Duk-Hee Kang1,2, Yuji Lee1,3, Carola Ellen Kleine1, Yong Kyu Lee1,4, Christina Park1, Jui-Ting Hsiung1, Connie M Rhee1, Csaba P Kovesdy5,6, Kamyar Kalantar-Zadeh1,7, Elani Streja1,7. 1. Harold Simmons Center for Kidney Disease Research and Epidemiology, School of Medicine, University of California Irvine, Orange, CA, USA. 2. Division of Nephrology, Department of Internal Medicine, Ewha Medical Research Center, Ewha Womans University College of Medicine, Seoul, South Korea. 3. Division of Nephrology, Department of Medicine, Samsung Changwon Hospital, Sungkyunkwan University School of Medicine, Changwon, South Korea. 4. Nephrology Division, Department of Internal Medicine, NHIS Ilsan Hospital, Goyang, South Korea. 5. Nephrology Section, Memphis Veterans Affairs Medical Center, Memphis, TN, USA. 6. Division of Nephrology, University of Tennessee Health Science Center, Memphis, TN, USA. 7. Nephrology Section, Tibor Rubin VA Medical Center, Long Beach, CA, USA.
Abstract
BACKGROUND: Eosinophils are traditionally known as moderators of allergic reactions; however, they have now emerged as one of the principal immune-regulating cells as well as predictors of vascular disease and mortality in the general population. Although eosinophilia has been demonstrated in hemodialysis (HD) patients, associations of eosinophil count (EOC) and its changes with mortality in HD patients are still unknown. METHODS: In 107 506 incident HD patients treated by a large dialysis organization during 2007-11, we examined the relationships of baseline and time-varying EOC and its changes (ΔEOC) over the first 3 months with all-cause mortality using Cox proportional hazards models with three levels of hierarchical adjustment. RESULTS: Baseline median EOC was 231 (interquartile range 155-339) cells/μL and eosinophilia (>350 cells/μL) was observed in 23.4% of patients. There was a gradual increase in EOC over time after HD initiation with a median ΔEOC of 5.1 (IQR -53-199) cells/μL, which did not parallel the changes in white blood cell count. In fully adjusted models, mortality risk was highest in subjects with lower baseline and time-varying EOC (<100 cells/μL) and was also slightly higher in patients with higher levels (≥550 cells/μL), resulting in a reverse J-shaped relationship. The relationship of ΔEOC with all-cause mortality risk was also a reverse J-shape where both an increase and decrease exhibited a higher mortality risk. CONCLUSIONS: Both lower and higher EOCs and changes in EOC over the first 3 months after HD initiation were associated with higher all-cause mortality in incident HD patients. Published by Oxford University Press on behalf of ERA-EDTA 2020. This work is written by US Government employees and is in the public domain in the US.
BACKGROUND: Eosinophils are traditionally known as moderators of allergic reactions; however, they have now emerged as one of the principal immune-regulating cells as well as predictors of vascular disease and mortality in the general population. Although eosinophilia has been demonstrated in hemodialysis (HD) patients, associations of eosinophil count (EOC) and its changes with mortality in HDpatients are still unknown. METHODS: In 107 506 incident HDpatients treated by a large dialysis organization during 2007-11, we examined the relationships of baseline and time-varying EOC and its changes (ΔEOC) over the first 3 months with all-cause mortality using Cox proportional hazards models with three levels of hierarchical adjustment. RESULTS: Baseline median EOC was 231 (interquartile range 155-339) cells/μL and eosinophilia (>350 cells/μL) was observed in 23.4% of patients. There was a gradual increase in EOC over time after HD initiation with a median ΔEOC of 5.1 (IQR -53-199) cells/μL, which did not parallel the changes in white blood cell count. In fully adjusted models, mortality risk was highest in subjects with lower baseline and time-varying EOC (<100 cells/μL) and was also slightly higher in patients with higher levels (≥550 cells/μL), resulting in a reverse J-shaped relationship. The relationship of ΔEOC with all-cause mortality risk was also a reverse J-shape where both an increase and decrease exhibited a higher mortality risk. CONCLUSIONS: Both lower and higher EOCs and changes in EOC over the first 3 months after HD initiation were associated with higher all-cause mortality in incident HDpatients. Published by Oxford University Press on behalf of ERA-EDTA 2020. This work is written by US Government employees and is in the public domain in the US.
Authors: Yoshitsugu Obi; Connie M Rhee; Anna T Mathew; Gaurang Shah; Elani Streja; Steven M Brunelli; Csaba P Kovesdy; Rajnish Mehrotra; Kamyar Kalantar-Zadeh Journal: J Am Soc Nephrol Date: 2016-05-11 Impact factor: 10.121
Authors: Charles J Diskin; Thomas J Stokes; Linda M Dansby; Lautrec Radcliff; Thomas B Carter Journal: Nephrol Dial Transplant Date: 2011-01-14 Impact factor: 5.992
Authors: Sarah B Withers; Ruth Forman; Selene Meza-Perez; Daniel Sorobetea; Kasia Sitnik; Thomas Hopwood; Catherine B Lawrence; William W Agace; Kathryn J Else; Anthony M Heagerty; Marcus Svensson-Frej; Sheena M Cruickshank Journal: Sci Rep Date: 2017-03-17 Impact factor: 4.379