The single surgeon learning curve of laparoscopic liver resection

The aim of the study was to evaluate the single-surgeon learning curve (SSLC) in laparoscopic liver surgery over an 11-year period with risk-adjusted (RA) cumulative sum control chart analysis.Laparoscopic liver resection (LLR) is a challenging and highly demanding procedure. No specific data are available for defining the feasibility and reproducibility of the SSLC regarding a consistent and consecutive caseload volume over a specified time period.A total of 319 LLR performed by a single surgeon between June 2003 and May 2014 were retrospectively analyzed. A difficulty scale (DS) ranging from 1 to 10 was created to rate the technical difficulty of each LLR. The risk-adjusted cumulative sum control chart (RA-CUSUM) analysis evaluated conversion rate (CR), operative time (OT) and blood loss (BL). Perioperative morbidity and mortality were also analyzed.The RA-CUSUM analysis of the DS identified 3 different periods: P1 (n = 91 cases), with a mean DS of 3.8; P2 (cases 92-159), with a mean DS of 5.3; and P3 (cases 160-319), with a mean DS of 4.7. P2 presented the highest conversion and morbidity rates with a longer OT, whereas P3 showed the best results (P < 0.001). Fifty cases were needed to achieve a significant decrease in BL. The overall morbidity rate was 13.8%; no perioperative mortality was observed.According to our analysis, at least 160 cases (P3) are needed to complete the SSLC performing safely different types of LLR. A minimum of 50 cases can provide a significant decrease in BL. Based on these findings, a longer learning curve should be anticipated to broaden the indications for LLR.