Introduction

Intestinal atresia represent a spectrum of isolated or syndromic congenital malformations whose consequence is an interruption in the continuity of the digestive tract with possibly a lack of certain segments of the gut [1]. The annual incidence of intestinal atresia is variable from 2 to 4 cases per year per million people [2, 3]. The diagnosis is difficult prenatally and is often done at birth [4]. This is a neonatal emergency care which requires surgical correction  to provide a temporary diversion or a sleeve resection of the atretic area in the neonatal period [1]. The prognosis of intestinal atresia has improved over the years in developed countries with a survival rate which rose from 9.1% in 1951 to 90% the last two decades [5, 6]. In countries in the developing world the management of this pathology is facing multiple problems such as: late diagnosis, precarious technical facilities, lack of unity of neonatal resuscitation, and unavailability of parenteral nutrition. These factors are real obstacles to achieve cure rates comparable to those of developed countries. [7] In Senegal, no work had so far been devoted to this subject which justifies the implementation of this retrospective study whose purpose was to report the diagnostic, therapeutic and evolutionary aspects of this pathology.

Patients and methods

Our work is a retrospective study of descriptive and cross-type nature. We collected the information from January 1st 2009 to December 31th 2012 and found that 10 neonates were supported for intestinal atresia in the pediatric surgery department of the Hospital Aristide Le Dantec in Dakar, Senegal. They had a mean age of 4.2 days and a sex ratio of 0.67. We have not included in this study patients with incomplete or default or initially operated in another health facilities. We studied the clinical, therapeutic and outcome data. We used the classification of Grosfeld to determine the anatomical type of the atresia. Analysis of data was performed using Microsoft Office Excel 2007 software.

Results

Clinical signs were dominated by signs of intestinal obstruction (Table 1). Rectal probe was negative in two children with abdominal distension. Examination of the anal margin was normal in all patients. We noted a case of Prune Belly Syndrome in a neonate with ileal atresia.
Thoraco-abdominal radiography was performed in 10 neonates. The air-fluid levels were found in the photographs of seven newborns of which in four cases it was of colonic nature and three cases at the level of the small bowel. The classic image of "double stomach" or "double bubble" was found in three infants with duodenal atresia and jejunal atresia respectively.
In our series, six pregnancies were properly followed with regular antenatal and prenatal ultrasound. No prenatal diagnosis of atresia was made.
Anemia was found in three patients with ileal atresia. They presented with hemoglobin rates of 7.4; 8.3 and 10.2 g / dl. Seven patients had electrolyte disorders with an average serum sodium of 122,3meq / l and an average serum potassium of 2.41 meq / l.
As part of the search for associated malformations Doppler echocardiography was performed in two newborns. This examination revealed a large patent foramen ovale and patent ductus arteriosus in one of the newborn. We also found situs solitus with a patent ductus arteriosus in a newborn with Prune Belly syndrome. Both children had an ileal atresia.
All patients received intravenous antibiotics. They received a combination of cefotaxime (50-100mg / kg / day) and gentamicin (3-5mg / kg / day). All patients had received intravenous infusions with paracetamol 15 mg / kg / 6 hours. They also received resuscitation facilities based fluid input in form of 100-120ml / kg / d of 10% SG and electrolytes in an amount of 3-5 meq / kg / day for the NaCl and 2- 3meq / kg / day for KCl. The introduction of the nasogastric tube was done routinely in our series.
We found 1 case of duodenal atresia, 2 cases jejunal atresia, 5 cases of ileal atresia (Figure 1) and one case of colon atresia. The newborn had Prune Belly syndrome and a heart defect also had intestinal malrotation position of the common mesentery (Table 2).
Nine patients with intestinal atresia were treated surgically. One newborn with duodenal atresia was treated with a duodenal diamond-shaped duodenostomy. Out of the two cases of jejunal atresia, one received a resection anastomosis end-to end and the other a resection stoma. The restoration of digestive continuity was performed one week after the first surgery. A stoma resection was performed in five infants with ileal atresia. The restoration of digestive continuity was performed at 14 days after the newborns survived the initial response. Side colostomy terminalization was performed in neonates with atresia of the colon.

The average time was 5.8 days (3-9 days) for the removal of the nasogastric tube, 4 days (3-5jours) for the resumption of transit, and 6.4 days (4-10 days) for the initiation of food. The average hospital stay was 14.5 days (1-51 days). It was 5 days for duodenal atresia and 16.85 days for the jejuno-ileal atresia. For the two patients who survived, the operating setbacks were 12 and 15 months. It was 12 months for a newborn who presented an ileal atresia type III according to the classification of Grosfeld with malformations (heart defects and intestinal malrotation) who received a stoma resection with restoration of continuity after one week. It was 15 months for a newborn with duodenal atresia type III who received a duodéno- duodenostomy diamond-shaped suregry.
Early complications were marked by six cases of electrolyte disturbances, two cases of retraction of the stump of the stoma, a case of necrosis, and four cases of parietal abscesses. We have not found of late complications in our study.
We recorded one death in our series preoperatively (ileal atresia autopsy Type IIIa) and four deaths in the immediate postoperative period. For the latter, it was a case of atresia coli, a case of jejunal atresia and two cases of ileal atresia who all received a stoma. Long-term outcome was marked by three other deaths including a case of jejunal atresia and two cases of ileal atresia. A total of eight out of 10 children have died in our series.

Discussion

The diagnosis of intestinal atresia can be done prenatally through the use of imaging techniques including prenatal ultrasound [4, 8]. In developing countries, the diagnosis is often made after birth [9]. In our study no prenatal diagnosis had been made due to the lack of experienced sonographers in the neonatal surgical field. The value of prenatal diagnosis is to enable better management of the newborn, conduct birth in a specialized center with a neonatal surgery expertise. At birth, intestinal atresia carries an array of neonatal intestinal obstruction. [11] The symptoms vary depending on the site of obstruction. The main signs found postnatally in our study were the signs of bowel obstruction. Our results were comparable to those of several authors (Table 3). Thoraco-abdominal radiography provides highly suggestive indications for the diagnosis or the level of obstruction in the digestive tract by showing pictures hydroaeric (atresia of the small intestine or colon) or double bubble (duodenal or jejunal atresia proximal). In our series, radiological examination was the only diagnostic imaging tool used. However, other imaging modalities such as ultrasound or œsogastroduodénal transit (TΟGD) examinations were performed by some authors [7, 9, 12].
In the literature, intestinal atresia is associated in 25-50% of the cases with other defects as well as chromosomal abnormalities [8, 11]. The malformations vary by the site of atresia and are more common seen in duodenal atresia [8, 9, 11, 12]. Our study had less defects associated with intestinal atresia in comparison to those of other authors [8, 11]. This is probably related to the small size of our series and the fact that the search for any abnormalities was not systematic. There are tests that are sent to France at high costs (120,000 CFA francs) and this is why we did not request them. The same situation is found in most developing countries. However, some authors practiced karyotyping and found certain chromosomal abnormalities [9, 11].

The surgical procedure in intestinal atresia is variable depending on the site of the atresia and the anatomical findings after laparotomy and associated malformations [8]. Duodenal atresia and duodenal stenosis were initially supported using a duodeno-jejunostomy. Most authors now recommend duodenal diamond-shaped duodenostomy [11], a technique we used while some African authors performed a duodeno-jejunostomy [7, 9, 12]. In case of jejunal ileal atresia the choice of surgical procedure depends on the associated malformations (malrotation, volvulus, meconium peritonitis, gastroschisis, omphalocele) and the length of the remaining [11] intestine. When the length of the intestine is appropriate a resection of the atretic area followed by end to end anastomosis is performed. [13] In the presence of a short intestine, in cases of meconium peritonitis resulting from a perforation, or when the viability of the intestine is questionable a temporary stoma is done. The treatment of atresia of the jejunum in our study included resection anastomosis end to end in one case and resection stoma in the other while that of ileal atresia in five newborns was made by a resection combined with stoma. Some African authors conducted jejunal or ileal atresia resection followed by end to end anastomosis [7, 9, 72]. The choice of treatment in one or two stages of atresia coli depends for most authors on two main parameters: the general condition of the newborn and the topography of the atresia [14]. A first suregry can quickly remove the obstacle and delay the cureof the atresia for a later time (two to three months) before a severe deterioration of general condition occurs [15]. The treatment is performed in a single session in pre angulo spleen atresia with a newborn in good condition. [16] Resection of the ileocecal valve is not desirable because of its role in the motility of the small intestine and the ileal content protection against contamination by colonic bacteria [17]. However, in recent years some authors have advocated the use of a resection with primary anastomosis, regardless of the location of the atresia whenever the general condition of the patient permits. [16] The treatment of atresia coli was done in this study as a first step wusing a terminalized side colostomy. We preferred this method because the patient's general condition was not good.

Survival of patients undergoing intestinal atresia also depends on nutritional support. Parenteral nutrition plays an important role in the nutritional management of neonates with short bowel after surgery [9]. Dalla Vecchia et al [11]promote  the early use of growth factors to maximize intestinal adaptation and dietary changes can improve the condition of patients with short bowel syndrome. The mean duration of parenteral nutritional support in the study of Bouchard [18] was 20 days. This period may be even longer in severe cases of atresia or malformations including gastroschisis. The average time from onset of feeding was five days in Gouli et al [7], seven days in Chirdan et al [12] and 13 days for Merkel. [8] In our study, the average time from start of feeding was 6.4 days. Our results are comparable to Gouli and Chirdan that feed patients early because of the lack of effective resuscitation.
The hospital stay is variable according to the literature and it depends on the location and type of atresia, associated malformations, and the type of surgery [7, 8, 9, 12, 18].
To avoid early complications it is important that the newborns received a stoma resection [19].  Complications mainly consist of gutting and infections that can range from the parietal suppuration peritonitis to dehiscence digestive problems. Late complications are dominated by the rupture and occlusion flange [8, 11]. In our study and those of Chirdan [12], Gouli [7], and Merkel [8], early complications were dominated by electrolyte disorders and parietal suppuration. In studies by Merkel and Dalla Vecchia early complications were more frequent in jejunal and ileal atresia whereas in duodenal atresia late complications were ore likely [8, 11].
The overall mortality of intestinal atresia depends on several factors that may be prematurity, congenital anomalies, the site of atresia, and also diagnostic delay [7, 9, 12]. In 1951, Evans et al reported a survival rate of 9.3% among patients with intestinal atresia [20]. In the last two decades, most studies from developed countries reported survival rates approaching or being greater than 90% for patients with jejunal or ileal [3] atresia. The prognosis of colonic atresia is usually good [17]. The prognosis of intestinal atresia is better in developed countries (Table 4) that in developing countries where there are survival rates barely reaching 50% [9, 72]. In our study, overall mortality was 80%. The improved prognosis in developed countries is largely due to a better understanding of the causative factors of intestinal atresia, changes in surgical procedures, improving anastomotic suture techniques, improving various aspects of neonatal care and anesthesia and the development of total parenteral nutrition. The high mortality in developing countries is due to several factors including  associated malformations, the unavailability of neonatal intensive care unit, the unavailability of total parenteral nutrition, complications related to surgical techniques including ostomy and low social level with economic problems for the parents who do not allow them to meet certain medical requirements. According to Dalla Vecchia and Shakya, the mortality also depends on the site of the atresia [11, 21]. It is much higher for ileal and jejunal atresia while it is almost zero for colonic atresia. In our study, only seven patients who presented with jejunal or ileal atresia survived after 12 months.

Conclusion

In our context the diagnosis of intestinal atresia is done when  post natal intestinal obstruction occurs. Surgical mortality remains high in developing countries. Improvement of prognosis could be achieved through the establishment of neonatal intensive care units, the availability of total parenteral nutrition, and strengthening of the technical platform for diagnosis and treatment.