Twelve RCTs (7,277 patients, range 55 to 2,027) were included in the review. Trials were considered good quality according to the Jadad scale (average score 3.58) and Cochrane allocation concealment item (six trials scored A and six scored B).
Pain was statistically significantly less with valaciclovir compared to acyclovir at time points between one and 112 days: one to 10 days (RR 0.92, 95% CI 0.88 to 0.97; three RCTs) and 61 to 112 days (RR 0.79, 95% CI 0.63 to 0.98; two RCTs). There was no evidence of statistical heterogeneity. NNT ranged from three to 17. The difference was not statistically significant between 113 and 170 days.
Pain was statistically significantly less in patients who received famciclovir compared to acyclovir at 28 to 30 days (RR 0.54, 95% CI 0.43 to 0.68; two RCTs; NNT=three). There was no evidence of statistical heterogeneity. There were no statistically significant differences between famciclovir and acyclovir in adverse events or lesion healing (three RCTs).
There were no statistically significant differences between valaciclovir or famciclovir versus acyclovir at any doses or time points in loss of vesicles, cessation of new lesions, full crusting or loss of crusts and adverse effects.
Results from dose response or single trials were reported in the review.