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


Oral mesalamine prevented relapse in patients with Crohn disease who were in remission for < 3 months

ACP J Club. 1993 July-Aug;119:11. doi:10.7326/ACPJC-1993-119-1-011

Related Content in the Archives
Cyclosporine Was Effective for Corticosteroid-resistant Ulcerative Colitis
Low-Dose Cyclosporine Did Not Improve Symptoms in Crohn Disease

Source Citation

Gendre JP, Mary JY, Florent C, et al. Oral mesalamine (Pentasa) as maintenance treatment in Crohn's disease: a multicenter placebo-controlled study. Gastroenterology. 1993 Feb;104:435-9.



To study the efficacy and safety of slow-release mesalamine to prevent relapse of Crohn disease in adults.


Randomized, double-blind, placebo-controlled trial with 2-year follow-up.


16 clinical centers in France.


161 patients (mean age 33 y, 85 women) were > 15 years old, had scores < 150 on the Crohn Disease Activity Index (CDAI), had taken no steroids or immunosuppressive agents in the previous month, and had been in clinical remission for < 24 months. 64 patients had been in remission for < 3 months (high-relapse risk stratum [HRRS]), and 97 patients had been in remission for 3 to 24 months (low-relapse risk stratum [LRRS]). Follow-up was 84%.


Randomization was stratified by center and by remission < 3 months. 80 patients (49 in LRRS, and 31 in HRRS) were allocated to receive mesalamine, two 250-mg tablets 4 times per day. 81 patients (48 in LRRS and 33 in HRRS) were allocated to receive placebo. Patients in both groups could take only antispasmodics, antidiarrheal agents, cholestyramine, and sedatives as additional medications.

Main outcome measures

Surgery for acute complications of Crohn disease and clinical relapse (CDAI of > 250 or CDAI > 150 that exceeded baseline by > 50 points). Patients kept diary cards of symptoms and were examined every 3 months for 2 years.

Main results

Analysis was by intention to treat. Patients who were in remission for < 3 months (HRRS) taking placebo compared with those taking mesalamine had a higher relapse rate at 1 year (68% vs 28%, {95% CI for the 40% difference 18% to 63%, P < 0.01}*). At 2 years the trend for lower relapse continued but was not statistically significant (71% vs 55%). For patients in remission for ≥ 3 months (LRRS), the groups did not differ for relapse. Patients in remission for < 3 months who took mesalamine had a 2.4-fold decrease (CI 1.4 to 4.0) in risk for relapse or acute surgical complications compared with the other 3 groups (P < 0.003). The groups did not differ for surgery, side effects, or change in mean serum creatinine levels.


Patients with Crohn disease in remission for < 3 months who were given oral mesalamine had fewer relapses at 1 year than did patients given placebo. For patients in remission for ≥ 3 months, the mesalamine and placebo groups did not differ.

Sources of funding: Institut National de la Santé et de la Recherche Médicale and Ferring SA (mesalamine).

For article reprint: Dr. J.P. Gendre, Hôpital Rothschild, Service d'Hépato-Gastroentérologie, 33, Boulevard de Picpus, 75571 Paris Cedex 12, France. FAX 33-1-40-193436.

*95% CI calculated from data in article.


Patients with Crohn disease and their physicians have been frustrated by the inability to prevent the predictable annual 10% relapse incidence (1). Relapse compromises quality of life, virtually ensuring dependency on medications that have major side effects. The studies by Gendre and colleagues and several others (2, 3) show that mesalamine reduces the risk for relapse.

Many variables influence relapse: disease location, type, previous surgeries for Crohn disease, evidence of aggressive "perforating" disease, nutritional status, laboratory evidence of active inflammation, and duration of remission. The last of these may be the most important in predicting relapse rates as well as a favorable response to maintenance therapy. The longer the clinical remission, the greater the likelihood of remaining in remission. Gendre and colleagues, proceeding on the assumption that a documented remission of < 3 months increases the likelihood of disease relapse, have shown the superiority of mesalamine (Pentasa) over placebo in reducing relapse frequency at 1 year for this group. The patients in the placebo and mesalamine groups were evenly matched, and the analysis of 19 risk factors for relapse did not alter the findings. The authors suggest that a higher dose may further improve outcome for both high- and low-risk recurrence subgroups.

Physicians will have difficulty resisting informed patients who wish to take mesalamine to "cure" their Crohn disease. We have the obligation to inform them that remission cannot be equated with cure and that the likelihood of recurrence will still increase with time. Because much is still to be learned about the variables that affect recurrence, we should encourage patients, when possible, to participate in well-designed institutional and community-based studies.

Arvey I. Rogers, MD
University of Miami School of MedicineMiami, Florida, USA


1. Sachar DB. The problem of postoperative recurrence of Crohn's disease. Med Clin North Am. 1990;74:183-8.

2. International Mesalamine Study Group. Coated oral 5-aminosalicylic acid versus placebo in maintaining remission of inactive Crohn's disease.Ailment Pharmacol Ther. 1990;4:55-64.

3. Prantera C, Pallone F, Brunetti G, et al. Oral 5-aminosalicylic acid (Asacol) in the maintenance treatment of Crohn's disease. The Italian IBD Study Group. Gastroenterology. 1992;103:363-8.

Updated Commentary

Recently, Camma and colleagues (1) systematically reviewed randomized controlled trials (RCTs) of mesalamine for maintenance treatment of Crohn disease, including the study by Gendre and colleagues. This review is commendable for including 15 RCTs (13 full-length reports and 2 abstracts) published from 1988 to 1997 involving 2097 patients (1048 receiving nonactive treatment) and for undertaking the complex statistical analyses necessary to adjust for confounding variables.

The major limitation of the Camma meta-analysis is that it is based on summarized data obtained retrospectively from trials which differed in study design, duration of remission before study entry, drug dosing and formulation, and length of follow-up; involved patients in whom a disease-free remission had been obtained surgically as well as induced medically; and did not control for such variables as type of Crohn disease (inflammatory, fistulizing, or stricturing), nutritional state, smoking history, or use of nonsteroidal anti-inflammatory drugs. Only when such variables are controlled for and individual patient demographics are applied to subgroup patients will a meta-analysis be flawless.

Despite the shortcomings of the meta-analysis which, to the credit of the authors and the journal's editorial board, are detailed in the discussion portion of the report, Camma and colleagues were able to conclude that mesalamine is most effective in preventing symptomatic relapses in subsets of patients with Crohn disease with 1) ileitis, 2) longer disease duration, and 3) surgically-induced remissions. A similar, though less statistically significant benefit, was achieved in other categories of patients who had not undergone surgery and were in a quiescent stage (medical remission).

A perfect meta-analysis is an unlikely scenario. Clinicians and their patients will continue to make clinical decisions on the basis of reasonable data published in reputable journals and discussed by reputable and reasonable clinicians, and they will likely select a maintenance medication with a relatively low side-effect and long-term safety profile. For the moment, mesalamine appears to fit this profile. We agree with the cautious conclusion of Camma and colleagues: “Mesalamine may be recommended for maintaining remission of quiescent Crohn's disease.”

Arvey I. Rogers, MD
Jeffrey Goldstein, MDUniversity of Miami School of MedicineMiami, Florida, USA

1. Camma C, Giunta M, Rosselli M, Cottone M. Mesalamine in the maintenance treatment of Crohn's disease: a meta-analysis adjusted for confounding variables. Gastroenterology. 1997;113:1465-73.