To test if different leucine tracers behave in an indistinguishable manner and, by implication, that their metabolism is identical to that of natural leucine, we measured whole body leucine turnover in dogs and humans and fibrinogen synthesis in dogs by simultaneously infusing either [1-14C]leucine or [4,5-3H]leucine or [1-14C]alpha-ketoisocaproate (KIC) and [4,5-3H]KIC. Whole body leucine fluxes calculated from the plasma specific activity of the transaminated product of the infused tracer (reciprocal pool model) were lower (dogs by 5.7%; humans by 6.4%, both P < 0.02) when the plasma 3H specific activity compared to 14C specific activity were used with leucine tracers and were also lower (dogs by 4.4%, P < 0.02; humans by 8.6%, P < 0.06) using the KIC tracers. Using leucine or KIC tracers in dogs, the fractional rate of fibrinogen synthesis was 6.7% or 9.4% lower, respectively, (P < 0.02) using the 3H versus the 14C tracer. The apparently lower incorporation of 3H into protein was only in part accounted for by detritiation (2.1%, P = 0.05) of [3H]leucine during acid hydrolysis of proteins. These results suggest that in vivo and/or in vitro differential isotope effects are small (approximately 5%), but should be considered when dual isotopes infusions are employed to partition amino acid metabolism.