Intraoperative hypotension (IOH) is associated with increased morbidity and mortality. Hypotension Prediction Index (HPI) is a machine learning derived algorithm that predicts IOH shortly before it occurs. We tested the hypothesis that the application of the HPI in combination with a pre-defined Goal Directed Therapy (GDT) hemodynamic protocol reduces IOH during major gynaecologic oncologic surgery. We enrolled women scheduled for major gynaecologic oncologic surgery under general anesthesia with invasive arterial pressure monitoring. Patients were randomized to a GDT protocol aimed at optimizing stroke volume index (SVI) or hemodynamic management based on HPI guidance in addition to GDT. The primary outcome was the amount of IOH, defined as the timeweighted average (TWA) mean arterial pressure (MAP) < 65 mmHg. Secondary outcome was the TWA-MAP < 65 mmHg during the first 20 min after induction of GA. After exclusion of 10 patients the final analysis included 60 patients (30 in each group). The median (25–75th IQR) TWA-MAP < 65 mmHg was 0.14 (0.04–0.66) mmHg in HPI group versus 0.77 (0.36–1.30) mmHg in Control group, P < 0.001. During the first 20 min after induction of GA, the median TWA-MAP < 65 mmHg was 0.53 (0.06–1.8) mmHg in the HPI group and 2.15 (0.65–4.2) mmHg in the Control group, P = 0.001. Compared to a GDT protocol aimed to SVI optimization, a machine learning-derived algorithm for prediction of IOH combined with a GDT hemodynamic protocol, reduced IOH and hypotension after induction of general anesthesia in patients undergoing major gynaecologic oncologic surgery.Trial registration number: NCT04547491. Date of registration: 10/09/2020.

BACKGROUND: Arterial hypotension is common after spinal anesthesia (SA) for cesarean delivery (CD), and to date, there is no definitive method to predict it. The hypotension prediction index (HPI) is an algorithm that uses the arterial waveform to predict early phases of intraoperative hypotension. The aims of this study were to assess the diagnostic ability of HPI working with arterial waveforms detected by ClearSight system in predicting impending hypotension in awake patients, and the agreement of pressure values recorded by ClearSight with conventional noninvasive blood pressure (NIBP) monitoring in patients undergoing CD under SA. METHODS: In this retrospective analysis of pregnant patients scheduled for elective CD under SA, continuous hemodynamic data measured with the ClearSight monitor until delivery were downloaded from an Edwards Lifesciences HemoSphere platform and analyzed. Receiver operating characteristic (ROC) curves were constructed to evaluate the performance of HPI algorithm working on the ClearSight pressure waveform in predicting hypotensive events, defined as mean arterial pressure (MAP) 1 minute. The sensitivity, specificity, positive predictive value, and negative predictive value were computed at the optimal cutpoint, selected as the value that minimizes the difference between sensitivity and specificity. ClearSight MAP values were compared to NIBP MAP values by linear regression and Bland-Altman analysis corrected for repeated measurements. RESULTS: Fifty patients undergoing CD were included in the analysis. Hypotension occurred in 23 patients (48%). Among patients experiencing hypotension, the HPI disclosed 71 alerts. The HPI predicted hypotensive events with a sensitivity of 83% (95% confidence interval [CI], 69–97) and specificity of 83% (95% CI, 70–95) at 3 minutes before the event (area under the curve [AUC] 0.913 [95% CI, 0.837–0.99]); with a sensitivity of 97% (95% CI, 92–100) and specificity of 97% (95% CI, 92–100) at 2 minutes before the event (AUC 0.995 [95% CI, 0.979–1.0]); and with a sensitivity of 100% (95% CI, 100–100) and specificity 100% (95% CI, 100–100) 1 minute before the event (AUC 1.0 [95% CI, 1.0–1.0]). A total of 2280 paired NIBP MAP and ClearSight MAP values were assessed. The mean of the differences between the ClearSight and NIBP assessed using Bland-Altman analysis (±standard deviation [SD]; 95% limits of agreement with respective 95% CI) was −0.97 mm Hg (±4.8; −10.5 [−10.8 to −10.1] to 8.5 [8.1–8.8]). CONCLUSIONS: HPI provides an accurate real time and continuous prediction of impending intraoperative hypotension before its occurrence in awake patients under SA. We found acceptable agreement between ClearSight MAP and NIBP MAP.

Intraoperative hypotension (IOH) is common during major surgery and is associated with a poor postoperative outcome. Hypotension Prediction Index (HPI) is an algorithm derived from machine learning that uses the arterial waveform to predict IOH. The aim of this study was to assess the diagnostic ability of HPI working with non-invasive ClearSight system in predicting impending hypotension in patients undergoing major gynaecologic oncologic surgery (GOS). In this retrospective analysis hemodynamic data were downloaded from an Edwards Lifesciences HemoSphere platform and analysed. Receiver operating characteristic curves were constructed to evaluate the performance of HPI working on the ClearSight pressure waveform in predicting hypotensive events, defined as mean arterial pressure < 65 mmHg for > 1 min. Sensitivity, specificity, positive predictive value and negative predictive value were computed at a cutpoint (the value which minimizes the difference between sensitivity and specificity). Thirty-one patients undergoing GOS were included in the analysis, 28 of which had complete data set. The HPI predicted hypotensive events with a sensitivity of 0.85 [95% confidence interval (CI) 0.73–0.94] and specificity of 0.85 (95% CI 0.74–0.95) 15 min before the event [area under the curve (AUC) 0.95 (95% CI 0.89–0.99)]; with a sensitivity of 0.82 (95% CI 0.71–0.92) and specificity of 0.83 (95% CI 0.71–0.93) 10 min before the event [AUC 0.9 (95% CI 0.83–0.97)]; and with a sensitivity of 0.86 (95% CI 0.78–0.93) and specificity 0.86 (95% CI 0.77–0.94) 5 min before the event [AUC 0.93 (95% CI 0.89–0.97)]. HPI provides accurate and continuous prediction of impending IOH before its occurrence in patients undergoing GOS in general anesthesia.