Cortisol concentrations and mortality from COVID-19

In their report on the association between high serum total cortisol concentrations and mortality from COVID-19, Tan and colleagues¹ appear to have overlooked the pulsatility of cortisol secretion. A recent study by Gibbison and colleagues² showed that both adrenocorticotropic hormone and cortisol are pulsatile in critical illness. The nadir of cortisol pulses in 35% of the patients in their study dropped below the threshold for deficiency in critical illness, as defined by international guidelines, although values shortly before and shortly after were well above this threshold.² On this basis, Gibbison and colleagues argue that the practice of point-testing of cortisol in critical illness should be stopped.² The use of aggregated, static models of serum total cortisol to predict mortality from COVID-19 should proceed with caution. Given the pulsatility, a significantly different serum total cortisol result might have been obtained if a sample was collected an hour earlier or later.²

Tan and colleagues¹ have not reported on plasma adrenocorticotropic hormone concentrations, which would have been crucial in understanding the secretory dynamics of the hypothalamic–pituitary–adrenal axis function in patients with and without COVID-19.² During critical illness, reduced cortisol breakdown, related to suppressed expression and activity of cortisol-metabolising enzymes, contributes to hypercortisolaemia and hence adrenocorticotropic hormone suppression.³

Because more than 90% of circulating cortisol in human serum is protein-bound, changes in the binding proteins can alter measured serum total cortisol concentrations without affecting free concentrations of the hormone.⁴ A study by Hamrahian and colleagues⁴ found that nearly 40% of critically ill patients with hypoproteinaemia had subnormal serum total cortisol concentrations, even though their adrenal function was normal. While serum free cortisol measurement is not routinely available, serum total protein concentrations, or conditions that affect serum proteins (eg, liver disease), should be considered when interpreting serum total cortisol concentrations.⁵

Lastly, the likely positive bias of the cortisol immunoassay (compared with the reference method gas chromatography–mass spectrometry), which has previously been reported in renal and ICU patients,⁶ should be considered. This bias is related to the specificity of the immunoassay.⁶ During critical illness and in renal impairment, changes in cortisol metabolism could lead to an accumulation of other steroids, which, because of similarities to the structure of cortisol, are recognised by the immunoassay.⁶ Differences have been reported in the bias of the immunoassay compared with gas chromatography–mass spectrometry in the serum of patients with sepsis compared with patients routinely attending outpatient clinics.⁶