Interstitial cortisol measurements aligned by wake time, healthy volunteers
Data files
Nov 24, 2024 version files 141.98 KB
-
healthy_cohort__Cortisol_LME_R_n_201.csv
54.76 KB
-
healthy_cohort_aligned_wake_Cortisol_LME_R_n_135.csv
36.96 KB
-
healthy_cohort_longest_quartile_Cortisol_LME_R_n_51.csv
13.82 KB
-
healthy_cohort_misaligned_wake_Cortisol_LME_R_n_64.csv
17.30 KB
-
healthy_cohort_shortest_quartile_Cortisol_LME_R_n_51.csv
13.97 KB
-
README.md
5.16 KB
Abstract
Background: Cortisol is released upon activation of the hypothalamic-pituitary-adrenal axis, varies across the day, and possesses an underlying diurnal rhythm reactive and responsive to stressors. While the endogenous peak of cortisol occurs in the morning, the observed cortisol rise post-awakening has been named the cortisol awakening response (CAR) based on the belief that the act of waking up stimulates cortisol secretion. However, objective evidence that awakening universally induces cortisol secretion is limited.
Methods: We used a mixed effects model with a linear spline fitted to the data to examine tissue free cortisol measurements obtained from 201 healthy volunteers by automated ambulatory microdialysis before and after awakening in their home environments. We also examined rate of change of cortisol depending on sleep duration and relative timing.
Results: We found no evidence for a change in the rate of cortisol increase in the hour after waking when compared with the hour prior to waking. We instead observed substantial interindividual variability in absolute concentration and rate of change, and differences in dynamics that may be attributable to duration and relative timing of sleep.
Conclusion: Based on these results, we strongly suggest caution is needed when interpreting cortisol measurements solely obtained in the hour after waking.
https://doi.org/10.5061/dryad.2280gb62w
Description of the data and file structure
The aim of this analysis was to investigate whether waking was associated with an increase in tissue free cortisol in a cohort of night-sleeping healthy volunteers, free from active medical diagnoses or regular medication, and who chose their own sleep period. To do this, we analysed data collected as part of the Dynamic Hormone Diagnostics (ULTRADIAN) observational cohort study (NCT02934399) and available in a public repository (UiB Open Research Data, https://doi.org/10.18710/5TW8YF).
The methods of data collection and preparation are described in detail in Upton, Zavala, Methlie et al. (2023) but briefly, involved the use of a portable microdialysis-based sampling system that enabled continuous sampling in subcutaneous tissue. Each participant completed a 24-hour sampling session generating 72 individual samples, representing a 20-minute sampling resolution. During the sampling period participants continued normal activities, and the sleep period occurred in the participant’s own home. Participants were free to select their own sleep and wake times, which were self-reported in an activity diary for both the night prior to and night of sampling. Concentrations of free cortisol and other adrenal steroids were quantified in each sample using liquid chromatography–tandem mass spectrometry.
In the dataset, each microdialysis measurement was timestamped as the midpoint of cumulative sample collected over a 20-minute period (for example sample collected between 06:00-06:20 would be labelled 06:10), and then the time of each sample was realigned relative to individual wake time. Since wake time did not necessarily coincide exactly with this midpoint sample time, we included any samples collected within the period of 70 minutes before and after wake time.
Files and variables
File: healthy_cohort__Cortisol_LME_R_n_201.csv
Description: Wake-aligned cortisol measurements for the entire cohort, n=201
Variables
- : empty
- value: microdialysate cortisol concentration, in nmol/L
- Minutes: time in minutes, relative to wake time
- bins: time bin in which the observation occurred, not used for analysis
- SID: unique anonymous participant identified
File: healthy_cohort_longest_quartile_Cortisol_LME_R_n_51.csv
Description: Subset of cohort with the longest sleep duration (n=51)
Variables
- :
- value: microdialysate cortisol concentration, in nmol/L
- Minutes: time in minutes, relative to wake time
- bins: time bin in which the observation occurred, not used for analysis
- SID: unique anonymous participant identified
File: healthy_cohort_aligned_wake_Cortisol_LME_R_n_135.csv
Description: Subset of cohort with aligned wake time (relative to previous morning wake) n=135
Variables
- :
- value: microdialysate cortisol concentration, in nmol/L
- Minutes: time in minutes, relative to wake time
- bins:
- SID: unique anonymous participant identified
File: healthy_cohort_misaligned_wake_Cortisol_LME_R_n_64.csv
Description:
Variables
- :
- value: microdialysate cortisol concentration, in nmol/L
- Minutes: time in minutes, relative to wake time
- bins:
- SID: unique anonymous participant identified
File: healthy_cohort_shortest_quartile_Cortisol_LME_R_n_51.csv
Description:
Variables
- :
- value: microdialysate cortisol concentration, in nmol/L
- Minutes: time in minutes, relative to wake time
- bins:
- SID: microdialysate cortisol concentration, in nmol/L
Code/software
Included is the R script used to run the mixed linear models described in the paper.
Within the code the version number of R and associated packages is included and reproduced below
platform x86_64-w64-mingw32
arch x86_64
os mingw32
crt ucrt
system x86_64, mingw32
status
major 4
minor 3.1
year 2023
month 06
day 16
svn rev 84548
language R
version.string R version 4.3.1 (2023-06-16 ucrt)
nickname Beagle Scouts
report_0.5.7 nlme_3.1-162 lspline_1.0-0
Access information
Other publicly accessible locations of the data:
*
Data was derived from the following sources:
- UiB Open Research Data https://doi.org/10.18710/5TW8YF.
This dataset contains serial measurements of cortisol collected by ambulatory microdialysis as described in https://doi.org/10.1126/scitranslmed.adg8464
The original data available at https://doi.org/10.18710/5TW8YF have been processed using Python code to realign time series for each participant relative to wake time.
There are 5 CSV files containing the following fields:
"value" : measurements of cortisol in nmol/L
"minutes" : time (in minutes) of the observation relative to wake time
"bins" : the 20 minute time range the observation occured within, placeholder only, not used for analysis
"SID" : subject identifier - anonymised, identifies each unique participant
There is one .R code file:
contains code to run mixed linear models using the CSV files in the repository. Version of R and R packages included in the code file