Background: Effects of dialyzer membrane characteristics on morbidity and mortality are highly controversial.
Methods: Post hoc, we analyzed data from the German Diabetes and Dialysis Study that evaluated atorvastatin in high-risk patients. Four groups were identified being constantly dialyzed with high-flux synthetic (n = 241), low-flux synthetic (n = 247), low-flux semisynthetic (n = 119), or cellulosic low-flux membranes (n = 41). Two end points were investigated: (1) a cardiovascular end point consisting of cardiac death, nonfatal myocardial infarction, and stroke and (2) death.
Results: After 4 years of follow-up, adjusted multivariate relative risks (RRs) were calculated. The RR to reach a cardiovascular end point was greater for patients dialyzed with cellulosic low-flux (RR, 2.33; 95% confidence interval [CI], 1.38 to 3.94; P = 0.002), low-flux semisynthetic (RR, 1.92; 95% CI, 1.35 to 2.73; P = 0.0003), or low-flux synthetic membranes (RR, 1.35; 95% CI, 0.99 to 1.85; P = 0.06) than for those treated with high-flux synthetic dialyzers. The likelihood to die was greater with cellulosic low-flux (RR, 4.14; 95% CI, 2.79 to 6.15; P < 0.0001), low-flux semisynthetic (RR, 2.24; 95% CI, 1.66 to 3.02; P < 0.0001), and low-flux synthetic membranes (RR, 1.59; 95% CI, 1.22 to 2.07; P = 0.0006) than with high-flux synthetic membranes. With respect to low-flux synthetic membranes, RRs of mortality for patients using cellulosic low-flux and low-flux semisynthetic membranes were 161% (RR, 2.61; 95 % CI, 1.80 to 3.79; P < 0.0001) and 41% (RR, 1.41; 95% CI, 1.07 to 1.86; P = 0.016) greater. Cellulosic low-flux membrane use was associated with an 85% (RR, 1.85; 95% CI, 1.24 to 2.76; P = 0.0025) greater RR of death than low-flux semisynthetic membranes.
Conclusion: These data suggest that biocompatibility and permeability may impact on death and cardiovascular events in hemodialysis patients with type 2 diabetes mellitus.