Incident in-hospital sarcopenia after pancreaticoduodenectomy: A single-center retrospective observational study

Incident sarcopenia after pancreaticoduodenectomy (PD) can negatively impact postoperative recovery, but its occurrence during hospitalization has not been well documented. This study aimed to describe operationally defined incident in-hospital sarcopenia after PD and to explore associated preoperative characteristics.

We conducted a retrospective study of patients undergoing PD who did not have sarcopenia preoperatively. Incident in-hospital sarcopenia was defined using the Asian Working Group for Sarcopenia 2019 criteria. This postoperative definition was operational and descriptive and was not intended to indicate established chronic sarcopenia. Analyses were performed on two prespecified descriptive subcohorts: patients discharged within 30 postoperative days and assessed before discharge, and patients hospitalized for ≥ 30 days and assessed around postoperative days 28-35. Baseline, operative, and perioperative characteristics were summarized descriptively. Exploratory logistic regression models, including age, sex, a composite body-composition index, and prognostic nutritional index, were fitted separately.

Among 49 patients, operationally defined incident in-hospital sarcopenia occurred in 15 patients (30.6%) overall, including 26.3% in the discharge dataset and 33.3% in the postoperative day 28-35 dataset. Vascular resection was performed in 5 patients (10.2%). The mean operative time was 495.5 ± 85.5 minutes, and mean blood loss was 654.8 ± 415.1 mL. Exploratory models showed moderate in-sample discrimination, with area under the curves of 0.745 and 0.771, but coefficient estimates were imprecise.

Operationally defined incident in-hospital sarcopenia was observed in approximately one-third of patients after PD. These findings should be interpreted cautiously due to the small sample size, limited events, and the operational nature of the outcome definition.