Metoprolol in dilated cardiomyopathy: is it possible to identify factors predictive of improvement? The Heart Muscle Disease Study Group

Abstract

Background: Some controlled clinical trials showed a beneficial effect of beta-blockers on symptoms, exercise tolerance, and left ventricular function in dilated cardiomyopathy. The purpose of this study was to investigate if there are clinical variables at baseline that could predict a favorable response to long-term metoprolol therapy.

Methods and results: Since November 1987, 94 consecutive patients with dilated cardiomyopathy and left ventricular ejection fraction less than 0.40 were treated with metoprolol (mean final dosage, 136 +/- 32 mg) associated with tailored medical therapy with digitalis, diuretics, and angiotensin-converting enzyme inhibitors. Eighty-four surviving patients had a complete 2-year noninvasive follow-up period. Ten patients died or were transplanted before the final assessment. Improvement was defined according to a clinical score based on left ventricular ejection fraction (increase > or = 10 U), left ventricular end-diastolic diameter (decrease > or = 10%), regression of restrictive filling pattern, New York Heart Association functional class, exercise tolerance (increase > or = 2 minutes), and cardiothoracic ratio (decrease > or = 10%). According to these criteria, 48 patients (51.1%) were classified as improved. Multivariate analysis identified a group of patients with a history of mild hypertension (blood pressure between 140/90 and 170/100 mmHg) and significantly higher probability of improvement with longterm metoprolol (odds ratio [OR], 2.22; 95% confidence interval, 1.25-3.94; P = .007). Among the 71 patients with normal blood pressure (< 140/90 mmHg), heart rate in upright position (100 vs 75 beats/min: OR, 2; 95% confidence interval, 1.38-4.94; P = .003), left ventricular ejection fraction 0.20-0.33 versus less than 0.20 (OR, 4.72; 95% confidence interval, 1.06-21.04; P = .042), and New York Heart Association class I-II versus III-IV (OR, 2.74; 95% confidence interval, 0.97-7.75; P = .05) were significantly associated with a positive response to metoprolol. At baseline, both supine and upright heart rate were significantly higher in patients who improved, but heart rate in the upright position was the most significant predictor of improvement in patients with normal blood pressure at multivariate analysis.

Conclusions: According to the authors' logit model, patients with a history of mild hypertension or with a higher resting heart rate, associated with controlled symptoms of heart failure (New York Heart Association class I-II) or moderate to severe left ventricular ejection fraction (range, 0.20-0.33) showed a remarkable probability of long-term (2-year) improvement on metoprolol.