[Evaluation of automatic sphygmomanometers]

In clinical evaluation of a new type of automatic sphygmomanometer, a strict comparability of results with the standard indirect auscultatory method is required. We have compared the oscillometric apparatus BP 103 N Nippon Colin (B) with the Hawksley random zero mercury sphygmomanometer (A) in a stratified sample of 26 subjects using a cross-over design of four simultaneous pairs of measurements in both arms by two observers. Mean values obtained by machine B were higher than values obtained by machine A, by 0.56 kPa systolic and by 0.62 kPa diastolic. Analysis of variance showed that the differences between both machines were within the random variability of blood pressure values obtained, and therefore irrelevant at single measurements. In case of blood pressure measurements in an epidemiological study, a correction of values should be done using the regression model B = K. Ac.