The timing hypothesis and hormone replacement therapy: a paradigm shift in the primary prevention of coronary heart disease in women. part 1: comparison of therapeutic efficacy

The Timing Hypothesis and Hormone Replacement Therapy:A Paradigm Shift in the Primary Prevention of Coronary HeartDisease in Women. Part 1: Comparison of Therapeutic Efficacy Howard N. Hodis, MD,*†‡ and Wendy J. Mack, PhD*‡ :The long-held belief that outcome data from interven- Sex-Specific Primary Prevention Therapy: Statins and tion trials in men are generalizable to women has cre- ated the framework in which the primary prevention ofcoronary heart disease (CHD) in women is viewed, but 3-hydroxy-3-methylglutaryl coenzyme A reductase over the past decade, data have accumulated to refute such a supposition of generalizability. These lines ofevidence concern the sex-specific efficacy of CHD pri- Statins are the most commonly used medications for mary prevention therapies and timing of postmeno- lipid lowering and for the primary prevention of CHD pausal hormone replacement therapy (HRT) initiation in women and men. The prevailing belief is that statins according to age and time since menopause as modifiers reduce CHD events and mortality under primary and sec- of efficacy and risk. Although the standard primary pre- ondary prevention conditions in women and men. How- vention therapies of statins and aspirin reduce CHD in ever, careful examination of randomized control trial men, neither therapy reduces CHD and, more impor- (RCT) data does not provide clear evidence that statins tantly, mortality in women under primary prevention reduce CHD events or total mortality in women under pri- CHD and mortality in primary prevention when it is In the first large meta-analysis of RCTs of statin ther- initiated in women who are younger than 60 or are apy in which primary (six RCTs; n = 11,435 women) and less than 10 years since menopause. Herein, the efficacy secondary (eight RCTs; n = 8,272 women) CHD preven- of the commonly used therapies for the primary preven- tion trials were analyzed separately in women, the data tion of CHD in women, statins, aspirin, and postmeno- support a significant reduction in CHD events (hazard pausal HRT is discussed. The comparative risks of ratio (HR) = 0.80; 95% confidence interval (CI) = 0.71– these therapies will be discussed in Part 2 of this series.
0.91) but not in total mortality (HR = 1.00, 95% CI = 0.77–1.29) in women under secondary preventionconditions, although neither CHD events (HR = 0.89, 95% CI = 0.69–1.09) nor total mortality (HR = 0.95, 95% CI = 0.62–1.46) was reduced in women under pri- In the first sex-specific meta-analysis to analyze pri- mary CHD prevention trials independently from secondaryprevention trials, statin therapy reduced CHD in men butnot in women under primary prevention conditions,2 evenwith inclusion of the Management of Elevated Cholesterolin the Primary Prevention Group of Adult Japanese(MEGA) trial in which 5,356 women were enrolled and From the *Atherosclerosis Research Unit, University of Southern followed for more than 5 years.3 Statin therapy did not California, Los Angeles, California; †Department of Medicine, Keck significantly reduce total mortality in women or men under School of Medicine, University of Southern California, Los Angeles, primary prevention conditions (Table 1).2 These results California; and ‡Department of Preventive Medicine, Keck School ofMedicine, University of Southern California, Los Angeles, California.
were confirmed in another sex-specific meta-analysis ofstatins and CHD prevention4 (Table 1) that included Address correspondence to Howard N. Hodis, Harry J. Bauer andDorothy Bauer Rawlins Professor of Cardiology, Atherosclerosis Research RCTs conducted in individuals with diabetes mellitus with- Unit, Keck School of Medicine, University of Southern California, 2250 out a history of cardiovascular disease (CVD), as well as Alcazar Street, CSC 132, Los Angeles, CA 90033. E-mail: athero@usc.edu in MEGA and the Justification for the Use of Statins in 2013, Copyright the AuthorsJournal compilation 2013, The American Geriatrics Society trials does not alter the conclusion that statin therapy has Table 1. Comparison of Statin Therapy on Coronary a null effect on CHD events and total mortality in primary Heart Disease and Total Mortality in Primary Preven- The sex-specific efficacy of primary prevention of CHD reported for statins is concordant with that of aspirin ther- apy.8 In meta-analyses of primary CHD prevention trials, aspirin significantly reduced MI by approximately 32% with a null effect on stroke in men, whereas in women, aspirin had a null effect on MI but significantly reduced ischemic stroke by approximately 17%.8,9 In women and men, aspirin therapy has a null effect on total mortality under primary CHD prevention conditions. Consistent with statin therapy, the null effect of aspirin therapy on CHD extends to high-risk women with diabetes mellitus without a history of CVD. In the Japanese Primary Preven- tion of Atherosclerosis with Aspirin for Diabetes (JPAD) trial, the largest RCT of aspirin therapy and primary CHD prevention in individuals aged 30 to 85 with type 2 diabe-tes mellitus (1,152 women), the effect of aspirin therapy HR = hazard ratio; CI = confidence interval.
on CHD was null relative to placebo (relative risk(RR) = 0.88, 95% CI = 0.53–1.44) after a median treat- Prevention: An Intervention Trial Evaluating Rosuvastatin ment of 4.4 years.10 The consistency across individual pri- (JUPITER), another primary CHD prevention RCT with a mary CHD prevention trials and sex-specific meta-analyses large cohort of 6,801 women.5 Even though this meta- show no evidence that aspirin therapy reduces CHD events analysis included a greater number of women at greater or total mortality in women without preexisting CVD, risk for CHD than the previous meta-analyses, the risk including those with diabetes mellitus who are at high risk reduction reported for CHD in women was statistically of CHD. The sex-specific nature of CVD benefits restricted nonsignificant, and total mortality was no different than in to men under primary prevention conditions is seen with other interventions such as angiotensin-converting enzyme As the consistency across individual primary CHD inhibitor therapy, with men having less CVD and mortal- prevention trials and sex-specific meta-analyses shows, there is no clear evidence that statins reduce CHD eventsor total mortality in women without preexisting CVD. Theone possible exception is JUPITER in which the primary Timing of Initiation of Postmenopausal HRT trial endpoint for CHD was lower (HR = 0.54, 95%CI = 0.37–0.80)5 in women but JUPITER was stopped early, after a median follow-up of 1.9 years, and it is Over the last decade, cumulated data from RCTs of HRT unclear whether findings from JUPITER were due to the clearly demonstrate two distinct populations of women unique characteristics of the cohort (women aged  60 who respond differentially to HRT according to timing of with low-density lipoprotein cholesterol <130 mg/dL and HRT initiation relative to age and time since menopause.13 high-sensitivity C-reactive protein  2 mg/dL),5 the trial Specifically, CHD events and total mortality benefits occur biases and flaws6,7 or to the subjective nature of certain when HRT is initiated in younger women (<60) in close components of the primary endpoint.5 The JUPITER pri- proximity to menopause (<10 years since menopause) and mary cardiovascular endpoint was a composite comprising a null and possible adverse effect when initiated in older “hard endpoints” (nonfatal myocardial infarction (MI), women (  60) remote from menopause (>20 years since nonfatal stroke, or confirmed death resulting from cardio- menopause).13 The beneficial effect of HRT on CHD vascular causes) and “soft endpoints,” whose occurrence according to timing of HRT initiation has been shown in a rely on medical decisions (arterial revascularization or hos- large meta-analysis of 23 RCTs with 191,340 women- pitalization for unstable angina pectoris). In men, all of years of follow-up.14 When analyzed over all ages and in the hard and soft components of the composite primary women who initiate HRT when aged 60 and older or endpoint were significantly lower in the rosuvastatin than more than 10 years after menopause, the effect of HRT on in the placebo arm. In women, only the soft endpoints CHD is null. In women who initiate HRT when younger (revascularizations and hospitalizations) were significantly than 60 or less than 10 years since menopause, the risk of lower and clearly drove the primary endpoint to statistical CHD is statistically significantly 32% less than with significance because none of the hard endpoints in the placebo (Figure 1). The magnitude of CHD reduction for women differed significantly (P > .10) between the rosu- women younger than 60 or less than 10 years since meno- vastatin and placebo arms.5 Total mortality was not statis- pause when randomized to HRT is similar to observational tically different between the rosuvastatin and placebo arms studies of populations of women who initiated HRT at the in women (P = .12) or men (P = .08).5 Including JUPITER in meta-analyses along with other primary prevention THE TIMING HYPOTHESIS AND PREVENTION IN WOMEN A Coronary Heart Disease
The 11-year WHI CEE trial follow-up (7 years of randomized treatment and 4 years of postintervention fol- 0.99 (0.88-1.11)
low-up) showed that women aged 50 to 59 given CEE had a >60 years old
>10 years since menopause,
1.03 (0.91-1.16)
CI = 0.38–0.90), total MI (HR = 0.54, 95% CI = 0.34– <60 years old
0.86), and total mortality (HR = 0.73, 95% CI = 0.53– <10 years since menopause,
0.68 (0.48-0.96)
1.00) than with placebo. Compared with women aged 60 to69 and 70 to 79, the P-value for interaction was statistically Relative Risk (95% CI)
B Total Mortality
P = .04, respectively18), indicating that the CEE effect onthese outcomes differs according to age. Invasive breast can- 0.98 (0.87-1.18)
cer was statistically significantly 23% lower (HR = 0.77, >60 years old,
95% CI = 0.62–0.95) in women who received CEE than Mean age = 66 years
1.03 (0.91-1.16)
with placebo regardless of age at randomization.18 <60 years old,
Of immense importance to understanding early initia- Mean age = 54 years
0.61 (0.39-0.95)
tion of HRT, long-term use, and clinical outcomes inhealthy young women is the Danish Osteoporosis Preven- tion Study (DOPS)19, the only prospective longitudinal Relative Risk (95% CI)
randomized trial designed to examine clinical outcomes in Figure 1. (A) Relative risks (RRs) and 95% confidence inter- women who were specifically a priori randomized to HRT vals (CIs) for coronary heart disease events associated with in the perimenopausal or early postmenopausal period.
hormone replacement therapy (HRT) from a meta-analysis of DOPS included 1,006 women who were on average 23 randomized controlled trials (RCTs) in 39,049 women 50 years old (range 45–58) and 7 months postmenopausal (followed for 191,340 women-years). Results are shown for when randomized for 10 years to oral 17b-estradiol plus all ages and for women aged 60 and older or more than sequential norethisterone acetate or to an untreated group.
10 years after menopause or younger than 60 years old or less Hysterectomized women received oral 17b-estradiol 2 mg than 10 years after menopause when randomized and HRT daily. After randomized treatment, the women were fol- initiated. (B) RRs and 95% CIs for total mortality associated lowed for another 6 years for a total follow-up of with HRT from a meta-analysis of 30 RCTs in 26,708 16 years. After 10 years of randomized treatment, the women (followed for 119,118 women-years). Results are composite primary trial endpoint of mortality, MI, or shown for all ages and for women aged 60 and older or more than 10 years after menopause or younger than 60 years old (HR = 0.48, 95% CI = 0.27–0.89), and total mortality or less than 10 years after menopause when randomized and was 43% (HR = 0.57, 95% CI = 0.30–1.08) lower in the HRT group than in the control group. After a total fol-low-up of 16 years, the composite primary trial endpointremained significantly 49% lower (HR = 0.61, 95% The Women’s Health Initiative (WHI) trial data also CI = 0.39–0.94) and total mortality was 34% (HR = 0.66, support the “timing” hypothesis, showing significant 95% CI = 0.41–1.08) lower in the women originally ran- trends of an HRT effect on CHD according to time since domized to HRT than in those randomized to the control menopause.17 Women randomized to conjugated equine group. There were no statistically significant differences in estrogen (CEE) less than 10 years after menopause had a incident breast cancer, stroke, or venous thromboembolism 52% lower risk of CHD than with placebo (HR = 0.48, between treatment groups in DOPS. DOPS results are sim- 95% CI = 0.20–1.17), whereas women who were 10 to ilar to the 11-year WHI CEE trial follow-up data of the 19 years since menopause (HR = 0.96, 95% CI = 0.64– women aged 50 to 59 when randomized to CEE18 and to 1.44) and 20 or more years since menopause (HR = 1.12, the 32% lower CHD14 and 39% lower total mortality20 95% CI = 0.86–1.46) showed no benefit of CEE on CHD.
shown in meta-analyses of RCTs of women younger than Women randomized to CEE plus medroxyprogesterone 60 or less than 10 years since menopause with HRT than acetate (MPA) within 10 years after menopause had a 12% lower risk of CHD than with placebo (HR = 0.88,95% CI = 0.54–1.43), whereas women 10 to 19 years after menopause (HR = 1.23, 95% CI = 0.85–1.77) and Although the risks and benefits of HRT continue to be 20 or more years after menopause (HR = 1.66, 95% debated, postmenopausal HRT is the only primary preven- CI = 1.14–2.41) showed no benefit and possibly a greater tion therapy in women that has been demonstrated to risk with CEE plus MPA than with placebo. With both tri- reduce total mortality and extend life.13 The beneficial als combined, women randomized to CEE and CEE plus effect of HRT on total mortality according to age was MPA within 10 years of menopause had a 24% lower risk shown in a large meta-analysis of 30 RCTs with 119,118 of CHD than with placebo (HR = 0.76, 95% CI = 0.50– women-years of follow-up.20 When analyzed across all ages 1.16), whereas women 10 to 19 years since menopause and in women aged 60 and older when initiating HRT, the (HR = 1.10, 95% CI = 0.84–1.45) and 20 or more years effect on total mortality was null, whereas there was signi- ficant 39% lower total mortality in women younger than showed no benefit and a possibly greater risk of CEE and 60 (mean age 54) when with HRT than with placebo CEE plus MPA on CHD than with placebo.
(Figure 1), a difference similar to that fond in observational quality-adjusted life years (QALYs) at a cost of $2,438 per studies. Age at HRT initiation in women in observational QALY gained.22 Net gains gradually increase with treat- studies and age of younger women randomized to RCTs ment durations of 5 to 30 years, and results for younger examined in the meta-analysis was similar.15,16 women are robust to all sensitivity analyses, with HRT Similar to CHD trends, total mortality in WHI was remaining highly cost effective. At $2,438 per QALY 30% lower with CEE plus MPA and CEE therapies than gained, these data indicate that HRT is a highly cost-effec- with placebo in women aged 50 to 59 when randomized.17 tive strategy for improving quality-adjusted life. Alterna- Women randomized to CEE when younger than 60 had a tively, there is a smaller net gain of 0.11 QALYs at a cost of 29% lower risk of total mortality than with placebo $27,953 per QALY gained for 65-year-old postmenopausal (HR = 0.71, 95% CI = 0.46–1.11), whereas women aged women initiating HRT.22 Cost effectiveness ratios of less 60 to 69 (HR = 1.02, 95% CI = 0.64–1.44) and 70 to 79 than $50,000 per QALY are considered worthwhile, those (HR = 1.20, 95% CI = 0.93–1.56) showed no effect of less than $5,000 per QALY gained are considered highly CEE on total mortality. Women randomized to CEE plus cost effective, and a cost-effectiveness ratio greater than MPA when younger than 60 had a 31% lower risk of total $100,000 per QALY is considered unattractive.23 mortality than with placebo (HR = 0.69, 95% CI = 0.44–1.07), whereas women aged 60 to 69 (HR = 1.09, 95% CI = 0.83–1.44) and 70 to 79 (HR = 1.06, 95% CI = 0.80 –1.41) showed no benefit of CEE plus MPA on total mor- The cumulated data and their implication for women’s tality. In both trials combined, women randomized to CEE health in the primary prevention of CHD have become plus MPA or CEE when younger than 60 had a statisti- clearer over the past decade as the sex-specificity of statins cally significant 30% lower risk of total mortality than and aspirin and timing of initiation of HRT as modifiers of with placebo (HR = 0.70, 95% CI = 0.51–0.96), whereas efficacy and risk in women have become more fully eluci- women aged 60 to 69 (HR = 1.05, 95% CI = 0.87–1.26) dated. The data clearly show that standard primary CHD and 70 to 79 (HR = 1.14, 95% CI = 0.94–1.37) showed prevention therapies that are presumably efficacious in men no benefit of CEE and CEE plus MPA on total mortality.
(statins and aspirin) do not statistically significantly reduce To address the risks and benefits of HRT, a Bayesian CHD events or total mortality in women.
meta-analysis was conducted using RCTs and observa- Data are consistent in showing that, when initiated in tional studies to evaluate the effect of HRT on total mor- women younger than 60 or less than 10 years after meno- tality in postmenopausal women younger than 60 who pause, HRT reduces CHD events and total mortality, initiated HRT in close proximity to menopause.21 Results extends life, and is highly cost effective. The risks associ- from this meta-analysis using 19 RCTs with 16,283 ated with HRT, such as breast cancer, are considered rare women (mean age 54.5) followed for 83,043 women-years (<1 event per 10,000 women treated per year) (see Part 2).
over 5.1 years (range 1–6.8 years) showed total mortality The evidence-based data are large and consistent across approximately 40 observational studies and meta-analyses (CrI) = 0.52–0.96) in women randomized to HRT than in encompassing 20 to 30 RCTs. In addition, the type and those who received placebo. Using pooled data from eight magnitude of risks associated with the standard primary prospective observational studies in which 212,717 women CHD prevention therapies are similar to those associated were followed for 2,935,495 patient-years over a mean of 13.8 years (range 6–22 years), total mortality was 22% Although the data are clear in showing sex-specific (RR = 0.78, 95% CrI = 0.69–0.90) lower in HRT users benefits restricted to men under primary prevention condi- than in nonusers.21 Total mortality was 28% (RR = 0.72, tions with statin and aspirin therapy,2–5 the old paradigm 95% CrI = 0.62–0.82) lower with the RCT and prospec- that “what works in men must work in women” continues tive observational data combined. Results from this study to be promulgated as evidence-based data indicating the indicate a convergence of evidence from several sources fallacy of this paradigm are ignored. Data supporting the that support a beneficial effect of HRT on total mortality use of statin and aspirin therapy for a significant reduction in women who initiate HRT in close proximity to meno- in CHD events and total mortality under primary preven- pause. Results from this meta-analysis also indicate that tion conditions for women are lacking (Table 2). The data RCTs and observational studies are similar, each with a consistently support that, when initiated in women youn- total mortality reduction of approximately 25%, similar to ger than 60 or less than 10 years after menopause, HRT the 10-year randomized trial follow-up (43% lower total reduces CHD and total mortality (Table 2). The statisti- mortality) and 16-year total follow-up (34% lower total cally significantly greater risk of diabetes mellitus, risk of mortality) of DOPS,19 the 11-year follow-up of WHI CEE breast cancer similar to that with CEE plus MPA therapy, (27% lower total mortality),18 and the significant 30% and potential for the increase in new cancer cases in older lower (HR = 0.70, 95% CI = 0.51–0.96) total mortality persons24 associated with statin therapy have important shown in postmenopausal women younger than 60 with implications for the balance of benefit and risk of statin HRT than with placebo in the WHI trials.17 therapy under primary prevention conditions in whichindividual RCTs and meta-analyses show no benefit in reducing CHD events or total mortality in women (seePart 2). In light of the accumulating data and safety label A cost-effectiveness analysis indicates that, compared with changes to include warnings of diabetes and cognitive no therapy, HRT given to postmenopausal women in their impairment that the Food and Drug Administration 50s for 5 to 30 years results in a substantial increase of 1.5 THE TIMING HYPOTHESIS AND PREVENTION IN WOMEN Comparison of the Effect of Initiation of Hormone Replacement Therapy in Young Postmenopausal Women with that of Statin and Aspirin Therapy on Coronary Heart Disease and Total Mortality in PrimaryPrevention aInitiation in women younger than 60 or less than 10 years after menopause.
requires,25 the use of statin therapy for the primary pre- age of 50 and an average of 7 months after menopause, vention of CHD in men has also been questioned.6,26 DOPS provides strong evidence of the long-term efficacy DOPS, the only prospective longitudinal randomized and safety of HRT for reducing CHD and total mortality.
trial conducted specifically in women younger than 60 Inconsistencies in presentation and interpretation of HRT (average age 50) and less than 10 years after menopause data has created great confusion for healthcare providers (average 7 months), provides direct and compelling evi- and patients alike, culminating in the call for an indepen- dence for up to 16 years that the benefits of prevention of dent commission to evaluate the interpretation and dissem- chronic diseases outweigh the risks.19 Women included in ination of the evidence-based data in relation to public DOPS specifically represent women studied in previous observational studies, and hence, DOPS is the only ran- The totality of evidence indicates that the benefits of domized trial to appropriately test the “estrogen cardio- the initiation of HRT at or near menopause outweigh the protective” hypothesis in the same population of women risks, with the weight of evidence supporting downstream in which this hypothesis was generated. In addition, few prevention of morbidity and mortality. Healthcare provid- prevention therapies other than HRT have been studied ers and patients can be confident in applying the cumula- under randomized conditions for 10 years.
tive data in making clinical decisions concerning chronic Timing of initiation of primary prevention therapy disease prevention, keeping in mind that any prevention appears to have significant biological and clinical conse- strategy must be personalized. Cumulative data provide quences for women. The timing of initiation of primary not only strong evidence of the beneficial effects of HRT prevention therapies provides opportunity for reduction of when initiated in women in close proximity to menopause, CHD events and total mortality throughout the postmeno- but also reassurance of their safety.
pausal period and forges a new paradigm in the primaryprevention of CHD in women. It is important to rethink the appropriate clinical application of the evidence-based Funded in part by National Institutes of Health, National data: reduction of CHD events and total mortality in Institute on Aging Grant R01AG-024154.
women who initiate HRT when younger than 60 or less Conflict of Interest: The editor in chief has reviewed than 10 years after menopause versus no reduction of the conflict of interest checklist provided by the authors CHD events or total mortality with statin and aspirin ther- and has determined that the authors have no financial or apy (Table 2). Women aged 60 and older or more than any other kind of personal conflicts with this paper.
10 years after menopause have likely missed the window Author Contributions: Both authors took part in all of opportunity for benefit from HRT and should probably not initiate HRT for the primary prevention of CHD.
In conclusion, a large body of RCT data converges with results from observational studies, animal studies,and basic science supporting that HRT health outcomes vary according to age or time since menopause. Focused in 1. Walsh JME, Pignone M. Drug treatment of hyperlipidemia in women.
young healthy postmenopausal women with an average 2. Petretta M, Costanzo P, Perrone-Filardi P et al. Impact of gender in pri- 14. Salpeter SR, Walsh JME, Greyber E et al. Coronary heart disease events mary prevention of coronary heart disease with stain therapy: A meta-anal- associated with hormone therapy in younger and older women: A meta- ysis. Int J Cardiol 2010;138:25–31.
analysis. J Gen Intern Med 2006;21:363–366.
3. Mizuno K, Nakaya N, Ohashi Y et al. Usefulness of pravastatin in primary 15. Grodstein F, Stampfer M. The epidemiology of coronary heart disease and prevention of cardiovascular events in women: Analysis of the Management estrogen replacement in postmenopausal women. Prog Cardiol Dis of Elevated Cholesterol in the Primary Prevention Group of Adult Japanese (MEGA Study). Circulation 2008;117:494–502.
16. Grodstein F, Stampfer M. Estrogen for women at varying risk of coronary 4. Brugts JJ, Yetgin T, Hoeks SE et al. The benefits of statins in people with- out established cardiovascular disease but with cardiovascular risk factors: 17. Rossouw JE, Prentice RL, Manson JE et al. Postmenopausal hormone ther- Meta-analysis of randomised controlled trials. BMJ 2009;338:b2376.
apy and risk of cardiovascular disease by age and years since menopause.
5. Mora S, Glynn RJ, Hsia J et al. Statins for the primary prevention of car- diovascular events in women with elevated high-sensitivity c-reactive pro- 18. LaCroix AZ, Chlebowski RT, Manson JE et al. Health outcomes after tein or dyslipidemia: Results from the Justification for the Use of Statins in stopping conjugated equine estrogens among postmenopausal women Prevention: An Intervention Trial Evaluating Rosuvastatin (JUPITER) and with hysterectomy: A randomized controlled trial. JAMA 2011;305: meta-analysis of women from primary prevention trials. Circulation 19. Schierbeck LL, Rejnmark L, Tofteng CL et al. Effect of hormone replace- 6. de Lorgeril M, Salen P, Abramson J et al. Cholesterol lowering, cardiovas- ment treatment on cardiovascular events in recently postmenopausal cular diseases, and rosuvastatin-JUPITER controversy. Arch Intern Med women: Randomized trial. BMJ 2012;345:e6409.
20. Salpeter SR, Walsh JME, Greyber E et al. Mortality associated with hor- 7. Kaul S, Morrissey RP, Diamond GA. By Jove! What is a clinician to make mone replacement therapy in younger and older women: A meta-analysis.
of JUPITER? Arch Intern Med 2010;170:1073–1077.
8. Berger JS, Roncaglioni MC, Avanzini F et al. Aspirin for the primary pre- 21. Salpeter SR, Cheng J, Thabane L et al. Bayesian meta-analysis of hormone vention of cardiovascular events in women and men: A sex-specific meta- therapy and mortality in younger postmenopausal women. Am J Med analysis of randomized controlled trials. JAMA 2006;295:306–313.
9. Ridker PM, Cook NR, Lee IM et al. A randomized trial of low-dose aspi- 22. Salpeter SR, Buckley NS, Liu H et al. The cost-effectiveness of hormone rin in the primary prevention of cardiovascular disease in women. N Engl J therapy in younger and older postmenopausal women. Am J Med 10. Ogawa H, Nakayama M, Morimoto T et al. Low-dose aspirin for primary 23. Cohen DJ, Reynolds MR. Interpreting the results of cost-effectiveness stud- prevention of atherosclerotic events in patients with type 2 diabetes: A ran- ies. J Am Coll Cardiol 2008;52:2119–2126.
domized controlled trial. JAMA 2008;300:2134–2141.
24. Shepherd J, Blauw GJ, Murphy MB et al. Pravastatin in Elderly Individuals 11. Wing LMH, Reid CM, Ryan P et al. A comparison of outcomes with at Risk of Vascular Disease (PROSPER): A randomized controlled trial.
angiotensin-converting-enzyme inhibitors and diuretics for hypertension in the elderly. N Engl J Med 2003;348:583–592.
25. US Food and Drug Administration. FDA drug safety communication: 12. Shekelle PG, Rich MW, Morton SC et al. Efficacy of angiotensin-converting Important safety label changes to cholesterol lowering statin drugs enzyme inhibitors and beta-blockers in the management of left ventricular [on-line]. Available at http://www.fda.gov/drugs/drugsafety/ucm293101.
systolic dysfunction according to race, gender, and diabetic status: A meta- analysis of major clinical trials. J Am Coll Cardiol 2003;41:1529–1538.
26. Redberg RF, Katz MH. Healthy men should not take statins. JAMA 13. Hodis HN, Mack WJ. A window of opportunity: The reduction of coro- nary heart disease and total mortality with menopausal therapies is age 27. Utian WH. A decade post WHI, menopausal hormone therapy comes full and time dependent. Brain Res 2011;1379:244–252.
circle—need for independent commission. Climacteric 2012;15:320–325.

Source: http://www.hotflashhavoc.net/userfiles/files/JAGS_Part-1_2013.pdf