Risk Factors for Anastomotic Leakage after Low Anterior Resection

Introduction: Low anterior resection (LAR) is a standard surgical procedure for distal rectal carcinoma that allows creation of a colorectal anastomosis, with anal sphincter preservation and permanent colostomy avoidance. Anastomotic leakage (AL) is a potential complication of LAR present in 3% to 20% of cases leading to a significant increase in postoperative morbidity. AL represents a communication between the intra luminal and extra luminal compartments caused by a violation of the integrity of the intestinal wall. The risk factors of this procedure have been discussed and still remain a controversial issue. Prevention is the best management. Patients with predisposing factors should be paid special attention. Aim: Our objective was to identify and analyze the risk factors of anastomotic leakage after low anterior resection of rectal carcinoma. Patients and methods: This study included all patients who underwent low anterior resection for rectal cancer at the Department of Surgery of University Hospital Kaspela between 2011 and 2016. Results: Low anterior resections were performed in 141 patients during the study period. Due to a positive air leak test on 16 patients, a protective ileostomy was created and these patients were eliminated from the study. The sex distribution of the remaining 125 patients was 69 men (55.2%) and 56 women (44.8%). The height of the anastomosis from the dentate line was 3.0 cm on an average (range 2.5 to 4 cm). Conclusion: Anastomotic leakage remains the most feared and serious complication after low anterior resection surgery. The evidence suggests that the main risk factors for anastomotic leakage at low anterior resection are the height of the anastomosis and the high ASA score.


INTRODUCTION
Low anterior resection (LAR) is a standard surgical procedure for distal rectal carcinoma that allows the creation of anastomosis and anal sphincter preservation. 1 Anastomotic leakage (AL) is a potential complication of LAR that is present in 3% to 20% of cases, leading to a significant increase in postoperative morbidity. 2 AL is a communication between intraluminal and extraluminal compartments of the intestinal wall caused by a violation of the integrity of the intestinal wall in the zone of attachment between the colon and the rectum. 3 There are still controversies over the risk factors of this complication. 4 Prevention is believed to be the best approach when close attention is paid to patients with risk factors. 5

AIM
Our objective was to identify and analyze the risk factors for anastomotic leakage after low anterior resection for treatment of rectal carcinoma

PATIENTS AND METHODS
This study includes all patients who underwent low anterior resection for rectal cancer at the Department of Surgery of University Hospital Kaspela between 2011 and 2016. The patients were monitored from their hospital discharge day to one month after.
In all patients, intestinal tract preparation and bowel cleaning was achieved using four sachets of "Fortrans". Pre-operative corrections of the imbalance in fluids, electrolytes and protein were accomplished, as well as prophylaxis with low molecular heparin and antibiotics of the cephalosporin group. Surgeries were performed by highly qualified team of surgeons. Conventional LAR was performed in 37 patients and laparoscopic LAR in 104 patients. The surgical technique was consistent with the oncological requirements. In all cases we performed a total mesorectal excision (TME) with medial-to-lateral approach, low ligation of the inferior mesenteric vessels and a partial or full mobilization of the left flexure to reach the required colon length avoiding the occurrence of tension on the anastomosis. The distal rectum was divided by articulating Endo GIA™ Articulating Medium/Thick Reload with Tri-Staple™ (Cincinnati, OH, USA) in laparoscopic surgery or with the ТА-45 TA™. Single use reloadable staplers and reloads were used in conventional or laparoscopic assisted procedures. All anastomoses were performed in termino-lateral fashion using circular staplers No 29-32 depending on the diameter of the intestine. Anastomoses were routinely tested with an 'air leakage test' . In cases where the test was positive, a protective ileostomy was performed, the defect was sutured and these patients were excluded from the study. During the post-operative period, patients were monitored daily for the following clinical signs suggesting the presence of anastomotic leakage: fever, increased WBC, abdominal pain, prolonged post-operative ileus, tachycardia, and tachypnea, increased secretion from abdominal drainages, elevation of CRP (C-reactive protein) and renal function impairment. Definitive signs of anastomotic leakage were established as the presence of peritoneal irritation and/or extravasation of feces at the anastomotic site. In addition to physical examination and laboratory tests, x-rays, computed tomography and endorectal ultrasonography were performed for visual confirmation and assessment of the leakage (Fig. 1).

RESULTS
During the study period, low anterior resections were performed in 141 patients. Due to a positive air leak test on 16 patients, a protective ileostomy was created and these patients were excluded from the study. Of the remaining 125 patients, gender distribution was as follows: 69 males (55.2%) and 56 females (44.8%). The height of the anastomosis from the dentate line was 3.0 cm on the average, varying from 4 to 2.5 cm. Of these patients, 41 (38.8%) were referred to non-adjuvant chemo-radiation therapy. Seventy-eight (62.4%) patients were smokers. The average BMI of the patients was 24.43. Perioperative blood loss was on average 80 ml. A high ASA score was found in 27 patients (21.6%). Perioperative transfusion of blood and blood products was required in 20 patients (16%). Anastomotic leakage was observed in 11 patients (8.8%) (Fig. 2).
The gender distribution in this group is as follows: 6 males and 5 females.
Anastomotic leakage most commonly occurred in patients with a risk factor combination of high ASA score  and a BMI> 27 and it was observed in 6 patients (54.5%) distributed equally between males and females. In addition to the height of anastomosis and smoking, non-adjuvant chemo/radiation therapy was associated with an increased risk of anastomotic leak in 3 patients (27.3%). Two patients (18.2%) had a combination of anastomosis height, male gender and increased operating time> 200 minutes due to intraoperative hemorrhage requiring perioperative transfusion of blood products (Fig. 3).

DISCUSSION
Our results show that a high ASA score and the height of anastomosis are the most common risk factors for anastomotic leakage. Comorbidity as a factor for AL has been supported by a number of studies that show a significant increase in the incidence of anastomotic dehiscence in patients presenting with more than two concomitant diseases. 6 On the other hand, the height of anastomosis is an independent predictor of AL. According to Vignali it is <7 cm, Karanjia -<6 cm, Rullier -<5 cm, and Eriksen -<3 cm from the dentate line. Rose et al. 7 reported AL complication in 16.8% of cases at a distance of LAR less than 10 cm from the dentate line.
The present study shows that anastomotic leakage is more than twice as common in male gender. This is confirmed by some studies and is due to the narrow male pelvis and poor visualization during dissection. 8 Patients with intra-operative adverse events leading to prolongation of the surgery, such as excessive blood loss requiring preoperative blood transfusion and fecal contamination of the abdominal cavity showed more than 4 times an increased risk of developing AL. 9,10 The question remains whether neo-adjuvant chemo-radiation therapy can increase the risk of dehiscence in low rectal anastomosis. Several studies have been published reporting that it significantly increases the incidence of AL. 11 Other authors reported no difference in both patient groups. 12,13 In our study, non-adjuvant therapy was not observed to be a major risk factor associated with anastomotic complications.
A high BMI was a proven risk factor of anastomotic complications, especially in combination with other risk factors. 14 Smoking as an element causing vascular wall damage and other factors affecting blood vessel disorders such as diabetes mellitus and atherosclerosis have a significant impact on the severity of complications of low rectal anastomoses. 15,16 Many systems including 'colon leakage score' have been developed for the assessment of the risk factors for anastomotic leakage and its prevention (Fig. 4).

CONCLUSIONS
Anastomotic leakage after low anterior resection remains the most feared and serious complication of this surgery. Based on our own experience with the procedure, we can conclude that the main risk factors for anastomotic leakage are the height of the anastomosis and a high ASA score.