Comparison of effects of different anticoagulants and sample handling procedures on rat insulin radioimmunoassay

1. The effects of several potential inducers of artifacts on insulin RIA: use of anticoagulant (heparin, EDTA) versus clotting , repeated freezing/thawing and the presence of added ions (Ca2+, Zn2+) has been studied on a large uniform pooled sample of rat blood. 2. Repeated freezing and thawing resulted in a very significant loss of RIA insulin. 3. The use of heparin or added lipid (Intralipid) to serum samples resulted in lower insulin measurements, whereas low protein or EDTA addition resulted in higher values. The effect of EDTA was not a consequence of its sequestering of calcium or zinc, as their combined addition did not change the EDTA effect. 4. The use of individualized non-specific binding for actual insulin estimations did not correct most of the artifacts induced by the different treatments studied. 5. As a general conclusion, utmost care must be taken to prevent the incidence of these sources of error by referring the data to controls, avoiding the use of EDTA as well as taking into account the actual composition and handling of the samples when measuring their insulin levels.