MARKERS FOR PREDICTING THE MANIFESTATION OF GLOMERULONEPHRITIS IN PATIENTS AFTER COVID

Abstract

Recently, there have been reports in the scientifi c- medical literature about the occurrence of renal pathology after COVID-19 of moderate severity, detected 1 year after this viral infection. Often such changes are detected for the fi rst time. It is now known that after a severe course of COVID-19, interstitial lesions are found in the kidneys, and glomerular changes are added over time. Today, there is no clear answer as to which changes predominate in patients with moderate COVID-19. Of great interest is the probability of glomerulonephritis manifestation 1 year after moderate COVID-19.  The aim of the study was to identify a possible relationship between moderate COVID-19 and the manifestation of glomerulonephritis. Materials and methods. We studied 25 patients with stage I-II chronic glomerulonephritis (GN) who had moderate COVID-19 1 year ago, 27 patients who had moderate COVID 1 year ago without the onset of glomerulonephritis, and 20 healthy people with no history of COVID-19. Cluster and classifi cation methods of data mining, enzyme- linked immunosorbent assay were used to study the levels of IL-1 β, IL-6, IL-10 in blood and urine. We investigated possible relationships between moderate- severe COVID-19 and the manifestation of glomerulonephritis. For this purpose, three diff erent clustering methods were used, which are based on completely diff erent mathematical theories: the classical method of k-means, Kohonen maps (neural networks), fuzzy k-means (fuzzy logic). The research was carried out within the framework of the National Research Council of the Department of Internal Medicine of Bukovyna State Medical University «Clinical- pathogenetic justifi cation of diff erentiated treatment of patients with combined pathology of internal organs» (0122U002209 02.2022-12.2026).Results. A signifi cant increase in the content of IL-1β (p<0.05), IL-6 (p<0.05) and a decrease in the content of IL-10 in blood and urine (p<0.05) were found in patients with chronic glomerulonephritis who were sick with COVID-19 of moderate severity 1 year ago (p<0. 05), and a decrease in the level of IL-10 only in blood in patients who had COVID-19 of moderate severity 1 year ago without further development of glomerulonephritis (p<0.05). The clustering technique confi rmed the association between moderate COVID-19 and glomerulonephritis. Signifi cant indicators were found: IL-1β in urine, IL-6 in urine, IL-6 in blood. The main factor that characterizes the patient’s belonging to the healthy cluster is urine IL-1β with the same threshold –4.953≤ urine IL-1β ≤ 9.431, but the factor that determines the belonging to the disease type is urine IL-6 factor (urine IL-6> 0.994). Such an analysis gives reason to believe thatCOVID-19 of moderate severity may have signs of a factor causing the manifestation of glomerulonephritis. Conclusion. In clinical practice, the method of clustering and classifi cation analysis can be used to determine the main factors for diagnosis and prognosis (urine IL-1β, urine IL-6, blood IL-6). This diagnostic method, which diff ers from the usual regression methods, is based on the application of clustering and classifi cation methods of data mining and allows to establish the relationship between the diagnosis of COVID-19 of moderate severity and glomerulonephritis manifesting after 1 year.