A 45-year experience with surgical treatment of peptic ulcer disease in children

A 45-Year Experience With Surgical Treatment of Peptic Ulcer Disease in Children By K. Azarow,

P. Kim, B. Shandling, Toronto, Ontario

l Peptic ulcer disease (PUD) requiring surgical treatment has become rare with the availability of modern medical management. A retrospective study of all patients who required operations for PUD between 1949 and 1994 (n = 43) was done. The patients were classified into 3 groups: A (n = 38): pre-histamine-2 (H2) blocker era (1949-1975); B (n = 3): pre-hydrogen-potassium (H-K+) ATPase inhibitor era (1976-1968); C (n = 2): H-K+ ATPase inhibitor era (1989. 1994). Data, analyzed using 2 analysis (P c .Ol), included preoperative medical therapy, surgical indications, type of operation performed, complications, and postoperative medical therapy. The indication for surgery in group A was bleeding (26), perforation (8). or obstruction (4); in group B the indication was obstruction (2) or perforation (1); in group C the indication was obstruction (1) or bleeding (1). The incidence of obstruction as an indication for surgery did not differ among the groups (P < .Ol). Two of the three patients who had surgery for obstruction in groups B and C had biopsy-proven Helicobacterpylori. The postoperative morbidity rate was lower for groups B and C, although not significantly. The relative mortality among the groups did not change (P > .Ol). Children with PUD can have complications similar to those of adults with PUD. Since the introduction of H2 antagonists, the recognition and treatment of H pylorl and the use of H-K+ ATPase inhibition, the incidence of operations for bleeding and perforation has decreased dramatically. However, the incidence of surgery for obstruction remains the same. Copyright o 1996 by W.B. Saunders Company

INDEX WORDS: perforation.

Peptic

ulcer,

surgery,

obstruction,

bleeding,

HE INCIDENCE of peptic ulcer disease (PUD) among children is approximately 5 in 10,OOO.l Traditionally, the treatment for these patients is primarily medical, with surgery being reserved for the complications of the disease. With the use of H2 blockers, antibiotic and bismuth combinations, and hydrogen-potassium (H-K+) ATPase inhibition, the complications of PUD are becoming fewer, resulting in fewer operations in spite of the advent of minimally invasive surgery.2 Presently, there are no long-term reports on the

T

From the Division of General Surgery, The Hospital for Sick Children and University of Toronto, Toronto, Ontario. Presented at the 27fh Annual Meeting of the Canadian Association of Paediafric Surgeons, Montreal, Quebec, September 2-4, 1995. Address reprint requests to Dr B. Shandling, Division of General Surgery, The Hospital for Sick Children, Toronto, Ontario, Canada MSG 1X8. Copyright o 1996 by WB. Saunders Company

0022-3468/96/3106-0002$03.00/O 750

and S. Ein

sequelae of extended H-K+ ATPase inhibition. In addition, no information exists concerning the prognosis of a child who has long-term antral colonization of Helicobacter pylon’ (formerly Campylobacter pylori). Reports of surgical trends since the advent of current medical therapy are few in the pediatric literature. If available, this information would allow one to evaluate the efficacy of current management in preventing complications that require surgery over the long term. In 1973, Seagram et al published their 20-year experience with surgery for this disease.3 At that time, approximately 1.6 operations were performed each year. This report updates that experience, and spans a 45-year period that covers the recent changes in the management of this disease. MATERIALS

AND

METHODS

We reviewed the charts of all patrents who underwent surgery because of complications associated with PUD. The study included 1949 through 1994. The patients were categorized according to the three distinct eras in the evolution of PUD medical management: A: pre-H2 blocker era (1949-1975); B: pre-H-K+ ATPase inhibitor era (1976-1988); C: H-K+ ATPase inhibitor era (1989-1994). The groups were determined by the availability of these particular medicines at the Hospital for Sick Children, Toronto, Ontario. Data collected included method of diagnosis, type of medical therapy, surgical indication, type of surgical procedure, complications of the surgical procedure, and type of postoperative medical therapy. Data were analyzed using x2 analysis (P < .Ol) where appropriate. The incidence of PUD in Toronto was estimated to be six new cases per year.3,4 The total number of patients operated on was 43: (group A: n = 37; group B: n = 4; group C: n = 2). RESULTS

The methods used to diagnose PUD are listed in Table 1. Preoperative medical therapy for all patients in group A consisted of an “ulcer diet,” antacids, and anticholinergics. Those who had bleeding underwent iced saline treatment. All group B patients were treated with H2 blocker therapy. Those who had biopsy-proven Hpylori had antibiotics and bismuth as well. All patients in group C were treated with H2 blocker therapy. Anticholinergics, antibiotics with bismuth, and H-K+ ATPase inhibition were used in the patient with obstructive symptoms. H2 blocker therapy, when implemented, lasted for an average of 2 months before surgery. Antibiotic treatment for H pylon’ consisted of ampicillin or erythromycin plus flagyl for at least a 2-week course. JournalofPed/affic

Surgery,

Vol 31, No 6 (June),

1996: pp 750-753

PEPTIC

ULCER

751

Table Group

n

A

32

B

c

1. Methods

of Diagnosis

Table 3. Operations

Method

Group

5

Positive upper gastrorntestinal contrast Clinrcal diagnosis based on hematemesis

2

phragmatic air on chest x-ray Subdiaphragmatic air on chest x-ray

2

Endoscopic

1

pathology for Hpylori Positive upper GI (confirming endoscopy)

1 1

Endoscopic Endoscopy

confirmation

study (upper or subdia-

GI)

by

stricture

In 1975, H2 receptor blockers were placed on the formulary at the Hospital for Sick Children. Since then, the incidence of surgical intervention for complications of PUD has decreased.2,5Jj Campylobacter pylori was isolated in 1983.‘,* It soon became associated with patients who had chronic gastritisgJO As a result, patients with PUD and biopsy-proven Hpylori for Surgery

A

B

C

Subtotal gastrectomy Other (5) Omental patch (6)

and Yearly

Incidence

of Surgery

”

Yearly Incidence

Bleeding Perforation

26

.96

7

26

Obstruction Bleedrng Perforation

4

.I5

0 2

Obstructron

B

DISCUSSION

2. indication

Procedure

Perforation

diagnosed

C

0

2

.I5

1

.I7

0

0

Obstruction

1

17

(2)

and pyloroplasty

(1)

Gastrolejunostomy (2) Vagotomy and pyloroplasty Vagotomy and antrectomy

(I) (I)

Perforatron

Omental patch (I) Gastrostomy and duodenostomy

Obstructron Bleeding

Vagotomy Vagotomy

and pyloroplasty and pyloroplasty

Obstruction

Vagotomy

and gastrojejunostomy

(I)

(2) (I) (I)

were treated with bismuth and antibiotics to cover gram-positive organisms. These results were published in Lancet in 1988.” In organizing our series, patients were grouped according to the medical treatment modalities available at the time of diagnosis. Omeprazole first became available in Ontario in 1988. Although its FDA approval in the United States was for the treatment of gastroesophageal reflux, it also has been used to treat severe ulcer disease because of its effect on gastric acid secretions and gastrin levels.’ Children, as opposed to adults, tend to present to physicians with long-standing disease.12J3 The endoscopic manifestation of this is duodenal scarring with multiple ulceration sites that, if untreated, develop strictures. Our data agree with the literature that shows a decline in the incidence of surgery after the introduction of H2 blockers. However, when the indications for surgery are analyzed, the incidence of surgery for obstruction does not decline despite the advances in medical management. This supports the theory that obstruction is a complication that occurs due to scarring from chronic disease. Most of these children in our report had long-standing symptoms. In this instance, medical therapy for PUD cannot be expected to relieve the obstructive component. The average length of medical therapy before surgery in the three patients operated on for obstruction since 1975 was 15 months. All patients in groups B and C operated on for obstruction received H2 blocker -+ omeprazole and were treated for H pylon’. In each instance, weight loss with failure to thrive was docuTable

.I5

Obstruction Bleeding Perforation

Indications

Vagotomy and pyloroplasty (14) Suture of bleeding ulcer (5)

Vagotomy stricture

localrzation of bleeding and upper GI to confirm

Group

on Specific

indmtlon

Bleeding

of PUD and positive

The indications and yearly incidence of surgery for the indications are noted in Table 2. The incidence of obstruction as an indication for surgery did not differ significantly different between the groups. However, the incidence of bleeding was significantly lower in groups B and C in comparison to A (P < .Ol). Perforation was not significantly lower in B or C than in A. The surgical procedures performed are described in Table 3. The results of surgery were assessed by the postoperative complications and the need for postoperative medical therapy. Follow-up medical treatment in group A consisted of continuation of preoperative therapy for pain in nine (24%) patients. In group B, two (50%) were placed on an H2 blocker postoperatively. In group C, one (50%) was placed on H2 blockers postoperatively. The complications are noted in Table 4. Groups B and C were combined for statistical analysis. There was no difference in the mortality rate or the complication rate before and after the H2 blocker era.

Table

A

Based

4. Postoperative

Group

Rebleeding

A

2of21 0 of 0 0 of 1

B C

Complications PaIn

Other

Death

9of37 Oof4

1 1

4of37 1 of4

0 of 2

0

0 of 2

AZAROW

752

mented during therapy. Postoperatively, each patient regained his/her lost weight and resumed normal growth. It was noted that only the patients with obstruction were treated for Hpyloti. Whether or not untreated H pylon’ predisposes to the development of gastric outlet obstruction as a complication of PUD is unknown. It has been reported that 25% of normal children have H pylon’ colonization. I3 It is also known that adults with PUD and positive colonization of the antrum with H pylon’ have a higher relapse rate when the peptic ulcer is treated with H2 blockers.14 If obstruction is believed to result from chronic disease, then colonization with Hpylori may be a risk factor for the development of gastric outlet obstruction. For patients with gastric outlet obstruction, balloon dilation of the pylorus is offered by some.15 Its long-term efficacy has not been proven.16-l9 Age is an exclusion criterion in many adult studies, and the short-term failure rate is high-30%.17 For these reasons we did not offer it to the children who had obstruction. In our study, the decrease in the use of surgery for bleeding was associated with the use of H2 blockers. Although interventional endoscopy was available, it was not attempted in the group C patient who had bleeding. He had multiple medical problems (coagulopathy, chronic renal failure, steroid therapy, and neurological impairment). Diagnostic esophagogastroduodenoscopy (EGD) showed that his duodenal bleeding could be dealt with more expeditiously by laparotomy. Although the incidence of surgery for perforation decreased between the A and C eras, there was no significant difference between the A and B eras. The cause of the perforations in group B was believed to be related to steroid use. One patient was on dexamethasone for an astrocytoma; the other was on prednisone for pulmonary disease. Drugs that affect the gastric mucosal barrier are thought to play much less of a role in children than in adults.13 However, there is support for the use of H2 blockers as prophylaxis when the drug used is a corticosteroid.20,21 Neither of our patients was on an H2 blocker before the perforation occurred.

ET AL

Our operations to treat the complications of PUD have not changed over the past 45 years. Vagotomy plus pyloroplasty remains our most popular procedure to manage bleeding. Simple suturing was used in five instances in group A, and two patients in group A had subtotal gastrectomies. Omental patch was the standard for a perforation. Vagotomy with a drainage procedure has been most often used in instances of obstruction. There was no difference in the rebleeding or mortality rate among the different procedures performed for any single indication. The postoperative complication and mortality rates were not significantly lower for groups B and C. However, the cause of death for the patients in group A were bleeding in two, and sepsis after duodenal leak in the other two. The patient in group B who died within 30 days of surgery, died of complications related to his astrocytoma. This represents the only complication or death in the post-H2 blocker era. Other than newer antibiotics, the only difference in the management of these patients was the use of newer ulcer medicines. Thus, despite the lack of statistical significance owing to the small patient numbers, the decrease in complications over time is believed to be of clinical significance. The complaint of postoperative pain is subjective. However, the fact that it was documented in the charts of 24% of group A patients and not noted for any group B or C patient is noteworthy. We attribute the persistent pain to those who had surgery for a complication and then did not receive an “ulcer operation” (ie, omental patch without vagotomy and drainage, suture of bleeding vessel without vagotomy and drainage, or bypass of obstruction without vagotomy). In such patients, if an ulcer operation is not performed, postoperative ulcer treatment is imperative.22,23This was not available before 1975. We conclude that although the incidence of surgery for PUD has declined, the incidence of surgery for obstruction secondary to PUD has not. The obstruction probably is related to scarring from long-standing disease. HpyZori may be a risk factor in the development of obstruction. Lesser procedures to treat the complications of PUD may be sufficient with appropriate ulcer medical treatment postoperatively.

REFERENCES 1. years 2. ulcer 3.

ation period 4.

Karlstron F: Peptic ulcer in children m Sweden during the 1953-1962. Ann Paediatr 202:218-221,1964 Fineberg HV, Pearlman LA: Surgical treatment of peptic in the United States. Lancet 1305-1307,198l Seagram CGF, Stephens CA, Cumming WA: Peptrc ulcerat the Hospital for Sick Children, Toronto, during the 20-year 1949-1969. J Pediatr Surg 8:407-413,1973 Drumm B, Rhoads JM, Stringer DA, et al: Peptic ulcer

disease in children: Etiology, climcal Pediatrics 82:410-414, 1988 5. Smith MP: Decline in duodenal

findings,

and clinical

surgery.

JAMA

course.

237:987-988,

1977

6. Wyllie JH, Williams JA, Kennedy TL, et al: Effect of cimetidine on surgery for duodenal ulcer. Lancet 1307-1308. 1981 7. Williamson RCN: Endocrine and other unusual pancreatic tumors. Curr Opin Gastroenterol3:749-755,1987

753

PEPTIC ULCER

8. Warren JR: Unidentified curved bactlli on gastrtc epithelium in active chronic gastritis. Lancet 2:1273-1275, 1983 9. Rathbone BJ, Wyatt J. Heatly RV: Cmzpyiobactev pyloridis-A new factor in peptic ulcer disease? Gut 27:635-641,1986 10. Drumm B, Sherman P, Cutz E, et al: Association of Campylobacterpyloti on the gastric mucosa with antral gastritis in children. N Engl J Med 316:1557-1561,1987 11. McNulty CAM: The treatment of Campylobacter-associated gastritis. Am J Gastroenterol82:245-247, 1987 12. Ravitch MM, Duremdes GD: Operative treatment of chronic duodenal ulcer in childhood. Ann Surg 171:641-646.1970 13. Eastham EJ: Peptic ulcer, in Walker WA, Durie PR, Hamilton JR, et al. (eds): Gastrointestinal Disease: Pathophysiology, Diagnosis, and Management, chap 26, part 3. Philadelphia, PA, Decker, 1991, pp 438-451 14. Coughlan G, Gilligan D, Humphries H, et al: Cumpylobucter pylon’ and recurrent duodenal ulcers-A 12-month follow-up study. Lancet 2:1109-1111, 1987 15. Hogan RB, Hamilton JK, Polter DE: Preliminary experience with hydrostatic balloon dilation of gastric outlet obstruction. Gastrointest Endosc 32:71-74,1986 16. Kozarek

RA,

Botoman

VA,

Patterson

DJ:

Long-term

fol-

low-up in patients who have undergone balloon dilation for gastric outlet obstruction. Gastrointest Endosc 36:558-561, 1990 17. McLean GK, Cooper GS, Hartz WH, et al: Radiologically guided balloon dilation of gastrointestinal strictures. Part I. Technique and factors influencing procedural success. Radiology 165:3540.1987 18. McLean GK, Cooper GS, Hartz WH, et al: Radiologically guided balloon dilation of gastrointestinal strictures. Part II. Results of long-term follow-up. Radiology 165:41-43,1987 19. Griffin SM, Chung SC, Leung JW, et al: Peptic pyloric stenosis treated by endoscopic balloon dilation. Br J Surg 76:11471148,1989 20. Cates J, Chavez M, Laks H. et al: Gastrointestinal complications after cardiac transplantation: A spectrum of diseases. Am J Gastrenterol 86:412-416,199l 21. Bickler SW. Harrison MW, Campbell JR: Perforated peptic ulcer disease in children: Association of corticosteroid therapy. J Pediatr Surg 28:785-787, 1993 22. Qureshi H, Maher M, Ahmed W, et al: Is the vagotomy complete? J Pak Med Assoc 41:279-280,199l 23. Aboula M, Boukheloua B, Salem A, et al: Chrome duodenal ulcer in children and response to highly selective vagotomy. Eur J Pedratr Surg 2:5-7. 1992