**4. Studies comparing the effect of hemodiafiltration versus hemodialysis on mortality**

The above contradictory results of clinical studies that evaluated the comparative effectiveness of hemodialfiltration versus standard hemodialysis in causing regression of arterial stiffness are in line with the uncertain superiority of hemodiafiltration on mortality across large-scaled randomized controlled trials that included "hard" cardiovascular outcomes as primary endpoints. For example, in the primary analysis of the aforementioned CONTRAST study [40], 714 ESRD patients were randomly assigned to switch from low-flux hemodialysis to on-line hemodiafiltration or continue renal replacement therapy with low-flux hemodialysis. After a mean follow-up period of 3 years, incidence of all-cause mortality [Hazard Ratio (HR): 0.95; 95% CIs: 0.75–1.20] and occurrence of fatal and non-fatal cardiovascular events (HR: 1.07; 95% CIs: 0.83–1.39) were not different between the two dialytic modalities [40]. The subsequent Turkish on-line hemodiafiltration (OL-HDF) study enrolled 782 ESRD patients receiving standard thrice-weekly hemodialysis who were randomly assigned in a 1:1 ratio to postdilution on-line hemodiafiltration or high-flux conventional hemodialysis [41]. Over a mean follow-up period of 22.7 ± 10.9 months, the occurrence of the composite primary outcome of all-cause mortality and non-fatal cardiovascular event was identical in both study arms (eventfree survival of 77.6% in hemodiafiltration versus 74.8% in the high-flux group, P = 0.28) [41]. Contrary to the above results, the Estudio de Supervivencia de Hemodiafiltración On-Line (ESHOL) study supports the notion that on-line hemodiafiltration is superior over hemodial‐ ysis in reducing all-cause and cardiovascular mortality [42]. In this open-label, randomized controlled trial, 906 prevalent hemodialysis patients were randomly assigned either to switch to high-efficiency post-dilution on-line hemodiafiltration or to remain on standard low-flux hemodialysis. Switching from low-flux hemodialysis to on-line hemodiafiltration was associated with a 30% risk reduction for all-cause mortality (HR: 0.70; 95% CI: 0.53–0.92), 33% risk reduction for cardiovascular mortality (HR: 0.67; 95% CIs: 0.44–1.02) and 55% risk reduction for infection-related mortality (HR: 0.45; 95% CIs: 0.21–0.96) [42].
