Learn more about the benefits of registering on the new BJS website

Long‐term prognostic impact of surgical complications in the German Rectal Cancer Trial CAO/ARO/AIO‐94. BJS 2018; 105: 1510-1518.

Published: 30th May 2018

Authors: T. Sprenger, T. Beißbarth, R. Sauer, J. Tschmelitsch, R. Fietkau, T. Liersch et al.


The influence of postoperative complications on survival in patients with locally advanced rectal cancer undergoing combined modality treatment is debatable. This study evaluated the impact of surgical complications on oncological outcomes in patients with locally advanced rectal cancer treated within the randomized CAO/ARO/AIO‐94 (Working Group of Surgical Oncology/Working Group of Radiation Oncology/Working Group of Medical Oncology of the Germany Cancer Society) trial.


Patients were assigned randomly to either preoperative chemoradiotherapy (CRT) followed by total mesorectal excision (TME) or postoperative CRT between 1995 and 2002. Anastomotic leakage and wound healing disorders were evaluated prospectively, and their associations with overall survival, and distant metastasis and local recurrence rates after a long‐term follow‐up of more than 10 years were determined. Medical complications (such as cardiopulmonary events) were not analysed in this study.


A total of 799 patients were included in the analysis. Patients who had anterior or intersphincteric resection had better 10‐year overall survival than those treated with abdominoperineal resection (63·1 versus 51·3 per cent; P < 0·001). Anastomotic leakage was associated with worse 10‐year overall survival (51 versus 65·2 per cent; P = 0·020). Overall survival was reduced in patients with impaired wound healing (45·7 versus 62·2 per cent; P = 0·009). At 10 years after treatment, patients developing any surgical complication (anastomotic leakage and/or wound healing disorder) had impaired overall survival (46·6 versus 63·8 per cent; P < 0·001), a lower distant metastasis‐free survival rate (63·2 versus 72·0 per cent; P = 0·030) and more local recurrences (15·5 versus 6·4 per cent; P < 0·001). In a multivariable Cox regression model, lymph node metastases (P < 0·001) and surgical complications (P = 0·008) were the only independent predictors of reduced overall survival.


Surgical complications were associated with adverse oncological outcomes in this trial.

Full text

Your comments