Prehabilitative parenteral nutrition and surgery-related muscle loss in patients undergoing major upper gastrointestinal surgery: a single-centre observational cohort study.

INTRODUCTION: Surgery-related muscle loss (SRML) is common after major upper gastrointestinal (GI) surgery and is associated with worse recovery and survival, and prehabilitative parenteral nutrition (PN) has been proposed to attenuate it, yet data from Southeast Asian populations are absent. We set out to characterise perioperative muscle change and to test whether the amount of protein and energy delivered per kilogram body weight was associated with muscle preservation. METHODS: We conducted a single-centre observational cohort study of 115 adults who underwent elective major upper GI surgery at Hospital Sultanah Aminah, Johor Bahru, and received 800 kcal/day premixed prehabilitative PN (38 g amino acids), as a supplement to oral or enteral intake, for 7 days before surgery. Body composition was measured by segmental bioimpedance analysis (InBody 770) at admission and on postoperative days (PODs) 3, 7, 30, and 90. The primary outcome was clinically relevant SRML, defined as a decrease in skeletal muscle mass (SMM) of 10% or more from admission to POD 7; 111 patients had an evaluable POD 7 measurement. RESULTS: Patients had a median age of 61 years, 58.3% were male, 70.4% had malignant disease, and 19.1% were underweight; median total protein delivery (PN plus oral or enteral intake) was 1.4 g/kg/day. From admission to POD 7, SMM decreased by a mean of 2.6%, phase angle by 6.2% and body mass index by 3.1% (all p < 0.001). Clinically relevant SRML occurred in 18 of 111 evaluable patients (16.2%). Protein delivered per kilogram was not associated with the change in SMM (Spearman rho = 0.05, p = 0.59), and energy per kilogram showed only a weak, non-significant trend (Spearman rho = 0.16, p = 0.09). In exploratory multivariable models limited by only 18 events, only a higher baseline SMM predicted greater relative loss, and no delivery variable predicted SRML. DISCUSSION: In a real-world prehabilitative PN cohort, clinically relevant muscle loss was relatively infrequent, but protein delivery was not independently associated with muscle preservation, consistent with anabolic resistance during surgical stress. The single-arm design precludes causal inference, and the narrow range of protein exposure may have left the study underpowered to detect a dose-response relationship. Controlled studies with contemporaneous non-PN comparators, wider exposure contrasts and functional endpoints are required.