Predictive value of the electrical impedance tomography-derived COVmax for weaning outcomes and 60-day mortality in mechanically ventilated patients: A single-center retrospective pilot study of 43 patients

Background To evaluate the prognostic performance of an electrical impedance tomography (EIT)-derived parameter—the maximum center of ventilation (COVmax)—for predicting weaning outcomes from invasive mechanical ventilation (IMV) and 60-day mortality in critically ill adults. Methods In this single-center retrospective cohort (December 2023–August 2025), 43 adults receiving IMV were included. EIT monitoring yielded the center of ventilation (COV) and global inhomogeneity (GI) indices. Associations between EIT parameters and (1) weaning outcome and (2) 60-day survival status were examined using binary logistic regression and Cox proportional hazards models, respectively. Multivariable models were adjusted separately for age, Charlson Comorbidity Index (CCI), and Acute Physiology and Chronic Health Evaluation II (APACHE II) score. Discrimination was assessed with receiver operating characteristic (ROC) analysis, including time-dependent ROC curve analysis for mortality. Results COVmax independently predicted both weaning failure and 60-day mortality. After adjustment for age, CCI, and APACHE II score, each 1-unit increase in COVmax was associated with lower odds of successful weaning (OR 0.86; 95% CI 0.75–0.99) and a higher risk of death at 60 days (HR 1.133; 95% CI 1.039–1.236). For the weaning outcome, the area under the curve (AUC) for COVmax combined with age was 0.687. In the time-dependent ROC curve analysis, the COVmax-plus-age model achieved an AUC of 0.754 for 60-day mortality. Conclusions EIT-derived COVmax is an independent predictor of weaning failure and 60-day mortality in mechanically ventilated patients. A higher COVmax suggests a compensatory shift of ventilation toward dependent (dorsal) lung regions, potentially reflecting severe underlying parenchymal disease. These pilot data support further large-scale prospective validation of COVmax as a prognostic marker in IMV and its integration into multimodal risk stratification. Trial registration This is a single-center, retrospective, observational study. As such, it did not require registration in a public clinical trials registry. The study protocol was approved by the Institutional Ethics Committee of Dongguan Hospital of Traditional Chinese Medicine (Approval No: SL [2025] 134).