Dynamics of Nutritional Competence in the Last Year Before Death in a Large Cohort of US Hemodialysis Patients

Xiaoling Ye*, Marijke J. E. Dekker, Franklin W. Maddux, Peter Kotanko, Constantijn J. A. M. Konings, Jochen G. Raimann, Frank M. van der Sande, Len A. Usvyat, Jeroen P. Kooman, Stephan Thijssen

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review


Objectives: Recently, a new Nutritional Competence Score (NCS) has been shown to associate with hospitalization and outcome in hemodialysis patients. The aim of this study was to investigate the dynamics, the individual components, and the impact of hospitalizations of this score's trajectory in the year before death. In addition, we investigated whether dynamics in the NCS add additional independent prognostic value over a single cross-sectional assessment.

Design: We included all Fresenius Medical Care North America patients who initiated hemodialysis between January 1, 2006, and December 31, 2011 with data on all 5 NCS components (serum albumin, creatinine, phosphate, equilibrated normalized protein catabolic rate, and interdialytic weight gain) in at least 1 month. NCS was quantified monthly, and trajectories were compared between non-survivors and survivors across different dialysis vintage strata. Survivors and nonsurvivors were matched by dialysis vintage. The association of baseline NCS and NCS dynamics with mortality risk were assessed with Cox proportional hazards models.

Results: In this cohort of 110,794 patients, we found that across all vintage groups, NCS was lower in patients who died than in survivors. NCS was found to significantly decline before death, whereas survivors showed no decline in NCS. The rate of NCS decline before death was not materially influenced by hospitalization in the months before death. Cox models showed that NCS dynamics over time carry significant predictive power above a cross-sectional NCS assessment.

Conclusions: There are distinct differences in NCS values and their trajectories between patients who die and vintage-matched controls. These differences may be able to be exploited for implementation of a routine, prospective monitoring tool for early detection of patients at increased risk of death. Prospective studies are required to validate such an approach. (C) 2017 by the National Kidney Foundation, Inc. All rights reserved.

Original languageEnglish
Pages (from-to)412-420
Number of pages9
JournalJournal of Renal Nutrition
Issue number6
Publication statusPublished - Nov 2017



Cite this