Differences in oscillometric non-invasive blood pressure measurements recorded by different revisions of the Philips Component Monitoring System

Abstract

To investigate whether a simulator can detect differences in oscillometric noninvasive blood pressure measurements in different revisions of the Philips Component Monitoring System. Philips revised the intraarterial-based algorithm with software revision C.0, and added an auscultation-based algorithm in revision C.1. Three methods were adopted. Firstly, differences were assessed using the Dynatech Nevada CuffLink simulator. Five noninvasive blood pressure modules were tested on monitors with software revision C (software revision prior to C.0) and revision C.0. Monitor C.1 allows the selection of the original intraarterial algorithm (as for revision C), the improved intraarterial (revision C.0) or auscultation; two noninvasive blood pressure modules were tested on it.Secondly, pressures were recorded simultaneously from a participant's left and right arms by monitors C and C.0. The monitors were then swapped between arms and a second set of six pressures were recorded. The two sets were subtracted to calculate the differences between monitors, removing interarm differences. The experiment was carried out on two participants.Thirdly, pressures were recorded from three participants with a simultaneous reference auscultatory measurement. Six sets of simultaneous oscillometric and auscultatory determinations were recorded from monitors C and C.0. All three methods revealed that the 'original' intraarterial-based algorithm in revision C recorded lower systolic, diastolic and mean arterial pressures than revision C.0, with the difference in diastolic pressure being the greatest. The difference in diastolic pressures between C and C.0 was up to 15 mmHg. The simulator evaluation finding of higher systolic and diastolic pressures recorded by the revised intraarterial algorithm (C.1 option 3) than by the auscultation-based algorithm (C.1 option 2) is in qualitative agreement with the regression equations of the clinical validation studies. The results suggest that simulators can identify the existence of differing algorithms used to measure blood pressure by the CMS monitor. Philips has continued to develop the algorithms used to determine the pressure by the oscillometric method. While significantly different pressures were recorded by the different revisions, it is interesting that each revision passed clinical validation studies.