Partial Versus Complete Cholecystectomy: A Propensity Score-Matched Analysis of Complications

INTRODUCTION: Partial cholecystectomy is increasingly performed for complex gallbladder disease, yet comparative outcomes data remain limited by selection bias and inconsistent procedural coding. To address this gap, we compared postoperative complications between propensity-matched patients undergoing partial versus complete cholecystectomy. METHODS: This retrospective cohort study included 15,122 adult patients who underwent cholecystectomy from October 2010 to January 2025. Partial cholecystectomy cases (n = 107) were identified using a validated BioClinicalBERT natural language processing classifier applied to operative reports (95.9% test set accuracy). Propensity score matching (4:1) was adjusted for age, sex, race, ethnicity, and comorbidities, yielding a matched cohort of 518 patients (104 partial, 414 complete). Primary outcomes were overall complications, intensive care unit admission, and 30-d mortality. Secondary outcomes included bile duct injury, drain placement, postoperative endoscopic retrograde cholangiopancreatography, and ileus. Fisher's exact test was used for comparisons. RESULTS: Partial cholecystectomy was associated with five-fold higher 30-d mortality (3.8% versus 0.7%; 95% confidence interval [1.28, 20.21]; P = 0.033), 2.8-fold higher overall complication rates (75.0% versus 26.6%; P < 0.001), and 3.3-fold higher intensive care unit admission (8.7% versus 2.7%; P = 0.009). Bile duct injury (2.9% versus 0.0%, P = 0.008), drain placement (68.3% versus 19.8%, P < 0.001), postoperative endoscopic retrograde cholangiopancreatography (13.5% versus 1.4%, P < 0.001), and ileus (17.3% versus 2.9%, P < 0.001) were all significantly elevated. CONCLUSIONS: Partial cholecystectomy was associated with substantially higher postoperative complication rates even after propensity matching. These findings provide evidence-based information for surgical decision-making and informed consent discussions.