L-Carnitine, a naturally occurring compound, is indicated in the treatment of primary systemic carnitine deficiency. To assess the differences in pharmacokinetic parameters calculated from data corrected for baseline versus those from "uncorrected" data, compartmental fitting was carried out for baseline corrected and original plasma concentration data obtained following a single intravenous (iv) dose of 20 mg/kg. For free L-carnitine, mean volumes of distribution at steady state (Vdss) of the central compartment were similar using either approach (9.86 versus 11.2 L). However, Vdss (54.0 versus 29.0 L) and apparent elimination half-life (17.4 versus 5.0 h) were significantly different between the two data bases. Similar observations were noted for pharmacokinetic parameters based on plasma concentrations of total L-carnitine. Although the pharmacokinetic parameters obtained after baseline correction may represent the kinetics of a bolus dose, the pharmacokinetic parameters from uncorrected plasma data probably represent the clinical settings for patients. Baseline correction also probably has its greatest value in attempting to determine and/or define the biological half-life and Vdss for the "exogenously" administered dose and uncorrected data best describes the pharmacokinetics of composite endogenous and exogenous L-carnitine levels.