Twelve RCTs were included for review (3,644 patients analysed; sample sizes ranged from 40 to 1,161). No studies scored 5 on the Jadad; 11 scored 3 or 4 and one scored 2. Drop-outs ranged from three to 76 patients (total was 294). Follow-up ranged from six to 12 weeks.
Systolic blood pressure (WMD -2.97 95% CI -4.18 to -1.77; 12 studies, 3,644 patients) and diastolic blood pressure (WMD -1.76 95% CI -2.57 to -0.96; 12 studies 3,644 patients) significantly reduced with candesartan compared to losartan at trough after 24 hour follow up. There was evidence of moderate statistical heterogeneity for both of these outcomes (Ι²= 40.4% for systolic and Ι²= 55.6% for diastolic blood pressure). Subgroup analyses demonstrated that candesartan was superior to losartan in reducing blood pressure at all dosages.
Candesartan demonstrated better control (RR 1.26 95% CI 1.06 to 1.50) and response rates (RR 1.12 95% CI 1.06 to 1.18) than losartan.
Candesartan was associated with significantly lower incidence of serious adverse events compared to losartan (RR 0.48 95% CI 0.25 to 0.92; six studies). There was no significant difference between candesartan and losartan in the occurrence of common adverse events (RR 0.98 95% CI 0.86 to 1.12; eight studies). There was no statistical heterogeneity for the outcomes of adverse events (common adverse events Ι²=0% and serious adverse events Ι²=1%). There were no significant differences between candesartan and losartan in the occurrence of adverse events were headache, dizziness, respiratory infections, fatigue and gastroenteritis.
There was no evidence of publication bias. The results of the meta-regression were reported in the review.