Correspondence

Beta-Blockers after Myocardial Infarction

N Engl J Med 1998; 339:2022-2023December 31, 1998

Article

To the Editor:

In their report on an observational study of beta-blockade after myocardial infarction, Gottlieb et al. (Aug. 20 issue)1 firmly conclude that beta-blockers should be used for high-risk patients after acute myocardial infarction, including patients with congestive heart failure. This conclusion is not supported by the data presented. The randomized studies of beta-blockers excluded patients with severe heart failure — and all physicians are trained in such contraindications. Therefore, we know that physicians will be reluctant to use beta-blockers in high-risk patients, and we can readily anticipate that the willingness to use a beta-blocker is a marker of low risk, both overall and in subgroups of patients. The authors' demonstration that beta-blockers are associated with a dramatic reduction in risk confirms a self-evident truth, regardless of the huge sample. The authors could have confirmed this explanation by demonstrating that beta-blockers are a marker of good exercise performance, a short hospital stay, and higher social status — other unspecific indicators of low risk.

The value of beta-blockers in high-risk patients must be considered uncertain until proper randomized trials have proved otherwise. Such studies are under way in our country and elsewhere, and their completion must be awaited before firm conclusions are drawn.

Christian Torp-Pedersen, M.D., D.Sc.
Lars Køber, M.D.
Gentofte University Hospital, DK 2900 Hellerup, Denmark

1 References

1. 1

   Gottlieb SS, McCarter RJ, Vogel RA. Effect of beta-blockade on mortality among high-risk and low-risk patients after myocardial infarction. N Engl J Med 1998;339:489-497
   Full Text | Web of Science | Medline

Author/Editor Response

The authors reply:

To the Editor: Drs. Torp-Pedersen and Køber correctly point out that patients who receive beta-blockers may differ from those who do not, and this may in part explain the benefit of beta-blockade observed in our study. However, by neglecting the consistency and magnitude of the results and by ignoring our ability to control for multiple variables, they would have us disregard substantial evidence of a survival benefit in patients receiving beta-blockers.

We controlled for many factors known to affect mortality, including length of hospital stay, ejection fraction, sex, age, race or ethnic group, indexes of the severity of illness (Acute Physiology and Chronic Health Evaluation [APACHE II] and Killip score), concomitant illnesses (asthma, chronic obstructive pulmonary disease, and diabetes mellitus), hemodynamic variables such as heart rate and blood pressure, and metabolic factors. We also controlled for interventions performed at the time of the infarction and for the medical history. The patients who received beta-blockers were slightly healthier than those who did not, as indicated by these factors, for which we controlled. It is difficult to hypothesize what other factors a patient's physician would use in defining a much healthier subgroup of patients to receive beta-blockers.

Our analysis showed that in the groups of patients with perceived contraindications, the mortality rate (even among those who received beta-blockers) was higher than among patients without complications, suggesting that these patients were truly ill. Nevertheless, the magnitude of the survival benefit in these groups was the same (approximately 40 percent) as that in groups of patients in whom beta-blockade has proved efficacious in controlled studies. To state that the survival benefit in some groups was due to the intervention but that the same benefit in other groups (when the analysis was controlled for all other variables) was dependent on some unspecified variable is a highly improbable explanation of these results.

Our results are also consistent with subgroup analyses (in blinded, controlled, and randomized studies) of patients with heart failure1 and diabetes mellitus.2 These studies have shown that the improvement in survival, expressed as a percentage, is equal to that in “ideal” patients.

We, too, place the highest value on the results of randomized clinical trials. In the absence of randomized studies involving all groups of patients (an impossibility), our data provide the next best evidence. Such a large, detailed observational study can adequately take into account numerous confounding variables while focusing on the intervention of interest.

In the light of the consistency of the benefit seen in both randomized clinical trials and observational studies, it is no longer acceptable that only one third of patients receive beta-blockers after myocardial infarction.

Stephen S. Gottlieb, M.D.
Robert J. McCarter, Sc.D.
Robert A. Vogel, M.D.
University of Maryland School of Medicine, Baltimore, MD 21201-1595

2 References

1. 1

   Chadda K, Goldstein S, Byington R, Curb JD. Effect of propranolol after acute myocardial infarction in patients with congestive heart failure. Circulation 1986;73:503-510
   CrossRef | Web of Science | Medline

2. 2

   Zuanetti G, Latini R, Maggioni A, Santoro L, Franzosi MG. Influence of diabetes on mortality in acute myocardial infarction: data from theGISSI-2 study. J Am Coll Cardiol 1993;22:1788-1794
   CrossRef | Web of Science | Medline