Geographical variability of patient characteristics and treatment patterns affect outcomes for incident hemodialysis patients

Background: Geographical differences in disease prevalence and mortality have been described in the general population and in chronic kidney disease patients in Europe. In this secondary analysis of the Membrane Permeability Outcome (MPO) study, we addressed differences in patient and treatment patterns, and whether these affect patient outcomes. Methods: Participating countries were grouped according to geographical location; thus study centers in France, Greece, Italy, Portugal and Spain were allocated to southern Europe (n=499), and those in all other countries (Belgium, Germany, Poland and Sweden) to northern Europe (n=148). Descriptive analysis of patient and treatment patterns at study start, as well as survival analysis, was performed. Results: In patients from the northern European countries, a higher prevalence of diabetes mellitus and of cardiovascular disease was observed than in those from southern Europe (diabetes 35.1% vs. 21.0%, p=0.0007; cardiovascular disease 40.5% vs. 22.8%, p<0.0001). In northern Europe, 23% of patients started hemodialysis with a catheter for vascular access, while in southern European centers, only 13% did so (p=0.0042). Kaplan-Meier survival analysis revealed a lower probability for both all-cause and cardiovascular mortality in southern Europe (log-rank test p<0.001). In a Cox proportional hazards model, a higher mortality risk was estimated for the northern European patients after adjustment for age, sex, membrane permeability, comorbidity index and vascular access (hazard ratio = 1.831; 95% confidence interval, 1.282-2.615; p=0.0009). Conclusions: Our study patients from northern Europe showed a higher risk profile than those from southern Europe. However, only some of the factors can be modified in attempts to lower the mortality risk in this geographical area.