ABSTRACT
Motion artifact tends to degrade oscillometric noninvasive blood pressure measurement (NIBP) accuracy and other aspects of performance (measurement time, patient comfort, false-positive readings). Medical personnel generally have not fully appreciated the extent of these degradations, in part because NIBP provides no waveform display to allow visualization of artifact disruption (unlike the electrocardiography (ECG) and pulse oximetry (SpO2) patient channels). More importantly, the magnitude and frequency of NIBP errors has also gone unappreciated because the auditory noise produced by transport vibration prevents accurate quantification of NIBP accuracy by the traditional auscultatory method. To overcome these problems, a commercially available NIBP simulator was modified to permit the superimposition of repeatable motion artifact waveforms from a function generator onto known patient blood pressure profiles available in the NIBP simulator. The superimposed artifact waveforms had been collected under transport conditions. This methodology enabled comparisons between artifact-free NIBP readings, on the one hand, and artifact-contaminated readings on the other. Monitors under test were subjected to multiple combinations of patient and artifact profiles. Measurement errors were expressed as a percent deviation of the artifact-contaminated readings from the expected (artifact-free) readings. Statistical analyses of the data compared the performance of the different monitor types with nonparametric tests of inference (Kruskal-Wallis H test, Mann-Whitney U test, and chi-squared test). These analyses demonstrated statistically significant differences in performance including accuracy, yield (incidence of values within various error categories), retries, measurement time, and false-positive readings under artifact-only conditions. The method further demonstrated that the monitor using ECG synchronization to filter motion artifact achieved statistically and clinically significant improvements in accuracy without compromising clinical expectations for measurement time. This approach provided a reproducible and quantifiable method by which to assess and differentiate the artifact tolerance of different NIBP technologies.
