Safety evaluation of valproate on bone mineral density and bone metabolism in children with epilepsy

Objective To evaluate the safety of valproate (VPA) on bone mineral density and bone metabolism in children with epilepsy. Methods Retrieve relevant research from online databases (January 1, 1980-March 1, 2018) as PubMed, EBMASE/SCOPUS and Cochrane Library with key words: epilepsy, bone, child. Selection of studies was performed according to pre-designed inclusion and exclusion criteria. Quality of studies was evaluated by using Newcastle-Ottawa Scale (NOS). All data were pooled by RevMan 5.2 software for Meta?analysis. Results The research enrolled 1455 articles, from which 13 studies with NOS score ≥ 6 were chosen after excluding duplicates and those not meeting the inclusion criteria. A total of 683 children with epilepsy using VPA and 436 children in control group were included. Meta-analysis showed that comparing with control group, bone mineral density of lumbar (BMDL) in children treated with VPA has no significant difference (MD = -0.019, 95% CI: -0.044-0.006; P = 0.140), but there was a significant decrease in bone mineral density of femur (BMDF; MD = - 0.037, 95%CI: -0.069- -0.005, P = 0.020). There was no significant change in serum calcium (SMD = -0.284, 95% CI: -0.951-0.382; P = 0.400), phosphorus (SMD = - 0.164, 95%CI: -0.457-0.129; P = 0.270), alkaline phosphatase (ALP; SMD = 0.363, 95%CI: -0.294-1.020, P = 0.280), parathyroid hormone (PTH; SMD = 0.102, 95%CI: -0.291-0.495, P = 0.610) and 25-hydroxy vitamin D [25(OH)D; SMD = -0.104, 95%CI：-0.620-0.413, P = 0.690] in bone metabolic markers. Sensitivity analysis showed Meta-anlaysis of BMDF, serum phosphorus and 25(OH)D was unstable, therefore, it should be cautious when explaining the results. Conclusions VPA has less effect on bone mineral density and bone metabolism, but for children with epilepsy using VPA, their BMDF should be monitored. DOI: 10.3969/j.issn.1672-6731.2018.06.005