The HPA axis mediates the response of the body to stress; as such, it has been a natural focus of biological research into a disorder with a close link to stress. A schematic representation of the HPA axis is shown in Figure 2.1. The HPA axis has been extensively studied in depression; about 50% of depressed patients show apicture of hypercortisolaemia. However, this varies with the symptomatic picture: rates are higher in those with features of DSM-IV melancholic depression, strong somatic symptoms, or psychosis (Schatzberg et al., 2002).
Assessing the HPA axis can be problematic. Cortisol is a pulsatile hormone, has a strong diurnal rhythm, and is released in stressful circumstances, such as blood sampling. For this reason, more detailed methods of endocrinological assessment are needed. A widely used method in depression has been the dexamethasone-suppression test (DST). Dexametha-sone is a synthetic glucocorticoid which suppresses hypothalamic corticotrophin-releasing hormone (CRH) and pituitary adrenocorticotrophic hormone (ACTH) via glucocorticoid receptors (Figure 2.1). In a proportion of depressed individuals, such suppression fails to occur, averaging around 60-70% in melancholic depression and 30-40% in 'neurotic' depression. The test is not specific to depression, as non-suppression can be seen in other
GR/MR receptors j A
hippocampus c hippocampal I I I neurons T T T
v acth adrenal cortex i cortisol-
metabolic effects depression hippocampal atrophy impaired 5-HT neurotransmission
Figure 2.1 Schematic representation of the control of the hypothalamo-pituitary-adrenal (HPA) axis. CRH: corticotrophin-releasing hormone; ACTH: corticotropin; AVP: argenine vasopressin; MR: mineralocorticoid receptors; GR: glucocorticoid receptors. Abnormalities in this axis in depression are shown in Table 2.1.
conditions. A further refinement of this test is the combined dexamethasone-CRH test. Dexamethasone pre-administration usually attenuates the cortisol response to CRH; in depression, this effect is less pronounced due to downregulated glucocorticoid receptors. This test has proved more able to distinguish depressed subjects from normals that the simple DST (Heuser et al., 1994). Table 2.1 outlines the specific findings of the various tests applied to components of the HPA axis.
Table 2.1 HPA axis abnormalities in depression relating to Figure 2.1
A Increased incidence of life events and chronic stress B Impaired hippocampal fast (rate-sensitive) feedback C Hippocampal atrophy
D Raised CRH levels in the cerebrospinal fluid; impaired negative feedback by dexamethasone E Impaired pituitary ACTH response to CRH administration; this may represent downregulated CRH receptors or negative feedback from high cortisol levels. Pituitary hypertrophy F Hypertrophied adrenal cortices G Hypercortisolism
Thus, in a substantial proportion of depressed patients, there is oversecretion of cortisol and reduced negative feedback at the hypothalamus and pituitary. Can cortisol hypersecretion be considered a plausible biological mechanism for depression? First, since cortisol is the main stress hormone, it is easy to see how it might mediate between life events and biological changes in depression. Evidence that raised cortisol levels may be driving depression rather than vice versa comes from studies showing that lowering cortisol levels, as for example, by administering cortisol synthesis inhibitor drugs such as metyrapone or ketoconazole, can alleviate depression (Murphy, 1997). However, raised cortisol secretion in endogenous Cushing's disease is associated with depression in 50-85% of cases. Furthermore, abnormally high cortisol levels have been shown to be associated with other biological changes, such as inhibitory effects on neuronal 5-HT neurotransmission; given the links between 5-HT neurotransmission and mood changes, this is a feasible mechanism by which 5-HT neurotransmission could become dysregulated. There are also suggestions that prolonged periods of high cortisol can lead to hippocampal atrophy—indeed, in Cushing's disease, the decreased hippocampus size can be correlated with plasma cortisol levels and cognitive impairment. Recent studies also suggest hippocampal atrophy in depression (Sheline & Minyun, 2002).
However, the role of the HPA may also be seen in other ways. Adverse circumstances in childhood, such as losing parents or suffering abuse, are well known to predispose an individual to depression. Recent work suggests that the HPA axis may provide some further understanding of the mechanism of this link. Experiencing childhood abuse leads to a long-term alteration of the stress response (Heim et al., 2000). Thus, it is also possible to see the HPA axis changes as a biological link between early life stresses and an increased vulnerability to stress and depression.
Others have noted that changes elsewhere in the HPA axis may mediate symptoms. For example, CRH may also act as a neurotransmitter, and it produces symptoms of agitation, insomnia, and reduced feeding in animals. The amount of CRH expressed in cells, and the co-occurrence of CRH with its synergistic ACTH releaser, vasopressin, is increased in depressed suicide victims. CRH levels in the cerebrospinal fluid (CSF) are increased. Furthermore, CRH receptors are found in the cortex, and show a reduced density in suicide victims, consistent with high levels of CRH release. Thus, increased CRH in depression could contribute to some symptoms (Nemeroff, 1996).
The importance of the HPA axis changes in depression go beyond the apparent ability to provide a neat mediator between stressful events and symptoms, and also concerns prognostic indicators. There are suggestions that DST non-suppression is associated with a poorer response to placebo (though not a superior response to medication) and a poorer response to cognitive therapy (Thase et al., 1996). More strikingly, if clinical response to treatment is associated with continued non-suppression, there is a fourfold increase in the risk of short-term relapse or suicide attempt. Recently, the long-term risk of suicide was also found to be more closely linked to DST non-suppression than any other factor, more so even than past suicide attempts (Coryell & Sehlesser, 2001).
In summary, there is no doubt that HPA axis dysfunction is present in a large proportion of depressed patients, particularly those with more melancholic symptom patterns. The HPA axis provides a plausible biological mechanism for some of the most replicated causal theories of depression. Thus, the links between depressive symptoms and stressful life
events, chronic social adversity, or traumatic or abusive childhoods could all be explained by the mediating role of the HPA axis.
Was this article helpful?