Colonic ulcerations associated with diclof enac treatment

Nonsteroidal anti-inflammatory drugs frequently exert toxic effects on the upper gastrointestinal mucosa. However, inflammation and ulcerations in the colon secondary to these agents have been increasingly recognized in recent years. A patient presenting with abdominal pain, weight loss and diarrhea while taking a 
slow release form of diclofenac sodium (Voltaren SR; Ciba-Geigy) is described. Extensive ulcerations in the right colon and healing upon discontinuation of the drug are documented by colonoscopy.

N ONSTEROIOAL ANTI-INFLAMMA- tory drugs (NSA IDs) arc frequently prescribed med ications with widely known adverse effects on the upper gastrointest inal mucosa.In recent years evidence has accumulated that NSAIDs a rc also potentia ll y tox ic 10 the lower imcstinal tract and rhat c hron ic use may lead to sma ll intestinal inflammat ion in a la rge numher o( patients ( I ).It appears t hat colonic inju ry caused by NSAIDs may account for a significa nt number of cases of newly diagnosed coli tis ( 2).
The aut hors describe the occurrence of colonic ulce ration in a patient caused by a slow release preparation of <liclofcnac sod ium (Voltaren SR; C iha-Gcigy).

CASE HISTORY
A 73-ycar-ol<l ma le initiall y presented wirh a 'flu-like illness and increasing a rth ralgias whi le on vacation .He was prescribed <l iclofcnac sodi um (Voltarcn SR; C iba-Geigy) at a <lose of 75 mg The patient had no pre\'ious gastromtcstinal complaints and he reported always having had regular howcl hahits.Past medical history mcluded mild hypertension, pacemaker dependency and srahlc angina pectons tor which he was takmg atenolol and nu roglycenn.There was no recent change or detcrioraunn of cardiovascu lar status.
Physical exammmion revealed n nmmntemive afcbrde male.There were nn signs of heart fodure.The abdomen was sl 1ghtl y distended and tL•nder 111 hot h lower quadrants.Apart from diffuse osteoart hrn ic ch anges 111 the L'Xt remit ies and a pacemaker in the left upper chest, the examtnatinn was unremarkabk.
A barium cm:ma shmved small sigmoid di,•erticula.An upper gast ro1ntestinal series and small howel barium study was normal wil h good \'iSualizatton of the terminal ileum.
A colnnoscopy revealed two partially digested pilb in the cecal tip as well Figure 2) Endoscopiet•iew of linear ukmuion nl'ar the hepaut fie \"ltrt!as erythema and a small area of ulcernuon just proximal to the ileocccal valve (Figure l ).However, there was extensive linear ulceration at the hcpati<.flexure extcnJing for almoM 10 cm into the proximal transverse colon (Figure 2).The surround ing mucosa was normal in appearatKL' as was the Jistal <.olon.B1llps1es obtained from rhc ul-<.ersshowed inflammatory infiltrallon ,md granulation tissue with no evidence of granulomas.
L ) iclofenac sodium was d1scont1nued and the patient was started on 5amimlsalicylic acid (Asam!; Norwich Eaton) tid 800 mg.Symptoms 1111-pro\'cJ anJ disappeared after approximately two weeks; he stopped taking Asacol two months later.A repeat colonu:,copy six months later was normal except for an obvious scar a t the pre-\'ious site of ukeration near the hcpatt<.flexure (Figure 3 ).Hemoglohin was 139 g/L.The pauent has remained asymptomatic after one year of follow-up.

DISCUSSION
The association bet ween NSAlns and lower gastrointestinal inf1ammarnm is well described in the literature and was rL'\'teWL'd recently (3 ).
The c.lin1cal course of the pm,cnt patient, the lack of prcd1sposmg factors except for diclofcnac, and the colllnoscopic finding of medication 111 the vicinity of the ulcerated mucosa arc in keeping with NSAll)-111duccd col 1tb.The authors could nor identify the pilb with certainty as Voltarcn SR bur their presence 1n the cecum suggests rclati\'e stasis in this part of the colon facd1tat-111g prolonged direct coniact bet ween the mucosa and the injurious agent.