Effect of ramipril on mortality and morbidity of survivors of acute myocardial infarction with clinical evidence of heart failure

Articles

Effect of ramipril acute

myocardial

on

mortality and morbidity of survivors of

infarction with clinical evidence of

heart failure The Acute Infarction

Ramipril Efficacy (AIRE) Study Investigators

Summary

Introduction

Survival after acute myocardial infarction has been enhanced by treatment with thrombolytic agents, aspirin, and &bgr;-adrenoceptor blockade. However, there remains a substantial subgroup of patients who manifest clinical evidence of heart failure despite the first two of these treatments, and for whom &bgr;-adrenoceptor antagonists are relatively or absolutely contraindicated. These patients have a greatly increased risk of fatal and non-fatal ischaemic, arrhythmic, and haemodynamic events. In this selected high-risk subset of patients we investigated the effect of therapy with the angiotensin converting enzyme (ACE) inhibitor ramipril, postulating that it would lengthen survival. 2006 patients who had shown clinical evidence of heart failure at any time after an acute myocardial infarction (AMI) were recruited from 144 centres in 14 countries. Patients were randomly allocated to double-blind treatment with either placebo (992 patients) or ramipril (1014 patients) on day 3 to day 10 after AMI (day 1). Patients with severe heart failure resistant to conventional therapy, in whom the attending physician considered the use of an ACE inhibitor to be mandatory, were excluded. Follow-up was continued for a minimum of 6 months and an average of 15 months. On intention-to-treat analysis mortality from all causes was significantly lower for patients randomised to receive ramipril (170 deaths; 17%) than for those randomised to receive placebo (222 deaths; 23%). The observed risk reduction was 27% (95% Cl 11% to 40%; p=0·002). Analysis of prespecified secondary outcomes revealed a risk reduction of 19% for the first validated outcome (ie, first event in an individual patient)—namely, death, severe/resistant heart failure, myocardial infarction, or stroke (95% CI 5% to 31%;

Patients with clinical evidence of heart failure after myocardial infarction have a poor outcome even if the manifestations of failure resolve within the first 24 hours. 1-3 The AIRE Study was designed to determine whether administration of the angiotensin converting enzyme (ACE) inhibitor, ramipril, to these patients would improve survival. After acute myocardial infarction (AMI) the degree of ventricular damage, the extent of myocardium at further ischaemic risk, and an arrhythmic tendency contribute substantially to the risk of subsequent fatal and non-fatal cardiac events.5 Because of this heterogeneity, therapy beneficial to one aspect may worsen another and thereby produce a neutral or negative overall effect on survival.6-8 The findings of the second CONSENSUS Study,8 in which patients were given the ACE inhibitor enalapril after acute myocardial infarction, are particularly relevant in this regard and in the context of the AIRE

p=0·008). Oral administration of ramipril to patients with clinical evidence of either transient or ongoing heart failure, initiated between the second and ninth day after myocardial infarction, resulted in a substantial reduction in premature death from all causes. This benefit was apparent as early as 30 days and was consistent across a range of subgroups. Lancet 1993; 342: 821-28

Members of the AIRE Study Group and listed at the end of the article

participants In the study are

Correspondence to: Prof S G Ball, Academic Unit of Cardiovascular Studies, University of Leeds, Leeds LS2 9JT, UK

Study. At present diuretics are used extensively in the treatment of patients with cardiac failure after AMI and their efficacy in acute heart failure is well established. However, in the long-term there is a tendency for diuretics to increase afterload and deplete electrolytes, and in many patients

ventricular function continues to deteriorate.9 After AMI, particularly where thombolysis has failed and there has been a full thickness infarction, the infarcted area thins and elongates-the so-called infarct expansion.1O The subsequent, and for many patients presumed adverse, "remodelling" of the heart, in which the ventricle enlarges and becomes distorted with hypertrophy of the remaining viable myocardium, continues despite treatment with diuretic therapy." ACE inhibitors reduce afterload, preserve electrolytes, and reduce the ventricular enlargement that ensues in some patients after infarction.9,11 Several major trials indicate benefit from this group of drugs in patients with symptomatic chronic heart failure not adequately controlled with digoxin and diuretic, and in those likely to develop infarct expansion and ventricular dilation after AMI, but do not address the patient population studied here.12-15 At the inception of the AIRE Study we postulated that, after myocardial infarction, patients with a high risk of premature death (denoted by clinically observed signs of either transient or ongoing heart failure) would benefit from receiving appropriately timed oral ramipril.4,16 There was no restriction on the use of concomitant therapies, but we expected most of the patients to have received diuretic or a nitrate and few to be taking a P-adrenoceptor antagonist 821

(though such treatment would not be absolutely contraindicated). The first CONSENSUS trial had demonstrated particular benefit from ACE inhibitor therapy in patients with severe heart failure (New York Heart Association classification grade IV).12 However, those with recent myocardial infarction or unstable angina were excluded. We too were concerned about potential adverse effects of ACE inhibition, particularly when used early after acute myocardial infarction.16 Therefore only patients who were haemodynamically stable and without overt evidence of ongoing ischaemia were randomised, and not before the second day after myocardial infarction. Patients with severe cardiac failure (NYHA classification grade IV) were excluded from the study. Ramipril has high affinity for, and binds tightly to, converting enzyme in both circulation and tissuesp,18 It also has strong bradykinin potentiating effects. 19 The AIRE Study tests the hypothesis that patients with acute myocardial infarction complicated by clinical evidence of heart failure will live longer if they receive long-term ramipril treatment, initiated between the second and ninth days after infarction. Randomisation between ramipril and matching placebo was arranged in blocks of 10 patients, and was stratified by centre. The prespecified primary statistical endpoint was all-case ("intention-to-treat"), all-cause mortality. Methods The rationale, design, organisation, and outcome definitions of the AIRE Study were described prospectively.4 In brief it was a

multicentre, multinational, double-blind, randomised, placebocontrolled study. 2006 patients with acute myocardial infarction and clinical evidence of heart failure were recruited in 144 centres in 14 countries. All patients aged at least 18 years admitted to coronary care, intensive care, or general medical units with a definite AMI and clinical evidence of heart failure at any time after the index AMI (usually sufficient to justify at least short-term diuretic or vasodilator treatment), were eligible. AMI was defined as an evolving electrocardiogram (ECG) diagnostic of myocardial infarction (ie, progressive changes in the ST segment and T wave compatible with AMI with or without the presence of pathological Q waves) and abnormal cardiac enzymes (ie, elevation of creatine phosphokinase and/or aspartate aminotransferase and/or lactic dehydrogenase to greater than twice the upper limit of the laboratory reference range or equivalent increases in their isoenzymes). Clinical evidence of heart failure was defined as at least one of the following: evidence of left ventricular failure (pulmonary venous congestion with interstitial or alveolar oedema on at least one chest radiograph); evidence of pulmonary oedema (bilateral post-tussive crackles extending at least one-third of the way up the lung fields in the absence of chronic pulmonary disease); or auscultatory evidence of a third heart sound with persistent tachycardia. Although clinical evidence of heart failure following the index AMI was mandatory for trial entry, this could be transient and not necessarily present at the time of randomisation. Patients with severe heart failure (usually NYHA grade IV), heart failure of primary valvular or congenital aetiology, unstable angina, or any of the recognised contraindications to ACE inhibitor treatment were excluded from the study. All participating centres were asked to complete a simple screening form for each patient admitted with a suspected or definite AMI, to describe the population from which the eventual study group was selected. The strategy was then to select a group of patients after myocardial infarction thought to have a particularly poor prognosis and in the majority of whom, because of evidence of left ventricular dysfunction, treatment with a diuretic or vasodilator drug would have been instituted though not necessarily continued. Most

822

Table 1: Patient characteristics, trial therapy, and concomitant medication at baseline

clinicians

would hesitate to prescribe a cardioprotective p-adrenoceptor antagonist to patients in this group. However a spectrum was envisaged. At one end would be patients with no evidence of impaired ventricular function, who would receive a p-adrenoceptor antagonist, and at the other those with severe

failure treated with diuretics and an open ACE inhibitor. Clinical judgment would be used to select patients between these extremes to be randomised to ACE inhibitor or placebo. Clinicians were encouraged at all times to continue their normal clinical practice in terms of overall management and drug therapy.

Initiation of treatment Treatment was initiated in hospital between day 3 and day 10 after AMI (day 1 = day of index infarction). Patients initially received 2-5mg of ramipril or placebo twice daily. Those who tolerated this dose received it for 2 days and then were given 5 mg of ramipril or placebo twice daily thereafter. Patients who could not tolerate the

Summary of primary endpoints, validated secondary events, and secondary endpoints

Table 2:

Figure 1: Mortality curves Illustrating the primary endpoint of all-cause mortality analysed by intention-to-treat Most patients were followed for less than 18 months, and the curves have been terminated at 30 months because of the small numbers of patients with prolonged follow-up.

higher 5 mg dose were discharged on 2mg or placebo twice daily. Ramipril 1-25 mg or placebo was provided for those patients who could not tolerate the initial 2-5 mg dose. These patients began again on the lower dose 1-25 mg twice daily for 2 days before increasing to 2-5 mg twice daily and then 5 mg twice daily. When therapy was started or dosage was changed, blood pressure was monitored before and at 2, 4, and 6 hours, all adverse events being recorded. If a patient was unable to tolerate ramipril at least 2-5 mg twice daily or matching placebo he or she was withdrawn from the study treatment but followed at the prescribed visit intervals for intention-to-treat analysis. The protocol did not allow discharge of a patient on the low dose of 1-25 mg twice daily.

Follow-up Patient follow-up was designed to minimise any interference with usual clinical practice. All patients, including those withdrawn from randomised treatment, were seen at 4 and 12 weeks after randomisation and thereafter every 12 weeks until study close. At each visit, the occurrence of a study outcome, other adverse events, compliance, and concomitant therapy were recorded. Renal function (serum electrolytes, creatinine, and urea) was reviewed in accordance with the investigator’s normal clinical practice. Patients could continue or begin any other necessary treatment except an ACE inhibitor while on randomised treatment. Monitoring of serum potassium was strongly advised, particularly if potassium-sparing diuretics or potassium supplements were

judged necessary. The last day in the analysis of the mortality data was Feb 28, 1993, six months after the 2000th patient had been recruited. As soon as possible after this date, the final status of all patients was

assessed.

Sample size On the following assumptions the trial was estimated to require about 2000 patients: predicted average patient follow-up 15 months; predicted placebo mortality 20% at 15 months; a "clinically relevant improvement" defined as a 25% reduction in all-cause mortality, resulting in an expected mortality of 15 % in the active treatment group at 15 months; and statistical power of at least 80% at a significance level of 5% (two-tailed test, log rank test).4

Study organisation An independent adjudicating panel (IAP) acted as the overall ethical supervisory body and had access to the randomisation code. The IAP performed the interim analysis. The IAP was also responsible for transmitting data on serious adverse events to the relevant regulatory authorities. An international steering

committee met regularly to review progress and was responsible, inter alia, for the clinical definition of the secondary endpoints. The executive committee, chaired by the principal investigator (SGB) and including representatives of the sponsor, the study managers, and the data manager, was responsible for the day-today decisions on the conduct of the study and the operation of the AIRE Study Co-ordinating Centre. All endpoints were validated by a subcommittee of the international steering committee. An independent group was responsible for conducting a series of prospective audits of study procedures, to ensure that the study conduct adhered closely to the European Guidelines for Good Clinical Research Practice.

Statistical methods The primary endpoint was all-cause mortality, analysed by intention-to-treat. Survival curves were obtained by use of the Kaplan-Meier estimate, and the log rank test was used for comparison of the two groups (ramipril and placebo). The relative hazard and corresponding 95% confidence interval (CI) was obtained by fitting a Cox proportional hazards regression model with a single covariate-namely, the allocated treatment. Subgroup effects were examined by including interaction terms in the Cox proportional hazards regression model. The secondary endpoint was the time to the first validated secondary event-namely, death, progression to severe resistant heart failure, reinfarction, or stroke. These data were analysed in the same way as those for the primary endpoint. A planned interim analysis was performed after 175 deaths, half way through the projected total of 350 deaths. A very stringent stopping rule was specified in the protocol (p < 0-001 was the guideline for early termination) so that no adjustment was required to the final analysis to compensate for this interim analysis. In addition, during the early stages, the independent adjudicating panel requested a safety analysis after a total of 40 deaths had been reported. This analysis could not have led to early termination of the study with a claim of efficacy, and so again no adjustment was made to the final analysis. All statistical endpoints were reviewed and validated by a subcommittee of the international steering committee and the procedures used are being reported elsewhere .20

Patient population The study began in Leeds, and in its first year (1989) two amendments were made to produce the protocol under which the vast majority of patients were randomised. The inclusion criteria were extended to include patients with non-Q as well as Q-wave infarction and the protocol was clarified to emphasise that patients with transient cardiac failure after myocardial infarction were eligible to be randomised. The study was extended to include numerous other centres. Most of the patients were recruited between April, 1991, and the close of recruitment on Aug 27, 1992. A total of 52 019 patients had been reviewed. The main reason for exclusion was no firm evidence of myocardial infarction in 21 302

823

Figure 2: Curves illustrating proportion of surviving patients who had withdrawn from study medication during follow-up Patients withdrawn from study medication continued to be followed for inclusion in the intention-to-treat analysis.

and no evidence of cardiac failure in 16 989. Of the remaining 13 728 screened patients, exclusions in order of number were: died before possible inclusion 2784, unstable angina 2314, no consent or follow-up impracticable 1962, on or about to start an ACE inhibitor 1404, severe heart failure 1329, renal failure 289, sustained hypotension 279, and "other reasons" 1361. We estimate that 15-20% of the patients with definite myocardial infarction had evidence of transient or ongoing cardiac failure and were eligible for the trial, of whom one-half were randomised.

Exclusion of one centre of the study it became apparent that data from inconsistent. After discussion with the members of the independent adjucating panel it was agreed to remove the 20 patients apparently recruited and randomised at this centre. This was done before study close. Analyses that included the information available on these patients had no important qualitative impact on the description of the patient population or outcome of the study. The data presented are those of the

During the one

course

centre were

remaining

1986

patients.

Heiative hazara

Figure 3: Relative hazards with corresponding 95% Cis, illustrating consistent benefit with ramipril over a wide range of subgroups In no case level.

Results

was

the test for interaction statistically

significant at the 5%

Baseline

demographic data patients were randomised to ramipril and 982 to placebo; randomisation to drug or placebo was well 1004

balanced within the 14 countries. Table 1 shows the demographic data of the patients and the relevant conditions for which they were receiving treatment before the index infarction. Some 22-6% of patients had received treatment for a previous myocardial infarction but only 8-2% had a history of previous heart failure. The mean time to randomisation was 54 (SD 2-1) days after AMI for ramipril and 5-4 (SD 2-2) for placebo. The groups were well matched in all aspects at baseline (table 1). Overall, 58% of patients received thrombolytic treatment. Concomitant medication was similar in the two groups. Trial therapy More than 90% of patients in each treatment group were discharged from hospital on study medication. A greater percentage of patients receiving placebo achieved the "higher dose level" (table 1).

All-cause

mortality Figure 1 shows the findings for all-cause mortality. Separation of the curves occurred early and they continued to diverge throughout the study. There were 170 deaths (17%) in the ramipril group and 222 (23%) in the placebo group, with overall a 27% reduction in the risk of death (95% CI 11% to 40%), which was highly significant statistically (p 0-002). =

Secondary endpoints Table 2 shows all-cause mortality (primary endpoint) and secondary events validated by the outcomes subcommittee.2° For the formal analysis we used only the findings for the first validated event in any individual

patient-namely, death, reinfarction, stroke, or development of severe/resistant heart failure (table 2). Again the reduction, 19% (95% CI5% to 31 %), was highly significant statistically (p 0-008). =

,

Follow-up The average time of follow-up was 15 months with a minimum of 6 months. Only 1 patient was lost to follow-up-last seen 12 weeks after randomisation, at which time the data for this patient were censored. 824

Withdrawal from study medication The withdrawal rates from study drug, excluding patients who had died, are shown in figure 2. In total there were 352 premature withdrawals from the ramipril group and 318 from the placebo group. Intolerance was given as the

primary

or a

contributory factor

in 126 of the

ramipril

withdrawals and in 68 of the placebo withdrawals, whereas

progression to severe/resistant heart failure was the stated reason for 58 ramipril withdrawals and 92 placebo withdrawals. Serious adverse events There were fewer patients with reported serious adverse events on ramipril, 581 (58%), than placebo, 625 (64%). Serious adverse events included the endpoints of the trial

(death, progression to severe/resistant heart failure, reinfarction, and stroke) as well as possible adverse effects of treatment. Syncope was reported for 24 (2-4%) patients on ramipril and 17 (1-7%) on placebo with a similarly increased occurrence of hypotension in the ramipril treated group-42 (4-2%) compared with placebo 23 (2-3%). Renal failure occurred with a similar frequency in the two groups: 15 (15%) on the drug and 12 (1-2%) on placebo. Angina, which it was thought might be worsened in some patients prescribed an ACE inhibitor, was reported as a serious adverse event in 181 patients (18%) taking ramipril and 171 (17%) taking the placebo. Interactions-effect of age, sex, delay to treatment, and

adjuvant therapy Several interactions were examined (figure 3). All the estimated relative hazards were below 1 and in no case was the test for interaction significant at the 5% level.

Discussion

Ramipril, administered to patients with clinical evidence of heart failure on the second to ninth days after myocardial infarction for an average of 15 months, caused a highly significant and substantial reduction in all-cause mortality. The mortality curves for ramipril and placebo were clearly separating within weeks of starting treatment and continued to diverge. There was a significant, though smaller, benefit from drug on the secondary endpointnamely, time to occurrence of the first validated outcome of death, progression to severe/resistant heart failure, reinfarction, or stroke. These positive findings based on intention-to-treat analysis are even more striking when we note that by 18 months 40% of patients (excluding those who had already died) were no longer taking randomised medication and that most of the patients withdrawn from placebo were taking open-label ACE inhibitor. Ramipril was well tolerated and more than 90% of the patients in the ACE-inhibitor group were taking it when they left hospital. Importantly there was no pre-trial run-in phase to exclude patients who could not tolerate the drug; this strategy gives a simple appraisal of the benefit of treatment with ramipril, increasing the relevance of the findings to normal clinical practice. Subsequent withdrawal rate over the first 2 years was only marginally higher with ramipril than with placebo (withdrawals caused by minor adverse effects of the active agent were offset by the lower incidence of heart failure necessitating open label treatment). The findings of this study should be compared with those of the two other published mortality trials undertaken with ACE inhibitors in patients after myocardial infarction. In the CONSENSUS II Study8 all patients with presumed myocardial infarction were randomised within 24 hours to treatment with intravenous enalaprilat or placebo followed by oral treatment with enalapril or placebo. The study was

stopped by the ethical review committee after recruiting just over 6000 of the intended 9000 patients. At this time more patients had died on drug than on placebo though the difference was not statistically significant. In contrast, the SAVE Study15.21 included 2231 patients 3-16 days (average 11 days) after myocardial infarction. These patients were carefully selected in that they were symptomless at randomisation, had a radionuclide ejection fraction less than or equal to 40%, and had undergone an exercise test to exclude ischaemia. Only 33% had been given a thrombolytic, but almost 60% had coronary angiography and 25% had a revascularisation procedure before entry into the study. All patients had a dose of captopril prerandomisation and only those who tolerated this were entered into the study. Follow-up was for an average of 42 months and mortality was reduced by 19%. However, no effect on mortality was apparent until almost 1 year after the start of treatment. This contrasts strikingly with the observations reported here, with benefit apparent early and after a much shorter follow-up. ACE inhibitors have become established treatment, when added to diuretic, in chronic congestive cardiac failure. Two major trials, V-HeFT 11 and the treatment and prevention arms of SOLVD,13,22 investigating the effect of this group of drugs on mortality in patients with chronic heart failure, were published during the course of the AIRE Study. In both the entry criteria were based on a measurement of ejection fraction and both excluded patients with recent myocardial infarction (within the previous one or three months and the great majority of patients were entered much later). The V-HeFT IIStudy’4 required evidence that exercise capacity was limited by fatigue/breathlessness but not by chest pain, again like SAVE excluding overt ongoing ischaemic problems. In the SOLVD treatment and prevention studies patients were given enalapril for up to 1 week then placebo for 2 weeks and only those patients tolerating the drug and its withdrawal were randomised.13,22 V-HeFT 1114 and the SOLVD13 treatment arm showed that enalapril reduced mortality in patients with symptomatic heart failure. The SOLVD prevention arm did not reveal a statistically significant reduction in mortality.22 Whilst ejection fractions in patients entering the SOLVD treatment and prevention arms differed by only 3%, mortality in the symptom-free prevention group was less than half that of the symptomatic patients in the placebo group of the treatment arm.13.22 AIRE Study investigators were informed of the findings of the SOLVD treatment arm and those of the SAVE Study since it was thought that patients selected on the criteria of these trials would benefit from treatment with an ACE inhibitor. The AIRE Study was designed to take account of changing clinical practice. Patients with severe heart failure (NYHA grade IV) were known to have a particularly poor prognosis at the beginning of the study and although the first CONSENSUS Study had excluded patients less than 2 months post myocardial infarction (and for the majority probably much longer), this group with severe heart failure was specifically excluded since it was felt that most physicians would wish to consider prescribing ACE inhibitors to these patients.4.12 However, the difficulty of defining different grades of heart failure was recognised and clinicians were able to decide individually what they regarded as sufficiently severe heart failure unresponsive to conventional therapy to justify open prescription of this group of drugs rather than to enter such patients into the 825

trial. Similarly, the subsequent progression to severe/ resistant heart failure, usually NYHA grade IV, but again at the discretion of the physician, necessitating prescription of open ACE inhibitor, was used as a secondary endpoint in this study. During the course of the study, increasing numbers of patients with moderate heart failure were treated with open label ACE inhibitor in preference to

continuing study drug. The early effect within weeks

was very similar to that observed in the first CONSENSUS Study of patients with severe failure. An analysis at 30 days, prespecified before the randomisation code was broken, showed that the difference between drug and placebo was close to significance (p=0-053). Furthermore when a second analysis was done on mortality only between 30 days and study close, thus biasing the findings considerably against the drug, a further substantial benefit was apparent

(p=0016). In a recent further analysis of the SOLVD data23 and in the SAVE Study15 administration of ACE inhibitor was claimed to reduced myocardial infarction rates. Whilst there was a trend to fewer such events on ramipril than on placebo in the AIRE Study no clear reduction was apparent. However, the number of validated reinfarctions was small, largely because of the much shorter average follow-up period than those in SOLVD and SAVE. The favourable trend observed here would be in keeping with the observations made after a similar period of follow-up in these studies. The incidence of stroke was higher in the active drug group but the numbers were small and an adverse effect of the drug can be neither supported nor refuted. Fewer patients progressed to severe/resistant heart failure in the active drug group and there were fewer admissions to hospital (data not shown), as would be expected from the mode of action of ramipril and the findings of other trials; but hospital admission was not specified as part of the primary or secondary analyses. Both sudden death, presumed arrhythmic, and progression to severe failure were reduced by administration of ramipril. The classification of death in heart failure is complex and the approach used in the AIRE Study will be detailed elsewhere.2O The findings on mode of death and adverse events including stroke and myocardial infarction will also be described in further publications. 40% of patients were not taking diuretic therapy at randomisation, indicating only short lived evidence of heart failure in a substantial proportion. Whilst the subgroup analysis (figure 3) showed remarkable consistency with all relative hazards below 1, patients not taking diuretics at randomisation may have derived less benefit, though it must be emphasised that the difference was not statistically significant. A similar trend was seen for patients under 65 years of age. Benefit accrued irrespective of administration of a thrombolytic agent. The proportion of patients receiving such treatment was similar to that in the CONSENSUS II Study8 and in excess of that in the SAVE tria1.15 It is likely, however, that the inclusion criteria for both the AIRE and SAVE studies will have produced a bias towards the selection of patients in whom attempts at reperfusion were unsuccessful. Almost 80% of patients were taking aspirin at the time of randomisation. Experimental evidence indicates that ACE inhibitors and aspirin alter prostaglandin metabolism. Interaction between the effects of the two agents has been reported in severe heart failure.24 In AIRE we saw a clear benefit with ramipril in aspirin-treated patients; but there was a trend 826

towards

an even

greater benefit in those

not

receiving

aspirin. ramipril was additive to that of P-adrenoceptor antagonists. Most studies with (3-blockers, which in general reduce morbidity and mortality after myocardial infarction, have specifically excluded patients The

action

of

with clinical evidence of left ventricular failure.25 One notable exception was the Beta-blocker Heart Attack Trial, which included patients with either mild or transient clinical signs and symptoms of heart failure.26 A retrospective analysis of this subgroup of patients indicated that propranolol had a beneficial effect on

survival, though mortality rates remained high.26 Ramipril offers clear benefit

complementary to current conventional

treatment.

Patient screening data and extrapolation from many other studies both before1 and after the widespread use of thrombolytic agents27-29 indicate that as many as 20% of patients with myocardial infarction might fall into this group. Ramipril brought about a striking reduction in mortality and this seemed independent of treatment with thrombolysis, 0-adrenoceptor antagonists, and oral nitrates. Furthermore and importantly, women3o benefited equally to men. Clinicians were able to identify by simple everyday measures a group who can expect substantial benefit from treatment with the ACE inhibitor

ramipril. The AIRE Study group thank Hoechst for their support. Prof S G Ball is British Heart Foundation Chairholder and thanks colleagues in Leeds (Dr A S Hall, Dr A F Mackintosh, Dr C Winter, Dr L B Tan, Dr J C Cowan, Dr G W Reynolds) and representatives of the sponsor (Dr M A Akbary, Dr N Bender, Dr B Rangoonwala, and Prof P Stonier) for support throughout the study. a

AIRE Study administration

Independent adjudicating panel D G Julian (UK, chairman); A J Moss (USA); G D Murray (UK); P A Poole-Wilson (UK); M L Simoons (Netherlands). International steering committee S G Ball (chairman, principal investigator); N Bender (sponsor); J G F Cleland (UK); D L Clement (Belgium); P J Commerford (South Africa); S Dalla Volta (Italy); C Delagardelle, (Luxembourg); L Erhardt (Sweden); W Fennell (Ireland); A S Hall (advisor to principal investigator); R Hopf (Germany); S Jagerholm (Finland); M Jolly (coordinating centre); W Klein (Austria); G D Murray (independent statistician); M G Niemeyer (Netherlands); S Rasmussen (Denmark); M A Riccitelli (Argentina); L Richardson (coordinating centre); P D Stonier (sponsor); H-P Vogelin (Switzerland); C Winter (independent safety coordinator). Outcome subcommittee J G F Cleland (UK), L Erhardt (Sweden). Authorship subcommittee S G Ball, D G Julian, G D Murray. Coordinating personnel M A Akbary (sponsor), J Clinch, M Jolly, N Raw, L Richardson, H Twigg, P J Williamson. Data centre G D Murray, A Lawrence, J Love, T Welby, C A Woods. Parttapants Countries and centres listed in order of number of patients recruited; principal investigators denoted by *. UK: General Infirmary at Leeds (S G Ball*, J C Cowan, C Winter); St James’s University Hospital, Leeds (A F MacKintosh*, L B Tan); Royal Hallamshire Hospital, Sheffield (L Caldicott, K Channer*, S K Hawley, A Wilson); Oldchurch Hospital, Romford (I Dews, H Kadr, J Stanton, C Statuch, J D Stephens*); Wharfedale General Hospital, Otley (K E Berkin*); Good Hope District General Hospital, Sutton Coldfield (S Davies, M V J Raj*); Stracathro Hospital, Brechin (T S Callaghan*); Mayday Hospital, Croydon (S Joseph*, A C Rao); Walton Hospital, Liverpool (D Hambleton, R S Hornung*, E Rodrigues*); Royal South Hants Hospital, Southampton, and Southampton General Hospital (H Roberts, D Waller*, N Warner); Sunderland District Hospital (S E Pugh*, N Ramaknshnan, M Sekar); Wexham Park Hospital, Slough (R A Blackwood*, P Robinson); Cumberland Infirmary, Carlisle (M Clark, R H Robson*); General Hospital, Hartlepool (P T Pickens, G Tildesley*, R W B White); Pinderfields General Hospital, Wakefield (P Kelly, J I Wilson*); Whittington Hospital, London (S Jones, C Kingdom, D L H Patterson*); Wycombe General Hospital, High Wycombe (A Bentley, S. Gupta, W G Hendry*); Royal Sussex County Hospital, Brighton (C Davidson*, P Glennon); Ashford Hospital (A Bishop, D Kluth, A Mooney, P Wilkinson*); Pontefract General Infirmary (T P Anthony, A K Ghosh*); Greenwich District Hospital (D Robson*); Erne Hospital, Enniskillen (B McAleer, M P S Varma*); Harrogate District Hospital (H Larkin*); Hillingdon Hospital (S Kaddoura, C Knight, G Sutton*); Scarborough Hospital (R S Clark*); Walsgrave Hospital, Coventry (M Been*, A Raman); Royal Lancaster Infirmary (A K Brown*); St Mary’s Hospital, London (R Foale*, J Shahi); Hinchingbrooke Hospital, Huntingdon (C Borland*); Friarage Hospital, Northallerton (T G Jayasuriya, U Somasundram*).

Sweden: Hoglandssjukhuset, Eksjö (S Ekdahl*, S Hansen); UniversitetssJukhuset, Lmkoping (M Broqvist, U Dahlstrom*, 0 Kongstad); Allmanna Sjukhuset, Malmo (C Cline, L Erhardt*, R Willenheimer); Huddinge Sjukhus (H Berglund, 0 Nyquist*); Sjukhuset, Angelholm (P Nyman, D Ursing*); Lanslasarettet, Karlshamn (S Jensen*); Landskrona Lasarett (F Gyland, R Linne, T Ragnartz*); Sjukhuset, Kristinehamn (R Watz*); Centrallasarettet, Karlskrona (M Freitag, L Olsson*); Simrishamns Sjukhus (J Hallgren*, P Lindvall).

James’ Hospttal, Dublin (C Bergin, P Crean*, N El-Guylam, B McAdams, Sheahan); Sligo General Hospital (J McKenna, D Murray*); Mallow General Hospital (P Sullivan*); Cork Regional Hospital (W Fennell*, P Kearney, C Vaughan); Mater Misericordiae Hospital, Dublin (J Galvin, D Sugrue*); Limerick Regional Hospital (T Peirce*); Louth County Hospital, Dundalk (N Ghaisas, E Harkin, T O’Callaghan*); County Hospital, Roscommon (P McHugh*, S Nisar); University College Hospital, Galway (K Daly*, F Lavin, P Shah); Our Lady’s Hospital, Navan (R Juneja, K McGarry*); Monaghan General Hospital (S Junejo, B MacMahon*, N P Pillay*); Our Lady of Lourdes Hospital, Drogheda (D Long, C Maguire, B C Muldoon*, P Shiels); Letterkenny General Hospital (L Bannan*, E Connelly); Portiuncula Hospital, Ballinasloe (J Barton*); St Vincents, Dublin (C McCreery, P Quigley*). Ireland: St

D Morais, R

South 4/nca; Addington Hospital, Durban (S Cassim, D P Naidoo*); H F Verwoerd Hospital, Pretoria (D P Myburgh*); Universitas Hospital, Bloemfontein (J D Marx*); Groote Schuur Hospital, Cape Town (P J Commerford*, J Lawrenson); Hydromed Hospital, Bloemfontein (H Du T Theron*); R K Khan Hospital, Chatsworth (M C Rajput*); Umtas Hospital, Verwoerdburg (J M Bennett*); Baragwanath Hospital, Soweto (P Sareli*).

Argentina: Hospital Argerich, Buenos Aires (P Arce, M A Riccitelli*); Hospital Santojanni, Buenos Aires (D Ryba*); Hospital Carlos Durand, Buenos Aires (E Beck, A Demartini*); Hospital Pirovano, Buenos Aires (N Goncalves Borrega,J Lazzari*); Sanatorio Mitre, Buenos Aires (M Gonzalez, A Sosa Liprandi*); Hospital de Clinicas Jose de San Martin, Buenos Aires (J Martinez Martinez, E A Sampo*); Hospital Espaflol, Buenos Aires (G Bortman*, H Grancelli); Hospital Ramos Mejia, Buenos Aires (J Carbajales, A Girotti*); Hospital Juan A Fernandez, Buenos Aires (B Maumer*, 0 Gabrielli); Hospital Italiano, Buenos Aires (A Cagide*); Hospital Interzonal de Agudos Eva Peron, Buenos Aires (A Lapuente*); Policlinico Bancario, Buenos Aires (R Esper*). Netherlands: Martini Ziekenhuis, Gromngen (L J Takens*, M Y van der Heijden); De Wever Ziekenhuis, Heerlen (J A Kragten*, R Renkens); Het Nieuwe Spittaal, Zutphen (A C Tans*); Spaarne Ziekenhuis, Heemstede (MJ W Grosfeld, H H Kruyswijk*); St Streekziekenhuis Hilversum (A de Groot, K L Liem*); Elkerliek Ziekenhuis, Helmond (H Olthofk); Canisius-Wilhelmina Ziekenhuis, Nijmegen (H E Haerkens-Arends, T E H Hooghoudt*); Merwede Ziekenhuis, Sliedrecht en Dordrecht (A Kuypers, M G Niemeyer*); St Ziekenhuis Amstelveen (J P van Mantgem*); St Sint Joseph Ziekenhuis, Veldhoven (E Aelfers, L C Slegers*); St Clara Ziekenhuis, Rotterdam (M G Scheffer*); Martini Ziekenhuis, Groningen (B Oedit Doebe, L E J M Schrijvers*); St Sint Joseph Ziekenhuis, Veghel (L Relik van Wely*); Drechtsteden Ziekenhuis, Zwijndrecht (A H Herweijer*).

Belgium: Hopital Braine-I’Alleud (M DeMyttenaere*); Sint-Annaziekenhuis, SintTruiden (P Peerenboom*); A Z Sint-Jozef, Turnhout (I Bekaert*); Onze Lieve Vrouw Middelares Ziekenhuis, Deume (P Surmont*); Heilig Hartziekenhuis, Neerpelt (A Van Dorpe*); Onze Lieve Vrouwziekenhuis, Mechelen (J Beys*); Heilig Hartziekenhuis, Roeselare (D Clement*); Clinique des Deux Alice, Brussels (E Benit*, M Beyloos, C Durieux); Kliniek Zwarte Zusters, leper (M Bayart*); Sint-Jozefkliniek, Bornem (M Herssens*); Hopital de Samte Therese, Bastogne (S Cremers*); A Z Sint-Norbertus, Duffel (F De Keyser, U Van Walleghem*). Germany: Krankenhaus Sachsenhausen, Frankfurt (R Hopf*, A Moller); Klinik am Eichert, Goppingen (J Hauber, F Hofgartner, H A Sigel*); Dreieich-Krankenhaus, Langen (M Grieshaber, M Neubauer*, K Rudolph); Evangelisches Krankenhaus, Herne (W Sehnert*); Albert Schweitzer Krankenhaus, Northeim (R Cartsburg, P Kleine*); Kreiskrankenhaus, Wurselen (T H Hennecke,J Kindler*, P Ramme); St Josefs Hospital, Cloppenburg (D Hellmann, D A Hemmen-Funk*); Kreiskrankenhaus Gunzenhausen (F Freytag*); Stadtkrankenhaus, Hanau (H J Becker*, A Eisenmenger, W A Fach); Evangelisches Krankenhaus, Essen (P Bernhardt*); B K H Ernst Scheffler, Aue (H Jacob, K Malinowski*); Stadtisches Krankenhaus, Duren (H Roth, H Simon*); Kreiskrankenhaus, Waldbrol (H J Bias, K 0 Bischoff*); Stadtische Kliniken, Chemnitz (U Gerner*); Stadtische Kliniken, Oldburg (P Tomow*). Denmark: F A C Hillerod (S Galatius Jensen, K Mellemgaard*); Braedstrup Sygehus (A Amtoft*, A Axelsen, C Christensen); Rigshospitalet, Copenhagen (S Haunso, P Jorgensen, S Rasmussen*); Horsens Sygehus (P Strunge*); Naestved Centralsygehus (T Lysbo Svendsen*, H Madsen, W Nielsen); Svendborg Sygehus (F Egede*,J Rasmussen,J Markenvard); Centralsygehuset I Nykøbing F (H Mollerup, B Sigurd*); Frederikssund Sygehus (M Kirchhoff, A McNair, P E Nielsen*) ; K A S Glostrup, Copenhagen (J E Rokkedal Nielsen*); Sundby Hospital, Copenhagen (M P

Andersen*). Austria: Krankenanstalt Rudolfstiftung, Vienna (E Bucher, P Elliott, J Slany*); Medizinische Universitatsklimk Graz (M Grisold, W Klein*); Hanusch-Krankenhaus, Vienna (E Aldor*, C Domaus, W Kainz); Krankenhaus Lainz, Vienna (J Mlczoch, H Nobis*, S Trinks). Fmland. Helsinki

University Central Hospital (S Pohjola-Sintonen*); Jorvi Hospital, Espoo (V Naukkarinen*, R Siplla); Malmi District Hospital, Pietarsaari (K Nikus*, P E Wingren); Porvoo District Hospital (M Harkbnen*, S Jagerholm*); Lappi Central Hospital, Rovaniemi (M Eloranta*,J Isojarvi); Lohja District Hospital (0 Suhonen*); Lounais-Hame District Hospital, Forsa (J Koskelainen*). Italy’Polichnico Umversilano, Padova (F De Cian, S Dalla Volta*); Ospedale Generale, Bassano del Grappa (G Berton, F Cucchini*); Ospedale Civile, Piacenza (I Abelli, U Gazzola*, G Q Villani); Ospedale Generale Regionale, Vicenza (P Centofante, G M Mosele, M Vincenzi*); Ospedale Generale Provinciale, Conegliano (F Accorsi*, G A Deitos).

Luxembourg: Centre Hospitalier De Luxembourg (C Delagardelle*). mtMernd;

Kreisspital, Bulach (U Munch, H-P Vogelin*).

References 1

2

Gottlieb S, Moss AJ, McDermott M, Eberly S. Interrelation of left ventricular ejection fraction, pulmonary congestion and outcome in acute myocardial infarction. Am J Cardiol 1992; 69: 977-84. Killip T, Kimball JT. Treatment of myocardial infarction in a coronary care unit: a two year experience in 250 patients. Am J Cardiol

1967; 20: 457-64. 3

4

5

Bigger JT, Coromilas J, Weld FM, Reiffel JA, Rolmitzkey LM. Prognosis after recovery from acute myocardial infarction. Annu Rev Med 1984; 35: 127-47. Hall AS, Winter C, Bogle SM, Mackintosh AF, Murray GD, Ball SG. The Acute Infarction Ramipril Efficacy (AIRE) Study: rationale, design, organization, and outcome definitions. J Cardiovasc Pharmacol 1991; 18 (suppl 2): S105-S109. The Multicenter Postinfarction Research Group. Risk stratification and survival after acute myocardial infarction. N Engl J Med 1983; 309:

331-36. The CAST Investigators. Preliminary report: effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. N Engl J Med 1989; 321: 406-12. 7 The MDPIT research group. The effect of diltiazem on mortality and reinfarction after myocardial infarction. N Engl J Med 1988; 319: 385-92. 8 Swedburg K, Held P, Kjekshus J, et al. On behalf of the CONSENSUS II Study Group. Effects of early administration of enalapril on mortality in patients with acute myocardial infarction. Results of the Cooperative North Scandinavian Enalapril Survival Study II (CONSENSUS II). N Engl J Med 1992; 327: 678-84. 9 Sharpe N, Murphy J, Smith H, Hannan S. Treatment of patients with symptomless left ventricular dysfunction after myocardial infarction. Lancet 1988; 334: 255-59. 10 Hutchins GM, Bulkley BH. Infarct expansion versus extension: two different complications of acute myocardial infarction. Am J Cardiol

6

1978; 41: 1127-32. 11 Pfeffer MA, Braunwald E. Ventricular remodeling after myocardial infarction. Circulation 1990; 81: 1161-72. 12 The CONSENSUS trial study group. Effects of enalapril on mortality in severe congestive heart failure; results of the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS). N Engl J Med 1987; 316: 1429-35. 13 The SOLVD Investigators. Effect of enalapril on survival in patients with reduced ejection fractions and congestive heart failure. N Engl J Med 1991; 325: 293-302. 14 Cohn JN, Johnson G, Ziesche S, et al. A comparison of enalapril with hydralazine-isosorbide dinitrate in the treatment of chronic congestive heart failure. N Engl J Med 1991; 325: 303-10. 15 Pfeffer MA, Braunwald E, Moye LA, et al. Effect of captopril on mortality and morbidity in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med 1992; 327: 669-77. 16 Hall AS, Ball SG. ACE inhibitors after AMI. Lancet 1989; 337: 1527. 17 Bunning P. Kinetic properties of the angiotensin converting enzyme inhibitor ramiprilat. J Cardiovasc Pharmacol 1987; 10 (suppl 7): S31-S35. 18 Erman A, Winkler J, Chen-Gal B, Rabinov M, Zelykovski A, Tadjer S, Shmueli J, Levi E, Akbary A, Rosenfeld B. Inhibition of angiotensin converting enzyme by ramipril in serum and tissue of man. J Hypertens 1991; 9: 1057-62. 19 Linz W, Wiemer G, Scholkens BA. ACE-inhibition induces NOformation in cultured bovine endothelial cells and protects isolated ischaemic rat hearts. J Mol Cell Cardiol 1992; 24: 909-10. 20 Cleland JGF, Erhardt L, Hall AS, Winter C, Ball SG. Validation of primary and secondary outcomes and classification of mode of death among patients with clinical evidence of heart failure after myocardial infarction. A report from the Acute Infarction Ramipril Efficacy (AIRE) Study Investigators. J Cardiovasc Pharmacol (in press). 21 Moye LA, Pfeffer MA, Braunwald E, for the SAVE Investigators. Rationale, design and baseline characteristics of the survival and ventricular enlargement trial. Am J Cardiol 1991; 68: 70D79D. 22 The SOLVD Investigators. Effect of enalapril on mortality and the development of heart failure in asymptomatic patients with reduced left ventricular ejection fractions. N Engl J Med 1992; 327: 685-91. 23 Yusuf S, Pepine CJ, Garces C, et al. Effect of enalapril on myocardial infarction and unstable angina in patients with low ejection fractions. Lancet 1992; 340: 1173-78. 24 Hall D, Zeitler H, Rudolph W. Counteraction of the vasodilator effects of enalapril by aspirin in severe heart failure. J Am Coll Cardiol 1992; 20: 1549-55. 25 Yusuf S, Peto R, Lewis J, Collins R, Sleight P. Beta-blockade during and after myocardial infarction: an overview of the randomized trials. Prog Cardiovasc Dis 1985; 27 (5): 335-71.

827

26

27

Furberg CD, Hawkins CM, Lichstein E. Effect of propranolol in postinfarction patients with mechanical or electrical complications. Circulation 1984; 69 (4): 761-65. The GISSI-2 Investigators. GISSIS-2: a factorial randomised trial of alteplase versus streptokinase and heparin versus no heparin among 12 490 patients with acute myocardial infarction. Lancet 1990; 336:

65-71. 28 The International

Study Group. In-hospital mortality and clinical

of 20 891 patients with suspected acute myocardial infarction randomised between alteplase and streptokinase with or without heparin. Lancet 1990; 336: 71-75. 29 Stevenson R, Ranjadaylan K, Wilkinson P, Roberts R, Timmis AD. Short and long term prognosis of acute myocardial infarction since introduction of thrombolysis. BMJ 1993; 307: 349-53. 30 Lorell BH. ACE inhibitors after myocardial infarcton. N Engl J Med course

1993; 328: 966-67.

Aminoacid polymorphisms of insulin receptor substrate-1 in non-insulin-dependent diabetes mellitus

Introduction

Summary absolute insulin deficiency and insulin in the aetiology of non-insulininvolved insensitivity dependent diabetes mellitus (NIDDM), we examined whether patients with NIDDM exhibit genetic variability in the coding region of insulin receptor substrate-1 (IRS-1), a candidate gene that is ubiquitous in insulin-sensitive and insulin-like growth factor 1 (IGF1) sensitive tissues, including those that determine glucose production and clearance and those with regulatory effects on pancreatic &bgr;-cell function. IRS-1 has a central role as an adaptor molecule that links the insulinreceptor and IGF1-receptor kinases with enzymes that regulate cellular metabolism and growth. Single-stranded conformation polymorphism analysis and direct nucleotide sequencing were applied to genomic DNA from 86 unrelated patients with NIDDM and 76 normoglycaemic controls. 10 of the patients with NIDDM and 3 of the controls were heterozygous at codon 972 for a polymorphism in which glycine was substituted with arginine. Moreover, at codon 513, 6 patients with NIDDM and 2 controls had a heterozygous polymorphism with a transition from alanine to proline. None of the polymorphism carriers had both aminoacid variants and the total allelic frequency of IRS-1 polymorphisms was about three times higher in patients with NIDDM than in controls (p=0·02). Both aminoacid substitutions were located close to tyrosine phosphorylation motifs that are putative recognition sites for insulin and IGF1 signal transmission proteins. Analysis of the phenotypes showed that patients with NIDDM who had IRS-1 variants did not differ in their degree of insulin resistance compared with patients without known IRS-1 polymorphisms. However, carriers of the codon 972 variant had significantly lower plasma levels of fasting insulin and C-peptide. Our results suggest that aminoacid polymorphisms in IRS-1 may be involved in the aetiology of a subset of late-onset NIDDM. Since relative

or

are

Lancet 1993; 342: 828-32

Steno Diabetes Center and Hagedorn Research Institute (K Almind, C Bjørbaek PhD, H Vestergaard MD, T Hansen MD, S Echwald PhD, O Pedersen DMSc), Copenhagen, DK-2820 Gentofte, Denmark

Correspondence to: 828

Dr Oluf Pedersen

A considerable body of evidence suggests that genetic factors contribute to the pathogenesis of non-insulindependent diabetes mellitus (NIDDM).1 Studies of patients with overt NIDDM and of individuals at high risk of NIDDM reveal abnormalities of insulin secretion and insulin action.2 However, a genetic defect at one or more steps in the cellular action of insulin and insulin-like growth factor 1 (IGF1) might well involve the biochemical pathways of tissues that regulate insulin secretion and insulin-sensitive hepatic and extrahepatic tissues that

produce or extract glucose. Insulin initiates its cellular effects by binding to the a-subunit of its tetrameric plasma membrane receptor.3 The kinase in the P-subunit is thereby activated, which in turn catalyses the intramolecular autophosphorylation of specific tyrosine residues on the 0-subunit, which further stimulates the tyrosine kinase activity of the receptor towards other protein substrates in the cascade of insulin action. Recently, the first endogenous substrate for the insulin receptor kinase, insulin receptor substrate-1 (IRS-1) was cloned and sequenced .1-6 The complementary DNA sequence encodes a hydrophilic protein that contains multiple phosphorylation sites. During insulin exposure, this protein undergoes rapid tyrosine phosphorylation and the phosphorylated sites of IRS-1 associate with high affinity to cellular proteins that contain SH2 domains (src-homology-2 domains). Several SH2 domaincontaining proteins that are activated by IRS-1 have recently been identified which suggests that IRS-1acts as a multisite "docking" protein to bind signal proteins thereby linking the receptor kinase to the variety of cellular functions that are regulated by insulin.7 Besides being a substrate for the insulin receptor kinase, IRS-1 is phosphorylated after activation of the IGF1 receptor kinase.8 Although more than twenty different mutations of the insulin receptor gene have been reported in syndromes of severe insulin resistance frequently associated with acanthosis nigricans or ovarian hyperandrogenism,9 mutations in the insulin receptor molecule do not explain the genetic basis of the common form of NIDDM. Because of the key function of IRS-1 in mediating the early steps in the action of insulin and IGF1, we addressed whether patients with late-onset NIDDM exhibit variability in the coding region of the IRS-1gene.