Predictive factors of stenosis after stapled colorectal anastomosis: prospective analysis of 179 consecutive patients.

ABSTRACT

The incidence, risk factors, and clinical relevance of stenosis of stapled colorectal anastomosis (CRA) were studied prospectively. Anastomotic stricture was defined as the inability of traversing the anastomosis with the rigid proctoscope. The population studied consisted of 179 patients (94 males) with an average age of 59.3 years (range 20 to 91 years). The main indication for surgery was colorectal cancer in 59% of the cases, followed by diverticular disease in 23%. The first endoscopic control was performed before 4 months in 25% of the patients, between 5 and 10 months in 50%, and during the following 10 months in 25%. Stenosis was verified with the rigid instrument in 21.1% of the cases and with the flexible colonoscope in 4.4%. The barium enema performed in 12 cases confirmed a punctiform stenosis in 5 patients, 4 of whom had been asymptomatic. An endoscopic dilatation was performed on 5 of the 8 symptomatic patients, with one relapse that required an additional dilatation. In the univariate analysis only the lesser 4-month interval was statistically significant (p = 0.033; odds ratio (OR) = 2.3; confidence interval (CI) 95% = 1.06 to 4.97). Male patients (p = 0.057; OR = 2.08; IC 95% = 0.97-4.44) show a tendency to CRA stricture that does not reach statistically significant levels. In the multivariate analysis, only sex (p = 0.04; OR = 4.11; IC 95% = 1.03 to 5.41) and the time interval (p = 0.012; OR = 2.87; IC 95% = 1.25 to 6.57) appear as independent variables in stenosis risk of a stapled CRA. The incidence of this complication depends on the criteria used for defining it. It is clinically relevant in no more than 5% of the patients. Five out of eight patients in category II were treated successfully with an endoscopic dilatation, while the other three improved spontaneously. Early stenosis, although frequent, is generally asymptomatic and disappears spontaneously. Considering the lack of correlation between the degree of stricture and its symptomatology, it is convenient to combine both the anatomic and the clinical criteria in the selection of candidates for an eventual therapeutic procedure.