24-hour saliva cortisol measurements in twins and siblings selected to be at high or low genetic risk for anxious depression

Mireille van den Berg¹, Eco de Geus¹, Dorret I. Boomsma¹, Connor V. Dolan², and Clemens Kirschbaum<sup>3

1</sup> Department of Biological Psychology, Vrije Universiteit, Amsterdam, The Netherlands
² Department of developmental psychology, University of Amsterdam, The Netherlands
³ Institute of Physiological Psychology II, University of Düsseldorf, Germany

Multiple factors contribute to the onset of major depression, including a dysregulation of the HPA-axis resulting in high cortisol levels. The purpose of this paper is to: 1) determine whether the relation between heightened cortisol levels and depression is also found in non-clinical samples, 2) examine the nature of this correlation in a genetic design. 309 twins and siblings, ascertained from 160 families registered in the Netherlands Twin Register, participated in saliva cortisol sampling during a 24-hour period. Selection of families is based on a multivariate genetic analysis, on Beck depression, the Young Adult Self-Report, ABV Neuroticism and Spielberger Trait anxiety, derived from a longitudinal study on health and lifestyle. Data of up to 4 repeated assessments, over the period 1991-1997, was available. Subjects were selected to form extreme concordant or discordant sib pairs. In selected offspring, the Composed International Diagnostic Interview (CIDI) was performed, from which DSM-IV diagnosis can be derived. The subjects collected cortisol during a 24-hour period: directly after arrival of the researcher between 9-10 (1st morning sample), 11.00, 15.00, 20.00 and at 22.30. The sixth sample was collected the following morning immediately after awakening. Questionnaire data on general health, health behaviour and mood are available for the measurement day. A normal diurnal rhythm was observed. The high trait depression (htd) group had higher cortisol levels during the day than the low trait depression (ltd) group, according to prediction. However, subjects with a DSM-IV diagnosis had lower cortisol levels than the non-depressed. First analysis suggest that MZ correlations for cortisol are higher than DZ correlations on all measurement points, indicating heritability, save the 1st morning sample. Results of MX modelling will be presented.