Eight RCTs (n=743) were included in the review. Sample sizes ranged from 51 to 184 participants. The quality of the included studies was reported to be high. All trials were double-blinded and all had adequate allocation concealment and baseline comparability.
Corticosteroids were significantly more effective than placebo in resolving pain completely at 24 hours (relative risk 3.16, 95% CI 1.97 to 5.08, p<0.001; four RCTs) and 48 hours (relative risk 1.65, 95% CI 1.32 to 2.06, p<0.001; three RCTs). There was no evidence of statistical heterogeneity. The number needed to treat was 3.7 at 24 hours and 3.3 at 48 hours. Subgroup analyses indicated significant effects in adult patients only at 24 and 48 hours (both p<0.001) and significant effects for oral administration of corticosteroids at 24 hours (p<0.001) and 48 hours (p=0.004).
The mean time to onset of pain relief occurred on average 6.3 hours earlier in patients treated with corticosteroids compared to controls (95% CI 9.29 to 3.35, p<0.001; six RCTs). There was evidence of significant statistical heterogeneity (I2=73%). Sensitivity analyses did not significantly alter the findings. Subgroup analyses indicated a significant reduction in mean time to onset of pain relief in patients with exudative sore throat (weighted mean difference 5.5 hours, 95% CI 8.0 to 3.0; two RCTs), sore throat that was bacterial pathogen positive (weighted mean difference 5.3, 95% CI 8.0 to 2.6, p<0.001; four RCTs) and trials of patients with severe sore throat (weighted mean difference 7.2 hours, 95% CI 10.0 to 4.3, p<0.001; two RCTs). There were some discrepancies in the results reported in the text and tables for mean time to onset of pain relief; the data reported here were presented in the tables. Subgroup analyses in trials with children only showed no significant differences between intervention and controls in mean time to onset of pain relief.
The findings for trials that measured time to complete resolution of symptoms were inconsistent and could not be pooled due to heterogeneity. Secondary outcomes were reported in the review.