Objective: To establish a scoring system to predict the risk of anastomotic leakage in patients with rectal cancer older than 60 years.
Methods: The study included 995 patients (≥60 years) with rectal cancer locating 3-12 cm from the anal verge who underwent anterior resection or intersphincteric resection at the Department of General Surgery, Henan Cancer Hospital from January 2012 to December 2016. Potential risk factors for leakage were subjected to univariate analysis. Multivariate logistic regression analysis was used to identify the independent risk factors for anastomotic leakage. The scoring system was developed based on regression coefficient for each significant risk factor. One point was allocated to the risk factor with a regression coefficient β<1, and two points were allocated to the risk factor with β>1. The proposed scoring system was tested by the area under curve (AUC) of the receiver operating characteristic curve (ROC).
Results: Surgery was successfully performed in all 995 patients. The incidence of anastomotic fistula was 4.6%(46/995). Among these 46 patients, 31 recovered after conventional treatment, and 13 patients underwent transverse colostomy,and 2 died of multiple organ failure. Independent risk factors included age (β=0.643, OR=1.902, 95%CI: 1.020-3.614, P=0.048), body mass index(BMI) (β=1.218, OR=3.379, 95%CI: 1.607-7.105, P=0.001), albumin levels (β=0.986, OR=2.681, 95%CI: 1.432-5.021, P=0.002), and level of anastomosis from the anal verge (β=1.395, OR=4.034, 95%CI: 2.086-7.801, P=0.000). The scoring system was created base on coefficient β of the independent risk factors (age≥70 years for 1, BMI≥25 kg/m2 for 2, albumin levels <35 g/L for 1, level of anastomosis from anal verge <4.0 cm for 2). All the scores were added up, and all patients were divided into the high-risk group(4-6 points, n=71) and intermediate-low-risk group(0-3 points, n=924) based on the scoring system. The incidence of anastomotic leakage in the two groups was 23.9%(17/71) and 3.1%(29/924), respectively (χ2=60.092, P=0.000). The AUC of age, BMI, albumin levels, and level of anastomosis from the anal verge were 0.598, 0.591, 0.622, and 0.635 respectively. The AUC of the scoring system was 0.656, which was higher than above parameters with a sensitivety of 0.37 and specificity of 0.94.
Conclusions: The scoring system is effective and accurate for identifying a subgroup at high risk for postoperative anastomotic leakage in rectal cancer patients over 60 years old.