|
Meta-analyses of cisapride, omeprazole and ranitidine in the treatment of GORD: implications for treating patient subgroups |
Iskedjian M, Einarson T R |
|
|
Authors' objectives To examine the efficacy of the pharmocological treatment of gastro-oesophageal reflux disease (GORD) with cisapride, omeprazole or ranitidine.
Searching MEDLINE was searched from 1984 to 1995 using the following keywords: 'GERD' (gastroesophageal reflux disease), 'GORD', 'gastroesophageal', 'reflux', 'cisapride', 'ranitidine' and 'omeprazole'. The search was limited to studies in humans, but was not limited to articles published in the English language; articles in French, German and Italian were also included. The bibliographies of the retrieved articles and reviews located in the search were examined for additional studies.
Study selection Study designs of evaluations included in the reviewDouble-blind randomised controlled trials comparing at least two regimens of the same drug, or comparing one drug with other drugs, regimens or placebo, were included in the analyses. However, the authors then stated that in order to broaden the selection of studies, all placebo- or active-controlled trials of each comparator drug regimen were retrieved. Thus, it was unclear what the exact study design inclusion criteria were.
Specific interventions included in the reviewThe classes and dosages of drugs included in this study were the histamine H2 receptor antagonists (ranitidine: 150 mg twice daily, 300 mg daily, or 300 mg twice daily), proton-pump inhibitors (omeprazole: 20 or 40 mg daily), and prokinetic agents (cisapride: 10 or 20 mg four times daily, or 20 mg twice daily.
Maintenance therapy included cisapride (20 or 40 mg/day; most frequently 20 mg/day), omeprazole (20 or 40 mg/day) and ranitidine (300 mg/day).
Participants included in the reviewThe patients were otherwise healthy adults with a diagnosis of GORD confirmed (preferably by endoscopy) by the physician researchers, and who were not receiving concomitant drug therapy. Studies that only included geriatric patients, i.e. those aged above 65 years, were excluded.
Outcomes assessed in the reviewEfficacy and relapse rates were assessed.
The efficacy rates (cured, failed, improved) were derived from the results reported by the investigators in the original research studies. Cure was defined as the proportion of patients who were evaluated as 'cured' or rated 'excellent'. 'Failure' was defined as the proportion of patients who either did not respond to therapy, responded poorly (e.g. minimal improvement), or whose condition deteriorated. All other patients were defined as 'improved'.
Among patients experiencing a cure during initial therapy, a 'relapse' was defined as the recurrence, verified clinically or by endoscopy, of GORD after discontinuation of therapy, substitution of placebo, or during maintenance therapy.
How were decisions on the relevance of primary studies made?The authors do not state how the papers were selected for the review, or how many of the authors performed the selection.
Assessment of study quality The authors do not state that they assessed validity.
Data extraction The efficacy rates were determined by two independent reviewers, their figures compared, and any discrepancies resolved through consensus.
The authors did not state whether the data were extracted in an intention to treat format, or how those who dropped out of therapy were counted in the data.
Methods of synthesis How were the studies combined?The random-effects meta-analytical method of DerSimonian and Laird (see Other Publications of Related Interest no.1) was used to summarise rates across all studies for each comparator or therapy duration. However, this method was not used for the data regarding relapse rates, because of the small number of data points and disparity among the reported research methods. Thus, these data were calculated as simple weighted averages.
How were differences between studies investigated?The authors did not present a test for heterogeneity.
The studies were stratified into three groups based on treatment duration (4, 8 and 12 weeks). Studies of less than 4 weeks' duration were not considered, while 6-week regimens were pooled with the 8-week regimens, and 10-week regimens were pooled with the 12-week regimens.
A subset analysis was performed for severity of disease, which was confirmed by endoscopy examination rates. Mild GORD was considered to include grades 0 to 1 and severe GORD included grades 2 to 4.
Results of the review Forty-eight papers reporting on 69 clinical trials were included. There were 56 trials of acute treatment: 30 with ranitidine, 13 with omeprazole, 11 with cisapride, and 2 with a combination of cisapride plus cimetidine. There were 13 reports of maintenance therapy: 5 with cisapride, 5 with omeprazole, and 3 with ranitidine. There were 5,060 participants involved in studies examining efficacy rates, and 1,525 participants in studies examining relapse rates.
The 4-week cure rates were higher with omeprazole than with ranitidine. The cure rates were 0.626 (95% confidence interval, CI: 0.477, 0.775) and 0.655 (95% CI: 0.469, 0.842) for doses of 20 and 40 mg omeprazole, respectively, compared with 0.329 (95% CI: 0.257, 0.400) and 0.452 (95% CI 0.385, 0.520) for doses of 300 and 600 mg rantidine, respectively. The lack of 4-week cure raw data for cisapride prevented a comparison.
After 12 weeks of therapy, the cure rates were greater than 50% for all GORD regimens tested, with omeprazole again associated with the highest cure rate (95% CI not reported for omeprazole 40 mg/day). The cure rate was 0.811 (95% CI: 0.757, 0.866) for ranitidine 600 mg/day. and 0.578 (95% CI: 0.442, 0.714) for cisapride 40 mg/day; these were similar to ranitidine 300 mg/day (0.619, 95% CI: 0.502, 0.735).
Eight studies reported data separated according to the severity of the disease, and were included in the subset analysis. One study assessed 12 weeks of therapy and 7 assessed 8 weeks of therapy. For mild GORD, the cure rates were 38% (n=30) with ranitidine 300 mg/day, 56% (n=18) with cisapride 40 mg/day, 75% (n=16) with omeprazole 20 mg/day, and 82% (n=11) with cisapride 80 mg/day. For severe GORD, the cure rates were 33% (n=33) with cisapride 40 mg/day, 43% (n=40), with cisapride 80 mg/day, 87% (n=289) with omeprazole 20 mg/day and 50% (n=194) with ranitidine 300 mg/day.
Seven papers included 13 reports of relapse rates for post-treatment or during maintenance treatment for GORD. The relapse rate after 6 months was lowest with omeprazole maintenance therapy (17%). This was approximately half of that associated with cisapride maintenance therapy (33%) and one third of that associated with ranitidine maintenance therapy (49%). Among patients receiving no maintenance therapy, the relapse rates at 6 months' post-treatment were lowest among those treated with cisapride 40 mg/day (40%). This rate was approximately half that of the relapse rate observed for patients treated acutely with omeprazole 20 mg/day (80%) or omeprazole 40 mg/day (76%).
Authors' conclusions The present meta-analysis indicated that omeprazole is the most effective agent overall of the three acute GORD treatments analysed, and that cisapride 40 mg/day is similar in efficacy to ranitidine 300 mg/day. However, differences emerged regarding the comparative efficacy of cisapride, omeprazole and ranitidine in mild versus severe GORD, and in the frequency of relapse. These differences indicated that omeprazole is clearly the treatment of choice for severe GORD, and suggested that cisapride may be the treatment of choice for mild GORD. These differences also indicated that either of these two treatments is superior to ranitidine for the prevention of relapse.
CRD commentary The review focused on a well-defined question.
The search was fairly narrow and could have been extended to include other databases such as EMBASE. No attempt was made to identify unpublished literature, so a publication bias cannot be ruled out. The inclusion and exclusion criteria were reported, although it was unclear what the inclusion criteria were regarding study design. Some details of the primary studies were included, but it would have been useful to have included participant characteristics such as age and gender. No validity assessment was undertaken. Tests for heterogeneity were not performed before the studies were combined.
The authors' conclusions follow logically from the results.
Perras and Otten have performed a meta-analysis of randomised controlled trials in adults with severe oesophagitis (see Other Publications of Related Interest no.2).
Implications of the review for practice and research The authors suggest that further comparative studies, designed specifically to delineate the most appropriate drug therapy for various subgroups of GORD patients, are needed.
They also point out that there is a paucity of solid and reliable reports of symptomatic results, indicating the need for additional well-designed studies using standardised symptomatic outcome measures.
Bibliographic details Iskedjian M, Einarson T R. Meta-analyses of cisapride, omeprazole and ranitidine in the treatment of GORD: implications for treating patient subgroups. Clinical Drug Investigation 1998; 16(1): 9-18 Other publications of related interest 1. DerSimonian R, Laird N. Metal-analysis in clinical trials. Control Clin Trials 1986;7:177-88. 2. Perras C, Otten N. Pharmacological management of gastroesophageal reflux disease. Ottawa: Canadian Coordinating Office for Health Technology Assessment (CCOHTA); 1996.
Indexing Status Subject indexing assigned by NLM AccessionNumber 11998001284 Date bibliographic record published 31/12/1999 Date abstract record published 31/12/1999 Record Status This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn. |
|
|
|