P1437RELATIONS BETWEEN IMPLEMENTATIONS OF EVIDENCE-BASED TREATMENTS AND IMPROVED OUTCOMES IN PATIENTS STARTING HEMODIALYSIS DURING THE LAST 10 YEARS: EXPERIENCES FROM THE SWEDISH RENAL REGISTRY 2006-2015

Abstract

Abstract Background and Aims Although few randomized controlled trials have been able to demonstrate benefits of single treatments in improving outcomes of patients undergoing hemodialysis (HD), the last decades have witnessed significant therapeutic advances. This relates both to the HD treatment itself and to the management of patient complications. We here evaluated temporal changes in the use of evidence-based treatments and survival rates in incident HD patients. Method We included all patients initiating HD in Sweden between 2006-2015 and followed them until the end of 2017. Data were linked to national registries to retrieve information on date of death, cardiovascular events, comorbidity, and drug prescriptions. We first evaluated trends in the use of different HD related therapies and selected key therapeutic targets and medications in 2-year blocks. We then evaluated the incidence of death and major cardiovascular events (MACE) within one and two years from start of dialysis through standardized incidence rates via logistic regression models to account for differences in patient characteristics over time. Via Cox regression models, we explored whether adjustment for implementation of evidence-based treatments (e.g. hemodiafiltration (HDF), frequent HD, working fistula, drugs to control CKD-MBD, anemia) modified the observed survival and MACE risks. Finally, survival trends were also compared against an age-sex-calendar year-matched general Swedish population using standardized incidence ratios. Results We identified 6,612 patients starting HD in Sweden during the study period. There was no difference in mean age or proportion of women over time, but body mass index, serum parathyroid hormone levels and the proportion of patients with cerebrovascular disease, atrial fibrillation and cancer increased. Conversely, mean serum hemoglobin, phosphate and albumin values decreased over time. The proportion of patients who underwent HDF, had more than three HD sessions/week, and had a working fistula increased progressively, as well as the use of phosphate binders (particularly non-calcium), cinacalcet, and vitamin D3. After standardization for differences in demography and comorbidities, the one-year risk of mortality or MACE risk decreased by 7% and 16%, respectively, in 2014/15 compared to 2006/07. Similarly, the two-year risk of death or MACE decreased by 16 and 21%. In multivariable Cox models, we explored the linear association between calendar year blocks and study outcomes. Per 2-year period, the risk of death within one year decreased by 6% (HR 0.94, 95% CI 0.92-1.00), and of MACE by 5% (0.94, 95% CI 0.92-0.98). Adjustment for changes in the evidence-based treatments over time abrogated these associations (HR 1.00, 95% CI 0.91-1.09 for death and 1.00, 95% CI 0.94-1.06 for MACE). Similar results were obtained for 2-year outcomes. Compared with the general population, the one-year risk of death for a HD patient was 6 times higher in 2006/2007 [standardized incidence rate ratio, SIR 6.05 (5.30–6.91)], but decreased to 5.5 times higher in 2014/15 [SIR 5.57 (4.82–6.44)], corresponding to a SIR reduction of 8%. Conclusion In patients initiating HD therapy in Sweden, there has been a gradual implementation of new and established evidence-based treatments during the last 10 years, which was associated with a parallel reduction in the risk of death and MACE.