To determine the following from a comprehensive review of published comparative clinical trials of roxithromycin and erythromycin in patients with lower respiratory tract infections (LRTIs).

1. The proportions of patients who experienced adverse events during the course of antibacterial therapy.

2. The proportion of patients who withdrew from the trials as a result of adverse events associated with antibacterial drug therapy.

MEDLINE was searched from 1987 to 1996, and AdisBase from 1989 to 1996. The review was limited to comparative clinical studies published in the English language (or with an English language summary) involving the use of roxithromycin and/or erythromycin in the treatment of LRTI. Details of the search strategy were not presented.

Further comparative studies of roxithromycin, published over this period, were identified by examining the reference lists of all retrieved articles and that of a major review of roxithromycin.

Study designs of evaluations included in the review

Clinical trials involving the use of roxithromycin and/or erythromycin in the treatment of LRTI were included. the length of treatment and the minimum follow-up period were not stated.

Specific interventions included in the review

Clinical studies that compared roxithromycin directly with erythromycin, or studies that compared either roxithromycin or erythromycin with any other comparator antibacterial drug, in the treatment of LRTI.

Roxithromycin was administered at dosages of 150 mg twice daily or 300 mg once daily. Various oral formulations of erythromycin were used; the dosages ranged from 1 to 2 g/day, either in 4, 3 or 2 divided doses.

Participants included in the review

Adult patients with LRTI. The majority of the included patients were diagnosed as having community-acquired pneumonia. Other patients were diagnosed with acute infective exacerbation of chronic bronchitis, isolated acute bronchitis, or even bronchiectasis. Roughly equivalent proportions of patients receiving roxithromycin were diagnosed with pneumonia, acute infective exacerbation of chronic bronchitis, or acute bronchitis.

Outcomes assessed in the review

The outcomes assessed were the incidence of adverse events possibly or probably associated with drug therapy, and the number of patients who withdrew from therapy because of adverse events.

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.

The authors do not report the criteria used to assess validity, or how the validity assessment was performed.

The authors do not state how the data were extracted for the review, or how many of the authors performed the data extraction.

How were the studies combined?

The studies were pooled using a non-parametric (Wilcoxon test) approach for continuous covariates and a parametric (chi-squared test) for categorical covariates. The main effects for potential confounders were adjusted by categorical modelling techniques.

How were differences between studies investigated?

One of the included studies had a greater number of withdrawals and adverse events than the remaining trials. The analyses were therefore performed with and without this study. However, heterogeneity of the studies within each of the drug groups was formally tested, and this suggested they were homogeneous and could be treated as a whole (chi-squared p>0.05).

There were 942 patients (13 studies) who received roxithromycin and 1,253 patients (15 studies) who received erythromycin in various formulations and dosages. This included 3 direct comparative studies of roxithromycin versus erythromycin, and one study that compared two different formulations of erythromycin. The remainder compared either roxithromycin (10 trials) or erythromycin (11 trials) with another macrolide or another agent.

The proportion of patients reporting adverse events possibly or probably related to therapy was 10.1% (95% confidence interval, CI: 8, 12) for roxithromycin and 24.8% (95% CI: 22, 27) for erythromycin; the difference was 14% (standard error 1.5). The proportion of patient withdrawal due to adverse events was 2.0% (95% CI: 1, 3) for roxithromycin and 7.1% (95% CI: 6, 9) for erythromycin; the difference was 5% (standard error 0.045).

This pooled analysis of clinical trials showed that roxithromycin was considerably better tolerated than erythromycin, and that fewer patients withdrew from therapy because of adverse events related to drug therapy. Most of these events were gastrointestinal in nature. Roxithromycin is, therefore, likely to be more effective than erythromycin in community practice because of higher compliance rates, in addition to causing fewer intolerable adverse effects for patients with an LRTI. Consequently, it is also likely to be associated with lower reconsultation rates for treatment failure and for management of troublesome gastrointestinal events.

The literature search was rather limited and, therefore, relevant information may have been missed. The authors only searched MEDLINE and AdisBase, covering the period 1989 to 1996, and only checked the references of retrieved articles looking at roxithromycin. The search was also limited to English language articles and there was no investigation of unpublished data. The authors failed to state the method used to make decisions about including or excluding studies, and the criteria used to assess the validity of the included studies.

The main findings of this review should be treated with caution, because in all but three studies, indirect comparisons were made. If the results of the three comparative studies are compared with those from the indirect comparisons, a different pattern appears. The outcomes for the three comparative trials show a similar pattern of adverse events and withdrawal for both roxithromycin and erythromycin, with the exception of one study, where adverse events were reported by 51% of the patients receiving erythromycin. However, this particular study evaluated the use of erythromycin stearate which, as the authors state, seems to be associated with a high incidence of adverse events.

The authors stated that further research is needed on the following: the role of antibacterial therapy in patients with acute isolated bronchitis; rates of withdrawal from therapy; reconsultation rates in community practice; and the tolerability of other first-line drug therapies for LRTIs.

Hoechst Marion Roussel.

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.