Dialysis Therapies Hemodialysis Timing, Survival, and Cardiovascular Outcomes in the Hemodialysis (HEMO) Study Yue-Harn Ng, MD, Klemens B. Meyer, MD, John W. Kusek, PhD, Guofen Yan, PhD, Michael V. Rocco, MD, Paul L. Kimmel, MD, Robert L. Benz, MD, Srinivasan Beddhu, MD, Johanna T. Dwyer, DSc, RD, Robert D. Toto, MD, Garabed Eknoyan, MD, and Mark L. Unruh, MD ● Background: The timing of medical therapies has been shown to influence the outcomes and side effects of treatments for disease. This report examines the extent to which hemodialysis treatment time of day was associated with cardiovascular mortality and morbidity and all-cause mortality in a secondary analysis of the Hemodialysis Study. Methods: Dialysis start time defined dialysis shift: morning beginning between 0400 and 0930 hours (n  822); midday, between 0930 and 1530 hours (n  851); and evening, between 1530 and 2200 hours (n  172). Outcome measures included all-cause mortality, cardiac death, composite end point of all-cause mortality or first cardiac hospitalization, and composite end point of first cardiac hospitalization or cardiac death. Results: Morning hemodialysis was associated with a lower likelihood of cardiovascular events compared with the evening shift in all-cause mortality or first cardiac hospitalization (evening versus morning, relative risk [RR], 1.29; 95% confidence interval [CI], 1.01 to 1.65; P  0.043), as well as first cardiac hospitalization or cardiac death (evening versus morning, RR, 1.44; 95% CI, 1.11 to 1.89; P  0.007). No differences were noted in the other 2 outcomes, and there was no statistically significant difference between the morning and midday shifts. Although crude mortality rates were greater in the midday compared with morning (RR, 1.21; 95% CI, 1.05 to 1.39; P  0.008), this association was attenuated after adjustment (RR, 1.04; 95% CI, 0.89 to 1.22; P  0.64). Conclusion: Making extensive adjustment for patient characteristics, this report does not support the association of lower all-cause mortality with morning hemodialysis or a particular benefit for older patients. Am J Kidney Dis 47:614-624. © 2006 by the National Kidney Foundation, Inc. INDEX WORDS: Hemodialysis (HD); outcomes; circadian rhythms; cardiovascular disease. I N THE PAST 2 decades, it was suggested that circadian rhythm may influence the incidence of many diseases, including acute myocardial infarc- tion, sudden cardiac death, and stroke. 1-6 Further- more, the timing of medical therapy was shown to impact on outcomes in those treated for malig- nancies, 7-9 hyperparathyroidism, 10,11 and hyper- tension. 12-15 It is of note that chronic kidney disease is a well-known risk factor for subse- quent cardiovascular events; more than half the deaths in this population are ascribed to cardio- vascular disease. 16-19 In patients with end-stage renal disease under- going hemodialysis, the timing of hemodialysis provides a thrice-weekly interaction of medical therapy with a modifiable risk caused by circa- dian rhythms. Bliwise et al 20 and Abbott et al 21 showed decreased risk for death in older patients with kidney failure undergoing morning hemodi- alysis compared with afternoon initiation of he- modialysis. In the study by Bliwise et al, 20 expo- sure to hemodialysis by time of day was examined in 242 elderly hemodialysis patients in Georgia assessed for hemodialysis treatment times in 1988 and 1991. Although elderly patients under- going hemodialysis in the morning had a 29% lower risk for death, this study was unable to examine differences in dialysis clearance be- tween shifts or effects of timing of hemodialysis on those younger than 60 years. 20 In a larger From the University of Pittsburgh Medical Center, Pitts- burgh; Lankenau Hospital and Medical Research Center, Wynnewood, PA; Tufts-New England Medical Center, Bos- ton, MA; National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD; The Cleveland Clinic Foun- dation, Cleveland, OH; Wake Forest University, Winston- Salem, NC; George Washington University, Washington, DC; University of Utah and the Veterans Affairs Salt Lake City Health Care System, Salt Lake City, UT; and University of Texas Southwestern Medical Center; Baylor College of Medicine, Houston, TX. Received August 26, 2005; accepted in revised form December 12, 2005. Originally published online as doi:10.1053/j.ajkd.2005.12.024 on February 10, 2006. Support: Supported by cooperative agreement U01 DK49271 from the National Institute of Diabetes and Diges- tive and Kidney Diseases; ASN-Hartford-ASP Junior Devel- opment Grant in Geriatric Nephrology and DK-66006 (M.L.U.). Potential conflicts of interest: None. Address reprint requests to Mark Unruh, MD, A909 Scaife Hall, 3550 Terrace St, Pittsburgh, PA 15261. E-mail: unruh@ pitt.edu © 2006 by the National Kidney Foundation, Inc. 0272-6386/06/4704-0006$32.00/0 doi:10.1053/j.ajkd.2005.12.024 American Journal of Kidney Diseases, Vol 47, No 4 (April), 2006: pp 614-624 614