Due to inadequacies in analytical methodology of total antioxidant capacity (TAC) measurement, proteins are initially separated from the human serum matrix by precipitation and are left unmeasured, thereby causing an important "antioxidant gap.'' The aim of this work is to measure the TAC of serum with the modified CUPric Reducing Antioxidant Capacity (CUPRAC) method, and to identify the contribution of serum proteins, especially thiol-containing proteins, to TAC. CUPRAC results were statistically compared to those found by reference methods, namely ABTS (2,2'-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid), FRAP (ferric reducing antioxidant power), and Ellman thiols assay. The curves of absorbance vs thiol concentration, as well as of absorbance vs diluted serum (whole serum, trichloroacetic acid (TCA)-precipitated and redissolved serum protein solution, and TCA supernatant fractions) volume, of three distinct serum samples showed excellent linearity and low intercept values only with the modified CUPRAC method. The proposed method will help characterize the "antioxidant gap'' of serum TAC originating from protein components which should not be neglected in future antioxidant measurements.