Measurement of muscle protein synthesis using stable isotopically labeled tracers usually requires isotope ratio mass spectrometry (IRMS) because of the need to measure very low enrichments of stable isotopically labeled tracers (tracer to tracee ratio [TTR], 0.005% to 0.10%). This approach is laborious, requiring purification of the metabolite of interest and combustion to a gas for IRMS analysis, and is best suited for use with 13C tracers. We have developed an approach whereby low enrichments can be conveniently measured by a conventional gas chromatography/mass spectrometry (GC/MS) instrument. The approach includes three critical elements: (1) use of a highly substituted tracer containing three or more labeled atoms, to measure enrichment above a very low natural abundance of highly substituted isotopomers; (2) use of a highly substituted natural abundance isotopomer as a base ion for comparison rather than the most abundant m + 0 isotopomer, to reduce the dynamic range of the isotopomer ratio measurement; and (3) a sensitive mass spectrometric analysis that measures the natural abundance of the isotopomer used as a tracer with a high signal to noise ratio (> 100:1). This approach was used to measure the rate of synthesis of muscle protein following a primed continuous infusion of L-[13C6]-phenylalanine (PHE) in eight fasted dogs and L-[2H3]-leucine in five fasted human subjects. Values for [13C6]-PHE enrichment by GC/MS rates were virtually identical to those obtained by a conventional approach using high-performance liquid chromatography (HPLC) to isolate PHE, combustion to CO2, and measurement of 13CO2 enrichment by IRMS (IRMS enrichment = 0.9988 x GC/MS enrichment, R2 = .891), resulting in identical values for muscle fractional synthesis rates ([FSRs] mean +/- SEM: 2.7 +/- 0.2 and 2.5 +/- 0.2%/d for GC/MS and IRMS, respectively). Human muscle synthesis rates measured by GC/MS analysis of [2H3]-leucine enrichment (1.90 +/- 0.17%/d) were similar to published values based on IRMS analysis using a 1- 13C-leucine tracer. We conclude that compared with the IRMS approach, the GC/MS approach offers faster throughput, has a lower sample requirement, and is suitable for a wider variety of tracers such as 2H. The principles outlined here should be applicable to the measurement of low enrichments by GC/MS in a wide variety of stable isotope tracer applications.