Objectives: For some time, the inaccuracies of non-invasive blood pressure measurement in critically ill patients have been recognised. Measurement difficulties can occur even in optimal conditions, but in prehospital transportation vehicles, problems are exacerbated. Intra-arterial pressures must be used as the reference against which to compare the performance of non-invasive methods in the critically ill patient population. Intra-arterial manometer data observed from the patient monitor has frequently been used as the reference against which to assess the accuracy of noninvasive devices in the emergency setting. To test this method's validity, this study aimed to determine whether numerical monitor pressures can be considered interchangeable with independently sampled intra-arterial pressures.
Methods: Intensive Care Unit nurses were asked to document arterial systolic, diastolic and mean pressures numerically displayed on the patient monitor. Observed pressures were compared to reference intra-arterial pressures independently recorded to a computer following analogue to digital conversion. Differences between observed and recorded pressures were evaluated using the Association for the Advancement of Medical Instrumentation (AAMI) protocol. Additionally, two-level linear mixed effects analyses and Bland-Altman comparisons were undertaken.
Results: Systolic, diastolic and integrated mean pressures observed during 60 data collection sessions (n = 600) fulfilled AAMI protocol criteria. Integrated mean pressures were the most robust. For these pressures, mean error (reference minus observed) was 0.5 mm Hg (SD 1.4 mm Hg); 95% CI (two-level linear mixed effects analysis) 0.4-0.6 mm Hg; P < 0.001. Bland-Altman plots demonstrated tight 95% limits of agreement (-2.3 to 3.2 mm Hg), and uniform agreement across the range of mean blood pressures.
Conclusions: Integrated mean arterial pressures observed from a well maintained patient monitor can be considered interchangeable with independently sampled intra-arterial pressures and may be confidently used as the reference against which to test the accuracy of non-invasive blood pressure measuring methods in the prehospital or emergency setting.