Robotic-Assisted Versus Thoracoscopic Lobectomy Outcomes From High-Volume Thoracic Surgeons

Background: Reports of surgical outcomes comparing proficient surgeons who perform either robotic-assisted or video-assisted thoracoscopic lobectomy are lacking. We evaluate the comparative effectiveness of robotic-assisted and video-assisted thoracoscopic lobectomies by surgeons who performed 20 or more annual surgical procedures in a national database. Methods: Patients 18 years or older, who underwent elective lobectomy by surgeons who performed 20 or more annual lobectomies by robotic-assisted or thoracoscopic approach from January 2011 through September 2015, were identified in the Premier Healthcare database with the use of codes from the ninth revision of the International Statistical Classification of Diseases and Related Health Problems. Propensity-score matching based on patient and hospital characteristics and by year was performed 1:1 to identify comparable cohorts for analysis (n = 838 in each cohort). All tests were two-sided, with statistical significance set at p less than 0.05. Results: A total of 23,779 patients received an elective lobectomy during the study period: 9,360 were performed by video-assisted thoracoscopic approach and 2,994 were by robotic-assisted approach. Propensity-matched comparison of lobectomies performed by surgeons who performed 20 or more procedures annually (n = 838) showed that robotic-assisted procedures had a longer mean operative time by 25 minutes (mean 247.1 minutes vs 222.6 minutes, p < 0.0001) but had a lower conversion-to-open rate (4.8% vs 8.0%, p = 0.007) and a lower 30-day complication rate (33.4% vs 39.2%, p = 0.0128). Transfusion rates and 30-day mortality rates were similar between the two cohorts. Conclusions: When surgical outcomes are limited to surgeons who perform 20 or more annual procedures, the robotic-assisted approach is associated with a lower conversion-to-open rate and lower 30-day complication rate when than video-assisted thoracoscopic surgeons, with a mean operative time difference of 25 minutes.