S. Groene1, C. Chandrasekera1, T. Prasad1, A. Lincourt1, B. T. Heniford1, V. Augenstein1 1Carolinas Medical Center,Division Of Gastrointestinal And Minimally Invasive Surgery,Charlotte, NC, USA
Introduction: Laparoscopy has become a common technique for many colorectal resections. Robotic-assisted colorectal surgery is also gaining traction. The aim of this study was to compare outcomes following colorectal resections based on surgical technique.
Methods: Review of the Carolinas Medical Center (CMC) NSQIP data for colorectal resections performed from 2013 to 2015 was conducted. Data were classified as open (O), laparoscopic (L) and robot-assisted (R) cases based on CPT codes and chart review. Demographics, pre-operative co-morbidities, minor and major complications were evaluated using standard statistical methods.
Results: There were 616 patients: 371 in L, 212 in O and 33 in R groups. There were no differences in age, BMI, race or gender. The groups differed in rates of pre-op sepsis (20.8% vs 0.8% vs 0% for O, L and R, respectively; p<0.001), contaminated wounds (53.3% vs 11.9% vs 3%; p<0.001), ASA class 4-5 (29.7% vs 8.7% vs 0%; p<0.001), emergent cases (31.1% vs 1.9% vs 0%; p<0.001), smokers (25.9 vs 15.6 vs 15.2; p=0.001), pre-op ventilator use (11.3% vs 0% vs 0%; p<0.001), and pre-op transfusions (12.7% vs 3.2% vs 0%; p<0.001). Also, there was a significant difference among the groups in post-op transfusions (34% vs 10.2% vs 3%; p<0.001), superficial SSI (9% vs 4.9% vs 0%; p=0.05), post-op patients on ventilators at 48 hours (18.4% vs 1.1% vs 3%; p<0.001) acute renal failure (1.9% vs 0% vs 3%; p=0.001), UTI (9% vs 4% vs 0%; p=0.02), DVTs (3.8% vs 0.8% vs 0; p=0.04), 30 day mortality (10.4% vs 0.5% vs 0%; p<0.001), post-op sepsis (17% vs 4% vs 6.1%; p<0.001) and LOS (13.2±10.1 vs 7.4±7.3 vs 8.7±8.6d; p<0.001). After controlling for wound and ASA class and pre-op sepsis, multivariate analysis indicated that O cases had significantly higher rates of superficial SSI [OR 0.5 (95% CI 0.2,0.9)], organ space infection [OR 0.2 (95%CI 0.07,0.5)], post-op transfusions [OR 0.5(95%CI 0.3,0.8)] post-op ventilator use [OR0.3 (95%CI 0.7,0.8)], sepsis [OR0.4 (95%CI 0.2,0.8)] and LOS (p<0.001) compared to L cases. There was not a significant difference in anastomotic leaks among the groups.
Conclusion: Patients who undergo open colorectal resections are sicker and have more post-operative complications, even when controlling for ASA, wound classification and pre-op sepsis. Laparoscopic and robotic colorectal resections have similar outcomes.