Rebecca C. Thurston, PhD Director, Women’s Biobehavioral Health Research Program University of Pittsburgh North American Menopause Society 2019 Annual Meeting, Chicago, IL, Sept 26, 2019
Funding and Disclosures Funding: Supported by the National Institutes of Health (NIH), National Institute on Aging (NIA), National Institute of Nursing Research (NINR), and the Office of Women’s Health Research (OWHR, Grants U01NR004061; U01AG012505, U01AG012535, U01AG012531, U01AG012539, U01AG012546, U01AG012553, U01AG012554, U01AG012495). Content is the responsibility of authors and does not necessarily represent the views of the NIH, NIA, NINR, or ORWH. Disclosures (Thurston): Astellas, Pfizer, Procter & Gamble (consulting)
Vasomotor Symptoms (VMS) VMS: Hot flashes / night sweats >70% of women experience during menopause transition ➢ 30% frequent or severe Frequent VMS persist for 7-10 years (Avis…Thurston et al, 2015) o VMS important for quality of life ➢ Physical health?
VMS and Cardiovascular Disease in Women o CVD leading cause of death in women o Newer research links VMS to CVD risk indicators ➢ CVD risk factors ➢ Subclinical CVD indices o Research with “hard” clinical CVD outcomes limited ➢ Recalled VMS over years ➢ Reporting and memory biases
VMS and Cardiovascular Disease in Women o Few studies are well designed to address this question ➢ Longitudinal study of menopause ➢ Regular, prospective assessment of VMS and CVD events ➢ Capture midlife and years when women have CVD events: Follow women for 20+ years o Study of Women’s Health Across the Nation ➢ Uniquely positioned to address this question
Vasomotor Cardiovascular Symptoms Disease Events Demographic and CVD risk factors
Baseline Study of Women’s Health Across the 1 2 Nation (SWAN), N=3302 3 4 16 visits over 22 years: 5 • Demographics 6 • VMS: Hot flashes, night sweats 7 • SBP, DBP, BMI 8 • Phlebotomy: lipids, glucose, etc 9 • Medication use 10 11 • Health behaviors 12 • CVD events (myocardial infarction, cerebrovascular 13 accident, heart failure, revascularization), adjudication 14 • Death certificates, cause of death (NDI search) 15
Methods o VMS at baseline ➢ None, 1- 5 days, ≥6 days; past 2 weeks o Persistent frequent VMS over the menopause ➢ Proportion of visits reporting VMS ≥6 days / 2 weeks o CVD events & CVD mortality: First event o Control for range of risk factors
Baseline Participant Characteristics (N=3272) No VMS† 1- 5 Days† ≥6 Days† (N=1986) (N=921) (N=365) 46 (44 - 48) 46 (44 - 48) 47 (45 - 49) Age, Median (IQR)* Race, N (%)* 489 (24) 286 (31) 144 (39) Black 981 (49) 406 (44) 155 (42) White 171 (9) 58 (6) 17 (5) Chinese 152 (8) 100 (11) 32 (9) Hispanic 193 (10) 71 (8) 17 (5) Japanese 26 (22 - 31) 27 (23 - 33) 29 (25 - 36) BMI, M (SD)* 113 (105 - 124) 118 (108 - 129) 120 (109 - 130) SBP, M (SD)* 74 (69 - 80) 76 (69 - 83) 76 (70 - 83) DBP, M (SD)* 770 (39) 493 (54) 222 (62) Menopause stage, N (%) early peri- (vs Pre-)* † VMS in past two weeks; *Varies by VMS category, p<.05
Baseline VMS in relation to CVD events / CVD mortality 0.4 No VMS 0.35 VMS 1-5 days/2weeks P=.0001 0.3 VMS >=6 days/2weeks 0.25 Event Rates 0.2 0.15 0.1 0.05 0 0 5 10 15 20 Years from Baseline N=3272, 231 events
Baseline VMS in Relation to CVD events/CVD mortality Model 1 Model 2 HR (95% CI) HR (95% CI) Frequency of VMS 1.33 (.99-1.80) † 1-5 days / 2 weeks 1.04 (.76-1.41) 6 days / 2 weeks 2.16 (1.53-3.03)*** 1.51 (1.05, 2.17)* Relative to no VMS in prior two weeks † p<.10, * p<.05, ** p<.01, *** p<.001 Model 1: Site, baseline age, race/ethnicity Model 2: Model 1 + Education, financial strain, menopause stage, SBP, BMI, LDL-C, triglycerides, HOMA-IR, medication use (BP-lowering, lipid-lowering, anti-diabetic), smoking, physical activity
Persistent frequent VMS in relation to CVD events / CVD mortality 0.4 0.35 [0-33)% of the visits (7.7%) 0-33% of the visits P=.0001 0.3 [33-66)% of the visits (15.3%) 33-66% of the visits Event Rates 0.25 66-100% of the visits [66-100]% of the visits (21.4%) 0.2 0.15 0.1 0.05 0 0 5 10 15 20 Years from Baseline N=3272, 231 events
Persistent VMS in Relation to CVD events/CVD mortality Model 1 Model 2 HR (95% CI) HR (95% CI) >25% (vs. ≤25%) of attended 1.79 (1.37-2.34)**** 1.80 (1.36-2.36)**** visits with frequent VMS † p<.10, *p<.05, **p<.01, ***p<.001, ****p<.0001 Model 1: Site, baseline age, race/ethnicity Model 2: Model 1 + Education, financial strain, menopause stage, SBP, BMI, LDL-C, triglycerides, HOMA-IR, medication use (HT, BP-lowering, lipid-lowering, anti-diabetic), smoking, physical activity, number of attended visits
Additional Findings o Findings similar: ➢ For adjudicated CVD events ➢ Timing of VMS (when the VMS occurred) ➢ Censoring women at bilateral oophorectomy ➢ Dropping visits with hormone therapy use
Summary o SWAN: Frequent VMS at baseline or persistently frequent VMS over the transition associated with increased risk of CVD events later in life ➢ Not explained by standard CVD risk factors ➢ 50-80% increased risk o Frequent or persistent VMS as novel marker of midlife CVD risk ➢ Risk prediction for CVD in midlife women?
Acknowledgements Co-Authors: Helen Vlachos, PhD Samar El Khoudary, PhD Study Participants Maria Brooks, PhD Carol Derby, PhD Karen Matthews, PhD Hadine Joffe, MD Elizabeth Jackson, PhD Sioban Harlow, PhD
Recommend
More recommend