Hugh Mackenzie, MBBS, BSc, MRCS, Melody Ni, Dr, Danilo Miskovic, Mr, Mark G Coleman, Mr, George B Hanna, Professor
Imperial College London
Introduction
A sign-off assessment was introduced to ensure that delegates of the English National Training Programme (lapco) are reaching a high operative standard before starting independent practice in laparoscopic colorectal surgery. This involves assessment of two video-recorded laparoscopic colorectal resections by at least two independent, blinded experts. The aim of this study was to identify types of operative errors predicting failure to pass the sign-off process.
Methods
Content analysis of the expert assessor comments from all sign-off assessments, submitted between November 2008 to May 2012 was performed. Two reviewers independently reviewed the assessors’ written comments in two rounds and determined a broad, primary error classification system. Each script was then examined for the existence of each type of error. Unifactorial analysis with Pearson’s chi-square test was performed to identify individual predictors of failure. Those with a p-value of <0.05 were included in a binomial multivariate logistic regression model, to predict whether the assessor deemed the case to be a pass (‘safe performance’) or a failure (‘needs more training’). Significant primary errors from the regression model were re-assigned into a more specific secondary error classification system (see table for details). The comments were then analysed for the presence of these ‘specific technical errors’ and these were included in a second level of unifactorial and regression analysis.
Results
Of the 208 submitted cases, 119 (57%) passed the assessment successfully, 89 (43%) failed. A total of 632 errors were identified; cases that passed sign-off had significantly fewer errors (1.82 vs 4.83, p<0.001). The most significant predictor of unsafe performance was if the video did not show the entire case, hence relevant information was missing [HR (hazard ratio) 8.724]. Other significant primary error categories are shown in the table below. The more specific secondary error categories that predicted failure were; poor stapling technique of the anastomosis, too distal dissection of vascular pedicle, inadequate extent of resection, damage to the mesorectum/ mesocolon, unclear tissue planes around the pedicle, being consistently out of tissue planes and iatrogenic small bowel injuries (see table for details).
Conclusions
This qualitative analysis has identified the significant critical errors that caused failure of the Lapco sign-off assessment procedure as those with potentially serious clinical and oncological consequences. These data have important implications on the design of future assessment systems for the evaluation of operative quality.
Table
| Primary Classification (only significant categories) | Secondary Classification | hazard ratio (HR) | p-value |
|---|---|---|---|
| Anastomosis Error | Poor Stapling of Anastomosis Tension on Anastomosis No check for proximal rotation | 11.556 6.087 * | 0.040 0.164 * |
| Oncologically Unsafe Dissection | Too Distal on the Pedicle Inadequate Resection | 2.858 5.761 | 0.048 0.021 |
| Ineffective Dissection | Ineffective/laborious dissection Insufficient Mobilisation | * 2.438 | * 0.375 |
| Out Of Plane | Damage Mesocolon/Mesorectum Unclear plane around Pedicle Breach Toldt’s fascia Consistently Out of Plane | 3.404 4.065 2.287 3.790 | 0.043 0.006 0.322 0.004 |
| Tissue Damage | Small Bowel Injury Rough Dissection | 4.823 2.500 | 0.001 0.235 |
* Chi-square value > 0.05 – therefore not included in the regression analysis
Session: Poster Presentation
Program Number: P173
