Impact of Thoracic Duct Resection on Postoperative Body Composition Trajectory After Oesophagectomy: A Prospective Cohort Study

Abstract

Background: Thoracic duct resection (TDR) is frequently performed during radical oesophagectomy to improve locoregional control in oesophageal squamous cell carcinoma (ESCC). However, its impact on postoperative body composition—particularly skeletal muscle mass—remains unclear. This study aimed to evaluate the extent and temporal pattern of postoperative changes in adiposity- and sarcopenia-related indices following TDR. Methods: In this prospective cohort study, 347 patients with ESCC who underwent curative oesophagectomy between May 2018 and June 2022 were included. Patients were classified into a TDR group (n = 288) and a thoracic duct preservation group (n = 59). Body composition was assessed using bioelectrical impedance analysis (BIA) at six time points: preoperatively and 1, 2, 3, 6 and 12 months postoperatively, yielding 1925 measurements. Metrics analysed included body mass index (BMI), fat mass index (FMI), skeletal muscle mass index (SMI) and fat-free mass index (FFMI). Sensitivity analysis was performed using 1:1 propensity score matching (PSM), based on key clinical variables. Results: Median age was 64 years, and ~90% of patients were male in both groups. Baseline operative variables were comparable between groups, including operative time (485 vs. 478 min), total lymph nodes (66 vs. 63) and complication rates (30% vs. 32%). BMI and FMI declined gradually over 12 months with no significant between-group differences (BMI at 12 months: TDR vs. preservation, 21.0 vs. 20.6 kg/m2; p = 0.809). In contrast, SMI and FFMI showed significant early postoperative declines, with more pronounced reductions in the TDR group during the first 3 months (SMI: −11.2% vs. −8.1%, p = 0.036). These early differences attenuated after PSM but remained directionally consistent. Recovery of muscle mass began around postoperative month 3, and by 12 months, sarcopenia-related indices were comparable between groups (SMI: p = 0.343; FFMI: p = 0.733). Subgroup analysis in patients with clinical stage I disease revealed similar patterns, suggesting that the observed muscle loss may reflect procedure-related effects, independent of tumour burden. Exploratory nutritional markers—including albumin, lymphocyte count and cholesterol—showed no significant intergroup differences at any interval. Conclusions: This is the largest study to date to assess longitudinal body composition changes after TDR using serial BIA. TDR was associated with a greater decline in sarcopenia-related indices, particularly within the first 3 months. These effects were transient, reversible and reproducible in early-stage patients. Our findings support the oncologic role of TDR while underscoring the importance of early nutritional and rehabilitative care. © 2026 The Author(s). Journal of Cachexia, Sarcopenia and Muscle published by Wiley Periodicals LLC.