Twenty-one RCTs were included: six RCTs (n=227 patients) were double-blinded and 15 RCTs (n=1,064 patients) were not. Fifteen trials had a quality score of 1 point, two had a quality score of 3 points, three had a score of 4 points and one had a score of 5 points. The mean length of follow-up was 13.8 weeks (SD 19.6).
There was a statistically significant benefit with a clozapine combination compared to clozapine monotherapy on symptom scale score for the open trials (SMD -0.80, 95% CI: -1.14 to -0.46; 14 trials), but not the double-blind trials (SMD -0.12, 95% CI: -0.57 to 0.32; six trials). Statistical heterogeneity was high for both analyses. The findings were not substantially altered when they were further subgrouped by trial duration.
Significantly fewer participants in the clozapine combination group failed to show an improvement compared to monotherapy in the pooling of the open trials (RR 0.64, 95% CI: 0.42 to 0.97; 10 trials), but not the double-blind trials (RR 0.91, 95% CI: 0.75 to 1.11; six trials). Statistical heterogeneity was high for both analyses. The findings for the open trials were not substantially altered when further subgrouped by trial duration, but for the double-blind studies the subgroup of longer duration trials showed a statistically significant benefit with combination therapy.
The funnel plot did not suggest publication bias.