THE LEARNING-CURVE FOR LAPAROSCOPIC COLORECTAL SURGERY - PRELIMINARY-RESULTS FROM A PROSPECTIVE ANALYSIS OF 1194 LAPAROSCOPIC-ASSISTED COLECTOMIES

Background: Laparoscopic-assisted colectomy is an emerging technology for patients with cancer, polyps, inflammation, and other types of pat hologic conditions. While previous studies have shown better outcomes for laparoscopic cholecystectomies when surgeons per form more procedu res, there is no information on the relationship between surgeon volum e and outcomes for laparoscopic-assisted colectomy. Objective: To eval uate whether better clinical outcomes are found for surgeons who perfo rm higher numbers of laparoscopic-assisted colectomies and whether suc h a relationship, if it exists, applies to both intraoperative and pos toperative outcomes. Design: Analysis of a data set of 1194 patients, operated on by 114 surgeons, from a prospective registry sponsored by the American Society of Colon and Rectal Surgeons, from May 1991 to Oc tober 1994. Main Outcome Measures: Completion rate, intraoperative and postoperative complications, and length of hospital stay. Results: In 75% of cases, surgery was completed laparoscopically, with no differe nce between high-volume surgeons (greater than or equal to 40 cases) a nd low-volume surgeons. Length of stay (average, 6 days) did not vary according to surgeon volume. Postoperative complications occurred in 1 5% of cases, with a significantly lower rate for high-volume surgeons (10% vs 19%; P<.001). Intraoperative complications occurred in 5% of c ases, with a nonsignificant trend toward a lower rate for high-volume surgeons (3.7% vs 6.3%). A multivariate regression analysis, adjusting for type of disease (cancer vs inflammation vs polyps) and for level of difficulty of the procedure (high vs low) showed that for high-volu me surgeons there is a lower probability of both intraoperative compli cations (adjusted odds ratio, 0.56; 95% confidence interval, 0.32-0.97 ; P=.04) and postoperative complications (adjusted odds ratio, 0.48; 9 5% confidence interval, 0.34-0.68; P<.001). Conclusions: There is a le arning curve for laparoscopic-assisted colectomy with respect to intra operative and postoperative outcomes. As with other laparoscopic proce dures, surgeons who perform higher volumes of laparoscopic-assisted co lectomy have lower rates of intraoperative and postoperative complicat ions.