Perioperative outcomes of robotic-assisted gastrointestinal surgery during early program implementation: a retrospective single-center study
摘要
Robotic-assisted gastrointestinal (GI) surgery offers technical advantages over conventional minimally invasive approaches; however, evidence from low- and middle-income countries remains limited. This study evaluated the safety, feasibility, and early perioperative outcomes of implementing a robotic-assisted GI surgery program in a resource-constrained setting. A retrospective observational cohort study was conducted at a tertiary care center in Pakistan. Consecutive adult patients who underwent robotic-assisted GI procedures between January 2022 and October 2025 were included. Clinical, operative, and postoperative data were extracted retrospectively, with analysis performed in October 2025 following institutional ethical approval. Primary outcomes included perioperative morbidity, conversion rates, operative parameters, and short-term mortality. Postoperative complications were graded using the Clavien–Dindo classification, and multivariable logistic regression was used to identify predictors of postoperative complications. A total of 104 patients were analyzed (mean age 52.4 ± 15.2 years; 62.5% male). Procedures included cholecystectomy (40.4%), colectomy (26.0%), gastrectomy (24.0%), and hepatectomy (9.6%). Mean operative time was 158 ± 41 min, and mean estimated blood loss was 218 ± 98 mL. Conversion to open or laparoscopic surgery occurred in 5.8% of cases. Postoperative complications occurred in 15.4% of patients, predominantly Clavien–Dindo grades I–II. Seven patients (6.7%) required reoperation (Clavien–Dindo Grade IIIb). The 30-day readmission rate was 3.8%, reoperation rate was 6.7%, and 90-day mortality was 1.0%. No independent predictors of postoperative complications were identified. In this early, heterogeneous single-center cohort, robotic-assisted gastrointestinal surgery was implemented without excess perioperative morbidity or mortality. These findings represent preliminary feasibility data rather than definitive performance benchmarks.