Effect of short- and long-term treatment with valproate on carnitine homeostasis in humans.

Aims: The aim of this study was to identify the mechanisms of hypocarnitinemia in patients treated with valproate.
Methods: Plasma concentrations and urinary excretion of carnitine, acetylcarnitine, propionylcarnitine, valproylcarnitine, and butyrobetaine were determined in a patient starting valproate treatment and in 10 patients on long-term valproate treatment. Transport of carnitine and valproylcarnitine by the proximal tubular carnitine transporter OCTN2 was assessed in vitro.
Results: In the patient starting valproate, the plasma carnitine and acetylcarnitine levels dropped for 1-3 weeks and had recovered after 3-5 weeks, whereas the plasma levels of propionyl and valproylcarnitine increased steadily over 5 weeks. The renal excretion and excretion fractions (EFs) of carnitine, acetylcarnitine, propionylcarnitine, and butyrobetaine decreased substantially after starting valproate. Compared with controls, patients on long-term valproate treatment had similar plasma levels of carnitine, acetylcarnitine, and propionylcarnitine, whereas valproylcarnitine was found only in patients. Urinary excretion and renal clearance of carnitine, acetylcarnitine, propionylcarnitine, and butyrobetaine were decreased in valproate-treated compared with that in control patients, reaching statistical significance for carnitine. The EFs of carnitine, acetylcarnitine, and propionylcarnitine were <5% of the filtered load in controls and were lower in valproate-treated patients. In contrast, the EF for valproylcarnitine approached 100%, resulting from a low affinity of valproylcarnitine for the carnitine transporter OCTN2 and competition with concomitantly filtered carnitine.
Conclusions: The initial drop in plasma carnitine levels of valproate-treated patients is most likely due to impaired carnitine biosynthesis, whereas the recovery of the plasma carnitine levels is explainable by an increased renal expression of OCTN2. Renally excreted valproylcarnitine does not affect renal handling of carnitine in vivo.