A poor correlation exists between oscillometric and radial arterial blood pressure as measured by the Philips MP90 monitor.

In anesthesia and critical care, invasive arterial blood pressure monitoring is the gold standard against which other methods of monitoring are compared. In this assessment of the Philips MP90 monitor, the objective was to determine whether or not oscillometric measurements were within the accuracy standards set by the Association for the Advancement of Medical Instrumentation (AAMI) and the British Hypertension Society (BHS). Three hundred and one invasive and noninvasive paired measurements were obtained from eleven adult patients on the neurosurgical service at Stanford University Medical Center. Bland-Altman plots were created to assess agreement between the two measurement systems. Paired correlation analysis, bias and precision calculations were performed. Oscillometric blood pressure measurements correlated with arterial measurements yielding Pearson r values of 0.68, 0.67 and 0.62 for systolic, diastolic and mean pressures, respectively (P < 0.01.) Mean differences with 95% confidence intervals were -3.8 mmHg +/- 13.6, -2.4 mmHg +/- 10.0, and 4.0 mmHg +/- 13.1 for systolic, diastolic and mean pressures, respectively. The mean difference for these measurements was <or=5 mmHg as stipulated by the AAMI guidelines, but the standard deviation was greater than the 8 mmHg allowed by the AAMI guidelines. When the BHS guidelines were applied, the device merited a grade "D" for systolic and mean arterial pressure, and a grade "C" for diastolic pressure, with the highest possible grade level being "A." There was a poor correlation between noninvasive and invasive measurements of arterial blood pressure as measured with a cuff and radial arterial cannula using the Philips MP90 monitor. These inaccuracies could lead to unnecessary interventions, or lack of appropriate interventions in anesthetic management. Further study is needed to specify the absolute inaccuracy of the monitor, and to determine if accuracy between the two methods varies with patient co-morbidities, surgical procedures, or anesthetic management.