LAUSR

Background Pericardial fat (PF) has not been considered a prognostic biomarker for overall survival (OS) in lung cancer. This study was designed to elucidate the impact of PF on prognosis of resected non-small cell lung cancer patients. Methods We retrospectively reviewed a total of 349 patients who underwent lung resection and received high-resolution computed tomography in our institute. PF volume was calculated. PF extended vertically from the diaphragm to the bifurcation of the right main pulmonary artery. Propensity score matched analysis was used to compare OS between the high- and low-PF groups. Results PF volume increased according to body mass index ( p  < 0.001). Receiver operating characteristics (ROC) curve analysis for 3-year OS showed the possibility of better predictivity of PF than body-mass index (area under the curve, 0.66 vs. 0.61, p  = 0.010). Cutoff level of PF volume was determined based on the ROC with 122 cm 3 . Five-year OS was poorer in the low-PF group (63.5% vs. 73.4%; p  = 0.002). After propensity score matching, each group consisted of 89 cases. Five-year OS was poorer in the low-PF group (66.5% vs. 82.7%; p  = 0.008). A Cox proportional hazards model showed low-PF volume was associated with poorer OS (hazard ratio, 2.14; p  = 0.009). The number of respiratory-related deaths was higher in the low-PF group (10/89 vs. 2/89, p  = 0.032). Conclusions Low-PF volume may be associated with poor OS with an increase in the number of respiratory-related deaths. Patients with low-PF volume require careful follow-up after surgery.