POS0626 CREATININE-BASED CKD-EPI EQUATION LEADS TO OVERESTIMATION OF GLOMERULAR FILTRATION RATE IN PATIENTS WITH RHEUMATOID ARTHRITIS

Abstract

BackgroundAccurate calculation of glomerular filtration rate (GFR) is an important aspect of clinical care for rheumatologic patients [1, 2, 3].ObjectivesTo choose the optimal method of determining the GFR to assess the severity of renal dysfunction in patients with rheumatoid arthritis (RA).MethodsAn open cross-sectional study was performed involving 96 patients with a reliable diagnosis of RA (mean age 54.4±11.6 years, duration of disease 10.7±8.56 years, 57.3% with moderate RA activity).For comparative assessment of renal function, we used the estimated glomerular filtration rate (eGFR) using the CKD-EPI formulas on the National Kidney Foundation website (USA): creatinine-based equation (eGFRcr), cystatin C-based equation (eGFRcyst) and calculated creatinine and cystatin C equation (eGFRcr-cyst). The 2009 CKD-EPI creatinine equation (eGFRcr) was used as a reference for comparative calculations of GFR. Based on the eGFRcr measurements, patients with RA were divided into four groups: I, >90 ml/min/1.73 m2; II, 89-60 ml/min/1.73 m2; III, 45-59 ml/min/1.73 m2; and IV, <45 ml/min/1.73 m2.ResultsThe mean eGFRcr, presented as a reference in this study, was 70.0±18.7 ml/min/1.73 m2. Signs of hyperfiltration using eGFRcr (>90 ml/min/1.73 m2) were observed in 16 (10.1%) patients with RA, mild decrease of renal function (60-89 ml/min/1.73 m2) was registered in 52 (32.9%), moderate/severe decrease (<59 ml/min/1.73 m2) - in 28 (17.8%) patients with RA. Decreased eGFRcr was differentially associated with increased patient age (r=0.46; p=0.003), disease duration (r=0.24; p=0.017), cumulative dose of hormones (r=0.66; p=0.007), lower height (r= 0.35; p=0.001).Analysis of eGFR values demonstrated significant differences using the selected methods (χ2=9.91, p= 0.007). Intergroup differences (in degree of decrease in eGFR) were statistically significant for all variants of eGFR calculation (eGFRcr, eGFRcyst, eGFRcr-cyst; H-test and Median-test, p<0.001). According to the eGFRcr formula, an absolute majority of patients with RA (83.3%) had a decrease in GFR of varying severity (a slight decrease was registered in 54.2% of cases) (Table 1).Table 1.Group distribution of RA patients according to eGFR, n(%)Group IGroup IIGroup IIIGroup IVeGFRcr16 (17.0)52 (54.2)20 (20.8)8 (8.33)eGFRcr-cys22 (22.9)39 (40.6)20 (20.8)15 (15.6)eGFRcyst29 (30.2)29 (30.2)22 (22.9)16 (16.7)The use of eGFRcyst showed that only 12 of 29 people in the first group had optimal (>90 ml/min/1.73 m2) eGFRcr (p=0.031), and 17 patients entered the group with slightly decreased (60-89 ml/min/1.73 m2) eGFRcr. A similar, but less significant situation (12 of 22 people and 10 patients, p=0.02) was also observed with eGFRcr-cyst. In the second group of RA patients 19 patients corresponded to the chosen criteria in determination of eGFRcyst, and 8 patients entered groups (III and IV) with more severe decrease of renal function (2 patients were included into group I) (p=0,011). Significant differences in this group were also noted when comparing the proportions according to eGFRcr with eGFRcr-cyst (p=0.044). A probable decrease in renal filtration function with eGFRcr (compared with the alternative use of eGFRcr-cyst or eGFRcyst) can be observed in 11-18% of RA patients in group 1 (high/optimal renal function) and up to 10% of RA patients in group 2 (slight decrease). No significant differences were found using the three estimated CKD-EPI formulas in RA patients with moderate/significant decrease in GFR (p>0.05).ConclusionCurrently, the overall diagnostic performance of the CKD-EPI equation based on creatinine and cystatin C may be the most optimal (in comparison with other calculated CKD-EPI formulas) in patients with RA, and may also be useful for confirming eGFRcr results >60 ml/min/1.73m2.References[1]Saisho K. et al. Mod Rheumatol. 2016;26(3):331-5.[2]Couderc M. et al. Arthritis Care Res (Hoboken). 2016;68(5):638-44.[3]Aleksandrov V. et al. Ann Rheum Dis. 2021;80(S1):1059. doi:10.1136/annrheumdis-2021-eular.2275.Disclosure of InterestsNone declared