FC 106HIGHER MAGNESIUM DIALYSATE CONCENTRATION SIGNIFICANTLY IMPROVE SURVIVAL AND CEREBRAL OUTCOME IN HD-PATIENTS WITH ATRIAL FIBRILLATION: LONG-TERM STUDY ON GERMAN NETWORK DATA

Abstract Background and Aims Hemodialysis (HD) patients with atrial fibrillation (AF) are at high risk for cardio-vascular events, severe bleeding and rapid vascular/valvular calcification. Thus, higher than low standard dialysate Mg (d-Mg) may improve outcome by less arrhythmic or calcification impact, but clinical data are missing. Our study evaluated applied d-Mg, risk-factors and antithrombotic therapies on long-term outcome in a large representative German HD-cohort. Method We used pseudonymized benchmarking data (2013-2018) of 16226 adult chronic HD patients (informed consent) from DNeV dialysis network. Diagnoses were coded by “International Classification of Diseases (ICD)” and drugs via “Anatomical Therapeutical Chemical (ATC)” codes. Risk scores (Carlson Comorbidity Index=CCI, CHA2DS2-VASc and HAS-BLED) were tested for de-novo outcome prediction. Results At baseline, 2752 (17%) HD-patients had coded AF. CHA2DS2-VASc (4.0/SD1.5) and HAS-BLED (3.2/0.9) estimated high risk for embolism/bleeding. Standard dialysate-Mg (sd-Mg; 0.5 mmol/L) was used by 1317 (48%), d-Mg 0.75 had 331 (12%), d-Mg >1.0 had 134 (5%) and 970 (35%) patients changed from 0.5 to 0.75 during the study period (change group). Median study time was 2.1 yrs (Range=R: 0.01–6 yrs.). Overall 6-yr mortality was high (63%; Kaplan Meier median survival of 2.9 yrs. Unchanged d-Mg levels were significantly (p<0.02) related to survival: Patients on sd-Mg had lower median survival (2.7 yrs.) than on 0.75 (3.1 yrs; p<0.05) or >1.0 (3.4 yrs; p=0.02). The change group had the same survival (3.1 yrs) as the 0.75 group (p<0.03 vs. 0.5). Cox-Regression (multivariate, sd-Mg=ref.) revealed d-Mg >1.0 (hazard ratio=HR 0.74), d-Mg 0.75 (HR 0.79), serum albumin (HR 0.93), age (HR 1.04) and CCI (HR 1.06) as independently related to mortality (p=0.002). Sd-Mg had higher (p<0.05) cerebral adverse events (5.2%) than 0.75 (1.8%) and >1.0 group (3.7%). Apart from dialysis-related heparin-supply four main approaches regarding anti-coagulation were identified: No therapy, VK-OAC, Heparin or only Aspirin/Clopidogrel (Asp/Clop): VK-OAC and Asp/Clop had same median survival (2.8 yrs) both better (p<0.001) than no therapy (1.3 yrs) or Heparin (1.6 yrs), but VK-OAC had higher bleeding rates (6.4%; p<0.001) than Asp/Clop (3.5%). Cerebral adverse events (3,8% in 6 yrs) were much lower than estimated and similar for all four regimes (R: 3.9-4.4%). Conclusion Use of higher d-Mg in HD-patients with AF significantly improved survival and cerebral outcome, is a feasible cost-effective approach and has more relative impact than well established survival risk-factors such as age, comorbidity (=CCI) and serum albumin. Our data warrant prospective trials comparing higher d-Mg levels with anti-thrombotic drugs and/or left atrial appendage occlusion for better evidence. So far, therapy of HD patients with AF should base on implementation of higher d-Mg, prefer Asp/Clop as best anti-thrombotic drugs and clearly avoid more harmful VK-OAC.