Treatment of children with infantile spasms: A network meta-analysis.

Abstract

We performed a systematic review and network meta-analysis (NMA) to obtain comparative effectiveness estimates and rankings of non-surgical interventions used to treat infantile spasms. All randomized controlled trials (RCTs) including children 2 months to 3 years of age with infantile spasms (with hypsarrhythmia or hypsarrhythmia variants on electroencephalography) receiving appropriate first-line medical treatment were included. Electroclinical and clinical remissions within 1 month of starting treatment were analyzed. Twenty-two RCTs comparing first-line treatments for infantile spasms were reviewed; of these, 17 were included in the NMA. Both frequentist and Bayesian network rankings for electroclinical remission showed that high dose adrenocorticotropic hormone (ACTH), methylprednisolone, low dose ACTH and magnesium sulfate (MgSO4 ) combination, lowdose ACTH, and highdose prednisolone were most likely to be the 'best' interventions, although these were not significantly different from each other. For clinical remission, low dose ACTH/MgSO4 combination, highdose ACTH (with/without vitamin B6 ), highdose prednisolone, and lowdose ACTH were 'best'. Treatments including ACTH and highdose prednisolone are more effective in achieving electroclinical and clinical remissions for infantile spasms. Adrenocorticotropic hormone and highdose prednisolone are more effective than other medications for infantile spasms. Symptomatic etiology decreases the likelihood of remission even after adjusting for treatment lag.