Determination of serum carbamazepine concentration using dried blood spot specimens for resource-limited settings. Issue 2 (15th March 2017)

• Record Type: Journal Article
• Authors: Das, Saibal; Fleming, Denise H; Mathew, Binu S; Winston A, Blessed; Prabhakar, Appaswamy T; Alexander, Mathew
• Is Part Of: Hospital practice. Volume 45:Issue 2(2017)
• Page Start: 46
• Page End: 50
• Publication Date: 2017-03-15
• Subjects: Carbamazepine (CBZ) -- anti-epileptic -- dried blood spot (DBS) -- therapeutic drug monitoring (TDM) -- high performance liquid chromatography (HPLC); Medicine -- Periodicals; Hospital care -- Periodicals; Hospitals; Medicine; Médecine -- Périodiques; Soins hospitaliers -- Périodiques; Hôpitaux -- Périodiques; Hospital care; Medicine; Periodicals; 616.005
• DOI: 10.1080/21548331.2017.1296318
• Languages: English
• ISSNs: 2154-8331

Abstract

ABSTRACT: Objectives : Carbamazepine (CBZ) is a commonly used anti-epileptic in rural hospitals in India. These hospitals lack the facilities to measure CBZ concentration; however, in larger hospitals this is performed using high performance liquid chromatography (HPLC). Dried blood spot (DBS) represents a feasible matrix for safe transportation by post/courier. This study was to determine whether the concentration of CBZ in serum can be predicted from that measured in DBS using an inexpensive HPLC method and inexpensive standard filter paper. Methods : CBZ in serum and DBS from 80 epileptic patients were measured using a validated HPLC assay. The data was then randomly divided into two groups; simple Deming regression was performed with the first group and validation was performed using the second. Results : There was a good correlation between the serum and DBS concentrations (r = 0.932) in the first group. The regression equation obtained was: predicted serum concentration = DBS concentration x 0.83 + 1.09. In the validation group, the correlation between the predicted and actual serum concentrations was also good (r = 0.958), and the mean difference between them was only 0.28 μg/ml ( p = 0.8062). The imprecision and bias in both the groups were acceptable. Conclusion : Using inexpensive materials, serum CBZ concentrations can be accurately predicted from DBS specimens. This method can be recommended for the therapeutic drug monitoring of CBZ in resource-limited settings.