The validity of the Dinamap automated oscillometric blood pressure monitor, an instrument frequently used for continuous monitoring in intensive care settings, was evaluated in comparison to two standard methods of blood pressure measurement. Oscillometric monitoring was compared to random-zero sphygmomanometry in 28 ambulatory subjects, using the instruments in a random order. The oscillometric monitor overestimated random-zero systolic pressure (mean difference = 5.7 mm Hg; P less than 0.001), but did not differ in diastolic pressure (difference = 1.3 mm Hg, P greater than 0.3). Correlation (r) between methods was 0.94 for systolic pressure and 0.83 for diastolic pressure. Repeatability of sequential measures did not differ between methods. To evaluate the oscillometric monitor in the intraoperative setting, it was used simultaneously with a radial artery catheter to measure blood pressure in 14 patients undergoing peripheral vascular or other major surgery. The oscillometric monitor underestimated intra-arterial systolic pressure (mean difference = -9.4 mm Hg; P less than 0.002), but overestimated intraarterial diastolic pressure (difference = 5.7 mm Hg; P less than 0.005). There was no difference in mean arterial pressure (P greater than 0.1). Correlation between methods was 0.88 for systolic, 0.71 for diastolic, and 0.79 for mean arterial pressure. The oscillometric instrument had a sensitivity of 63% and specificity of 97% in detecting intraarterial systolic pressures of 160 mm Hg or greater, and a sensitivity of 89% and specificity of 89% in detecting intraarterial systolic pressures of 100 mm Hg or less. These findings demonstrate differences between the oscillometric monitor and other standard methods of blood pressure measurement.(ABSTRACT TRUNCATED AT 250 WORDS)