The natural history of chronotropic incompetence is not clear. To assess this, we evaluated corrected sinus node recovery time (cSNRT) and sinus node chronotropy at rest and during exercise in two groups of syncopal patients with sinus node disease. Group A comprised patients with resting bradycardia but normal cSNRT and group B had resting bradycardia and prolonged cSNRT (greater than 1000 ms). An additional two groups (C and D) were studied. Group C comprised patients with complete AV (CAVB) and no evidence of sinus node disease and group D were asymptomatic controls of similar age. At diagnosis, patients with symptomatic bradycardia but normal cSNRT and no evidence of carotid sinus syndrome (group A) had resting bradycardia and impaired peak heart rate (PHR-I) on exercise compared to controls (P less than 0.001 and P less than 0.05, respectively), but no reduction in exercise duration. At follow-up group A patients demonstrated an increase in resting rate that was significantly slower than the controls (P less than 0.01). Peak heart rate (PHR-II) also remained significantly slower (P less than 0.05). There was no difference in exercise duration between groups A and D at follow-up. Group B was further subdivided according to follow-up findings of preservation of atrial activity in seven patients (group B-1) and junctional rhythm without any atrial activity in four patients (group B-2). Retrospective analysis showed no significant difference in resting heart rate at initial examination but group B-2 showed a significantly lower peak heart rate on exercise compared with B-1 (P less than 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)