#2906 INTEGRATED KIDNEY MARKERS AND CARDIOVASCULAR MORTALITY: A SURVIVAL ANALYSIS IN UK BIOBANK

Abstract Background and Aims The estimated glomerular filtration rate (eGFR) other kidney function markers are associated with cardiovascular disease (CVD) mortality [1]. It remains unclear whether integrating multiple together can improve CVD risk prediction what would be an appropriate method of integration. In a small general population sample, we recently showed that confirmatory factor analysis (CFA) may predict better than single markers, but it did not outperform cystatin C-based eGFR (eGFRcys) [2]. To assess our findings were context-dependent to which extent they extend assessment, applied CFA exploratory (EFA), five in the UK Biobank (UKBB) study, comparing discrimination for renal failure versus established formulas. Method We analyzed data from 366,758 UKBB participants (mean age 56.6 years; females 53.7%) without clinical history at baseline. Information on participants’ was collected National Health System registry, using ICD-10 codes I00-I99 N17-N19 identify mortality. EFA creatinine-based (eGFRcre), eGFRcys, blood urea nitrogen (BUN), uric acid (UA), serum albumin (Alb). fitted maximum likelihood. Promax rotation then applied. Cox regression models examine associations markers: CFA-based index [CFA]; 1st EFA-based [EFA1]; 2nd [EFA2]; eGFRcre; eGFRcys; creatinine- (eGFRcrecys). Models adjusted sex, age, body mass index, education, self-reported ancestry, hypertension, diabetes, tobacco smoking. receiver operating characteristics (ROC) curve DeLong test used compare discriminatory ability each index. Results standardized loadings (λ) eGFRcre, BUN, UA Alb 0.81, 0.73, -0.55, -0.39, 0.12, respectively. identified two factors: EFA1, largely dependent eGFRcys (λ = 0.85), EFA2, reflecting BUN 0.88), eGFRcre -0.55) 0.36). Over median follow-up 12.5 years, observed 26,327mortality cases, 5,376 45 related failure. hazard ratios (HR) 95% confidence intervals (CI) per standard deviation change 1.22 (1.19–1.25) 3.53 (2.98–4.18) 1.62 (1.57–1.67) 7.34 (5.82–9.27) 1.47 (1.43–1.51) 4.82 (4.04–5.76) eGFRcrecys, 1.38 (1.34–1.41) 2.94 (2.63–3.29) CFA, 1.61 (1.56–1.66) 7.54 (5.90–9.64) 1.33 (1.30–1.35) 1.98 (1.85–2.12) EFA2. area under (AUC) higher EFA1 (0.706, 0.699–0.713) any marker, except (0.709, 0.702–0.716). AUC (0.936, 0.901–0.972) similar (0.939, 0.905–0.972). Conclusion contribution CVD-related studies mainly healthy individuals, is way markers. However, does which, being based simpler calculation, choice prediction.