Introduction: The stress response to critical illness includes activation of the hypothalamic-pituitary-adrenal (HPA) axis, and increased release of cortisol and diminished or lost physiological diurnal variation. Details of this response were analyzed in consecutive burn patients.
Methods: Forty-nine patients, 15 women and 34 men, median age 41(range 19-60), median burn size 31 % (range 10-96)n and treated at Uppsala Burn Center were investigated during their first week after injury. Clinical parameters were registered daily, serum cortisol concentrations were analyzed four times daily for seven days, and the daily diurnal slope starting from the daily maximum was calculated. Relevant confounding variables were identified by means of a directed acyclic graph (DAG).
Results: The circadian zenith for free cortisol frequently occurred at noon rather that in early morning, it was not related to burn size, ventilator care, or surgery, but was weakly related to the SOFA score. Multilevel modeling revealed that burn size explained cortisol slope. After adjustments for covariates, including the SOFA score, the only significant covariate was the SOFA score itself. Ventilator care explained cortisol slope, but surgery the preceding day did not. Differences between large and small burns were only noticed for the free cortisol concentration at 6pm, suggesting that the flattening of the slope was primarily due to a slower decline in free cortisol from morning to evening. When adjusting for all covariates, slope at 6pm was explained by burn size and P-CgA as well as SOFA score. Both burn size and SOFA score explained the daily coefficient of variability (CV) in free cortisol concentration. Similarly, burn size, P-CgA and SOFA explained AUC.
Conclusions: Free cortisol concentration was related to the size of burns, as was the circadian cortisol rhythm. This effect of burn size was related at least in part to its effect on organ function.