Prediction of ventilator weaning failure in postoperative cardiac surgery patients using vasoactive-ventilation-renal score and nomogram analysis

Front Cardiovasc Med. 2024 Mar 14:11:1364211. doi: 10.3389/fcvm.2024.1364211. eCollection 2024.

Abstract

Objective: This study evaluated the predictive value of the vasoactive-ventilation-renal (VVR) score in identifying the risk of weaning failure after cardiac surgery and developing a nomogram model to help physicians improve the success rate of weaning from mechanical ventilation in adult patients undergoing postoperative cardiac surgery.

Methods: Clinical data were retrospectively collected from adult patients who underwent extracorporeal circulation cardiac surgery at the First Affiliated Hospital of Nanjing Medical University between August 2022 and April 2023 and who were subsequently transferred to the Intensive Care Unit (ICU) and treated with vasoactive drugs. Patients were divided into successful and unsuccessful weaning groups based on first-attempt weaning success. Variable selection was regularized using univariate logistic regression and Least absolute shrinkage and selection operator (LASSO) regularization. Multivariate logistic regression was performed to identify predictors and a nomogram was created to predict the risk of weaning failure.

Results: A total of 519 patients were included in the study. After selecting multiple stepwise variables, the VVR score before weaning, the modified Sequential Organ Failure Assessment (mSOFA) score on weaning day, and mechanical ventilation duration before weaning were determined as predictive indicators of weaning failure in adult patients after cardiac surgery. The optimal cut-off values for these indicators were 18.46 points, 4.33 points, and 20.50 h, respectively. The predictive model constructed using these three factors demonstrated good predictive performance.

Conclusions: The VVR score before weaning accurately predicts the probability of weaning failure in adult patients after cardiac surgery. The weaning risk-predictive nomogram model, established based on the VVR score, mSOFA score, and mechanical ventilation duration before weaning, demonstrated robust predictive ability.

Keywords: cardiac surgery; mechanical ventilation; predictive modelling; vasoactive-inotropic score; vasoactive-ventilation-renal score; weaning failure.

Grants and funding

The authors declare financial support was received for the research, authorship, and/or publication of this article.