The Mean Machine; Accurate Non-invasive Blood Pressure Measurement in the Critically Ill Patient

Abstract

Accurate indirect prehospital blood pressure measurement in the critically ill patient remains an important challenge to both patient management and prehospital research. Ambulatory blood pressure measuring devices have not been trialled for prehospital use in critically ill patients. Prior to prehospital validation where conditions are suboptimal, we aimed to test under favourable conditions in the Intensive Care Unit, a selection of ambulatory devices that may be suitable for use in the field. Systolic, diastolic and mean pressures of three ambulatory devices were compared to the average of 1 min of independently recorded, high fidelity intra-arterial reference pressures. Eighteen critically ill patients were recruited. Device performance was required to fulfil the Association for the Advancement of Medical Instrumentation (AAMI) protocol. Additionally, agreement between measurement methods was examined using Bland-Altman plots. Two-level linear mixed model analyses were under- taken. For each device, 150 paired measurements (arterial reference and device) were analysed. According to the AAMI protocol, no device measured systolic pressures accurately. One device measured diastolic pressures accurately. Integrated mean pressures were accurately measured by all devices. Overall, SunTech Medical's Oscar 2 performed best with mean pressure error not exceeding 17 mmHg. For this device, Bland-Altman plots showed uniform agreement across a wide range of mean pressures. Two-level linear mixed effects analyses showed that Oscar 2 mean error reduced during vasopressor use by (-) 3.9 mmHg (95% CI -5.9, -1.9; P < 0.001), and clinically, performance was little affected during mechanical ventilation. For the Oscar 2, there was up to (-) 7.0 mmHg (95% CI -10.3, -3.5; P < 0.001) l ess mean error during hypotension compared to normo- tension. In the Intensive Care Unit, the performance of one device, the Oscar 2, surpassed the others and fulfilled the AAMI protocol criteria for mean pressure measurement. This device is suitable for prehospital validation.