Novel insights in the HPA-axis during critical illness
Critical illness is defined as any condition that requires support of failing vital organ functions, without which death would rapidly ensue. As such, it represents an extreme example of physical stress, where stress comprises the normal physical response of the human body that ariseswhen confronted with a threat. In healthy individuals, experiencing stress causes an immediate activation of the hypothalamic-pituitary-adrenalaxis, which induces secretion of the hypothalamic corticotropin-releasing hormone, followed by the release of adrenocorticotropic hormone (ACTH) from the anterior pituitary gland. ACTH causes production of cortisol from cholesterol in the adrenal gland. Cortisol itself exerts feedback inhibition at the pituitary and the hypothalamic level, as such regulating its own release. Under normal conditions, these hormones are secreted in a pulsatile manner according to a fixed daily rhythm. Whenever stressoccurs, activation of the hypothalamic-pituitary-adrenal axis causes anadditional episodic release of cortisol.
Critical illness is therefore hallmarked by the presence of high circulating cortisol levels, proportionate to the severity of illness. This was traditionally attributed to activation of the hypothalamic-pituitary-adrenal axis and increased ACTH-driven cortisol production. However, ACTH levels were observed to be low, suggesting that non-ACTH dependent regulators of cortisol production are important during critical illness. Cortisol production has even never been quantified during critical illness. Furthermore, it is assumed that activation of the hypothalamic-pituitary-adrenal axis can sometimesbe insufficient to cope with severe illness. In the presence of high plasma cortisol, this condition has been labeled relative adrenal insufficiency. However, the underlying mechanisms remain unknown. Also, it remains controversial whether treating such patients with corticosteroids is required, as interventional studies provided conflicting results.
The main objective of this doctoral thesis was to explore the regulation of cortisol secretion and metabolism during critical illness in order to gain more insight in the pathophysiology of adrenal insufficiency. The general hypothesis of this doctoral thesis postulated that during critical illness reduced cortisol breakdown, rather than continuously stimulated cortisol secretion, contributes to maintain the elevated plasmacortisol concentrations. Negative feedback inhibition, exerted by high circulating cortisol levels would then explain low ACTH levels. When ACTH deprivation sustains, adrenal integrity and function could be negatively affected, predominantly in the prolonged phase of critical illness.
In a first part, we investigated in critically ill patients and matched healthy control subjects daily ACTH and cortisol levels in plasma during the first week of illness; plasma cortisol clearance and production during infusion of cortisol isotopes as tracers; plasma clearance of 100 mg of synthetic cortisol and the urinary cortisol metabolites and enzymeexpression at tissue level to assess the major cortisol-metabolizing enzymes. This showed that while circulating cortisol levels were consistently higher in patients than in controls, ACTH levels were decreased during critical illness. Cortisol production was less than doubled in patients. Furthermore, critically ill patients showed a more than 50% reductionin cortisol clearance during tracer infusion and after the administration of exogenous cortisol. All these factors accounted for the 4-fold increased plasma cortisol levels in patients, as compared with controls. Reduced cortisol metabolism was explained by reduced inactivation of cortisol in the liver and kidney of patients.
In the first study, only single sample ACTH and cortisol concentrations were reported. Since both ACTH and cortisol are secreted in pulses, this precludes analysis of theirsecretory dynamics and their relation during critical illness. In a second study, we therefore analyzed the dynamics of ACTH and cortisol secretion in critically ill patients and matched healthy control subjects, via nocturnal time series of repeated blood samples. We documented that hypercortisolemia during critical illness coincided with suppressed pulsatile ACTH and cortisol secretion, whereas the cortisol secretory response to a given plasma ACTH concentration was unaltered. These findings speak against the classical dogma of an activated hypothalamic-pituitary-adrenal axis in critical illness and instead suggest feedback-inhibition on ACTH exerted by circulating cortisol.
Considering the important function of ACTH in ensuring adrenal structure and function, theobserved low ACTH levels could negatively affect the adrenal glands, possibly predominantly in patients who stayed in the ICU for a prolonged time. In the third part of this thesis, we therefore investigated post-mortem adrenal glands from patients who died in the intensive care unit compared to adrenal glands from sudden out-of-hospital deaths as controls.Adrenal glands from patients who stayed long in the intensive care unitwere clearly more distorted in structure. Furthermore, the amount of stored cholesterol droplets was reduced in the adrenal glands from these patients. We also showed that the genes responsible for cholesterol uptake and synthesis as well as crucial genes for the production of cortisol were downregulated only in prolonged critically ill patients. Since cortisol cannot be stored in the adrenal gland and its production is dependent on rapid synthesis from cholesterol, these alterations limit cortisolproduction. Given that all the observed alterations are regulated by ACTH, one could speculate that the observed low ACTH levels might play a role and, when ACTH deprivation sustains, may help to explain the increased incidence of adrenal failure in the prolonged phase of critical illness.
In conclusion, reduced cortisol metabolism contributes to high cortisol levels during critical illness. Consequently, these high cortisollevels could suppress pulsatile ACTH secretion by negative feedback inhibition, causing acutely a reduced pulsatile cortisol secretion. Maintaining cortisol by not breaking it down seems a highly energy efficient way. However, the concurrent low ACTH levels could explain the observed disruption in structure and function of the adrenal gland in the prolongedphase of critical illness. These novel insights help to understand adrenal failure during critical illness and may have important implications for its diagnosis and treatment.