Assessment of oxidative stress is an important but technically challenging procedure in medical and biological research. The reactive oxygen metabolites (d-ROMs) test is a simple assay marketed for analyzing the total amount of hydroperoxides in serum via the Fenton's reaction. Earlier reports have raised a suspicion that a part of the signal detected in the assay comes from sources other than metabolites generated by oxidative stress. The aim of this study was to identify which serum components interfere with the d-ROMs signal. By application of sodium azide, ethylenediaminetetraacetic acid, sodium dodecylsulphate, varying temperature, and spiking endogenous substances we demonstrate that in the case of mammalian sera the assay determines ceruloplasmin (CP) activity with potential interferences from hydroperoxides, iron level, thiols, and albumin. In sera of avian species hydroperoxides contribute more to the test outcome, but the CP part is insensitive to inhibition by azide. In conclusion, this assay has deficiencies in terms of detecting realistic concentrations of hydroperoxides, is mostly measuring CP and is also interfered with other serum components, making it very difficult to interpret in most biological systems.