Abstract
INTRODUCTION: Stress activates the hypothalamic-pituitary-adrenal (HPA) axis of which cortisol is an end product. 'Allostatic load' is where systems including the HPA axis are exposed to high, cumulative, physiologic burdens (such as chronic breathlessness) leading to flatter diurnal cortisol slopes and poorer health outcomes. The aim of this hypothesis-generating study explored longitudinal changes in cortisol secretion and any associated changes in breathlessness after introducing regular, low dose morphine or placebo. METHODS: This was an optional, hypothesis-generating sub-study embedded in a multi-site, randomised, double-blind, placebo-controlled trial (RCT) of regular, low-dose morphine for chronic breathlessness and chronic obstructive pulmonary disease. In a blinded dose-increment algorithm by week three, doses were 0 mg-32 mg. Participants in the RCT could elect to continue in a six-month blinded extension. This sub-study excluded people who used non-inhaled corticosteroids in the previous month or were on subcutaneous insulin. Participants collected saliva for cortisol assays for two days at baseline, and ends of weeks 1, 3 and 12 at 3,6 and 12 h after waking, generating sufficient data to calculate diurnal cortisol slopes and areas under the curve (AUC). Samples were analysed using ELISA. Correlations between diurnal cortisol profiles (slope and AUC) and a range of measures were explored. RESULTS: Twenty mostly female former smokers were in this sub-study. At baseline and the end of week 1, one-way ANOVA between-group analyses showed no significant differences in the log-transformed cortisol slope or ln-AUC. There was a strong correlation between the age-adjusted Charlson Comorbidity Index (CCI) and ln-AUC (r=-0.70, p < 0.001) and moderate correlation with age (r=-0.43, p = 0.06). In the blinded extension study, there was a self-selecting blinded group (n = 7) all on active medication. Global impression of change (GIC) was highly correlated with the diurnal cortisol slope (rs = 0.98, p = 0.01), and with decrease in average breathlessness (r = 0.89, p = 0.04). DISCUSSION: This hypothesis-generating study did not show a relationship between the diurnal cortisol profile and morphine in people with chronic breathlessness and COPD. For the sub-group still on study at 12weeks, the cortisol curves became steeper as average breathlessness decreased and as global impression of change (GIC) improved, suggesting that reducing breathlessness may potentially positively impact the HPA axis in a sub-group of people. TRIAL REGISTRATION: Registration Number NCT02720822 date registered 28/03/2016.