Abstract
BACKGROUND: The mortality of dialysis patients greatly exceeds that of the general population and identifying predictive factors for mortality may provide opportunities for earlier intervention. This study assessed the influence of sarcopenia on mortality in patients on haemodialysis. METHODS: This prospective, observational study enrolled 77 haemodialysis patients aged 60 years and over, of whom 33 (43%) were female, from two community dialysis centres. Baseline demographic and laboratory data were collected, and sarcopenia was diagnosed using grip strength, muscle mass by bioimpedance analysis (BIA) and muscle function by timed up-and-go according to European Working Group on Sarcopenia in Older People criteria. Nutritional status was assessed using a subjective nutritional assessment score, comprising functional changes in weight, appetite, gastrointestinal symptoms and energy.. A comorbidity score (maximum 7 points) was derived from the presence or absence of hypertension, ischaemic heart disease, vascular disease (cerebrovascular disease, peripheral vascular disease, and abdominal aortic aneurysm), diabetes mellitus, respiratory disease, a history of malignancy and psychiatric disease. Outcomes over six years were linked to the Australian and New Zealand Dialysis and Transplant Registry. RESULTS: The median participant age was 71 years (range 60-87). Probable and confirmed sarcopenia was present in 55.9% and severe sarcopenia with reduced functional testing in 11.7%. Over 6 years, overall mortality was 50 of the 77 patients (65%), principally from cardiovascular events, dialysis withdrawal and infection. There were no significant survival differences between patients with no, probable, confirmed, or severe sarcopenia, or between tertiles of the nutritional assessment score. After adjustment for age, dialysis vintage, mean arterial pressure (MAP) and the total comorbidity score, no sarcopenia category predicted mortality. However, the total comorbidity score [Hazard Ratio (HR) 1.27, Confidence Intervals (CI) 1.02, 1.58, p = 0.03] and MAP (HR 0.96, CI 0.94, 0.99, P = < 0.01) predicted mortality. CONCLUSION: Sarcopenia is highly prevalent in elderly haemodialysis patients but is not an independent predictor of mortality. Haemodialysis patients have multiple competing risks for mortality which, in this study, was predicted by a lower MAP and a higher total comorbidity score. TRIAL REGISTRATION: Recruitment commenced December 2011. The study was registered 10.01.2012 with the Australian New Zealand Clinical Trials Registry (ACTRN12612000048886).