Longitudinal Changes in CT-Derived Vertebral Trabecular Attenuation and Psoas Muscle Density After Liver Transplantation for Hepatocellular Carcinoma

Background and Objectives: Hepatocellular carcinoma (HCC) is a major indication for liver transplantation (LT); however, longitudinal musculoskeletal changes detectable by computed tomography (CT) during the first postoperative year remain insufficiently characterized. The aim of this study is to compare baseline CT-derived bone and muscle parameters between patients undergoing LT for HCC and controls and to evaluate longitudinal changes during the first postoperative year. Materials and Methods: This retrospective study included 64 patients who underwent LT for HCC and had technically adequate non-contrast abdominal CT examinations at baseline, postoperative day 180 (POD180), and postoperative day 365 (POD365). The control group comprised 64 patients undergoing non-contrast abdominal CT for urinary stone evaluation. CT-derived measures included L1–L4 vertebral trabecular attenuation, psoas muscle area (PMA), the psoas muscle index (PMI), and psoas muscle density (PMD). Repeated-measures analyses and adjusted linear mixed-effects models evaluated longitudinal changes within the HCC cohort. Results: At baseline, vertebral trabecular attenuation was higher in the HCC group than in controls (151 ± 36 vs. 137 ± 39, p = 0.037), whereas crude PMI, PMA, and PMD did not differ significantly. In hierarchical multivariable analysis, the between-group difference in vertebral attenuation was attenuated after adjustment, while age and baseline PMI remained independently associated with L1–L4 attenuation. In adjusted multivariate analysis, baseline PMD was higher in the HCC group than in controls (51.5 vs. 44.4, p < 0.001). Longitudinally, L1–L4 attenuation declined from baseline to POD180 and remained low at POD365 (overall p < 0.001), with no difference between POD180 and POD365. PMD also decreased over time (overall p < 0.001), whereas PMI and PMA showed no significant change. Adjusted mixed-effects analyses showed that these postoperative reductions persisted after adjustment for age, BMI, albumin, calcium, and ALP. The adjusted PMD model showed an additional decrease between POD180 and POD365 (mean difference, 1.411 HU; 95% CI, 0.065–2.757; Bonferroni-adjusted p = 0.037). Older age remained independently associated with lower vertebral attenuation and PMD. Conclusions: During the first year after LT for HCC, CT-derived vertebral trabecular attenuation and PMD showed significant longitudinal decreases, while psoas muscle quantity remained relatively stable. Muscle quality and vertebral attenuation may therefore provide earlier imaging signals of post-transplant musculoskeletal deterioration than area-based muscle-quantity measures.