Endocrine Abstracts

Serum total cortisol has been linked to increased mortality in patients with Covid-19, but its reliability in critically ill patients is limited. We examined the association between serum free cortisol (SFC) levels and clinical outcomes in patients hospitalized with Covid-19 between 5/5/2020 and 1/3/2021 in our institution.

Methods: SFC was measured in blood samples collected at patient’s admission, prior to any medical treatment. Patients’ files were reviewed retrospectively.

Results: There were 241 patients (78% female), mean (SD) age 67.4 (18.5), of whom 47.3% received dexamethasone treatment (DT). According to the NIH severity index, 46.9% had asymptomatic or mild disease, 17.4% moderate, and 35.7% had severe or critical disease. The in-hospital mortality, 30-day mortality and the need for assisted ventilation were 8.7%, 14.9% and 18.3% respectively. SFC levels were higher in patients who died in hospital [3.74 (2.8) vs 1.4 (0.85) μg/dl, P<0.0001), or within 30 days [3.01 (2.3) vs 1.32 (0.77) μg/dl, P<0.0001] or who required assisted ventilation [2.77 (2.4) vs 1.4 (0.8) μg/dl, P<0.0001]. SFC levels were significantly higher in patients with diabetes, hypertension, cardiovascular disease and chronic renal failure. There was a positive correlation between SFC and IL-6, CRP, ferritin, LDH, D-dimers, neutrophil/lymphocyte ratio (NLR) and a negative correlation with GFR and oxygen saturation at admission (P<0.0001 for all pair comparisons). The area under the ROC curve (AUC) to discriminate 30-day mortality was significantly higher for SFC (0.837) compared with IL-6 (0.733, P=0.012), CRP (0.634, P<0.0001), ferritin (0.618, P<0.0001), LDH (0.667, p-0.001) and NLR (0.759, P=0.039). The AUC to discriminate in-hospital mortality for SFC (0.837) was similar to IL-6 (0.811), LDH (0.790) and ferritin (0.716), but higher than CRP (0.665, P=0.004) and NLR (0.715, P=0.008). The SFC AUC for the need for assisted ventilation was 0.723, not significantly different from the other parameters. Among patients who received DT, 71.1% had severe or critical disease, 17.5% had moderate and 11.4% had mild disease. SFC levels were higher [1.88 (1.5) μg/dl] in patients who received dexamethasone compared with those that did not [1.36 (1) μg/dl, P<0.0001]. Within treated patients, SFC levels were higher in those who died in hospital (3.58 (2.8) vs 1.56 (0.9) μg/dl, P< 0.0001] or within 30 days [3.1(2.6) vs 1.56 (0.9) μg/dl, P<0.0001], compared with those that survived.

Conclusions: SFC levels strongly predict in hospital and 30-day mortality as well as the need for ventilation support in hospitalized patients with Covid-19, irrespective of DT.