[Evaluation of bone mineral density using digital image processing in children receiving anticonvulsants].

Bone mineral density (BMD) increases rapidly in a biphasic manner in childhood. During and after adolescence, BMD correlates more closely with bone age than chronological age. Digital image processing (DIP) allows the rapid assessment of BMD and bone age on one X-ray film. Herein, using DIP methods, the effects of various anticonvulsants on chronological and bone age were evaluated in 98 epilepsy patients (age range, 3-15 years) with no intellectual or motor disorders or diseases affecting bone metabolism. All patients were taking one or a combination of the following anticonvulsants: valproate sodium (VPA); carbamazepine (CBZ); and phenobarbital (PB). Bone maturation scores for radius-ulnar-short bones (RUS) were calculated using Tanner-Whitehouse 2 methods. Bone age was determined based on standard Japanese bone-maturation scores. In each patient, Z-scores for chronological and bone ages were calculated by subtracting standard BMD for gender and age from each BMD, then dividing the result by the standard deviation. The Z-score for each drug in relation to the administration period was analyzed using the Mann-Whitney test. For chronological age, significant differences in BMD were observed regarding the administration periods in children taking multiple drugs, but not in children on VPA, CBZ, or PB monotherapy. For bone age, no significant differences in BMD were observed regarding the administration periods for all drugs. Children taking multiple drugs showed a significant negative correlation between administration period and Z-scores for BMD calculated based on chronological age (Spearman rank correlation: - 0.457, p = 0.008), but not bone age. Among children receiving long-term VPA administration, bone age was delayed approximately 1 year, and bone maturation may have been delayed. No delay in bone age was noted among children receiving long-term administration of multiple drugs, suggesting that these anticonvulsants do not influence bone maturity. These findings indicate that bone age should be considered when assessing BMD.