US Haemodiafiltration, Where Art Thou?

This blog post was made by Dr. John Agar on December 17, 2015.
US Haemodiafiltration, Where Art Thou?

US dialysis is clearly very different than dialysis as it is practiced everywhere else. I have long tried to understand how and why some of these differences have evolved, yet chief among these difference are the very US-specific funding and business models. Congress has made successive attempts to ‘correct’ a pallet-load of summative distortions of the US model since its first misstep in 1972(1)*, but each attempt has only deformed and buckled the funding streams and business models more.

* NOTE: The Rettig book should be a required read for every trainee nephrologist.

Let me put the business models aside; Mel Hodge addressed this issue well earlier this year in his post, It’s (Past) Time to Destroy the Dialysis Business Model . The treatment model is broken too. Everywhere one looks within the US dialysis-sphere, treatment is too short, too fast, too brutal, too facility-based, too system-oriented, too profit-focused, too suffering-ignorant, too patient-disrespective. Yet, these are the damning criticisms that arise as a cri-de-coeur, day upon day, from US dialysis patients as they sadly give an un-heeded voice to their frustrations. But these, too, are not the intent of this post.

There is another yawning difference I find increasingly difficult to understand – and it is this: Why is it that we, in the rest of the world are turning, more and more, to the use of the several options of haemodiafiltration (HDF) when—inexplicably—when (or if) HDF is mentioned in the US, the response from the overwhelming majority is “What is HDF?

HDF has many advantages over HD—in particular, it is a significantly less symptomatic treatment with a much more rapid recovery time, and consistent survival advantages of between 30 and 35% have now been confirmed in a large number of studies. The data have been summarized in the Italian report(2) and by Bernaud Canaud,(3) a long-time French advocate of HDF well known in the US as a regular invited speaker at the excellent Annual Dialysis Conference (ADC; to be held in Seattle next year) and at the ASN. As an aside, perhaps a trip to Seattle in February 27-March 1, 2016 might be instructive, and the website is referenced here for those who may want to consider a very useful registration.

Europe long-ago turned towards HDF. A detailed analysis from 2012 that reports data to the end of 2011, showed a pan-European penetrance of HDF in favour of HD of 15% with ¼ of Northern Europeans on an HDF modality(2). A full 67% of Swiss dialysis patients were on an HDF modality at the time the Italian report appeared in 2012. The pan-European uptakes are now likely even more.

Australia and New Zealand appear to be following suit: see the graph below.(4) In our own regional Australian unit, we converted all but our 45 home high frequency, extended hour nocturnal patients to online HDF a couple of years ago. Canada, too, has a growing incidence of on-line HDF. Only in the US is HDF almost non-existent, a recent incidence analysis being tagged at just 0.1%.

A thinking-man’s dialysis-expert from the US, nephrologist Tom Golper, recently addressed this odd disparity between US practice and “the rest,” in a commentary in Kidney International.(5) The subtext conclusion to the question he posed in his title: “Is Hemodiafiltration Ready for Broader Use?’ being—and I hope I have read you right, Tom—a clear “Yes.”

Why then this additional difference in US dialysis compared to ‘the rest’? If the purported benefits of the HDF modalities over conventional hemodialysis are as many and as clear as those described in the multitude of papers over the last 15 years in the European literature, then in addition to my earlier examples where US dialysis seems out-of-step, (see my figure below) this denial of an HDF option for US patients is inexplicable.

I, and others, have pointed out the “option drought” that is Arbor’s heavily funded but intellectually barren PCORI-supported EPOCH-RRT study(6). A bare two options were recognised by Arbor as they devised an educational program to alert US dialysis to the smorgasbord of choice available to them: facility-based in-centre HD and home PD.

Somewhere along their truncated thought-line, they forgot home HD (short daily, alternate nocturnal and/or frequent nocturnal – and anywhere in between); they forgot centre-based nocturnal HD; they forgot the option to transition from failing home PD directly into home HD while the home fires are burning; and I have to be suspicious that they may not even have heard of the multiple modalities—both in-centre and at home—of online HDF.

As many who read this blog may be unfamiliar with the process differences between HD and HDF, I will post a follow-up in the New Year that explains just how HDF works, and why it has been shown to have a range of outcome advantages.


  1. Rettig RA. Origins of the Medicare Kidney Disease Entitlement: The Social Security Amendments of 1972. Available at:
  2. Del Vecchio M et al. Elements for Economic Evaluation on Online HDF vs Standard HD to Treat Patients with End Stage Renal Disease (ESRD). Italian Journal of Public Health 9(4) Supplement 1: Page S21. 2012. Available at:
  3. Canaud B et al. Optimal convection volume for improving patient outcomes in an international incident dialysis cohort treated with online hemodiafiltration. Kidney Int. (2015) 88, 1108. doi:10.1038/ki.2015.251. Available at:
  4. ANZDATA Registry. 37th Report, Chapter 4: Haemodialysis. Table 4.12. Australia and New Zealand Dialysis and Transplant Registry, Adelaide, Australia. 2015. Available at:
  5. Golper T. Is hemodiafiltration ready for broader use? Kidney Int. (2015) 88, 940–942.
  6. Agar JWM. Absurd! EPOCH-RRT Study “Decision Aid” Won’t Include Home Hemodialysis


  • John Agar

    Dec 27, 2015 2:14 AM

    David, you responded with the comment: “I believe one of the obstacles to HDF in the US is cost.” However, there are now a number of costing comparisons in the literature, all of which suggest that the all-in, rolled-up, per-treatment costs of HDF are either marginally cheaper, or marginally more expensive, but mostly line-ball with the costs of HD. Either way, the savings or extra costs appear measured only in cents/treatment.
    In one of these trials (1), this one from the UK with costs measured in British pounds (£), the differential came down to whether standard or cuvette-included lines were used. In the case of standard line use, HDF cost £0.78/session less than high-flux HD while when more expensive lines were used for HDF, HDF costs were £1.16/session greater. Of note, within the rolled-up comparison were savings from the lower requirement for and use of phosphate binders in HDF compared with HD of £1.20/session. These authors concluded that, depending upon choice of blood lines, on-line HDF was either slightly more expensive or slightly cheaper than high flux HD, while the reduced need for phosphate binders proved a useful internal cost saving.

    A dissertation paper from the CONTRAST trial (2), an RCT of 714 patients from 29 centres in Netherlands, Norway and Canada who were randomly assignment to on-line HDF or high flux HD, concluded that despite a survival benefit to HDF when convection volumes were more than 20.3L/session, and despite an improvement in the QoL domains for HDF, HDF was marginally more expensive. This led to the author to conclude that, for the clinical value gained, HDF dis not appear cost effective.

    These are but two of the trials that evaluate cost – there are others – yet while they conclude differently, it always comes down differences of a few cents/session, either way. Perhaps if or when profit margins matter, these small differences might seem more important, and it is true that HDF has largely prospered in jurisdictions where when private profit is not a factor. But, even if taking only those studies that have found HDF to be more expensive, it seems hard to argue strongly that that any marginal cost impost (if true) should be seen as a significant obstacle when the reported benefits to QoL, middle molecule clearance, and symptom improvement are taken into account.


    1. Oates T, Cross J, Davenport A. Cost Comparison of On-line HDF vs. High Flux HD. J Nephrol 2012. 25(2) 192-197

    2. Mazairac AH. A‘Dissertation Paper’: Published Utrecht University, The Netherlands
    Reply to a Comment
    *All fields are required.
    Your email will not be displayed publicly
  • David Baker

    Dec 23, 2015 1:16 PM

    I believe one of the obstacles to HDF in the US is cost. I believe a Company by the name of Nephros has designed a device that can foster the use of HDF in a cost-efficient manner. This device was approved by the FDA over two years ago. I recall when the device was first entering FDA trials it was met with much fanfare. However since approval it has not received much attention. Davita began a trial of the device but promptly abandoned it without explanation. Fresenius is performing an ongoing trial of the device. Meanwhile US dialysis patients continue to suffer.
    Reply to a Comment
    *All fields are required.
    Your email will not be displayed publicly
Leave a New Comment
*All fields are required.
Your email will not be displayed publicly