Aim: Functional recovery after cardiovascular surgery is closely linked to long-term outcomes and readmission risk, but predicting it remains difficult in aging populations. This study aimed to assess how preoperative psoas muscle density (PMD), obtained from routine computed tomography (CT), adds predictive value beyond existing models. Methods: This retrospective study included 231 patients undergoing elective cardiovascular surgery. PMD was calculated as the mean CT attenuation of bilateral psoas muscles. Primary outcomes were delayed independent ambulation within 4 postoperative days and a ≥50-m decline in 6-minute walk distance (6 MWD) at postoperative day 14. The predictive benefit of adding PMD to a baseline model (sex, EuroSCORE II, handgrip strength, and Short Physical Performance Battery) was assessed using the area under the curve (AUC), cNRI, and IDI. Results: Delayed ambulation occurred in 47.1%, and 6 MWD decline in 38.5%. Lower PMD independently predicted delayed ambulation (odds ratio [OR], 1.15, 95% CI 1.07–1.24) and 6 MWD decline (OR 1.09, 95% CI 1.03–1.15). Adding PMD significantly increased the AUC for predicting delayed ambulation, and although the AUC improvement for 6 MWD decline was non-significant, cNRI and IDI improved significantly (P<0.05). Conclusions: Preoperative PMD provides strong, complementary predictive power for postoperative functional results and can be obtained without extra burden from routine CT scans. Including it in preoperative evaluation offers a practical pathway toward accurate risk stratification and targeted prehabilitation, ultimately enhancing patient-centered care.
View full abstract