Learning curve of robotic transversus abdominis release in ventral hernia repair: a cumulative sum (CUSUM) analysis

Abstract

Listen

Translate article

Despite heightened interest in robotic transversus abdominis release (rTAR), concerns over its steep learning curve (LC) and associated challenges may limit its adoption. This study defines the operative time and morbidity-based LC of a single surgeon's experience with rTAR. A retrospective analysis of patients undergoing rTAR over an 8-year period was conducted. Consecutive ventral and incisional hernia repairs were stratified into four sub-categories based on bilaterality and complexity, with complex hernias being defined as those > 10cm. Cumulative sum analyses (CUSUM) were used to evaluate skin-to-skin time and morbidity LCs. This study included a total of 156 rTARs with a mean skin-to-skin time of 222.8min. Mean skin-to-skin times (min) for sub-categories were as follows: unilateral non-complex (137.6), bilateral non-complex (206.8), unilateral complex (241.9), and bilateral complex (298.6). The CUSUM-LC was obtained by summing the differences between each procedure's operative time and its sub-category mean, revealing a quadratic best-fit line maximum at case 49 and a transition point between early and late phases at case 75. Although skin-to-skin times between early and late phases did not differ significantly (235.3 vs 211.2, respectively; p = 0.12), a significant difference was found in console times. Overall postoperative complications also decreased significantly from early to late phases (41.3% vs 25.9%; p = 0.041). Postoperative complications were predicted by a history of wound infection (c = 0.61). This study reveals that the rTAR LC was overcome between 49 and 75 cases, after which, console time and postoperative complications decreased significantly.