Losartan for cardiovascular disease in patients with and without diabetes in the LIFE study

Abstract

We write to propose that a more standard format is adopted for the reporting of major trials. The latest study that does not present complete and interpretable findings is the Losartan Intervention for Endpoint reduction (LIFE) study by Björn Dahlöf and colleagues (March 23, p 995).1Dahlöf B Devereux RB Kjeldsen SE et al.for the LIFE study groupCardiovascular morbidity and mortality in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol.Lancet. 2002; 359: 995-1003Summary Full Text Full Text PDF PubMed Scopus (4767) Google ScholarMany examples can be identified in leading journals.1Dahlöf B Devereux RB Kjeldsen SE et al.for the LIFE study groupCardiovascular morbidity and mortality in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol.Lancet. 2002; 359: 995-1003Summary Full Text Full Text PDF PubMed Scopus (4767) Google Scholar, 2Yusuf S Sleight P Pogue J Bosch J Davies R Dagenais G Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in highrisk patients: the Heart Outcomes Prevention Evaluation study investigators.N Engl J Med. 2000; 342: 145-153Crossref PubMed Scopus (8126) Google Scholar, 3Bosch J Yusuf S Pogue J et al.on behalf of the HOPE investigatorsUse of ramipril in preventing stroke: double blind randomised trial.BMJ. 2002; 324: 699-702Crossref PubMed Google Scholar The three major deficiencies are a failure to present the details of the endpoints contributing to a composite endpoint, failure to provide information on survival free of an event in analyses of the occurrence of an event, and failure to provide event rates.An example of the first deficiency is seen in table 3 of the LIFE study. Numbers are provided for the composite endpoint of cardiovascular death, myocardial infarction, and stroke, but in the three subsequent lines information is presented on the total number of events for each outcome over the duration of the trial. Thus, the number and nature of each first event that contributes to the composite endpoint cannot be established.In the Heart Outcomes Prevention Evaluation (HOPE) study2Yusuf S Sleight P Pogue J Bosch J Davies R Dagenais G Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in highrisk patients: the Heart Outcomes Prevention Evaluation study investigators.N Engl J Med. 2000; 342: 145-153Crossref PubMed Scopus (8126) Google Scholar the same error arises for incidence of the primary outcome and of deaths from any cause. And yet Cohn and Tognoni,4Cohn JN Tognoni G A randomized trial of the angiotensin-receptor blocker valsartan in chronic heart failure.N Engl J Med. 2001; 345: 1667-1675Crossref PubMed Scopus (2741) Google Scholar in their trial of valsartan in chronic heart failure, provide the contribution of the endpoints to the combined endpoint, but the total events over the period of the trial for each of the endpoints is reported only in a cursory way.An example of the second deficiency is seen in a further HOPE study report,3Bosch J Yusuf S Pogue J et al.on behalf of the HOPE investigatorsUse of ramipril in preventing stroke: double blind randomised trial.BMJ. 2002; 324: 699-702Crossref PubMed Google Scholar in which the investigators focus on the prevention of stroke. In their table 1, they show total number of strokes. For an unknown reason, the total number of strokes differs from the total of subtypes of stroke classified as ischaemic or non-ischaemic. Their Kaplan-Meier estimate of stroke alone does not include an analysis of strokefree survival. Dead patients cannot have strokes.A similar deficiency can be found in most trial reports when a composite endpoint is used, such as cardiovascular death, myocardial infarction, or stroke. It is generally impossible to identify how many patients survived the trial free of myocardial infarction or stroke. The LIFE and HOPE reports are examples.1Dahlöf B Devereux RB Kjeldsen SE et al.for the LIFE study groupCardiovascular morbidity and mortality in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol.Lancet. 2002; 359: 995-1003Summary Full Text Full Text PDF PubMed Scopus (4767) Google Scholar, 2Yusuf S Sleight P Pogue J Bosch J Davies R Dagenais G Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in highrisk patients: the Heart Outcomes Prevention Evaluation study investigators.N Engl J Med. 2000; 342: 145-153Crossref PubMed Scopus (8126) Google ScholarThe third deficiency relates to the frequent absence of information on absolute rates of events expressed as number of events per unit of persontime of follow-up. Relative risks and p values mean little for individuals. What matters is the risk difference. Numbers-needed-to-treat have become popular, but few publications report the relevant event rates directly or give data on treatment-group specific mean durations of follow-up required to calculate these numbers.5Lubsen J Hoes A Grobbee DE Implications of trial results: the potentially misleading notions of number needed to treat and average duration of life gained.Lancet. 2000; 356: 1757-1759Summary Full Text Full Text PDF PubMed Scopus (48) Google ScholarDespite previous efforts with the CONSORT guidelines to standardise the presentation of trials, more guidance is needed so readers are not at a disadvantage or completely uninformed when trying to interpret the results of trials for clinical practice. Lack of space is a destitute excuse, since a more interpretable presentation of results can usually be achieved without increasing the length of a report. We write to propose that a more standard format is adopted for the reporting of major trials. The latest study that does not present complete and interpretable findings is the Losartan Intervention for Endpoint reduction (LIFE) study by Björn Dahlöf and colleagues (March 23, p 995).1Dahlöf B Devereux RB Kjeldsen SE et al.for the LIFE study groupCardiovascular morbidity and mortality in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol.Lancet. 2002; 359: 995-1003Summary Full Text Full Text PDF PubMed Scopus (4767) Google Scholar Many examples can be identified in leading journals.1Dahlöf B Devereux RB Kjeldsen SE et al.for the LIFE study groupCardiovascular morbidity and mortality in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol.Lancet. 2002; 359: 995-1003Summary Full Text Full Text PDF PubMed Scopus (4767) Google Scholar, 2Yusuf S Sleight P Pogue J Bosch J Davies R Dagenais G Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in highrisk patients: the Heart Outcomes Prevention Evaluation study investigators.N Engl J Med. 2000; 342: 145-153Crossref PubMed Scopus (8126) Google Scholar, 3Bosch J Yusuf S Pogue J et al.on behalf of the HOPE investigatorsUse of ramipril in preventing stroke: double blind randomised trial.BMJ. 2002; 324: 699-702Crossref PubMed Google Scholar The three major deficiencies are a failure to present the details of the endpoints contributing to a composite endpoint, failure to provide information on survival free of an event in analyses of the occurrence of an event, and failure to provide event rates. An example of the first deficiency is seen in table 3 of the LIFE study. Numbers are provided for the composite endpoint of cardiovascular death, myocardial infarction, and stroke, but in the three subsequent lines information is presented on the total number of events for each outcome over the duration of the trial. Thus, the number and nature of each first event that contributes to the composite endpoint cannot be established. In the Heart Outcomes Prevention Evaluation (HOPE) study2Yusuf S Sleight P Pogue J Bosch J Davies R Dagenais G Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in highrisk patients: the Heart Outcomes Prevention Evaluation study investigators.N Engl J Med. 2000; 342: 145-153Crossref PubMed Scopus (8126) Google Scholar the same error arises for incidence of the primary outcome and of deaths from any cause. And yet Cohn and Tognoni,4Cohn JN Tognoni G A randomized trial of the angiotensin-receptor blocker valsartan in chronic heart failure.N Engl J Med. 2001; 345: 1667-1675Crossref PubMed Scopus (2741) Google Scholar in their trial of valsartan in chronic heart failure, provide the contribution of the endpoints to the combined endpoint, but the total events over the period of the trial for each of the endpoints is reported only in a cursory way. An example of the second deficiency is seen in a further HOPE study report,3Bosch J Yusuf S Pogue J et al.on behalf of the HOPE investigatorsUse of ramipril in preventing stroke: double blind randomised trial.BMJ. 2002; 324: 699-702Crossref PubMed Google Scholar in which the investigators focus on the prevention of stroke. In their table 1, they show total number of strokes. For an unknown reason, the total number of strokes differs from the total of subtypes of stroke classified as ischaemic or non-ischaemic. Their Kaplan-Meier estimate of stroke alone does not include an analysis of strokefree survival. Dead patients cannot have strokes. A similar deficiency can be found in most trial reports when a composite endpoint is used, such as cardiovascular death, myocardial infarction, or stroke. It is generally impossible to identify how many patients survived the trial free of myocardial infarction or stroke. The LIFE and HOPE reports are examples.1Dahlöf B Devereux RB Kjeldsen SE et al.for the LIFE study groupCardiovascular morbidity and mortality in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol.Lancet. 2002; 359: 995-1003Summary Full Text Full Text PDF PubMed Scopus (4767) Google Scholar, 2Yusuf S Sleight P Pogue J Bosch J Davies R Dagenais G Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in highrisk patients: the Heart Outcomes Prevention Evaluation study investigators.N Engl J Med. 2000; 342: 145-153Crossref PubMed Scopus (8126) Google Scholar The third deficiency relates to the frequent absence of information on absolute rates of events expressed as number of events per unit of persontime of follow-up. Relative risks and p values mean little for individuals. What matters is the risk difference. Numbers-needed-to-treat have become popular, but few publications report the relevant event rates directly or give data on treatment-group specific mean durations of follow-up required to calculate these numbers.5Lubsen J Hoes A Grobbee DE Implications of trial results: the potentially misleading notions of number needed to treat and average duration of life gained.Lancet. 2000; 356: 1757-1759Summary Full Text Full Text PDF PubMed Scopus (48) Google Scholar Despite previous efforts with the CONSORT guidelines to standardise the presentation of trials, more guidance is needed so readers are not at a disadvantage or completely uninformed when trying to interpret the results of trials for clinical practice. Lack of space is a destitute excuse, since a more interpretable presentation of results can usually be achieved without increasing the length of a report. Losartan for cardiovascular disease in patients with and without diabetes in the LIFE studyAuthors' reply Full-Text PDF