Abstract: FR-PO888

Associations between Dialysate Magnesium, Serum Magnesium, and Mortality: A Retrospective Cohort Study of the Monitoring Dialysis Outcomes (MONDO) Initiative

Session Information

Category: Dialysis

  • 607 Dialysis: Epidemiology, Outcomes, Clinical Trials - Non-Cardiovascular


  • Ye, Xiaoling, Renal Research Institute, New York, New York, United States
  • Usvyat, Len A., Fresenius Medical Care North America, Melrose, Massachusetts, United States
  • Wang, Yuedong, University of California - Santa Barbara, Santa Barbara, California, United States
  • Daugirdas, John T., University of Illinois College of Medicine, Burr Ridge, Illinois, United States
  • Kotanko, Peter, Renal Research Institute, New York, New York, United States
  • Raimann, Jochen G., Renal Research Institute, New York, New York, United States
  • Guinsburg, Adrian M., Fresenius Medical Care, Moron, Argentina
  • Marelli, Cristina, Fresenius Medical Care Argentina, buenos Aires, Argentina
  • Canaud, Bernard J., FMC Deutschland GmbH, Bad Homburg, Germany
  • Stuard, Stefano, Fresenius Medical Care, Moron, Argentina
  • Xu, Xiaoqi, Fresenius Medical Care Asia Pacific, Hong Kong, China
  • Kooman, Jeroen, Maastricht University Medical Centre , Maastricht, Netherlands
  • van der Sande, Frank, Maastricht University Medical Centre , Maastricht, Netherlands
  • Power, Albert J., Richard Bright Renal Unit, Bristol, United Kingdom

Group or Team Name

  • MONDO initiative

Serum magnesium (SMg) associated with mortality and particularly its deficiency substantially increases the risk of adverse outcomes. We studied the relationship between dialysate magnesium (DMg) and SMg and the effects DMg on all-cause mortality in a large global cohort.


All the patients(pts) started in-center hemodialysis(HD) between 2000 and 2012 were included. Following the first available DMg data point we established a 3 months baseline. All the value were average during baseline. Follow-up defined as 1 year after that. A multivariable regression model was applied to study the association of 1.0 versus 0.75 mEq/L DMg on SMg. Then we used 1:1 propensity score matching (age, gender, catheter, and vintage) to create two cohorts with DMg of 1.0 and 1.5 mEq/L, respectively. We compared survival times between these 2 cohorts using KM analysis, log rank-test and Cox regression analysis adjusted for age, gender, and catheter.


We studied 15,211 pts (57.4 yrs, 58% males, 41% DM, 24% catheter; DMg 0.75: 2481 (16%), 1.0: 12,508 (82%) and 1.5: 222 (1%)]. In multivariate regression accounting for age, nPCR, NLR and albumin, a DMg increase by 0.25 mEq/L (from 0.75 to 1.0 mEq/L) was associated with a SMg increase by 0.09 (95%CI 0.03 to 0.14) mEq/L.Propensity score-matching created 2 well balanced cohorts with DMg of 1.5 and 1.0. Uni- and multivariate survival analysis did not show significant differences between the two DMg groups [Figure 1; adjusted HR of DMg 1.5: 1.1 (95% CI 0.6 to 2.0)].


Our results indicate a direct association between DMg and SMg. This finding is of importance, since higher SMg are associated with better outcomes in observational studies.Prospective studies are warranted to further delineate the complex interaction between DMg, SMg, and patient outcomes.