Blog entry by Arif Khwaja

Anyone in the world

For a number of years some nephrologists have advocated online hemodiafiltration (HDF) as the therapy of choice in chronic dialysis patients. The theory being that enhanced convective removal of middle molecules allows the removal of uraemic toxins that contribute to the inflammatory state in haemodialysis and promote cardiovascular disease.  A number of observational studies have shown a survival benefit of HDF.

This months JASN publishes the results of the Dutch CONTRAST RCT that compared HDF to low flux dialysis in 714 RRT patients. Ultra pure water was used in both arms of the study. There was good separation of beta two micro globulin between the two groups indicating good convective removal with HDF. It appears to be a very  well conducted study.
The results are disappointing. There was no difference in the primary endpoint of death after three years of therapy. Nor was there any difference in fatal and non-fatal cardiovascular outcomes between the two groups. In a post-hoc analysis of patients who achieved a delivered convective volume of > 21.95litres there was an associated reduction in mortality. However as the authors clearly state this kind of post-hoc analysis does not support the use of HDF. Indeed it maybe those that achieved the higher convective volumes simply had better vascular access and few patients achieved these volumes.
What may explain the lack of effect of HDF... A number of explanations are possible including i) the beneficial effects of HDF in observational studies is due to the use of ultra pure water.. In this study the control groups also dialysed with ultra pure water.ii) "beneficial" mediators as well as uraemia toxins are removed by HDF iii) the follow up of 3 years is too short to realise any benefit iv) the HDF dose delivered was simply too low... In fact few patients achieved the intended convective volume of 24 litres v) the wrong end-point was chosen - perhaps intradialtic hypotension would be a more appropriate endpoint or a quality of life measure. vi) the treatment is no more effective than modern-day conventional HD!!
The latter may well be true and once the Turkish HDF study and others are published in the next year or so we will have the answer for sure. ... Of course HDF was non-inferior to standard dialysis so if cost is not an issue the it would be perfectly reasonable to use HDF - though the environmental cost may be significantly higher with HDF.
For now as is so often the case expert opinion supported by observational data once again has been found wanting...

[ Modified: Thursday, 1 January 1970, 1:00 AM ]