In 28 patients (pts) (5 normal subjects (N), 12 pts with valvular heart disease, 6 with cardiomyopathy, and 5 with ischemic heart disease), left ventricular (LV) pressure was measured using Millar's catheter tip transducer simultaneously with LV posterior wall thickness (W) by echocardiography. After the control recording, LV peak systolic pressure was altered by about 30 mmHg from the control level by angiotensin (20 ng/kg/min) or nitroprusside (1 gamma/kg/min) to attain two (18 pts) or three (10 pts) different end-systolic pressure points, determined after the steady state was achieved. LV end-systolic pressure and wall thickness (P-W) relation a the three different end-systolic pressure points was found to be linear (r = 0.996) in each of the 10 pts, suggesting that end-systolic P-W relation is independent of afterload. Isoproterenol infused in 4 pts, which led to an increase in peak positive (+) dp/dt by a mean of 28%, increased the slope of the P-W relation by 43.4% of the control value (P less than 0.001), while propranolol, which led to a decrease in peak (+) dp/dt by a mean of 16.7%, reduced the slope by 27.3% (p less than 0.001). This indicates the sensitiveness of the slope of P-W relation to the inotropic change of the myocardium. In N, the slope of the P-W relation was -29.7 +/0 6.0 mmHg/mm (mean +/- SD), while the value in 4 pts with decreased functional capacity (NYHA III or IV) was significantly lower (-13.7 +/- 2.9 mmHg/mm; p less than 0.001). Thus, the present study suggests that the slope of the end-systolic P-W relation is useful in assessing regional myocardial contractility in man.