Effect of oxidizing adulterants on human urinary steroid profiles

Steroid profiling is the most versatile and informative technique adapted by doping control laboratories for detection of steroid abuse. The absolute concentrations and ratios of endogenous steroids including testosterone, epitestosterone, androsterone, etiocholanolone, 5α-androstane-3α,17β-diol and 5β-androstane-3α,17β-diol constitute the significant characteristics of a steroid profile. In the present study we report the influence of various oxidizing adulterants on the steroid profile of human urine. Gas chromatography–mass spectrometry analysis was carried out to develop the steroid profile of human male and female urine. Oxidants potassium nitrite, sodium hypochlorite, potassium permanganate, cerium ammonium nitrate, sodium metaperiodate, pyridinium chlorochromate, potassium dichromate and potassium perchlorate were reacted with urine at various concentrations and conditions and the effect of these oxidants on the steroid profile were analyzed. Most of the oxidizing chemicals led to significant changes in endogenous steroid profile parameters which were considered stable under normal conditions. These oxidizing chemicals can cause serious problems regarding the interpretation of steroid profiles and have the potential to act as masking agents that can complicate or prevent the detection of the steroid abuse.     

Development of a liquid chromatography-tandem mass spectrometry method for the determination of 23 endogenous steroids in small quantities of primate urine

A quantitative method using liquid chromatography-tandem mass spectrometry (LC-MS-MS) was developed for the simultaneous determination of 23 endogenous steroids in primate urine. The introduced method includes estrone, pregnandiol, cortisol, testosterone and several human urinary glucocorticoid and androgen metabolites. As the method is intended for the analysis of steroid hormones in behavioral studies on wild-living primates, it was adapted for a sample volume of 200microL urine. The sample preparation consisted of an enzymatic hydrolysis of steroid glucuronides using beta-glucuronidase from E. coli followed by a solvolytic cleavage of steroid sulfates employing sulfuric acid/ethyl acetate. The extraction of steroids from urine was optimized with respect to pH during extraction, type of ether and the amount of enzyme necessary for complete hydrolysis of glucuronides. The recovery of steroids spiked into urine before hydrolysis was 58.9-103.7% with an intra-day precision of 2.7-14.3% and an inter-day precision of 2.9-14.8%. Detection limits ranged from 0.1-0.5ng/mL. The reproducibility of the whole sample preparation process was also demonstrated for unspiked urine (CV 1.2-16.5%). The proportion of steroid hormone excreted as sulfate was determined for 21 steroids in chimpanzee urine. The solvolysis proved to be essential for all investigated steroids except for pregnandiol, tetrahydrocortisol and tetrahydrocortisone, which were found to be less then 10% in the solvolysis fraction.     

Direct determination of anabolic steroid conjugates in human urine by combined high-performance liquid chromatography and tandem mass spectrometry

A novel screening procedure for the sulfate and glucuronide conjugates of testosterone (T) and epitestosterone (E) in human urine was developed based on liquid-solid extraction and microbore high-performance liquid chromatography combined on-line with ion-spray tandem mass spectrometry. Confirmation of the sulfate and glucuronide conjugates of testosterone and epitestosterone isolated frrm normal human urine was acheived by selected reaction monitoring of characteristic product ions of the parent compounds. Endogenous levels of the steroid conjugates are detected in normal male urine and an increase is observed when the sample is fortified with authentic analytical standards of the conjugates. Calibration curves of all steroid conjugates in urine are linear over a range of twenty. Deuterated internal standards of testosterone glucuronide and epitestosterone sulfate were used for quantitation of the endogenous conjugates. T/E ratios were determined based on the glucuronide fractions of six replicates from a normal male and were shown to be statistically reproducible and below the accepted T/E threshold of 6:1. Sulfate conjugates were shown to be present at significantly lower levels in the urine. The method has potential as an alternative for monitoring anabolic steroid conjugates in human urin.     

A new method for hydrolyzing sulfate and glucuronyl conjugates of steroids

A new method for hydrolyzing steroid conjugates (both sulfates and glucuronides conjugates) that is efficient, effective, and inexpensive is described. This method comprises incubation of the conjugates--after salting-out into ethyl acetate or elution from a C18 cartridge--with anhydrous methanolic hydrogen chloride (methanolysis) for 10 min. It has been successfully applied to our routine radioimmunoassay screening and GC/MS confirmation studies of steroids in prerace and postrace equine urine samples. Comparative GC/MS studies on entire (male horse) urine samples showed that methanolysis gave amounts of free steroids (estrone, estradiols, testosterone, estrenediols, nandrolone, androstanediols) at least as large as those obtained by solvolysis. Similar studies on urine samples from a gelding that had been administered nandrolone phenylpropionate showed that methanolysis gave larger amounts of free steroids (nandrolone, estranediols) than Helix pomatia enzymatic hydrolysis or solvolysis. Also, TLC studies on methanolysis of corticosteroid conjugates such as hydrocortisone 21-sulfate and hydrocortisone 21-phosphate showed that free corticosteroid was released in 5 min.     

A radioimmuno—chromatographic scanning method for the analysis of testosterone conjugates in urine and serum

A method for the analysis of testosterone (and 5α-dihydrotestosterone) conjugates in human serum and urine samples is described. The samples were brought to pH 1 and extracted with a diethyl ether—methanol mixture. After evaporation the residues were run in a thin-layer chromatography system, individual samples' paths were cut into 1-cm long pieces and eluted with methanol. The methanol was evaporated and the residue subjected to acid hydrolysis. The released steroid was extracted by diethyl ether and measured by radioimmunoassay. The methodology described represents a new approach to the qualitative and quantitative study of steroid conjugates in serum and urine, and can easily be applied to the study of steroid conjugates in other biological material.     

Urine sex steroid profile in healthy volunteers during isolation in a pressurized compartment

In order to study the time course of the parameters of urine sex steroid profile and its potential boundary changes, the quantitative determination of a number of endogenous steroids and their metabolites in healthy human urine has been carried out by the gas chromatography method using mass-selective detection. The samples were obtained from six volunteers under the conditions of total monitoring of vital activity factors affecting urine steroid profile (diet, water consumption, physical activity, temperature and air composition, day-night rhythm, and the psychoemotional state) in an experimental study using a pressurized compartment. The healthy human profile parameters of urine steroids were found, which were affected by conditions of controlled vital activity in a pressurized compartment. Parameters of the individual and group variability of the parameters of steroid profile and their dependence on experimental factors, salt consumption mode, and autonomous vital activity, were found.     

Profile of sex steroids in healthy volunteers' urine during experiment in isolated object

The quantitative determination a number of endogenous steroids and their metabolites in urine of healthy volunteers by means of gas chromatography - mass spectrometry was performed. The dynamic of steroid profile of healthy individuals as well as possible ranges of several endogenous steroid parameters have been investigated. Samples were obtained during 105-days experiment with 6 volunteers in isolated on ground modules where were modeling the main life conditions which could influence the steroid profile: meal volume and composition, water consumption, motion activity, air composition and temperature, rate sleep - wakefulness and emotional tension. The parameters of urine steroid profile of healthy volunteer which were affected by life conditions in isolated object were revealed. The parameters of individual and group variability of steroid profile and its dependence from definite experiment conditions - change of salt consumption periods, autonomy of vital activity were detected.     

Sodium ascorbate improves yield of urinary steroids during hydrolysis with Helix pomatia juice

Urinary steroid profile analysis requires enzymatic hydrolysis of glucuronide and sulfate conjugates and this is achieved simultaneously using Helix pomatia juice (HPJ), but steroids with 3beta-hydroxy-5-ene structure undergo transformation and yield of 5alpha-reduced corticosteroid glucuronides is poor. We describe the use of sodium ascorbate to solve these problems and provide a basis for its mode of action. Steroid conjugates were extracted from urine, hydrolyzed in acetate buffer with HPJ and sodium ascorbate and analyzed as methyloxime-trimethylsilylether derivatives by gas chromatography-mass spectrometry. Ranges of temperature, pH and ascorbate, substrate and HPJ concentrations were compared for urine and pure standards. Activity of other antioxidants and that of bacterial cholesterol oxidase were examined. Helix pomatia enzyme preparations from different commercial sources were compared. Loss of 3beta-hydroxy-5-ene steroids was enzyme-dependant, since it required HPJ, was saturable, subject to substrate competition and heat-inactivated. Products were 3-oxo-4-ene steroids and 4,6-diene and 6-oxy derivatives of these but the latter were not formed from 3-oxo-4-ene precursors. Ascorbate, other antioxidants or oxygen exclusion diminished activity. These characteristics were shared by cholesterol oxidase. Yield of 5alpha-reduced steroids was diminished by pre-incubation of HPJ before ascorbate addition and this was reversed if ascorbate was added to the pre-incubation mixture. We conclude that transformation of 3beta-hydroxy-5-ene steroids by HPJ is due to cholesterol oxidase and is diminished by antioxidants or oxygen denial. Yield of 5alpha-reduced steroids is low due to oxidative damage of beta-glucuronidase during hydrolysis, prevented by ascorbate. These features are shared by most commercial Helix pomatia enzyme preparations tested.     

Enzyme immunoassay of 17 beta-estradiol, estrone conjugates, and testosterone in urinary and fecal samples from male and female mice.

ELISA measures of 17 beta-estradiol, estrone conjugates, and testosterone were adapted for fecal and urinary samples from laboratory mice. We will report on validations of these assays and data from interacting males and females. Unconjugated gonadal steroids were consistently measurable in urine and feces of both males and females. Females that were parturient following insemination excreted relatively low levels of urinary testosterone compared to non-parturient females. The results are consistent with evidence that elevated androgens and estrogens are incompatible with intrauterine implantation of fertilized ova, and suggest that steroids in male urine could contribute to pheromonal action. These methods permit repeated noninvasive measurement of steroid activity in this species.     

Method for combined analysis of profiles of conjugated progesterone metabolites and bile acids in serum and urine of pregnant women

A method for analysis of profiles of conjugated progesterone metabolites and bile acids in 10 ml of urine and 1–4 ml of serum from pregnant women is described. Total bile acids and neutral steroids from serum and urine were extracted with octadecylsilane-bonded silica. Groups of conjugates were separated on the lipophilic ion-exchanger triethylaminohydroxypropyl Sephadex LH-20 (TEAP-LH-20). Fractions were divided for steroid or bile acid analyses. Sequences of hydrolysis/ solvolysis and separations on TEAP-LH-20 permitted separate analyses of steroid glucuronides, monosulfates and disulfates and bile acid aminoacyl amidates, sulfates, glucuronides and sulfate-glucuronides. Radiolabelled compounds were added at different steps to monitor recoveries and completeness of separation, and hydrolysis/solvolysis of conjugates was monitored by fast-atom bombardment mass spectrometry. The extraction and solvolysis of steroid disulfates in urine were studied in detail, and extraction recoveries were found to be pH-dependent. Following methylation of bile acids, all compounds were analysed by capillary gas chromatography and gas chromatography—mass spectrometry of their trimethylsilyl ether derivatives. Semiquantification of individual compounds in each profile by gas—liquid chromatography had a coefficient of variation of less than 30%. The total analysis required 3 days for serum and 4 days for urine.     

Development of a GC/C/IRMS method - Confirmation of a novel steroid profiling approach in doping control

In doping control, an athlete can only be convicted with the misuse with endogenous steroids like testosterone (T), if abnormal values of steroid metabolites and steroid ratios are observed and if the subsequent analysis with isotope ratios mass spectrometry (IRMS) confirms the presence of exogenously administered androgens. In this work, we compare the results of a novel steroid profiling approach with the performance an in-house developed IRMS method. The developed IRMS has the advantage over other methods to be relatively short in time and with target compounds androsterone, etiocholanolone, 5β-androstane 3α,17β-diol and 5α-androstane 3α,17β-diol. Pregnanediol was used as an endogenous reference compound (ERC). Reference limits for the IRMS values were established and applied as decision limits for the evaluation of excretion urine from administration with oral T, T-gel, dihydrotestosterone (DHT) - gel and dehydroepiandrosterone (DHEA). Results indicated the importance of both androstanediols as important IRMS markers where relative values compared to an ERC (Δδ(13)C) yielded better detection accuracy than absolute δ(13)C-values. The detection times of all administered endogenous steroids were evaluated using the proposed thresholds. The results of traditional steroid profiling and a new approach based upon minor steroid metabolites monitoring introduced in a longitudinal framework were evaluated with IRMS. With traditional steroid profiling methods, 95% of the atypical samples could be confirmed whereas an additional 74% of IRMS confirmed was provided by a new biomarkers strategy. These results prove that the other steroid profiling strategies can improve the efficiency in detection of misuse with endogenous steroids.     

Relevance of the selective oestrogen receptor modulators tamoxifen, toremifene and clomiphene in doping field: endogenous steroids urinary profile after multiple oral doses

The present study was performed to investigate the influence of the intake of selective oestrogen receptor modulators on the urinary endogenous steroids profile. For this purpose the circadian variability of luteinizing hormone, follicle-stimulating hormone, testosterone, 5α-androstan-3α,17β-diol, 5β-androstan-3α,17β-diol, epitestosterone, 4-androstenedione, androsterone and etiocholanolone were measured on eight subjects (four males and four females) by gas chromatography–mass spectrometry and chemiluminescent immunometric assay techniques before and after oral administration of multiple doses of either tamoxifen (80 mg for 2 days) or toremifene (120 mg for 2 days) or clomiphene (100 mg for 2 days). The individual baseline variability of the steroids studied was set up by collecting the urine samples every 3 h, for 3 days prior to the treatment; whereas the evaluation of the effects of the oral administration of multiple doses of selective oestrogen receptor modulators on the steroid urinary profile was assessed by collecting urine samples every three hours for at least five days from the first administration.The results of our measurements showed that, only in male subjects, the relative urinary concentrations of testosterone, epitestosterone and 4-androstenedione were significantly altered generally after the second day of drug administration. While no significant effects were recorded in both sexes on the luteinizing hormone, follicle-stimulating hormone, androsterone, etiocholanolone, 5α-androstan-3α,17β-diol and 5β-androstan-3α,17β-diol urinary levels and on testosterone/epitestosterone, 5α-androstan-3α,17β-diol/5β-androstan-3α,17β-diol and androsterone/etiocholanolone ratios.     

Enzyme-assisted synthesis and characterization of glucuronide conjugates of neuroactive steroids

Synthesis of reference standards is needed to determine the presence and function of steroid glucuronides in the brain or other tissues, because commercial sources of steroid glucuronide standards are limited or unavailable. In the present study porcine, rat, and bovine liver microsomes were tested to evaluate their ability to glucuronidate eight neurosteroids and neuroactive steroids of various types: dehydroepiandrosterone, pregnenolone, isopregnanolone, 5alpha-tetrahydrodeoxycorticosterone, corticosterone, cortisol, beta-estradiol, and testosterone. In general, the glucuronidation efficiency of rat liver was rather poor compared with that of bovine and porcine liver microsomes. Since porcine liver apparently has a relatively large amount of dehydrogenase, its microsomes also produced dehydrogenated steroids and their glucuronides, as well as various regioisomers in which the site of glucuronidation varied. In contrast, bovine liver microsomes produced mainly a single major glucuronidation product and few dehydrogenation products and gave the best overall yield for two-third of the steroids tested. The enzymatic synthesis of five glucuronides of four steroids was carried out and the conditions, purification, and analytical methods for the glucuronidation products were optimized. The steroid glucuronides synthesized were characterized by nuclear magnetic resonance spectroscopy (NMR) and liquid chromatography-mass spectrometry (LC-MS). The stereochemically pure steroid glucuronide conjugates were recovered in milligram amounts (yield 10-78%) and good purity (>85-90%), which is sufficient for LC-MS/MS method development and analyses of steroid glucuronides in biological matrices such as brain, urine, or plasma.     

Influences of β-HCG administration on carbon isotope ratios of endogenous urinary steroids

Several factors influencing the carbon isotope ratios (CIR) of endogenous urinary steroids have been identified in recent years. One of these should be the metabolism of steroids inside the body involving numerous different enzymes. A detailed look at this metabolism taking into account differences found between steroids excreted as glucuronides or as sulphates and hydrogen isotope ratios of different steroids pointed out possibility of unequal CIR at the main production sites inside the male body - the testes and the adrenal glands. By administration of β-HCG it is possible to strongly stimulate the steroid production within the testes without influencing the production at the adrenal glands. Therefore, this treatment should result in changed CIR of urinary androgens in contrast to the undisturbed pre-treatment values. Four male volunteers received three injections of β-HCG over a time course of 5 days and collected their urine samples at defined intervals after the last administration. Those samples showing the largest response in contrast to the pre-administration urines were identified by steroid profile measurements and subsequent analysed by GC/C/IRMS. CIR of androsterone, etiocholanolone, testosterone, 5α- and 5β-androstanediol and pregnanediol were compared. While pregnanediol was not influenced, most of the investigated androgens showed depleted values after treatment. The majority of differences were found to be statistically significant and nearly all showed the expected trend towards more depleted δ(13)C-values. These results support the hypothesis of different CIR at different production sites inside the human body. The impact of these findings on doping control analysis will be discussed.     

Steroid metabolism in human and boar testis tissue. Steroid concentrations and the position of the sulfate group in steroid sulfates

The steroid composition of and steroid conjugation in human cadaver and boar testes were investigated by analyzing the endogenous steroids. Gas-liquid chromatography and gas chromatography-mass spectrometry were used to identify the steroids and to determine the position of the sulfate group in sulfate conjugates. For the latter purpose, the steroids were first acetylated and subsequently solvolyzed and converted to trimethylsilyl ethers.In addition to the compounds previously identified as endogenous components, human testis was also found to contain progesterone, 17_α-hydroxyprogesterone and 20_α-hydroxy-4-pregnen-3-one in the free steroid fraction and 3_β-hydroxy-17_α-5-pregnen-20-one in the monosulfate fraction.     

Capillary electrophoresis with UV detection and mass spectrometry in method development for profiling metabolites of steroid hormone metabolism

The aim of this study was to develop a method for comprehensive profiling of metabolites involved in mammalian steroid metabolism. The study was performed using the partial filling micellar electrokinetic chromatography (PF-MEKC) technique for determination of endogenous low-hydrophilic steroids. The detection techniques in capillary electrophoresis were UV absorption and electrospray mass spectrometry (ESI-MS). Thirteen steroids were included in the method development, and the selected were metabolites involved in major pathways of steroid biosynthesis. Although only eight of them could be separated and detected with UV, they could be identified by ESI-MS using selected ion monitoring (SIM) technique. Tandem MS spectra were also collected. UV detection was more sensitive than MS due to better separation of compounds and the selective signal sensitivity. The lowest limits of detection were 10-100ng/mL for cortisone, corticosterone, hydrocortisone and testosterone. The other steroids could be detected at 500-1000ng/mL. The identification of cortisone, corticosterone, hydrocortisone, estrogen and testosterone were made in patient urine samples and their concentrations were 1-40mug/L.     

Estimation of immunoreactive testicular androgen metabolites in the urine of saddle-back tamarins

Hydrolysis, extraction, and radioimmunoassay techniques for the estimation of excreted testosterone metabolites in the urine of saddle-back tamarins have been validated and are described. The steroids measured with the testosterone antiserum used are mostly present as glucuronides and sulfates. Immunoreactivity in high-performance liquid chromatography (HPLC) fractions of urinary extracts and in a standard mixture of cortisol, testosterone, and dihydrotestosterone (DHT) were compared. The fractions with the same retention data as testosterone accounted for the major part of the immunoreactivity. Several other immunoreactive compounds of unknown identity were present in low concentrations. These results suggest that testosterone conjugates are the major steroid metabolites measured with this method in Saguinus fuscicollis. Urinary testosterone levels of castrated males were much lower than those of intact males. Testosterone treatment of castrated males resulted in a temporary superphysiological increase in the levels of urinary testosterone and in an individually variable increase in the levels of the minor immunoreactive compounds. These results suggest that estimation of testosterone metabolite levels in urine is a valid method for the assessment of testicular activity in Saguinus fuscicollis.     

Alcohol intake, androgen and glucocorticoid steroids in premenopausal women using oral contraceptives: an interventional study

Long-term heavy alcohol intake is associated with endocrinological abnormalities the mechanisms of which are still unclear. The objective of the present study was to investigate the effect of alcohol intake on plasma and urine glucocorticoid and androgen steroid levels in healthy premenopausal women using oral contraceptives. In a placebo-controlled interventional study with a cross-over design including nine premenopausal women using oral contraceptives no effect of tolerance was observed with regard to the magnitude of the acute transient alcohol-induced testosterone elevation after a 1-week alcohol drinking period (0.8 g/kg per day). At non-intoxicated time points elevated plasma testosterone and androstenedione levels were found in the afternoon but not in the morning during the alcohol drinking period compared with placebo. An increase in plasma cortisol levels was observed after the discontinuation of alcohol drinking. No effects were observed in total glucocorticoid conjugates in morning urine spot samples. An increase during the alcohol period relative to placebo was, however, observed in the urine etiocholanolone/androsterone, tetrahydrocortisol/allotetrahydrocortisol as well as the 20-hydroxy-/20-ketosteroid ratios. No consistent effect was observed in the urine (tetrahydrocortisol+allotetrahydrocortisol)/tetrahydrocortisone ratio. It is suggested that the alcohol-induced alterations in plasma glucocorticoid and androgen levels during non-intoxicated conditions are due to a change in the hypothalamic-pituitary-adrenal function. The effects observed in the conjugated urine glucocorticoid and androgen ratios are likely to be mediated by a change in the metabolism of these steroids in the liver. The present results may be of relevance in the development of disturbances in the glucocorticoid as well as sex steroid balance among heavy female drinkers.     

Steroid sulfates in testis tissue of prostatic cancer patients treated for six months with the gonadotropin releasing hormone agonist buserelin

In order to assess the extent of inhibition of testicular steroidogenesis during long-term treatment of prostatic cancer with GnRH agonist, we measured the intratesticular levels of 5 steroid sulfate conjugates in human testis tissue removed from patients after 6 months of intranasal treatment with buserelin. The most pronounced decreases were found in testosterone and pregnenolone sulfates, to 1.6 and 7.1%, respectively, of concentrations measured in testis tissue from primarily orchiectomized prostatic cancer patients. In contrast, clearly smaller decreases were found in three other steroid sulfates measured, those of dehydroepiandrosterone (to 26%), 17-hydroxyprogesterone (to 27%) and 5-androstene-3 beta, 17 beta-diol (to 62%). These results are in keeping with our previous analyses of unconjugated steroids in similar tissue samples, and indicate that testicular steroidogenesis per se is not totally blocked by long-term intranasal treatment with GnRH agonist. Testicular steroid sulfate conjugation may be specifically suppressed since the total concentration of these conjugates decreased more than free steroid levels in our earlier measurements.     

Studies on anabolic steroids--12. Epimerization and degradation of anabolic 17 beta-sulfate-17 alpha-methyl steroids in human: qualitative and quantitative GC/MS analysis.

The epimerization and dehydration reactions of the 17 beta-hydroxy group of anabolic 17 beta-hydroxy-17 alpha-methyl steroids have been investigated using the pyridinium salts of 17 beta-sulfate derivatives of methandienone 1, methyltestosterone 4, oxandrolone 7, mestanolone 10 and stanozolol 11 as model compounds. Rearrangement of the sulfate conjugates in buffered urine (pH 5.2) afforded the corresponding 17-epimers and 18-nor-17,17-dimethyl-13(14)-enes in a ratio of 0.8:1. These data indicated that both epimerization and dehydration of the 17 beta-sulfate derivatives were not dependent upon the respective chemical features of the steroids studied, but were instead inherent to the chemistry of the tertiary 17 beta-hydroxy group of these steroids. Interestingly, in vivo studies carried out with human male volunteers showed that only methandienone 1, methyltestosterone 4 and oxandrolone 7 yielded the corresponding 17-epimers 2, 5 and 8 and the 18-nor-17,17-dimethyl-13(14)-enes 3, 6 and 9 in ratios of 0.5:1, 2:1 and 2.7:1, respectively. No trace of the corresponding 17-epimers and 18-nor-17,17-dimethyl-13(14)-enes derivatives of mestanolone 10 and stanozolol 11 was detected in urine samples collected after administration of these steroids. These data suggested that the in vivo formation of the 17-epimers and 18-nor-17,17-dimethyl-13(14)-enes derivatives of 17 beta-hydroxy-17 alpha-methyl steroids is also dependent upon phase I and phase II metabolic reactions other than sulfation of the tertiary 17 beta-hydroxyl group, which are probably modulated by the respective chemical features of the steroidal substrates. The data reported in this study demonstrate that the 17-epimers and 18-nor-17,17-dimethyl-13(14)-enes are not artifacts resulting from the acidic or microbial degradation of the parent steroids in the gut as previously suggested by other authors, but arise from the rearrangement of their 17 beta-sulfate derivatives. Unchanged oxandrolone 7 was solely detected in the unconjugated steroid fraction whereas unchanged steroids 1, 4 and 11 were recovered from the glucuronide fraction. These data are indirect evidences suggesting that the glucuronide conjugates of compounds 1 and 4 are probably enol glucuronides and that of compound 11 is excreted in urine as a N-glucuronide involving its pyrazole moiety. The urinary excretion profiles of the epimeric and 18-nor-17,17-dimethyl-13(14)-ene steroids are presented and discussed on the basis of their structural features.