Hypotension Prediction Index versus High Mean Arterial Pressure Target for Preventing Intraoperative Hypotension: A Randomized Controlled Trial

BACKGROUND: The Hypotension Prediction Index (HPI) was developed to predict intraoperative hypotension several minutes before its occurrence. Although previous randomized trials reported reduced hypotension with HPI-guided management, these findings may partly reflect performance bias from open-label designs and selection bias in algorithm validation. Recent observational evidence suggests that much of HPI's apparent benefit may result from earlier treatment at higher mean arterial pressure (MAP) thresholds, supporting direct randomized comparison with a proactive higher MAP target. METHODS: In this open-label, parallel-group randomized controlled trial at two centers, 100 adults undergoing major noncardiac surgery requiring invasive arterial monitoring were equally randomized to HPI-guided management (treatment triggered when HPI ≥85) or a higher MAP-targeted strategy (treatment triggered when MAP ≤ 73 mmHg), both following the same predefined hemodynamic protocol. The primary endpoint was time-weighted average hypotension (MAP <65 mmHg), used to test superiority of HPI-guided management over a proactive higher MAP target. Secondary outcomes included area under the threshold for hypotension, hypertension burden (MAP >100 mmHg), norepinephrine dose, length of hospital stay, and 30-day mortality. RESULTS: Median (interquartile range) time-weighted average hypotension was 0.07 (0-0.20) mmHg in the HPI group and 0.16 (0.02-0.50) mmHg in the higher MAP target group (P = 0.119). The area under the threshold for hypotension was 22 (0-96) mmHg·min and 59.7 (5.7-119.3) mmHg·min, respectively (P = 0.172). Time-weighted average hypertension did not differ significantly between groups: 0.80 (0.50-1.80) versus 0.70 (0.20-1.98) mmHg (P = 0.555). Pre-specified secondary clinical outcomes, including norepinephrine dose, length of hospital stay and 30-day mortality, did not differ significantly. CONCLUSIONS: HPI-guided management was not superior to a proactive MAP ≤73 mmHg strategy. Because this trial was not designed to test equivalence or noninferiority, the findings indicate no demonstrated superiority rather than clinical equivalence between the trigger strategies.