Hair has long been used in toxicology, forensic science, doping control and other fields as a biological specimen for the detection of environmental agents, drugs, or toxins. Most recent evidence suggests that also hormones are incorporated and trapped inside the growing hair. This has led to the hypothesis that cortisol measurement of distinct hair segments could provide a retrospective calendar of cortisol production for the individual. In this first proof-of-concept study in humans, we analyzed cortisol in hair donated by mothers with a neonate child (n-Mothers; N=103), mothers with toddlers 3-9 months of age (t-Mothers; N=19), and control women (N=20). We cut hair strands from each women into at least three 3-cm segments, which, based on an average hair growth rate of 1cm per month, would represent hair grown over the past three, six, and nine months, respectively. Since in the third trimester of pregnancy there is a well-documented increased production of cortisol, we expected to see elevated levels of cortisol in the most proximal hair segment of women who had just given birth to a child (n-Mothers) compared with the control women. Likewise, we expected to see elevated levels in the second, third, or fourth segment of mothers of 3-month olds, 6-months olds, and 9-months olds, respectively. These hair segments, cut at 4-12 cm from the scalp, would represent hair grown throughout the third trimester of pregnancy. Results showed that there was a strong monotonic decline in cortisol concentration from the segment closest to the scalp to the most distal hair segment (p<0.0001). Cortisol levels decreased by 30-40% from one segment to the next for the most recent four hair segments. Segments from hair older than one year had similarly, low levels of cortisol. Comparisons of cortisol levels in hair between n-Mothers and control women yielded the expected results: cortisol levels in the first 3-cm hair segment (i.e., closest to the scalp) of n-Mothers were two-fold higher than in controls (p<0.0001), probably reflecting increased cortisol levels throughout the third trimester of pregnancy. No differences in cortisol content were apparent for the second or third 3-cm segments in n-Mothers (p>0.2). When hair from mothers with 6-9 months old toddlers was analyzed, the hair segment representing the third trimester period contained the same amount of cortisol as the hair grown more recently in mothers with 3-4 months old toddlers only. Age of the women, hair curvature, hair color, and frequency of hair washes per week were unrelated to cortisol levels. We conclude that cortisol measured in human hair can be a valid reflection of increased cortisol production for a period of up to six months. Due to a rapid decline of cortisol levels in human adult hair, a retrospective calendar of cortisol exposure may be limited to the past six months.
