Cytokine testing may improve prediction of poor outcomes in COVID-19

Tristan Manalac
26 Oct 2020

Testing for cytokines at the point of diagnosis improves the identification of patients at risk of poor outcomes from the novel coronavirus disease (COVID-19), according to a recent study.

“A combined cytokine panel improves the accuracy of the predictive value for adverse outcome beyond standard clinical data alone,” the researchers said. “Identification of specific cytokines that predict poor outcome … may help stratify patients towards trials of specific immunomodulatory treatments to improve outcomes for these patients.”

In 100 COVID-19 patients (mean age, 61.55±17.28; 61 percent male) enrolled, the mean National Early Warning Score 2 (NEWS2) value was 5.2, indicating a medium risk of deterioration. Inflammatory markers, such as the C-reactive protein (CRP), lactate dehydrogenase (LDH), ferritin, and D-dimer, were all elevated. [Respir Res 2020;21:245]

Nearly half (43.6 percent) of the patients achieved the composite endpoint of intensive care admission, mechanical ventilation, and/or death. Eighteen patients died, yielding a mortality rate of 20.7 percent.

Being aged 70 years or older emerged as a significant risk factor for death in univariable analysis, as did comorbidities, such as hypertension, cardiac disease, and chronic obstructive pulmonary disease, and neutrophil count at admission. NEWS2 was the only significant univariate predictor of the composite outcome.

In the adjusted models, serum cytokine levels, measured at admission, were also significantly correlated with outcomes. The interleukins (IL) 1β (odds ratio [OR], 10.21, 95 percent confidence interval [CI], 1.11–93.37) and 33 (OR, 4.78, 95 percent CI, 1.18–19.30) were both significantly and strongly correlated with the risk of the composite endpoint. Cytokines such as IL-6, IL-8, and tumour necrosis factor (TNF) were also moderately associated with inflammatory markers.

The researchers then sought to optimize predictive models for COVID-19 outcomes. Alone, demography and the NEWS2 score showed acceptable accuracy for identifying adverse clinical outcomes, with an area under the receiver operating characteristic (AUROC) curve of 0.71.

Incorporating blood panel resultswhich measured levels of CRP, LDH, and neutrophils, among others—improved the predictive value, yielding an AUROC of 0.85. Finally, taking cytokine concentrations into account further boosted the accuracy and bumped the AUROC up to 0.91.

Despite the promising results, the researchers pointed out important limitations that need to be taken into account. “Our cohort was a small, single-centre study, and further work should include validation in a larger multicentre trial.”

Future studies should also focus on the respiratory tract, as the primary site of COVID-19 infection, and how the cytokine levels in this area may be of predictive benefit, as well as how they can improve the understanding of the pathophysiology of the disease, they added.

“The results suggest that cytokine analysis at the point of diagnosis has potential to improve clinical care in COVID-19, and shift management of this heterogenous disease towards precision medicine,” the researchers said.

Editor's Recommendations