Latent renal tubular dysfunction in patients with epilepsy treated with valproic acid

Background: Valproic acid (VPA)-induced Fanconi's syndrome has been reported, especially in patients with severe motor and intellectual disabilities, with carnitine deficiency identified as a possible factor. Latent renal tubular dysfunction (RTD) has also been noted in patients with epilepsy treated with VPA. We aimed to evaluate latent RTD in patients treated with and without VPA by examining urine RTD markers and their relationships with clinical variables. Methods: Urine N-acetyl-β-D-glucosaminidase/creatinine (NAG/Cr) and β2-microglobulin/creatinine (BMG/Cr) were evaluated in 147 patients with childhood-onset epilepsy. NAG/Cr and BMG/Cr were compared between 90 patients treated with VPA (VPA group) and 57 patients treated with other anti-seizure medications (non-VPA group). Single correlations and multiple regression analyses were conducted between RTD markers and clinical variables in the VPA group. Results: NAG/Cr was significantly higher in the VPA group than in the non-VPA group, whereas BMG/Cr was not significantly different. High or borderline NAG/Cr (>4.0 IU/g*Cr) was seen in 85 (94 %) patients on VPA, and high NAG/Cr (≥10.0 IU/g*Cr) was seen in 21 (23 %) on VPA. In the VPA group, NAG/Cr was significantly correlated with free carnitine (FC), VPA dose adjusted for body weight, and duration of treatment with VPA. BMG/Cr was significantly correlated with the number of antiseizure medications. Multiple regression analysis showed that NAG/Cr was significantly correlated with FC. Conclusions: Latent RTD was found in many patients on VPA, and elevated RTD markers were correlated with carnitine deficiency. Physicians should be aware of latent RTD in patients with childhood-onset epilepsy treated with VPA.