Impact of hypotension prediction index-guided management on intraoperative hypotension and postoperative outcomes in abdominal surgery: A meta-analysis of randomized controlled trials

Abstract

Background: The Hypotension Prediction Index (HPI) is a machine-learning algorithm designed to predict hypotension. by maintaining mean arterial pressure (MAP) above 65 mmHg. This meta-analysis evaluated whether HPI-guided management improves postoperative outcomes and included post hoc analyses of intraoperative hypotension (IOH) metrics in adults undergoing major abdominal surgery. Methods: A comprehensive search of PubMed, EMBASE, and Cochrane databases identified randomized controlled trials comparing HPI-guided management with standard care. Primary outcomes were postoperative complications, acute kidney injury (AKI), perioperative mortality, and hospital length of stay (LOS). Post hoc analyses assessed IOH metrics, including time-weighted average (TWA) of MAP < 65 mmHg, area under the threshold (AUT), total time with MAP < 65 mmHg, and intraoperative fluid use. Meta-analyses were conducted using random-effects models to calculate pooled standardized mean differences (SMDs), odds ratios (ORs), and mean differences (MDs). Results: Eight trials involving 1534 patients were included. No significant differences were observed for AKI (OR: 0.85; 95% CI: 0.64-1.13), postoperative complications (OR: 1.10; 95% CI: 0.83-1.46), mortality (OR: 0.96; 95% CI: 0.32-2.83), LOS (SMD: -0.15; 95% CI: -0.73 to 0.42), or fluid use (SMD: -0.06; 95% CI: -0.35 to 0.24). HPI reduced TWA MAP < 65 mmHg (SMD: -0.25; MD: -20.5 min), AUT (SMD: -0.83), and total time with MAP < 65 mmHg (SMD: -0.74). Conclusions: HPI-guided management did not significantly improve patient-centered outcomes. Post hoc analyses indicated a reduction in IOH metrics, but the clinical relevance of these findings remains uncertain given the lack of blinding and high risk of bias.