Anxiety and depressive symptoms effects on cortisol trajectories from pregnancy to postpartum: Differences and similarities between women and men

Highlights

• The pattern of cortisol variations in the perinatal period is similar in women and men.

• Anxiety and depressive symptoms are associated with flattened cortisol variations in the perinatal period.

• Effects of anxiety and depressive symptoms on cortisol variations are sex-specific.

Abstract

Anxiety and depressive symptoms may influence cortisol trajectories in women and men during pregnancy and the postpartum period. Using a multilevel approach, anxiety and depressive symptoms effects on 24-hour urinary free cortisol trajectories from the 2nd trimester to 3-months postpartum were examined in a sample of 66 women and 65 men with no known psychosocial or medical risk (N = 131; 33 (50%) of them were couples that participated in the same assessment waves). Results showed that both anxiety and depressive symptoms influence women's and men's cortisol trajectories from mid-pregnancy to 3-months postpartum. Women with high depressive symptoms and men with high anxiety or high depressive symptoms exhibited less accentuated variations in the 24-hour urinary free cortisol trajectories compared with women with low depressive symptoms and men with low anxiety or depressive symptoms, respectively. These effects were significant for women's cortisol trajectories from the 2nd to the 3rd pregnancy trimester and for men's cortisol trajectories throughout the entire period. The effect of anxiety and depressive symptoms on HPA axis functioning and cortisol production during pregnancy and postpartum, seems to be sex-specific. Reproductive-related alterations (associated with gestation, parturition and lactation) in women's HPA axis functioning may explain these sex-specific effects.

Introduction

Pregnancy and the postpartum are challenging periods for both women and men, associated with both anxiety and depressive symptoms (e.g., Don et al., 2014; Figueiredo and Conde, 2011a; Underwood et al., 2016) and hormonal changes (e.g., Conde and Figueiredo, 2014; Kane et al., 2014). Anxiety and depressive symptoms paths over pregnancy and the postpartum periods are generally similar in women and men: higher levels of anxiety at the 1st and 3rd pregnancy trimesters compared to the 2nd trimester (e.g., Figueiredo and Conde, 2011a; Lee et al., 2007), followed by a decline after childbirth (e.g., Andersson et al., 2006; Canário and Figueiredo, 2017; Don et al., 2014), and a decrease in depressive symptoms from the 1st to the 3rd pregnancy trimester (e.g., Figueiredo and Conde, 2011a; Tendais and Figueiredo, 2016) and at 3 to 6 months postpartum compared to mid-pregnancy (e.g., Andersson et al., 2006; Figueiredo and Conde, 2011a; Heron et al., 2009). Nevertheless, there are sex differences in the severity of psychopathological symptoms (e.g., Figueiredo and Conde, 2011a; Parfitt and Ayers, 2014; Rollè et al., 2017). Women have increased anxiety and depressive symptoms during pregnancy and childbirth (Figueiredo and Conde, 2011a; Teixeira et al., 2009) but not in the postpartum (Figueiredo and Conde, 2011b) compared to men.

Stage-specific differences in cortisol levels are also generally similar in women and men: increase during pregnancy with a peak in labor and decrease during the postpartum period (Cheng and Pickler, 2010; Conde and Figueiredo, 2014; Jung et al., 2011; Storey et al., 2000). Evidence suggests that in women the mechanisms underlying cortisol changes are influenced by the reproductive status (Handa and Weiser, 2014). Alterations in the Hypothalamic-pituitary-adrenal (HPA) axis functioning can be triggered by gestation, fetal development, and parturition phenomena (Glynn et al., 2013; Kammerer et al., 2006; Murphy et al., 2006). During the postpartum period, the decline in cortisol release is associated with the placenta expulsion and continuity of hypertrophy of the adrenal glands (Cheng and Pickler, 2010; Glynn et al., 2013; Jung et al., 2011). These changes may have adaptive functions including: regulation of parental behavior (Carter and Porges, 2012), facilitation of energy conservation required for lactation, buffer from stress-associated inhibition of lactation, and improvement of the immune function (e.g., Glynn et al., 2013; Hahn-Holbrook et al., 2011). Modifications in HPA axis function persist over the postpartum period and best estimates suggest that it takes at least two to three months for the HPA axis to return to pre-pregnancy functioning pattern (Glynn et al., 2013; Jung et al., 2011; Magiakou et al., 1996; Owens et al., 1987).

Reports from neuroendocrine research conducted in adults, although not in the perinatal period, show an association between anxiety and depressive symptoms and a dysregulation of the HPA axis (Kofman et al., 2019), determining alterations in cortisol release (Elnazer and Baldwin, 2014; Oldehinkel and Bouma, 2011). In individuals with anxiety and depressive symptoms, the HPA axis seems to be hyperactive, with increased 24-hour urinary free cortisol levels (Hess et al., 2007; Kathol et al., 1989; Maes et al., 1998). These alterations seem to be sex-specific, since women with major depressive disorders or anxiety disorders exhibit a blunted cortisol response to psychosocial stress compared to controls, whereas men with major depressive disorders or anxiety disorders exhibit an elevated cortisol response (Zorn et al., 2017). Specifically, during pregnancy and postpartum periods, changes in gonadal steroid hormones can compromise the HPA axis ability to respond to environmental stimuli, since reproduction is regulated by a neuroendocrine axis, the Hypothalamo–pituitary–gonadal (HPG) axis, involving the hypothalamus, anterior pituitary gland and gonads (Handa and Weiser, 2014). Blunted plasma cortisol to acute stressors driven by anxiety during pregnancy suggests that the rising circulating levels of HPA axis hormones and placental corticotropin-releasing hormone (CRH) influence the endocrine response to challenges (De Weerth and Buitelaar, 2005; Serati et al., 2016). This is also consistent with clinical study reports showing non-significant or low associations between stress and cortisol levels in women (e.g., Giesbrecht et al., 2012; O'Connor et al., 2014). Depressed women during pregnancy exhibit higher cortisol levels, lower wake-up, flattened, less sharply declining diurnal pattern (O'Connor et al., 2014), and elevated night-time (Peer et al., 2013) cortisol levels. In the postpartum period depression is associated with a dysregulated HPA axis activity (Glynn et al., 2013), reflected in a lack of a cortisol morning rise (Taylor et al., 2009), reduced cortisol diurnal variation (de Rezende et al., 2016) and elevated evening salivary cortisol levels when compared to healthy controls (Iliadis et al., 2015).

The majority of the studies conducted so far are cross-sectional, focused on women and on the association between psychological symptoms and cortisol levels at specific time points, without controlling for individual differences in cortisol levels. Recent studies suggest that a longitudinal approach and multilevel analysis are necessary to clearly describe the potential effects of psychological symptoms on cortisol levels (Giesbrecht et al., 2012; Kane et al., 2014). Cortisol-related research also involves a variety of methods for biological samples collection, including blood, urine and saliva (Sarkar et al., 2013), depending on the aim of the study. The 24-hour urinary free cortisol is considered a moderately stable indicator of adrenocortical output, therefore the best way to assess the overall or long-term HPA axis activity during pregnancy and postpartum, namely in non-acute experimental situations (Yehuda et al., 2003). As far as we know, no studies have been published on the association between men's anxiety or depressive symptoms and cortisol levels during the prenatal and postpartum period.

Using a multilevel approach, the current study examines anxiety and depressive symptoms effects on women's and men's 24-hour urinary free cortisol trajectories from the 2nd pregnancy trimester (20 to 24 weeks) to 3-months postpartum, adjusting for parity. Parity was adjusted in the analysis due to differences between first and second-time parents (2nd child or beyond) in anxiety and depressive symptoms (Chen et al., 2019; Figueiredo and Conde, 2011a; Skari et al., 2002) and in cortisol levels (Conde and Figueiredo, 2014) over pregnancy and postpartum period. We hypothesize that both anxiety and depressive symptoms effects on cortisol trajectories in women may be different from those observed in men, not only because the HPA axis functioning is different in women and men (Helpman et al., 2017), namely in the presence of a psychiatric disorder (Zorn et al., 2017), but also because the reproductive and lactation phenomena interferes with the HPA axis functioning (Altemus et al., 1995; Handa and Weiser, 2014).