Cytokines as predictors of COVID-19 severity: evidence from a meta-analysis

To the editor

With great interest, we read the articles by Kosmaczewska et al¹ and Marcinkiewicz et al,² showing the potential role of cytokines, including interleukin (IL) 6, as predictors of coronavirus disease 2019 (COVID-19) severity. In most cases of COVID-19, the course of the disease is asymptomatic, or the symptoms are so mild that the patient does not require hospitalization. However, some patients experience a sudden deterioration in health. This phenomenon is called a cytokine storm and is caused by an abnormal overresponse of the immune system. Cytokine storms are common complications of COVID-19, influenza, and other coronavirus-induced respiratory diseases such as severe acute respiratory syndrome and Middle East respiratory syndrome. During the course of the disease there is a rapid release of cytokines, that is, proteins which stimulate other immune system cells to specific reactions. Cytokines, including ILs, coordinate the body’s response to infection and cause inflammation. Interleukin 6 exerts a multidirectional effect on the cells of the innate and gained immune systems. It plays a key role in initiating and developing an acute inflammatory response by activating cells via the classical route. It also facilitates the development of gained response and directs its course. As shown by Musselman et al,³ an increase in the concentration of IL-6 and other proinflammatory cytokines also affects the central nervous system, contributing to the occurrence or intensification of symptoms characteristic of both neuropsychological and somatic depression.

To confirm the effect of cytokines as predictors of COVID-19 severity, we performed a meta-analysis following the Meta-analysis of Observational Studies in Epidemiology (MOOSE) guidelines for reporting systematic reviews and meta-analyses of observational studies.⁴ The detailed methodology of the study is described in Supplementary material.

Twenty-four studies reported IL-6 levels in the severe and nonsevere COVID-19. Pooled analysis showed significantly higher IL-6 concentrations in patients with severe disease compared with the nonsevere group (mean difference, 21.9; 95% CI, 17.64–26.16; P <⁠0.001; Figure 1). The analysis also showed that higher levels of tumor necrosis factor, IL-2R, IL-4, IL-8, and IL-10 were associated with a more severe course of COVID-19 (Supplementary material). Moreover, the pooled analysis revealed that IL-6 and IL-10 concentrations were higher in patients admitted to intensive care units than in patients hospitalized in other units. Additional analysis showed that significantly higher levels of tumor necrosis factor, IL-2R, IL-6, IL-8, and IL-10 were observed among those who died from COVID-19 than among those who survived.

In summary, this meta-analysis is the most up-to-date presentation of the use of ILs as predictors of COVID-19 severity. Simultaneously, the obtained results show that cytokines (including ILs) can be used as independent predictors of the severity of a patient’s condition.