Quantitative determination of 2-hydroxy-3-methoxy-6 beta-naltrexol (HMN), naltrexone, and 6 beta-naltrexol in human plasma, red blood cells, saliva, and urine by gas liquid chromatography.

Two gas liquid chromatographic methods differing mainly in sensitivity are described for the quantitative determination of 2-hydroxy-3-methoxy-6 beta-naltrexol (HMN), a minor metabolite of naltrexone (NT), in human body-fluids. The methods also incorporate simultaneous determinations of naltrexone and its major human metabolite, 6 beta-naltrexol (beta-OL), in urine, serum (or plasma), red blood cells (RBC), and saliva. Flame ionization detection of the bis-(trimethylsylil) trifluoroacetamide (BSTFA) derivatives provided sufficient sensitivity for quantitation of the bases in urine. However, lower levels in serum, RBC and saliva necessitated the use of more sensitive electron capture detection of the pentafluoropropionate (PFPA) derivatives of the bases. Because HMN and 6 beta-naltrexol PFPA derivatives have nearly identical gas chromatographic retention times, their separation was achieved by differential extraction, based on their different partition characteristics between aqueous and organic solvents. In the plasma of 4 subjects, 16 and 24 hrs. after 2 X 200 mg NT doses, the relative percentages were 23.1% HMN, 3.4% NT and 73.5% beta-OL. In urine samples collected at the same time as the blood samples the relative percentages were 14.4% HMN, 9.0% NT and 76.6% beta-OL. The nonpolar nature of HMN and the greater polarity of beta-OL may have influenced their differential distribution into RBCs and saliva. In the RBCs, 96.1% HMN and no significant amount of beta-OL was found, in saliva, 92.3% of beta-OL and no HMN was found.

Evaluation of the radioimmunoassay for benzoylecgonine (a cocaine metabolite) in human urine.

The 125I-radioimmunoassay (RIA) for benzylecgonine (cocaine metabolite) in urine was evaluated by comparison with gas-liquid chromatography (GLC) and thin-layer chromatography (TLC) and the Enzyme-Multiplied Immunoassay Technique (EMIT). By RIA, a statistically significant concentration, 2 microng/liter, was observed for urinary benzoylecgonine. The coefficient of variation for the RIA was 2.58+/-0.38% inter-assay and 2.20+/-0.14% intraassay. There was cross-reactivity with cocaine (more reactive than benzoylecgonine) and other members of the tropane family of alkaloids. There was agreement between results by RIA and GLC in 95.5% of the samples, between RIA and TLC in 87.0%, and between RIA and EMIT in 84.5%. The percentage of true false-positives was 3.5% for the RIA in comparison to GLC, 8.8% in comparison to TLC, and 9.1% in comparison to EMIT. True false-negatives were insignificant (0 to 1.0%). GLC and RIA results correlated highly (phi=0.908). GLC, therefore, was the best comparison method for this evaluation study. RIA for benzoylecgonine is sensitive, reproducible, and reliable for the detection of cocaine in urine.