Comparing the effect of intermittent diazepam and continuous phenobarbital in preventing recurrent febrile seizures among children under 6 years old: A systematic review and meta-analysis

Background: Febrile convulsion (FC) is the most common and preventable seizure in children. This study aimed to assess the effectiveness of the diazepam and phenobarbital for preventing recurrent FC. Materials and Methods: In this systematic review study, literature published in English language were carefully searched in biological databases (Cochrane Library, Medline, Scopus, CINHAL, Psycoinfo, and Proquest) by February 2020.Randomized clinical trials (RCTs) and Quasi randomized trial were included in the review. Two researchers checked the literature independently. The quality of studies was assessed using the JADAD score. The potential risk for publication bias was assessed by Funnel plot and Egger's test. Meta regression test and sensitivity analysis were used to identify the reasons for heterogeneity. Given the results of assessing heterogeneity, the random effect model in RevMan5.1 software was used for meta analysis. Results: Four out of 17 studies had compared the effect of diazepam and phenobarbital in preventing recurrent FC. The result of the meta analysis showed that the use of diazepam in comparison with phenobarbital reduces the risk of recurrence FC by 34% (risk ratio = 0.66, 95% confidence interval [CI] = [0.36-1.21]), but the relationship was not statistically significant. In assessing the effect of diazepam or phenobarbital versus placebo, the results showed that the use of diazepam and phenobarbital has reduced the risk of recurrent FC by 49% (risk ratio = 0.51, 95% CI = [0.32-0.79]) and 37% (risk ratio = 0.63, 95% CI = [0.42-0.96)]), respectively, and these relationships were statistically significant (P < 0.05). Results of the meta regression test showed that the follow up time can be a reason for the heterogeneity between trials with the comparison of diazepam versus phenobarbital (r = 0.047, P = 0.049) and Phenobarbital versus placebo (r = 0.022, P = 0.016). According to the results of Funnel plot and Egger's test, there was evidence of publication bias (P = 0.0584 for comparison of diazepam vs. phenobarbital; P = 0.0421 for comparison of diazepam vs. placebo; P = 0.0402 for comparison of phenobarbital vs. placebo). Conclusion: The results of this meta analysis indicated that preventive anticonvulsants can be useful in preventing recurrent convulsions in cases of febrile seizures.