The ratio of serum LL-37 levels to blood leucocyte count correlates with COVID-19 severity | Scientific Reports

Studies have described the beneficial effect of vitamin D on the disease course of COVID-19, although randomized controlled trials are still pending21,22,23. Cathelicidin, the precursor of LL-37, is strongly upregulated by vitamin D24,25. In addition, LL-37 binds with high affinity to the SARS-CoV-2 spike protein, preventing the spike protein from binding to the host cell9. LL-37 also binds to host cell ACE2 and prevents the spike protein from binding to the host cell, thereby preventing virus entry9. Those studies led us to believe that serum LL-37 levels and COVID-19 severity might be related. Therefore, we examined the serum LL-37 levels of COVID-19 patients. We found a decrease in mean LL-37 serum levels from 87 ng/l in group 1 to 57 ng/l in group 4 (Table1). Surprisingly, this difference was not statistically significant and there was no statistically significant correlation between serum LL-37 levels and COVID-19 severity.

A recent study compared serum cathelicidin levels of healthy persons and COVID-19 patients26. It found a significant difference between healthy and SARS-CoV-2 infected persons but no correlation between serum cathelicidin levels and COVID-19 severity. In our study, we examined serum levels of free (unbound) LL-37. LL-37 is known to bind to lipids and apolipoproteins18,19. Because LL-37 is cytotoxic to human cells at higher concentrations, binding of LL-37 to lipids and apolipoproteins may regulate serum levels of LL-37 and help avoid cytotoxicity19,27. In addition, the binding of LL-37 may serve as an LL-37 reservoir for rapid LL-37 accumulation in the event of infection. LL-37 is produced by various cells, such as epithelial cells, and by white blood cells in response to infection and after stimulation with vitamin D28,29. Hence, serum LL-37 levels are also dependent on leukocyte count, and leukocyte count varies during a COVID-19 infection. Therefore, we calculated serum LL-37 levels as a function of leukocyte count. The ratio indicates how much LL-37 is produced per leukocyte. LL-37 is part of the innate immune system, and the release of LL-37 and the LL-37/leukocyte count ratio are indicators of the activity of the innate immune system. We found a highly significant inversely proportional correlation between the LL-37/leukocyte count ratio and the severity of COVID-19. Since blood samples had been collected at the beginning of the inpatient stay, the LL-37/leukocyte count ratio could indicate a greater likelihood of a severe course of COVID-19 early. In our study population, the mean LL-37/leukocyte count ratio was less than 10 in the mechanical ventilated and deceased patients (groups 4 and 5, Table 1). Our results provide evidence that the LL-37/leukocyte count ratio could be used as a marker for early prognosis of the COVID-19 severity.

Because white blood cells are recruited from vessels to extravascular inflamed tissues, local extravascular concentrations of LL-37 may differ from serum LL-37 levels. In addition, LL-37 is released by neutrophil degranulation, and the local LL-37 concentration may be much higher30. During the course of COVID-19, some patients develop severe pneumonia, and leukocytes are recruited to the airways and lungs31. Granulocytes and macrophages locally secrete large amounts of LL-37, and local LL-37 levels in the airways and lungs may differ from serum LL-37 levels. Examination of LL-37 concentrations in the broncho-alveolar lavage fluid (BALF) of COVID-19 patients may reveal a correlation between LL-37 levels and COVID-19 severity. The presence of LL-37 in BALF is mainly due to the degranulation of neutrophil granulocytes. This process means that activated leukocytes are surrounded by a gradient of cathelicidins at the site of infection. Because pneumonia in COVID-19 patients is associated with massive mucus production with leukocytes32,33, it is likely that high concentrations of LL-37 are present in the BALF of COVID-19 patients. As a comparison, the LL-37 concentration of 30 µg/ml in BALF of cystic fibrosis patients is sixfold higher than an LL-37 concentration of 5 µg/ml in BALF of healthy individuals34. In addition, a significant correlation has been found between the concentration of LL-37 in nasal secretions and the severity of acute bronchiolitis in hospitalized infants and young children35.

COVID-19 has a mortality rate that is currently higher in northern latitudes above 35°36. Latitude 35° north is the latitude above which most people do not receive sufficient sunlight in winter to maintain adequate vitamin D levels36. Vitamin D could thus have a beneficial effect on COVID-19 severity. Therefore, we examined the serum vitamin D levels of COVID-19 patients in relation to the severity of COVID-19. We found no correlation between serum vitamin D levels and COVID-19 severity. Some studies support our findings but others do not. A recent study showed that low calcidiol levels in hospitalized patients with COVID-19 were associated with severe disease and increased ICU admission and mortality13. The results of a pilot study in Spain suggest that vitamin D supplementation may reduce disease severity52. In contrast, a retrospective cohort study of 231 patients found that serum calcidiol levels were not significant in predicting mortality in patients with SARS-CoV-237. Another cohort study suggested a similar risk of mortality in COVID-19, independent of vitamin D concentration38. Further interventional studies are needed on the potential beneficial effect of vitamin D on COVID-19 severity.

Vitamin D has been shown to upregulate the CAMP gene encoding LL-37 and stimulate LL-37 production39. This upregulation suggests a positive relationship between serum vitamin D and serum LL-37 levels. Nevertheless, we found no correlation between serum vitamin D and serum LL-37 levels. Again, this finding may be explained by the binding of LL-37 to serum lipids and apolipoproteins, or because we measured free (unbound) serum LL-37 levels.

Measurement of calcidiol, the precursor of the vitamin D hormone (calcitriol), is recommended for assessing vitamin D deficiency. In our study, all 5 patient groups showed a deficiency of calcidiol. In contrast, mean calcitriol concentrations were within the normal range (Table 1). In recent years, vitamin D deficiency has been very present in the media in Germany, and calcitriol intake is common. Because we did not ascertain whether our patients had taken calcitriol supplements before their hospitalization, we cannot exclude a supplemental calcitriol intake as a reason for the calcidiol deficiency and the normal serum calcitriol levels of our study population. However, the discrepancy could arise from inaccuracy in the current reference values for calcidiol and calcitriol. New studies may be needed to determine the validity of the stated reference values.

Studies have already shown an association between age and COVID-19 severity. The older the patient, the higher the risk of a more severe course of COVID-19 and mortality40,41. In general, older patients have a weaker immune system and are more likely to have preexisting conditions than younger patients. Surprisingly, we found no correlation between age and COVID-19 severity (Table 2). However, the mean age of all our groups was comparatively high, ranging between 52 and 68 (Table 1). Potentially, older asymptomatic COVID-19 patients or older COVID-19 patients with mild symptoms were hospitalized out of caution, in contrast to younger COVID-19 patients, who were not concerned about a severe course of COVID-19 and did not go to the hospital.

Determination of inflammatory markers may be helpful in assessing the severity and prognosis of COVID-19. CRP is elevated in acute inflammatory processes, which are common in SARS-CoV-2 infections. We found a positive correlation between serum CRP levels and COVID-19 severity (Table 2). We found the highest average serum CRP levels in patient groups 3, 4, and 5 (Table 1). All patients in those groups required ventilation and some died from COVID-19. In particular, the patients who died, group 5, had the highest serum CRP levels, with a mean serum CRP value of more than 156 mg/l, indicating high-grade systemic inflammation. A recent study showed similar results and that high serum CRP levels were associated with a more severe disease course in COVID-19 patients42. Our results confirm that serum CRP level is an important marker for early detection of severe COVID-19 disease and timely intensive medical intervention.

In our study, COVID-19 severity correlated positively with serum levels of IL-6 (Table 2), and groups 4 and 5 had mean serum IL-6 levels above 150 ng/l (Table 1). Serum IL-6 levels above 150 ng/l are indicative of severe inflammation. A recent study shows that serum IL-6 levels can effectively assess the severity of COVID-19 and predict disease progression in patients43. That finding confirms our findings and highlights the importance of serum IL-6 as an indicator of disease severity. Regular monitoring of inflammatory parameters IL-6 and CRP is recommended44 and clinical studies assessed a possible therapeutic benefit of an IL-6 receptor blockade in COVID-1945,46.

We demonstrated a positive correlation between serum PCT levels and COVID-19 severity (Table 2). A recent study demonstrated higher serum PCT levels in patients who required intensive care47. In our study, the mean serum PCT concentration was less than 2.0 µg/l in all groups (Table 1). Values above 2.0 µg/l indicate systemic bacterial infection. Values between 0.5 and 2.0 µg/l are considered pathological but are in the borderline range where systemic, bacterial infection can neither be reliably diagnosed nor excluded. In another study, PCT concentrations between 2 µg/l and 0.5 µg/l were detected in COVID-19 patients43. Thus, concurrent bacterial involvement in addition to infection with SARS-CoV-2 may occur but does not seem to be the rule.

Pneumonia and pulmonary fibrosis are common findings in severe COVID-19 cases14 and are associated with tissue destruction and cell death. Studies have shown that an increase in serum LDH level is associated with severe COVID-19 and increased mortality14,48. In our study, the mean serum LDH level was above the reference range of 250 U/l in all 5 groups. We found a significant correlation between serum LDH levels and COVID-19 severity. In a previous study, we demonstrated an increase in serum LDH levels in moderate and severe courses of COVID-19 above 400 U/l48. In the current study, only group 4 achieved a comparably high mean serum LDH level. The mean serum LDH levels of the other groups were below 400 U/l. The difference in serum LDH levels between the two studies may be explained by improved treatment options. Treatment with dexamethasone, which has an anti-inflammatory effect, has been shown to reduce the extent of ensued airway and lung inflammation and mortality in COVID-19 patients requiring oxygen supplementation49.

In addition to markers of inflammation, the white blood cell count can also be an important indicator of infections. Leukocytopenia can occur when the consumption of leukocytes to ward off the infection outweighs the new synthesis of leukocytes in the bone marrow. Leukocytosis can be caused by lymphocytosis or granulocytosis as a cytokine-induced immune response to viral or bacterial infection. The COVID-19 patients of our study had both leukocytopenia and leukocytosis. The white blood cell count correlated positively with COVID-19 severity. Our findings are in line with a recent study that found a positive correlation between white blood cell count and disease severity11: Patients with a severe course of COVID-19 had leukocytosis more frequently than patients with a mild course11. Recently, the white blood cell count of 1099 Chinese patients with COVID-19 was investigated and leukopenia was found in 34% of the cases50. These results support our findings that both leukocytopenia and leukocytosis can occur in patients with COVID-19.

In our study, in addition to CRP, IL-6, LDH, and leukocyte count, we identified the ratio of LL-37 to leukocyte count as another prognostic laboratory parameter that correlates with COVID-19 severity. All parameters were determined on days 1–3 after hospitalization. Severity grading was performed after recovery from COVID-19 or patient death. All parameters, CRP, IL-6, LDH, leukocyte count, and LL-37/leukocyte count, can be considered early prognostic parameters. Currently, no decision limits have been defined for the parameters, e.g., limits above which CRP concentration or below which LL-37/leukocyte ratio patients will die. Without decision limits, it is not possible to assess which parameter correctly predicts the course of COVID-19 more often. Therefore, their prognostic significance cannot be compared. Further, each parameter has a different meaning. CRP and IL-6 are inflammatory markers reflecting the extent of the systemic proinflammatory response. Leukocytes play a central role in defense against infection and leukocyte production is generally stimulated during infection. However, it can also decrease due to consumption during the course of infection. LDH, in turn, reflects the extent of cell and tissue damage. LL-37 is part of the innate immune system, and the release of LL-37 and the LL-37/leukocyte count ratio indicate the activity of the innate immune system. Because the parameters indicate different things, we advise considering all the parameters.