Gastric-outlet obstruction is divided into two types: pyloric atresia and antral webs. The first case of a patient with congenital pyloric atresia (CPA) was described by Calder in 1749. Touroff and colleagues performed the first successful operation in 1940 1–7. CPA is a very rare condition. The estimated incidence of CPA is 1 : 100 000 live births, and it accounts for less than 1% of all upper-gastrointestinal-tract atresias 1,4,5. It affects male and female patients equally. CPA is known to be associated with low birth weight and polyhydramnios, both of which are reported in up to 60% of the patients 8. Commonly, CPA occurs as an isolated lesion, which has an excellent prognosis. It can, however, also occur in association with other malformations, which can have a negative impact on the final outcome 8–16. Carmi et al. 10 suggested this pathophysiology of the disorder; therefore, it is also referred to as ‘Carmi syndrome’.
Aetiology and embryology
The exact aetiology of CPA is unknown. Embryologically, however, it is thought to result from developmental arrest between the 5th and 12th weeks of intrauterine life 3,14. According to Tandler 17, CPA results from the failure of the pyloric tube to canalise during embryonic development. This is the most accepted theory for the development of CPA. In contrast, Lowe and Bernard 18 proposed a mechanical cause or vascular accident as an aetiology for CPA. Weber 19 proposed that CPA associated with epidermolysis bullosa (EB) results from an intrauterine mucosal injury within the junctional EB, with subsequent ulceration and an inflammatory scarring reaction. This is an attractive theory in CPA patients with associated EB, but does not explain the occurrence either of isolated CPA or CPA with a gap. The occurrence of CPA and EB is interesting, and appears to be more than the result of chance association, as suggested by Chang and colleagues, who also proposed that CPA in association with EB occurs secondary to an intrauterine complication of EB, in which ulceration and sloughing of the pyloric mucosa leads to fibrosis, scarring and obstruction of the pyloric canal 20,21. The sometimes familial occurrence of CPA, with a high frequency of consanguinity and an equal sex incidence, suggest a genetic predisposition, with an autosomal recessive mode of inheritance 22,23. The reason for the high incidence of CPA in some parts of the world is not known; however, as these patients are genetically similar, a high incidence of consanguinity in these parts of the world may be a contributing factor 8,24.
Classification and associated anomalies
CPA is classified anatomically and also clinically 8. CPA is classified, depending on the presence or absence of associated anomalies, into three groups:
Isolated CPA 1
In these patients, CPA occurs in isolation without other anomalies. These patients have an excellent prognosis.
CPA associated with other genetic disorders like EB and aplasia cutis congenita (ACC) 6,9–11,25–30
Patients with this combination have the worst prognosis. Although CPA commonly occurs in isolation, it is frequently reported in association with other anomalies, which are commonly seen in 30–45% of the patients. Ilce et al. 4 reported associated anomalies in 43.8% of their patients. This study has reported associated anomalies in 54.5% of the patients 8. EB and intestinal atresias are the most common associated anomalies. EB in association with CPA was first described by Swinbume and Kohler 31. It is believed that CPA in association with EB results from the intrauterine sloughing of the mucous membrane, with subsequent healing by fibrous tissue, leading to the obliteration of the pylorus lumen 21. This is supported by the occasional occurrence of atresias in other parts of the gastrointestinal tract, such as the oesophagus, larynx and anus. The occurrence of acquired pyloric atresia in association with EB is another supporting factor 32–36. Although this can explain the occurrence of the pyloric membrane, it does not explain the occurrence of CPA with a gap between the two ends 24. A more acceptable theory is that both conditions are genetically determined, and their occurrence together is due either to the pleiotropic expression of a single gene, or the action of two closely related genes. This is also the case with ACC, which is seen in association with CPA. More commonly, all three conditions may occur in the same patient 10,11,35. ACC is also a hereditary condition, and this study proposes that the occurrence of CPA, together with EB and ACC, is the result of closely linked abnormal genes.
EB consists of a group of hereditary bullous skin disorders. On the basis of histology, it is divided into three types: EB simplex, junctional EB and dystrophic EB. All three types have been reported in association with CPA, although junctional EB is the most frequently reported. Evidence suggests that the CPA–EB association is a distinct clinical entity, which is now referred to as the CPA–EB syndrome. This clinical entity may result from a mutation in one of the integrin genes. The absence of detectable α-6 integrin, but not β-4 integrin, in these patients raises the possibility that α-6 integrin or its ligands are responsible for the CPA–junctional EB association 20–22,37. In eight of the patients, CPA was associated with EB, four being from the same family. One of them also had associated ACC 8.
CPA associated with other intestinal atresias 8,12,13,38–41
The associated intestinal atresias are usually multiple, and can form part of the hereditary multiple intestinal atresias (HMIA) 4,5,14,15. These atresias can affect any part of the gastrointestinal tract, from the stomach to the rectum. The presence of these associated anomalies is a contributing factor for the reported high mortality in these patients 8. The possibility of associated intestinal atresias must always be kept in mind, and, to exclude or locate associated colonic atresia, a preoperative barium enema is advocated. The presence of calcification on plain abdominal radiography should raise the possibility of associated HMIA 42. Intraoperatively, it is also important to check the patency of the remaining intestines, using a catheter that is passed distally and injecting saline, to exclude associated intestinal atresias, which are often multiple. HMIA is a very rare condition transmitted as an autosomal recessive inheritance. The atresias can affect any part of the gastrointestinal tract, from the oesophagus to the rectum. Although the atresias are multiple, there are variations in the affected parts of the gastrointestinal tract among the reported patients, even among affected brothers 5,12,13. This may be attributed either to gene mutations or to incomplete gene penetration. Lambrecht and Kluth 43 collected 35 well-documented patients of HMIA from the literature. An interesting finding in some of these patients is the occurrence of CPA with distal duodenal atresia. This occurrence forms a closed duodenal loop, in which biliary and pancreatic secretions accumulate, leading to a massive distention of the duodenum. This distention increases the danger of perforation, or of the secretions refluxing back into the biliary tree and leading to its dilatation 7,12–14. The presence of associated pure oesophageal atresia makes the preoperative diagnosis of associated CPA impossible. It is important to exclude this at the time of gastrostomy. This, however, is not the case in patients presenting with associated oesophageal atresia and tracheoesophageal fistula. In the latter two conditions, a plain abdominal radiography shows the stomach to be dilated, with no air seen distally 8. Other reported associated anomalies with CPA include Down’s syndrome, Meckel’s diverticulum, ureterovesical-junction obstruction, pelviureteric-junction obstruction, agenesis of the gallbladder and malrotation, high anorectal agenesis, congenital heart disease, cleft palate and pylorocholedochal fistula 7,36,41,44,45.
A review of the literature in 1989 reported 125 patients of CPA, three of whom presented with associated oesophageal atresia, and 18 with EB 1. The researchers in this study treated 20 newborn babies with CPA, in seven (35%) of whom, CPA occurred in isolation; in eight (40%) of whom, it was associated with EB; and in five (25%) of whom, it was associated with other gastrointestinal atresias 8.
CPA is also classified anatomically into three types 5,8,24:
- Type 1: pyloric membrane (57%), which may be multiple.
- Type 2: pyloric atresia without a gap (34%).
- Type 3: pyloric atresia with a gap between the stomach and the duodenum (9%).
Moore 1 proposed a pathophysiological classification of congenital gastric-outlet obstruction due to pyloric atresia or gastric antral web. In most patients of CPA, the defect consists of a mucosal or submucosal membrane, without a muscular component (pyloric membrane or diaphragm). Intraoperatively, it is important to make sure that only one pyloric diaphragm is present, as there may be multiple occurrences 8. Less frequently, the pylorus is a fibrous string; else, a complete segmental defect may be present. CPA with a gap is the least common type.
Infants with CPA usually present with nonbilious vomiting soon after birth, feeding difficulties and, sometimes, upper-abdominal distention. The upper-abdominal distention is secondary to the dilated stomach (Fig. 1). A history of polyhydramnios is present in the majority of these patients. A delayed diagnosis may lead to pulmonary aspiration, severe metabolic derangement and gastric perforation, all of which can be fatal. The presence of associated anomalies is a contributing factor for delayed diagnosis 8. This must be kept in mind. Clinically, the presence of EB and or ACC can be detected. This evidence should raise the possibility of associated CPA, in those who present with nonbilious vomiting (Fig. 2). In patients with associated pure oesophageal atresia, the preoperative diagnosis of CPA is difficult, as there is no gastrointestinal air. This must be kept in mind at the time of gastrostomy 8. In patients with oesophageal atresia and distal tracheo-oesophageal fistula, the stomach appears dilated on plain abdominal radiography, with no air distally (Fig. 3). Patients with CPA and intestinal atresia involving the duodenojejunal junction have been reported 12,13. This forms a closed duodenal loop that can dilate as a result of accumulation of pancreaticobiliary secretions and, sometimes, perforate. The diagnosis in these patients is usually made intraoperatively.
Congenital anomalies of the gastrointestinal tract are relatively common, but CPA is extremely rare, constituting less than 1% of all upper gastrointestinal atresias 1–5. Commonly, it occurs as an isolated lesion, which presents with upper-abdominal distention and nonbilious vomiting. The familial occurrence of isolated CPA supports an autosomal recessive mode of inheritance 46,47. CPA can be diagnosed antenatally by an ultrasound in the second trimester, which can show the presence of polyhydramnios, associated with a dilated stomach (single air-bubble) 11,48,49 (Fig. 4). The diagnosis of CPA is simple and can be made with a plain abdominal radiography. This classically shows a single large gastric air-bubble, and a gasless nondistended abdomen distally (single air-bubble) (Fig. 5). The diagnosis can be confirmed by a barium meal, although it is often not necessary (Fig. 6). Reports of perforated pyloric webs are available. These show scanty gas distally in the gastrointestinal tract 50. These are not simple to diagnose, and contrast enema or endoscopy may be necessary to confirm the diagnosis. In patients who are suspected to have associated multiple intestinal atresias, a contrast enema is done to locate or roll out associated colonic atresia. The presence of calcification in plain abdominal radiography should alert the physician to the possibility of associated multiple intestinal atresias 42,43. It is important to inspect the whole intestines intraoperatively, to roll out associated intestinal atresia. The physician needs to ensure the patency of the remaining intestines through distally injected saline, also known as a saline test. The presence or absence of associated oesophageal atresia can be confirmed by passing a nasogastric tube. Neonatal gastric perforation is also a rare and unusual presentation of CPA, and these patients present with pneumoperitoneum.
The treatment of CPA involves surgical correction, and this depends on the type of pyloric atresia. Different procedures are followed, depending on the anatomic type. Pyloric atresias of types 1 and 2 include pyloric diaphragms and pyloric atresias without a gap. These are treated by the excision of the membrane and Heineke-Mikulicz or Finney pyloroplasty. Intraoperatively, it is important to make sure that only one pyloric diaphragm is present, as these manifestations can be multiple 8. It is also important to check for the patency of the remaining intestines using saline injection, to exclude associated intestinal atresias, which are often multiple. For those with pyloric atresia with a gap, the treatment is pyloroduodenostomy. Dessanti et al. 51 describe pyloric-sphincter reconstruction for patients having CPA without a gap. Gastrojejunostomy should be avoided, as it is associated with high morbidity, including anastomotic ulcers. Gastrostomy should not be part of the operative treatment, but can be carried out in those with associated pure oesophageal atresia.
The association of CPA and EB has almost fatal outcomes. In consequence, surgeons advocate no surgical intervention for CPA in these patients 27. Others feel that the association of CPA with EB should not preclude surgical treatment, as, recently, encouraging reports of survivors have surfaced, and new treatments of EB are being developed 25,45,52. These patients usually die because of fatal infections with sepsis and severe electrolyte imbalance. The CPA–EB syndrome is a hereditary condition, which carries a 25% risk of recurrence in future siblings. This is important, and parents should undergo genetic counselling. This is specially so in this era, in which a prenatal diagnosis of CPA can be made as early as 10 weeks of gestation, and EB can be diagnosed by the electron microscopy of a foetoscopic skin biopsy 11,48,49.
The prognosis of CPA is variable, depending on the presence or absence of severe associated anomalies 8,53. The overall mortality is very high, exceeding 50%, but this is due to the high incidence of severe, and often fatal, associated anomalies. The prognosis of isolated pyloric atresia, and of pyloric atresia associated with other non-hereditary intestinal atresias, is excellent. Early diagnosis and surgery, together with neonatal supportive care, have significantly improved the survival rate in these patients. CPA associated with HMIA has a poor prognosis, and some of these patients may have an associated combined immunodeficiency syndrome. Recently a combined immunodeficiency syndrome was reported in patients with HMIA 38–40. CPA, in association with HMIA, is universally fatal, as so far none of the reported patients have survived 7,13. The cause of death in the majority of them was sepsis. The association of CPA and EB has a poor prognosis, and the majority of these patients die because of sepsis and dehydration. EB is generally a fatal disease, and the cause of death is sepsis and fluid loss. As a result of this, some surgeons advocate non-surgical intervention in those with EB associated with CPA 27. Okoye et al. 16 and others feel that the association of CPA with EB should not preclude surgical treatment 16. Recently, the use of steroids and phenytoin has been reported to give favourable results 20,35,52. Prenatally, it is now possible to diagnose CPA on ultrasound, as early as 10 weeks, and EB can be also diagnosed by the electron microscopy of a foetal-skin biopsy, carried out at 19 weeks’ gestation 11,48,49. The decision regarding the prognosis and the continuance of that pregnancy, as well as future pregnancies, should be discussed with the parents, considering both the 25% risk of recurrence and the encouraging results with EB, using phenytoin and steroids.
In conclusion, CPA is a very rare malformation that can be familial and inherited as an autosomal recessive. It can either occur as an isolated lesion with an excellent prognosis, or be associated with other genetically determined conditions, such as EB and/or ACC. Its occurrence in association with other intestinal atresias is interesting and must be kept in mind. These atresias can affect any part of the gastrointestinal tract, from the oesophagus to the rectum. Whereas the managemnet of CPA is simple, the managment of associated intestinal atresias is more complex. Every case should be managed individually. The overall prognosis of CPA, however, is still poor, and this is due to the frequent – and often fatal – associated anomalies.
Conflicts of interest
There are no conflicts of interest.
1. Moore CM. Congenital gastric outlet obstruction. J Pediatr Surg 1989; 24:1241–1246.
2. Kume K, Ikeda K, Hayashida Y, Naito K, Teraoka H, Nishibayashi Y, Nagai N. Congenital pyloric atresia: a report of three cases and review of the literature. J Jpn Soc Pediatr Surg 1980; 16:259–268.
3. Bronsther B, Nadeau MR, Abrams MW. Congenital pyloric atresia. A report of three cases and a review of the literature. Surgery 1971; 69:130–136.
4. Ilce BZ, Erdogan E, Kara C, Celayir S, Sarimurat N, Senyüz OF, Yeker D. Pyloric atresia: 15-year review from a single institution. J Pediatr Surg 2003; 38:1581–1584.
5. Al-Salem A, Nawaz A, Matta H. Congenital pyloric atresia: the spectrum. Int Surg 2002; 87:147–151.
6. Nawaz A, Matta H, Jacobsz A, Al-Salem A. Congenital pyloric atresia and junctional epidermolysis bullosa: a report of two cases. Pediatr Surg Int 2000; 16:206–208.
7. Sencan A, Mir E, Karaca I, Günşar C, Sencan A, Topçu K. Pyloric atresia associated with multiple intestinal atresias and pylorocholedochal fistula. J Pediatr Surg 2002; 37:1223–1224.
8. Al-Salem AH, Abdulla MR, Kothari MR, Naga MI. Congenital pyloric atresia, presentation, management, and outcome: a report of 20 cases. J Pediatr Surg 2014; 49:1078–1082.
9. Tomá P, Mengozzi E, Dell'Acqua A, Mattioli G, Pieroni G, Fabrizzi G. Pyloric atresia: report of two patients (one associated with epidermolysis bullosa and one associated with multiple intestinal atresias). Pediatr Radiol 2002; 32:552–555.
10. Carmi R, Sofer S, Karplus M, Ben-Yakar Y, Mahler D, Zirkin H, Bar-Ziv J. Aplasia cutis congenita in two sibs discordant for pyloric atresia. Am J Med Genet 1982; 11:319–328.
11. Achiron R, Hamiel-Pinchas O, Engelberg S, Barkai G, Reichman B, Mashiach S. Aplasia cutis congenita associated with epidermolysis bullosa and pyloric atresia: the diagnostic role of prenatal ultrasonography. Prenat Diagn 1992; 12:765–771.
12. Al-Salem AH. Pyloric atresia associated with jejunal and duodenal atresia and duplication. Pediatr Surg Int 1999; 15:512–514.
13. Al-Salem AH, Qaissruddin S, Varma KK. Pyloric atresia associated with intestinal atresia. J Pediatr Surg 1997; 32:1262–1263.
14. Snyder CL, Mancini ML, Kennedy AP, Amoury RA. Multiple gastrointestinal atresias with cystic dilatation of biliary ducts. Pediatr Surg Int 2000; 16:211–213.
15. Chittmittrapap S. Pyloric atresia associated with ileal and rectal atresias. Pediatr Surg Int 1989; 3:426–430.
16. Okoye BO, Parikh DH, Buick RG. Pyloric atresia: five new cases, a new association, and a review of the literature with guidelines. J Pediatr Surg 2000; 35:1242–1245.
17. Tandler J. Zur Entwicklungsegeschichte des menschlichen Duodenum im fruhen embryonens stadium [On the developmental history of the human duodenum in the early embryonic stage]. Gerenbaur Morph Gahng 1900; 29:187–216.
18. Lowe JH, Bernard CN. Congenital intestinal atresia: observations on its origin. Lancet 1953; 269:1065–1067.
19. Weber M. Hemidesmosome deficiency of gastrointestinal mucosa, demonstrated in a child with Herlitz syndrome and pyloric atresia. Acta Derm Venereol 1987; 67:360–362.
20. Hayashi AH, Galliani CA, Gillis DA. Congenital pyloric atresia and Junctional epidermolysis bullosa: a report of longterm survival and a review of the literature. J Pediatr Surg 1991; 26:1341–1345.
21. Chang CH, Perrin EV, Bove KE. Pyloric atresia associated with epidermolysis bullosa: special reference to pathogenesis. Pediatr Pathol 1983; 1:449–457.
22. Bar-Maor JA, Nissan S, Nero S. Pyloric atresia. J Med Genet 1972; 9:70–72.
23. Olsen L, Grotte G. Congenital pyloric atresia: report of a familial occurrence. J Pediatr Surg 1976; 11:181–184.
24. Al-Salem AH. Congenital pyloric atresia and associated anomalies. Pediatr Surg Int 2007; 23:559–563.
25. Alam-Sahebpoor A, Ghaffari V, Shokoohi L. Pyloric atresia associated with epidermolysis bullosa: a report of 4 survivals in 5 cases. Iran J Pediatr 2007; 17:369–374.
26. Pulkkinen L, Kim DU, Uitto J. Epidermolysis bullosa with pyloric atresia: novel mutations in the beta4 integrin gene (ITGB4). Am J Pathol 1998; 152:157–166.
27. Rosenbloom MS, Ratner M. Congenital pyloric atresia and epidermolysis bullosa in premature siblings. J Pediatr Surg 1987; 22:374–376.
28. Pfendner E, Uitto J. Plectin gene mutations can cause epidermolysis bullosa with pyloric atresia. J Invest Dermatol 2005; 124:111–115.
29. Valari MD, Phillips RJ, Lake BD, Harper JI. Junctional epidermolysis bullosa and pyloric atresia: a distinct entity. Clinical and pathological studies in five patients. Br J Dermatol 1995; 133:732–736.
30. Samad L, Siddiqui EF, Arain MA, Atif M, Parkash J, Ahmed S, et al. Pyloric atresia associated with epidermolysis bullosa – three cases presenting in three mm. J Pediatr Surg 2004; 39:1267–1269.
31. Swinbume I, Kohler HG. Familial pyloric atresia associated with epidermolysis bullosa. J Pediatr 1979; 94:162.
32. Cetinkurşun S, Oztürk H, Celasun B, Sakarya MT, Alpasian F. Epidermolysis bullosa associated with pyloric, esophageal and anal stenosis: a case report. J Pediatr Surg 1995; 30:1477–1478.
33. Kenna MA, Stool SE, Mallorg SB. Junctional epidermolysis bullosa of the larynx. Pediatrics 1986; 78:172–174.
34. Honig PG, Yoder M, Ziegler M. Acquired pyloric obstruction in a patient with epidermolysis bullosa letalis. J Pediatr 1983; 102:598–600.
35. Egan N, Ward R, Olmstead M, Marks JG. Junctional epidermolysis bullosa and pyloric atresia in two siblings. Arch Dermatol 1985; 121:1186–1188.
36. Agarwala S, Goswami JK, Mitra DK. Pyloric atresia associated with epidermolysis bullosa, malrotation, and high anorectal malformations with rectourethral fistula: a report of successful management. Pediatr Surg Int 1999; 15:264–265.
37. Varki R, Sadowski S, Pfendner E, Uitto J. Epidermolysis bullosa. I. Molecular genetics of the junctional and hemidesmosomal variants. J Med Genet 2006; 43:641–652.
38. Bass J. Pyloric atresia associated with multiple intestinal atresias and immune deficiency. J Pediatr Surg 2002; 37:941–942.
39. Moreno LA, Gottrand F, Turck D, Manouvrier-Hanu S, Mazingue F, Morisot C, et al. Severe combined immunodeficiency syndrome associated with autosomal familial multiple gastrointestinal atresias: a study of a family. Am J Med Genet 1990; 37:143–146.
40. Bass J. Pyloric atresia associated with multiple intestinal atresias and immune deficiency. J Pediatr Surg 2002; 37:941–942.
41. Sencan A, Mir E, Karaca I. Pyloric atresia associated with multiple intestinal atresias and pylorocho-ledochal fistula. J Pediatr Surg 2002; 37:1223–1224.
42. Martin CE, Leonidas JC, Amoury RA. Multiple gastrointestinal atresias with intraluminal calcifications and cystic dilatation of bile ducts: a newly recognized entity resembling ‘a string of pearls’. Pediatrics 1976; 57:268–271.
43. Lambrecht W, Kluth D. Hereditary multiple intestinal atresias of the gastrointestinal tract: report of a case and review of literature. J Pediatr Surg 1998; 33:794–797.
44. Ratan KN, Pandit SK, Budhiraja S. Pyloric membrane and congenital hydronephrosis – a rare association. Indian J Gastroenterol 1996; 15:76.
45. Okoye BO, Parikh DH, Buick RG, Landa AD. Pyloric atresia: five new cases, a new association, and a review of the literature with guidelines. J Pediatr Surg 2000; 35:1242–1245.
46. Wallerstein R, Klein MK, Genieser N, Pulkkinen L, Uitto J. Epidermolysis bullosa, pyloric atresia, and obstructive uropathy: a report of two cases with molecular correlation and clinical management. Pediatr Dermatol 2000; 17:286–289.
47. Usta IM. Familial pyloric atresia: report of a family and review of the literature. J Matern Fetal Med 2000; 9:190–193.
48. Rodeck CM, Eady RAJ, Gorden CM. Prenatal diagnosis of epidermolysis bullosa letalis. Lancet 1980; 1:949–952.
49. Peled Y, Hod M, Friedman S, Mashiach R, Greenberg N, Ovadia J. Prenatal diagnosis of familial congenital pyloric atresia. Prenat Diagn 1992; 12:151–154.
50. Hanif A, Hasina K, Hassan K, Hasauzzaman A, Islam N, Hossain AZ, Tlukder MK. Pyloric atresia – a case report. Bangladesh J Child Health 2009; 33:35–36.
51. Dessanti A, Iannuccelli M, Dore A, Meloni GB, Niolu P. Pyloric artesia an attempt of anatomic pyloric sphincter reconstruction. J Pediatr Surg 2000; 35:1372–1374.
52. Bauer EA, Cooper TW, Tucker DR, Esterly NB. Phentoin therapy of recessive dysmorphic epidermolysis bullosa: clinical trial and proposed mechanism of action on collagenase. N Engl J Med 1980; 303:776–781.
53. Gupta R, Soni V, Mathur P, Babu Goyal R. Congenital pyloric atresia and associated anomalies: a case series. J Neonatal Surg 2013; 2:40.