Rapid spectrophotometric determination of plasma carnitine concentrations

A spectrophotometric enzymatic assay for plasma carnitine concentrations has been automated on the Monarch 2000. Prior to the assay, each plasma sample was divided into three fractions, ie, free carnitine, acid-soluble carnitine and total carnitine, in order to determine the concentration of both free and esterified carnitine. Using the method developed by Tachikawa et al (Seikagaku 56:998, 1984), each of the samples was then chromatographed on an anion exchange resin to eliminate those compounds which could adversely impact the accuracy of the enzymatic assay for carnitine. After the completion of these preparatory steps, 32 specimens were assayed in less than 16 min on the Monarch 2000 with a high degree of both accuracy and precision. The assay was linear over a wide concentration range (5.0-80 mumol/liter), with the lower limit of sensitivity being 5.0 mumol/liter. The coefficient of variation (CV%) of the within run precision was 2.1%, 2.8%, and 6.7% for the determinations of free carnitine, acid-soluble carnitine, and total carnitine, and 17.4% and 27.8% for the calculated values of short-chain and long-chain acylcarnitines, respectively. The CV% of the between run precision for the same fractions was 6.5%, 2.7%, 3.8%, 14.2%, and 14.3%, respectively. When authentic L-carnitine was added to the plasma, the mean recovery rate was 94.7 +/- 11.0%. Reference values were determined using plasma obtained from 40 healthy adult volunteers.