Clinical features and prognosis of anastomotic leak after anterior resection for rectal cancer following neoadjuvant chemoradiotherapy

Objective: To explore clinical features and prognosis of anastomotic leak (AL) after anterior resection following neoadjuvant chemoradiotherapy for rectal cancer patients. Methods: A retrospective cohort study was performed. Data were retrieved from colorectal cancer database of the Sixth Affiliated Hospital, Sun Yat-sen University. The clinical data of 470 patients with rectal cancer who underwent anterior resection after neoadjuvant chemoradiotherapy at our department from September 2010 to December 2018 were enrolled. Clinical features and outcome of postoperative AL were analyzed. The primary outcomes were the short-term and long-term incidence and severity of AL (ISREC grading standard was adopted). The secondary outcomes were the prognostic indicators of AL, including the secondary chronic presacral sinus, anastomotic stenosis and persistent stoma. Patients received regular follow-up every 3-6 months after surgery, including physical examination, blood test, colonoscopy and image; those received follow-up once a year after postoperative 2-year; those who did not return to our hospital received telephone follow-up. Data of this study were retrieved up to January 2020. Univariate χ(2) test and multivariate logistic analysis were used to identify risk factors of AL and prognostic factors of persistent stoma. Results: There were 331 males (70.4%) with the average age of (53.5±11.6) years. Distance from tumor to anal verge ≤ 5 cm was found in 228 (48.5%) patients. The diverting stoma was performed in 440 (93.6%) patients. After a median follow-up of 28 months, AL was found in 129 (27.4%) patients, including 67 (14.3%) patients with clinical leak (ISREC grade B-C). The median time for diagnosis of AL was 70 days (2-515 days) after index surgery. Common symptoms included sacrococcygeal pain (27.9%, 36/129), purulent discharge through anus (25.6%, 33/129), and rectal irritation (17.8%, 23/129). Sixty five point one percent (84/129) of the defect site was at the posterior wall of the anastomosis. Transanal incision and drainage or lavage (27.9%, 36/129) and percutaneous drainage under ultrasound or CT (17.1%, 22/129) were the most common management. Chronic presacral sinus tract could not be evaluated in 12 patients because imaging was performed more than 1 year after the operation. Evaluation beyond 1 year showed that 73 of 458 eligible patients (15.9%) were found with chronic presacral sinus, accounting for 62.4% (73/117) of patients with AL; 69 of 454 (15.2%) were diagnosed with anastomotic stenosis, of whom 49 were secondary to AL; 59 of 470 (12.6%) had persistent stoma due to AL. Univariate analysis showed that male, operative duration > 180 minutes, intraoperative blood loss >150 ml, and pelvic radiation injury were associated with AL (all P 150 ml (OR=1.82, 95% CI: 1.11-2.97, P=0.017), and pelvic radiation injury (OR=4.90, 95% CI: 3.09-7.76, P<0.001) were independent risk factors of AL after anterior resection. For patients with AL, clinical leak (ISREC grade B-C) (OR=9.59, 95% CI: 3.73-24.69, P<0.001), age ≤55 years (OR=3.35, 95% CI: 1.35-8.30, P=0.009), distance from tumor to anal verge ≤ 5 cm (OR=3.33, 95% CI: 1.25-8.92, P=0.017), and pelvic radiation injury (OR=3.29, 95% CI: 1.33-8.14, P=0.010) were independent risk factors of persistent stoma. Conclusions: AL after anterior resection following neoadjuvant chemoradiotherapy for rectal cancer patients is common. Among patients with AL, the proportion of those needing persistent stoma is high. Pelvic radiation injury is significantly associated with occurrence of AL and subsequent persistent stoma. Sphincter-preserving surgery for rectal cancer should be selectively used based on the risk of pelvic radiation injury, which is beneficial to reduce the incidence of AL and improve the quality of life.