Answer Happy

1. State the Population, Intervention, Control , Outcome
(PICO) for the aspirin component of this study. For Population
provide three descriptors (occupation, gender, age range)?
(Answer in 4 bullet points)
2. Was the rate of cardiovascular
deaths or risk of cardiovascular deaths
measured in this study. Justify your answer in one
sentence.
3.There were 254.8 per 100,000 per year Myocardial deaths in the
aspirin group compared with 439.7 per 100,000 per year in the
placebo group. How many participants need to be treated with
aspirin for 1 year to result in one (1) fewer myocardial
death. Show calculations.
les e for brief summ scribing informail The New England Journal of Medicine ©Copyright, 1989, by the Massachusetts Medical Society JULY 20, 1989 FINAL REPORT ON THE ASPIRIN COMPONENT OF THE ONGOING PHYSICIANS' HEALTH STUDY STEERING COMMITTEE OF THE PHYSICIANS' HEALTH STUDY RESEARCH GROUP* Volume 321 Abstract The Physicians' Health Study is a randomized, double-blind, placebo-controlled trial designed to deter- mine whether low-dose aspirin (325 mg every other day) decreases cardiovascular mortality and whether beta car- otene reduces the incidence of cancer. The aspirin. component was terminated earlier than scheduled, and the preliminary findings were published. We now pre- sent detailed analyses of the cardiovascular component for 22,071 participants, at an average follow-up time of 60.2 months. There was a 44 percent reduction in the risk of myocar- dial infarction (relative risk, 0.56; 95 percent confidence. interval, 0.45 to 0.70; P<0.00001) in the aspirin group (254.8 per 100,000 per year as compared with 439.7 in the placebo group). A slightly increased risk of stroke among those taking aspirin was not statistically significant; this trend was observed primarily in the subgroup with hemor- rhagic stroke (relative risk, 2.14; 95 percent confidence interval, 0.96 to 4.77; P = 0.06). No reduction in mortality LTHOUGH A aspirin-like properties, was prescribed for pain relief by Hippocrates in the fifth century B.C., the possible role of aspirin in reducing the risk of cardio- From the Channing Laboratory, Departments of Medicine and Preventive Med- icine, Harvard Medical School and Brigham and Women's Hospital, Boston. Address reprint requests to Dr. Charles H. Hennekens at 55 Pond Ave., Brook- line, MA 02146 Supported by grants (HL-26490, HL-34595, CA-34944, and CA-40360) from the National Institutes of Health. *The members of the Steering Committee and of the Physicians' Health Study Research Group are as follows: Steering Committee: Charlene Belanger, M.A.. Julie E. Buring, Sc.D., Nancy Cook, Sc.D., Kimberley Eherlein, M.P.H., Sam- uel Z 1 Z. Goldhaber, M.D., David Gordon, M.A., Charles H. Hennekens, M.D. (chairman), Sherry L. Mayrent, Ph.D., Richard Peto, F.R.S., Bernard Rosner, Ph.D., Meir J. Stampfer, M.D., Fran Stubblefield, B.S., and Walter C. Willett. M.D. The Steering Committee also includes the members of the Data Monitoring Board, as well as Thomas Ryan, M.D., Thomas Blaszkowski, Ph.D., (ex offi- cio), and Andrew Vargosko, Ph.D. (ex officio). Physicians' Health Study Re- search Group: Charlene Belanger, M.A., Eugene Braunwald. M.D., Julie E. Buring, Sc.D., Nancy Cook, Sc.D., Richard Doll, F.R.S., Kimberley Eberlein, M.P.H.. Samuel Z. Goldhaber, M.D., David Gordon, M.A., Charles H. Henne- kens, M.D. (principal investigator), Edward 11. Kass, M.D., Janet Lang, Sc.D., JoAnn Manson, M.D., Micheline Mathews-Roth, M.D., Sherry L. Mayreat, Ph.D.. Richard Peto, F.R.S., Bernard Rosner, Ph.D., Suzanne Satterfield, M.D., Frank E. Speize M.D., Meir J. Stampfer, M.D., Fran Stubblefield, B.S., James O. Taylor, M.D., and Walter C. Willett, M.D. End Points Commit tee: Harris Funkenstein, M.D., Samuel Z. Goldhaber, M.D., Meir 1. Stampfer, M.D., and James O. Taylor, M.D. (chairman). Dana Monitoring Board 5. Cohen. nel De Theodore Colton, Sc.D., David DeMets, Ph.D. Deykin, M.D., Lawrence Friedman, M.D., and Bernard Rosner, Ph.D. Recently added members are John Cairns, M.D., Peter Green- wald, M.D., and Charles H. Hennekens, M.D. The retired chairman is George B. Hutchison, M.D. Number 3 from all cardiovascular causes was associated with aspirin (relative risk, 0.96; 95 percent confidence interval, 0.60 to 1.54). Further analyses showed that the reduction in the risk of myocardial infarction was apparent only among those who were 50 years of age and older. The benefit was present at all levels of cholesterol, but appeared greatest at low lev- els. The relative risk of ulcer in the aspirin group was 1.22 (169 in the aspirin group as compared with 138 in the placebo group: 95 percent confidence interval, 0.98 to 1.53; P=0.08), and the relative risk of requiring a blood transfusion was 1.71. This trial of aspirin for the primary prevention of cardio- vascular disease demonstrates a conclusive reduction in the risk of myocardial infarction, but the evidence concern- ing stroke and total cardiovascular deaths remains in- conclusive because of the inadequate numbers of physi- cians with these end points. (N Engl J Med 1989; 321: 129-35.) vascular disease has been recognized only very recent- ly. Such a possibility derives from the capacity of aspi- rin in low doses to inhibit cyclooxygenase-dependent platelet enzymes virtually completely, resulting in the inhibition of aggregability for the life of the platelet.' These effects are so profound that higher doses add little benefit but do increase the risk of side effects. Although an early case-control study raised the pos- sibility of a large benefit, most observational studies have suggested a cardiovascular benefit of about 20 percent. In such circumstances, the amount of uncon- trolled confounding in case-control or cohort studies may be as large as the small-to-moderate effects being sought"; consequently, conclusive data can result only from a randomized trial whose sample is sufficiently large. The Physicians' Health Study is a double-blind, placebo-controlled, randomized trial designed to test two primary-prevention hypotheses in a population of healthy male physicians: whether aspirin in low doses. (Bufferin, Bristol-Myers Products, 325 mg every other day) reduces mortality from cardiovascular disease, and whether beta carotene (Lurotin, BASF, 50 mg on alternate days) decreases the incidence of cancer. Al- though the beta carotene component of the trial is continuing at least through 1990, the Data Monitoring Board recommended the early termination of the
July 20, 1989 cause was documented by convincing evidence of a cardiovascular mechanism from all available sources, including death certificates, hospital records, and for death outside the hospital-observers impressions. For the end points of myocardial infarction and stroke (Tables 1 and 2), only the first event within each category was counted. For cardiovascular mortality (Table 3), all deaths were included in the analyses. Thus, for the 15 subjects who had both a nonfatal myocar- dial infarction and a nonfatal stroke, both events were counted as end points. For the 23 who had a nonfatal myocardial infarction (or stroke) followed by death from a cardiovascular cause, the nonfatal event was included in our analysis of myocardial infarction (or stroke), and the fatal event was included in cardiovascular deaths. In addition, we performed analyses using as end points only the first cardiovascular event experienced by the participants myocardial infarction, stroke, or cardiovascular death and this method yield- ed virtually identical results. For the combined end point of nonfa tal myocardial infarction, nondatal stroke, and death from a cardio- vascular cause, only a participant's first cardiovascular event was counted. The relative risk was calculated as the number of events per person-year of observation in the aspirin group divided by the cor responding number in the placebo group after adjustment for age and for assignment to beta carotene treatment," even though no significant effect of beta carotene was observed on any of the major cardiovascular end points. Since the trial was so large, the adjusted ration were, in practice, never materially different from the ratio of y different from the racios of the crude numbers affected in each group." For each relative risk, the two-sided P value and the 95 percent confidence interval were calculated. Multiple logistic-regression analysis was used in con- trol simultaneously for the joint effects of any small differences in base-line cardiovascular risk factors, although no individual factor differed significantly between the aspirin and placebo groups. RESULTS There were 139 myocardial infarctions among those assigned to aspirin and 239 among those assigned to aspirin placebo (relative risk, 0.56; 95 percent confi- dence interval, 0.45 to 0.70; P<0.00001) (Table 1). This represents a 44 percent reduction in risk, and the benefits of aspirin were significant for both fatal and nonfatal myocardial infarction. In terms of ab- solute rates of events, the figures can be extrap- olated to 254.8 per 100,000 per year in the aspirin group and 439.7 per 100,000 per year in the placebo group. As for total stroke, there were 119 events in the aspirin group and 98 in the placebo group- an in- crease in risk that was not statistically significant (rel- ative risk, 1.22; 95 percent confidence interval, 0.93 to 1.60; P 0.15). Strokes were then subdivided into is- chemic and hemorrhagic events and, further, into those resulting in mild, moderate, or severe disability or death (Table 2). In the subgroup with hemorrhagic strokes, aspirin was associated with an increased risk that was of borderline statistical significance (relative risk, 2.14, 95 percent confidence interval, 0.96 to 4.77; P=0.06). This subgroup included 10 mild hemor- rhagic strokes in the aspirin group and 6 in the pla cebo group (relative risk, 1.67; 95 percent confidence interval, 0.61 to 4.57; P 0.32), as well as 13 moder- ate-to-severe or fatal hemorrhagic strokes in the aspi- rin group and 6 in the placebo group (relative risk, 2.19, 95 percent confidence interval, 0.84 to 5.69; P-0.11). With respect to total cardiovascular mortality (Ta ble 3), there were 81 deaths among those assigned to aspirin and 83 among those given placebo (relative risk, 0.96; 95 percent confidence interval, 0.60 to 1.54; THE NEW ENGLAND JOURNAL OF MEDICINE 1:30 blinded aspirin component of the trial on December 18, 1987. This decision was based on all available evi- dence, including three major considerations: the pres- ence of a significant (P<0.00001) reduction in the risk of total myocardial infarction among those in the aspi- rin group; the fact that no effect of aspirin on cardio- vascular mortality could be detected in the trial until the year 2000 or later, because of the exception- ally low cardiovascular death rates among the par ticipating physicians; and the fact that aspirin was subsequently prescribed for more than 85 percent of the participants who experienced nonfatal vascular events, which made any finding about cardiovascular. mortality particularly difficult to interpret. The trial's preliminary findings were published on January 28, 1988. We report here the results of the final analyses of the cardiovascular component up to January 25, 1988, when the participants were told whether they had been assigned to the aspirin or the placebo group. METHODS The subjects and methods of the Physicians' Health Study were described in detail in the preliminary repon." Briefly, 22,071 physi cians were randomly assigned, according to two-by-two factorial design to one of four treatment groups aspirin and beta caro tene, aspirin and beta carotene placebo, aspirin placebo and beta carstent, or aspirin placebo and beta carotene placebo. Alto- gether, 11,037 physicians were assigned at random to receive aspirin and 11,034 to receive aspirin placebo. All 22,071 participants ran- domly assigned to a treatment group have been included in all analyses. Every six months for the first year and annually thereafter, the participants were sent a supply of monthly calendar packs (pre- vided by Bristol-Myers Products) containing white tablets (aspirin or placebo) for odd-numbered days and red capsules (beta carotene or placebo) for even-numbered days. They were also sent brief questionnaires asking about their compliance with the treatment regimen and the occurrence of any relevant events. By January 25, 1988, the participants had been followed for an average of 60.2 months (range, 45.8 10 77.0); 99.7 percent were still providing information on morbidity, and the vital status of all 22,071 doctors was known. The reported consumption of aspirin or other platelet-active drugs was 85.71 percent in the aspirin group and 14.23 percent in the placebo group A total of 1269 physicians (624 taking aspirin and 645 taking aspirin placebo) requested an enteric-coated preparation (supplied by Bristol-Myers Products), and an additional 29 (16 assigned to aspirio and 13 assigned to placebo) specifically requested Ecotrin or its placebo (supplied by SmithKline Beckman). When a participant reported a relevant outcome event, written consent for the review of his medical records was obtained. The formation was requested from hospitals and responsible physi- cians. Reported diagnoses of cardiovascular disease or deaths were considered confirmed only after the examination of all available information by an End Points Committee of physician that included two internists, a cardiologist, and a neurologist, all blinded to the assigned treatment. When written consent or the relevant reconds could not be obtained, a reported event could not be confirmed. Records were available for review for 95.6 percent of the reported myocardial indarctions, 95.2 percent of the strokes, and 54.8 percent af all deaths. All our analyses were based on confirmed FOLL The diagnoses of nonfatal myocardial infarction were confirmed with use of the criteria of the World Health Organization Nonfa tal struke was defined as a typical omurologic deficit that was soddes rapid in onet, lasted more than 24 hours, and was attributable to a cerebrovascular event, Strokes were further classified according to the severity of the residual impairment at the time of hospital dia charge (mild, moderate, or severe) and according to the probable cause ischemic or hemorrhagie) on the basis of medical records and the judgment of the neurologist. Drath due to a cardiovascular Vol 321 No. 3 Table 1. Confirmed Cardiovascular End Points in the Aspirin Component of the Physicians Health Study. According to Treatment Group. A • 359 PLACE RELATIVE CONFIDENCE Coor G கொடர் R INTERVAL De Pr PVAL svocential infarction Fatal Nonfatal Total 10 26 129 213 139 239 54,560.0 54.355.7 034 0.15-0.75 0.007 0.39 0.47-0.74 0.00001 0.45-0.70 0.00001 HE 0.55 Person years of obrvarinn Stroke Fatal Notaral Total Penon-years of observation 9 6 1.51 0.54-4.28 0.43 110 92 1.20 0.91-1.59 0.20 119 98 1.21 54.650.3 54.635.8 093-140 0.15 eve Ation that could not be cred becco were nailable included became not 17aycandal function (30 in the plein group and 7 is the plache group and 11 strokes C aspirin in pid 8 placbe P 0.87). In the five categories of death from specific cardiovascular causes, there was only one statistically significant finding; a reduction in the rate of fatal myocardial infarction (10 in the aspirin group and 28 in the placebo group; P= 0.004). For sudden death, there was an apparent increase in risk that did not achieve statistical significance (22 in the aspirin group and 12 in the placebo group; P=0.09). Combining the category of fatal acute myocardial infarction (ICD International Classification of Diseases number] 410) with the categories of all other fatal ischemic heart disease (ICD 411 to 414) yielded 34 deaths in the aspirin group and 53 in the placebo group (relative risk, 0.60; 95 percent confidence interval, 0.37 to 0.98; P=0.04). The addition of sudden death (ICD 798) resulted in 56 deaths in the aspirin group and 65 in the placebo group (relative risk, 0.86; 95 percent confi- dence interval, 0.60 to 1.22; P=0.41). There was no reduction in the risk of all deaths from noncardiovas- cular causes (124 in the aspirin group vs. 133 in the placebo group; relative risk, 0.93; 95 percent confi- dence interval, 0.72 to 1.20; P=0.59). Thus, there were 205 deaths with a confirmed cause in the aspirin group and 216 in the placebo group (relative risk, 0.95; 95 percent confidence interval, 0.79 to 1.15; P-0.60). To help clarify a risk-to-benefit ratio, we considered. a combined end point consisting of nonfatal myocar- dial infarction, nonfatal stroke, and death from a car- diovascular cause. There were 307 important vascular events among those assigned to aspirin and 370 among those assigned to placebo (relative risk, 0.82; 95 per- cent confidence interval, 0.70 to 0.96; P=0.01). This represents a statistically significant, 18 percent reduc- tion in important vascular events among those as- signed to aspirin. As expected with a sample of 22,071 participants randomly assigned to treatment groups, there were to differences in the base-line characteristics-age, tigarette smoking, incidence of diabetes mellitus, parental history of myocardial infarction, cholesterol level, systolic blood pressure, diastolic blood pressure, alcohol use, amount of vigorous exercise, and body- 131 mass index. When the possible effects of any differ ences in the joint distributions of these risk factors. were taken into account through logistic regression, the relative risks for each cardiovascular end point were unchanged. We also examined the possible effects of aspirin in subgroups of physicians with various cardiovascu- lar risk factors. As shown in Table 4, the effects of t aspirin on the risk of myocardial infarction were modified by two coronary risk factors-age and blood cholesterol level. The reduction in the risk of myo-i cardial infarction associated with the use of aspirin was apparent only in those 50 years of age or old- er (P=0.02). No consistent effect of age on the relation between aspirin and either stroke or cardio- vascular mortality was observed. For cholesterol, the beneficial effects of aspirin on myocardial infarction were apparent at all levels but appeared greatest at low levels (P = 0.04). For cigarette smoking, the reduction in the risk of myocardial infarction associ- ated with the use of aspirin was similar among those who had never smoked, past smokers, and current smokers. For stroke, cigarette smoking did not mod- ify the effect of aspirin, but for cardiovascular mor- tality, it appeared to do so (P=0.05). However, nei- ther the observed reduction in risk among nonsmokers (P=0.18) nor the apparently increased risk among current smokers (P-0.20) was significant. Final- ly, blood-pressure levels had no consistent effect on the association between aspirin and myocardial infarction, stroke, or mortality from cardiovascular causes. During the 60.2 months of follow-up, gastrointesti- nal discomfort was reported by 26.1 percent of the aspirin group and 25.6 percent of the placebo group; 0.5 percent was therefore attributable to the active drug, a nonsignificant excess (P=0.45) (Table 5).. For all gastrointestinal symptoms except ulcer, the corresponding figures were 34.8 percent and 34.2 per- cent, respectively (P=0.48, also not significant). Table 2. Subcategories of Stroke, According to Treatment Group, 91% A PLACERO RELATIVE CONFIDENCE GROUP C R INTERVAL PYALIN TYPE OF STO Ischemic 69 1.13 0.80-1.00 0.48 61 20 1.05 0.57-1.95 0.38 Mad Moderate, severe. or fatal Unknown severity Total Hemontagic Mild 21 1 82 1.01 0.82-1.50 1.67 0.61-4.57 0.32 0.54-5.09 0.11 2.19 Moderate, sevent of fatal Toal Unknown cause Mid 2.14 0.96-4.77 0.06 - P Moderats, severe, or fatal 1 Unknown severity Total Total 119 98 1.22 0.93-1.40 0.15 affecting facing. "Severity was defined as follows mild, impen functional impairment and a change in way of the depends ASPIRIN COMPONENT OF PHYSICIANS HEALTH STUDY-HENNEKENS ET AL. 91 10 13 23 4 I H 6 16 12 t 2 R 1
D BI TON 10 24 132 Table 3. Confirmed Deaths, According to Treatment Group. A Ca 159 PLACE RELATIE C PLAC RELATE CY OHI Rox 8.96 0.80-1.34 0.37 031 014-868 0004 Tout canfrancalar death 21 Aux madal fac (410) 25 0.97 0.60-1.55 0.89 Oper ischemic heart isme (411-414) Sudden death (798) 22 17 1.96 0.91-4.22 0.00 1.44 0.54-3.88 047 10 Stroke 1430, 431, 434, 403601 Oter cardiovascolar (400, 15 11 1.38 0.62-3.05 0.43 421, 424, 425, 428,429. 437, 440, 441) Total mandiocular 1244 133 091 072-1.30 0.50 deste Tond deaths with contred 205 216 0.95 0.79-1.15 0.60 ce Traldat 217 227 0.96 080-1.14 0.64 Pen years of churvation 54,7046 54,3842 - "of the her 143 face mo ay pets and 2 placebl inde apari hemorag This co bed tecassed wwelable inchad Til Cof included, gends of previo model e (3 the place greep approp bel, k 21 112 11 place of which were speed to be cardiovascatar (7 11 do nad 12 cartela ( 7 pla mpiris There were 169 participants with ulcer in the as pirin group and 138 in the placebo group (relative risk, 1.22; 95 percent confidence interval, 0.98 to 1.53; P=0.08). Among those with ulcer, 38 of the participants taking aspirin experienced some hemor- rhage, as compared with 22 taking placebo (relative risk, 1.77; 95 percent confidence interval, 1.07 to 2.94; P=0.04). With respect to bleeding, 2979 of the aspirin group and 2248 of those taking placebo reported problems such as easy bruising, hematemesis, melena, non- specific gastrointestinal bleeding, epistaxis, or other bleeding (relative risk, 1.32; 95 percent confidence in- terval, 1.25 to 1.40; P<0.00001). Furthermore, 48 in the aspirin group and 28 in the placebo group required transfusion (relative risk, 1.71; 95 percent confidence interval, 1.09 to 2.69; P=0.02). One death from gastrointestinal hemorrhage was reported. This oc- curred in the aspirin group, and the event was con- firmed. DISCUSSION The results just described were virtually identical to the findings presented in our preliminary report." Overall, there was a statistically significant, 44 per- cent reduction in the risk of myocardial infarction that included significant benefits of aspirin for both Fatal and nonfatal events. There continued to be an apparent but not significantly increased risk of stroke -primarily in the subgroup of all hemorrhagic i strokes associated with the use of aspirin. In the subgroup of moderate-to-severe or fatal hemorrhagic stroke, the increased risk that we had observed pre- viously was no longer statistically significant (13) w July 20, 1989 events in the aspirin group and 6 in the placebo group; P0.11). No reduction in the risk of mortality from all cardiovascular causes was associated with aspirin. Our findings regarding stroke and cardiovascular mortality must be viewed in the context of the fact that the trial had too few events to evaluate either of these end points. Even in a trial with our large sample, the ability to identify particular subgroups of participants more or less likely to benefit from aspirin is limited. We ob served no significant modification of the effects of as- pirin on any of the major manifestations of cardiovas- cular disease across the various risk-factor subgroups classified according to cigarette smoking, the history. of diabetes mellitus, parental myocardial infarction, diastolic blood pressure, systolic blood pressure, alco- hol use, amount of exercise, and body-mass index. The reduction in the risk of myocardial infarction as sociated with aspirin was apparent only among those 50 or older. More important, however, was the low absolute risk of myocardial infarction at younger ages, suggesting that any net benefit of aspirin would be greatest among physicians 50 years of age and older. For cholesterol, the beneficial effects of aspirin were apparent at all levels. The observation that the bene- fits of aspirin for myocardial infarction were greatest at low levels of cholesterol was unexpected. This ob- servation may be correct or may reflect random fluctu- ations in the data from which it was derived. It is not possible on the basis of the present data to decide between these explanations. For stroke and cardiovas cular mortality, there was no apparent modification of the effects of aspirin according to risk factors, with the possible exception of cigarette smoking (in relation only to cardiovascular mortality). The number of these end points was too small for us to detect whether overall results were meaningful; the analysis of these end points according to subgroup was therefore par- ticularly difficult. The only other randomized trial of the role of aspi- rin in the primary prevention of cardiovascular dis ease is a smaller study of British doctors. In contrast to the Physicians' Health Study, the British Doctors Trial showed no significant differences for fatal or non- fatal myocardial infarction, but the 95 percent confi- dence intervals were very wide. As in the Physicians Health Study, there were more strokes among those assigned to aspirin, although the difference was not significant, and there was no significant difference in mortality from all cardiovascular causes. There were differences in the design of the U.S. and British trials, including the doses (325 mg on alternate days vs. 500 mg daily), blinding (double-blind with placebo con- trol vs. single-blind), compliance (85.71 percent in the aspirin group and 85.74 percent in the placebo group after 60.2 months vs. 70 percent in the aspirin group and 98 percent in the group without aspirin after 36 months of a 72-month follow-up), and definition of end points. Perhaps most relevant, however, was the difference in the sample size; 5139 subjects were ran domly assigned to aspirin or control in the British 7 THE NEW ENGLAND JOURNAL OF MEDICINE 133 Vol. 321 No. 3 Table 4. Risk of Total Myocardial Infarction Associated with Aspirin Use, According to Level of Coronary Risk Factors. RELATIVE PVALUE OF THENG RELATIV Rox AG Airnew Claras PLACE GROUP of lon 27/4527 (0.6) 51/3725 (1.4) 39/2015 (19) 22/40 C3.0 1.12 0.58 0.46 0.02 0.49 Age (y) 40-49 50-59 60-69 70-84 Cigarette smoking Never Past Current Diabetes mellitus Yes 24/4524 (0.5) 87/3725 (23) 84/2045 (4.1) 44/740 (6 00 96/5488 (1.8) 105/4301 (24) 37/1225 (30) 55/5431 (1.01) 634373 (14) 21/1213 (17) 0.58 0.39 0.57 0.99 cardiovascular disease demonstrat- ed a 25 percent reduction in the in- cidence of subsequent important vascular events (nonfatal myocar- dial infarction, nonfatal stroke, or death from cardiovascular disease), a 32 percent decrease in nonfatal. myocardial infarction, a 27 percent reduction in nonfatal stroke, and a 15 percent reduction in cardiovas- cular mortality. All these reduc- tions were statistically significant. Moreover, aspirin was at least as effective as dipyridamole or sulfin- pyrazone. Most recently, the Sec- ond International Study of Infarct Survival evaluated the role of as pirin in evolving myocardial infarc- tion in a randomized trial with 17,187 participants and demon- strated a 49 percent reduction in nonfatal myocardial infarction, a 46 percent decrease in nonfatal stroke, and a 23 percent reduction in car- diovascular death after five weeks. All these reductions were statisti- cally significant. 11/275 14.00 26/258 (10.1) 0.39 0.60 0.22 128/10.750 (12) 213/10,763 (2.0) 23/1420 (16) 112/9505 (12) 39/1432 (27) 1929481 (20) 0.59 0.58 0.97 No Parental history of myocardial infarction Yes Wan No Cholesterol level (eg per 100 ml) <199 160-209 210-259 2/382 (0.4) 12/1587 (0.8) 26/1435 (1.8) 14/582 (24) 9406 (2.2) 37/1511 (25) 43/1444 (3.01 23/570 (4.0) 0.23 0.29 0.61 0.39 0.04 200 0.21 Diamalic blood pressure (mm Hg) *.69 70-79 80-89 2/583 (0.3) 24/2999 (0.8) 71/5061 (1.4) 26/1017 (25) 9/562 (1.6) 40/3076 (1.3) 1285083 (2.5) 43/970 (4.4) 0.88 0.01 0.53 0.56 90 Systolic blood pressure (mm Hg) 100 0.22 0.52 110-129 130-149 4/296 (1.4) 75/5129 (15) 115/386 (30) 26-412 (6.3) 0.48 0.55 150 Alonbol use Daily 1/330 (0.3) 40/5072 (0.8) 63/7829 (13) 19/454 (4.2) 26/2718 (10) 70/5419 (13) 40/2802 (1.4) 0.65 045 0.61 0.63 55/2727 (2.0) 112/5313 (2.1) 65/2897 (22) Weekly Rarely 0.26 Vigorous exercise at least once a week Yes 91/7910 (12) 45/2997 (1.3) 0.21 No propisim Body-mass $71.00EAT 1407861 (18) 92/3060 (3.05 41/2807 (15) 46/2627 (1.8) 75/2823 (2.7) 76/2776 (27) 26/2872 (0.9) 32/2700 (12) 32/2713 (12) 49/2750 (18) "To nel value to milles per er, aply by 0 Thody-index is the weight in kilogramo mes the bightsand 0.65 0.49 0.61 0.68 0.44 0.65 23.0127-24.4075 24.4076-26.3865 263866 0.90 trial, and 22,071 in the U.S. study. To minimize the effect of this difference in sample size, we undertook an overview of the two trials of primary prevention." Since the U.S. trial was so much larger, this over- view showed a significant, 33 percent reduction in nonfatal myocardial infarction associated with aspirin (P<0.0002). A repeated analysis, in which data from this final report were used, gave virtually identical results. Aspirin's benefits in reducing the incidence of subsequent myocardi- al infarction have been shown con- clusively in the survivors of myo- cardial infarction and stroke and in patients with unstable angina, as well as in those with an evolving heart attack. Our trial demon- strates conclusively a benefit of as- pirin in reducing the incidence of first myocardial infarction and thus extends the previous findings to healthy people. Benefits in reducing the incidence of subsequent stroke have been shown conclusively in survivors of myocardial infarction and stroke and in patients with unstable angina, as well as in those with an evolving heart attack. The findings of our trial of primary prevention, although not statistically significant, are compatible with an increase in the number of all strokes among aspirin users. It seems important to distinguish between ischemic stroke, in which one might expect aspirin to be of benefit, and hemorrhagic stroke, in which one might expect an ad- verse effect. The possibility of an increase in the inci- dence of hemorrhagic stroke among aspirin users is not unexpected, since any agent that decreases clot- ting may help prevent ischemic events but increase bleeding. To evaluate this matter further, future trials in which the sample size is adequate to distinguish between ischemic and hemorrhagic stroke are re- quired. The findings of this study of aspirin in the primary prevention of cardiovascular disease should be viewed in the context of all the evidence concerning the possi- ble role of aspirin. In 1985 the Food and Drug Admin- istration approved the labeling of aspirin as an agent to be prescribed for the treatment of patients with a previous myocardial infarction or unstable angina." A recent overview of 25 randomized trials of antiplate- let therapy (aspirin, dipyridamole, or sulfinpyrazone, alone or in combination) in patients with a history of Finally, aspirin's benefits in reducing the incidence of cardiovascular mortality have been shown con- ASPIRIN COMPONENT OF PHYSICIANS HEALTH STUDY-HENNEKENS ET AL
O II was no longer statistically significant ( 6 THE NEW ENGLAND JOURNAL OF MEDICINE ET 0 29 M PHYSICIANS' HEALTH STUDY-HENNEKENS ET AL. Vel. 321 N the drug should ided important cost. With the In addition, the between plants and blood vessel walls. N Engl J Med 1979, 3001142 spirin trial bar results. Br Mod 1 1988, 26.516-25 C RA 5 3 ♡ de pes the operating penticol of prophylactic deity 18921-4 is, bet, receved from com ENG 9:06 PM I