Hormonal Therapy for Infantile Spasms: A Systematic Review and Meta-Analysis

OBJECTIVE: The limitations of adrenocorticotrophic hormone (ACTH) treatment for infantile spasms (ISs), such as high costs, limited availability, and adverse effects (AEs), make it necessary to explore whether corticosteroids are optimal alternatives. Many other compelling treatments have gone through trials due to the suboptimal effectiveness of hormonal therapy. A systematic review and meta-analysis were performed to evaluate the effectiveness and safety of hormonal therapy for patients with ISs. METHODS: EMBASE, Ovid MEDLINE, Cochrane Central Register of Controlled Trials (CENTRAL), and online registers were searched through April 2021 for randomized controlled trials (RCTs). RESULTS: A total of 19 RCTs (N = 1,279) were included. There was no significant difference in the effectiveness of oral corticosteroids and ACTH in electro-clinical response (risk ratio [RR] = 0.85, 95% CI 0.41–1.76). Low-dose ACTH had similar effectiveness in electro-clinical response compared to usual-dose group (RR = 0.94, 95% CI 0.60–1.47) but conferred a lower risk of AEs (RR = 1.71, 95% CI 1.08–2.71). ACTH was more beneficial in controlling spasms than vigabatrin (VGB) (RR = 1.31, 95% CI 1.05–1.64) for patients without tuberous sclerosis complex (TSC). All RCTs were connected through network meta-analysis, and we found that ketogenic diet (KD), zonisamide, methylprednisolone, or combined treatment of hormonal therapy with topiramate (TPM) or pyridoxine was not different in electro-clinical response compared to usual-dose ACTH. CONCLUSION: Our analysis showed that oral corticosteroids could be optional alternatives when ACTH is not applicable, and ACTH is more beneficial for patients without TSC. Moreover, low-dose ACTH is recommended due to comparative effectiveness but lower risk of AEs. However, due to the high heterogeneity of included patients and treatment protocols, these results must be interpreted with caution. RCTs with multicentric involvement and larger sample size are needed for solid evaluation of other alternative treatments.