ORIGINAL ARTICLE

The Detection of Cortisol in Human Sweat: Implications for Measurement of Cortisol in Hair Evan Russell, MSc,* Gideon Koren, MD,*†‡§ Michael Rieder, MD, PhD,†¶ and Stan H. M. Van Uum, MD, PhD†

Background: Hair cortisol analysis has been shown to be an effective measure of chronic stress. Cortisol is assumed to incorporate into hair via serum, sebum, and sweat sources; however, the extent to which sweat contributes to hair cortisol content is unknown.

Methods: Sweat and saliva samples were collected from 17 subjects after a period of intensive exercise and analyzed by salivary enzyme-linked immunosorbent assay (ELISA). Subsequently, an in vitro test on exposure of hair to hydrocortisone was conducted. Residual hair samples were immersed in a 50-ng/mL hydrocortisone solution for periods lasting 15 minutes to 24 hours, followed by a wash or no-wash condition. Hair cortisol content was determined using our modified protocol for a salivary ELISA.

Results: Postexercise control sweat cortisol concentrations ranged from 8.16 to 141.7 ng/mL and correlated significantly with the logtransformed time of day. Sweat cortisol levels significantly correlated with salivary cortisol concentrations. In vitro hair exposure to a 50-ng/mL hydrocortisone solution (mimicking sweat) for 60 minutes or more resulted in significantly increased hair cortisol concentrations. Washing with isopropanol did not affect immersion-increased hair cortisol concentrations. Conclusions: Human sweat contains cortisol in concentrations comparable with salivary cortisol levels. This study suggests that perfuse sweating after intense exercise may increase cortisol concentrations detected in hair. This increase likely cannot be effectively decreased with conventional washing procedures and should be considered carefully in studies using hair cortisol as a biomarker of chronic stress. Key Words: sweat, cortisol, hair, stress, HPA (Ther Drug Monit 2014;36:30–34)

Received for publication September 24, 2012; accepted May 27, 2013. From the *Department of Physiology and Pharmacology, †Department of Medicine, Schulich School of Medicine and Dentistry, and ‡Ivey Chair in Molecular Toxicology, University of Western Ontario, London, Canada; §Department of Clinical Pharmacology/Toxicology, Hospital for Sick Children, Toronto, Ontario, Canada; and ¶CIHR-GSK Chair in Pediatric Clinical Pharmacology, Children’s Health Research Institute, London, Ontario, Canada. The authors declare no conflict of interest. Correspondence: Stan H. M. Van Uum, MD, PhD, St Joseph’s Health Care, Room B5-120, 268 Grosvenor St, London, Ontario N6A 4V2, Canada (e-mail: [email protected]). Copyright © 2014 by Lippincott Williams & Wilkins

30

INTRODUCTION Cortisol is a glucocorticoid hormone released from the hypothalamo-pituitary-adrenal axis in response to stress.1 Traditionally, it has been quantified in serum, urine, and saliva. These matrices are useful for assessing acute stress, but the intra- and interday fluctuations in cortisol secretion make these matrices less suitable for measuring chronic stress. Hair analysis has been used for decades to monitor chronic exposure to exogenous compounds, with particular emphasis on detecting drugs of abuse.2 Recently, hair analysis has shown to be an effective measure of long-term exogenous and endogenous cortisol exposure.3,4 Because hair grows at a fairly predictable rate of 1 cm/mo, month-by-month changes in cortisol exposure can be identified.5 This ability to monitor cortisol exposure over months at a time makes hair analysis an excellent matrix for monitoring chronic stress. Our laboratory has applied this tool clinically to monitor cortisol concentrations in patients with the Cushing syndrome,6 severe chronic pain,7 stress related to risk for myocardial infarction,8 and to monitor glucocorticoid replacement therapy in patients with adrenal insufficiency.9 Hair cortisol analysis is also useful in identifying psychosocial stresses including general anxiety disorder,10 unemployment,11 shift work,12 posttraumatic stress disorder,13 and bipolar disorder.14 Although hair cortisol analysis may be a useful biomarker of long-term cortisol exposure, some methodological questions remain. One such question relates to the exact mechanism(s) by which cortisol is incorporated into hair. The paradigm for cortisol incorporation into hair assumes that, as hair grows, cortisol enters the medulla of the hair shaft via blood. Therefore, each centimeter of hair reflects an integral of a month’s worth of serum cortisol concentrations. It has also been postulated that sebaceous and eccrine secretions may contribute to hair cortisol concentrations, but supporting evidence is lacking.15,16 It is presently unknown what cortisol concentrations are found in human sweat. Because individuals have various levels of perspiration, personal hygiene, and hair washing frequencies, it would be important to assess to what extent sweat affects hair cortisol concentrations. The objective of this study was to determine if cortisol is present in human sweat and to characterize its concentrations relative to salivary cortisol levels, an established point measure approximation of free plasma cortisol concentrations. Furthermore, this study sought to examine whether hair cortisol concentrations are altered when hair is exposed to sweat-like cortisol concentrations and if this effect can be mitigated with isopropanol washes. Ther Drug Monit  Volume 36, Number 1, February 2014

Ther Drug Monit  Volume 36, Number 1, February 2014

SUBJECTS AND METHODS Ethical Considerations and Subjects Recruitment The study protocol was approved by the University of Western Ontario Health Sciences Research Ethics Board. Each subject gave written informed consent. Subjects received no monetary compensation for their participation. Healthy athletic subjects 18 years and older were recruited. Subjects were excluded if they used any form of glucocorticoids, had the Cushing syndrome, or if they were pregnant. Age, smoking status, alcohol consumption, height, and weight were obtained.

Sample Collection and Analysis Saliva and sweat samples were collected immediately after 15–60 minutes of vigorous exercise. Vigorous exercise entailed either rowing on an ergometer or running for a minimum of 10 minutes. Saliva and sweat samples were collected immediately after the period of exercise. Samples were collected in the morning, afternoon, and evening to examine whether sweat cortisol concentrations approximated the wellestablished circadian cortisol rhythm as measured in saliva and serum. Both saliva and sweat samples were collected using Alpco Diagnostics Saliv-Saver salivettes (Alpco Diagnostics, Salem, NH). Saliva was collected by placing a cotton swab under the tongue, allowing it to saturate, and then inserting it into a 5-mL salivette. For sweat collection, a cotton swab was rubbed over the scalp hair and neck, allowed to saturate, and placed it in a 5-mL salivette. Salivettes were centrifuged at 2218g for 5 minutes and 1 mL of the supernatant was pipetted into a 1.5-mL Eppendorf tube. Samples were stored at 2208C until analysis. Of note, iontophoresis not used for sweat collection because the sweating due to exercise was considered to be more representative of normal physiological sweating. Sweat and saliva samples were analyzed with a salivary enzyme-linked immunosorbent assay (ELISA) (Alpco Diagnostics). The intraassay and interday coefficients of variation were determined to be 4.13% and 10.31%, respectively. The manufacturer of the ELISA reports a lower limit of detection and lower limit of quantification to be 1.0 and 4.0 ng/mL, respectively.

Hydrocortisone Immersion and Hair Washing Experiments Preparation of a Sweat-Mimicking Hydrocortisone Solution There are a variety of protocols for the preparations of artificial sweat.17 Preparations vary in their pH, sodium chloride concentrations, and the presence or absence of phosphates, lactic acid, urea, and L-histidine monochloride. In choosing a solution to approximate sweat, phosphate buffered saline (PBS) was selected because its pH, sodium chloride concentration, and phosphate composition are similar to those of the various artificial sweat preparations. An 18.12-mg/mL hydrocortisone standard (Sigma-Aldrich, Oakville, ON, Canada) was dissolved in PBS to a concentration of 50 ng/mL (confirmed via salivary ELISA). This concentration was consistent Ó 2014 Lippincott Williams & Wilkins

Detection of Cortisol in Human Sweat

with the low to middle range of sweat cortisol concentrations observed in healthy controls.

Isopropanol Wash Isopropranol is the standard wash solvent when preparing hair for cortisol analysis.18 Isopropanol is thought to be the most effective solvent for removing glucocorticoid contaminants while maintaining the endogenous cortisol content found inside the medulla of the hair shaft. To wash the hair, the hair segments are placed inside a scintillation vial, 3 mL of isopropanol is added, and the vials are then slowly rotated at 0.28 g for 3 minutes. The isopropanol is then decanted, and the wash is repeated once. Hair is left to dry for at least 5 hours under a fume hood.

Proof of Concept: Prolonged Hydrocortisone Immersion To determine whether immersion in a hydrocortisone solution had any effect on hair cortisol concentration, residual hair samples (n = 14) from past studies were used. All hair samples had been collected from the vertex posterior of the scalp as per our standard protocol, and these subjects had not been sweating before hair collection. The most proximal 3 cm of hair of each sample was segmented and divided into 5 subgroups. Each subgroup was then subjected to 1 of 5 conditions. The control condition consisted of a standard isopropanol wash. The 4 treatment conditions included a 12-hour hydrocortisone immersion, a 24-hour hydrocortisone immersion, a 12-hour hydrocortisone immersion with an isopropanol wash, and a 24-hour hydrocortisone immersion with an isopropanol wash. Of note, before the isopropanol washes, hair that had been immersed in the hydrocortisone solution was allowed to dry at room temperature for 12 hours on paper towel. Hair cortisol analysis was performed in accordance with the protocol established in our laboratory.4 Briefly, 10–15 mg of each hair sample was placed in a scintillation vial, 1 mL of methanol was added, and hair segments were minced with surgical scissors until granular in appearance. The vials were sealed and incubated at 508C for 16 hours and gently rotated. The cortisol-containing methanol solution was pipetted into a 5-mL test tube, placed on a hot plate at 508C, and evaporated under a stream of nitrogen gas. The remaining residue was reconstituted with 250 mL of PBS. The reconstituted samples were analyzed with the Alpco Diagnostics salivary ELISA.

Physiologically Relevant Hydrocortisone Immersions To assess the effect of physiologically relevant durations of immersion on hair cortisol content, we used residual hair samples (n = 7) from past studies. The most proximal 3 cm of hair were segmented and divided into 5 subgroups. The control condition consisted of a brief 2-second immersion in the hydrocortisone solution followed by an isopropanol wash. The 4 treatment conditions consisted of 15, 30, 60, and 120-minute hydrocortisone immersions followed by an isopropanol wash. Of note, after each hydrocortisone immersion, the hair segments were dried at room temperature for 12 hours on paper towel before being washed with

31

Ther Drug Monit  Volume 36, Number 1, February 2014

Russell et al

isopropanol. Hair cortisol analysis was then performed as described above.

Statistical Analysis

Results are presented as mean 6 SD unless indicated otherwise. Data were assessed for normality using Kolmogorov–Smirnov test; non-normally distributed data were log transformed before statistical analysis. Correlation coefficients between different parameters were calculated with the Pearson correlation coefficient for normally distributed data. A repeated measures analysis of variance with a posthoc Bonferroni test was used to examine if differences existed among the control and treatment conditions for the hydrocortisone immersion experiments. GraphPad Prism version 4.0b was used for all statistical analyses (GraphPad Software Inc, La Jolla, CA).

FIGURE 1. Effect of prolonged hair immersion for 12 or 24 hours in a 50-ng/mL hydrocortisone PBS solution (mimicking sweat-containing cortisol) and washing on hair cortisol concentrations. W: preanalysis isopropanol wash. *P , 0.001 compared with control.

RESULTS We recruited 17 subjects; their characteristics are displayed in Table 1. All subjects had done intensive exercise making them sweat profusely. Sweat cortisol concentrations were 74.62 6 41.51 ng/mL, ranging from 8.16–141.7 ng/mL, with the highest concentrations in the morning samples. The concentration of sweat cortisol determined was dependent on the time of day collected, with the highest concentrations being found in the morning samples and lower concentrations in evening samples. Sweat cortisol concentrations were significantly correlated with the time of day (r2 = 20.44, P , 0.01). Sweat cortisol concentrations were significantly correlated with salivary cortisol concentrations (r2 = 0.30, P , 0.05). Twelve- and 24-hour hair exposure to the cortisolcontaining solution resulted in increased hair cortisol content compared with control (P , 0.001) and were not affected by isopropanol washes (Fig. 1). The effect of duration of immersion to the cortisolcontaining solution is shown in Figure 2. Incubation for 1 hour or more resulted in significantly increased hair cortisol content (P , 0.001; analysis of variance). Posthoc analysis identified that hair cortisol concentrations for hair samples that had incubated for 60 and 120 minutes were significantly increased from the control concentration (P , 0.01 and P , 0.001, respectively). Furthermore, hair samples incubated for 120 minutes had significantly increased cortisol relative to those that had only incubated for 15 and 30 minutes (P , 0.001).

DISCUSSION

To our knowledge, this is the first study that investigated cortisol in sweat. The presence of a quantifiable amount of cortisol in sweat supports the hypothesis that sweat contributes to hair cortisol content. Similar to salivary cortisol concentrations, which represent the free unbound fraction of cortisol in the blood,19 it is very likely that sweat cortisol concentrations also represent free cortisol. Indeed, the significant correlation between cortisol in sweat and saliva suggests that sweat cortisol, like salivary cortisol, may reflect acute hypothalamo-pituitary-adrenal activity.20 The hydrocortisone immersion experiments underscore the need for studies using hair cortisol analysis to record personal hygiene habits (eg, hair washing frequency and frequency of extensive sweating). Our results indicate that the hair cortisol content may increase significantly after exposure to sweat-containing cortisol. Importantly, the standard isopropanol wash procedure used to remove external contaminants proved ineffective in removing cortisol that was

TABLE 1. Subject Characteristics (n = 17) Age 6 SD Gender (M:F) BMI 6 SD Smokers Alcohol drinks per day, median (range) Subjects taking prescription drugs Time of sampling, median (range) BMI, body mass index.

32

25 6 9.3 12 (71%):5 (29%) 24.2 6 2.94 — 0 (023) — 10:40 h (07:40–22:00 h)

FIGURE 2. Effect of immersion time on hair cortisol concentrations. Hair samples were immersed a 50-ng/mL hydrocortisone PBS solution for 15–120 minutes. *P , 0.01, **P , 0.001, for hair cortisol compared with control. Ó 2014 Lippincott Williams & Wilkins

Ther Drug Monit  Volume 36, Number 1, February 2014

absorbed from the sweat-mimicking hydrocortisone solution. This suggests that variation in sweating and hair hygiene could be a significant confounder and that profuse sweating at the time of hair collection should be carefully noted. These results contrast a study in which repeatedly exposing hair samples to shampoo washes decreased hair cortisol concentrations.21 A possible reason for this contrast is that surfactants are potentially more effective than alcohol washes at removing sweat and sebum from hair. Although shampoo is unlikely to be an appropriate laboratory wash solvent, perhaps exploring more conventional wash solvents such as dichloromethane would be useful. Future work should test dichloromethane ability to remove sweat-deposited cortisol and exogenous glucocorticoid contamination while maintaining cortisol found in the medulla of the hair shaft. When hair is exposed to hydrocortisone for shorter periods of time, more in line with real-life durations of hair exposure to sweat, the relationship between exposure time and hair cortisol content becomes apparent. Exposure of hair to the hydrocortisone solution for 60 minutes resulted in increased hair cortisol concentrations compared with shorter exposure. Thus, if hair samples were collected when a subject had been sweating for 60 minutes, hair cortisol concentration could be increased compared with if the individual had not been sweating. A previous article found increased hair cortisol concentrations in athletes compared with control subjects. In light of the current findings, it is conceivable that some of this increase in hair cortisol content may be due to prolonged sweating.22 A limitation to this study was that only perspiration induced through intensive exercise was examined. Cortisol concentrations increase during periods of physical stress such as exercise23; therefore, it is possible that the sweat cortisol levels were temporarily elevated in our treatment conditions. However, increased plasma cortisol concentrations have been noted in cases of passive hyperthermia.24 Furthermore, in conditions such as acute myocardial infarctions or fever, where diaphoresis is a common symptom, elevated serum cortisol concentrations are also noted.25 Thus, most conditions that induce sweating are likely associated with elevated cortisol concentrations. It would have also been preferable to examine interday variations in sweat cortisol concentrations to try to establish its variability over time. Additionally, due to the nature of how sweat was collected, it is possible that this would have also collected some sebum, and the sweat cortisol concentration may have also represented some sebum cortisol. However, given that sebum is much more viscous than sweat, it is unlikely that it was a major contributing source to the cortisol measures because it would not have been so easily centrifuged from the cotton swab.

CONCLUSIONS Previously, our group and others have shown that the patterns of change in hair cortisol, when segmented to reflect month-by-month exposure, closely follow the changes measure in blood and saliva. This was commonly interpreted that most of the hair load of cortisol enters the shaft from the Ó 2014 Lippincott Williams & Wilkins

Detection of Cortisol in Human Sweat

blood. Our observation, that sweat contains cortisol in levels reflecting systemic concentrations and that sweat cortisol is not easily washable from the hair shaft, indicate that at least some of the measured cortisol in hair may stem from sweat. Because sweat cortisol significantly correlates with systemic levels of the hormone, sweat-derived cortisol is not likely to disrupt the correlation between hair cortisol and chronic stress. ACKNOWLEDGMENT We thank all athletes who participated.

REFERENCES 1. McEwen BS. Stress, adaptation, and disease. Allostasis and allostatic load. Ann N Y Acad Sci. 1998;840:33–44. 2. Gaillard Y, Vayssette F, Balland A, et al. Gas chromatographic-tandem mass spectrometric determination of anabolic steroids and their esters in hair. Application in doping control and meat quality control. J Chromatogr B Biomed Sci Appl. 1999;735:189–205. 3. Raul JS, Cirimele V, Ludes B, et al. Detection of physiological concentrations of cortisol and cortisone in human hair. Clin Biochem. 2004;37: 1105–1111. 4. Sauve B, Koren G, Walsh G, et al. Measurement of cortisol in human hair as a biomarker of systemic exposure. Clin Invest Med. 2007;30: E183–E191. 5. Wennig R. Potential problems with the interpretation of hair analysis results. Forensic Sci Int. 2000;107:5–12. 6. Thomson S, Koren G, Fraser LA, et al. Hair analysis provides a historical record of cortisol levels in Cushing’s syndrome. Exp Clin Endocrinol Diabetes. 2010;118:133–138. 7. Van Uum SH, Sauve B, Fraser LA, et al. Elevated content of cortisol in hair of patients with severe chronic pain: a novel biomarker for stress. Stress. 2008;11:483–488. 8. Pereg D, Gow R, Mosseri M, et al. Hair cortisol and the risk for acute myocardial infarction in adult men. Stress. 2011;14:73–81. 9. Gow R, Koren G, Rieder M, et al. Hair cortisol content in patients with adrenal insufficiency on hydrocortisone replacement therapy. Clin Endocrinol (Oxf). 2011;74:687–693. 10. Steudte S, Stalder T, Dettenborn L, et al. Decreased hair cortisol concentrations in generalised anxiety disorder. Psychiatry Res. 2011;186: 310–314. 11. Dettenborn L, Tietze A, Bruckner F, et al. Higher cortisol content in hair among long-term unemployed individuals compared to controls. Psychoneuroendocrinology. 2010;35:1404–1409. 12. Manenschijn L, van Kruysbergen RG, de Jong FH, et al. Shift work at young age is associated with elevated long-term cortisol levels and body mass index. J Clin Endocrinol Metab. 2011;96: E1862–E1865. 13. Steudte S, Kolassa IT, Stalder T, et al. Increased cortisol concentrations in hair of severely traumatized Ugandan individuals with PTSD. Psychoneuroendocrinology. 2011;36:1193–1200. 14. Manenschijn L, Spijker AT, Koper JW, et al. Long-term cortisol in bipolar disorder: associations with age of onset and psychiatric comorbidity. Psychoneuroendocrinology. 2012;37:1960–1968. 15. Pragst F, Balikova MA. State of the art in hair analysis for detection of drug and alcohol abuse. Clin Chim Acta. 2006;370:17–49. 16. Russell E, Koren K, Rieder M, et al. Hair cortisol as a biological marker of chronic stress: current status, future directions, and unanswered questions. Psychoneuroendocrinology. 2012;37:589–601. 17. Kulthong K, Srisung S, Boonpavanitchakul K, et al. Determination of silver nanoparticle release from antibacterial fabrics into artificial sweat. Part Fibre Toxicol. 2010. Available at:doi: 10.1186/1743-8977-7-8. 18. Davenport MD, Tiefenbacher S, Lutz CK, et al. Analysis of endogenous cortisol concentrations in the hair of rhesus macaques. Gen Comp Endocrinol. 2006;147:255–261. 19. Jessop DS, Turner-Cobb JM. Measurement and meaning of salivary cortisol: a focus on health and disease in children. Stress. 2008;11:1–14.

33

Russell et al

20. Kudielka BM, Hellhammer DH, Wust S. Why do we respond so differently? Reviewing determinants of human salivary cortisol responses to challenge. Psychoneuroendocrinology. 2009;34:2–18. 21. Hamel AF, Meyer JS, Henchey E, et al. Effects of shampoo and water washing on hair cortisol concentrations. Clin Chim Acta. 2011;412: 382–385. 22. Skoluda N, Dettenborn L, Stalder T, et al. Elevated hair cortisol concentrations in endurance athletes. Psychoneuroendocrinology. 2012;37: 611–617.

34

Ther Drug Monit  Volume 36, Number 1, February 2014

23. Shojaei EA, Farajov A, Jafari A. Effect of moderate aerobic cycling on some systemic inflammatory markers in healthy active collegiate men. Int J Gen Med. 2011;4:79–84. 24. Jimenez C, Melin B, Savourey G, et al. Effects of passive hyperthermia versus exercise-induced hyperthermia on immune responses: hormonal implications. Eur Cytokine Netw. 2007;18:154–161. 25. Vallance BD, Hume R, Weyers E. Reassessment of changes in leucocyte and serum ascorbic acid after acute myocardial infarction. Br Heart J. 1978;40:64–68.

Ó 2014 Lippincott Williams & Wilkins

The detection of cortisol in human sweat: implications for measurement of cortisol in hair.

Hair cortisol analysis has been shown to be an effective measure of chronic stress. Cortisol is assumed to incorporate into hair via serum, sebum, and...
113KB Sizes 0 Downloads 0 Views