|

 

1ridk.oa

The ne w engl and journal of medicine A Randomized Trial of Low-Dose Aspirin in the Primary Prevention of Cardiovascular Disease in Women Paul M Ridker, M.D., Nancy R. Cook, Sc.D., I-Min Lee, M.B., B.S., David Gordon, M.A., J. Michael Gaziano, M.D., JoAnn E. Manson, M.D., Charles H. Hennekens, M.D., and Julie E. Buring, Sc.D.
b a c k g r o u n d
Randomized trials have shown that low-dose aspirin decreases the risk of a first myo- From the Divisions of Preventive Medicine (P.MR., N.R.C., I-M.L., D.G., J.M.G., J.E.M., cardial infarction in men, with little effect on the risk of ischemic stroke. There are few J.E.B.), Cardiovascular Medicine (P.MR., similar data in women.
J.M.G.), and Aging (J.M.G., J.E.B.), Depart-ment of Medicine, Brigham and Women's Hospital, Harvard Medical School; the De- We randomly assigned 39,876 initially healthy women 45 years of age or older to re- partment of Epidemiology, Harvard School of Public Health (P.MR., N.R.C., I-M.L., ceive 100 mg of aspirin on alternate days or placebo and then monitored them for 10 J.E.M., J.E.B.); Veterans Affairs Boston years for a first major cardiovascular event (i.e., nonfatal myocardial infarction, non- Healthcare System (J.M.G.); and the De-fatal stroke, or death from cardiovascular causes).
partment of Ambulatory Care and Preven-tion, Harvard Medical School (J.E.B.) — allin Boston; and the Departments of Medi- cine and Epidemiology and Public Health, During follow-up, 477 major cardiovascular events were confirmed in the aspirin group, University of Miami School of Medicine,as compared with 522 in the placebo group, for a nonsignificant reduction in risk with and the Department of Biomedical Sci- ence, Center of Excellence, Florida Atlan- aspirin of 9 percent (relative risk, 0.91; 95 percent confidence interval, 0.80 to 1.03; tic University, Miami (C.H.H.). Address P=0.13). With regard to individual end points, there was a 17 percent reduction in the reprint requests to Dr. Buring at therisk of stroke in the aspirin group, as compared with the placebo group (relative risk, Division of Preventive Medicine, Brigham and Women's Hospital, 900 Common- 0.83; 95 percent confidence interval, 0.69 to 0.99; P=0.04), owing to a 24 percent re- wealth Ave. East, Boston, MA 02215, or at duction in the risk of ischemic stroke (relative risk, 0.76; 95 percent confidence inter- [email protected].
val, 0.63 to 0.93; P=0.009) and a nonsignificant increase in the risk of hemorrhagic This article was published at www.nejm.org stroke (relative risk, 1.24; 95 percent confidence interval, 0.82 to 1.87; P=0.31). As on March 7, 2005.
compared with placebo, aspirin had no significant effect on the risk of fatal or nonfatalmyocardial infarction (relative risk, 1.02; 95 percent confidence interval, 0.84 to 1.25; N Engl J Med 2005;352:1293-304.
Copyright 2005 Massachusetts Medical Society. P=0.83) or death from cardiovascular causes (relative risk, 0.95; 95 percent confidenceinterval, 0.74 to 1.22; P=0.68). Gastrointestinal bleeding requiring transfusion wasmore frequent in the aspirin group than in the placebo group (relative risk, 1.40; 95percent confidence interval, 1.07 to 1.83; P=0.02). Subgroup analyses showed that as-pirin significantly reduced the risk of major cardiovascular events, ischemic stroke,and myocardial infarction among women 65 years of age or older.
In this large, primary-prevention trial among women, aspirin lowered the risk of strokewithout affecting the risk of myocardial infarction or death from cardiovascular causes,leading to a nonsignificant finding with respect to the primary end point.
n engl j med 352;13 Downloaded from www.nejm.org at TORONTO WESTERN HOSP on June 11, 2005 . Copyright 2005 Massachusetts Medical Society. All rights reserved. The new england journal of medicine lthough aspirin is effective in tective effect, while minimizing gastrointestinal the treatment of acute myocardial infarc- side effects through the use of a low dose and al- a tion and in the secondary prevention of ternate-day administration. The design of the study cardiovascular disease among both men and wom- has previously been described in detail.14,15 In brief,en,1 its use in primary prevention remains contro- between September 1992 and May 1995, letters ofversial. To date, five randomized trials involving invitation were mailed to more than 1.7 million fe-55,580 participants have evaluated aspirin in the male health professionals. A total of 453,787 com-primary prevention of cardiovascular disease.2-6 In pleted the questionnaires, with 65,169 initially will-aggregate, these trials indicate that, as compared ing and eligible to enroll. Women were eligible ifwith placebo, aspirin therapy was associated with a they were 45 years of age or older; had no historysignificant, 32 percent reduction in the risk of my- of coronary heart disease, cerebrovascular disease,ocardial infarction, but the data on the risk of stroke cancer (except nonmelanoma skin cancer), or otherand death from cardiovascular disease remain in- major chronic illness; had no history of side effectsconclusive.7 Moreover, three of these trials evaluat- to any of the study medications; were not taking as-ed men exclusively, and fewer than 180 of the 2402 pirin or nonsteroidal antiinflammatory medicationsvascular events occurred in women. Thus, at this (NSAIDs) more than once a week (or were willingtime, the current recommendations for the use of to forego their use during the trial); were not takingaspirin in primary prevention in women are based anticoagulants or corticosteroids; and were not tak-on limited direct data from women.8-10 ing individual supplements of vitamin A, E, or beta Direct evidence regarding the effects of aspirin carotene more than once a week. For the purposes in women is necessary because cardiovascular dis- of this study, inhibitors of cyclooxygenase, wheth-ease is the leading cause of death among both wom- er selective or nonselective, were considered to been and men. Direct evidence is also relevant be- NSAIDs.
cause of the potential for sex-based differences in Eligible women were enrolled in a three-month salicylate metabolism11 and continuing uncertain- run-in period of placebo administration to identifyty regarding the cardiovascular effects of hormone- a group likely to be compliant with long-term treat-replacement therapy.12 Moreover, in addition to a ment. A total of 39,876 women were willing, eligi-paucity of data on women, the prophylactic use of ble, and compliant during the run-in period and un-aspirin in both sexes has prompted concern owing derwent randomization: 19,934 were assigned toto the potentially increased risk of hemorrhagic receive aspirin and 19,942 to receive placebo. Writ-stroke.13 This issue is particularly complex, since ten informed consent was obtained from all partic-the relative proportion of stroke to myocardial in- ipants. The trial was approved by the institutionalfarction differs between women and men.
review board of Brigham and Women's Hospital, We addressed these questions in the Women's Boston, and was monitored by an external data and Health Study, a large randomized, double-blind, pla- safety monitoring board.
cebo-controlled trial of low-dose aspirin in the pri- Every 12 months, the women were sent a year's mary prevention of cardiovascular disease among supply of monthly calendar packs containing ac-39,876 apparently healthy women followed for a tive agents or placebo as well as questionnaires onmean of 10 years for the major cardiovascular events compliance, side effects, the occurrence of relevantof myocardial infarction, stroke, and death from car- clinical end points, and risk factors. Study medica-diovascular causes.
tions and end-point ascertainment were continuedin a blinded fashion through the scheduled end ofthe trial (March 31, 2004). Follow-up and validation of reported end points were completed in February 2005. Rates of follow-up with respect to morbidity The Women's Health Study is a two-by-two facto- and mortality were 97.2 percent complete and 99.4rial trial evaluating the balance of risks and bene- percent complete, respectively.
fits of low-dose aspirin (100 mg every other day; All the women were followed for myocardial in- Bayer HealthCare) and vitamin E (600 IU every farction, stroke, or death from cardiovascular caus-other day; Natural Source Vitamin E Association), es. Medical records were obtained for all women inin the primary prevention of cardiovascular disease whom a cardiovascular end point was reported toand cancer. The trial was designed to evaluate the occur and were reviewed in a blinded fashion bylowest dose of aspirin that would have a cardiopro- an end-points committee of physicians. Myocardial Downloaded from www.nejm.org at TORONTO WESTERN HOSP on June 11, 2005 . Copyright 2005 Massachusetts Medical Society. All rights reserved. infarction was confirmed if symptoms met World and other randomized treatment assignments (vi-Health Organization criteria and if the event was as- tamin E and beta carotene, which was a componentsociated with abnormal levels of cardiac enzymes or of the trial for a median of 2.1 years17). Prespecifieddiagnostic electrocardiograms. A confirmed stroke subgroup analyses were performed according towas defined as a new neurologic deficit of sudden the presence or absence of major cardiovascularonset that persisted for at least 24 hours. Clinical in- risk factors. Modification of the effect of aspirinformation, computed tomographic scans, and mag- by the risk factors was assessed with the use of in-netic resonance images were used to distinguish teraction terms between subgroup indicators andhemorrhagic from ischemic events.16 Death was aspirin assignment, with tests for trend performedconfirmed to be from cardiovascular causes on the when subgroup categories were ordinal. To exam-basis of an examination of autopsy reports, death ine effects among women who were compliant,certificates, medical records, and information ob- we performed a sensitivity analysis in which fol-tained from the next of kin or other family mem- low-up data were censored at the time a woman re-bers. The use of coronary revascularization (bypass ported having taken less than two thirds of thesurgery or percutaneous coronary angioplasty) was study medication during the previous year. In ad-confirmed by a review of the medical records. A con- ditional analyses, data were censored on womenfirmed transient ischemic attack was defined as a if and when they started taking NSAIDs more thanneurologic deficit of sudden onset that lasted for three times a month.
less than 24 hours. Death from any cause was con-firmed by the end-points committee or on the basis of a death certificate. Only confirmed end points
were included in this analysis. For women with a primary analyses
reported myocardial infarction, the most common As shown in Table 1, the aspirin and placebo groups
diagnoses among those in whom the diagnosis was were similar with respect to baseline characteris-
not confirmed were stable or unstable angina or tics. The average duration of follow-up from ran-
chest pain without evidence of infarction. For wom- domization to the end of the trial was 10.1 years
en with a reported stroke, the most common alter- (range, 8.2 to 10.9). At the completion of the trial,
native diagnosis was transient cerebral ischemia.
999 women had had a first major cardiovascularevent (Table 2), for an absolute event rate of 253 per 100,000 person-years. Of these women, 477 were All primary analyses were performed on an inten- in the aspirin group and 522 were in the placebotion-to-treat basis. The primary end point was a group, indicating that there was a nonsignificantcombination of major cardiovascular events, in- reduction in risk of 9 percent (relative risk, 0.91; 95cluding nonfatal myocardial infarction, nonfatal percent confidence interval, 0.80 to 1.03; P=0.13).
stroke, and death from cardiovascular causes, and Regarding individual end points, women in the the trial was initially designed to have a statistical aspirin group had a 17 percent reduction in the riskpower of 86 percent to detect a 25 percent reduc- of stroke (relative risk, 0.83; 95 percent confidencetion in this end point. Secondary end points includ- interval, 0.69 to 0.99; P=0.04), as compared withed the individual end points of fatal or nonfatal myo- women in the placebo group; a 24 percent reductioncardial infarction, fatal or nonfatal stroke, ischemic in the risk of ischemic stroke (relative risk, 0.76; 95stroke, hemorrhagic stroke, and death from cardio- percent confidence interval, 0.63 to 0.93; P=0.009);vascular causes. Additional analyses included the and a nonsignificant increase in the risk of hem-incidence of death from any cause, transient ische- orrhagic stroke (relative risk, 1.24; 95 percent con-mic attack, and the need for coronary revascular- fidence interval, 0.82 to 1.87; P=0.31) (Table 2).
ization. If more than one end point occurred in a There was no significant difference between thegiven woman, only the first event within each cate- groups in the risk of fatal stroke (relative risk in thegory was counted; for the primary combined end aspirin group, 1.04; 95 percent confidence interval,point, the first event in each woman was counted.
0.58 to 1.86; P=0.90), but the aspirin group had a Cox proportional-hazards models were used to decreased risk of nonfatal strokes (relative risk, calculate relative risks and 95 percent confidence 0.81; 95 percent confidence interval, 0.67 to 0.97;intervals for the comparison of event rates in the as- P=0.02), as compared with the placebo group.
pirin and placebo groups after adjustment for age There was no evidence that, as compared with n engl j med 352;13 Downloaded from www.nejm.org at TORONTO WESTERN HOSP on June 11, 2005 . Copyright 2005 Massachusetts Medical Society. All rights reserved. The new england journal of medicine Table 1. Baseline Characteristics of the Women.
Smoking status (%) 25.0 to 29.9 (%) Menopausal status and use of HRT(%)† Postmenopausal and current HRT Postmenopausal and no HRT Hypertension (%)‡ Blood pressure (%) <120/<75 mm Hg 120–129/75–84 mm Hg 130–139/85–89 mm Hg ≥140/≥90 mm Hg placebo, aspirin reduced the overall risk of myocar- myocardial infarction, ischemic stroke, and hem-dial infarction (relative risk, 1.02; 95 percent con- orrhagic stroke according to the year of follow-up.
fidence interval, 0.84 to 1.25; P=0.83), fatal myo- Because it has been suggested that the ability of as-cardial infarction (relative risk, 1.16; 95 percent pirin to inhibit platelet function diminishes overconfidence interval, 0.54 to 2.51; P=0.70), nonfa- time,18 we also evaluated incidence rates accordingtal myocardial infarction (relative risk, 1.01; 95 per- to the length of follow-up. A beneficial effect of as-cent confidence interval, 0.83 to 1.24; P=0.90), or pirin on stroke was observed early in the trial anddeath from cardiovascular causes (relative risk, persisted throughout the trial, with no apparent ben-0.95; 95 percent confidence interval, 0.74 to 1.22; efit of aspirin on myocardial infarction at any pointP=0.68). However, aspirin therapy was associated during follow-up.
with a 22 percent reduction in the risk of transient Neither treatment with vitamin E nor treatment ischemic attack (relative risk, 0.78; 95 percent con- with beta carotene significantly modified the effect
fidence interval, 0.64 to 0.94; P=0.01), with no sig- of aspirin on the primary or secondary end points.
nificant effects on the risk of coronary revascular-
ization (relative risk, 1.04; 95 percent confidence subgroup analyses
interval, 0.90 to 1.20; P=0.61) or death from any There was no evidence that any of the cardiovas-
cause (relative risk, 0.95; 95 percent confidence in- cular risk factors considered, except smoking sta-
terval, 0.85 to 1.06; P=0.32).
tus and age, modified the effect of aspirin on the Figures 1 and 2 present the cumulative inci- primary end point of major cardiovascular events dence rates of major cardiovascular events, stroke, (Table 3). We observed a greater benefit of aspirin Downloaded from www.nejm.org at TORONTO WESTERN HOSP on June 11, 2005 . Copyright 2005 Massachusetts Medical Society. All rights reserved. Table 1. (Continued.)
Hyperlipidemia (%)§ Parental history of myocardial infarction before 60 yr of age (%) 10-yr risk of coronary heart disease (%)¶ No. of risk factors (%)¿ * The body-mass index is the weight in kilograms divided by the square of the height in meters.
† HRT denotes hormone-replacement therapy.
‡ Hypertension was defined as a systolic blood pressure of at least 140 mm Hg, a diastolic blood pressure of at least 90 mm Hg, or self-reported physician-diagnosed hypertension. § Hyperlipidemia was defined as a total cholesterol level of at least 240 mg per deciliter (6.2 mmol per liter) or self-report- ed physician-diagnosed high cholesterol levels.
¶ This variable was calculated with the Framingham risk score for those with blood specimens.
¿ Risk factors were smoking, hypertension, hyperlipidemia, diabetes, and obesity.
Table 2. Incidence and Relative Risk of Confirmed Cardiovascular End Points.
Relative Risk
End Point
(95% CI)*
no. of events Major cardiovascular event† 0.91 (0.80–1.03) 0.83 (0.69–0.99) 0.76 (0.63–0.93) 1.24 (0.82–1.87) 1.04 (0.58–1.86) 0.81 (0.67–0.97) Myocardial infarction 1.02 (0.84–1.25) 1.16 (0.54–2.51) 1.01 (0.83–1.24) Death from cardiovascular causes 0.95 (0.74–1.22) Transient ischemic attack 0.78 (0.64–0.94) 1.04 (0.90–1.20) Death from any cause 0.95 (0.85–1.06) * CI denotes confidence interval.
† A major cardiovascular event was defined as a nonfatal myocardial infarction, a nonfatal stroke, or death from cardiovas- cular causes.
n engl j med 352;13 Downloaded from www.nejm.org at TORONTO WESTERN HOSP on June 11, 2005 . Copyright 2005 Massachusetts Medical Society. All rights reserved. The new england journal of medicine sis in which follow-up data were censored at the time a woman reported having taken less than twothirds of the study medication during the preced-ing year. In this analysis, aspirin, as compared with placebo, reduced the risk of major cardiovascularevents by 13 percent, reduced the risk of stroke by 26 percent, reduced the risk of ischemic stroke by33 percent, and had no significant effect on the riskof myocardial infarction (relative risk, 1.03; 95 per-cent confidence interval, 0.81 to 1.32; P=0.79).
Some women started taking NSAIDs during the trial — a potentially important issue with respect tothe risk of thrombotic events, since certain NSAIDs can compete with aspirin for receptors on plate-lets.19 According to our data, however, the lack of Cumulative Incidence of Major Cardiovascular Events
an effect of aspirin therapy on the risk of myocar- dial infarction was not explained by concomitant Years of Follow-up
Figure 1. Cumulative Incidence Rates of the Primary End Point of Major Car-
diovascular Events.
Reports of gastrointestinal bleeding and peptic A major cardiovascular event was defined as a nonfatal myocardial infarction, ulcer were confirmed by means of follow-up ques- a nonfatal stroke, or death from cardiovascular causes.
tionnaires. These side effects were significantlymore common among women in the aspirin groupthan among women in the placebo group (Table 4).
There were 127 episodes of gastrointestinal bleed- on the risk of major cardiovascular events among ing requiring transfusion in the aspirin group, asformer smokers and those who had never smoked, compared with 91 in the placebo group (relative risk,with an apparent increased risk among current 1.40; 95 percent confidence interval, 1.07 to 1.83;smokers (P for interaction <0.001), although it is P=0.02). Self-reported hematuria, easy bruising,important to interpret this information in the con- and epistaxis were frequent among women in bothtext of multiple comparisons. In addition, age sig- groups, with small but significant excesses amongnificantly modified the effect of aspirin on the risk those in the aspirin group. The percentage of wom-of both major cardiovascular events and myocar- en reporting any symptoms suggestive of gastric up-dial infarction (P for interaction=0.05 and 0.03, re- set was virtually identical in the two groups. Therespectively). The most consistent benefit of aspirin were five fatal gastrointestinal hemorrhages, two inwas observed among the subgroup of women 65 the aspirin group and three in the placebo group.
years of age or older at study entry; in this sub-group, the risk of major cardiovascular events was reduced by 26 percent among those who took aspi-rin as compared with those who took placebo (rel- In this large, placebo-controlled, primary-preven-ative risk, 0.74; 95 percent confidence interval, 0.59 tion trial involving 39,876 initially healthy wom-to 0.92; P=0.008), and the risk of ischemic stroke en, prophylactic aspirin at a dose of 100 mg everywas reduced by 30 percent (relative risk, 0.70; 95 other day was associated with a nonsignificant re-percent confidence interval, 0.49 to 1.00; P=0.05). duction in the risk of major cardiovascular events,This was also the only subgroup in which aspirin, a reduced risk of total stroke and of ischemic stroke,as compared with placebo, significantly reduced a nonsignificant increase in the risk of hemorrhagicthe risk of myocardial infarction (relative risk, stroke, and no significant effect on the risk of my-0.66; 95 percent confidence interval, 0.44 to 0.97; ocardial infarction or death from cardiovascularP=0.04). causes. With respect to the primary end point of To address whether compliance may have af- major cardiovascular events as well as the individ- fected our results, we performed a sensitivity analy- ual end points of fatal or nonfatal stroke and my- Downloaded from www.nejm.org at TORONTO WESTERN HOSP on June 11, 2005 . Copyright 2005 Massachusetts Medical Society. All rights reserved. of Stroke
of Myocardial Infarction
Years of Follow-up
Years of Follow-up
of Ischemic Stroke
of Hemorrhagic Stroke
Years of Follow-up
Years of Follow-up
Figure 2. Cumulative Incidence Rates of Stroke, Myocardial Infarction, Ischemic Stroke, and Hemorrhagic Stroke.
ocardial infarction, consistent benefits of aspirin (relative risk, 0.76; 95 percent confidence inter-were observed among the subgroup of women val, 0.62 to 0.95; P=0.01) but had no significantwho were 65 years of age or older. We found no ev- effect on the risk of stroke (relative risk, 0.97; 95idence that menopausal status, the use or nonuse percent confidence interval, 0.83 to 1.13; P=0.69).
of hormone-replacement therapy after menopause, In analyses stratified according to sex (Fig. 3), com-or global cardiovascular-risk status modified the bined data on women from the Women's Healtheffect of aspirin. As expected, the frequency of side Study, the Hypertension Optimal Treatment (HOT)effects related to bleeding and ulcers was increased study,5,21 and the Primary Prevention Project6 (andamong women who received aspirin.
Roncaglioni MC: personal communication) indi- Our findings must be interpreted in the context cate that aspirin therapy was associated with a sig- of those of other completed, randomized trials of nificant, 19 percent reduction in the risk of strokeaspirin in the primary and secondary prevention of (relative risk, 0.81; 95 percent confidence interval,cardiovascular disease. In secondary prevention, the 0.69 to 0.96; P=0.01), with no reduction in the riskAntithrombotic Trialists' Collaboration showed that of myocardial infarction (relative risk, 0.99; 95aspirin clearly reduced the risk of cardiovascular percent confidence interval, 0.83 to 1.19; P=0.95).
events, myocardial infarction, and ischemic stroke By contrast, the aggregate data on men from thein both men and women.1 To address the effects Physicians' Health Study,2 the British Doctors'of aspirin in primary prevention, we performed a Trial,3 the Thrombosis Prevention Trial,4 the HOTrandom-effects meta-analysis that included current study,5,21 and the Primary Prevention Project6 indi-data from the Women's Health Study, as well as cate that aspirin therapy was associated with a sig-data from five prior trials involving 55,580 partici- nificant, 32 percent reduction in the risk of myo-pants with no history of heart disease.2-6,21 Over- cardial infarction (relative risk, 0.68; 95 percentall, as compared with placebo, aspirin therapy sig- confidence interval, 0.54 to 0.86; P=0.001) and anificantly reduced the risk of myocardial infarction nonsignificant increase in the risk of stroke (rela- n engl j med 352;13 Downloaded from www.nejm.org at TORONTO WESTERN HOSP on June 11, 2005 . Copyright 2005 Massachusetts Medical Society. All rights reserved. The new england journal of medicine 1.23 (0.87–1.75) 1.17 (0.86–1.59) 0.66 (0.44–0.97) 1.50 (1.06–2.13) 0.87 (0.68–1.10) 0.92 (0.66–1.28) 1.11 (0.80–1.54) 0.98 (0.65–1.47) 0.69 (0.38–1.26) 1.49 (0.91–2.46) 1.12 (0.77–1.63) 0.9 (0.67–1.21) 0.95 (0.72–1.25) 1.10 (0.83–1.46) 1.28 (0.70–2.34) 1.02 (0.66–1.56) 1.11 (0.78–1.58) 0.92 (0.66–1.30) 0.80 (0.57–1.14) 0.80 (0.57–1.12) 0.70 (0.49–1.00) 1.14 (0.76–1.70) 0.67 (0.53–0.85) 0.69 (0.50–0.94) 0.78 (0.55–1.11) 0.85 (0.56–1.29) 1.01 (0.56–1.79) 0.47 (0.26–0.85) 0.71 (0.52–0.96) 0.73 (0.56–0.96) 0.78 (0.58–1.05) 0.78 (0.47–1.29) 0.82 (0.53–1.26) 0.74 (0.48–1.15) 0.85 (0.63–1.16) 0.84 (0.62–1.14) 0.78 (0.57–1.08) 1.14 (0.80–1.63) 0.75 (0.61–0.92) 0.75 (0.58–0.98) 0.83 (0.60–1.13) 0.96 (0.65–1.41) 0.97 (0.59–1.60) 0.57 (0.35–0.94) 0.98 (0.72–1.34) 0.76 (0.57–1.00) 0.76 (0.59–0.98) 0.88 (0.68–1.13) 0.94 (0.61–1.45) 0.89 (0.61–1.30) 0.75 (0.51–1.09) 0.73 (0.54–0.98) vents, According to Baseline Characteristics.*
1.01 (0.81–1.26) 0.98 (0.80–1.20) 0.74 (0.59–0.92) 1.30 (1.03–1.64) 0.80 (0.69–0.93) 0.82 (0.68–1.00) 0.89 (0.72–1.11) 1.05 (0.81–1.36) 0.88 (0.61–1.26) 0.91 (0.66–1.27) 0.98 (0.78–1.23) 0.86 (0.71–1.03) 0.84 (0.71–1.00) 0.96 (0.81–11.15) 1.04 (0.75–1.45) 0.94 (0.72–1.22) 0.94 (0.74–1.20) 0.81 (0.65–1.00) Major CV E
Total No.
Incidence and Relative Risk of C
120–129/75–84 130–139/85–89 Body-mass index† Menopause and HRT Downloaded from www.nejm.org at TORONTO WESTERN HOSP on June 11, 2005 . Copyright 2005 Massachusetts Medical Society. All rights reserved. 1.03 (0.77–1.39) 1.00 (0.77–1.31) 1.48 (0.88–2.49) 0.96 (0.77–1.18) 0.89 (0.55–1.43) 1.03 (0.81–1.30) 0.99 (0.69–1.42) 0.78 (0.50–1.20) 1.48 (0.86–2.53) 0.90 (0.53–1.53) 0.94 (0.65–1.34) 0.94 (0.64–1.36) 1.30 (0.86–1.98) risk, CI confidence interval, HRT hormone-replacement or as self-reported physician-diagnosed hypertension.
ular causes. The total number of cardiovascular events may 0.68 (0.50–0.93) 0.83 (0.64–1.08) 0.42 (0.22–0.82) 0.81 (0.66–1.00) 0.73 (0.41–1.31) 0.75 (0.60–0.95) 0.70 (0.50–1.00) 0.92 (0.58–1.46) 0.57 (0.31–1.04) 0.76 (0.48–1.22) 0.70 (0.49–1.01) 0.73 (0.50–1.06) 0.86 (0.54–1.35) ed high cholesterol levels.
0.62 (0.47–0.83) 0.99 (0.78–1.24) 0.46 (0.25–0.85) 0.87 (0.72–1.05) 0.74 (0.43–1.26) 0.82 (0.67–1.00) 0.80 (0.59–1.07) 1.11 (0.74–1.68) 0.54 (0.30–0.98) 0.86 (0.58–1.28) 0.88 (0.64–1.20) 0.70 (0.50–0.98) 0.83 (0.54–1.28) 0.80 (0.66–0.97) 0.98 (0.84–1.16) 0.9 (0.79–1.03) 0.86 (0.61–1.21) 0.90 (0.78–1.04) 0.86 (0.69–1.07) 0.96 (0.72–1.27) 1.06 (0.74–1.52) 0.86 (0.64–1.15) 0.92 (0.73–1.14) 0.80 (0.63–1.01) 1.07 (0.80–1.42) ramingham risk score for those with blood specimens (28,345 women).
before 60 yr of age , and CHD coronary heart disease.
rental history of myocardial in- 10-yr risk of CHD¶ No. of risk factors¿ A major cardiovascular (CV) event was defined as a nonfatal myocardial infarction, a nonfatal stroke, or death from cardiovasc not sum to 39,876 owing to missing data for some variables. Asp denotes aspirin, Pla placebo, no. number of women, RR relative The body-mass index is the weight in kilograms divided by the square of the height in meters.
Hypertension was defined as a systolic blood pressure of at least 140 mm Hg, a diastolic blood pressure of at least 90 mm Hg, Hyperlipidemia was defined as a total cholesterol level of at least 240 mg per deciliter or as self-reported physician-diagnos This variable was calculated with the F Risk factors were smoking, hypertension, hyperlipidemia, diabetes, and obesity n engl j med 352;13 Downloaded from www.nejm.org at TORONTO WESTERN HOSP on June 11, 2005 . Copyright 2005 Massachusetts Medical Society. All rights reserved. The new england journal of medicine Table 4. Incidence and Relative Risk of Side Effects.*
Relative Risk
Side Effect
no. of events (%) 1.22 (1.10–1.34) Requiring transfusion 1.40 (1.07–1.83) 1.32 (1.16–1.50) 1.06 (1.01–1.12) 1.40 (1.37–1.45) 1.16 (1.11–1.22) Any report of gastric upset 0.99 (0.97–1.02) * The presence of gastrointestinal bleeding or peptic ulcer was confirmed by a specific follow-up questionnaire. CI denotes tive risk, 1.13; 95 percent confidence interval, 0.96 also believe it unlikely that a reduction in the effica-to 1.33; P = 0.15). The differences between men cy of aspirin over time is a viable explanation, sinceand women were significant at the P=0.01 level for the cumulative incidence data presented in Figuresmyocardial infarction and at the P=0.005 level for 1 and 2 offer no support for this hypothesis. Fur-stroke.
thermore, suboptimal compliance is an unlikely ex- The reasons for any sex-based differences in the planation, since aspirin did not decrease the risk efficacy of aspirin for primary prevention are un- of myocardial infarction among women with highclear and require further exploration. Although the rates of compliance — an observation again in con-observed reductions in the risk of stroke could be trast to data on stroke among the same women.
due to chance, the reduced risk of transient ische- With regard to daily clinical practice, our data mic attack associated with aspirin therapy adds demonstrate that aspirin therapy was associatedsupport to the possibility of a causal interpretation. with a net reduction in the risk of stroke amongWe cannot rule out the possibility that our null find- women, with a reduction in the risk of the far moreing for the risk of myocardial infarction in women common ischemic stroke and an increase in the riskwas due to an insufficient dose of aspirin or to the of hemorrhagic stroke. This observation is partic-alternate-day regimen. However, we believe these ularly relevant, since as compared with men, wom-explanations to be unlikely for three reasons. en have a relatively greater proportion of strokes First, we have previously shown that the dose than of myocardial infarctions. Among women in of 100 mg every other day used in the Women's the placebo group, there were more strokes thanHealth Study reduces thromboxane levels by 93 per- myocardial infarctions (266 vs. 193), and thus, thecent and prostacyclin levels by 85 percent and that ratio of incident strokes to incident myocardial in-these effects are similar in men and women.22 Sec- farctions was 1.4:1, as compared with the ratio ofond, in the HOT study, a 75-mg daily dose of aspi- 0.4:1 among men in the Physicians' Health Study.2rin significantly lowered the risk of myocardial in- From a policy perspective, our findings clearly dem-farction overall, with a 42 percent reduction in the onstrate the importance of studying women as wellrisk among men but a far smaller and nonsignifi- as men in major cardiovascular clinical trials.
cant reduction among women.21 Third, since the An interesting finding in our subgroup analyses dose and alternate-day regimen of aspirin used in was that the most consistent benefit of aspirin wasour study were adequate to lower the risk of stroke, observed among women 65 years of age or older.
it is unlikely that any hypothesized sex-based dif- This group of 4097 women composed 10 percentferences in the resistance to aspirin were at play of the study population yet had almost one thirdoverall. However, resistance to aspirin may be more of the cardiovascular events. In this group, aspirinprevalent among smokers,23 and this resistance use, as compared with placebo use, led to 44 few-may have played some role in the increased risk er myocardial infarctions, strokes, or deaths fromwith aspirin observed among current smokers. We cardiovascular causes (P=0.008) but to 16 more Downloaded from www.nejm.org at TORONTO WESTERN HOSP on June 11, 2005 . Copyright 2005 Massachusetts Medical Society. All rights reserved. Relative Risk of Myocardial Infarction
Relative Risk of Stroke among Men
among Men
RR=0.68 (0.54–0.86) RR=1.13 (0.96–1.33) Relative Risk of Myocardial Infarction
Relative Risk of Stroke among Women
among Women
RR=0.99 (0.83–1.19) RR=0.81 (0.69–0.96) Figure 3. Aspirin in the Primary Prevention of Myocardial Infarction and Stroke among Men and Women.
The results of a sex-specific random-effects meta-analysis of data from six trials are shown: the British Doctors' Trial (BDT), the Physicians' Health Study (PHS), the Thrombosis Prevention Trial (TPT), the Hypertension Optimal Treatment (HOT) study, the Primary Prevention Project (PPP), and the current Women's Health Study (WHS). The relative risk (RR) and 95 percent confidence interval (in parentheses) are shown for each trial (indicated by the box and horizontal line through each box, respectively), and the relative risk is shown for the combined results (indicated by the diamond and the dashed line in each graph). For the relative risk of myocardial infarction among women, the dashed line is coincident with the solid line at 1.00. The size of the box is proportional to the amount of information in the corresponding trial.
gastrointestinal hemorrhages requiring transfu- woman consults her physician or health care pro-sion (P=0.05), emphasizing, as with any agent, the vider, so that the net absolute benefits and risks forimportance of balancing benefits and risks. This the individual patient can be ascertained.
age-based difference deserves further investigation.
Supported by grants (HL-43851 and CA-47988) from the Nation- With respect to guidelines in primary preven- al Heart, Lung, and Blood Institute and the National Cancer Insti- tute, Bethesda, Md. Aspirin and aspirin placebo were provided by tion, in 2002, the Preventive Services Task Force24 Bayer HealthCare. Vitamin E and vitamin E placebo were providedand the American Heart Association9 recommend- by the Natural Source Vitamin E Association.
ed aspirin for adults whose 10-year risks of a first Dr. Ridker reports having received grant support from Bayer. Dr.
Cook reports having served as a consultant to Bayer. Dr. Gaziano re- coronary-heart-disease event were at least 6 percent ports having served as a consultant to, and receiving grant supportand 10 percent, respectively. However, this may be from, Bayer and McNeil. Dr. Hennekens reports having served as acomplex for women, since in our study overall, as- consultant to Bayer and McNeil and receiving grant support from pirin lowered the risk of stroke without affecting We are indebted to the 39,876 participants in the Women's Health the risk of myocardial infarction or death from car- Study for their dedicated and conscientious collaboration; to the en-diovascular causes. Thus, as with men, any decision tire staff of the Women's Health Study, under the leadership of Da- vid Gordon, Maria Andrade, Susan Burt, Mary Breen, Marilyn Chown, about the use of aspirin in primary prevention Lisa Fields-Johnson, Georgina Friedenberg, Inge Judge, Jean Mac-among women must ultimately be made after a Fadyen, Geneva McNair, Laura Pestana, David Potter, Philomena n engl j med 352;13 Downloaded from www.nejm.org at TORONTO WESTERN HOSP on June 11, 2005 . Copyright 2005 Massachusetts Medical Society. All rights reserved. Quinn, Claire Ridge, Fred Schwerin, and Harriet Samuelson; to Pradhan, Kathryn Rexrode, Bernard Rosner, and H. Jacqueline Suk Christine Albert, Michelle Albert, Gavin Blake, Claudia Chae, Wendy for their assistance in the design and conduct of the trial; and espe- Chen, Richard Doll, Carlos Kase, Tobias Kurth, Richard Peto, Aruna cially to James Taylor for chairing the end-points committee.
a p p e n d i x
Members of the data and safety monitoring board included L. Cohen, R. Collins, T. Colton, D. DeMets, I.C. Henderson, A. La Croix, R. Pren-tice, and N. Wenger (chair) and M.F. Cotch, F. Ferris, L. Friedman, P. Greenwald, N. Kurinij, M. Perloff, E. Schron, and A. Zonderman (ex of-ficio members).
r e f e r e n c e s
Antithrombotic Trialists' Collaboration.
cardiovascular events: a summary of the evi- tation and incidence of cancer and cardio- Collaborative meta-analysis of randomised dence for the U.S. Preventive Services Task vascular disease: the Women's Health Study.
trials of antiplatelet therapy for prevention Force. Ann Intern Med 2002;136:161-72.
J Natl Cancer Inst 1999;91:2102-6.
of death, myocardial infarction, and stroke Pearson TA, Blair SN, Daniels SR, et al.
18. Pulcinelli FM, Pignatelli P, Celestini A,
in high risk patients. BMJ 2002;324:71-86.
AHA guidelines for primary prevention of Riondino S, Gazzaniga PP, Violi F. Inhibi- [Erratum, BMJ 2002;324:141.] cardiovascular disease and stroke: 2002 up- tion of platelet aggregation by aspirin pro- Steering Committee of the Physicians' date: consensus panel guide to comprehen- gressively decreases in long-term treated pa- Health Study Research Group. Final report sive risk reduction for adult patients without tients. J Am Coll Cardiol 2004;43:979-84.
on the aspirin component of the ongoing coronary or other atherosclerotic vascular 19. Catella-Lawson F, Reilly MP, Kapoor SC,
Physicians' Health Study. N Engl J Med 1989; diseases. Circulation 2002;106:388-91.
et al. Cyclooxygenase inhibitors and the an- 10. Mosca L, Appel LJ, Benjamin EJ, et al.
tiplatelet effects of aspirin. N Engl J Med Peto R, Gray R, Collins R, et al. Ran- Evidence-based guidelines for cardiovascu- domised trial of prophylactic daily aspirin in lar disease prevention in women. Circulation 20. Kurth T, Glynn RJ, Walker AM, et al. In-
British male doctors. Br Med J (Clin Res Ed) hibition of clinical benefits of aspirin on 11. Montgomery PR, Berger LG, Mitenko
first myocardial infarction by nonsteroidal The Medical Research Council's Gener- PA, Sitar DS. Salicylate metabolism: effects antiinflammatory drugs. Circulation 2003; al Practice Research Framework. Thrombo- of age and sex in adults. Clin Pharmacol Ther sis prevention trial: randomised trial of low- 21. Kjeldsen SE, Kolloch RE, Leonetti G, et
intensity oral anticoagulation with warfarin 12. Manson JE, Hsia J, Johnson KC, et al. Es-
al. Influence of gender and age on prevent- and low-dose aspirin in the primary preven- trogen plus progestin and the risk of coro- ing cardiovascular disease by antihyperten- tion of ischaemic heart disease in men at in- nary heart disease. N Engl J Med 2003;349: sive treatment and acetylsalicylic acid: the creased risk. Lancet 1998;351:233-41.
HOT study. J Hypertens 2000;18:629-42.
Hansson L, Zanchetti A, Carruthers SG, 13. Rodondi N, Bauer DC. Assessing the risk/
22. Ridker PM, Hennekens CH, Tofler GH,
et al. Effects of intensive blood-pressure low- benefit profile before recommending aspi- Lipinska I, Buring JE. Anti-platelet effects of ering and low-dose aspirin in patients with rin for the primary prevention of cardiovas- 100 mg alternate day oral aspirin: a random- hypertension: principal results of the Hyper- cular events. Am J Med 2004;117:528-30.
ized, double-blind, placebo-controlled trial tension Optimal Treatment (HOT) random- 14. Buring JE, Hennekens CH. The Wom-
of regular and enteric coated formulations ised trial. Lancet 1998;351:1755-62.
en's Health Study: summary of the study de- in men and women. J Cardiovasc Risk 1996; Collaborative Group of the Primary Pre- sign. J Myocardial Ischemia 1992;4:27-9.
vention Project. Low-dose aspirin and vita- 15. Rexrode KM, Lee IM, Cook NR, Henne-
23. Sanderson S, Emery J, Baglin T, Kin-
min E in people at cardiovascular risk: kens CH, Buring JE. Baseline characteristics month A-L. Narrative review: aspirin resis- a randomised trial in general practice. Lan- of participants in the Women's Health Study.
tance and its clinical implications. Ann In- cet 2001;357:89-95. [Erratum, Lancet 2001; J Womens Health Gend Based Med 2000;9: tern Med 2005;142:370-80.
24. Preventive Services Task Force. Aspirin
Eidelman RS, Hebert PR, Weisman SE, 16. Atiya M, Kurth T, Berger K, Buring JE,
for the primary prevention of cardiovascular Hennekens CH. An update on aspirin in the Kase CS. Interobserver agreement in the clas- events: recommendation and rationale. Ann primary prevention of cardiovascular dis- sification of stroke in the Women's Health Intern Med 2002;136:157-60.
ease. Arch Intern Med 2003;163:2006-10.
Study. Stroke 2003;34:565-7.
Copyright 2005 Massachusetts Medical Society. Hayden M, Pignone M, Phillips C, Mul- 17. Lee I-M, Cook NR, Manson JE, Buring
row C. Aspirin for the primary prevention of JE, Hennekens CH. b-Carotene supplemen- clinical problem-solving series
The Journal welcomes submissions of manuscripts for the Clinical Problem-Solving
series. This regular feature considers the step-by-step process of clinical decision
making. For more information, please see http://authors.nejm.org.
Downloaded from www.nejm.org at TORONTO WESTERN HOSP on June 11, 2005 . Copyright 2005 Massachusetts Medical Society. All rights reserved.

Source: http://www.cii.org.ar/epidemiology/modulo%203/1293.pdf