A gas chromatographic/mass spectrometric method based on the use of short capillary gas chromatograph columns (3-5 m) and electron ionization mass spectrometry has been optimized and evaluated for the determination of underivatized anabolic steroids. The short-column gas chromatographic/mass spectrometric method was shown to result in short analysis times and to require minimal sample preparation, but suffered from some loss in sensitivity and chromatographic resolution compared with conventional gas chromatographic/mass spectrometric techniques for derivatized steroids. Therefore, short-column gas chromatographic conditions were optimized to maximize the sample transfer efficiency (sensitivity) from the gas chromatograph into the ion source of the mass spectrometer, while maintaining chromatographic integrity and minimizing thermal decomposition. Mass spectrometric conditions were optimized to maximize ionization efficiency with respect to the intensity of the molecular ion and degree of fragmentation such that positive identification of each steroid could be made based on the resulting mass spectra. Under optimized conditions, we have shown that underivatized anabolic steroids spiked into urine samples can be determined at low-nanogram levels using short-column chromatography/full-scan electron ionization mass spectrometry.

Short-column (3.5 m) gas chromatography (GC)/tandem mass spectrometry (MS/MS) has been investigated for the detection of structurally related, underivatized anabolic steroids in urine. The approach described here demonstrates the ability to rapidly and qualitatively detect underivatized anabolic steroids in spiked urine matrices. In this approach, underivatized steroids are determined using a short-column GC separation, ionized by positive ion chemical ionization, and detected by selected reaction monitoring MS/MS. This approach permits positive identification of underivatized anabolic steroids based on retention time and the production of characteristic product ions. Preliminary detection limits studies in spiked urine samples showed quantitative results between 2 and 40 ng steroid per milliliter of uterine. The potential advantages of this approach compared to present screening methods based on conventional (30 m) GC/MS are its rapidity and selectivity. Reliable qualitative identification can be performed with a short-column GC/MS/MS analysis of less than 6 min with a reduction in sample preparation time due to the elimination of the derivatization step.