Pericoronary fat attenuation index as a novel predictor of delayed extubation following lung transplantation: a retrospective cohort study

Background To evaluate whether preoperative pericoronary fat attenuation index (FAI) measured by coronary computed tomography angiography (CCTA) predicts delayed extubation following lung transplantation. Methods This retrospective study analyzed 208 consecutive lung transplant recipients who underwent preoperative CCTA. FAI was quantified in proximal 40 mm segments of all three coronary arteries using dedicated software. Delayed extubation was defined as mechanical ventilation > 72 h postoperatively. Optimal FAI thresholds were determined using ROC analysis. Multivariable logistic regression assessed independent associations. Results Delayed extubation occurred in 38 patients (18.3%). These patients demonstrated significantly higher FAI values in the right coronary artery (RCA) (− 62.1 ± 11.4 vs − 87.8 ± 11.7 HU, P < 0.001) and the left anterior descending artery (LAD) (− 64.5 ± 10.1 vs − 85.9 ± 10.5 HU, P < 0.001). Optimal thresholds were − 72.5 HU for RCA-FAI and − 71.5 HU for LAD-FAI. After adjusting for APACHE II score and other variables, RCA-FAI remained independently associated with delayed extubation (OR 1.29, 95%CI 1.08–1.53, P < 0.01). A three-vessel high FAI model achieved superior discrimination (AUC: 0.86) compared to single-vessel (AUC: 0.66) or two-vessel models (AUC: 0.73). Decision curve analysis confirmed clinical utility across relevant probability thresholds. Conclusions Preoperative FAI independently predicts delayed extubation following lung transplantation. Multi-vessel FAI assessment provides exceptional discriminative performance, offering a novel imaging biomarker for enhanced preoperative risk stratification that could guide perioperative management strategies.