Copyright © 2004 by the European Society of Cardiology.
Letter to the Editor
Assessing the effect of ACE inhibitors on exercise tolerance: a question of study design
The Cardiothoracic Centre, Liverpool, UK
Unit of Molecular Cardiology, University of Leeds, UK
* Tel.: +44-151-228-1616; fax: +44-151-293-2254
E-mail address: drsgwilliams{at}hotmail.com
Dear Prof Van de Werf
We read with great interest the paper by Abdulla et al.1 investigating the effects of trandolapril on exercise tolerance in patients with left ventricular dysfunction post myocardial infarction. The investigators showed trandolapril to have a neutral effect on exercise tolerance at 1, 3 and 12 months following myocardial infarction. Although ACE inhibitors have been conclusively shown to be prognostically beneficial to patients with heart failure,2–4 their effects on exercise capacity have been rather variable. A potential reason for the observed inconsistencies is that the often used parallel-group study design, which is ideal for mortality studies, may not be suitable for investigating drug effects on exercise capacity.5,6
Of the 254 patients recruited for the exercise sub-study by Abdulla et al.,1 144 patients (57%) had either dropped out of the study or died at 1 year follow-up. Thus, it is uncertain whether the perceived (lack of) difference in exercise tolerance between the trandolapril and placebo groups were due to the dropping out of patients who were less able to exercise. In view of the high dropout rates, results from this and other parallel-group studies need to be interpreted with caution. It is important to note that the impact of parallel-group study design on trials assessing drug effects on exercise capacity is quite different from the impact on trials assessing prognosis.
The process of randomised allocation to the two parallel groups is to ensure that they are as comparable as possible in all aspects. However, even when using a mixed model analysis, as was the case in the study by Abdulla et al., the exercise capacity of dropouts cannot be ascertained. Those who died might be arbitrarily assigned an exercise capacity of zero,7 but those who refused to exercise or dropped out of study cannot be assigned an assumed exercise capacity. The arbitrarily assigned values are artefactual, prone to rendering the final results unreliable.
Removing the dropouts from the final analyses disturbs the original comparability (through randomisation) of the parallel treatment groups, and the resulting observed differences may therefore turn out to be due to selection bias, rather than true differences in treatment effects.8 Thus, although a parallel group study design is considered to be satisfactory for mortality studies, it has severe limitations when investigating the drug effects on exercise capacity.
References
- Abdulla J, Burchardt H, Abildstrom SZ et al. The angiotensin enzyme inhibitor trandolapril has neutral effect on exercise tolerance or functional class in patients with myocardial infarction and reduced left ventricular systolic function. Eur. Heart J. 2003;24:2116–2122.
[Abstract/Free Full Text] - 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). The CONSENSUS Trial Study Group. N Engl J Med 1987;316:1429–35.
- The SOLVD Investigators. Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med 1991;325:293–302.
- Swedberg K, Kjekshus J, Snapinn S, for the CONSENSUS Investigators. Longterm survival in severe heart failure in patients treated with enalapril. Ten year followup of CONSENSUS I. Eur Heart J 1999;20:1369.
- Williams SG, Cooke GA, Wright DJ et al. Disparate results of ACE inhibitor dosage on exercise capacity in heart failure: A reappraisal of vasodilator therapy and study design. Int. J. Cardiol. 2001;77:239–245.[CrossRef][ISI][Medline]
- Cooke GA, Williams SG, Marshall P et al. A mechanistic investigation of ACE inhibitor does effects on aerobic exercise capacity in heart failure patients. Eur. Heart J. 2002;23:1360–1368.
[Abstract/Free Full Text] - Lubsen J, Chadha DR, Yotof YT et al. Metaanalysis of morbidity and mortality in five exercise capacity trials evaluating ramipril in chronic congestive cardiac failure. Am. J. Cardiol. 1996;77:11916.
- Kaplan RM, Atkins CJ. Selective attrition causes overestimates of treatment effects in studies of weight loss. Addict. Behav. 1987;12:297–302.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||