Do not exclude your observations: Negative cortisol awakening responses (CAR) may be biologically relevant

Abstract

The Cortisol Awakening Response (CAR) is the change in cortisol concentrations within 30–40 minutes after waking from sleep and is frequently used in stress research. Since a positive CAR is expected, we hypothesized that negative values could be associated to an underlying health condition (reflected in hematological parameters) or to environmental exposures such as lead (Pb), which has neuroendocrine effects including altered cortisol diurnal rhythms. Our aim was to analyze the prevalence of negative CAR values and their association with hematological parameters and blood Pb (BPb) levels in pregnant women (n = 900). Cortisol was measured by luminescence immunoassay in two-day saliva samples. CAR was estimated as the difference between the first (time of awakening) and second (45 min after) cortisol concentrations for each collection day and was operationalized as: both days positive (CAR-PP, 23 %), either day with a negative (CAR-NP/PN, 40 %), and both negative (CAR-NN, 37 %). A complete blood count was done using a coulter hematology analyzer. BPb was analyzed by inductively coupled plasma-mass spectrometry. Associations between hematological variables and CAR groups were analyzed using adjusted multinomial logistic regression models. Probabilities were estimated to assess the influence of BPb and hematological variables between CAR groups. The median (25th, 75th) CAR for the first collection day was −2.76 nmol/L (-16.55, 14.62) and −4.14 nmol/L (-17.66, 13.24) for the second day. Women with higher concentrations of leukocytes, eosinophils, basophils, and BPb were more likely to belong to CAR-NN or CAR-NP/PN groups. Compared to women with CAR-PP, those with CAR-NP/PN and CAR-NN had inverse associations for leukocyte levels and higher BPb concentrations. We conclude that negative CAR values could be an indicator of an underlying health condition or associated with environmental exposures such as Pb. Research should consider a thorough assessment of negative CAR values before excluding them from analyses.