From the Hospices Civils de Lyon, Hôpital Louis Pradel, Service d'Anesthésie-Réanimation (MJ-L, EG, RS, PP, MR, J-LF), Laboratoire CarMeN, IHU OPERA, Inserm (MJ-L, J-LF), Université Claude Bernard Lyon 1, Faculté de médecine Lyon-Est, Lyon, France (MJ-L, MR, FA, J-LF), Hospices Civils de Lyon, Hôpital de la Croix-Rousse, Service d'Anesthésie-Réanimation.
Eur J Anaesthesiol. 2022 Jul 1;39(7):574-581. doi: 10.1097/EJA.0000000000001693. Epub 2022 Jun 10.
Hypotension prediction index (HPI) software is a proprietary machine learning-based algorithm used to predict intraoperative hypotension (IOH). HPI has shown superiority in predicting IOH when compared to the predictive value of changes in mean arterial pressure (ΔMAP) alone. However, the predictive value of ΔMAP alone, with no reference to the absolute level of MAP, is counterintuitive and poor at predicting IOH. A simple linear extrapolation of mean arterial pressure (LepMAP) is closer to the clinical approach.
Our primary objective was to investigate whether LepMAP better predicts IOH than ΔMAP alone.
Retrospective diagnostic accuracy study.
Two tertiary University Hospitals between May 2019 and December 2019.
A total of 83 adult patients undergoing high risk non-cardiac surgery.
Arterial pressure data were automatically extracted from the anaesthesia data collection software (one value per minute). IOH was defined as MAP < 65 mmHg.
Correlations for repeated measurements and the area under the curve (AUC) from receiver operating characteristics (ROC) were determined for the ability of LepMAP and ΔMAP to predict IOH at 1, 2 and 5 min before its occurrence (A-analysis, using the whole dataset). Data were also analysed after exclusion of MAP values between 65 and 75 mmHg (B-analysis).
A total of 24 318 segments of ten minutes duration were analysed. In the A-analysis, ROC AUCs to predict IOH at 1, 2 and 5 min before its occurrence by LepMAP were 0.87 (95% confidence interval, CI, 0.86 to 0.88), 0.81 (95% CI, 0.79 to 0.83) and 0.69 (95% CI, 0.66 to 0.71) and for ΔMAP alone 0.59 (95% CI, 0.57 to 0.62), 0.61 (95% CI, 0.59 to 0.64), 0.57 (95% CI, 0.54 to 0.69), respectively. In the B analysis for LepMAP these were 0.97 (95% CI, 0.9 to 0.98), 0.93 (95% CI, 0.92 to 0.95) and 0.86 (95% CI, 0.84 to 0.88), respectively, and for ΔMAP alone 0.59 (95% CI, 0.53 to 0.58), 0.56 (95% CI, 0.54 to 0.59), 0.54 (95% CI, 0.51 to 0.57), respectively. LepMAP ROC AUCs were significantly higher than ΔMAP ROC AUCs in all cases.
LepMAP provides reliable real-time and continuous prediction of IOH 1 and 2 min before its occurrence. LepMAP offers better discrimination than ΔMAP at 1, 2 and 5 min before its occurrence. Future studies evaluating machine learning algorithms to predict IOH should be compared with LepMAP rather than ΔMAP.
低血压预测指数（HPI）软件是一种专有的基于机器学习的算法，用于预测术中低血压（IOH）。与单独的平均动脉压（ΔMAP）变化的预测值相比，HPI 在预测 IOH 方面显示出优越性。然而，单独的ΔMAP 预测值，没有参考 MAP 的绝对水平，是违反直觉的，并且在预测 IOH 方面表现不佳。平均动脉压的简单线性外推（LepMAP）更接近临床方法。
我们的主要目的是研究 LepMAP 是否比单独的ΔMAP 更能预测 IOH。
回顾性诊断准确性研究。
2019 年 5 月至 2019 年 12 月期间的两家三级大学医院。
共 83 名接受高危非心脏手术的成年患者。
动脉压数据从麻醉数据采集软件中自动提取（每分钟一个值）。IOH 定义为 MAP<65mmHg。
为了预测 IOH，确定了 LepMAP 和 ΔMAP 在发生前 1、2 和 5 分钟的预测能力的重复测量相关性和接收器操作特征（ROC）曲线下面积（AUC）（A 分析，使用整个数据集）。还分析了排除 MAP 值在 65 和 75mmHg 之间的数据（B 分析）。
共分析了 24318 个十分钟长的片段。在 A 分析中，LepMAP 预测发生前 1、2 和 5 分钟 IOH 的 ROC AUC 分别为 0.87（95%置信区间，CI，0.86 至 0.88）、0.81（95% CI，0.79 至 0.83）和 0.69（95% CI，0.66 至 0.71），而单独的ΔMAP 为 0.59（95% CI，0.57 至 0.62）、0.61（95% CI，0.59 至 0.64）、0.57（95% CI，0.54 至 0.69）。在 LepMAP 的 B 分析中，这些分别为 0.97（95% CI，0.9 至 0.98）、0.93（95% CI，0.92 至 0.95）和 0.86（95% CI，0.84 至 0.88），而单独的ΔMAP 为 0.59（95% CI，0.53 至 0.58）、0.56（95% CI，0.54 至 0.59）、0.54（95% CI，0.51 至 0.57）。在所有情况下，LepMAP 的 ROC AUC 均显著高于ΔMAP 的 ROC AUC。
LepMAP 可在发生前 1 和 2 分钟可靠地实时连续预测 IOH。LepMAP 在发生前 1、2 和 5 分钟的鉴别能力优于ΔMAP。未来评估用于预测 IOH 的机器学习算法的研究应与 LepMAP 进行比较，而不是与ΔMAP 进行比较。
