Therapeutic drug monitoring of the anticonvulsant levetiracetam may be indicated in patients with conditions that may alter pharmacokinetic characteristics, for tailoring individual dosage regimens or to investigate patient compliance. In this study, the Bio-Rad high-performance liquid chromatography (HPLC) method (in-use method) and the ARK immunoassay method (new method) for levetiracetam monitoring in serum were compared.
Levetiracetam concentrations were determined in 63 samples using: (1) “Levetiracetam by HPLC” kit by Bio-Rad (Hercules, CA) on the Agilent 1100 HPLC system, and (2) “ARK Levetiracetam” immunoassay by ARK Diagnostics Inc (Fremont, CA) on the CDx90 platform by ThermoFisher Scientific Inc.
Within-laboratory imprecision and bias of the new method evaluated over a 20-day period were 7.4% and 0.5% at 7.5 mcg/mL, 4.5% and 1.9% at 30 mcg/mL, and 3.1% and 2.0% at 75 mcg/mL. Passing–Bablok regression analysis (X:Bio-Rad; Y:Ark) showed a nonsignificant intercept of 0.16 [95% confidence interval (CI), −0.55 to 0.72] and a slope marginally significantly different from unity of 0.95 (95% CI, 0.90–0.99), which suggested minimum proportional systematic error. In agreement, Bland–Altman analysis showed minimum systematic bias of 1.0 mcg/mL (95% CI, 0.32–1.69) with 95% of the HPLC–Ark differences ranging from −4.3 (95% CI, −5.52 to −3.16) to 6.3 (95% CI, 5.16–7.52). Our data showed that the 2 methods were identical both within inherent imprecision and analytical quality specifications (maximum allowable error 15%).
The new Ark method on the CDx platform is acceptable and may be used to measure serum levetiracetam concentrations routinely.