In order to evaluate the impact of blood sample handling processes on circulating TGF-β1 levels, blood specimens were obtained from 13 healthy volunteers using different handling processes (kept at room temperature (RT) or on ice before centrifugation, using different centrifugal forces). TGF-β1 levels were measured using an enzyme-linked immunosorbent assay. A paired-T test was used for statistical analysis. The TGF-β1 level in on-ice serum was significantly lower than that in room-temperature serum (P<0.001), and both were significantly higher than that found in on-ice plasma (P<0.001). Compared with on-ice plasma samples, the longer the samples were kept at RT, the higher the levels of TGF-β1 in plasma (P=0.268, 0.040, and 0.0015 for 5 min, 30 min, and 60 min in RT, respectively). Compared with plasma centrifuged at 2,500×g for 30 min, the TGF-β1 levels were much lower than those found in plasma centrifuged at 1,200×g for 10 min (P=0.003); and a double centrifugation before TGF-β1 detection, significantly decreased the level (P<0.001). It is suggested that the optimal sampling conditions for the detection of TGF-β1 should be plasma prepared on ice and spun down at a higher centrifugal force.