Is the cortisol awakening response truly a response to awakening? Replication and extension using overnight sampling

The cortisol awakening response (CAR) is a discrete component of the circadian cortisol profile. Evidence suggests that the CAR is a deviation from the pre-awakening increase in cortisol concentration, although this has yet to be replicated. Therefore, the purpose of this study was to replicate this finding and to investigate further the extent to which the CAR is distinct from the circadian profile. Twelve participants completed 2 overnight visits in a sleep laboratory. An intravenous catheter was used to sample blood every 15 min. Participants were allowed to sleep ad libitum beginning at 2030 h, and blood sampling continued throughout the night until 1 h post-awakening, generating a total of 802 serum samples. Selected serum samples were assayed for cortisol, and piece-wise linear mixed-effect models tested the extent to which the increase in cortisol concentrations post-awakening deviated from the rise in cortisol concentrations from various pre-awakening periods. Results demonstrated that the CAR only significantly deviates from the pre-awakening rise when the pre-awakening linear rise is considered 4 (β = 1.79, 95% CI = 0.49-3.09, p = 0.007) or 5 (β = 1.79, 95% CI = 0.49-3.09, p = 0.007) h before waking. When including a non-linear change in cortisol during the 4-h pre-awakening period, the deviation from the diurnal profile for the CAR was no longer significant (β = 0.96, 95% CI = - 0.74-2.66, p = 0.266). These results partly agree and replicate previous evidence for the CAR being superimposed on the diurnal profile but continue to extend our teleological understanding of the functional role of the CAR.