Abstract
INTRODUCTION: The adoption of age or individualized body surface area (i-BSA) estimated glomerular filtration rate (eGFR) thresholds could influence the prevalence and prognosis of chronic kidney disease (CKD). This longitudinal study with up to 15 years of follow-up in the general population, compares different eGFR thresholds for CKD definition: standard, corrected to i-BSA, and age-stratified. For each, we assessed the prevalence of CKD and the combined impact on rapid renal function decline (RRFD) and mortality. METHODS: Patients were classified as CKD according to the presence of significant albuminuria and/or different eGFR thresholds as follows: (i) < 60ml/min per 1.73 m(2); (ii) < 60ml/min corrected to i-BSA; (iii) stratified by age, that is, < 75, < 60 and < 45 ml/min per 1.73 m(2) if aged < 40 years, 40 to 65 years, and > 65 years, respectively. We performed adjusted Cox regression analyses to predict RRFD and global mortality. RESULTS: We analyzed 4952 participants (54% women; mean age: 52 years). Age-stratified definition resulted in 24 of 677 participants aged < 40 years reclassified as CKD, with no adverse outcomes; whereas 55 of 713 participants aged > 65 years were reclassified as non-CKD, with 12 deaths and 1 RRFD. After multivariate adjustment, the CKD group had a poorer prognosis compared with the non-CKD group independently of the definition used; hazard ratio (HR) and 95% confidence interval (CI) were 2.23 (1.59-3.12), 2.06 (1.46-2.90), and 1.64 (1.13-2.38) for the standard, corrected to i-BSA, and age-stratified definitions, respectively. CONCLUSION: In our study, classification of CKD by age or i-BSA does not appear to improve prediction of RRFD and mortality.