Construction of a nomogram to predict weaning-induced pulmonary edema in mechanically ventilated patients with cardiogenic respiratory failure.

Introduction: Weaning-induced pulmonary edema (WIPE) is a common but often underrecognized complication in patients undergoing mechanical ventilation for cardiogenic respiratory failure. This study aimed to construct and validate a risk prediction model to predict the occurrence of WIPE in this population. Methods: A retrospective analysis was conducted on 262 patients with cardiogenic respiratory failure who received invasive mechanical ventilation between April 2021 and December 2023. Clinical characteristics, physiological indicators, and comorbidities were compared to identify independent risk factors for WIPE. Results: Nine variables were identified as independent risk factors for WIPE: age, smoking history, APACHE II score, NYHA cardiac function grade, RSBI, duration of mechanical ventilation, history of hypertension, left ventricular diastolic dysfunction, and COPD. The constructed logistic regression model demonstrated excellent predictive performance (AUC = 0.884, sensitivity = 0.923, specificity = 0.811). The calibration curve showed good agreement between predicted and observed outcomes (MAE = 0.050). Discussion: This study developed a preliminary risk prediction model that has the potential to assist clinicians in early identification of patients at high risk of WIPE, but external validation in larger multicenter cohorts is warranted prior to clinical implementation.