A rapid AFB1 detection method by ELISA kit test was used on feedstuff samples, and compared to an HPLC method, to verify if the presence of clay-adsorbent (SA) could cause erroneous quantification of the toxin. Samples were obtained using two AFB1-contaminated feedstuffs (7.92 and 17.58 µg/kg for low and high contaminated feeds; LC and HC respectively), added either one of two commercial SAs (Atox® and Myco AD) and three different inclusion doses (0, 10 and 20 g/kg, respectively for CTR, 1% and 2% doses). The HPLC and ELISA data were compared in CTR samples with a paired t-test. The AFB1 recoveries, performed with ELISA, were analysed as a completely randomized design using a 2×2×3 factorial arrangement. The ELISA method tended to underestimate the AFB1 concentrations with respect to the HPLC method, both in HC (P=0.050) and in LC (P<0.001) feedstuffs. A more drastic reduction (P<0.001) was observed when SAs were included in the two feedstuffs. In particular, Atox® determined an AFB1 recovery of 15,5% in HC and 7,6% in LC (1% dose) and of 11,1% in HC and 8,4% in LC (2% dose). Less severe penalisation were observed when Myco AD was added to feeds.