Bleker LS, Roseboom TJ, Vrijkotte TG, Reynolds RM, de Rooij SR. Determinants of cortisol during pregnancy – The ABCD cohort. Psychoneuroendocrinology. 2017;83:172-81.
Bleker and colleagues conducted a study to examine the determinants of cortisol during pregnancy. They hypothesised that physiological and psychological stressors would contribute to maternal cortisol levels which would in turn provide an altered intra-uterine environment for the fetus and predispose to adverse neurodevelopment outcomes in the child.
They used data from the Amsterdam-Born Children and their Development (ABCD) cohort. Participants were all pregnant women in Amsterdam during 2003 – 2004 (n=12,373). Of these women, 67% returned the study questionniare and of these, n=4389 participated in the biomarker substudy. Women without information on the time of blood collection were excluded (28%) as were women with multiple pregnancies and those women who used steroid medication. The majority of women completed the questionnaire after the blood sample was taken, although the exact timing was quiet variable (mean 12 days; SD 17 days).
Of particular interest is that women were asked if they perceived the number of hours they slept as “sufficient”, “too long”, or “too little”. Lifestyle risk factors and psychological stress were also assessed via various questionnaire based measures.
The main factors correlated with cortisol levels were gestation age (r=0.3) and time of day (r=-0.4). With adjustment for these factors, the factors that contributed significantly to a multivariate regression model for cortisol level were (nmol/L): maternal age (-1.5), multiparity (-21.5), pre-pregnancy BMI (-1.3), CRP level (1.0), female fetus (9.2), smoking (-14.2), insufficient sleep (-8.5), and employment (-12.7).
The sensitivity analysis carried out only in women in the 2nd trimester and whose cortisol was collected between 11am – midday suggests that while there was no significant association between sleep and cortisol levels, the effects were in the same direction as the main analysis.
The authors note that none of the questionnaire measures of stress were significantly linked to cortisol level, however, that is unsurprising given that sleep and lifestyle and employment were likely strongly collinear with self-reported stressors.