Network identification of hormonal regulation

Characterizing Alterations in Cortisol Secretion During Cardiac Surgery

Cortisol is a neuroendocrine hormone of the hypothalamus-pituitary-adrenal (HPA) axis secreted from adrenal glands in response to stimulation by adrenocorticotropic hormone (ACTH) from the anterior pituitary and corticotropin releasing hormone (CRH) from the hypothalamus. Cortisol has multiple functionalities in maintaining bodily homeostasis - including anti-inflammatory influences - through its diurnal secretion pattern (which has been studied extensively); its secretion is also increased in response to major traumatic events such as surgery. Due to the adverse health consequences of an abnormal immune response, it is crucial to understand the effect of cortisol in modulating inflammation. To address this physiological issue, we characterize the secretion of cortisol using a high temporal resolution dataset of ten patients undergoing coronary arterial bypass grafting (CABG) surgery, in comparison with a control group not undergoing surgery. We find that cortisol exhibits different pulsatile dynamics in those undergoing cardiac surgery compared to the control subjects. We also summarize the causality of cortisol's relationship with different cytokines (which are one type of inflammatory markers) by performing Granger causality analysis.Clinical relevance- This work documents time-varying patterns of the HPA axis hormone cortisol in the inflammatory response to cardiac surgery and may eventually help improve patients' prognosis post-surgery (or in other conditions) by enabling early detection of an abnormal cortisol or inflammatory response and enabling patient specific remedial interventions.

Characterization of Leptin Secretion in Premenopausal Obese Women Treated with Bromocriptine

Leptin, a hormone secreted by adipose tissue, is primarily responsible for inhibiting hunger and maintaining energy balance. Improper leptin secretion may result in hyperleptinemia (excess secretion of leptin) or leptin resistance, both of which contribute to obesity. Diagnosing abnormal leptin secretion may help treat this underlying cause of obesity. Therefore, continuous monitoring of the level of leptin may help characterize its secretion dynamics and also help devise an appropriate treatment. In this research, we consider leptin hormone concentration data taken over a 24 hour time period from eighteen healthy premenopausal obese women before and after treatment with a dopamine agonist, bromocriptine, and deconvolve the observed leptin hormone levels to estimate the number, timing, and magnitude of the underlying leptin secretory pulses. We find that there is an overall decrease in leptin secretion, particularly during sleep, but the changes in the secretory and clearance rates, and the number of pulses underlying the secretion process are not statistically significant.Clinical relevance- This work seeks to understand the effect of bromocriptine on leptin secretory dynamics and will help further current understanding of the effect of bromocriptine in relation to obesity.

Interrelationships Between Pituitary Hormones as Assessed From 24-hour Serum Concentrations in Healthy Older Subjects

CONTEXT

Hormones of the hypothalamic-pituitary-target gland axes are mostly investigated separately, whereas the interplay between hormones might be as important as each separate hormonal axis.

OBJECTIVE

Our aim is to determine the interrelationships between GH, TSH, ACTH, and cortisol in healthy older individuals.

DESIGN

We made use of 24-hour hormone serum concentrations assessed with intervals of 10 minutes from 38 healthy older individuals with a mean age (SD) of 65.1 (5.1) years from the Leiden Longevity Study. Cross-correlation analyses were performed to assess the relative strength between 2 24-hour hormone serum concentration series for all possible time shifts. Cross-approximate entropy was used to assess pattern synchronicity between 2 24-hour hormone serum concentration series.

RESULTS

Within an interlinked hormonal axis, ACTH and cortisol were positively correlated with a mean (95% confidence interval) correlation coefficient of 0.78 (0.74-0.81) with cortisol following ACTH concentrations with a delay of 10 minutes. Between different hormonal axes, we observed a negative correlation coefficient between cortisol and TSH of -0.30 (-0.36 to -0.25) with TSH following cortisol concentrations with a delay of 170 minutes. Furthermore, a positive mean (95% confidence interval) correlation coefficient of 0.29 (0.22-0.37) was found between TSH and GH concentrations without any delay. Moreover, cross-approximate entropy analyses showed that GH and cortisol exhibit synchronous serum concentration patterns.

CONCLUSIONS

This study demonstrates that interrelations between hormones from interlinked as well as different hypothalamic-pituitary-target gland axes are observed in healthy older individuals. More research is needed to determine the biological meaning and clinical consequences of these observations.

Comparing Methods for Measurement Error Detection in Serial 24-h Hormonal Data

Measurement errors commonly occur in 24-h hormonal data and may affect the outcomes of such studies. Measurement errors often appear as outliers in such data sets; however, no well-established method is available for their automatic detection. In this study, we aimed to compare performances of different methods for outlier detection in hormonal serial data. Hormones (glucose, insulin, thyroid-stimulating hormone, cortisol, and growth hormone) were measured in blood sampled every 10 min for 24 h in 38 participants of the Leiden Longevity Study. Four methods for detecting outliers were compared: (1) eyeballing, (2) Tukey’s fences, (3) stepwise approach, and (4) the expectation-maximization (EM) algorithm. Eyeballing detects outliers based on experts’ knowledge, and the stepwise approach incorporates physiological knowledge with a statistical algorithm. Tukey’s fences and the EM algorithm are data-driven methods, using interquartile range and a mathematical algorithm to identify the underlying distribution, respectively. The performance of the methods was evaluated based on the number of outliers detected and the change in statistical outcomes after removing detected outliers. Eyeballing resulted in the lowest number of outliers detected (1.0% of all data points), followed by Tukey’s fences (2.3%), the stepwise approach (2.7%), and the EM algorithm (11.0%). In all methods, the mean hormone levels did not change materially after removing outliers. However, their minima were affected by outlier removal. Although removing outliers affected the correlation between glucose and insulin on the individual level, when averaged over all participants, none of the 4 methods influenced the correlation. Based on our results, the EM algorithm is not recommended given the high number of outliers detected, even where data points are physiologically plausible. Since Tukey’s fences is not suitable for all types of data and eyeballing is time-consuming, we recommend the stepwise approach for outlier detection, which combines physiological knowledge and an automated process.