The analysis of Δ⁹-tetrahydrocannabinol (Δ⁹-THC) and its main metabolites [11-hydroxy-Δ⁹-tetrahydrocannabinol (11-OH-Δ⁹-THC) and 11-nor-9-carboxy-Δ⁹-tetrahydrocannabinol] in serum is a routine assay in forensic toxicology in the case of drivers influenced by Cannabis abuse and in other cases. Analysis of the specimen may involve protein precipitation, although there are authors who do not use this step. In this study we investigated the effect of acetonitrile as protein precipitant added to the serum on the absolute extraction recoveries of the analytes. This is very important not only from a forensic point of view, but also from the aspect of impact of Δ⁹-THC therapy. Our results showed that in the case of spiked serum (2 ml), 80–87 % extraction recovery can be achieved if 4 ml of acetonitrile is added before solid phase extraction. The second best result could be reached if no acetonitrile was added (64–73 %). However, in the case of physiological sera of Cannabis consumers, no precipitation may be more advantageous in some cases. Matrix effects, which were studied by comparing the detectability and relative intensities of matrix peaks within the corresponding time windows of the analytes, were less influenced if the extraction was achieved with or without acetonitrile.