#2120 Impact on CKD classification and comorbidities using GFR estimated by European Kidney Function Consortium equation in kidney transplant recipients

Abstract

Abstract Background and Aims In 2020, European Kidney Function Consortium (EKFC) proposed a new equation that estimates glomerular filtration rate (GFR) in European population. Few studies have compared EKFC with CKDEPI, particularly in kidney transplant (KT) recipients. Chronic Kidney Disease (CKD) classification by KDIGO is used not only by nephrologists who evaluate more parameters of kidney function, but also by every other specialist who are going to be guided mainly by this stage. Our objective is to compare EKFC equation and CKDEPI in classifying CKD in KT patients and its relationship with associated comorbidities. Method We conduct a retrospective observational study including all KT performed at our centre from 1993 to 2022 with survival greater than one year. We collect creatinine, haemoglobin (Hb), phosphorus (P−), potassium (K+), and bicarbonate (Bic) values at three months and one-year post-transplant. We define anaemia as Hb < 11 g/dl, hyperphosphataemia as serum P− > 4.5 mg/dl, hyperkalaemia as serum K+ > 5 mEq/ml, and acidosis as Bic < 22 mEq/L. We calculate estimated GFR (eGFR) using CKDEPI and EKFC equations, classify CKD stage according to KDIGO, and compare both results in terms of correlation and differences. We use Kappa Coefficient and Chi-Square by contingency tables, and ROC curves with homogeneity test for prediction of these comorbidities. We use SPSS and MedCalc as statistics programs. Results We include 968 KT, with median age at the time of transplant 52.75 year [41.25-61.27]. 294 KT (30.4%) are in women. Comorbidity frequencies are depicted in Table 1. ROC analysis shows no significant differences in predicting acidosis at one year, hyperphosphataemia at any time, hyperkalaemia at 3rd month, anaemia at any time, and composite event at any time. We find significant differences in favour to CKDEPI for acidosis at 3rd month (ΔAUC 0.00575 95% CI [0.00182-0.00969]; p = 0.0042) and for hyperkalaemia at one year (ΔAUC = 0.00853 [0.00161-0.0154]; p = 0.0157). Through contingency tables we compare classification into stages, finding very good agreement based on the Kappa coefficient (κ) but significant differences by Chi-Square. With κ = 0.905, EKFC reclassifies 70 patients, 7.23% overall, but not significantly down nor upstaging (31 vs 39). It reclassifies more men than women: 8.46% vs 4.42%. In 10-year age groups, the largest reclassification occurs in the 70-80 years group, a 17.07% with κ = 0.774, all cases upstaged. At one-year eGFR, contingency tables also show a very good agreement based on κ but significant differences by Chi-Square. EKFC reclassifies 83 KT (κ = 0.887), 8.57% overall. It reclassifies 7.48% women and 9.05% men. In 10-year age groups, the largest reclassification occurs in 20-30 years group with 17.24% (κ = 0.767) and 14.81% in 70-80 years (κ = 0.796). All patients in these groups were upstaged. We resume graphically the reclassification of the groups in Fig. 1, observing how EKFC upstages CKD (worsen CKD stage) in high eGFR stages, and downstages CKD in low eGFR stages. Conclusion Estimation of GFR in KT recipients using the new EKFC equation yields similar results to CKDEPI, showing very good agreement for CKD classification although there is a non-negligible percentage of reclassification, which may be clinically relevant in terms of treatment indications and diagnostic tests. It is to a worse CKD stage in high eGFR and to a better stage in low eGFR, and when EKFC reclassification percentage is greater—in the extreme ages of our cohort—it is always by upstaging. Except for minimal differences which we consider clinically inconsequential, we find no significant differences between both equations in discriminating presence of hyperphosphatemia, hyperkalaemia, anaemia or acidosis.