Background: Acute respiratory and circulatory collapse might occasionally happen after pulmonary endarterectomy (PEA). We aimed to investigate the utilization of extracorporeal membrane oxygenation (ECMO) after PEA and potential risk factors.
Methods: Demographic characteristics, clinical and surgical data were collected for all patients who underwent PEA from December 2016 to June 2022. All factors were compared between patients in the ECMO group and those in the other group. The most characteristic risk factors were obtained by least absolute shrinkage and selection operator regression and support vector machine machine learning, and receiver operating characteristics (ROC) Curve analysis was performed to verify the diagnostic value of the obtained risk factors.
Results: A total of 117 patients underwent PEA, and 8 (6.8%) of them received ECMO treatment intraoperatively or postoperatively. There were significant differences between the two groups in terms of cardiac function, pulmonary vascular resistance (PVR), preoperative inflammation and cardiopulmonary bypass time. The PVR and neutrophil-to-lymphocyte ratio (N/L ratio) were the most characteristic risk factors with an area under the ROC curve of 0.847 (95% confidence interval [CI] = 0.7517-0.9420, p = .005) and 0.896 (95% CI = 0.803-0.989, p = .001), respectively. The ECMO group had higher PVR (1549.4 ± 600.7 vs. 952.9 ± 466.9 dyn.s.cm-5 , p = .004) and N/L ratio (6.3 ± 5.6 vs. 2.4 ± 1.7, p = .001).
Conclusions: PVR and N/L ratio can correctly predict who is likely to receive ECMO treatment after PEA. Therefore, addressing the preoperative inflammatory status might be beneficial but further research is needed.
Keywords: chronic thromboembolic pulmonary hypertension; extracorporeal membrane oxygenation; machine learning; pulmonary endarterectomy; risk factors.
© 2022 Wiley Periodicals LLC.