Our retrospective data analysis evaluated the potential role of biomarkers of inflammation, IL-6, CRP, procalcitonin, D-dimer and fibrinogen in the prediction of severe disease and need for oxygen therapy in residents of LTCF during Covid-19 outbreak. It suggests that IL-6 is a most robust predictor of hypoxemia requiring oxygen therapy. The concentration of IL-6 > 24 pg/mL at the initial assessment is showing the best combination of sensitivity and specificity in predicting the hypoxemia requiring oxygen therapy.
There is a substantial body of evidence linking the IL-6 concentration to the severity of disease and unfavorable outcome of Covid-19 [13, 14, 15, 16, 17, 18, 20, 21, 22]. However, to our knowledge, this study is the first study which evaluated the ability of IL-6 to predict the need for supplementary oxygen administration as a surrogate marker of severe Covid-19 requiring hospital admission in the population of high risk elderly patients. It also suggests clinical implications for intervention procedures for Covid-19 outbreak in LTCF. Han et al. examined the predictive value of various cytokines and concluded that IL-6 is the best predictor of severe Covid-19 . A metanalysis of 9 studies concluded that increased IL-6 is highly associated with severe disease. In this study, patients with severe Covid-19 had mean IL-6 58 pg/mL compared to 17 pg/mL in mild disease . A study by Herold et al. found that IL-6 > 80 pg/mL the predicts respiratory failure and need for mechanical ventilation in Covid-19. Like in our study, IL-6 was superior to CRP in ROC analysis. Chen et al. found cut-off 80 pg/mL of IL-6 differentiates the survivors from the non-survivors . We suggest that we found much lower optimal cut-off value (24 pg/mL compared to 80 pg/mL) because we choose less severe endpoint (need for supplementary oxygen). Overall, we can conclude that our study is in concordance with previous studies and adds new aspects to the known body of evidence.
Clinical significance and implications: The residents of LTCF are one of the most vulnerable populations of the Covid-19 pandemic [8, 9, 10]. The outbreak of Covid-19 in LTCF might significantly burden the local health care system. Timely and effective intervention is essential to reduce morbidity and mortality during such outbreak. According to the proposed guideline by Kim et al., the first phase of response should include broad testing a quick identification of cases and their clinical assessment and triage. In the next phase, monitoring of patients should be implemented in order to quickly identify the patients in need of hospital care . Identification of patients at high risk of deterioration during the initial assessment may significantly improve the monitoring process by allocating the resources to high-risk patients more effectively. Especially in the case of limited human resources, the focus on high-risk patients is might contribute to mortality reduction and improve the overall outcome of the outbreak.
In our study, all patients that developed hypoxemia requiring oxygen therapy during the follow up had the baseline concentration of IL-6 over 24 pg/mL. As a screening tool, it provides excellent sensitivity with an acceptable specificity over 88%. It may be used to identify the patients with a high risk of hypoxemia. It also identifies the patients with low risk for hypoxemia with great negative predictive value. This might help in the decision making for admission during the initial triage. Baseline CRP concentration with cut-off of 24 mg/L showed sensitivity and especially specificity inferior to IL-6, however, might be used as cheaper alternative in low resource setting.
For the purposes of our study, we defined that the residents that developed hypoxemia requiring oxygen therapy are the cases that needed close monitoring and early transfer to the hospital. The rationale is that patients who developed hypoxemia requiring oxygen therapy are those that might benefit from the early pharmacological treatment in order to reduce mortality. According to the body of evidence on remdesivir and dexamethasone, these drugs significantly improves the outcome in the patients needing conventional oxygen therapy [24, 25, 26]. Thus, the patients with early disease but in the high risk of development of hypoxemia are those patients who will profit from close observation and rapid initiation of remdesivir and dexamethasone in case of progression of Covid-19.
IL-6 and development of severe Covid-19: IL-6 is produced by stromal cells and virtually all immune system cells in the lungs and its secretion is stimulated by proinflammatory cytokines, especially interleukin 1β (IL-1β) and tumor necrosis factor α (TNFα). In the early stages of the infection, it is produced by lung macrophages after stimulation of toll-like receptors . An important trait of IL-6 upregulation in Covid-19 is that it precedes the development of acute lung injury that implicates its usability as an early marker of severe disease . However, there is controversy if excessive IL-6 synthesis is true a cornerstone of the pathogenesis of respiratory failure in Covid-19 or is just an epiphenomenon of increased IL-1β and TNFα in the cytokine storm . Predominant theory is that overexpression of IL-6 have a crucial role in the incitement and propagation of the so-called cytokine storm leading to lung injury and ARDS [20, 21]. It is believed that IL-6 increases the permeability of lung capillaries driving the ARDS development and also stimulates the coagulation pathway leading to microthrombi in lung circulation and increases the risk of thrombotic event . A study by Giamarellos-Bourboulis et al. suggests that patients with severe respiratory failure in Covid-19 suffer from distinct types of immune dysregulation which are mediated by IL-6 upregulation. This dysregulation is characterized by high production of proinflammatory cytokines by monocytes and macrophages and CD4 lymphocyte depletion that contributes to the progression of inflammation of lung parenchyma . The direct role of IL-6 in Covid-19 pathogenesis is further supported by findings that IL-6 inhibition improves the prognosis of severe Covid-19 [29, 27].
Other markers of severe disease: In our study, patients who developed hypoxemia had significantly higher serum concentrations of AST, ALT, CRP, serum glucose, creatinine, procalcitonin, and fibrinogen. They also had significantly higher blood count of neutrophils and lower count of lymphocytes and eosinophils. These variables were identified as markers of severe disease by the previous studies . Medians of other well-established markers of serious disease, D-dimer and ferritin were higher in patients who developed hypoxemia, however, the differences were not statistically significant. Using ROC analysis, we concluded that IL-6 is better marker of hypoxemia than CRP or other evaluated variables. Other proinflammatory cytokines, like IL-1β or TNF-α are also associated with severity of Covid-19 and might also serve as biomarkers [30, 32]. However, because the primary goal of our study was to improve the practical algorithm for intervention in LTCF, we focused on examination of biomarkers that are well established in clinical practice and are available for evaluation in most commercial biochemical laboratories in Slovakia.
Prognosis of LTCF residents suffering from Covid-19: According to an epidemiologic study by McMichael et al., during the outbreak in one LTCF in Washington, USA, 54.5% of residents required hospitalization and required hospital admission and 33.7% of infected residents died . In our study, we identified 59 residents with positive swabs for SARS-CoV-2. Of these patients. 32 patients (54%) were admitted to hospital which is similar to the study by McMichael et al. In our study, 13 patients (22%) died. That is less than in the study by McMichael et al., however, this difference of proportion might be attributed to potentially different age and comorbidity status of residents.
Limitations: The limitations of our study are the relatively low sample size and retrospective design. Larger prospective studies are needed to obtain more robust data and to evaluate if the examination of IL-6 during the initial assessment leads to better prognosis of LTCF residents and improves the management of the Covid-19 outbreaks in the LTCFs.
Because of the retrospective design of the study, there might be a concern of bias in the sensitivity of the diagnosis of hypoxemia between groups of hospitalized and outpatient residents. The hospitalized patients were naturally more closely monitored and therefore might be more likely to be diagnosed with hypoxemia. However, we regard this potential bias as insignificant because the outpatient residents were daily monitored for the symptoms and signs of respiratory failure, and patients suffering from dyspnea and patients with tachypnea and/or SpO2 below 90% were transferred to hospital. In normoxic patients, the bias of pulse oximetry comparing to SaO2 is regarding to be insignificant. It reliably identifies the patients with SaO2 below 90% and is a reliable screening tool for hypoxemia with very high negative predictive value .
Another limitation is that we did not evaluate the IL-1β as a possible biomarker of hypoxemia. IL-1β is believed to contribute to the cytokine storm in Covid-19. It is the cytokine which stimulates the IL-6 production and therefore is on the level above the IL-6 in the cascade of cytokine storm . The theory that IL-1β have an important role in the pathogenesis of Covid-19 is widely accepted . However, recent study by Mandel et al found that baseline concentration of IL-1β is not significantly higher in non-survivors . Giamarellos-Bourboulis et al. concluded that immune dysregulation in severe Covid-19 is directly driven by IL-6, not IL-1β upregulation .
Our study included more females than males. This proportion reflects the LTCF population in which female to male ratio was three to one and is not a result of deliberate selection.