The HPA Axis in Bipolar

In the last post, I threw around the term HPA axis without describing what it is.  I will try to rectify that here.  The HPA axis, more correctly called by its full name the hypothalamic-pituitary-adrenal axis, is a part of the  neuroendocrine system.  The HPA Axis is largely involved in responses to stress---sometimes termed the 'flight or flight' response.  Stress is the input to the system and can come in the form of either physiological or psychological demands.  In either event, these stresses are input to the HPA axis through the neurons that connect to it, including those involved in mood regulation.

There is a nice review of the HPA axis for bipolar

Duffy A et al. Biological indicators of illness risk in offspring of bipolar parents: targeting the hypothalamic-pituitary-adrenal axis and immune system. Early Interv Psychiatry. 6(2):128-37.  2012.

How does the HPA axis work?  There is a nice picture of this process at Wikipedia.  What we know at the moment says that there are three major organs (in italics) involved in signalling cascade and feedback loop whereby the system self-regulates.  The signal cascade originates in the neurons connected to the paraventricular nucleus in the  hypothalamus which in turn produces vasopressin and corticotropin-releasing hormone (CRH).  The vasopressin and CRH work together to cause the pituitary gland to produce adrenocorticotropic hormone (ACTH).  In response to ACTH, the adrenal cortex produces glucocorticoid hormones (mainly cortisol in humans).  These glucocorticoid hormones in sufficient quantity suppress the activity of the hypothalamus and the pituitary gland which generate less CRH and ACTH.  The glucocorticoid hormones by suppression work to extinguish the stress-response.  It is important to note that cortisol has many functions throughout the body (including functions for the immune system), and most tissues have glucocorticoid receptors.

The previous paragraph is fairly detailed.  Why should we go through all this?  There are various times at which the HPA axis can be disregulated.  Specifically, we are interested in the form of disregulation called hyperactivity which is characterized by inhibited response to CRH and increased levels of cortisol in the saliva and blood.  We know that Cushing's syndrome is a disease where hyperactivity of the HPA axis can result in euphoria or even psychosis.  Additionally, many studies show strong evidence of a link between HPA axis disregulation and bipolar.  For example

Watson S et al. Hypothalamic-pituitary-adrenal axis function in patients with bipolar disorder. Br J Psychiatry.  184:496-502. 2004.

which shows a p-value of 0.001 for the presence of increased cortisol in the saliva of individuals with bipolar.  Just to remind everyone, the lower the p-value the likely that the result is real, and values below 0.05 are typically considered strong.  The p-value is the probability of the ranked cortisol levels in the saliva of the two groups (non-bipolar, and bipolar) under a random model of ranked cortisol levels.  The main point is that there are many papers such as the one cited here, and the result of statistically elevated cortisol levels in people with bipolar is a robust finding regardless of whether the people have active or remitted bipolar.

There are similar findings of increase cortisol levels in people with depression.

Daban et al. Hypothalamic-pituitary-adrenal axis and bipolar disorder.  Psychiatr Clin North Am. Jun;28(2):469-80. 2005.

Similarly depression can remit, but the HPA axis function might not return to normal.  Indeed, for people in remission, if the HPA axis function improves, and then a return to hyperactivity can predict the return of depression.  This relationship causes one to wonder if HPA axis hyperactivity should be taken into account when diagnosing mental illnesses.

Duffy et al. suggest that the HPA hyperactivity is due to the over production of CRH which results in a failure of the suppressing effect of cortisol.  They say that successful treatment with fluoxetine, amitriptyline, desipramine or electroconvulsive treatment returns CRH levels to normal.  They also speculate that Lithium is effecting one of these pathways.  Notice in this paragraph all the tentative words such as 'suggest' and 'speculate'.  These words indicate that much of this is hypothetical and has not been tested.  However, this is the opinion of experts formed from the most recent available data.

Is the relationship between bipolar and the HPA axis overstated?  The relationship between the HPA axis and bipolar are not fully explained, yet.  We do not know if there is a causality to the link, or in what direction it might go.  We have some strong links between cortisol and bipolar, but we have yet to have solid data on the potential links between bipolar and the other hormones in the HPA axis.  We do not know whether it is the input to the HPA axis that is faulty or some part of the signal cascade in the HPA axis.  We do not know the temporal associations between HPA axis disregulation and bipolar mood states.  While it would be nice to predict that perturbations in cortisol levels correlates to extreme mood states, we simply do not have the data to draw any conclusions of this nature.