A quantitative evaluation of the Hewlett-Packard 78354A noninvasive blood pressure meter.

Oscillometrically determined brachial artery pressures were compared with simultaneous contralateral radial intraarterial pressures in 19 anesthetized adult cardiac surgical patients throughout their surgical procedures, interrupted only by nonpulsatile, low-pressure, low-flow cardiopulmonary bypass. Radial intraarterial pressure values ranged widely for systolic (55 to 207 torr), mean (43 to 141 torr), and diastolic (26 to 106 torr). Both error specification methods proposed by the Association for the Advancement of Medical Instrumentation were used and compared. As expected, error method 1 gave consistently lower mean errors, smaller error standard deviations, and higher correlation coefficients than did error method 2. The errors during time periods immediately before and after cardiopulmonary bypass were compared with those from more quiescent times. Higher mean errors, larger error standard deviations, and lower correlation and regression coefficients were found during those time periods surrounding cardiopulmonary bypass. In general, mean errors were lowest for systolic pressure, followed by mean and diastolic pressures in that order, whereas error standard deviations were smallest for mean pressure, followed by systolic and diastolic pressures. Correlation and regression coefficients were highest for systolic pressure, followed by mean and diastolic pressures. In summary, the oscillometric method provides convenient and reproducible estimates of radial intraarterial pressure during most clinical situations, typically with better accuracy than the auscultatory Korotkoff method. The accuracy and reproducibility are diminished during those periods immediately surrounding cardiopulmonary bypass, perhaps due to direct surgical manipulation of the heart with its attendant rapid changes in cardiac output and blood pressure.