Potosnak Willa, Dufendach Keith A, Nagpal Chirag, Kaczorowski David J, Yoon Pyongsoo, Bonatti Johannes, Miller James K, Dubrawski Artur W
Auton Lab, School of Computer Science, Carnegie Mellon University, Pittsburgh, Pennsylvania.
Division of Cardiac Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania.
Ann Thorac Surg Short Rep. 2024 Mar 7;2(3):336-340. doi: 10.1016/j.atssr.2024.02.005. eCollection 2024 Sep.
Intraoperative physiologic parameters could offer predictive utility in evaluating risk of adverse postoperative events yet are not included in current standard risk models. This study examined whether the inclusion of continuous intraoperative data improved machine learning model predictions for multiple outcomes after coronary artery bypass grafting, including 30-day mortality, renal failure, reoperation, prolonged ventilation, and combined morbidity and mortality (MM).
The Society of Thoracic Surgeons (STS) database features and risk scores were combined with retrospectively gathered continuous intraoperative data from patients. Risk models were developed for each outcome by training a logistic regression classifier on intraoperative data using 5-fold cross-validation. STS risk scores were included as offset terms in the models.
Compared with the STS Risk Calculator, models developed using a combination of the intraoperative features and the STS preoperative risk score had improved mean area under the receiver operating characteristic curve for prolonged ventilation (0.750 [95% CI, 0.690-0.809] vs 0.800 [95% CI, 0.750-0.851]) and MM (0.695 [95% CI, 0.644-0.746] vs 0.724 [95% CI, 0.673-0.775]). Additionally, models developed using intraoperative features had improved calibration, measured with Brier score, for prolonged ventilation (0.060 [95% CI, 0.050-0.070] vs 0.055 [95% CI, 0.045-0.065]) and MM (0.092 [95% CI, 0.081-0.103] vs 0.087 [95% CI, 0.075-0.098]).
The inclusion of time series intraoperative data in risk models may improve early postoperative care by identifying patients who require closer monitoring postoperatively.
术中生理参数在评估术后不良事件风险方面具有预测作用，但目前的标准风险模型中并未纳入这些参数。本研究探讨了纳入连续术中数据是否能改善机器学习模型对冠状动脉搭桥术后多种结局的预测，这些结局包括30天死亡率、肾衰竭、再次手术、通气时间延长以及合并症和死亡率（MM）。
将胸外科医师协会（STS）数据库特征和风险评分与回顾性收集的患者术中连续数据相结合。通过使用五折交叉验证在术中数据上训练逻辑回归分类器，为每个结局建立风险模型。STS风险评分作为偏移项纳入模型。
与STS风险计算器相比，使用术中特征和STS术前风险评分相结合开发的模型在通气时间延长（0.750 [95%CI，0.690 - 0.809] 对 0.800 [95%CI，0.750 - 0.851]）和MM（0.695 [95%CI，0.644 - 0.746] 对 0.724 [95%CI，0.673 - 0.775]）方面，受试者工作特征曲线下的平均面积有所改善。此外，使用术中特征开发的模型在通气时间延长（0.060 [95%CI，0.050 - 0.070] 对 0.055 [95%CI，0.045 - 0.065]）和MM（0.092 [95%CI，0.081 - 0.103] 对 0.087 [95%CI，0.075 - 0.098]）方面，以Brier评分衡量的校准有所改善。
在风险模型中纳入术中时间序列数据可能通过识别术后需要更密切监测的患者来改善术后早期护理。
