Nine male patients (mean age 65 yr) with chronic atrial fibrillation underwent maximal exercise testing during placebo, beta-adrenergic (celiprolol, 600 mg), or calcium (diltiazem, 30 or 60 mg four times daily) channel blockade. The results were analyzed to determine which factors most closely related to ratings of perceived exertion (RPE) during exercise. Heart rate (HR), blood pressure (BP), oxygen uptake (VO2), minute ventilation (VE), and carbon dioxide production (VCO2) were evaluated at rest, 3.0 mph/0% grade, the gas exchange anaerobic threshold (ATge), 80% of placebo maximal O2 uptake, and maximal exercise. Both beta-adrenergic and calcium channel blockade significantly reduced heart rate and systolic blood pressure relative to placebo; these effects were more profound during beta-adrenergic blockade and as exercise progressed. Correlation coefficients and estimates of slope were derived for changes in RPE during exercise vs. changes in HR, VO2, VE, and VCO2 during the three treatments (r = 0.76 to 0.92, P less than 0.001). Although RPE was significantly correlated with HR during placebo and diltiazem therapy (r = 0.45, P less than 0.01), this was not the case during beta-adrenergic blockade (r = 0.31, NS). Slope of the regression lines between RPE and VO2, VE, and VCO2 did not differ between the three treatments. Slope of the regression lines between RPE and HR differed only during calcium channel blockade. Because the presence of atrial fibrillation and beta-adrenergic blockade altered the associations between RPE, VO2, and HR, these results suggest that VE is more closely related to RPE than the other parameters.