,286	48.9	9,563	53.4	39,131	48.7
Moderate/severe	2,354	8.7	3,030	8.6	2,083	11.6	7,467	9.3
Mood swings
None	15,954	59.0	23,856	67.5	12,856	71.8	52,666	65.6
Mild	8,741	32.3	9,675	27.4	4,356	24.3	22,772	28.4
Moderate/severe	2,336	8.6	1,811	5.1	693	3.9	4,840	6.0
Restless or fidgety
None	17,526	64.8	23,827	67.4	12,524	69.9	53,877	67.1
Mild	7,787	28.8	9,738	27.6	4,563	25.5	22,088	27.5
Moderate/severe	1,718	6.4	1,777	5.0	818	4.6	4,313	5.4
Difficulty concentrating
None	17,738	65.6	24,943	70.6	12,490	69.8	55,171	68.7
Mild	7,717	28.5	9,019	25.5	4,791	26.8	21,527	26.8
Moderate/severe	1,576	5.8	1,380	3.9	624	3.5	3,580	4.5

^a>Summary statistics were unadjusted.

Table 3. Multiple-symptom adjusted hazard ratios ^a associating severity of menopausal symptoms with total cardiovascular disease ^b and all-cause death during the study period and cumulative follow-up ^c
Menopausal symptom Symptom severity	Total CVD
	Study period			Cumulative follow-up
	No. events (%^d)	HR	(95% CI)	No. events (%^d)	HR	(95% CI)
Joint pain or stiffness
None or mild	4,261 (0.89)		Reference	6,360 (0.92)		Reference
Moderate or severe	1986 (1.39)	1.12	(1.05 to 1.20)	2,756 (1.37)	1.10	(1.04 to 1.16)
Heart racing or skipping beats
None	4,733 (0.95)		Reference	7,010 (0.97)		Reference
Mild	1,177 (1.16)	1.15	(1.07 to 1.23)	1,659 (1.15)	1.10	(1.04 to 1.17)
Moderate or severe	337 (1.74)	1.39	(1.22 to 1.57)	447 (1.67)	1.34	(1.21 to 1.50)
Dizziness
None	4,889 (0.94)		Reference	7,209 (0.96)		Reference
Mild, moderate or severe	1,358 (1.35)	1.12	(1.04 to 1.20)	1,907 (1.35)	1.12	(1.06 to 1.19)
Feeling tired
None or mild	4,453 (0.92)		Reference	6,611 (0.95)		Reference
Moderate or severe	1794 (1.30)	1.19	(1.12 to 1.27)	2,505 (1.29)	1.20	(1.13 to 1.27)
Menopausal symptom Symptom severity	All-cause death
	Study period			Cumulative follow-up
	No. events (%^d)	HR	(95% CI)	No. events (%^d)	HR	(95% CI)
Dizziness
None or mild	4,347 (0.70)		Reference	31,733 (2.03)		Reference
Moderate or severe	196 (1.27)	1.39	(1.18 to 1.64)	1,031 (2.73)	1.14	(1.06 to 1.23)
Tremors
None	4,159 (0.69)		Reference	24,711 (1.98)		Reference
Mild, moderate or severe	384 (1.09)	1.21	(1.12 to 1.30)	8,053 (2.26)	1.15	(1.12 to 1.18)
Feeling tired
None or mild	3,261 (0.66)		Reference	30,457 (2.01)		Reference
Moderate or severe	1,282 (0.89)	1.27	(1.13 to 1.43)	2,307 (2.70)	1.16	(1.10 to 1.21)

CVD, cardiovascular disease; LASSO, least absolute shrinkage and selection operator.
^aMenopausal symptom variables were selected with LASSO using 10-fold cross-validation of prediction error. Hazard ratios were then estimated by including all selected variables in the core multivariable-adjusted regression model for the full cohort.
^bTotal CVD was a composite of coronary heart disease, stroke, congestive heart failure, angina, peripheral vascular disease, carotid artery disease, and coronary revascularization.
^cTotal CVD events were adjudicated through September 30, 2010, with median (Q1-Q3) = 12.8 (8.9–14.3) years of follow-up. All-cause death was enhanced by National Death Index queries through December 31, 2019, and is essentially complete with median (Q1-Q3) = 21.9 (18.2–23.5) years of follow-up.
^dIncidence rates were annualized, %.