Institutional members access full text with Ovid®

Share this article on:

Liquid Chromatography Tandem Mass Spectrometry Assay for Topiramate Analysis in Plasma and Cerebrospinal Fluid: Validation and Comparison With Fluorescence-Polarization Immunoassay

Christensen, Jakob*† ; Højskov, Carsten S.; Poulsen, Jørgen H.


Summary The authors report the development and validation of a liquid chromatography tandem mass spectrometry assay (LC/MS/MS assay) for the analysis of topiramate (2,3:4,5-bis-o-(-1-methyl)-β-D-fructopyranose sulfamate) in plasma and cerebrospinal fluid (CSF). Comparison is made with the commercially available fluorescence-polarization immunoassay (FPIA).

LC/MS/MS Assay Using the internal standard, 1,2:3,4-bis-o-(1-methylethylidene-α-D-galactopyranose sulfamate), a structural isomer, the calibration curve in plasma was linear in the concentration range of 0.02–20.0 mg/L (r2 = 0.9998). The coefficients of variation in plasma were ≤ 3%, and the accuracy ranged from 100% to 101% in the therapeutically relevant concentration range of 0.4–16.0 mg/L. In CSF, the mean recovery was 98%, and there was linearity between the nominal and the estimated concentration in the range of 1.5–20.0 mg/L (r2 = 0.9996).

FPIA The calibration curve was linear in the concentration interval of 1.6–24.3 mg/L (r2 = 0.9994), and the mean recovery was 96%. Accuracy in plasma was 99– 104%, and precision was 3.2–6.0%. In CSF, there was linearity between the nominal concentration and the estimated concentration in the range of 1.5–20.0 mg/L (r2 = 0.9995), and the mean recovery was 100%.

Comparison Between FPIA and LC/MS/MS There was a high correlation between the FPIA and the LC/MS/MS assay (r2 = 0.9965 in plasma and r2 = 0.9996 in CSF, P < 0.001 for both). In plasma and CSF, the two methods showed equal results, evaluated as the ratio between the two methods (plasma: median ratio = 1.00; 95% confidence interval [CI], 0.98–1.02, paired-sample t test, P = 0.79; and CSF: median ratio = 1.00, 95% CI, 0.99–1.02, paired-sample t test, P = 0.75). The coefficient of variation on the ratios between the two methods had similar levels: 5% in plasma and 3% in CSF.

Conclusion The new LC/MS/MS assay has favorable characteristics, being highly precise and accurate. FPIA also proved precise and accurate, and there was a high agreement with the LC/MS/MS assay in plasma and CSF. Either method displayed sufficient precision and accuracy and may thus be implemented in daily routine.

Departments of *Clinical Pharmacology, †Neurology, and ‡Clinical Biochemistry, Aarhus University Hospital, Aarhus, Denmark

Received November 14, 2001; accepted March 27, 2002.

Address correspondence and reprint requests to Jakob Christensen, Center for Clinical Pharmacology, Department of Pharmacology, The Bartholin Building, Aarhus University, 8000 Aarhus C, Denmark; E-mail:

© 2002 Lippincott Williams & Wilkins, Inc.