(1) Background: The Hypotension Prediction Index (HPI) is an algorithm that predicts hypotension, defined as mean arterial pressure (MAP) less than 65 mmHg for at least 1 min, based on arterial waveform features. We tested the hypothesis that the use of this index reduces the duration and severity of hypotension during noncardiac surgery. (2) Methods: We enrolled adults having moderate- or high-risk noncardiac surgery with invasive arterial pressure monitoring. Participating patients were randomized 1:1 to standard of care or hemodynamic management with HPI guidance with a goal directed hemodynamic treatment protocol. The trigger to initiate treatment (with fluids, vasopressors, or inotropes) was a value of HPI of 85 (range, 0–100) or higher in the intervention group. Primary outcome was the amount of hypotension, defined as time-weighted average (TWA) MAP less than 65 mmHg. Secondary outcomes were time spent in hypertension defined as MAP more than 100 mmHg for at least 1 min; medication and fluids administered and postoperative complications. (3) Results: We obtained data from 99 patients. The median (IQR) TWA of hypotension was 0.16 mmHg (IQR, 0.01–0.32 mmHg) in the intervention group versus 0.50 mmHg (IQR, 0.11–0.97 mmHg) in the control group, for a median difference of −0.28 (95% CI, −0.48 to −0.09 mmHg; p = 0.0003). We also observed an increase in hypertension in the intervention group as well as a higher weight-adjusted administration of phenylephrine in the intervention group. (4) Conclusions: In this single-center prospective study of patients undergoing elective noncardiac surgery, the use of this prediction model resulted in less intraoperative hypotension compared with standard care. An increase in the time spent in hypertension in the treatment group was also observed, probably as a result of overtreatment. This should provide an insight for refining the use of this prediction index in future studies to avoid excessive correction of blood pressure.

Intraoperative hypotension is common and has been associated with adverse events. Although association does not imply causation, predicting and preventing hypotension may improve postoperative outcomes. This review summarizes current evidence on the development and validation of an artificial intelligence predictive algorithm, the Hypotension Prediction (HPI) (formerly known as the Hypotension Probability Indicator). This machine learning model can arguably predict hypotension up to 15 min before its occurrence. Several validation studies, retrospective cohorts, as well as a few prospective randomized trials, have been published in the last years, reporting promising results. Larger trials are needed to definitively assess the usefulness of this algorithm in optimizing postoperative outcomes.