Cortisol perturbation in the pathophysiology of septicaemia, complicated pregnancy and weight loss/obesity.

Abstract

Cortisol, the principal glucocorticoid secreted from the adrenal glands, is essential for life. Healthy cortisol levels are maintained through negative feedback on the central nervous system (CNS) – pituitary stimulatory apparatus which regulates production of adrenocorticotropin (ACTH) and contains a light–entrained intrinsic CNS driven diurnal rhythm. Cortisol participates in a regulatory mechanism where inflammatory cytokines stimulate cortisol release and cortisol in turn suppresses cytokine release. The effects of cortisol in inflammatory states include elevating blood pressure and metabolic regulation. This thesis contains three exploratory studies examining circulating cortisolaemia using the best available methodologies (total and free cortisol and corticosteroid-binding globulin (CBG)) in clinical states characterized by immune activation/ inflammation and altered blood pressure. These clinical states include: (1) septic shock, (2) hypertensive disorders of pregnancy and (3) obesity-induced hypertension. Prior to the studies described here, little was know about cortisolaemia in these common pathological states. Septic shock is a life threatening condition that complicates severe infection and is characterized by systemic inflammation and refractory hypotension. High plasma total cortisol levels and attenuated responses to synthetic ACTH stimulation are associated with increased mortality. The use of corticosteroids in septic shock has been highly controversial for decades, however recent trials have reported haemodynamic and survival benefits associated with the use of physiologic steroid replacement in patients with relative adrenal insufficiency (RAI) – currently defined as a total cortisol increment of 248 nmol/L or less following ACTH (250 μg) stimulation. However, CBG and albumin levels fall by around 50% with an increase in plasma free cortisol in critical illness. Hence, total cortisol may not reflect the biologically active free (unbound) cortisol, suggesting that standard assays for plasma cortisol (which measure total plasma cortisol) underestimate HPA axis activity. In this study, we have showed that plasma free cortisol is a better guide to circulating glucocorticoid activity in systemic infection than total cortisol. We have also validated the use of Coolens’ method in estimating free cortisol in systemic infection, using plasma total cortisol and CBG measurements as plasma free cortisol is not performed in clinical laboratories. Free cortisol measurement allows better categorization of RAI and non-RAI groups with a free cortisol increment of 110 nmol/L as cut-off. Moreover, we have shown that survivors of RAI have normal adrenocortical function on follow-up testing suggesting a lack of functional adrenal reserve rather than adrenal damage during critical illness. Larger randomized controlled trials will be required to redefine RAI using free cortisol measurements and relate that to clinical outcomes and responses to corticosteroid therapy. Nitric oxide (NO) is normally produced in the endothelium by the constitutive form of the NO synthase and this physiologic production is important for blood pressure regulation and blood flow distribution. Studies have shown that an overproduction of NO by the inducible form of NO synthase (iNOS) may contribute to the hypotension, cardiodepression and vascular hyporeactivity in septic shock. Clinical studies of non-selective inhibitors of the L-arginine nitric oxide pathway showed increased mortality from cardiovascular complications. However, glucocorticoids, which improve vasopressor sensitivity, may act by partially suppressing NO synthesis through selective direct inhibition of iNOS, and suppression of inflammatory cytokine synthesis. Hence, plasma nitrate/ nitrite (NOx) levels may provide a titratable end point to individualize glucocorticoid therapy in sepsis. The NOx study in this thesis showed that cortisol (total and free), CBG and NOx correlated to illness severity. Free cortisol, and to a lesser extent total cortisol, but not NOx levels, predicted septic shock. NOx levels were characteristically stable within individuals but inter-individual differences were only partly accounted for by illness severity or renal dysfunction. NOx levels correlated weakly with cortisol, did not relate to the need for vasopressors and were not suppressed by hydrocortisone treatment. Thus, NOx is not a suitable target for glucocorticoid therapy in septic shock. Pregnancy is the only sustained physiologic state of hypercortisolism in humans. A large body of data suggests that excessive foetal and prenatal glucocorticoid exposure leads to reduced birth weight and adverse health in offspring such as elevated blood pressure and insulin resistance. Pre-eclampsia and gamete donor pregnancies are associated with immune activation, elevated inflammatory cytokines as well as elevated blood pressure. Prior to the study described in this thesis however, there was no prospective data on maternal cortisolaemia in these complicated pregnancies. My study has demonstrated for the first time that there was a substantial fall in plasma CBG levels in the last few weeks of gestation with a corresponding rise in free cortisol in normal pregnancy, a finding obscured for methodological reasons in past studies. This free cortisol elevation in late pregnancy may facilitate organ maturation in the foetus and perhaps prepare the mother for the metabolic demands of labour. In pre-eclampsia and gestational hypertension, plasma CBG, total and free cortisol levels were lower in late third trimester; and in IUGR, plasma CBG levels were suppressed from 28 weeks gestation until delivery but with no significant difference in plasma total and free cortisol. Women with assisted reproduction using donor gametes/ embryos had significantly lower plasma CBG, total and free cortisol levels even in those with normal pregnancy outcomes. Low CBG may be due to reduced synthesis or enhanced inflammation-driven degradation. Low maternal cortisol may be due to a lack of placental corticotropin-releasing hormone, or reduced maternal ACTH, driving cortisol production. This unanticipated maternal hypocortisolism in complicated pregnancies may trigger precocious activation of the foetal HPA axis and could have implications for postnatal and adult health. Speculatively, since excess prenatal GCs increase HPA axis activity, we proposed that maternal hypocortisolism may predispose to the hypocortisolaemic state characterized by fatigue, pain and stress sensitivity, in offspring. The third state of immune/ inflammatory activation associated with blood pressure dysregulation studied in this thesis is obesity. The epidemiologic relationship between obesity and hypertension is widely recognised. Central obesity in particular has been associated with exaggerated HPA responses to stimuli. Studies of severe dieting and starvation resulted in hypercortisolism and a significant decrease in CBG. The HPA axis and the renin-angiotensin-aldosterone system (RAAS) have been implicated in the pathophysiology of obesity-induced hypertension. However, there is little data on the effect of moderate weight loss (30% caloric restriction) on adrenocortical function, and the relation of adrenal hormones to altered blood pressure with weight loss. In this study, measures of HPA axis and RAAS and blood pressure monitoring were performed in twenty-five obese subjects before and after a 12-week diet program (6000kJ/day). Short-term, moderate weight loss (mean 8.5 kg) was associated with a small reduction in blood pressure (mean arterial pressure 6 mmHg) and significantly reduced levels of aldosterone and renin but not cortisol levels. These findings suggest that aldosterone may have an important role in the blood pressure reduction with weight loss via a renin mediated mechanism, perhaps involving renal sympathetic tone. In contrast to severe caloric restriction, HPA axis activation does not occur with moderate weight loss. This suggests a threshold effect of weight loss on the HPA axis where greater caloric restriction is required for HPA stimulation, or a counterbalancing of central and direct adrenal effects on HPA axis function. Overall, these three exploratory studies have provided novel data on HPA axis function in systemic infection, pregnancy and in diet-induced weight loss. Each study offers a basis for further studies of HPA axis function in these disorders.