Cortisol in hair, body mass index and stress-related measures

Abstract

Hair cortisol concentrations (HCC) are assumed to reflect integrated cortisol secretion over extended periods of time and may provide a sensitive marker for stress-associated endocrine changes. Here, we report data from two independent studies of 155 (study I) and 58 participants (study II) in which HCC associations with different stress-related measures and body mass index (BMI) were investigated. Consistent evidence for positive associations between HCC and BMI was seen across both studies (study I: r = .33, p < .001; study II: r = .42, p = .001). On the other hand, findings failed to reveal reliable HCC associations with psychosocial variables, showing only a positive relationship with self-reported social overload in study II (r = .29, p = .03) but not with other stress-related measures.

Introduction

Long-term alterations in the secretion of the glucocorticoid hormone cortisol via the hypothalamus–pituitary–adrenal (HPA) axis are assumed to play a crucial role in mediating the link between chronic stress exposure and vulnerability to a variety of diseases (see Chrousos and Kino, 2007, Chrousos, 2009). However, the valid assessment of long-term cortisol secretion has been challenging using previous measurement strategies reflecting momentary cortisol levels (blood, saliva) or cortisol secretion over relatively short time periods (urine). In this context, evidence suggesting that endogenous cortisol concentrations can be measured in hair (Koren et al., 2002, Raul et al., 2004), providing a retrospective index of integrated cortisol secretion over periods of several months, may present a major methodological advancement (see Gow et al., 2010). The validity of hair cortisol concentrations (HCC) as a measure of long-term cortisol secretion has now been supported by a range of studies, both in animals (Davenport et al., 2006, Accorsi et al., 2008, Fairbanks et al., 2011) and in human participants (Sauve et al., 2007, Kirschbaum et al., 2009, Thomson et al., 2009, Stalder et al., 2010, D’Anna-Hernandez et al., 2011, Manenschijn et al., 2011, Skoluda et al., 2012). In addition, recent evidence suggesting a high level of intraindividual stability in HCC (Stalder et al., 2012) and a relative robustness of HCC to a range of potential sociodemographic and hair-related confounding influences (Dettenborn et al., in press-b) further speak for the ease and usefulness of employing HCC as a measure in psychobiological research.

Building on these enquiries into more fundamental aspects of HCC, research over the past years has begun to investigate associations between HCC and stress-related conditions. In non-human primates, HCC have been shown to increase with the stress of relocation or move to a high stress environment (Davenport et al., 2006, Fairbanks et al., 2011, respectively) with this long-term response in HCC being moderated by the animals’ early rearing experience (Dettmer et al., 2012). Research in humans has predominantly focused on examining groups known to experience increased levels of psychosocial stress: elevated HCC have been reported in hospitalised and ventilated neonates (Yamada et al., 2007), chronic pain patients (Van Uum et al., 2008) and long-term unemployed individuals (Dettenborn et al., 2010). In line with this, in a sample of pregnant women, HCC were found to be positively associated with self-reported perceived stress (Kalra et al., 2007), a finding which, however, was not confirmed in another study on pregnant women (Kramer et al., 2009). Similarly, no associations between HCC and measures of perceived stress were seen across different clinical groups and respective control groups (Van Uum et al., 2008, Dettenborn et al., 2010, Stalder et al., 2010) or with regard to intraindividual associations with HCC (Stalder et al., 2012). Finally, altered HCC have also been associated with stress-related psychiatric diseases (Steudte et al., 2011a, Steudte et al., 2011b, Dettenborn et al., in press-a) and the stress-associated risk of suffering a myocardial infarction (Pereg et al., 2011).

Besides stress-related HCC enquiries, another interesting and potentially related recent observation comes from two reports showing positive associations between HCC and body fat associated measures: Manenschijn et al. (2011) obtained measurements of participants’ body mass index (BMI) and waist-to-hip ratio (WHR) in healthy participants and found positive HCC associations with WHR but not with BMI (Manenschijn et al., 2011). However, a later report by these authors also revealed positive HCC associations with BMI in a sample of shift and day time workers (Manenschijn et al., in press). These hair cortisol findings add to previous evidence showing complex patterns of HPA axis dysregulation in overweight and abdominally obese subjects. Specifically, increased adrenal cortisol production and secretion, acute hyperresponsivity and elevated 24-h urinary free cortisol (UFC) concentrations have often been found in obesity alongside normal or even decreased cortisol levels in blood serum or saliva (see Pasquali et al., 2006, Bose et al., 2009, Müssig et al., 2010). The corroboration of HCC associations with body fat-related measures is particularly of interest in terms of a potential three-way association with psychosocial stress; e.g. body fat or obesity-related measures have been found to be associated with work stress (Kouvonen et al., 2005, Kivimaki et al., 2006) or psychological well-being (Wardle and Cooke, 2005).

Based on the above literature, the current investigation utilises data from two independent studies on normal healthy individuals to examine two main topic areas: first, as previous research on associations between psychosocial stress and HCC has largely been carried out in clinical samples, we seek to examine HCC relationships with perceived chronic stress across the current samples of healthy individuals. To further enhance the information provided here, in study I we also investigate HCC associations with other stress-related variables reported to be associated with altered cortisol secretion patterns, such as perceived social support (see Uchino, 2006), work-related exhaustion (see Melamed et al., 2006) and general self-efficacy (e.g. Schwerdtfeger et al., 2008). As a second goal, we aim to corroborate previous evidence of HCC associations with BMI and to examine the possibility of a three-way relationship of these variables with psychosocial stress-related measures.