Introduction
Intestinal atresia (IA) is one of the most common causes of intestinal obstruction in newborns. The IA incidence is 2.5-3 per 10,000 live births (1,2). IA can be classified as follows. Type 1 cases have a transluminal septum accompanied by a proximal dilated bowel in continuity with a collapsed distal bowel (the bowel is usually of normal length). Type 2 refers to cases in which two blind-ending atretic ends are separated by a fibrous cord along the edge of an intact mesentery. Type 3A cases exhibit type 2 findings plus an additional mesenteric defect and a shortened bowel length, while type 3B cases have proximal jejunal atresia, often with malrotation, absence of most of the mesentery, and varying lengths of the ileum surviving after perfusion from the retrograde flow of a single supply artery. Type 4 refers to cases with multiple IAs of types 1, 2, and 3 in sequence (sausage appearance). However, in our study, we did not consider types 3A, B as separate entities; rather, we classified them together as type 3 IA (3). We used the expression “multiple IAs” for cases with more than one type of IA in different regions of the intestinal system.
Approximately 60 years ago, the mortality rate observed in newborns with IA was 30-50%, whereas the current survival rates are 90% or higher (4-6). Advancements in newborn care, total parenteral nutrition (TPN), maternal polyhydramnios/antenatal diagnoses, neonatal anesthesia, and surgical techniques have resulted in higher survival rates among affected patients (2). However, some studies have reported survival rates of 41.7-71.5% (7-9). Despite all efforts, newborns with IA still die due to additional congenital abnormalities, and those who survive may experience prolonged hospitalization due to the need for TPN and the presence of interfering infections.
The aims of this study were to determine the clinical and demographic characteristics of patients undergoing IA surgical treatment in a tertiary pediatric surgery and neonatal intensive care unit and to identify the factors affecting clinical outcome, as well as to share our experience.
Methods
The Fırat University Faculty of Medicine Ethics Committee approved the protocol before study initiation (approval number: 11, date: 16.11.2017). The study included pediatric patients undergoing surgical treatment for IA in the Pediatric Surgery and Newborn Intensive Care Unit of Fırat University Faculty of Medicine between January 1997 and December 2016. We excluded the data from patients with esophageal atresia (EA) only or with isolated anal atresia. We extracted the data from patient files retrospectively. We recorded data such as gestational age, sex, birth weight, age at diagnosis, 5-minute Apgar score, maternal age, prenatal diagnosis, symptoms and findings, direct abdominal radiographic findings, IA site, presence of multiple IAs, concomitant intestinal/systemic abnormalities, major cardiac defects, surgical technique, duration of hospital stay, short-term complications, clinical outcomes, and causes of mortality and morbidity. We compared the patient data according to the site of IA.
Statistical Analysis
For all statistical analyses, we used SPSS 21 for Windows (IBM SPSS Statistics, Armonk, NY, USA). Numerical variables are expressed as means ± standard deviation and categorical variables as percentages (%). We applied the Kruskal-Wallis test to compare variables with a non-normal distribution among more than two groups. We used the chi-square or Fisher’s exact test to compare categorical variables. We evaluated binary categorical variables using the binomial test. We applied binary logistic regression to compare patients with and those without mortality. A p-value <0.05 was indicative of statistical significance.
Results
We evaluated 74 patients undergoing surgical treatment for IA. The newborn male to female ratio was 1.2. The mean age at diagnosis was 6.4±8.3 days (range: 1-45 days), the mean birth weight was 2.3±0.6 kg, the mean gestational age was 35.6±2.79 weeks, the mean maternal age was 28.9±6.1 years, the mean hospital stay was 24.5±25.3 days, and the mean 5-minute Apgar score was 7.6±1.7. The IA site was classified as duodenal (n=31, 41.8%), jejunal (n=17, 22.9%), ileal (n=15, 20.2%), multiple (n=5, 6.7%), pyloric (n=4, 5.4%), or colonic (n=2, 2.7%). Table 1 presents the patient characteristics according to the site of the atresia. The most common site was the duodenum. The 5-minute Apgar scores, gestational ages, birth weights, hospital stay durations, and rates of maternal polyhydramnios, multiple IAs or additional intestinal/systemic abnormalities, prematurity, sepsis, and mortality did not differ according to IA site (p>0.05). The complication rate was significantly higher in patients with multiple IAs than in those with IA at other sites (p=0.05).
Clinically, non-bilious vomiting was the most frequent complaint in patients with pyloric or duodenal atresia, whereas bilious vomiting/nasogastric drainage was the most frequent complaint in all other patients. The most common physical examination finding was abdominal distension. Radiographic findings of the patients revealed massive gastric distension in all patients with pyloric atresia (n=22, 71%) and double-bubble signs (n=8, 26%) and free intraperitoneal air (n=1, 3.2%) in patients with duodenal atresia. Air-fluid levels (n=15, 88%), free intraperitoneal air (n=1, 6%), and ground glass appearance/calcifications (n=1, 6%) were observed in patients with jejunal atresia. Of the 15 patients with ileal atresia, 11 (73%) had air-fluid levels, and 3 (20%) had intraperitoneal free air. The patient with ascending colonic atresia had intraperitoneal free air due to perforation. In addition, the patient with rectal atresia appeared to have a pouch colon. Four of the five (80%) patients with multiple IAs had air-fluid levels, and the remaining one (20%) had a massive gastric appearance.
The incidence of additional abnormalities was 68% in patients with duodenal atresia, 41% in patients with jejunal atresia, 24% in patients with ileal atresia, 80% in patients with multiple IAs, and 100% in patients with colonic atresia. Table 2 lists the additional abnormalities detected in the patients.
Mortality was observed in 4 of 33 (12%) patients undergoing surgery within the first 2 days after birth and in 7 of 41 (17%) patients undergoing surgery after more than 2 days; however, the difference was not statistically significant (p=0.55). The complication and mortality rates in patients with type 4 IA (40% and 20%, respectively) were higher than those in patients with other types of IA. The survival rate of patients with pyloric atresia (n=4) who underwent pyloroplasty and web excision was 100%. The 31 patients with duodenal atresia underwent Kimura’s diamond-shaped duodeno-duodenostomy (n=24), duodenotomy + duodenal web excision (n=6), or Kimura’s diamond-shaped duodeno-duodenostomy + gastrostomy (n=1). The survival rate of the patients with duodenal atresia was 81%. The 17 patients with jejunal atresia underwent resection anastomosis (n=14), jejunostomy (n=2), or resection anastomosis + gastroschisis repair with a prosthetic patch (n=1). The survival rate of the patients with jejunal atresia was 88%. The 15 patients with ileal atresia underwent resection anastomosis (n=11), ileostomy (n=3), or resection anastomosis with gastrostomy (n=1). The survival rate of the patients with ileal atresia was 87%. The two patients with colonic atresia underwent either resection anastomosis (n=1) or resection anastomosis + colon pull-through (n=1); both patients survived. The patients with multiple IAs (n=5) underwent resection anastomosis (n=4) or resection anastomosis + ileostomy (n=1). Their survival rate was 80%. Table 3 lists the surgical interventions performed on patients with IA, and Table 4 presents the postoperative complications and related treatments.
The mean hospital stay duration was 17.2±7.9 days (range: 9-25 days) in patients with pyloric atresia, 20.9±17 days (range: 2-75 days) in patients with duodenal atresia, 32.8±41.6 days (range: 7-180 days) in patients with jejunal atresia, 21.2±13.2 days (range: 8-50 days) in patients with ileal atresia, 8±0 days in patients with colonic atresia, and 39.2±50 days (range: 16-90 days) in patients with multiple IA. The longest hospital stay was observed in patients with jejunal atresia. Of the 11 patients who died, 9 (82%) had additional congenital abnormalities (X2=8.461, p=0.004), which comprised major cardiac defects (n=3), Down syndrome (n=2), congenital biliary atresia (n=1), EA + tracheoesophageal fistula (TEF) + anal atresia + tracheal atresia (n=1), EA + TEF (n=1), and microcephaly (n=1). We found no independent effect of congenital abnormalities on patient mortality in our logistic regression analysis (R2=0.866, B=5.634, SE=4.161, Wald=1.833, df=1, p=0.176, Exp B (95% confidence interval, 279.722 0.80 to 974627.134). The 5-minute Apgar score was ≤5 in 5 (46%) patients. The age at diagnosis was older than 2 days in 7 (63.6%) patients, and the birth weight was <2 kg in 5 (45.5%) patients. The mean hospital stay of patients with and those without additional abnormalities were 26.4±21.4 and 23.04±28.04 days, respectively (p=0.207).
Discussion
Duodenal atresia is more common in males than females. However, our series included 13 males and 18 females. Prenatal ultrasonography (US) can detect duodenal atresia better than jejunal, ileal, or colonic atresia. Basu and Burge (10) were able to diagnose 31% of small IA cases by antenatal US. In our series, maternal polyhydramnios was detected in 7 of 31 patients with duodenal atresia (23%) via prenatal US. Approximately 50% of the patients with duodenal atresia showed a double-bubble sign via direct abdominal radiography (11). The sign was present in 22 patients (71%) in our series, of whom 26% had massive gastric appearance, and 3% had free intraperitoneal air.
Due to the high incidence of renal and cardiac abnormalities that accompany duodenal atresia, such cases should undergo echocardiography and abdominal US (2). Congenital abnormalities are present in more than 50% of patients with IA. In a series evaluated by Escobar et al. (12) that included 169 patients with duodenal atresia, the rate of Down syndrome was 27%, that of maternal polyhydramnios was 33-50%, and that of premature birth was 45%. Congenital abnormalities were detected in 46% of the patients and included congenital cardiac diseases (n=46), EA (n=14), anal atresia (n=6), anal atresia + EA (n=3), renal anomalies (n=8), biliary atresia (n=2), pyloric stenosis (n=1), and Hirschprung’s disease (n=1). The incidence of additional abnormalities in the patients with duodenal atresia was 68% in our series, and these abnormalities included primarily Down syndrome (32.2%), major cardiac defects (12.9%), microcephaly (3.2%), type 2 choledochal cyst (3.2%), biliary atresia (3.2%), gastroschisis (3.2%), cystic fibrosis (3.2%), penoscrotal hypospadias (3.2%), anal atresia (9.6%), EA + TEF (6.4%), EA + TEF + anal atresia (3.2%), and EA + TEF + tracheal atresia (3.2%). The most common surgical procedure used to treat duodenal atresia is Kimura’s diamond-shaped duodeno-duodenostomy. Concomitant duodenectomy with web excision is a less frequent surgical technique used for cases of type 1 duodenal atresia due to the risk of damage to the ampulla vateri (2). Likewise, Kimura’s diamond-shaped duodeno-duodenostomy was performed in 24 patients in our series, duodenotomy + duodenal web excision in 6 patients, and Kimura’s diamond-shaped duodeno-duodenostomy + gastrostomy in 1 patient. The mortality rate after duodenal obstruction have been reported at 9% by Rattan et al. (13), at 5.9% by Chen et al. (14), and at 58% by Zamir and Akhtar (15). In our series, the mortality rate was 19%. Sepsis is the most important risk factor for duodenal atresia, with a mortality rate of 50%. Congenital cardiac disease, prematurity, low birth weight, and Down syndrome are other risk factors (13). Sepsis (32%), Down syndrome (32.2%), prematurity (48%), birth weight <2 kg (48%), and major cardiac defects (12.9%) were the most important underlying risk factors for mortality in our patients with duodenal atresia.
The incidence of jejunoileal atresia is approximately 1 in every 5000 live births, and this rate is similar between males and females. Approximately one-third of patients are born prematurely (16,17). In our study, 11 males and 6 females had jejunal atresia, and 11 males and 4 females had ileal atresia. A premature birth history was observed in 35% and 27% of the patients with jejunal and ileal atresia, respectively.
Advancements in newborn care, surgical techniques, TPN, antenatal diagnosis, and neonatal anesthesia have increased the survival rate of patients with IA from 80% in the 1990s (16) to over 90% in the 21st century (18,19). The survival rates in our study (100% in patients with pyloric atresia, 81% in those with duodenal atresia, 88% in those with jejunal atresia, 87% in those with ileal atresia, 100% in those with colonic atresia, and 80% in those with multiple IAs) were similar to those observed in the literature. The overall mortality rate was 14.9% in our study.
The surgical technique to be performed depends on the gastrointestinal system abnormalities and on the length of the intestine left behind, rather than the type of atresia itself. Whenever possible, wide resection of the proximal intestine and primary anastomosis are recommended with resection of the atretic segment (19). The expanded segment of the intestine of 3-25 cm was excised and anastomosed in our patients. The use of stomas is not recommended because it increases the risks of mortality and morbidity. The practice of stomas reduced from 20% to as low as 10% from the 1970s to 1990s (6). In our series, a stoma was opened in only 7 patients with intestinal perforation (4.5%).
Colonic atresia occurs in 1 in every 40,000 live births and constitutes approximately 1.8-15% of all IAs (20). In our study, colonic atresia was observed in 3 patients (4.1%): isolated colonic atresia in 2 and multiple IAs in 1. In the literature, 47% of colonic atresia cases are accompanied by congenital abnormalities (21). In our series, we did not encounter additional congenital abnormalities in any of the patients with colonic atresia. The mortality rate of colonic atresia is reported to be 25.7% (21). However, none of the patients in our series died.
The mean hospital stay is reported to be 19 days in patients with duodenal atresia, 25 days in patients with jejunal/ileal atresia, and 12 days in patients with colonic atresia in the study by Piper et al. (22). In our study, the mean hospital stay was 21 days in patients with duodenal atresia, 33 days in those with jejunal atresia, 21 days in those with ileal atresia, 8 days in those with colonic atresia, and 39 days in those with multiple IAs.
Early diagnosis and treatment are thought to reduce the mortality of patients with IA (23,24). In our series, 4 of 33 (12%) patients undergoing surgery within the first 2 days died, whereas 7 of 41 (17%) undergoing surgery after 2 days died; however, the difference was not significant. In patients with IA, the abdominal re-operation rates vary between 14% and 25% for postoperative complications (22,25). In our series, the rate of abdominal reoperation was 4% for postoperative complications.
Conclusion
The mean hospital stay in our study was similar to those reported in the literature. Although the mortality rates of patients with multiple and duodenal atresias were consistent with those in the literature, we encountered no deaths in patients with pylorus or colonic atresia. Concomitant severe congenital abnormalities in these patients adversely affect the hospital stay and mortality rate.
Ethics
Ethics Committee Approval: The Fırat University Faculty of Medicine Ethics Committee approved the protocol before study initiation (approval number: 11, date: 16.11.2017).
Informed Consent: Retrospective study.
Peer-review: Externally and internally peer-reviewed.
Authorship Contributions: Surgical and Medical Practices - M.S., T.T., Ü.B., A.K.; Concept - M.S., T.T., Ü.B.; Design - M.S., T.T., M.A.; Data Collection or Processing - T.T., Ü.B., İ.A.; Analysis or Interpretation - Ü.B., M.A., İ.A., A.K.; Literature Search – M.S., M.A., İ.A., A.K.; Writing - M.S., T.T., M.A., A.K.
Conflict of Interest: No conflict of interest was declared by the authors.
Financial Disclosure: The authors declared that this study received no financial support.