Open longitudinal pyloromyotomy and its use in oesophago-gastric reconstruction

A.Rohatgi, A.Riga, G.H.Dickson

Abstract

Background

The standard pyloroplasty is closed transversely which by shortening the gastric remnant may cause tension at the oesophago-gastric anastomosis and a subsequent leak. In order to reduce this risk we describe a technique of open longitudinal pyloromyotomy and review its use in 138 patients undergoing oesophago-gastric reconstruction.

Method

The information was obtained from a detailed database of all oesophago-gastric resections performed in one unit over 21 years.

Results

Two out of 138 patients developed a pyloroplasty leak one of whom died. No pyloroplasty complication was found at post mortem in the remaining nine deaths (overall post-operative mortality of 7.2%). During follow-up no patient presented with the clinical picture of gastric retention in the absence of intra-abdominal tumour recurrence. There were two patients with an oesophago-gastric leak (1.5%) neither of whom died.

Conclusion

We feel that this technique is likely to allow a tension-free oesophago-gastric anastomosis and thus minimise complications.

Introduction

Oesophageal resection includes transection of the vagi which leads to delayed gastric emptying.(1,2)  Therefore most surgeons perform a gastric drainage procedure. The commonest pyloroplasty performed (Finney or Heineke-Mikulicz) shortens the gastric remnant during transverse closure making a tension-free oesophago-gastric anastomosis more difficult to fashion. We describe a technique of longitudinal closure and its use in 138 consecutive patients. We have been unable to find a previous description of this method.

Figure 1   Surgical Technique

Following a Kocher manoeuvre a longitudinal incision is made through all layers of the pylorus. The length of this incision must be adequate to obtain a complete division of the pyloric muscle and provide access for meticulous closure; in our experience this is a 4 cms incision, 2cms into both the stomach and duodenum. To start the closure a 4-0 nylon continuous stitch passes through all layers of the duodenal wall. This is continued picking up only the mucosa and avoiding the pyloric muscle. The final throw of this stitch at the gastric end again picks up all layers. We take two precautions to avoid inadvertently picking up the posterior wall: the first is to advance the tip of a 28Fr Salem naso-gastric tube (Sherwood Medical, Ireland), which had been previously placed through the oesophago-gastric anastomosis, into the second part of the duodenum whilst the closure is taking place. The second utilises the sliding property of nylon by placing the final throws carefully into position before pulling the pylorotomy closed. A further layer of interrupted 4-0 nylon stitches loosely approximates the serosa. This stitch picks up only the serosa and avoids the underlying pylorus muscle which spontaneously falls out of the way. Finally the naso-gastric tube is gently withdrawn until the tip is just proximal to the pyloroplasty.

Patients and methods

210 oesophageal resections were performed in our centre between 1979 and 2000. The first 200 cases formed the basis of a paper comparing our survival rates with other series. (20) 138 of these patients had a distal gastric tube fashioned and joined to the transected oesophagus. This was followed by a pyloroplasty performed as described above. A study of these 138 patients forms the basis of this study. The median age was 68 (31 - 83) and there were 98 males and 40 females. Table 1 shows that 99% of the cases were carcinoma of which 90% were sited in the peri-juctional (3) oesophago-gastric region and 77% were adenocarcinomas. Post-operative gastric emptying was assessed by measuring the oral intake and gastric aspirate. If the 4-hourly aspirate was 50% or less of the 24-hour intake and there were no worrying clinical features, then the nasogastric tube was removed. Long-term gastric retention was assessed clinically. The patients were followed up for a mean of 30 months (0 - 208).

Results

The 10 patients who died post-operatively (7.2% mortality rate) were all examined at post-mortem. One patient died with a complication from the pyloroplasty; he had a localized peritonitis from a small leak which was incompletely draining through an abdominal drain. He also had pneumonia which prevented consideration of re-opening his abdomen; in the view of the pathologist the pneumonia was the major cause of death. Out of the 138 patients there were 2 pyloroplasty leaks (1.5%); the patient who died described above and a second who, having made an excellent initial recovery, developed peritonitis on the 12th post-operative day. At laparotomy he was found to have a wide-open pyloroplasty, generalised fat necrosis and marked pancreatic oedema from pancreatitis. An end of a jejunal roux-en-y loop was anastomosed end-to-side to the reopened pylorus, and multiple abdominal drains were inserted after peritoneal lavage. He made a full recovery.

No case of delayed gastric retention occurred post-operatively and no patients during long-term follow up had clinical symptoms to suggest this apart from those patients who developed obstructive symptoms from intra-abdominal tumour recurrence.

There were two patients with oesophago-gastric leaks (1.5%). The first patient with an oesophageal leak presented on day 3 with a suspicion of gastric contents in the chest drain. A small leak was shown on contrast X-ray. He was placed on limited sips of water, antibiotics, his drains were left in place and he was fed through a jejunostomy inserted during surgery. He made a full recovery.

The second patient had undergone an emergency resection of a bulky tumour leaking into the left chest following endoscopic dilatation. On postoperative day 7 he developed respiratory problems, and when a pleural effusion was drained gastric contents were found. He was managed conservatively with antibiotics, drains and a feeding jejunostomy. He went home 6 weeks after his operation

Table 2 shows the details of the four patients with gastrointestinal leaks including the outcomes and complications. All the patients were male, underwent an abdominio-thoracic approach (two left and two Ivor-Lewis) and had a diagnosis of cancer.

Discussion

Some surgeons believe that gastric drainage is unnecessary and the complications and side effects of pyloroplasty outweigh the advantages. These include post-operative leaks, stenosis, incomplete myotomy, dumping and duodeno-gastric reflux.(4-6)

However Mannell et al in a randomised controlled study comparing oesophagectomy with and without pyloroplasty showed that patients with an intact pylorus were more liable to aspiration pneumonia, post prandial distress, vomiting and heartburn.(7) Also studies have shown that pyloroplasty prevents pylorospasm, dilation of the intra-gastric stomach and gastro-oesophageal reflux. (8-10)

A Pyloroplasty can be performed by a variety of techniques; Heineke-Mikulicz, Finney, an extramucosal "Ramstedt-type" myotomy (11), partial pylorectomy (12),Y-V, Z, Y-U(13),Mercedes-Benz (14) and stapled pyloroplasties (15). A finger bougie stretch method without the need of a separate incision has also been reported.(16)

In the infantile hypertrophic pyloric stenosis setting we found reviews of Ramstedt's procedure documenting a 0.5% to 4.0% peroperative duodenal perforation rate (17,18) and a 0.6% to 4.0% reoperation rate, which included reoperations for incomplete myotomy.(18,19) The corresponding figures following the more difficult procedure of dividing a normal pyloric sphincter in the oesophago-gastrectomy setting are unpublished.

The standard transverse closure of a pyloroplasty (Heineke-Mikulicz) tends to shorten the gastric remnant, which may alone, or in association with post-operative distension, lead to tension and impairment of blood flow at the oesophago-gastric anastomosis, which in turn may result in a leak. The reported leak rate ranges from 1-12% (20-21) and has a mortality of about 50%. (22,23)

We suggest that a longitudinal closure makes this sequence of events less likely. An alternative strategy would be to use small or large intestine to re-establish intestinal continuity. However, it is generally agreed that the stomach should be used if possible as its use shortens operating time and provides the safest and most physiological replacement.(24-26)

Conclusion

We feel that our method of open longitudinal pyloroplasty avoids gastric remnant shortening and is effective in avoiding post-vagotomy gastric retention. This procedure is more likely to allow a tension-free oesophago-gastric anastomosis thus minimising the possibility of a leak.

Reference

1. Cullen JJ. Kelly KA. Gastric motor physiology and pathophysiology. Surgical Clinics of North America. 1993; 73:1145-60.

2. Behrns KE. Sarr MG. Diagnosis and management of gastric emptying disorders. Advances in Surgery. 1994; 27:233-55.

3. Dickson GH, Singh KK, Escofet X, Kelley K. Validation of a modified GTNM classification in peri-junctional oesophago-gastric carcinoma and its use as a prognostic indicator. EJSO 2001;27:641-644

4. Wang LS , Huang MH, Haung BS, Chien KY. Gastric subsitution for resectable carcinoma of the oesophagus: an analysis of 368 cases. Annals of Thoracic Surgery 1992;53:289-94

5. Sinha S, Padhy AK, Chattopadhyay TK.Dumping syndrome in the intra-thoracic stomach. Trop Gastroenterol 1997;18:131-3

6. Zieren HU, Muller JM, Jacobi CA, Pichlmaier H. Should a pyloroplasty be carried out in stomach transposition after subtotal esophagectomy with esophago-gastric anastomosis at the neck? A prospective randomized study. Chirurg 1995;66:319-25

7. Mannell A, McKnight A, Esser JD. Role of pyloroplasty in the retrosternal stomach: result of a prospective, randomised, controlled trial. BJS 1990;77:57-9.

8. Fok M, Cheng SW, Wong J. Pyloroplasty versus no drainage in gastric replacement of the esophagus American Journal of Surgery.1991;162:447-52.

9. Cheung HC, Siu KF, Wong J. Is pyloroplasty necessary in esophageal replacement by stomach? A prospective, randomised controlled trial. Surgery. 1987;102:19-24

10. Zhao CW. Clinical studies of intra-thoracic stomach function in patients after esophagectomy and reconstruction by whole stomach. Chinese journal of Surgery 1993;31:115-7

11. Law S, Cheung MC, Fok M, Chu KM Wong J. Pyloroplasty and pyloromyotomy in gastric replacement of the oesophagus after esophagectomy: a randomised controlled trial. Journal of the American College of Surgeons 1997;184: 630-6

12. Zakaria MA. Anterior hemipylorectomy. A method of pyloroplasty. Journal of the Royal College of Surgeons of Edinburgh 1984;29:367-9

13. Randolph JG. Y-U advancement pyloroplasty. Annals of Surgery 1975;181:586-90

14. Vashney S. Somers SS. Mercedes-Benz pyloroplasty: a new technique. Tropical Gastroenterology. 2001; 22:232

15. Potter SD, Spiro SA, Nance FC. An alternative method to traditional pyloroplasty using a circular stapler. Journal of the American College of Surgeons 1995;180:742-4

16. Yamashita Y, Hirai T, Mukaida H, Yoshimoto A, Kuwahara M, Inoue H, Toge T. Finger bougie method compared with pyloroplasty in the gastric replacement of the esophagus. Surgery Today 1999; 29:107-10

17. Pranikoff T, Campbell BT,Travis J, Hirschl RB.Differences in outcome with subspeciality care:pyloromyotomy in North Carolina. Journal of Pediatric Surgery. 2002;37:352-6

18. Hulka F, Harrison MW, Campbell TJ, Campbell JR. Complications of pyloromyotomy for infantile hypertrophic pyloric stenosis. Am J Surg 1997;173:450-2

19. Rokke O, Stjernholm E, Saeboe-Larssen J,Due J. Infantile hypertrophic pyloric stenosis. 22 years' data. Tidsskr nor Laegeforen 1989;109:2419-21

20. Dickson GH, Waters R, Bull J, Sitzia J. Should we continue oesophageal surgery in a district general hospital ? A review of 200 consecutive cases. Ann R Coll Surg Engl 2001;83:167-171

21. Beitler AL. Urschel JD. Comparison of stapled and hand-sewn esophagogastric anastomoses. American Journal of Surgery.1998;175:337-40.

22. Griffin S, Desai J, Charlton M, Townsend E, Fountain SW. Factors influencing mortality and morbidity following oesophageal resection. European Journal of Cardio-Thoracic Surgery. 1989;3:419-23

23. Lorentz T, Fok M, Wong J, Anastomotic leakage after resection and bypass for esophageal cancer : Lessons learned from the past. World J. Surg. 1989;13:472-477

24. Urschel JD. Does the interponat affect outcome after esophagectomy for cancer? Diseases of the Esophagus 2001;14:124-30

25. Michael Griffen S. In: Michael Griffen S, Raimes SA (eds) Upper Gastrointestinal Surgery . London: Saunders, 1997: 124-128

26. Orringer MB. Palliative procedures for esophageal cancer. Surg Clin North Am 1983;63:941