Abstract
Prolonged mechanical ventilation (PMV) is commonly associated with increased post-operative complications and mortality. Nevertheless, the predictive factors of PMV after lung transplantation (LTx) using extracorporeal membrane oxygenation (ECMO) as a bridge remain unclear. The present study aimed to develop a novel nomogram for PMV prediction in patients using ECMO as a bridge to LTx. A total of 173 patients who used ECMO as a bridge following LTx from January 2022 to June 2023 were divided into the training (122) and validation sets (52). A mechanical ventilation density plot of patients after LTx was then performed. The training set was divided in two groups, namely PMV (95) and non-prolonged ventilation (NPMV) (27). For the survival analysis, the effect of PMV was assessed using the log-rank test. Univariate and multivariate logistic regression analyses were performed to assess factors associated with PMV. A risk nomogram was established based on the multivariate analysis, and model performance was further assessed in terms of calibration, discrimination, and clinical usefulness. Internal validation was additionally conducted. The difference in survival curves in PMV and NPMV groups was statistically significant (P < 0.001). The multivariate analysis and risk factors in the nomogram revealed four factors to be significantly associated with PMV, namely the body mass index (BMI), operation time, lactic acid at T0 (Lac), and driving pressure (DP) at T0. These four factors were used to develop a nomogram, with an area under the curve (AUC) of 0.852 and good calibration. After internal validation, AUC was 0.789 with good calibration. Furthermore, goodness-of-fit test and decision-curve analysis (DCA) indicated satisfactory performance in the training and internal validation sets. The proposed nomogram can reliably and accurately predict the risk of patients to develop PMV after LTx using ECMO as a bridge. Four modifiable factors including BMI, operation time, Lac, and DP were optimized, which may guide preventative measures and improve prognosis.