OBLIQUE VS. CIRCULAR ANASTOMOSIS IN THE CHILDREN UNDERWENT SOAVE’S PULL-THROUGH SURGERY FOR THE TREATMENT OF HIRSCHSPRUNG’S DISEASE: WHICH IS THE BEST?

- Background : Several types of complications including constipation, fecal soiling, perianal excoriation, were reported among different types of surgery for Hirschsprung’s disease. Aim : To compare circular and oblique anastomoses following Soave’s procedure for the treatment of Hirschsprung’s disease. Methods : Children who underwent Saove’s pull through procedure with oblique and circular anastomoses were included. Duration of the follow up was two years after surgery. Postoperative complications, such as wound infection, wound dehiscence, peritonitis, fecal soiling, perianal excoriation, were recorded for each patient. Results : Thirty-eight children underwent oblique anastomoses. Circular ones were done for 32 children. Perianal excoriation was seen in 57.89% and 46.87% of children in oblique and circular group, respectively. Enterocolitis was more frequent in circular (40.62%) than oblique (28.94%) group. Anastomotic stricture was more frequent in circular (15.62%) than oblique (7.89%). Conclusion : Perianal excoriation was the most common complication among patient in both groups. Oblique anastomoses had fewer complications than circular, and may be appropriate option for patient who underwent Soave’s procedure. rate of Hirschsprung associated enterocolitis after Soave pull-through was 10% in Prahita et al 13 . In the study by Vega Mata and colleagues 19 incidence of post- surgery enterocolitis was zero among patients underwent Soave procedure. The rate of enterocolitis in our study was higher than other researches. Peritonitis was seen 3.12% and 2.63% of children in circular and oblique anastomosis respectively. In the study by Matiolli et al 9 on children who underwent endorectal pull-through in country

. As seen there, enterocolitis was more frequent in circular than oblique type (p=0.004).
Perianal excoriation was the most common complication in both groups, although it was more frequent among cases in oblique group than in circular, but this difference was not statistically significant.
There was no significant difference between two groups in terms of wound infection, length of hospital admission, bleeding during operation, and length of surgery (Table 1). There was no mortality in both groups.
Perianal excoriation was seen in 57.89% and 46.87% of patients with oblique and circular anastomoses. In the study by Pratap et al. 15 perianal excoriation was seen in 34% of children in pull-through for Hirschsprung's disease. Perianal excoriation was found in 36.8% and 42% of Shakya et al 16 and Teitelbaum et al 18 studies. The higher rate of perianal excoriation in our study may be due to different management in stool frequency and perianal excoriation between centers. Shakya et al. 16 used coconut oil for perianal excoriation.
Enterocolitis was one of the most frequent complications after Soave's procedure regardless the type of anastomosis, which is similar to our previous study 1 . Jester et al 5 showed 12% single episode of enterocolitis after pull-through for Hirschsprung. In the study by Langer 6 on transanal Soave pulltrough cases, enterocolitis was found in 6%. In the study by Nasr et al. 11 four of 27 children showed enterocolitis following Soave procedure. The rate of Hirschsprung associated enterocolitis after Soave pull-through was 10% in Prahita et al 13 . In the study by Vega Mata and colleagues 19 incidence of post-surgery enterocolitis was zero among patients underwent Soave procedure. The rate of enterocolitis in our study was higher than other researches.
Peritonitis was seen 3.12% and 2.63% of children in circular and oblique anastomosis respectively. In the study by Matiolli et al 9 on children who underwent endorectal pull-through in country INTRODUCTION H irschsprung's disease which is characterized by the absence of ganglion cell and is a common cause of neonatal intestinal obstruction. Several type of procedures were developed for the treatment of Hirschsprung's disease such as Duhamel, Soave's and posterior neurectomy 2,7,10 . Recent study showed less complication using oblique anastomosis 14 .
The aim of this study was to compare complications and outcome of patients who underwent circular vs. oblique type of anastomoses for the patients with transabdominal Soave's procedure.

METHOD
This retrospective analysis was carried out on the children who underwent transabdominal Soave's procedure using circular or oblique anastomoses starting from 2013 for five years. Duration of post-surgery follow up was two year. This study was done in Imam Khomeini Hospital of Ahvaz Jundishapur University of Medical Sciences which is the referral center for pediatric and neonatal surgery. This study was approved by research affair of Ahvaz Jundishapur University of Medical Sciences (Registration number=U-98011) and ethical committee of the Ahvaz Jundishapur University of Medical Sciences (IR-AJUMS-1398-059). Patient consent form was signed by parents.
Patients with other perineal or gastrointestinal abnormality, total colonic aganglionosis, with poor follow up, and the ones who underwent laparotomy due to acute abdomen were excluded. Patients with body weight >=10 kg at time of pull-through were included.
Circular anastomosis was done routinely in Soave's procedure. In oblique type anastomosis there was 1.5 cm distance between anterior aspects of anastomosis from dentate line while 0.5 cm distance between posterior aspects of anastomosis and dentate line (Figure 1). In the circular type distance between anterior and posterior aspects of anastomosis from dentate line was 0.5 cm.
Patients were categorized into circular and oblique type anastomosis. They were studied in terms of complications such as enterocolitis, constipation, anastomotic stricture, wound infection, fecal incontinency, postoperative fistula, postoperative fever, urologic complication, pelvic infection, wound dehiscence, perianal excoriation, postoperative leukocytosis and mortality.
Duration of postoperative follow up was two years.

RESULTS
In the current study 38 children underwent oblique anastomosis and 32 circular anastomoses. Duration of follow up was two years after surgery. Complications of two anastomoses are shown in OriginAl Article
Constipation was seen in 13.15% and 9.37% of the children in oblique and circular group respectively. Constipation was reported as a common complication following pull-through in different studies. Widyasari 20 reported constipation in 24% of children who underwent Soave's procedure. In our previous publication 1 , constipation was seen in 15% of the cases. In another study constipation occurred in 11.7% of the cases 16 . Constipation may be due to prolonged colonic transit time, postoperative stricture or retained in aganglionic segment 16 .
Fecal soiling was seen in 5.26% and 6.26% of children who underwent oblique and circular anastomosis respectively. In the study by Onishi et al 12 with more duration follow up that reach 18 year, 18.7% of patients showed incontinence and soiling.
Three of 38 cases (7.89%) in oblique group developed anastomotic stricture. Paul et al 14 on 17 children, reported one (5.88%) of patients with postoperative anastomotic stricture.
Anastomotic leakage is one of the most serious complication following pull-through surgery. Rate of anastomotic leakage was reported between 1.3% and 8% in different studies 3,8,17 . In our study, anastomotic leakage was not reported among circular or oblique anastomoses. Anastomotic leakage may be due to technical problem and surgeons experience.
In our study, urologic complication following pull-through was not seen in oblique and circular type anastomoses, as well as mortality. In another study mortality was seen in 5% 16 .
As seen above, oblique anastomosis may reduce complication rates following Soave's. Similar findings were reported by Paul et al 14 . Here, the rate of some complications, such as anastomotic leakage, was lower than in other studies, but anastomotic stricture and enterocolitis were higher.
The main limitations of this paper are it was done in a single center and with limited sample size. Another multicentric study including a bigger number of patients is recommended