Ventricular pacing (with loss of normal atrioventricular synchrony) has been considered to have a role in the development or progression of congestive heart failure (CHF) in patients with sick sinus syndrome (SSS). No rigorous study has tested this hypothesis. Five hundred seven consecutive patients with SSS who received an initial pacemaker from January 1980 to December 1989 were studied. Atrial or dual-chamber pacemakers were implanted in 395 patients and ventricular pacemakers in 112. After a mean follow-up of 65 +/- 37 months, 97 patients (19%) developed new CHF or increased their New York Heart Association functional class. By univariate analysis, preimplant predictors for these events were left ventricular dysfunction (p < 0.001), valvular heart disease (p = 0.004), peripheral vascular disease (p = 0.005), diabetes (p = 0.02), coronary artery disease (p = 0.02), high New York Heart Association functional class (p = 0.03) and complex ventricular arrhythmia (p = 0.03). By multivariate analysis (logistic regression), the only predictors for CHF were valvular heart disease (p = 0.002; odds ratio [OR] 2.51), peripheral vascular disease (p = 0.003; OR 1.7) and complex ventricular arrhythmia (p = 0.027; OR 2.74). When the analysis was restricted to patients who had preimplant assessment of left ventricular function, independent predictors for CHF were left ventricular dysfunction (p < 0.001; OR 1.66), and complex ventricular arrhythmia (p < 0.001; OR 1.75). In conclusion, progressive or new-onset CHF is a consequence of the underlying cardiovascular disease. In the present population of patients with SSS, ventricular pacing mode was not associated with an increased incidence of CHF.