Clinical impact of postoperative changes in the psoas muscle mass and prognostic nutritional index for patients with biliary tract cancer: a retrospective cohort study

Abstract Background Low preoperative Psoas Muscle Index (PMI) and Prognostic Nutritional Index (PNI) values in patients with biliary tract cancer (BTC) indicate unfavorable prognoses. However, the impact of changes in PMI and PNI from pre- to postoperative on survival outcomes remains unclear. We sought to clarify whether changes in skeletal muscle mass and PNI could affect the long-term postoperative prognosis of patients with BTC who underwent curative surgery. Methods We retrospectively reviewed 168 patients who underwent BTC resection. The PMI was assessed within 1 month before surgery, and the PNI was calculated from blood samples collected within 2 weeks before surgery; postoperative assessments were performed at 4–6 months (PMI) and 6 months (PNI) after surgery. We classified patients into a large decrease and a small decrease/increase group, according to the percentage change in PMI and PNI, using the optimal cut-off point. Associations between clinicopathological characteristics, including changes in the PMI and PNI, and survival outcomes were analyzed. Results Preoperative PMI values were similar in both groups. The decrease in PNI was significantly greater in the large- than in the small-PMI-decrease group ( P = 0.004). T classification ≥ 3 (hazard ratio [95% confidence interval]) (2.11 [1.29–3.47]; P = 0.003), lymph node metastasis (1.84 [1.10–3.09]; P = 0.002), and a large PNI (≥ 11.0%; 2.14 [1.27–3.54]; P = 0.005), and PMI (≥ 6.6%; 2.00 [1.18–3.50]; P = 0.009) decrease independently predicted overall survival in multivariate analysis. T classification ≥ 3 (1.72 [1.02–2.91]; P = 0.038), lymph node metastasis (1.77 [1.03–3.05]; P = 0.039), and a large PMI decrease (≥ 6.6%; 1.90 [1.08–3.47]; P = 0.026) were independent predictors of RFS in the multivariate analysis. Conclusions Large postoperative reductions in the PMI and PNI were independent predictors of poor survival, regardless of the preoperative sarcopenia status of patients with resectable BTC.