Anti‐seizure efficacy and retention rate of carbamazepine is highly variable in randomized controlled trials: A meta‐analysis

We sought to assess the anti‐seizure efficacy of carbamazepine (CBZ) and retention rate (RR) in randomized, controlled trials (RCTs) in epilepsy. Our analysis was carried out according to the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) statement. Inclusion criteria were monotherapy of CBZ in adequate dosage for epilepsy treatment and RCT duration of ≥3 months. Outcome measures were seizure freedom rate (SFR) and RR. Random‐effects meta‐analyses were performed to allow for comparison with other anti‐seizure medications (ASMs). Thirty RCTs of 734 were included. SFR at last follow‐up ranged from 11% at 36 months to 85% at 3 months. The aggregated SFR at 6 months was 58% (CI 49–66%) and 48% (CI 40–57%) at 12 months. The 6‐month SFR among blinded studies was 55% (CI 43–66%), compared with 61% (CI 50–71%) in unblinded studies. The 12‐month SFR was not significantly linked to the age of study participants. RR varied from 36% at 24 months to 81% at 6 months. When adjusting for blinding, the aggregated 6‐month RR in blinded studies was 59% (CI 52–66%) vs 76% (CI 71–81%) in unblinded studies. The point estimates of SFR of all RCTs showed an upward time trend, with an increase of approximately 15% between the years 1981 and 2018. In conclusion, the SFR and RR of CBZ were highly variable in RCTs and especially affected by study duration and blinding. These results underscore the impact of the design of RCTs investigating ASM and may challenge the wide use of CBZ as a comparator.