K. Hu1, R. Wu2, A. Szabo2, T. Ridolfi1, K. Ludwig1, C. Peterson1 1Medical College Of Wisconsin,Surgery,Milwaukee, WI, USA 2Medical College Of Wisconsin,Biostatistics,Milwaukee, WI, USA
Introduction:
The robotic platform is increasingly utilized in colorectal surgery. Prior studies evaluating differences in outcomes associated with robotic versus laparoscopic approaches have been inconclusive, and debate remains regarding the use of robotics in colorectal surgery. In the past few years, there have been new upgrades in the robotic platform and associated technology, warranting updated evaluation. Additionally, in 2016, the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) introduced procedure-targeted reports for proctectomy, facilitating evaluation of proctectomy-specific outcomes. The aim of our study was to compare updated outcomes in laparoscopic versus robotic proctectomy using ACS-NSQIP data.
Methods:
Using the ACS-NSQIP dataset, we retrospectively identified adult patients undergoing elective robotic (RP) and laparoscopic proctectomy (LP) between 2016-2017. Demographics, baseline patient characteristics, intraoperative data, and post-operative outcomes were collected. The association between surgical approach and variables was assessed with Chi-square and Wilcoxon tests. Propensity scores were calculated for each outcome and weighted analysis was performed to estimate the effect of robotic versus laparoscopic surgery on outcomes.
Results:
We identified 3872 proctectomies (2701 LP and 1171 RP) that met inclusion criteria. Patients undergoing RP were more likely to have higher ASA scores of 3-4 (53.6% vs 48.2%, p=0.02), low rectal tumors (56.6% vs 48.8%, p<0.01), and have undergone chemotherapy (39.7% vs 27.5%, p<0.01) or radiation (37.9% vs 26.3%, p<0.01) prior to surgery. With propensity weighted analysis, RP was associated with decreased conversion to open operation (estimated mean difference (ED) -6.9% (95% CI -8.9 – -4.9), p<0.01), increased operative time (ED 20.3 minutes (12.2-28.5), p<0.01), decreased length of stay (ED -0.6 days (-1.0 – -0.2), p<0.01), decreased prolonged post-operative NPO status/need for nasogastric tube (ED -3.9% (-6.5 – -1.2), p<0.01), and decreased number of resected lymph nodes (ED -1.4 nodes (-2.4 – -0.3), p=0.01). There were no differences in other complication rates or surgical margin status after propensity adjustment.
Conclusion:
Using a national cohort, this study showed that compared to laparoscopy, RP is associated with decreased rates of conversion to open operation, longer operation time, decreased length of stay, and decreased post-operative prolonged NPO status. Robotic surgery was also associated with decreased number of resected lymph nodes, though this may not be clinically significant (mean 16.4 nodes (SD 11.7) for RP vs 18.4 (14.5) for LP). Limitations to our study include use of retrospective administrative data, possible selection bias, and lack of cost analysis, as ACS-NSQIP does not collect cost data. Further study is warranted to assess cost effectiveness in conjunction with clinical outcomes.