Elevated Free Phenytoin and Free Valproic Acid Concentrations in Sera of Patients Infected With Human Immunodeficiency Virus

Abstract
Seizures are common in patients infected with human immunodeficiency virus(HIV). Phenytoin and valproic acid are common anticonvulsants, and both drugs are strongly bound to serum albumin. Because patients infected with HIV are often on polytherapy, using homeopathic medicines, and may also have hypoalbuminemia, elevated free drug concentrations may occur in these patients. The authors prepared one serum pool from patients infected with HIV but receiving no bactrim and the other pool from HIV patients receiving bactrim. They supplemented both HIV pools and normal pool (diluted with 0.9% saline to mimic albumin concentration of HIV pools) with a known concentration of phenytoin or valproic acid. After incubation at 37°C for 3 hours, they measured free phenytoin and free valproic acid concentrations in the protein free ultrafiltrates using fluorescence polarization immunoassays. The total drug concentrations in original sera were measured by microparticle enzyme immunoassays. None of the patients had any significant liver or renal disease. The aliquots of HIV pools and normal pool were supplemented with the same concentration of phenytoin or valproic acid. The concentration of free phenytoin and free valproic acid were significantly elevated in patients with HIV (mean = 2.52, SD = 0.11 µg/ml for phenytoin; mean = 41.5, SD = 1.5µg/ml for valproate) compared to controls (mean = 1.50, SD = 0.0 7µg/ml for phenytoin; mean = 19.9, SD = 0.5 µg/ml for valproate). The concentrations of both free phenytoin and valproic acid were further elevated in patients prepared in the HIV pool who were receiving bactrim (mean= 2.81, SD = 0.09 µg/ml for phenytoin; mean = 44.0, SD = 1.1 µg/ml for valproate), but when normal serum pool was supplemented with 4.4 mg/dl of bactrim (concentration of bactrim in HIV pool) and supplemented with the same concentration of phenytoin or valproic acid, the observed free concentrations were much lower (mean = 1.65, SD = 0.05 µg/ml for phenytoin; mean = 26.1, SD = 1.4 µg/ml for valproate). This indicates that hypoalbuminemia and bactrim concentrations do not account for the observed free drug concentrations in patients with HIV. The authors also observed elevated free phenytoin and valproic acid in sera from three individual patients with AIDS compared to normals (normal serum diluted with 0.9% saline to mimic the albumin concentration of serum collected from a patient with HIV and then both specimens supplemented with the same concentration of phenytoin or valproic acid).