To assess the reliability of Doppler echocardiography (DE) in measuring the right ventricular systolic pressure (RVSP) in ventricular septal defect (VSD), both left and right heart catheterization (Cath) and DE were performed simultaneously in 59 cases with VSD. Systolic shunt velocities through VSD were recorded by DE and converted into the peak instantaneous (delta P-PD) and mid-systolic (delta P-MD) pressure gradients using the simplified Bernoulli equation RVSP was estimated by subtracting delta P-PD from the brachial artery systolic pressure (BASP) measured by a cuff sphygmomanometer (method A) and by subtracting delta P-MD from BASP (method B). The left ventricular systolic pressure (LVSP), RVSP, and the peak instantaneous (delta P-PC) and the peak-to-peak (delta P-PP) pressure gradients were measured from pressure c rves. The comparison between BASP and LVSP yielded a good correlation (r = 0.90, SEE = 0.76 kPa). There were also good correlation of interventricular pressure gradients measured by two techniques (r = 0.98, SEE = 0.83-0.93 kPa). Although RVSP estimated by method A correlated well with that measured by Cath, there was a significant underestimation (P < 0.05). On the other hand, RVSP estimated by method B agreed highly with that measured by Cath and there was no significant difference between the two means. We conclude that DE offers a reliable technique for estimating RVSP in VSD noninvasively.