Plasma steroid concentrations reflect acute disease severity and normalise during recovery in people hospitalised with COVID‐19
Devine K., Russell CD., Blanco GR., Walker BR., Homer NZM., Denham SG., Simpson JP., Leavy OC., Elneima O., McAuley HJC., Shikotra A., Singapuri A., Sereno M., Saunders RM., Harris VC., Houchen‐Wolloff L., Greening NJ., Lone NI., Thorpe M., Greenhalf W., Chalmers JD., Ho L., Horsley A., Marks M., Raman B., Moore SC., Dunning J., Semple MG., Andrew R., Wain LV., Evans RA., Brightling CE., Kenneth Baillie J., Reynolds RM.
AbstractObjectiveEndocrine systems are disrupted in acute illness, and symptoms reported following coronavirus disease 2019 (COVID‐19) are similar to those found with clinical hormone deficiencies. We hypothesised that people with severe acute COVID‐19 and with post‐COVID symptoms have glucocorticoid and sex hormone deficiencies.Design/PatientsSamples were obtained for analysis from two UK multicentre cohorts during hospitalisation with COVID‐19 (International Severe Acute Respiratory Infection Consortium/World Health Organisation [WHO] Clinical Characterization Protocol for Severe Emerging Infections in the UK study), and at follow‐up 5 months after hospitalisation (Post‐hospitalisation COVID‐19 study).MeasurementsPlasma steroids were quantified by liquid chromatography–mass spectrometry. Steroid concentrations were compared against disease severity (WHO ordinal scale) and validated symptom scores. Data are presented as geometric mean (SD).ResultsIn the acute cohort (n = 239, 66.5% male), plasma cortisol concentration increased with disease severity (cortisol 753.3 [1.6] vs. 429.2 [1.7] nmol/L in fatal vs. least severe, p < .001). In males, testosterone concentrations decreased with severity (testosterone 1.2 [2.2] vs. 6.9 [1.9] nmol/L in fatal vs. least severe, p < .001). In the follow‐up cohort (n = 198, 62.1% male, 68.9% ongoing symptoms, 165 [121–192] days postdischarge), plasma cortisol concentrations (275.6 [1.5] nmol/L) did not differ with in‐hospital severity, perception of recovery, or patient‐reported symptoms. Male testosterone concentrations (12.6 [1.5] nmol/L) were not related to in‐hospital severity, perception of recovery or symptom scores.ConclusionsCirculating glucocorticoids in patients hospitalised with COVID‐19 reflect acute illness, with a marked rise in cortisol and fall in male testosterone. These findings are not observed 5 months from discharge. The lack of association between hormone concentrations and common post‐COVID symptoms suggests steroid insufficiency does not play a causal role in this condition.