Surgical treatment of drug-induced gastroduodenal perforations

Military Medicine, Jan 2002 by Garner, Jeffrey P

The optimum method of managing perforated peptic ulcers remains controversial, particularly when associated with the ingestion of ulcerogenic drugs. We reviewed the notes for a series of patients treated surgically by simple closure of druginduced peptic ulcer perforations to identify whether simple closure and omental patching alone was an effective treatment in these patients. We also wrote to their primary care physicians to obtain details of any further dyspeptic symptoms, continued ulcerogenic drug use, or the need for further antiulcer therapy or surgery. Thirty patients underwent simple closure and omental patching. The overall mortality rate was 27% but increased to 38% in patients older than 80 years. Aspirin was implicated in perforations in only seven cases, and steroids alone were not implicated in any cases. Eight patients have subsequently restarted ulcerogenic drugs without further dyspeptic symptoms. We conclude that simple closure is a safe and effective means of treating perforated peptic ulcers associated with ulcerogenic drugs, even if these are restarted.

Introduction

It is well recognized that the profile of gastroduodenal perforation is changing, but debate still rages regarding the best management of this disease. Taylor's conservative management regimen seems to have lost favor at present,1 and the choice lies between simple closure of the perforation and closure of the perforation accompanied by some form of definitive anti-ulcer surgery. It is now clear that ingestion of ulcerogenic drugs (nonsteroidal anti-inflammatory drugs [NSAIDs]/aspirin/steroids) is an independent risk factor for perforation2 and may form a pathologically distinct subset of perforation patients.3 This study reviewed the trends in such perforations and confirmed that simple closure is a safe and effective method of treatment even when ulcerogenic drugs are reintroduced. The literature on this subject is reviewed.

Patients and Methods

This study evaluated 30 consecutive patients over 7 years (1991-1997) who were taking ulcerogenic drugs for a variable period of time before they were operated on for gastroduodenal perforation at Frimley Park Hospital, Surrey, England. We reviewed the demographic profiles of these patients, duration of ulcerogenic drug usage, sites of perforation, concurrent medical conditions, and eventual outcome by case note analysis. Their follow-up over a mean period of 29 months (range, 1-96 months) has been augmented by a questionnaire sent to the patients' general practitioners asking for details of postoperative persistence of symptoms, postoperative anti-ulcer drug treatment, continuation of ulcerogenic drug therapy, and the need for subsequent anti-ulcer surgery. Mortality within 30 days of operation and significant medical complications were recorded. Mortality figures were compared using Fisher's exact test calculated using a statistical software package on a personal computer. (SPSS version 10, SPSS Inc., Chicago, Illinois)

Results

Thirty patients were treated surgically by simple closure and omental patching for gastroduodenal perforations associated with recent ingestion of ulcerogenic drugs during the 7-year study period. There were 21 women and 9 men with a mean age of 76 years (range, 36-96 years). Twenty patients (67%) were older than 70 years, and 8 patients (27%) were older than 80 years (Table I). Information on the duration of ulcerogenic drug therapy was available for only 22 patients, but of these, 15 (68%) had been taking an ulcerogenic drug for more than 3 months and 12 (55%) had been taking them for more than 1 year. Twenty-six patients (87%) were taking an NSAID at the time of perforation; the combinations of ulcerogenic drugs are listed in Table II.

Twenty-five patients had a total of 30 serious concomitant medical problems, which are listed in Table III. Of the 30 patients who perforated while on ulcerogenic drugs, only 7 (23%) were also taking some form of anti-ulcer medication (5 patients on H2 receptor blockers, 1 patient on antacids, and 1 patient on a proton pump inhibitor). Table IV identifies the sites of perforation. Twenty-four significant complications were experienced by 15 patients (including those who died subsequently); these are listed in Table V. Eight patients (27%) died as a direct result of their perforation or its sequelae (Table VI). Six of the eight (75%) who died were female, which gives a female mortality rate in this study of 28% compared with a male mortality rate of 22% (p = 1.00). Of the eight deaths, only 1 (13%) occurred in a patient younger than 70 years, whereas 3 (38%) occurred in patients older than 90 years. Seven of the 20 patients (35%) older than 70 years died, compared with 1 of 10 patients (10%) younger than 70 years (p = 0.21).

Twenty two patients were discharged from the hospital after their perforations and were followed for an average of 29 months (range, 1-96 months). One patient reperforated 3 days after discharge but was then successfully managed conservatively. Five other patients subsequently developed dyspeptic symptoms between 1 and 64 months (mean, 22 months) after their perforations. All underwent upper gastrointestinal endoscopy and were found to have varying degrees of gastritis, which subsequently resolved after Helicobacter pylori eradication therapy; all of these patients were free of symptoms 3 months after eradication therapy. Seven of those who were discharged from the hospital have subsequently died from other diseases.