Fax +41 61 306 12 34 E-Mail karger@karger.ch www.karger.com Original Paper Nephron Clin Pract 2009;112:c268–c275 DOI: 10.1159/000224794 The Impact of Type 2 Diabetes on Mortality in End-Stage Renal Disease Patients Differs between Genders A. Karamé a, b M. Labeeuw a, b P. Trolliet a A. Caillette-Beaudoin c R. Cahen a R. Ecochard d, e R. Galland c P. Hallonet c C. Pouteil-Noble a, b E. Villar a a Nephrology and Renal Transplantation, Lyon Sud Hospital, Hospices Civils de Lyon, Pierre-Bénite, b Université de Lyon-1, Villeurbanne, c Centre Associatif Lyonnais de Dialyse, Lucien Hussel Hospital, Vienne, and d Department of Biostatistics, Hospices Civils de Lyon, and e UMR CNRS 5558, Université de Lyon-1, Lyon, France the first year. Evolutions with time of these aHR by gender were significantly different (p = 0.009). Conclusions: T2DM was associated with death only in females. This association was not constant over time after the first dialysis. Copyright © 2009 S. Karger AG, Basel Introduction In general and end-stage renal disease (ESRD) popula- tions, type 2 diabetes mellitus (T2DM) is associated with high mortality rates [1–6]. Over the last few decades, a dramatic increase in T2DM prevalence was observed worldwide among ESRD patients [4–8]. In Europe [4–6] and in Australia/New Zealand [7] , more than one quarter of incident ESRD patients had associated T2DM. In the Key Words Dialysis  End-stage renal disease  Type 2 diabetes mellitus  Survival, gender Abstract Background/Aims: In diabetics with end-stage renal dis- ease (ESRD), risk of death has been reported to be non-con- stant after the first dialysis, and different outcomes have been observed between genders. We assessed the impact of type 2 diabetes (T2DM) on mortality in dialysis regarding its differential effect by gender using time-dependent anal- yses. Methods: All T2DM and non-diabetic (no-DM) patients who started dialysis in two renal units in Lyon, France, be- tween January 1, 1995, and December 31, 2007, were includ- ed. In multivariate analyses, the Cox model and Shoenfeld residual approach were used to assess the effect of T2DM on dialysis mortality by gender. Results: We included 235 T2DM (males: 57.9%) and 480 no-DM (males: 65.6%) patients. In males, the adjusted hazard ratio (aHR) for death in T2DM ver- sus no-DM was 0.83 (p = 0.20) and was constant over time after the first renal replacement therapy (RRT) (p = 0.88). In females, aHR for death in T2DM versus no-DM patients was not constant over time (p = 0.002). It was 0.64 (p = 0.13) with- in the first year after the first RRT and 2.10 (p = 0.002) after Received: November 11, 2008 Accepted: January 23, 2009 Published online: June 16, 2009 Emmanuel Villar, MD, PhD Service de Néphrologie, Dialyse et Transplantation Rénale Centre Hospitalier Lyon-Sud, 165, chemin du Grand-Revoyet FR–69495 Pierre-Bénite Cedex (France) Tel. +33 4 7267 8702, Fax +33 4 7267 8710, E-Mail emmanuel.villar@chu-lyon.fr © 2009 S. Karger AG, Basel 1660–2110/09/1124–0268$26.00/0 Accessible online at: www.karger.com/nec The interpretation of reported data is the responsibility of the au- thors and in no way should be seen as an official interpretation of the Renal Epidemiology and Information Network Registry. The re- sults presented in this paper have not been published previously as a whole or in part, except in abstract form at the 9th Congress of the Société de Néphrologie and the Société Francophone de Dialyse in Lyon, France, 2007 (abstract O53), and at the 45th ERA-EDTA Con- gress in Stockholm, Sweden, 2008 (abstracts MP431 and MP433).