METHODS: In 35 severely obese patients (median body mass index, 47 kg/m2), we simultaneously recorded noninvasive and invasive arterial pressure measurements over a period of 45 minutes. We compared noninvasive (test method) and invasive (reference method) arterial pressure measurements (sampling rate 1 Hz = 1/s) using Bland-Altman analysis (accounting for multiple measurements per subject), 4-quadrant plot/concordance analysis (2-minute interval, 5 mm Hg exclusion zone), and error grid analysis (calculating the proportions of measurements in risk zones A–E with A indicating no risk, B low risk, C moderate risk, D significant risk, and E dangerous risk for the patient due to the risk of wrong clinical interventions because of measurement errors).
RESULTS: We observed a mean of the differences (±SD, 95% limits of agreement) between the noninvasively and invasively assessed arterial pressure values of 1.1 mm Hg (±7.4 mm Hg, −13.5 to 15.6 mm Hg) for mean arterial pressure (MAP), 6.8 mm Hg (±10.3 mm Hg, −14.4 to 27.9 mm Hg) for systolic arterial pressure, and 0.8 mm Hg (±6.9 mm Hg, −12.9 to 14.4 mm Hg) for diastolic arterial pressure. The 4-quadrant plot concordance rate (ie, the proportion of arterial pressure measurement pairs showing concordant changes to all changes) was 93% (CI, 89%–96%) for MAP, 93% (CI, 89%–97%) for systolic arterial pressure, and 88% (CI, 84%–92%) for diastolic arterial pressure. Error grid analysis showed that the proportions of measurements in risk zones A–E were 89.5%, 10.0%, 0.5%, 0%, and 0% for MAP and 93.7%, 6.0%, 0.3%, 0%, and 0% for systolic arterial pressure, respectively.
CONCLUSIONS: During laparoscopic bariatric surgery, the accuracy and precision of the vascular unloading technique (Clearsight system) was good for MAP and diastolic arterial pressure, but only moderate for systolic arterial pressure according to Bland-Altman analysis. The system showed good trending capabilities. In the error grid analysis, >99% of vascular unloading technique–derived arterial pressure measurements were categorized in no- or low-risk zones.