Received Date: October 29, 2012; Accepted Date: November 19, 2012; Published Date: November 21, 2012
Citation: Hostrup M, Kalsen A, Hemmersbach P, Backer V. (2012) Intra-Individual Variability in the Urine Concentrations of Inhaled Salmeterol in Male Subjects with Reference to Doping Analysis – Impact of Urine Specific Gravity Correction. J Sports Med Doping Stud 2:118. doi:10.4172/2161-0673.1000118
Copyright: © 2012 Hostrup M. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Sports Medicine & Doping Studies
Since 2010, the World Anti-Doping Agency (WADA) has introduced urinary thresholds for some beta2-agonists. In doping analysis urine samples of beta2-agonists are not corrected for the Urine Specific Gravity (USG) by the WADA laboratories. Several studies have observed high differences in the urine concentrations of beta2-agonists when correction for USG is compared with no correction, as well as high inter-individual variability between subjects. However, no studies have measured the intra-individual variability after inhalation of the long-acting beta2-agonist salmeterol. As such, the purpose of this study was to measure the intra-individual variability in the urine concentrations of salmeterol and its metabolite α-hydroxysalmeterol. Furthermore, to highlight the variability between corrected and uncorrected urine samples for USG. Urine samples from 20 subjects were analyzed for USG, urine excretion and urine concentrations of salmeterol and α-hydroxysalmeterol. Seven of the subjects underwent a second visit with the same procedures. At each visit 100 μg salmeterol was administered by inhalation. Urine samples were collected before administration of the drug (T0) and 4 (T4), 8 (T8) and 12 (T12) hours after administration. The mean relative differences in the urine concentrations of salmeterol and α-hydroxysalmeterol between USG corrected and uncorrected samples were 43 ± 44, 27 ± 42 and 56 ± 87% at T4, T8 and T12, respectively. The intra-individual variability in the urine excretion of salmeterol and α-hydroxysalmeterol during visits one and two were 12.6 and 21.8%, respectively. The intra-individual variability of salmeterol and α-hydroxysalmeterol in the urine concentrations were significantly
higher when uncorrected for USG with 43.0 and 43.7% versus 20.4% (p<0.01) and 28.0% (p<0.05), respectively. Correction for USG reduces inter-individual and intra-individual variability in urine concentrations of salmeterol and α-hydroxysalmeterol.
α-Hydroxysalmeterol; Salmeterol; Urine; Urine specific gravity; Beta2-agonists; Doping
Salmeterol is a long-acting selective beta2-agonist that is frequently used for the treatment of asthma . In competitive sports the prevalence of asthma and exercise-induced bronchoconstriction is high, and a large number of especially endurance athletes use inhaled beta2-agonists. At the Olympic Games, in the period 2000-2010, it was well-documented, that the asthmatic athletes outperformed their healthy counterparts by relatively winning more medals . Even though beta2-agonists are prevalently used by athletes, the World Antidoping agency (WADA) loosened their restrictions towards salbutamol in 2010 followed by a loosening in 2011 for salmeterol, which is now allowed in therapeutic doses by inhalation as long as the athlete declare the use of it in doping tests, i.e. “declaration of use” .
The potential performance-enhancing effects of beta2-agonists  stress the importance of urinary thresholds in order to distinguish between therapeutic use and potential ergogenic misuse. The recently released prohibited list of 2013 contains no urinary threshold for salmeterol, and at the moment athletes can inhale supratherapeutic doses of salmeterol without any consequences. This is problematic as salmeterol may have an anabolic and performance-enhancing effect . The prohibited list only contains urinary thresholds for salbutamol (1,000 ngmL-1) and formoterol (40 ngmL-1) and therefore, the need for a urinary threshold for salmeterol should not be underestimated in order to avoid misuse and promote fair competition. Several studies have investigated the urine concentrations after inhalation of salbutamol, terbutaline and formoterol, thus providing a basis for the current urinary thresholds on the prohibited list [6-8]. In this category, the scientific literature for salmeterol is sparse, and only a few studies have investigated the urinary response to inhaled salmeterol [9,10].
Furthermore, information about inter-individual and intraindividual variability in the urine concentrations of salmeterol would be valuable and could help strengthen a prospective consensus on a urinary threshold. To our knowledge, no study has examined the interindividual or intra-individual variability in the urine concentrations of salmeterol and its major metabolite α-hydroxysalmeterol. Another relevant issue is whether urine concentration measurements should be corrected for Urine Specific Gravity (USG), since WADA accredited laboratories not currently correct for USG. USG is a measurement for the concentration of the molecules in the urine and is directly proportional with the osmolarity of the urine . It is well-known that exercise affects body fluid balance in athletes  and that USG is a valid measurement of body fluid balance . However, little is known about the extent to which urine concentration measurements of beta2-agonists are affected by body fluid balance, but it is extremely relevant since the body fluid balance will be greatly affected by exercise, fluid and food intake. Despite the relevance, differences in the urine concentrations of salmeterol and its metabolite α-hydroxysalmeterol before and after correction for USG have never been investigated.
As such, the purpose of the current study is to determine the differences in the urine concentrations of salmeterol and its major metabolite α-hydroxysalmeterol before and after correction for USG after inhalation of 100 μg salmeterol in male subjects. Furthermore, it is a purpose of this study to measure the intra-individual and interindividual variability in the urine concentrations and urine excretion of salmeterol and α-hydroxysalmeterol.
We enrolled 20 non-smoking male subjects (10 with asthma and habitual users of beta2-agonists), aged 24.6 ± 3.9 years and 75.3 ± 6.3 kgbw. The participants were physically active less than six hours per week. Before the main study all participants were interviewed about their medical history of asthma. A methacholine challenge was performed to evaluate any diagnosis of asthma. All subjects were asked to discontinue any use of medication, including beta2-agonists, 14 days before each visit. Furthermore, the subjects were asked to refrain from alcohol and coffee 24 hours before each visit. Written informed consent was obtained from all subjects. The protocol was approved by the local Ethics Committee (No. H-C-2009-044). The study was conducted in accordance with the protocol and the guideline for Good Clinical Practice (GCP).
The participants reported to the lab in the morning and urine samples (T0) were collected. The study drug, 100 μg salmeterol (Serevent® Diskus 50 μg), was then administered by inhalation during supervision. Four (T4) and eight (T8) hours following the administration of the study drug further urine samples were collected. All urine samples were collected in 1 L bottles and measured for volume. Approximately 40 mL of the collected sample were stored in aliquots at -80°C for later analysis. An additional urine sample was obtained in the participants’ homes 12 hours (T12) after administration and frozen at -20°C and delivered the following day to the department and stored at -80°C. Furthermore, blood samples were drawn from the antecubital vein before and after the administration of the study drug for analysis of serum concentrations of salmeterol (data presented in Hostrup et al. ). To determine the intra-individual variability in the urine concentrations and urine excretion of salmeterol and α-hydroxysalmeterol, seven subjects came in for a second visit approximately 14 days after the first visit and underwent the same procedures again. The subjects had the same food and fluid intake as during the first visit and 24 hours before the visit, which were recorded .
Analysis of urine samples
The urine concentrations of salmeterol and its major metabolite, α-hydroxysalmeterol, were quantified by the WADA accredited Norwegian Doping Control Laboratory. The analysis of the urine concentrations of salmeterol and α-hydroxysalmeterol were performed by the method described by Hostrup et al.  and validated according to in-house procedures which are in agreement with WADAs regulations . A urine aliquot of 1 mL was diluted with an equal volume of water. Salmeterol-d3 (AH Diagnostics, Aarhus, Denmark) and ractopamine-d5 (RIVM, Bilthoven, the Netherlands) were selected as internal standards for the measurement of salmeterol and α-hydroxysalmeterol, respectively. α-hydroxysalmeterol reference material was kindly provided by GlaxoSmithKline (Brentford, UK) with assistance from WADA. The internal standard solution (25 μL, salmeterol-d3 and ractopamine-d5, each 0.1 μg•mL-1 in methanol) was added, and the samples were then subjected to enzymatic hydrolysis by β-glucuronidase. Instrumental analyses were performed by liquid chromatography tandem mass spectrometry (LC-MS/MS) on a Thermo Surveyor liquid chromatography system connected to a Thermo TSQ Quantum IonMax triple-quadrupole mass spectrometer (Thermo- Fisher Scientific, San Jose CA, USA). The Limits of Quantification (LOQ) for salmeterol and α-hydroxysalmeterol were 0.12 ngmL-1 and 0.14 ngmL-1, respectively . Acquisition parameters (ion cursors and collision energy) for the analysis of salmeterol and α-hydroxysalmeterol, as well as the intermediate precision and recovery in the urine samples are presented in table 1 by Hostrup et al. .
|Visit 1||Visit 2||Intra-individual variability|
|Urine volume (mL)||453±272||287±117||331±172||373±179||289±207||384±220||10.5|
|Concentration ofsalmeterol (ngmL-1)||0.36±0.19||0.15±0.09||0.10±0.07||0.41±0.28||0.14±0.07||0.08±0.03||43.0|
|Excretion ofsalmeterol (µg)||0.13±0.04||0.04±0.03||0.03±0.02||0.13±0.06||0.04±0.03||0.03±0.02||12.6|
|Concentration of a hydroxysalmeterol-(ngmL-1)||2.33±1.43||1.61±1.39||0.68±0.59||2.50±1.49||2.19±1.29||0.92±0.37||43.7|
|Excretion of α-hydroxysalmeterol(µg)||0.86±0.24||0.39±0.23||0.23±0.21||0.99±0.48||0.55±0.34||0.30±0.16||21.8|
Table 1: Urine excretions and urine concentrations of salmeterol and α-hydroxysalmeterol when uncorrected for USG in seven subjects during the first and second visit after inhalation of 100 μg salmeterol. T4: Four hours after administration, T8: Eight hours after administration, T12: Twelve hours after administration. Intra-individual variability is expressed as the mean intra-individual %CV between visits one and two.
The reported urinary concentrations represent the sum of the glucuronide conjugates (expressed as the free drug following hydrolysis) and free drug concentrations in accordance with the WADA technical document . To measure differences between uncorrected raw samples with corrected samples for USG, all obtained samples were corrected for USG using an adjustment equation in accordance with the WADA 2013 technical document:
Tadjusted = [(USGsample – 1) / (1.020 – 1)] × T
Where USGsample is the urine specific gravity of the sample, Tadjusted is the USG corrected urine concentration, and T is the raw urine concentration.
The urine excretions of salmeterol and α-hydroxysalmeterol were calculated as the raw urine concentration multiplied by the urine volume.
The statistical software program SPSS 19.0 (SPSS, Inc., Chicago, IL) was used. All data are expressed as mean ± SD. Normality was tested using the Shapiro-Wilk test. Pearson’s correlation was used to investigate relations between USG and uncorrected urine concentrations of salmeterol and α-hydroxysalmeterol, USG and corrected urine concentrations of salmeterol and α-hydroxysalmeterol, and bodyweight (kg) and urine excretion of salmeterol and α-hydroxysalmeterol. Differences in urine concentrations and coefficients of variations between uncorrected and corrected samples for USG were investigated with a paired t-test. Furthermore, differences in intra-individual urine concentrations were investigated with a paired t-test. A p-value<0.05 was considered statistically significant. Differences between USG corrected and uncorrected samples were calculated as the grand mean of the mean relative difference between each sample. The intra-individual variability was calculated by the coefficient of variation (CV%) within subjects between visit one and visit two: Standard deviation (SD)/ mean*100. The inter-individual variability was calculated as the CV% between subjects: SD/mean*100.
Urine concentrations of salmeterol and α-hydroxysalmeterol before and after correction for urine specific gravity
A total number of 108 urine samples were analyzed, 80 from 20 subjects during a single visit, and further 28 from seven of the 20 subjects that underwent a second visit, similar to the first. All samples collected at baseline (T0, n=27) contained no signs of salmeterol and α-hydroxysalmeterol. The mean relative differences in the urine concentrations of salmeterol and α-hydroxysalmeterol between USG corrected and uncorrected samples (T4, T8 and T12, n=81) were 43 ± 44, 27 ± 42 and 56 ± 87% at T4, T8 and T12, respectively. When corrected for USG 16, 13 and 14 samples had decreases in the urine concentrations at T4, T8 and T12, respectively. The absolute concentrations of salmeterol and α-hydroxysalmeterol of uncorrected and corrected samples for USG are shown in figure 1.
The average USG in all samples (n=108) was 1.020 ± 0.006 gmL-1, ranging from 1.004 to 1.031 gmL-1. There was a significant correlation between the USG and the uncorrected urine concentrations of both salmeterol (r=0.45, p<0.01) (Figure 2A) and α-hydroxysalmeterol (r=0.55, p<0.01) in the 81 samples collected 12 hours after administration of the study drug (Figure 2B). When the samples were corrected for USG, no correlation was observed between the USG and the urine concentrations of salmeterol (r=0.02, p=0.84), but a significant, yet lower, correlation between USG and urine concentrations of α-hydroxysalmeterol was still evident (r=0.26, p<0.05).
Inter-individual variability in the urine excretion and urine concentrations of salmeterol and α-hydroxysalmeterol
The inter-individual (n=20) variability expressed as the coefficient of variance (CV%) in the urine concentrations of salmeterol at T4, T8 and T12 were 55, 44 and 59% when uncorrected for USG and 39, 51 and 47% when corrected, respectively. The absolute concentrations were 0.36 ± 0.20, 0.17 ± 0.08 and 0.11 ± 0.07 ngmL-1 at T4, T8 and T12, when uncorrected and 0.37 ± 0.14, 0.18 ± 0.09 and 0.11 ± 0.05 ngmL-1, when corrected for USG. For α-hydroxysalmeterol, the inter-individual variability (CV%) was 81, 53 and 78% at T4, T8 and T12 when uncorrected and 54, 49 and 54% when corrected for USG, respectively. The absolute concentrations were 3.52 ± 2.86, 2.34 ± 1.24 and 1.48 ± 1.15 ngmL-1 when uncorrected and 3.36 ± 1.82, 2.33 ± 1.13 and 1.47 ± 0.80 ngmL-1 when corrected for USG at T4, T8 and T12, respectively. The total urine obtained during the 12 hours after administration of the study drug was 988 ± 381 mL giving a total excretion of 0.19 ± 0.06 μg salmeterol and 2.00 ± 0.79 μg α-hydroxysalmeterol. The excretions of salmeterol and α-hydroxysalmeterol were highest (p<0.01) at T4 with 0.11 ± 0.04 and 1.02 ± 0.48 μg, respectively. The inter-individual variability in the total urine excretions were 29% for salmeterol and 39% for α-hydroxysalmeterol. No correlations were observed between bodyweight (kg) and the urine excretion of salmeterol and α-hydroxysalmeterol.
Intra-individual variability in the urine concentrations and urine excretions of salmeterol and α-hydroxysalmeterol during visit one and visit two
No significant differences were observed in the USG or urine concentrations of salmeterol and α-hydroxysalmeterol in the subjects (n=7) between visits one and two. The USG at baseline (T0) was 1.022 ± 0.006 and 1.022 ± 0.005 gmL-1 at visits one and two. The USG at T4, T8 and T12 are presented in table 1. The USG corrected urine concentrations of salmeterol and α-hydroxysalmeterol were highest at T4 with no differences between visits one and two (Figure 3). The uncorrected urine concentrations of salmeterol and α-hydroxysalmeterol for USG are presented in table 1.
No significant differences were observed in the urine volume or urine excretion of salmeterol and α-hydroxysalmeterol between visits one and two. The total urine volume collected during the 12 hours after administration of salmeterol was 1071 ± 397 mL at visit one and 1046 ± 294 mL at visit two, corresponding to a total urine excretion of 0.21 ± 0.06 and 0.20 ± 0.07 μg salmeterol, respectively (Table 1). The total urine excretion of α-hydroxysalmeterol during the 12 hours after administration was 1.48 ± 0.28 μg at visit one and 1.83 ± 0.52 μg at visit two. The intra-individual variability in the urine excretion of salmeterol and α-hydroxysalmeterol calculated as the mean intra-individual CV% between the total excretions during visits one and two were 12.6 and 21.8%, respectively. If calculated from the urine concentrations, the intra-individual variability of salmeterol between visits one and two in all collected samples (n=42) paired for subject and sampling point was 43.0% when uncorrected for USG, which was significantly (p<0.01) higher than the 20.4% when corrected for USG. For α-hydroxysalmeterol, the variability was 43.7% when uncorrected and 28.0% when corrected for USG, which was significant lower (p<0.05) (Table 1).
Currently, WADA accredited laboratories do not correct for USG when analyzing urine samples for peptide substances such as beta2- agonists and salmeterol . The present study provides evidence that the intra-individual variability decreases significantly when urine samples of salmeterol are corrected for USG, and that the USG is significantly correlated with urine concentrations of both salmeterol and α-hydroxysalmeterol. The latter observation is in agreement with findings after inhalation of the short-acting beta2-agonist, salbutamol, in which positive correlations between the USG and the urine concentrations was observed one hour after administration . It is noteworthy that the correlation between the USG and the urine concentration of salmeterol was abolished in the present study when samples were corrected for USG.
From a doping perspective, the importance of correction for USG in urine samples has been illustrated in several recent studies conducted after inhalation of other beta2-agonists. As such, inhalation of only 800 μg salbutamol , which is half of the maximally allowed daily dose on the prohibited list, was shown to exceed the current urinary threshold of 1,000 ngmL-1 in one uncorrected sample for USG with 1,057 ngmL-1 .When corrected, the concentration fell to 661 ngmL-1, corresponding to a USG of 1.032 gmL-1. Similar observations were seen in another study, in which repetitive inhalation of salbutamol  to a total dose of 1,600 μg resulted in one sample exceeding the current threshold with 1,082 ngmL-1 when uncorrected for USG . The USG in the sample was 1.029 gmL-1, and when corrected to a USG of 1.020 gmL-1, the sample concentration of salbutamol was only 746 ngmL-1. Lastly, a single sample of salbutamol was found to be 562 ngmL-1 when left uncorrected for USG and 1,387 ngmL-1 when corrected, well above the current threshold. Similar large differences have been reported in the urine concentrations after inhalation of terbutaline and formoterol between uncorrected and corrected samples for USG [8,18]. Correction for USG may therefore have an impact when evaluating doping cases.
At the moment, corrections for USG are only adjusted in urine samples exceeding a USG of 1.020 gmL-1 in the presence of anabolic steroids . The recent findings in numerous studies concerning differences in USG corrected and uncorrected urine concentrations of beta2-agonists raise some concerns. The present study, along with prior ones, shows that the urine concentrations of beta2-agonists are partially related with the USG . This is a problem in two ways. Firstly, several samples may be concentrated and well above a USG of 1.020 gmL-1, which, in most cases result in a high urine concentration. Dehydration is common in several sports, and the consequence in an elevated USG. Secondly, as observed in the study by Sporer et al.,samples can go unnoticed through an adverse analytical finding if the urine is diluted . Correction for USG seems to be a reasonable way to adjust concentrated or diluted samples. Still, the current regulations allow athletes to prove that any abnormal result is a consequence of therapeutic use in a controlled pharmacokinetic study.
We observed high intra-individual and inter-individual variability in the urine concentrations of salmeterol and α-hydroxysalmeterol between male subjects, even when the samples were corrected for USG. The high inter-individual variability observed is consistent with previous studies conducted after inhalation of beta2-agonists [6-8]. However, it is notable that such high intra-individual differences were observed in the present study. When calculating the intra-individual variability in the urine concentrations of salmeterol and α-hydroxysalmeterol differences as high as 20.4% and 28.0% were observed, even when samples were corrected for USG. However, when calculated from the total excretions of salmeterol and α-hydroxysalmeterol during the 12 hours after administration of salmeterol the variations were only 12.6% and 21.8%, and similar to what has been shown for salbutamol. When the urine excretion was calculated for salbutamol an intra-individual variability as low as ~10% has been observed, following inhalation of 100 and 300 μg salbutamol . The high differences observed may arise from biological as well as technical variability. Biological variability such as differences in anthropometrics, basal metabolic rate and heat production along with differences in drug deposition, absorption, clearance and renal clearance will all affect the concentration in the urine. Technical variability includes inhalation technique of the study drug, fluid and food intake, as well as the analytical variability.
It is difficult to analyze salmeterol in biological fluids due to the very low concentrations. Although, the LOQ from the method applied in this study was 0.12 ngmL-1 and more sensitive than prior studies , the urine concentrations of salmeterol were below the LOQ in many cases 8 and 12 hours after administration, resulting in a higher variability. For the purpose of doping analysis, however, the LOQ for salmeterol does not seem to be a problem. If a threshold were to be introduced on the prohibited list, it would be expected to be high enough to ensure no or little risk of false positive tests after therapeutic use of inhaled salmeterol. As such, a urinary threshold would be well above the LOQ in the analytical method applied, and the current method could therefore easily be adapted for routine doping control. The significantly higher concentrations of the metabolite, α-hydroxysalmeterol, along with its relatively low LOQ, allowed detection of all obtained samples (n=81) after inhalation of 100 μg salmeterol, even 12 hours after administration. Thus, α-hydroxysalmeterol may be a suitable marker for excessive misuse of salmeterol in competitive sports, which has also been suggested by others . It is reasonable to assume that α-hydroxysalmeterol can be detected for more than 48 hours in most cases, even after inhalation of low doses of salmeterol.
The urine recovery of salmeterol and α-hydroxysalmeterol were very low during the 12 hours of sampling. As such, only 0.2 μg was recovered in the urine, corresponding to less than 1% of the administered dose, which is in agreement with recent findings . In a study by Deventer et al., the urine excretion of salmeterol during the first 12 hours was between 0.09 and 0.30 μg, corresponding to 0.1% and 0.27% of the administered dose of 100 μg inhaled salmeterol . The small amount of recovery in the urine is related to the pharmacokinetic properties of salmeterol. The long half-life of approximately 5.5 hours, along with a protein binding of 94-98% [1,20], gives a low glomerular filtration in the nephrons. Consistently, it has been observed that it may take as long as 24 and 72 hours before most of the administered dose is recovered, and that the excreted amount of unchanged salmeterol in the urine accounts for less than 5% of the administered dose . In the present study, the urine excretion of the major metabolite, α-hydroxysalmeterol, on the other hand, was more than 10 times higher during the 12 hours after administration.
The recently released prohibited list effective as of 2013 only has urinary thresholds for salbutamol and formoterol, whereas there is still no limit for salmeterol. Consequently, at the moment, athletes have the opportunity to inhale unlimited doses of salmeterol and potentially increase their performance during competition. However, if WADA decides to keep the current regulations for inhaled beta2- agonists, it would be expected that a urinary threshold for salmeterol will be introduced at some point in the coming years. If a threshold should be applied for salmeterol, it should be high enough to ensure no or small risk of false positive doping tests. Furthermore, a specific therapeutic threshold should be applied in conjunction with the urinary threshold. At the moment, the 2013 prohibited list only states that the inhalation of salmeterol is allowed in therapeutic doses according to the manufactures recommended therapeutic regime .
The introduction of therapeutic and urinary thresholds for some beta2-agonists brings several new questions to the doping debate. On one hand, the new regulations lessen the amount of administrative work and financial expenses from the former therapeutic use exemption, which required athletes to get a dispensation to use beta2-agonists. On the other hand, this may result in misuse by non-asthmatic athletes and misdiagnosis of asthma in other athletes . Lastly, inhaled beta2- agonists may be performance-enhancing when taken in the maximally allowed doses on the prohibited list. To our knowledge, no studies have investigated the effects of 1,600 μg inhaled salbutamol on performance in athletes. In any case, WADA should continue to evaluate the current thresholds for inhaled beta2-agonists, and if they decide to maintain the current regulations, we would recommend introducing a urinary threshold for salmeterol. Nevertheless, more studies are needed investigating the urine concentrations after inhalation of higher doses and during sports exertion.