Increasing the risk of late rectal bleeding after high-dose radiotherapy for prostate cancer: The case of previous abdominal surgery. Results from a prospective trial
Published on May 1, 2012in Radiotherapy and Oncology4.856
· DOI :10.1016/J.RADONC.2012.03.012
Purpose: To evaluate and discuss the role of specific types of abdominal surgery (SURG) before radical radiation therapy as a risk factor for late rectal toxicity in prostate cancer patients. Methods: Results concerning questionnaire-based scored late bleeding and faecal incontinence in 718 patients with a complete follow-up of 36 months were analysed, focusing on the impact of specific pre-radiotherapy abdominal/pelvic surgery procedures. Patients were accrued in the prospective study AIROPROS 0102. Different types of surgery (rectum-sigma resection, kidney resection, cholecystectomy or appendectomy) were considered as covariates together with a number of different parameters previously found to be predictive of late toxicity and including clinical as well as dosimetric parameters. Univariate (UVA) and multivariate (MVA) logistic analyses were carried out. Results: In total 69/718 patients were previously submitted to one or more surgical procedures, mostly cholecystectomy (n = 21) and appendectomy (n = 27). Actuarial incidences of G2–G3 and G3 bleeding were 52 (7.2%) and 24 (3.3%) respectively; 19 (2.6%) chronic incontinence events were registered. Results: Cholecystectomy was found to be highly correlated with late rectal bleeding at UVA: OR = 4.3 and p = 0.006 for G2–G3 and OR = 5.4 and p = 0.01 for G3. Considering MVA (including dosimetric and clinical factors), G2–G3 bleeding was significantly correlated to cholecystectomy (OR = 6.5, p = 0.002), V75Gy (OR = 1.074, p = 0.003) and secondarily with appendectomy (OR = 2.7, p = 0.10), presence of acute radioinduced rectal bleeding (OR = 1.70, p = 0.21) and androgen deprivation (OR = 0.67, p = 0.25). Results: Appendectomy (OR = 5.9, p = 0.004) and cholecystectomy (OR = 5.5, p = 0.016) were very strong predictors of G3 bleeding with V75Gy playing a less significant role (OR = 1.037, p = 0.26). Conversely, no specific surgery was correlated with actuarial or chronic incontinence. Conclusions: This analysis highlights previous SURG as the best predictor of late rectal bleeding. Among the different types of abdominal surgery, cholecystectomy and appendectomy play the major role, especially for severe late bleeding.