Cyclic voltammetry (CV) on a bare glassy carbon electrode (GCE) has been applied to measure human and horse plasma antioxidant activity. The CV response of human plasma consisted of two broad voltammetric peaks observed in the potential range from 0.2 to 0.6 V and from 0.6 to 0.9 V. Horse plasma showed no voltammetric response on the non-activated GCE. Electrochemical activation in 0.5 H2SO4 induced a response similar to that in human plasma. Parameters that indicate the antioxidant capacity (AC) of the samples, i.e. the peak potential E-p, the peak current density i(p) and the charge Q below voltammetric waves were calculated for both waves. CV results have been compared with the two spectrophotometric methods, the Brand-Williams assay using a stable free radical a,a-diphenyl-b-picrylhydrazyl (DPPH.) and the commercial Randox (Total Antioxidant Status) TAS assay utilizing 2,2'-azino-di-[3-ethyl-benzthiazoline sulphonate] (ABTS(.+)) radical. The DPPH assay results have been shown to correlate well with the peak current densities i(pl) and the charge Q(1) of the first voltammetric wave showing that both the CV and DPPH methods reflect the AC of the two major water-soluble low molecular weight antioxidants (LMWA) in plasma - ascorbic acid and uric acid. Both methods show linear correlation with respect to plasma sample dilution. Comparison of the CV with the TAS method has shown that these two methods yielded considerably different chemical information. The TAS assay gave significantly higher values of the AC with the overall lower spread of the results than the CV and DPPH methods. Based on the CV scans in human and horse plasma that have indicated considerably different concentration of low-redox potential LMWA it has been concluded that the TAS results include contribution of the high formal redox-potential LMWA comprising the second CV wave. (c) 2005 Elsevier B.V. All rights reserved.