Nonsteroidal Antiinflammatory Drug-Induced Colonic Strictures: A Case Report THOMAS
HUBER, CHARLES RUCHTI, and FRED HALTER
Gastrointestinal Unit and Department of Pathology, University of Berne, Berne, Switzerland
Adverse effects of nonsteroidal antiinflammatory drugs can occur throughout the whole gastrointestinal tract. Recently, several cases of “diaphragmlike” thin ileal strictures have been reported. These strictures seem to result from nonsteroidal antiinflammatory drug-induced inflammatory changes and apparently represent a newly recognized nosological entity. The case of a 61-year-old man who gradually developed similar inflammatory changes in the ascending colon during prolonged intake of a slow-release form of diclofenac is presented, and the literature on nonsteroidal antiinflammatory drug-induced intestinal strictures is briefly reviewed.
onsteroidal antiinflammatory drugs (NSAIDs) are well known to produce gastrointestinal lesions, most of which are in the stomach and duodenum. However, many observations in experimental animals (l,2) and in humans (3-10) indicate that both the small bowel and the colon are also important targets of adverse effects of NSAIDs. The spectrum of such lesions ranges from frank ulcerations, often complicated by perforations and acute bleeding, to changes indistinguishable from idiopathic inflammatory bowel disease (i.e., ulcerative colitis and Crohn’s disease). Several case reports have recently been published documenting an association of small bowel strictures with intake of NSAIDs (11 ,12). In some of these cases, strictures had a typical, uniform macroscopic and microscopic morphology, which is well characterized by the term “diaphragmlike disease” (12). This pathology seems to represent a nosologic entity (13). To our knowledge, it has previously been observed only in the small bowel. We present the case of a patient with a history of prolonged NSAID intake who developed macroscopic and microscopic lesions in the ascending colon closely resembling the morphological changes observed in diaphragmlike disease of the small bowel.
Case Report A 61-year-old man was referred to our unit by his general physician because of an iron deficiency anemia of z years’ duration. Hemoglobin values fluctuated between 9.3 and lo.5 g/100 mL. There had been no improvement following endoscopically verified amelioration of reflux esophagitis with Hz-blocker treatment. The patient had been on slow-release diclofenac therapy for several years because of low back pain and arthralgias, as a cause of which only degenerative arthritis had been found. In May 1986 he had been admitted to our hospital for evaluation of this anemia, having had positive results of occult blood tests of the stool and one questionable episode of melena. He was found to have an elevated sedimentation rate of 63 mm/h. Gastroscopy and colonosopy revealed no pathology. Despite thorough investigation, no diagnosis could be established and the patient was discharged with the recommendation of further observation. Some weeks thereafter, he was reevaluated because of diarrhea and anal discharge of bloody mucus. Colonoscopy revealed several partially confluent ulcers with nonspecific histology in the ascending colon near the ileocecal valve. These lesions had an aggregate diameter of about 2 x 2 cm. Because the signs mentioned above disappeared spontaneously, no further specific diagnostic or therapeutic measures were felt to be necessary. Two years later, in July 1988, the increasing irondeficiency anemia and persistently elevated sedimentation rate, in absence of abdominal complaints, led to a further clinical evaluation. A grade III reflux esophagitis (circular confluent erosions], in conjunction with a hiatal hernia, was seen at gastroscopy. There was no change in hemoglobin levels, despite endoscopically verified amelioration of esophagitis to grade II (linear longitudinal confluent erosions] after 3 months of therapy with Hz-receptor antagonists. At colonoscopy, the cecum and the proximal ascending colon showed the following picture: multiple diaphragmlike, rigid, thin circumferential septa divided the
Abbreviation used in this paper: NSAID, nonsteroidal antiinflammatory drug. Q 1991 by tbe American Gastroenterological Association 0016-5095/91/$3.00
lumen into multiple compartments. The orifices had a diameter of approximately z cm. The mucosal layer seemed to be normal, with the exception of two wartlike structures near the orifice of one septum. The colonic wall between these membranes was macroscopically normal (Figure 1). Multiple biopsies revealed submucosal fibrosis with dense apical collagen fibers, in part aligned toward the mucosal surface (Figure 2A). The mucosa either was intact or featured distorted crypts with irregularly arranged and damaged epithelial cells and mild to moderate chronic submucosal inflammation. There were some incomplete erosions (Figure 2B). The two described w&like structures histologically proved to be benign hyperplastic polyps. The rest of the colon looked normal, and therefore biopsies were not performed. The patient was subsequently advised to discontinue his NSAID treatment, a recommendation with which he was only partially compliant. Thereafter, hemoglobin stabilized within the normal range (14.3 g/l00 mL) and the elevated blood sedimentation rate decreased toward normal levels. In January 1990, another year later, gastroscopy revealed persistence of grade II esophagitis despite continuing H,blocker therapy and normalization of blood hemoglobin levels. A double-contrast examination of the colon showed slight infiltrative changes and rigidity of the wall in the ascending colon without signs of mucosal damage. No pathological changes were present in the terminal ileum. Unfortunately, further colonoscopic follow-up has been refused by the patient. Discussion The patient presented here developed diaphragmlike lesions in the ascending colon during
Figure 1. Endoscopic aspect of the ascending colon (Olympus Videocolonoscope). The lumen is divided into several compartments by diaphragmlike septa (arrows). Highly friable mucosa is demonstrated by bleeding after touching with biopsy forceps. When air was introduced, the strictures appeared to be fixed and the gut wall appeared to have a greatly reduced compliance.
GASTROENTEROLOGY Vol. 100,
regular intake of a slow-release NSAID. These lesions macroscopically and microscopically closely resembled those previously observed in the small bowel under NSAID therapy (12,131. Because of identical morphology to those described in the small bowel (12,13) and the associated history of chronic NSAID intake, we propose that these alterations are of the same nature. Most data on NSAID-induced pathology of the lower gastrointestinal tract are based on animal experiments (1,2), human in vivo studies (14-17), epidemiologic investigations (3), and clinical observations (4-10). Such lesions are probably more frequent than clinically suspected because experimental data in humans show inflammatory side effects in the small intestines in > 60% of subjects (14-17). The macroscopic and microscopic morphology of the observed lesions is rather nonspecific and ranges from simple ulcerations to changes indistinguishable from those seen in idiopathic inflammatory bowel disease. Diaphragmlike strictures of the small bowel associated with intake of NSAIDs (12,13) seem to represent a subgroup with a typical macroscopic and a more or less typical microscopic morphology. A localized, submucosal, fibrous reaction consequent to NSAID-mediated lesions has been proposed as a possible cause for this phenomenon by Lang et al. (13). This diaphragmlike formation could be a special type of mucosal reaction (e.g., by certain gut segments or to higher NSAID doses) or could be a later stage in the progression of the inflammatory response. The latter could have been the case in our patient, in whom histologically nonspecific ulcerations had been found at the same location 2 years before observation of the strictures. Such cecal ulcers have been found by others (18-20) and attributed to NSAIDs (19,20). The pathogenic mechanism leading to such inflammatory changes in the lower gastrointestinal tract is currently unknown. A possible mode of action may be drug-induced changes of local eicosanoid metabolism, resulting in enhanced permeability of the mucosa to toxins and luminal antigens (14-17). In addition, NSAIDs have been shown to affect chemotaxis and neutrophile function (21,22). The reason for the development of such a specific pathology in the ascending colon remains obscure. Nonsteroidal antiinflammatory drugs can theoretically exert their effect on gut mucosa from the luminal side or systemically. Topical damage is a more likely explanation for the observed changes, and one can assume that the intraluminal NSAID concentration is an important factor. The localization of the lesions in the cecum and proximal ascending colon in our patient could be explained by topical action because he was taking diclofenac in a slow-release form. Unfortunately, radiologic examination was possible
NSAID-INDUCED COLONIC STRICTURES
Figure 2. Histology of biopsy specimens
(H&E stain: calibration bar = 200 pm).
A. Apical submucosal fibrosis with preserved mucosa in one biopsy specimen. Fibrosis with dense apical collagen fibers, in part arranged toward the mucosal surface, is present. B. Another specimen showing distorted apical mucosal crypts with damaged and irregularly arranged epithelial cells and moderate chronic mucosal inflammation with incomplete erosions.
only in the terminal ileum and only one year after the diagnostic colonoscopy because of lacking patient cooperation. Because the patient had clinically improved by that time, we are unable to exclude the initial presence of pathological changes in the ileum. However, it should be kept in mind that in the majority of cases even expert radiological studies are unable to demonstrate clearly the specific small bowel changes (23). Accordingly, the radiological findings in this patient’s ascending colon were not specific and could easily have been overlooked by an unwary observer. Amelioration of anemia following restriction of NSAID therapy, despite unchanged reflux esophagitis, indirectly supports the hypothesis of a causal relationship between the intestinal pathology and NSAIDs.
2. Fang W, Broughton
a. 9. 10.
References 1. Kent TH, Cardelli RM, Stammler FW. Small intestinal ulcers and intestinal flora in rats given indomethacin. 1969;54:237-245.
Am J Path01 11.
A, Jacobson ED. Indomethacin-induced intestinal inflammation. Dig Dis Sci 1977;22:749-760. Langman MJ, Morgan L, Worrall A. Use of anti-inflammatory drugs by patients admitted with small or large bowel perforations and hemorrhage. BMJ 1985;290:347-349. Kaufmann HJ, Taubin HL. Nonsteroidal anti-inflammatory drugs activate quiescent inflammatory bowel disease. Am J Med 1987;107:513-516. Giiller R. Die Nebenwirkungen nicht-steroidaler Antirheumatika im unteren Gastrointestinaltrakt. Schw Med Wschr 1987; 117:1527-1533. NSAID and gut damage (editorial). Lancet 1989;2:600. Banerjee AK. Enteropathy induced by non-steroidal antiinflammatory drugs. Often subclinical, but may mimic Crohn’s disease. BMJ 1989;298:1539-1540, Rampton DS. Non-steroidal anti-inflammatory drugs and lower gastrointestinal tract. Stand J Gastroenterol1987;22:1-4. Ravi S, Keat AC, Keat ECB. Colitis caused by NSAID. Postgrad Med J 1986;62:773-776. Tanner AR, Raghunath AS. Colonic inflammation and nonsteroidal anti-inflammatory drug administration. An assessment of the frequency of the problem. Digestion 1988;41:116120. Saveryumuttu A, Thomas A, Grundy A, Maxwell JD. Jleal
stricturing after long-term indomethacin treatment. Postgrad Med J 1986;62:967-968. Bjarnason I, Price AB, Zanelli G, Smethurst P, Burke M, Gumpel JM, Levi AJ. Clinicopathological features of nonsteroidal anti-inflammatory drug-induced small intestinal strictures. Gastroenterology 1988;94:1070-1074. Lang J, Price AB, Levi AJ, Burke M, Gumpel JM, Bjarnason I. Diaphragm disease: pathology of disease of the small intestine induced by non-steroidal anti-inflammatory drugs. J Clin Path01 1988;41:516-526. Bjarnason I, Zanelli, Smith T, Rouse P, Williams P, Smethurst P, Delacey G, Gumpel MJ, Levi AJ. Nonsteroidal antiinflammatory drug-induced intestinal inflammation in humans. Gastroenterology 1987;93:480-489. Bjarnason I, Williams P, Smethurst P, Peters TJ, Levi AJ. The effect of non-steroidal anti-inflammatory drugs and prostaglandins on the permeability of the human small intestine. Gut 1986;27:1292-1297, Bjarnason I, Williams P, So A, Zanelli G, Levi AJ, Gumpel MJ, Peters TJ, Ansell B. Intestinal permeability and inflammation in rheumatoid arthritis: effects of non-steroidal antiinilammatory drugs. Lancet 1984;2:1171-1174.
GASTROENTEROLOGY Vol. 100, No. 4
17. Bjarnason I, Peters TJ, Levi AJ. Intestinal permeability: clinical correlates. Dig Dis 1986;4:83-92. 18. Shallman RW, Kuehner M, Williams GH, Sajjad S, Sautter R. Benign cecal ulcers. Spectrum of disease and selective management. Dis Co1 Rect 1985;28:732-737. 19. Charuzi I, Ovnat A, Zirkin H, Peiser J, Sukenik S. Ibuprofen and benign cecal ulcer. J Rheumatol1985;12:1:188-189. 20. Debenham GP. Ulcer of the cecum during oxyphenbutazone therapy. Can Med Assoc J 1966;94:1182-1184. 21. Warne PJ, West GB. Inhibition of leucocyte migration by salicylates and indomethacin. J Pharm Pharmacol1978;30:783785. 22. Dahinden C, Fehr J. Receptor-directed inhibition of chemotactic factor induced neutropic hyperactivity by pyrazolon derivatives. J Clin Invest 1980;66:884-891, 23. Bjarnason I, Gumpel JM. Enteropathy induced by non-steroidal anti-inflammatory drugs (letter). BMJ 1989;299:326.
Received March 5,199O. Accepted September 20,199O. Address requests for reprints to: Prof. Fred Halter, Gastrointestinal Unit, Inselspital3010 Bern, Switzerland.