Objective. Evaluation of the possibility of using various methods for calculating glomerular filtration rate (GFR) with the determination of the incidence of renal dysfunction depending on the nosological type of socially significant diseases.
Material and methods. The subjects of the study were represented by 728 patients suffering from various socially significant diseases, 330 (45.3%) men and 398 (54.7%) women aged 16 to 98 years (mean age 50.5±14.2 years). Serum creatinine and cystatin C levels were determined. The glomerular filtration rate (GFR) was calculated using the following formulas: CKD-EPI (2011), MDRD (2000), Cokcroft-Gault (1976) and F.J. Hoek (2003). The entire sample was divided into 9 subgroups depending on the nosological type of the disease: 1 – obesity; 2 – arterial hypertension (AH); 3 – coronary artery disease (CAD); 4 – diabetes mellitus (DM); 5 – primary nephropathy (chronic glomerulonephritis and pyelonephritis); 6 – chronic obstructive pulmonary disease (COPD); 7 – cerebrovascular diseases (CVDs); 8 – comorbid diseases, and 9 – general group. A comparative analysis of the frequency of occurrence of renal dysfunction depending on the method of calculating the GFR was performed. The median and interquartile range of GFR were calculated depending on the formula for calculating the GFR in different clinical subgroups. A correlation analysis of the relationship between GFR and blood serum creatinine and cystatin C levels depending on the calculation formula in different clinical subgroups was carried out.
Results. A decrease in GFR in the range of 90–60 ml/min was detected in 29.3% according to the CKD-EPI formula, in 36.6% – according MDRD formula, in 24.3% – according Cokcroft-Gault, and in 71.2% according the method of F.J. Hoek. The largest number of patients with reduced GFR less than 60 ml/min was noted according to the method of F.J. Hoek (48.2%) and MDRD formula (25.1%). Relatively lower GFR value according to F.J. Hoek was recorded in a subgroup of patients with diabetes and primary renal pathology (glomerulonephritis/pyelonephritis). In patients with obesity, AH, COPD, CVDs and in the total sample, there was a significant decrease in GFR according to the method of F.J. Hoek compared to CKD-EPI, MDRD and Cokcroft-Gault formulas. GFR calculated on the basis of serum creatinine level according CKD-EPI formula gave a statistically highly significant relationship in the subgroup of patients with COPD (R=-0.756; P=0.001) and primary nephropathies (R=-0.781; P=0.001). A similar significant strong correlation according to the MDRD equation was observed in the subgroup of patients with COPD (R=-0.852; P=0.001). Compared to other subgroups, the correlation coefficient between serum creatinine level and estimated GFR using the Cokcroft-Gault formula was not as strong among patients with CAD (R=-0.484; P=0.005). GFR calculated using the F.J. Hoek method in all the subgroups gave a significant correlation. The correlation of GFR calculated on the basis of serum cystatin C level was stronger in the subgroup of patients with COPD
(R=-0.935; P=0.001).
Conclusion. The prevalence of a mild decrease in GFR according to the F.J. Hoek method in patients with various socially significant diseases was 71.2 and 36.6% according to the MDRD formula. Moderate reduction in renal function in the examined individuals according F.J. Hoek method was revealed in 48.2%, according to the MDRD equation – in 25.1%. The frequency of occurrence of renal dysfunction according to the CKD-EPI and Cokcroft-Gault formulas was equivalent: 29.3; 21.2% and 24.3; 20.1%, respectively.

Keywords: F.J. Hoek, Cokcroft-Gault, CKD-EPI, estimated glomerular filtration rate, renal dysfunction, MDRD, cystatin C, creatinine