Current issues of ACP Journal Club are published in Annals of Internal Medicine


Etodolac and naproxen had similar efficacy in rheumatic arthritis but etodolac had lower gastric toxicity

ACP J Club. 1991 July-Aug;115:10. doi:10.7326/ACPJC-1991-115-1-010

Source Citation

Bianchi Porro G, Caruso I, Petrillo M, Montrone F, Ardizzone S. A double-blind gastroscopic evaluation of the effects of etodolac and naproxen on the gastrointestinal mucosa of rheumatic patients. J Intern Med. 1991 Jan;229:5-8.



To evaluate the clinical efficacy and gastric toxicity of etodolac and naproxen in patients with rheumatoid arthritis.


Randomized, double-blind, controlled trial.


Gastroenterology and rheumatology units of a hospital in Milan, Italy.


108 patients were identified who met the American Rheumatism Association diagnostic criteria for classic or definite rheumatoid arthritis and who had received long-term therapy with nonsteroidal anti-inflammatory drugs (NSAIDs). Of these 108 patients, 42 women and 6 men showed normal mucosa during upper gastrointestinal endoscopy and were randomly assigned to treatment. 44 patients completed the trial.


After at least 1 week off all anti-inflammatory therapy, patients received either etodolac, 200 mg twice a day (n = 22), or naproxen, 500 mg twice a day, for 4 weeks (n = 26). Patients were reassessed by endoscopy after therapy. Endoscopy could also be done during the study for painful dyspepsia lasting ≥ 48 hours, unrelieved by antacids. Clinical and laboratory evaluations were done at the beginning and end of the study. Compliance was measured by pill count.

Main outcome measures

The number and grade of endoscopic gastric lesions were assessed (grade 0 = normal to grade 4 = active ulcer). Endoscopy was done in all cases by the same 2 investigators. Four clinical parameters were evaluated: pain, morning stiffness, Ritchie Index, and grip strength.

Main results

The 2 study groups were similar for age, severity of disease, incidence of previous indigestion, and history of proven ulcer. 3 of the 20 patients (15%) receiving etodolac developed grade 2 endoscopic lesions, and 11 of the 24 patients (46%) receiving naproxen developed grade 2, 3, or 4 lesions (P < 0.05). {This absolute risk reduction of 31% means that 3 patients would need to be treated with etodolac (rather than naproxen) for 4 weeks to prevent 1 additional endoscopic lesion, 95% CI 2 to 30; the relative risk reduction was 67%, CI 10% to 89%}.* 3 of the latter 11 patients (13%) had active ulcers. There was no correlation between endoscopic lesions and dyspeptic symptoms. 2 patients receiving naproxen required endoscopy during treatment because of painful dyspepsia. Patients in both treatment groups showed improvement for each of the clinical parameters (all P < 0.01), with no difference observed in the effectiveness of the 2 drugs.


Etodolac and naproxen have similar clinical efficacy in patients with rheumatoid arthritis, but etodolac produces substantially less gastric toxicity.

Source of funding: Not stated.

Address for article reprint: Professor G. Bianchi Porro, Gastrointestinal Unit, L. Saccho Hospital, Via G.B. Grassi 74, 20157 Milan, Italy.

*Numbers calculated from data in article.


NSAIDs are widely prescribed, and their most common serious side effects are gastrointestinal. Because there is little difference in efficacy among the NSAIDs, the choice of NSAID in most situations depends on the frequency of side effects. Thus, the study by Porro and colleagues addresses an important question.

The investigators did endoscopy in patients with rheumatoid arthritis who had received 1 month of therapy with either etodolac or naproxen and found significantly more gastric lesions in the naproxen than in the etodolac group. Further, active ulcers were found in only 3 patients, all from the naproxen group. However, the importance of this work is limited. First, the clinical significance of grades 1, 2, and 3 lesions is uncertain because evidence that they lead to ulcer or bleeding is lacking. Second, the study was not large enough to show a difference in the risk for ulcers or in the clinical efficacy of the drugs. Third, the clinical significance of endoscopy studies is uncertain because of the dissociation between endoscopic findings and symptoms. The current epidemiologic evidence does not consistently show that any of the drugs is safer (1, 2).

As suggested by the authors, further work is needed in large samples of patients with emphasis on outcomes of clinical importance. In the meantime, current data suggest that NSAIDs should be prescribed with the lowest effective dosage for the shortest possible time and that alternative drugs such as acetaminophen should be used whenever clinically appropriate.

Jeffrey L. Carson, MD
University of Medicine & Dentistry of New JerseyNew Brunswick, New Jersey, USA


1. Kurata JH. An assessment of non-steroidal anti-inflammatory drugs as a risk factor in ulcer disease. In: Soll AH, moderator. Nonsteroidal anti-inflammatory drugs and peptic ulcer disease. Ann Intern Med. 1991;114:307-19.

2. Carson JL, Strom BL. The gastrointestinal side-effects of the non-steroidal anti-inflammatory drugs. J Clin Pharmacol. 1988:28;554-9.