RESUMEN
AIM: Despite transanal endoscopic microsurgery (TEM) being used for over 30 years, there has been slow adoption of this modality in many centres. There remains a paucity of research regarding the learning curve and early performance of surgeons who begin to offer TEM. We sought to determine predictors of longer rates of tumour excision and improvements in operative time in a newly established TEM programme. METHOD: All patients who underwent TEM at the Ottawa Hospital, Ottawa, Canada, between October 2009 and September 2014 were included. Data were abstracted through a retrospective chart review. The average rate of lesion excision (ARE) was calculated to standardize the operation time by size of the pathological specimen (min/cm3 ), representing a measure of surgical efficiency. Surgical efficiency was plotted using restricted cubic splines. Predictors of higher ARE were determined using multivariable regression. RESULTS: During the study period 108 patients underwent TEM. ARE was available for 95 patients of mean age 67.2 years. The mean ARE was 18.6 min/cm3 . On adjusting for important covariates, the ARE improved with each additional case until 16 cases were completed. Significant predictors of higher ARE on multivariable analysis were age < 50 years, experience of fewer than five cases, and carcinoid/gastrointestinal stromal tumour or scar histology. CONCLUSION: Operative efficiency appears to improve as surgeons completed 16 TEM cases. We have identified important factors that result in longer operating time. The study has important implications with regard to surgical training and operative planning for new TEM programmes.