Predictors of Prolonged Length of Stay and Perioperative Antibiotic Use Patterns in General Surgery: A Single-Centre Observational Study.

BACKGROUND: Prolonged length of stay (LOS) is a key indicator of surgical quality, efficiency of care, and resource utilization. This study aimed to identify predictors of prolonged LOS and evaluate perioperative antibiotic use patterns among general surgical patients. METHODS: This prospective observational study was conducted at a Malaysian general hospital between January and October 2023. Adult patients undergoing elective or emergency general surgical procedures were consecutively enrolled. Prolonged LOS was defined as total hospitalization exceeding the cohort median (>3 days). Multivariate logistic regression was used to identify independent predictors, with full model diagnostics evaluated. RESULTS: Among 405 patients, 35.8% experienced prolonged LOS. Gastrointestinal (GI) surgery was the strongest independent predictor of prolonged stay (adjusted odds ratio [AOR] 85.76, 95% CI: 10.94-672.34). Other independent predictors included C-reactive protein (CRP) testing (AOR 5.03, 95% CI: 1.99-12.69), positive culture and sensitivity (C&S) results (AOR 2.52, 95% CI: 1.22-5.23), longer operative duration (AOR 1.02 per minute, 95% CI: 1.01-1.03), and greater intraoperative blood loss (IBL) (AOR 1.00 per mL, 95% CI: 1.00-1.01). In contrast, intraoperative cefazolin use (AOR 0.08, 95% CI: 0.02-0.39) and absence of postoperative antibiotics (AOR 0.41, 95% CI: 0.21-0.79) were favourable markers associated with shorter hospitalization. CONCLUSIONS: Prolonged LOS in general surgery is primarily driven by disease severity and surgical complexity. Adherence to guideline-recommended surgical antimicrobial prophylaxis and the absence of need for postoperative antibiotics are favourable markers of uncomplicated recovery, supporting health system efficiency and antimicrobial stewardship goals.