#4404 KIDNEY FUNCTION TRAJECTORY BEFORE AND AFTER ARTERIOVENOUS ACCESS CREATION IN PATIENTS WITH PREDIALYSIS CKD

Abstract Background and Aims Timely arteriovenous (AV) access creation in view of starting hemodialysis is challenged by non-linear kidney function decline the prospect competing mortality. In addition, some studies have shown slower CKD progression following AV patients not on dialysis. While pathophysiological mechanisms, such as ischemic preconditioning improved perfusion, been put forward to explain apparent influence estimated glomerular filtration rate (eGFR) trajectory, it cannot be ruled out that this finding resulted from an artefact induced use models assuming linear before after creation. Our aim was describe kinetics eGFR around period identify different trajectory profiles using relaxing hypothesis. Method From 2013 through 2016, CKD-REIN cohort included 3033 with stages 3 5 40 nationally representative outpatient nephrology clinics France. Participants were followed for years or until initiation replacement therapy (KRT), death, loss follow-up, whichever came first. This study focused who underwent their first during follow-up. Linear mixed restricted cubic spline functions (two internal knots, one at date, other, year before) used model a potential over time, based routine labs. Random effects intercept components allowed us deal individual variations trajectory. Instantaneous rates then extracted. we performed latent class (LCMM) distinct trajectories. Results During median follow-up 5.0 (interquartile range [IQR], 4.6 5.2), 415 (14% total population) (32% women, 51% diabetes). The age 69 (IQR, 61 76), eGFR, 13 ml/min/1.73 m² 11 16). numbers measurements 12 8 19) 2 6) respectively. average creation, constant slopes each period, 5.2 ml/min/1.73m² (95% confidence interval [CI], 4.8 5.5) 3.4 CI, 3.1 3.7), respectively, mean difference −1.8 ml/min/1.73m2 −1.4 −2.1). Analysis instantaneous showed slowdown began 8.3 months 7.8 8.6) LCMM identified two trajectories which mostly differed (Figure). both trajectories, time preceded 9.1 7.2 fastest slowest Conclusion nondialysis patients, appears occur several findings do support causal biological effect progression, but favor alternative hypotheses including optimal management greater inaccuracy estimation advanced due muscle mass loss, simply regression mean.