Abstract
BACKGROUND: With its convective component, hemodiafiltration (HDF) provides better middle molecule clearance compared with hemodialysis (HD) and is postulated to improve survival. A previous analysis of Dialysis Outcomes and Practice Patterns Study (DOPPS) data in 1998-2001 found lower mortality rates for high replacement fluid volume HDF versus HD. Randomized controlled trials have not shown uniform survival advantage for HDF; in secondary (non-randomized) analyses, better outcomes were observed in patients receiving the highest convection volumes. METHODS: In a 'real-world' setting, we analyzed patients on dialysis >90 days from seven European countries in DOPPS Phases 4 and 5 (2009-15). Adjusted Cox regression was used to study HDF (versus HD) and mortality, overall and by replacement fluid volume. RESULTS: Among 8567 eligible patients, 2012 (23%) were on HDF, ranging from 42% in Sweden to 12% in Germany. Median follow-up was 1.5 years during which 1988 patients died. The adjusted mortality hazard ratio (95% confidence interval) was 1.14 (1.00-1.29) for any HDF versus HD and 1.08 (0.92-1.28) for HDF >20 L replacement fluid volume versus HD. Similar results were found for cardiovascular and infection-related mortality. In an additional analysis aiming to avoid treatment-by-indication bias, we did not observe lower mortality rates in facilities using more HDF (versus HD). CONCLUSIONS: Our results do not support the notion that HDF provides superior patient survival. Further trials designed to test the effect of high-volume HDF (versus lower volume HDF versus HD) on clinical outcomes are needed to adequately inform clinical practices.