Restless legs syndrome does not affect 3-year mortality in hemodialysis patients

Stefanidis, I., Vainas, A., Giannaki, C.D., Dardiotis, E., Spanoulis, A., Sounidaki, M., Eleftheriadis, T., Liakopoulos, V., Karatzaferi, Christina, Sakkas, Giorgos K., Zintzaras, E. and Hadjigeorgiou, G.M. (2015) Restless legs syndrome does not affect 3-year mortality in hemodialysis patients. Sleep Medicine, 16 (9). pp. 1131-1138. ISSN 13899457

Stefanidis 2015_uremic RLS and mortality sleep med.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (715kB) | Preview


Objective: In chronic hemodialysis (HD) patients, the prevalence of restless legs syndrome (RLS) is significantly higher than in the general population. Uremic RLS has been related to an enhanced mortality of HD patients. In the general population, however, studies of this association have yielded inconsistent results. The aim of the present study was to re-evaluate the relationship of RLS and mortality in HD patients. Methods: We recorded the 3-year mortality rate in a population of 579 HD patients after assessment for RLS symptoms. In addition we analyzed the 5-year mortality rate after end-stage renal disease onset (i.e., commencement of HD). This HD patient population has been previously included in a cross-sectional survey of RLS prevalence in which RLS diagnosis was based on the essential clinical criteria of the International RLS Study Group. Mortality data were acquired from the national endstage renal disease registry. Survival probability was calculated by means of the Kaplan–Meier method and analyzed by the log-rank (Mantel–Cox) test. For multivariate survival analysis, we implemented a proportional hazards regression (Cox regression) model. Results: During the 3-year follow-up period, we documented 118 deaths and an overall 3-year mortality of 20.6%. Mortality was 15.6% in patients with RLS and 22.3% in patients without RLS (p = 0.079). According to the Cox regression analysis,there was no significant association between RLS and 3-year mortality, either in an age- and gender-adjusted model (hazard ratio [HR] = 0.772, 95% confidence interval [CI] = 0.488–1.219, p = 0.267) or in a multivariate adjusted model (HR = 0.667, 95% CI = 0.417–1.069, p = 0.092). Similarly, there was no association between RLS and 5-year mortality after end-stage onset based on our univariate (HR = 0.653, 95% CI = 0.417–1.022, p = 0.063) and multivariate (HR = 0.644, 95% CI = 0.399–1.0380, p = 0.071) analyses. Conclusion: Diagnosis of RLS according to the essential clinical criteria of the International RLS Study Group does not seem to influence the 3-year mortality in HD patients. Our findings are in contrast to those in some previous reports, and reinforce the need for further studies of RLS and mortality in HD patients.

Item Type: Article
Keywords: End-stage renal disease, Epidemiological study, Hemodialysis, Mortality, Restless legs syndrome
Divisions: ?? UniversityCollegePlymouthMarkJohn ??
Depositing User: Users 134 not found.
Date Deposited: 23 Dec 2016 10:52
Last Modified: 17 Feb 2020 15:19

Actions (login required)

Edit Item Edit Item