Celecoxib-induced upper gastrointestinal hemorrhage and ulceration. (Case Reports).

ABSTRACT: COX-2 inhibitors are a new class of nonsteroidal anti-inflammatory drugs with a reported benefit of less gastric and duodenal ulceration and hemorrhage. We describe a 67-year-old man taking a higher than usual dose of celecoxib (Celebrex) for osteoarthritis with resultant gastric erosions, ulceration, and a significant gastrointestinal (GI) hemorrhage.

**********

THE NEWEST DRUGS in the class of nonsteroidal anti-inflammatory drugs (NSAIDs) are the COX-2 inhibitors, which have been designed to selectively inhibit cyclooxegenase-2 (COX-2), while sparing cyclooxegenase-1 (COX-1). The purported benefit of selective inhibition of COX-2 is effective anti-inflammatory activity with decreased gastric and duodenal ulceration and bleeding for patients with osteoarthritis and rheumatoid arthritis. We describe a patient who had a bleeding duodenal ulcer and significant gastric ulceration while taking supratherapeutic doses of celecoxib (Celebrex), the first of the COX-2 drugs to be approved in the United States. CASE REPORT

A 67-year-old white man was admitted after an influenza vaccine-induced cellulitis failed to resolve with outpatient antibiotics and antipyretics. The patient had received the vaccination 5 days earlier, followed by fever (T up to 101.5[degrees]F), confusion, left axillary lymphadenopathy, anorexia, nausea and vomiting, diarrhea, and general malaise. He was treated with cephalexin (500 mg tid) and acetaminophen on an outpatient basis, without resolution of symptoms, and was admitted with a diagnosis of left axillary and anterior chest wall cellulitis.

The patient's medical history included hyperlipidemia, hypertension, and coronary artery disease, with angioplasty, atherectomy, and stent placement within the past 6 months. The patient had no history of peptic ulcer disease, dyspepsia, or any other upper GI symptoms before this admission. He was taking aspirin (81 mg daily) but was not taking any other antiplatelet medications at the time of admission. He had also been taking celecoxib (600 mg daily) for the past 14 weeks. His other medications included glucosamine (500 mg tid), niacin (3 g/day), simvastatin (10 mg qhs), vitamin E, and a multivitamin supplement.

Admission vital signs were temperature 98.0[degrees]F, blood pressure 110/60 mm Hg, pulse rate 150/mm, and respiratory rate 20/mm. Physical examination on admission revealed a well-developed, well-nourished white man in no apparent distress. Cardiac evaluation revealed tachycardia (150 beats per minute), with atrial flutter 2:1 shown on electrocardiogram. No murmurs were heard. Examination of the anterior chest wall revealed an area of erythema extending medially from the left axilla to encompass the left nipple, inferiorly to the superior border of the iliac crest, and about 4 cm posteriorly, onto the patient's back. Physical examination was otherwise unremarkable. Admitting laboratory values were white blood cell count, 26,900/[mm.sup.3] hemoglobin, 11.9 g/dL (normal, 14 to 18 g/dL); hemotocrit, 34.4% (normal, 42% to 52%); platelet count, 152,000/[mm.sup.3] (normal, 150,000/[mm.sup.3] to 450,000/[mm.sup.3); and mean corpuscular volume, 95 [[micro]m.sup.3] (normal, 80 to 94 [[micro]m.sup.3]).

Blood cultures were obtained, and treatment was started with IV vancomycin and piperacillin/tazobactam. The newly diagnosed atrial flutter was treated with cardioversion, along with flecainide (50 mg bid) and digoxin (0.25 mg/day). The patient's usual medications were continued, including celecoxib, but the dose was decreased to the maximal recommended dose of 200 mg/day.

The patient's condition improved, and he was returned to normal sinus rhythm. One of two blood cultures was positive for gram-positive cocci in chains, subsequently shown to be Streptococcus salivarius. Sequential hemograms revealed worsening anemia, with a progressive decline to a hemoglobin value of 5 g/dL. Melenic stools and orthostatic symptoms also developed.

The patient was transferred to the intensive care unit and had emergency upper endoscopy. No esophageal mucosal abnormalities were found, but multiple 5 to 8 mm gastric ulcerations were seen, primarily in the antrum and also in the gastric body (Fig 1). A 1.5 cm clean-based ulceration was seen on the posterior wall of the duodenal bulb.

A second 1.0 cm duodenal ulcer near the junction of the first and second portions of the duodenum had an overlying clot and minor oozing (Fig 2), and 7 mL of 1:10,000 epinephrine was injected circumferentially into this ulcer. Complete hemostasis was not achieved, however, until a bipolar probe was used to treat the continued bleeding. Blood transfusion beginning at the time of endoscopy subsequently increased the hemoglobin level from 5 g/dL to 8 g/dL. Celecoxib therapy was discontinued, and lansoprazole therapy was started at a dosage of 30 mg/day. A test for Helicobacter pylori antibody was negative. The patient was discharged after an additional 48 hours of observation. The hemoglobin value was 9.7 g/dL.

DISCUSSION

Nonselective nonsteroidal anti-inflammatory drugs are the initial treatment for rheumatoid arthritis and have, until recently, been a primary drug for the treatment of pain associated with osteoarthritis. These agents are also commonly used for the relief of pain and inflammation associated with a wide array of conditions, but they have potentially serious GI side effects. Clinical trials have shown that approximately 20% of arthritic patients who used NSAIDs had endoscopically detected erosions. (1) An estimated 17 million people in the United States take prescription NSAIDs regularly, (2) and NSAID-induced GI complications account for approximately 107,000 hospitalizations each year. (3)

Nonsteroidal anti-inflammatory drugs reduce pain and inflammation by inhibiting cyclooxegenase, an enzyme in the pathway for formation of prostaglandins, and thromboxane. Needleman et al?? demonstrated two isoforms of cyclooxygenase, COX-1 and COX-2. It has since been determined that COX-1 products are responsible for maintenance roles in GI, renal, and platelet physiology, (4) and that inhibition of COX-1 is responsible for the majority of the gastric mucosal injury noted with nonselective NSAIDs. (5) COX-2 products have been shown to be responsible for modulating pain and inflammation. (6,7) From this information, it was hypothesized that if a drug could be synthesized that would target COX-2 without inhibiting COX-1, pain and inflammation could be reduced without the GI irritation brought on by inhibition of COX-1. Such drugs have been synthesized and are now known as the COX-2 selective NSAIDs (celecoxib, rofecoxib).

The COX-2 inhibitors are an exciting new class of drugs that may give much relief to patients with inflammatory conditions such as rheumatoid arthritis, while causing less GI toxicity than conventional NSAIDs. In fact, Simon et al (8) have shown in four phase II trials that celecoxib is effective in treating the signs and symptoms of osteoarthritis and rheumatoid arthritis, as well as showing that celecoxib is statistically significantly less likely than nonselective NSAIDs to cause gastric or duodenal erosions/ulcerations, (9) without any significant effect on platelet aggregation or thromboxane B2 levels. (12)

From this information, it is clear that there is a fine balance between COX-1 and COX-2 in gastric injury. Although evidence supports COX-2 inhibitors as being less ulcerogenic than nonselective NSAIDS, (2,6,8,10,11) the COX-2 inhibitors are not completely "ulcer-proof," and in fact can cause delayed ulcer healing and worsening of a preexisting inflammatory state. (12-15)

COX-2 has been found to be an inducible isoform of cyclooxygenase that may be necessary for the production of essential prostaglandins used in the repair process in some inflammatory conditions. It is in these states that inhibition of COX-2 could be troublesome. (12-15) Evidence also supports the notion that good anti-inflammatory efficacy from COX-2 inhibitors is achieved only at doses where the COX-2 specificity is lost. (16)

Our patient was in fact taking supratherapeutic doses to control his symptoms. His dosing may have been so great as to inhibit both isoforms of cyclooxygenase, leading to the GI hemorrhage. A second postulate is that he had an underlying stress gastritis/duodenitis from his illness, in which the COX-2 should have been producing essential repair prostaglandins, and when inhibited resulted in GI hemorrhage.

Our case illustrates the concept that even though the new COX-2 inhibitors may be safer, one should remain cautious when prescribing these drugs, since they too can cause erosions and ulcerations. These lesions, like those due to nonselective NSAIDs, may be asymptomatic until a significant GI hemorrhage occurs, as seen in our patient.

References

(1.) Lanza FL: A guideline for the treatment and prevention of NSAID-induced ulcers. Am J Gastroenterol 1998; 93:2037-2046

(2.) Geis GS: Update on clinical developments with celecoxib, a new specific COX-2 inhibitor: what can we expect? J Rheumatol 1999; 26(suppl 56):31-36

(3.) Singh C: Recent considerations in nonsteroidal anti-inflammatory drug gastropathy. Am J Med 1998; 105:31s-38s

(4.) Smith WL, Dewitt DL: Prostaglandin endoperoxide H synthases-1 and -2. Adv Immunol 1996; 62:167-215

(5.) Vane JR, Botting RM: New insights into the mode of action of anti-inflammatory drugs. Inflamm Res 1995; 44:1-10

(6.) Seibert K, Zhang Y, Leahy K, et al: Pharmacologic and biochemical demonstration of the role of cyclooxygenase 2 in inflammation and pain. Proc Nati Acad Sci USA 1994; 91:12013-120137

(7.) Vane JR, Mitchell JA, Appleton I, et al: Inducible isoforms of cyclooxygenase and nitric-oxide synthase in inflammation. Proc Natl Acad Sci USA 1994; 91:2046-2050

(8.) Simon LS, Lanza FL, Lipsky PE, et al: Preliminary study of the safety and efficacy of SC-58635, a novel cyclooxygenase 2 inhibitor. Arthritis Rheum 1998; 41:1591-1602

(9.) Masferer JL, Zweifel BS, Manning PT, et al: Selective inhibition of inducible cyclooxygenase 2 in vivo is antiinflammatory and nonulcerogenic. Proc Natl Acad Sci USA 1994; 91:3228-3232

(10.) Simon LS, Weaver AL, Graham DY, et al: Anti-inflammatory and upper gastrointestinal effects of celecoxib in rheumatoid arthritis. JAMA 1999; 282:1921-1928

(11.) Langman MJ, Jensen DM, watson DJ, et al: Adverse upper gastrointestinal effects of rofecoxib compared with NSAIDs. JAMA 1999; 282:1929-1933

(12.) Schmaussmann A, Peskar BM, Stettler C, et al: Effects of inhibition of prostaglandin endoperoxide synthase-2 in chronic gastrointestinal ulcer models in rats. Br J Pharmacol 1998; 123:795-804

(13.) Gretzer B, Ehrlich K, Maricic N, et al: Selective cyclo-oxygenase-2 inhibitors and their influence on the protective effect of a mild irritant in the rat stomach. Br J Pharmacol 1998; 123:927-935

(14.) Mizuno H, Sakamoto C, Matsuda K, et al: Induction of cyclooxygenase 2 in gastric mucosal lesions and its inhibition by the specific antagonist delays healing in mice. Gastroenterology 1997; 112:387-397

(15.) wallace JL, Reuter BK, McKnight W, et al: Selective inhibitors of cyclooxygenase-2: are they really effective, selective, and GI-safe? J Clin Gastroenterol 1998; 27(suppl 1):S28-S34

(16.) Wallace JL, Bak A, McKnight W, et al: Cyclooxygenase 1 contributes to inflammatory responses in rats and mice: implications for gastrointestinal toxicity. Gastroenterology 1998; 115:101-109

KEY POINTS

* Cox-2 inhibitors offer effective anti-inflammatory activity for patients with rheumatoid arthritis and osteoarthritis.

* Inhibition of Cox-1 is responsible for the majority of gastric mucosal injury associated with nonselective nonsteroidal anti-inflammatory drugs (NSAIDs).

* Cox-2 inhibitors have been shown to be less ulcerogenic than nonselective NSAIDs.

* Cox-2 inhibitors can delay ulcer healing and worsening of a preexisting inflammatory state.

* Cox-2 inhibitors are not ulcer-free NSAIDs.

From the Department of Gastroenterology, Texas A&M University/Scott & White Memorial Hospital, Temple.

Reprint requests to Andrew S. Crawford, DO, Clinic 3B, 2401 S 31st St, Temple, TX 76508.