31 - Pyloroduodenal Fistula or Acquired Double Pylorus

Pyloroduodenal Fistula or Acquired Double Pylorus #

Whereas a double pylorus had previously been presumed to be of congenital origin, Hansen et al. (1972) described 2 cases in which the clinical and endoscopic findings indicated that it was usually an acquired lesion. The first case was a 55 year old male with a known prepyloric gastric ulcer. Follow-up gastroscopy showed that the ulcer had perforated into the duodenum, forming a short fistulous communication between the stomach and duodenum, which had the appearance of a second pyloric aperture. In the second case a known duodenal ulcer had perforated through the pyloric ring into the stomach, with a similar result. In both cases a mucosal septum was situated between the 2 apertures.

Drapkin et al. (1974) described a case in which a deep ulcer on the lesser curvature of the "antrum" eventually perforated into the duodenal bulb, forming an acquired pyloro- duodenal fistula. At endoscopy two pyloro-duodenal openings, separated by a mucosal septum, were seen. A catheter inserted into one opening re-entered the "antrum" through the other.

Engle (1975) collected 7 cases from the literature and described another case in which a prepyloric gastric ulcer had penetrated into the duodenum. The radiological appearance was that of a short gastroduodenal fistula, extending from the distal stomach to the duodenal bulb on the lesser curvature side. The accessory canal was separated from the normal pylorus by a septum or bridge of mucosa, which conceivably could simulate a filling defect or mass lesion at a radiographic study. It was stated that cases had probably been misdiagnosed previously and that the condition was more common than the number of reported cases would suggest. A similar case was reported by Bender and Soffa (1975).

Hegedus et al. (1978) studied the developmental history as well as the clinical, endoscopic and radiological features of 11 cases of acquired double pylorus encountered among 7,932 consecutive radiographic studies over a 3 year period. They were able to show how a known prepyloric gastric ulcer penetrated the wall and eventually perforated into the duodenal bulb to form a second "pyloric canal". This left no doubt about the acquired nature of the lesion. In 7 of the patients peptic ulcer symptoms disappeared at the time of formation of the fistula, rendering surgical interference unnecessary. It was said that the gastric side of the fistula might not be visible endoscopically, as it could be covered by fibrin or necrotic material; it was easier to recognize the condition radiologically.

Tallman et al. (1979) reported another 4 cases. In the first case 2 constantly patent pyloric openings with intact mucosal margins were seen gastroscopically. The duodenum could be visualized through both channels. The patient had had a prepyloric gastric ulcer for the previous 8 years. In another case a radiographic study showed a short fistulous tract leading from the superior portion of the "antrum" to the superior fornix of the duodenal bulb. In a third case gastroscopy showed a rosary of oedematous mucosal folds; although a fistula was not visible initially, a pediatric endoscope inserted through the occluded opening entered the duodenum. Two subsequent radiographic studies failed to reveal the fistula. In the fourth case a lesser curvature prepyloric ulcer had led to 2 fistulous communications between the stomach and duodenum, resulting in a tri- channelled pylorus.

Thompson et al. (1982) estimated that approximately 60 cases had been described up to that time. The fistulous communication usually extended from the lesser curvature of the "distal antrum" to the superior fornix of the duodenal bulb; less commonly it was located on the greater curvature side. The radiological appearance was usually characteristic; endoscopy sometimes failed to diagnose the condition.

During 6810 consecutive radiographic studies over a 2 year period we encountered 5 cases of acquired double pylorus (Keet and Bezuidenhout 1984). Four will be described briefly:

Case Reports #

Case 31.1. S.K., 67 year old male, was admitted with a history of burning epigastric pain of 4 months' duration. It commenced a few hours after meals, woke him at night and was relieved by food and antacids. Occasionally it radiated to the back. There was no history of haematemesis or malaena. Twenty years prior to admission he had had a bout of similar symptoms.

Physical examination revealed epigastric tenderness. A radiographic study, done elsewhere, showed an irregular narrowing 3.5 cm in length at the pylorus (Fig. 31.1A); it had been interpreted as a carcinoma. Repeat radiological examination showed that the narrowing was in reality a fistulous connection between the distal end of the pyloric sphincteric cylinder and the base of the duodenal bulb on the lesser curvature side. It was adjacent to, and located on the posterior aspect of the pylorus. The sphincteric cylinder remained partially contracted as illustrated, neither maximal contraction nor maximal expansion occurring. The duodenal bulb was deformed. The condition was diagnosed as an acquired double pylorus, i.e. a pyloroduodenal fistula as a result of a perforating ulcer. Endoscopy showed a benign pyloric ulcer filled with necrotic material. It had perforated into the duodenum and the instrument could be manipulated into the duodenum through the pylorus as well as through the channel formed by the perforation.

AB
Fig. 31.1 A,B. Case S.K. A Narrow, irregular channel between pyloric sphincteric cylinder and duodenal bulb, diagnosed as carcinoma. Normal pyloric aperture not filled with barium. Cylinder partially contracted. B Resection specimen. Arrow through pyloroduodenal fistula. Pyloric aperture visible behind arrow

At operation considerable fibrotic reaction was encountered and the duodenum had to be dissected from the pancreas. A truncal vagotomy, antrectomy and Billroth I anastomosis was performed. The resection specimen showed a second aperture between the stomach and duodenum next to the normal pylorus, with a bridge of mucosal tissue between the 2 apertures (Fig. 31.1B) confirming the diagnosis of pyloroduodenal fistula or so-called double pylorus. Case 31.2. J.N., 55 year old male, complained of burning epigastric pain of 4 months' duration. It commenced an hour after meals and was relieved by food and antacids. There was intermittent nausea with some loss of appetite and weight. Previously he had been treated for ischaemic heart disease, gout, ethanol and aspirin abuse.

Physical examination revealed gouty tophi at various joints, moderate hypertension and epigastric tenderness. A radiographic study showed an extra channel extending from the distal end of the pyloric sphincteric cylinder to the superior fornix of the duodenal bulb on the lesser curvature side (Fig. 31.2). The sphincteric cylinder remained partially contracted throughout the examination, with a deep incisura on the greater curvature due to contraction of the left pyloric loop, as illustrated. There was some deformity of the duodenal bulb. The condition was diagnosed as an acquired double pylorus, presumably the result of penetration of a pyloric ulcer into the duodenum.

Fig. 31.2. Case J.N. Fistula between pyloric sphincteric cylinder and base of duodenal bulb on lesser curvature side (arrow). Constant contraction of sphincteric cylinder. Pyloric aperture patent

Anti-ulcer treatment was commenced. Initially the patient was lost to follow-up, but endoscopy 3 months later showed a posteriorly situated, deep prepyloric ulcer. No definite penetration into the duodenum could be demonstrated, but the examination was difficult and incomplete on account of fixation of tissues. After an episode of haematemesis a month later, a second endoscopy showed the same ulcer to be filled with blood clot. Because of increasing fixation it was not possible to manipulate the instrument through the pylorus into the duodenum.

The patient returned a year later after a massive haematemesis. Following ressuscitation, a laparotomy was performed at which an inflammatory mass was palpated on the posterior aspect of the pylorus. After a truncal vagotomy had been performed it was found that a pyloric ulcer had penetrated deeply into the pancreas, necessitating considerable dissection. The pancreas formed the base of the ulcer; from here it had burrowed into the duodenum. The pyloric ring was intact. The findings confirmed the presence of a gastro-duodenal fistula as a result of a penetrating pyloric ulcer. An antrectomy with Billroth I anastomosis was done. Histologically the ulcer was benign, with chronic gastritis and intestinal metaplasia in the surrounding gastric mucosa.

Case 31.3 S.D., 44 year old female, was a known case of pyloric ulceration. Three years before admission an active ulcer on the lesser curvature of the pyloric sphincteric cylinder had been diagnosed at radiological examination. Follow-up endoscopy after anti-ulcer therapy had shown healing of the ulcer with some residual deformity of the pyloric region.

She then presented with a recurrence of symptoms. Endoscopy showed pyloric deformity. Radiographic examination three weeks later revealed an ulcer on the lesser curvature of the pyloric sphincteric cylinder, on the immediate oral side of the ring, with a fistulous communication between the ulcer and the superior fornix of the duodenal bulb (Fig. 31.3). Prominent, permanent circular mucosal folds were present in the cylinder, which remained partially expanded throughout the examination with complete absence of cyclical contraction and relaxation. Repeat endoscopy the following week confirmed the presence of a prepyloric ulcer with a pyloro-duodenal fistula.

Fig. 31.3. Case S.D. Ulcer lesser curvature side of sphincteric cylinder with fistula (arrow) to base of duodenal bulb. Partial expansion of cylinder. Permanent, circular mucosal folds in cylinder

Case 31.4 W.S., 54 year old male, presented with a history of intermittent epigastric pain, aggravated by meals, relieved by alkalies and at times associated with vomiting, of 4 years' duration. Physical examination revealed epigastric tenderness. The first endoscopic examination showed a small, superficial, benign ulcer on the lesser curvature of the stomach approximately l.0cm proximal to the pylorus. After initial improvement the symptoms recurred. Radiographic examination three years later showed a narrow fistula extending from the lesser curvature of the pyloric sphincteric cylinder, 1.0cm proximal to the pylorus, to the superior fornix of the duodenal bulb. The cylinder remained in a state of partial contraction throughout the examination, while the pyloric aperture appeared normal. Control endoscopic examination confirmed the presence of a benign prepyloric ulcer, described as "deep"; the first part of the duodenum was deformed but a pyloro-duodenal fistula could not be identified. A second radiographic study confirmed the findings of the first. At the third radiological study (six months after the first) a fistulous communication was again noted between the prepyloric lesser curvature ulcer and the superior fornix of the duodenal bulb (Fig. 31.4A). The pyloric sphincteric cylinder remained partially contracted throughout the examination, occasionally reaching the pseudo-diverticulum stage, but never relaxed fully (Fig. 31.4B). Subsequently truncal vagotomy and antrectomy was done elsewhere for "chronic prepyloric ulcer". Unfortunately the resection specimen was discarded and was not available for examination.

AB
Fig. 31.4 A,B. Case W.S. A Fistula (arrow) between sphincteric cylinder and duodenal bulb on lesser curvature side. Cylinder partially contracted. Pyloric aperture patent. B Sphincteric cylinder contracted to pseudo-diverticulum stage. It never relaxed fully

Einhorn et al. (1984) reviewed the literature of acquired double pylorus and reported 4 new cases with long term follow-up. In 36 cases the fistula extended from the lesser curvature of the "antrum" to the superior portion of the duodenal bulb. In 28 of these it had originated from "antral" ulcers, in 2 from "pyloric channel" ulcers and in 6 from duodenal ulcers. In 9 cases the fistula was situated between the greater curvature and the inferior portion of the duodenal bulb, 7 originating from antral ulcers, one from a pyloric channel ulcer and one from a duodenal ulcer. The radiological appearance of a double pylorus was said to be characteristic, consisting of two channels of barium separated by a smooth band of soft tissue which represented the intervening mucosal septum. The endoscopic appearance was also typical, the aperture of the fistula being clearly visible in the majority of cases. With the endoscope in the "distal antrum", it was frequently possible to insert a biopsy instrument into the fistula and visualize its tip within the duodenum by viewing through the pylorus.

None of the cases of acquired double pylorus reported up to 1984 had been associated with malignancy (Einhorn et al. 1984). Friehling and Rosenthal (1985) subsequently described an unusual case of gastric carcinoma in which the radiological appearance of a double-channel pylorus was the result of partition of the pyloric aperture by the tumor, no fistula being present.

The radiological features of two more (unconfirmed) cases of benign, acquired pyloro- duodenal fistula were subsequently described by Hurwitz and Friedman (1987).

Discussion #

Acquired double pylorus or pyloro-duodenal fistula occurring as a result of perforation of a peptic ulcer has also been known by the terms pseudo-pylorus, tunnel ulcer, pyloric diaphragm, antral mucosal band and pyloric septum (Drapkin et al. 1974; Hegedus et al 1978). None of the authors quoted, with the exception of Keet and Bezuidenhout (1984), described the condition in relation to the concept of the pyloric sphincteric cylinder.

Our cases had the following features in common: In all the pyloro-duodenal fistula extended from the superior aspect of the distal part of the pyloric sphincteric cylinder to the superior fornix of the duodenal bulb, i.e. it was situated on the lesser curvature side. (As reported in the literature the fistula may occasionally occur on the greater curvature side).

In all 4 cases the sphincteric cylinder remained partially contracted (or, in other words, incompletely expanded) throughout the radiological examination; although minor degrees of contraction and relaxation did occur, these movements were never maximal. This is best illustrated in Fig. 31.2, in which a permanent contraction of the left pyloric loop on the greater curvature, opposite the commencement of the fistula, is seen. In Case 31.4 the contraction at times reached the pseudo-diverticulum stage (Fig. 31.4B), before reverting to incomplete expansion (Fig. 31.4A).

It is concluded that our cases of acquired pyloro-duodenal fistula were associated with motility disturbance of the pyloric sphincteric cylinder, consisting of a restriction in the range of contraction and relaxation. Theoretically this could have an effect on emptying and trituration of solids (Chap. 18). Emptying of liquid barium occurred without undue delay through both orifices.

References #

  1. Bender MD, Soffa DJ. Acquired double pylorus: a case report. Radiology l975, 116, 325-326.
  2. Drapkin RL, Otsuka AL, Castellanos HL et al. Acquisition of a pyloric septum or pyloric duodenal fistula. Gastroenterology l974, 66, 1234-1236.
  3. Einhorn RI, Grace ND, Banks PA. The clinical significance and natural history of the double pylorus. Dig Dis Sci l984, 29, 213-218.
  4. Engle R. Tunnel ulcer with double pyloric canal. Radiology l975, 116, 323-324.
  5. Friehling JS, Rosenthal LE. Gastric carcinoma presenting as double-channel pylorus on upper gastrointestinal series. Dig Dis Sci l985, 30, 269- 273.
  6. Hansen HO, Kronborg O, Pedersen T. The double pylorus. Scand J Gastroenterol l972, 7, 695-696.
  7. Hegedus V, Poulsen PE, Reichardt J. The natural history of the double pylorus. Radiology l978, 126, 29-34.
  8. Hurwitz J, Friedman L. Pyloroduodenal fistula: a benign complication of peptic ulcer disease. South Afr Med J 1987, 72, 56-58.
  9. Keet AD, Bezuidenhout DJJ. Dubbele pilorus en piloroduodenale fistels. South Afr Med J l984, 66, 740-742.
  10. Tallman JM, Clements JL, Gilliam JH, et al. The multi-channelled pylorus. Clin Rad l979, 30, 337-341.
  11. Thompson WM, Kelvin FM, Gedgaudas RK, et al. Radiologic investigation of peptic ulcer disease. Rad Clin North Amer l982, 20, 701-720.