A STATE-SPACE INVESTIGATION OF CORTISOL ALTERATIONS IN CHRONIC FATIGUE SYNDROME

Date

2018-10-18

Authors

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Chronic Fatigue Syndrome (CFS) causes prolonged and crippling fatigue in patients. Reports estimate a $17 to $24 billion loss in productivity and medical care costs due to CFS annually [1]. Since the initial observations in 1930s [2], experts have tried to study multiple considerations for potential etiopathogenic hypotheses [3]. Currently, there is a growing focus on the hypothalamic-pituitary-adrenal (HPA) axis dysfunction in CFS patients, more significantly on its end product, cortisol secretion. Along with the autonomic nervous system, cortisol is responsible in controlling blood sugar level, regulating metabolism, and modulating inflammation response accordingly, during either undisturbed or stressed condition of the human body. Previous studies reported a trend of lower cortisol levels in early morning, lag in the circadian rhythms’ acrophase, and decreased adrenal glands’ response to stressor stimuli [3], [4], [5], [6]. In order to study this trend, we analyze cortisol secretion patterns in terms of the input-output relationship. We incorporate the pulsatile architecture with the pharmacokinetics of cortisol secretion dynamics into a state-space model. Then, we use a sparse deconvolution approach similar to Faghih et al. [7] to deconvolve experimental cortisol data collected by Crofford et al.’s [4]. In particular, we deconvolve experimental cortisol data from 12 CFS patients and 12 matched healthy controls. We obtain numbers, timings, and amplitudes of cortisol secretory events along with cortisol infusion and clearance rates with R2 above 0.88. We performed hypothesis testing on the various features of the recovered cortisol secretory events between 2 AM and 9 AM; during this timeframe, cortisol’s circadian rhythm increases from nadir to peak [8]. Comparing the mean amplitudes of the secretory events during this time window, the Wilcoxon signed rank test at an =0.05 level of significance revealed that the difference between the CFS patients and control subjects was significant. The Wilcoxon signed-rank test failed to reject the hypothesis that mean inter-arrival times, the number of the secretory events, the overall amount of the secretory events, and the energy in the secretory events differ in the two groups. This study could be extended to design a systematic diagnostic and treatment approach for CFS based on the HPA axis function.

Description

Keywords

Citation