INTRODUCTION: Chronic kidney disease (CKD) is a significant health problem around the world. The number of patients suffering from this disease is growing, and the treatment of these patients requires significant financial resources. Easily available methods used to detect CKD, such as determining the level of creatinine and nitrogenous substances in the blood, as well as urine examination, indicate the presence of kidney damage relatively late. On the other hand, methods that can indicate the presence of kidney damage in the early course of the disease, such as the clearance methods of radiolabeled compounds, are not easily available and are not used in everyday practice. In order to enable the timely diagnosis of disease, numerous studies have been done to identify biomarkers of early kidney damage. Uromodulin, a protein produced in the thick ascending limb of the loop of Henle and the initial segment of the convoluted part of the distal tubule, is one of the potential biomarkers of early renal damage. Given that tubular function is impaired in the early stage of renal disease, even before visible glomerular dysfunction, uromodulin may represent a promising biomarker of tubular integrity, ie early renal dysfunction. AIM OF THE STUDY: The aim of the research was: a) to compare the serum concentrations of uromodulin between patients with chronic kidney disease and healthy subjects in the control group; b) analyze the relationship between the level of serum concentration of uromodulin and parameters of renal function within the examined group of patients; c) evaluate the importance of determining the serum concentration of uromodulin in assessing the degree of renal function impairment and predicting the progression of chronic kidney disease. MATERIAL AND METHODS: The research was conducted as a prospective cohort study at the Center for Laboratory Medicine of the Clinical Center of Vojvodina, Novi Sad in patients with chronic kidney disease of non-diabetic etiology and a control group of 30 healthy subjects. All subjects with chronic kidney disease were divided into four groups based on stages defined by the glomerular filtration rate (First group – stage I of CKD GFR ���90 ml���min���1.73m2, Second group – stage II of CKD GFR 60-89 ml���min���1.73m2, Third group ��� stage III of CKD divided in IIIa GFR 45-59 ml���min���1.73m2 and IIIb GFR 30-44 ml���min���1.73m2, Fourth group ��� stage IV GFR 15-29 ml/min/1.73m2), with each group consisting of at least 20 subjects. The study was conducted at two time points: baseline and after 24 months. Laboratory testing as well as kidney functional testing were carried out according to the standard methodology of analyzes performed at the Center for Laboratory Medicine of the Clinical Center of Vojvodina. Functional kidney tests were performed using radioisotopic methods of DTPA clearance and hippuran clearance, while the level of serum uromodulin was determined by the ELISA method on the RYTO apparatus, using commercial kits (CLOUD-CLONE CORP.). RESULTS: The average serum uromodulin concentration in subjects with CKD of non-diabetic etiology (48.5簣27.2 ng/ml) was significantly lower compared to the values in the control group of healthy subjects (80.6簣21.5 ng/ml; t= 4.938; p<0.001). A statistically significant difference was also found between the values measured in the control group and subjects with CKD in stage IIIa (34.8簣10.8 ng/ml; p<0.001), IIIb (25.7簣10.6 ng/ml; p <0.001) and IV (21.3簣10.5 ng/ml; p<0.001), while between the control group and subjects with CKD in stage I (72.18簣18.8 ng/ml; p=0.575) , stage II (70.9簣17.8 ng/ml; p=0.373) the difference was not significant. There was no significant difference between the concentrations of serum uromodulin measured in subjects in stages I and II, that is, stages IIIa, IIIb and IV among themselves. A statistically significant strong positive correlation was determined between the serum uromodulin concentration and GFR determined by the radioisotopic method of DTPA clearance (r=0.83; p<0.001), as well as serum concentration of uromodulin and effective renal plasma flow (r=0.71; p< 0.001). A negative correlation was determined between the value of serum uromodulin concentration and serum creatinine concentration (rs= -0.72; p<0.001), serum urea concentration (rs= -0.69; p<0.001), serum uric acid concentration (r= – 0.37; p<0.001) and serum concentrations of cystatin C (r= ���0.64; p<0.001). In the set linear regression model with GFR value as a dependent variable, uromodulin was a statistically significant predictor of GFR value both in first (p<0.001) and second (p<0.001) measurements, and the regression equations in the first and second measurements were almost identical. In the multivariate linear regression model with serum uromodulin concentration values as a dependent variable higher glycemic values (B=-15.939; p=0.003) and lower HDL cholesterol values (B=20.588; p =0.019) were identified as statistically significant predictors of lower serum uromodulin concentration. When assessing the diagnostic accuracy of serum uromodulin concentrations in order to discriminate patients with GFR values below and above 60 ml/min/1.73m2, a cut-off value of 42 ng/ml was determined for a GFR value <60 ml/min/1.73m2 (sensitivity=95 .3%; specificity=93.5%). CONCLUSION: Serum uromodulin concentrations are significantly lower in patients with CKD of non-diabetic etiology, compared to serum uromodulin concentrations in clinically healthy subjects. Serum concentrations of uromodulin are significantly lower in subjects in stage III and IV of CKD compared to subjects in stage I and II of CKD. In patients with CKD of non-diabetic etiology, there is a strong positive correlation between serum uromodulin concentrations and the measured values of GFR and effective renal plasma flow. On the other hand there is a strong negative correlation between serum uromodulin concentrations and serum concentrations of creatinine, i.e. medium negative correlation between serum uromodulin concentrations and serum concentrations of urea, uric acid and cystatin C. In patients with CKD of non-diabetic etiology, there is no statistically significant correlation between serum uromodulin concentrations and albumin and protein concentrations in urine. Statistically significant predictors of lower values of serum uromodulin concentration in this study were lower values of HDL and higher values of glycemia. In patients with CKD of non-diabetic etiology, the serum concentration of uromodulin <42 ng/ml indicates a GFR value <60 ml/min/1.73m2. Serum uromodulin concentration can be used in long-term monitoring of disease progression in patients with CKD of non-diabetic etiology.