A 59-year-old white man presented with a complaint of mild right iliac fossa pain of 2 weeks' duration. His medical history was significant for a herniated lumbar disk that had been confirmed on magnetic resonance imaging and treated medically 2 years earlier. This was followed by a recent episode of mild lower backache preceding the onset of the right iliac fossa pain, for which the patient took naproxen sodium, 440 mg, at nighttime. A serendipitously scheduled surveillance colonoscopy showed a solitary 1 × 1.2-cm, oval, sharply delineated, whitish ulcer in the ascending colon, located 2 cm above the ileocecal valve on the opposite wall and involving a haustral crest (Figure 1, right margin). The rest of the large bowel appeared normal. The endoscopist noted that the ulcer edge “felt soft” to the biopsy forceps. A biopsy of the ulcer edge was taken. Hematoxylin-eosin–stained histologic sections were prepared from formalin-fixed, paraffin-embedded tissue.

Microscopic examination of the lesion showed superficial ulceration with a surface fibrinopurulent exudate (Figures 2 and 3). The underlying mucosa showed fibrosis of the lamina propria as well as evidence of mild glandular regenerative atypia (Figure 4).

What is your diagnosis?

The histologic features of superficial epithelial loss, inflammatory pseudomembranous exudate, mild regenerative atypia, and mucosal fibroplasia were consistent with a focal area of low-grade or chronic ischemia.

Following the patient's discontinuation of the naproxen sodium, the right iliac fossa pain gradually subsided, and a repeat colonoscopy 3 months later showed no sign of a colon lesion. The patient remains symptom free 6 months later.

Naproxen sodium belongs to the nonsteroidal anti-inflammatory group of drugs (NSAIDs). Naproxen and other NSAIDs are a well-established cause of gastrointestinal tract injury.1–3 This may occur even after short-term use.1 The mechanism of NSAIDs injury is not definitively known. The stomach and duodenum bear the brunt of the injury.1,3 In this case, 2 mechanisms are postulated to be responsible.3,4 The first mechanism is the direct mucosal damaging effect of NSAIDs.3 Back diffusion of gastric acid occurs with resultant peptic ulceration.5 This is due to topical loss of cytoprotective and reparative mucosal epithelial prostaglandins as a result of inhibition by NSAIDs of the mucosal cyclooxygenase enzyme system responsible for their biosynthesis from arachidonic acid.1,5 Mucosal damage may be exacerbated as a result of the local loss of prostaglandin-mediated mucus and bicarbonate secretions normally providing a barrier to back diffusion.5 A similar topical effect may apply in the large bowel, since sustained-release NSAIDs tablets have been found within colonic diverticulae and in the abdominal cavity near a perforation.6 Rectal lesions have been described in patients using NSAIDs suppositories, which provides evidence of direct mucosal damage.7 Sustained-release and enteric-coated tablets reduce the incidence of gastroduodenal ulcerations but increase enterocolonic complications, possibly as a result of contact toxicity causing the uncoupling of mitochondrial adenosine triphosphate and the subsequent breakdown of intercellular junctions.8 It has been suggested that topical cyclooxygenase enzyme inhibition favors the production of inflammatory leukotrienes in the colonic mucosa.4,8 The second mechanism is an indirect mucosal damaging effect seen throughout the gastrointestinal tract due to systemic cyclooxygenase enzyme inhibition and a decrease in cytoprotective and reparative mucosal epithelial prostaglandin production.4,8 Compelling evidence for a systemic pathway is provided by examples of gastric and colonic ulcerogenesis following the intramuscular administration of NSAIDs.4 

Most accounts of NSAIDs-associated gastrointestinal tract disease have appeared in the clinical literature, with relatively little attention being given in the pathologic literature. Histologic descriptions of NSAIDs-associated colonic ulcers have generally been vague; however, focal mucosal fibrosis, similar to that observed in this case, has been reported.1,2,4,8 This is suggestive of an ischemic etiology, although the clinical and endoscopic findings do not support a diagnosis of conventional ischemic colitis.1,2,4 A vascular mechanism may indeed play a role in this localized mucosal fibroplasia. Prostaglandins synthesized within the mucosa not only exert a cytoprotective effect but also increase mucosal blood flow.5 Their inhibition may therefore contribute to mucosal injury through a reduction in local mucosal blood flow. This hypothesis is supported by the association of NSAIDs with renal dysfunction and even acute renal failure attributable to a loss of vasodilatory control as a consequence of prostaglandin inhibition.9 

It is becoming increasingly apparent that NSAIDs-induced lesions of the large intestine, although relatively uncommon, are a significant cause of abdominal pain, anemia due to colonic bleeding, diarrhea, ulceration, perforation, ulcerated and nonulcerated strictures, patchy erosive colitis, collagenous colitis, diffuse colitis, and even death.2 The right side of the colon, especially the ileocecal region, is a favored site for both solitary and multiple ulcers.4,7 Lesions may be localized to haustral crests.6 Typically, the ulcers are sharply demarcated and surrounded by normal mucosa, and the remaining part of the colon is normal in appearance.10 These ulcers heal rapidly after the cessation of NSAIDs therapy.4,6,7 Some patients present initially with concentric, diaphragm-like strictures apparently resulting from stenotic healing of haustral ulcerations.6–8 Such strictures may clinically mimic Crohn disease and colon carcinoma and probably result from the prolonged use of NSAIDs.6 

Since the presence of a sizeable, single colon ulcer may suggest a colon carcinoma, it is easy to imagine a situation in which fragmented biopsy material showing regenerative glandular atypia and fibroplasia might lead to a mistaken diagnosis of colon cancer, a pitfall that can be avoided by close interaction with the endoscopist, acquaintance with the endoscopic appearance of the lesion, and, above all, obtainment of a history of NSAIDs use.8 

Because NSAIDs are so frequently prescribed and extensively used as over-the-counter medication, there is a compelling need for a heightened awareness, among both clinicians and pathologists, of their association with a wide variety of colopathic changes. This awareness might help avoid inappropriate management, such as colon resection for suspected colon cancer and medical treatment for inflammatory bowel disease.10 Except for cases of strictures and perforations, in which surgery may be indicated, the cessation of NSAIDs use is the only therapy required.4,6–8 

Uribe
,
A.
,
C.
Johnson
,
P.
Slezak
, and
C.
Rubio
.
Ulcerations of the colon associated with naproxen.
Gastrointest Endosc
1986
.
32
:
242
244
.
Puspok
,
A.
,
H-P.
Kiener
, and
G.
Oberhuber
.
Clinical, endoscopic and histologic spectrum of nonsteroidal anti-inflammatory drug-induced lesions in the colon.
Dis Colon Rectum
2000
.
43
:
685
691
.
Dajani
,
E. Z.
Gastrointestinal toxicity of over-the-counter analgesics.
Am J Gastroenterol
1998
.
93
:
1020
1022
.
Buchman
,
A. L.
and
M. R.
Schwartz
.
Colonic ulceration associated with systemic use of nonsteroidal anti-inflammatory medication.
J Clin Gastroenterol
1996
.
22
:
224
226
.
Bowen
,
R. A.
The gastrointestinal barrier.
In: Pathophysiology of the Digestive System. Ft Collins: Colorado State University Hypertexts for Biomedical Sciences; 1998:1–6
.
Halter
,
F.
,
B.
Weber
,
T.
Huber
,
F.
Eigenman
,
M. P.
Frey
, and
C.
Ruchti
.
Diaphragm disease of the ascending colon.
J Clin Gastroenterol
1993
.
16
:
74
80
.
Kaufman
,
H. L.
,
A. H.
Fischer
,
M.
Carroll
, and
J. M.
Becker
.
Colonic ulceration associated with nonsteroidal anti-inflammatory drugs.
Dis Colon Rectum
1996
.
39
:
705
710
.
Stamm
,
C.
,
C. E.
Burkhalter
, and
W.
Pearce
.
et al
.
Benign colonic ulcers associated with nonsteroidal anti-inflammatory drug ingestion.
Am J Gastroenterol
2002
.
89
:
2230
2233
.
Koch
,
J-E.
,
J.
Plum
,
B.
Grabensee
, and
U.
Modder
.
Prostaglandin E1: a new agent for the prevention of renal dysfunction in high risk patients caused by radiocontrast media?
Nephrol Dial Transplant
2000
.
15
:
43
49
.
Kirsch
,
M.
Nonsteroidal anti-inflammatory drug colopathy: mimicry of Crohn's disease and colon carcinoma.
J Clin Gastroenterol
1997
.
24
:
121
123
.

Author notes

Corresponding author: Stanley Lipper, MD, 42 Carriage Rd, Great Neck, NY 11024 (lipper@med.cuny.edu)