Individual surgeon is an independent risk factor for leak after double-stapled colorectal anastomosis: An institutional analysis of 800 patients

Autores de IIS La Fe

• Eduardo García-Granero Ximénez

  Autor

• Carlos Cerdán Santacruz

  Autor

• 

  Matteo Frasson

  Autor

• Álvaro García-Granero García-Fuster

  Autor

• Franco Guillermo Marinello

  Autor

• Blas Flor Lorente

  Autor

• Alejandro Espi Macias

  Autor

Participantes ajenos a IIS La Fe

• Navarro, F

Grupos

• Cirugía digestiva y Cuidados perioperatorios

  

  Leader

  Matteo Frasson

• Grupo de Investigación Clínica y Traslacional en Cardiología

  

  Investigador Principal

  Esther Roselló Lletí

Abstract

Background. Our aim was to assess whether the individual surgeon is an independent risk factor for anastomotic leak in double-stapled colorectal anastomosis after left colon and rectal cancer resection. Methods. This retrospective analysis of a prospectively collected database consists of a consecutive series of 800 patients who underwent an elective left colon and rectal resection with a colorectal, double-stapled anastomosis between 1993 and 2009 in a specialized colorectal unit of a tertiary hospital with 7 participating surgeons. The main outcome variable was anastomotic leak, defined as leak of luminal contents from a colorectal anastomosis between 2 hollow viscera diagnosed radiologically, clinically, endoscopically, or intraoperatively. Pelvic abscesses were also considered to be an anastomotic leak. Radiologic examination was performed when there was clinical suspicion of leak. Results. Anastomotic leak occurred in 6.1 % of patients, of which 33 (67%) were treated operatively, 6 (12%) with radiologic drains, and 10 (21 %) by medical treatment. Postoperative mortality rate was 2.9% for the whole group of 800 patients. In patients with anastomotic leak, mortality rate increased up to 16% vs 2.0% in patients without anastomotic leak (P <.0001). At multivariate analysis, rectal location of tumor, male sex, bowel obstruction preoperatively, tobacco use, diabetes, perioperative transfusion, and the individual surgeon were independent risk factors for anastomotic leak. The surgeon was the most important factor (mean odds ratio 4.9; range 1.0 to 13.5). The variance of anastomotic leak between the different surgeons was 0.56 in the logit scale. Conclusion. The individual surgeon is an independent risk factor for leakage in double-stapled, colorectal, end-to-end anastomosis after oncologic left-sided colorectal resection.