Twenty-seven RCTs were included in the review. The total number of participants was not reported, but was more than 3,802.
Clinical response (27 RCTs) was statistically significantly higher for meropenem than imipenem plus cilastatin (RR 1.04, 95% CI: 1.01, 1.06).
Bacteriologic response (22 RCTs) was statistically significantly higher for meropenem than imipenem plus cilastatin (RR 1.05, 95% CI: 1.01, 1.08).
There was no statistically significant difference in mortality over the course of treatment between meropenem and imipenem plus cilastatin (9 RCTs; RR 0.98, 95% CI: 0.71, 1.35).
There were statistically significantly fewer adverse events with meropenem than with imipenem plus cilastatin (18 RCTs; RR 0.87, 95% CI: 0.77, 0.97). The most commonly reported adverse events were those identified by laboratory tests, such as thrombocytosis and increased hepatic enzymes. Other adverse events included diarrhoea, injection site inflammation, nausea or vomiting, rash and seizure.
No evidence of clinical heterogeneity was found. In all subgroup analyses that consisted of more than one trial, the direction of the result was consistent with the primary analyses. There was no evidence of statistically significant heterogeneity in any of the primary analyses.
There was no evidence of significant publication bias when using either funnel plots or an assessment of regression of normalised effect versus precision for any of the primary analyses (P>0.30 for all comparisons).
The effect of changing the inclusion criteria to include all comparative RCTs, regardless of the type of infection or treatment regimen, did not change the direction of the results for any of the primary analyses. The effect of using a random-effects model also did not change the direction of the results for any of the primary analyses, or make a significant difference non-significant.
The primary meta-analysis of clinical response for meropenem compared with imipenem plus cilastatin was reanalysed as a cumulative analysis, based on the year of publication of the clinical trials included, in order to determine at which point a statistically significant result could have been identified if a cumulative meta-analysis had been employed as new research emerged. A meta-analysis of the 13 trials available in 1995 demonstrated a significant benefit in favour of meropenem (RR 1.03, 95% CI: 1.01, 1.05).