Understanding Concepts in Dialysate Flow

This blog post was made by Dr. John Agar on April 20th, 2017.
Understanding Concepts in Dialysate Flow

By Dr. John Agar

A recent thread of posts at the Home Dialysis Central FaceBook site contained a couple of erroneous comments that suggested some underlying misconceptions I couldn't resist addressing.

One appeared to imply that current dialysis was either “single pass” or “multi-pass”... as if there were a choice, or two options available. The other misinterpreted the efficiencies (or otherwise) of flow rates - in particular, dialysate flow rate, but also of blood flow rate.

First, single pass vs. multi-pass: what exactly do these terms mean?

Though there has been some nice work from James Heaf from Denmark on the opportunities that a multi-pass system might bring (see a short summary), multi-pass haemodialysis is a much different concept than the conventional single pass dialysis systems we use in dialysis today. 

All current commercially available counter-current dialysate flow systems, including NxStage, are single pass systems. Countercurrent means that the blood flows in the dialyser in the opposite direction as the dialysis fluid.

The differences between commercial systems are to do with flow rate—blood flow rate and dialysate flow rate—not the number of times the dialysate travels past the blood compartment. 

Sorbent dialysate regeneration systems have generally served as the classic model for a multi-pass dialysis system, but there are no commercially available sorbent systems at the moment. In the past, there have been the REDY system, which effectively disappeared in the early 1990's, and, briefly between 2006 and 2007, the RenalSolution Inc. Allient Eagle. Both are now extinct. So, while James Heaf's wonderful work is a most intriguing and sensible application of the multi-pass concept, and more may yet come of it, let me now leave the concept of multi-pass behind, and think JUST of dialysate flow.

Second, dialysate flow and phosphate removal.

A comment suggested that a lower dialysate flow rate in a Fresenius Baby K system will remove more phosphate than a faster flow, given the same treatment time. This statement compared phosphate removal between two options of a single variable during a 6-8 hour x 4-5 times/wk dialysis regime, one a lower (300 ml/min) dialysate flow rate vs. a faster (500 ml/min) dialysate flow rate. It suggested that phosphate removal would be greater at the lower dialysate flow rate. 

Indeed, phosphate removal will be a little higher—not lower—with the faster rate than the slower flow rate, though there would not likely be a huge lot in it. No...it is treatment time that plays the greater role in phosphate removal, not dialysate speed.

In our own nocturnal service, we set a flow rate of 300 ml/min for our many dozens of long, slow (8-9 hr), frequent (4-5 nights/wk) nocturnal patients, but this is more to conserve water than to influence clearance, as the clearance benefit is primarily derived from time, not dialysate flow rate. We still remove so much phosphate that we usually need to give some back by adding phosphate to the dialysate!

Thinking, for a moment, about single pass conventional dialysis systems, the total dialysate volume needed for each treatment is clearly determined by flow rate and treatment time: - A dialysate flow rate of 300 ml/min will require (ie: use up) 300 ml x 60 min/hour = 24 litres/hour. - At a flow rate of 500 ml/min, the dialysate requirement would be 500 x 60 = 30 litre/hour.

Thus, in the setting of a “standard” 4 hour treatment, a total dialysate volume of 96 (@300) or 120 litres (@500) would be needed. Blow these out to an 8 hour nocturnal treatment and 192 (@300) or 240 litres (@500) would be needed. While that is ok, the higher the dialysate flow rate, the more water-greedy the system.  It then boils down to a trade-off decision between the clinical clearance benefit of a higher flow rate (if any) and the practical availability and cost of water.

Whichever way the bread is buttered, conventional dialysis is water greedy, and does require so much water, that travel-friendly dialysate volumes are not possible, at least not without a travel-friendly reverse osmosis system...and, as yet, there isn't one!

To facilitate a travel mode and minimise the required dialysate volume to a manageable 30 litres (ie: 6 bags holding 5 litres each ) = about the maximum bagged dialysate volume that can be 'reasonably transported' for a single treatment, NxStage HAD to come up with a novel and REALLY slow dialysate flow concept. This had to be one that would permit a low enough total per treatment dialysate volume to make portability feasible, yet still “clear” (i.e., remove) sufficient “stuff” to provide adequate dialysis. Clearly, 96 - 240 litre volumes—as above—were NOT consistent with a transportable system. 

So, they chose to lower the dialysate flow way down to 30 litres/treatment. For example, if dialysate flow rate could be lowered to 150 ml/minute (= 9 litres/hour) for a total treatment time of 3 hours, only 27 litres of dialysate would be needed per treatment.

But...and this can be the hard bit to get your head around…as a slower dialysate flow rate begins to “saturate” the dialysate with waste, and thus the concentration gradient between blood and dialysate begins to lessen, waste removal begins to slow down. As the amount of waste in the dialysate relative to the amount in blood increases, less and less removal results.  If the differential concentration gradient between blood (removing from) and dialysate (removing to) is reduced, the efficiency of waste removal is reduced.

Peter is beginning to pay Paul!

Recognising this clearance problem as the Achilles heel in their march to reduce the dialysate volume to a manageable, transportable, travelable volume, NxStage came up with a nifty plan to combat this efficiency problem.  At least in part, they compensated by pushing up their recommended blood flow rate. They “reversed” the usual ratio of blood flow (250 ml/min) to dialysate flow (500 ml/min)—> a ratio of 1:2 to a blood flow of (say) 450 vs. a dialysate flow of 150 —> a ratio of 3:1, or similar. 

They called this “new ratio” the “flow fraction.”

This flow fraction concept largely—though not fully—corrected and compensated for the lesser waste removal achieved by slowing down the dialysate flow rate. NxStage had worked out a way whereby they could reduce the required total per-treatment dialysate volume, maintain adequate clearance, and facilitate a portable system, all at the “small expense” of needing a very fast blood flow rate.

But, nothing is ever free. High flow rates can potentially have significant implications  for the vascular integrity of the fistula. I have dealt with this issue before, in a blog about fistula flow rates in 2014 at KidneyViews.

To recap, NxStage now had a model that achieved adequate (and I use that word intentionally) clearances—not “great” nor “optimum,” but “adequate”—and that achieved portability - a major goal. This came at a price, as everything does, and that price was: (1) The risk to fistula or central vein integrity, (2) The ability to achieve optimum clearance; this latter “risk” becoming especially problematic when considering the removal of time-dependent substances…like phosphate!

NxStage was designed for short, sharp, frequent runs (initially 2-2.5 hrs) to allow its low flow rate dialysate to eek adequate clearance from a limited total dialysate volume. But, while this seemed to satisfy and broaden the options for an already poorly-dialysed American dialysis population, it didn't appeal elsewhere where better dialysis was already the accepted norm. This has led to further iterations of the NxStage system, as their designers ever strive to increase the available dialysate volume back towards volumes that again preclude travel—the untransportable Pureflow—leaving the lower volume System One bag system as a lesser efficiency travel mode.

So, in summary, back to the original two statements: 

  1. All current hemodialysis systems are single pass, including NxStage. It is just the speed of pass that is different (i.e. the ratio between the blood flow rate and the dialysate flow rate). Multi-pass is limited either to (1) a sorbent system, though no current commercial sorbent systems are out there, despite that many are being R&D'ed, or (2) a research tool - like James Heaf's nifty multi-pass version of a conventional single pass system - though this is not (yet) an option for routine management.  
  2. The notion that using a slower dialysate flow rate in a Baby K will enhance phosphate removal is also a “no.” If anything, it is the opposite, though the length of time on dialysis is what influences phosphate clearance most, and not the dialysate flow rate.

Finally, this blog does not mean to be denigrate the NxStage system—not at all—but as with all current systems, it does have disadvantages and inefficiencies, and these should be understood. NxStage offers advantages of size and travel-ability; indeed, its original raison d'être…but all life is a series of compromises, and, just as with life, the NxStage system requires some compromises to achieve its central goal of portability and to answer the question: “How can the need for untransportably large volumes of dialysate be overcome?"

The NxStage compromises have been, for many, acceptable. But, while acknowledging the benefits that the NxStage delivers to many, there are some (and I am firmly in this camp) who, retain a preference for efficiency and clearance superiority, and will seek that goal through frequency, duration, and the use of conventional blood:dialysate ratios, until a better offers comes along.

Comments

  • Dori

    Apr 24, 8:56 AM

    Talker, if I had to guess, that may just be the BEST compliment you could possibly give Dr. Agar. :-)

    And, Mel, I found your experience very compelling, particularly the changes in the lab test results.

    Reply to this Comment

    * All fields are required.

    1. Your email will not be displayed publicly
  • Amanda Wilson

    Apr 22, 5:32 PM

    Thank you again John for your thoughtful blog. One observation in response to it. I use a NxStage machine and always have done since I started home hemo. Five and a half years ago (not home dialysis, because I did PD before that for eight years). I chose the machine for its portability, accepting a trade off that it did not provide the same efficiency for clearanc of a conventional machine. At the beginning I appreciate that NxStage facilitated shorter times and a lower total volume. Now that the machine is being used for nocturnal dialysis, at lower BFR and dialysate rate. When I switched I was running at a BFR of 250, and 7.5 dialysate rate for eight hours every other night. Every day I woke up feeling terrible despite that fact that my Kt/v (a poor adequacy of dialysis I know) went up to 4.7 and all other parameters were in range. This was irrespective of how much fluid I removed. UFR was never more than 0.3/hr. My head hurt, my body hurt and the tinnitus that I suffer moderately from became really bad. The only real physical benefit was that my chronic tendinitis abated. I am now back to doing five hourly daytime treatments three times a week.

    I also suffer from periodic episodes of depression. I appreciate an optimal goal for dialysis, but in my experience that does not necessarily equate to optimal quality of life. I believe that there needs to be a balance between the two because greater frequency and time can, and do have an impact on quality of life.

    What are your views on this? Have you looked into whether nocturnal dialysis using NxStage passes beyond adequate?

    Do you have any suggestions for making nocturnal work for me?

    Do you know why I feel so bad after a nocturnal session? (I sleep soundly for the first five hours and then fitfully after that, which is not necessarily unusual for me but does not leave me feeling so unwell).

    I am pretty sure that if I did not feel so bad physically afterwards, then I would also have optimal quality of life. As long as I feel good then I am willing to acept that dialysis with NxStage may not be optimal.

    Reply to this Comment

    * All fields are required.

    1. Your email will not be displayed publicly
    • John Agar

      Apr 25, 8:18 AM

      I am at a disadvantage here, as apart for a short patient trial, reading, much lecture exposure, and my acute awareness of reams and reams of written patient (but not personal) experience, I have not used nor had patients on NxStage. We do not use it here in Oz. This means that any response I give will be guesswork, at best, and possibly off the mark. Nor do I know stuff like your pre and post numbers. We use these a lot to guide decisions like dialyser size (eg: excessively aggressive urea removal) ... and my recall is that NxStage does not provide multi-sized m2 options in dialyser size. As larger membrane-area dialysers rip urea of the blood quite quickly (which, in turn, can rapidly lower the plasma urea relative to brain tissue urea ... and urea crosses ONE physiological barrier slowly = the blood-brain barrier) this can lead to the movement of water INTO the brain - the old 'disequilibrium syndrome' that we rarely now see, but that we have encountered in NHD patients using too large a dialyser. NxStage does, I think, use a biggish dialyser to aid better solute removal for its primarily model = short hour treatments ... and this can't be down-sized easily through physician choice as the dialyser is inbuilt into the click-in cassette. I have no idea what you pre and post ureas are, but it might be interesting to know. Ditto, for some of your other biochemistry! Other factors, body size, albumin levels ... the list is long ... need consideration, too. And, none of this is known to me.

      It doesn't sound like a UFR or volume/BP issue.

      As for your depression? ... that is a different beast! Depression is complex, and may relate in part to how you have been knocked about by your dialysis, in part to the long (seemingly interminable) period of dialysis dependence, in part to factors at home or in your social, physical, employment (or lack of it) environment etc. These are all beyond my ken. All I can advise here is that you let your team know when stuff overwhelms ... and let them know early, and clearly, that you are struggling. They can help.

      I am not sure if any of that is near or far from the mark ... but I DO suspect electrolyte or solute issues - like disequilibration - may underpin some of the symptoms you felt. If you could (but with NxStage's pre-chosen and fixed dialyser, you can't) change the surface area down, I think I would probably be looking at a smaller kidney ... or at least questioning that aspect.

      Reply to this Comment

      * All fields are required.

      1. Your email will not be displayed publicly
      • Amanda Wilson

        Apr 25, 10:19 AM

        Thank you, John. That has given me some food for thought. I believe a lot of my depression comes from the frustration of not being able to dialyze at night and then giving up waking hours to it. I am going to run your thoughts past my nephrologist, but it may be tricky finding a unit that doesn't use NxStage, since it was hard enough finding one that even does home nocturnal.

        Reply to this Comment

        * All fields are required.

        1. Your email will not be displayed publicly
  • Mel Hodge

    Apr 21, 2:47 PM

    I ran a decade-long "clinical trial" (n=1) of NxStage vs. Fresenius 2008K dialysis during my wife Jane's years on dialysis that completely substantiate John's argument We used the Fresenius machine at home and the NxStage machine at two different travel locations for either one or two weeks most months. We always ran nocturnal six nights/week and blood flow rate of 200 on both machines (we were unwilling to risk Jane's catheter at a higher rate), a dialysate rate of 300 ml/min on the Fresenius machine, but 7-5 liter bags on the NxStage machine -- about 70 ml/min.

    Jane invariably felt worse with NxStage treatment. After a night or so she would start to mutter, "I hate that machine," and the feeling ill continued until we returned home and back on the Fresenius machine. Our center even offered to support moving the Fresenius machine for us when we traveled, but it was an impractical idea.

    The difference was not just in how Jane felt; it also showed up in the lab results. Her serum creatinine was significantly higher after a week or two of NxStage dialysis. In fact, the difference was enough to raise her CKD classification from Stage 5 to Stage 3 with the Fresenius machine and 8 hour/6 day treatment when her creatinine was plugged into the MDRD equation (I know this is a bit of a stretch), but CKD patients are nearly symptom-free at Stage 3 and so was Jane on the Fresenius machine.

    A second noticeable lab difference was serum phosphate. Jane had to take a phosphate supplement on the Fresenius machine (as John noted was typical for 8 hour patients), but not on the NxStage machine (but serum phosphate was still never high enough to require binders because her treatments were so long).

    Our experience convinced me that John's arguments about a saturated dialysate system are exactly right and very important and fully justify not using the NxStage machine "down under." Unfortunately, in the U.S. it is about the only available home option.

    The NxStage machine has minor issues, mostly from its foundation in turn-of-the-century technology which I'm sure will be fixed in a new machine. But a new machine must provide a dialysate rate of at least 150% of the blood flow rate if it hopes to provide more than "adequate" dialysis.

    I should acknowledge the NxStage System One has two great attributes. First, its "luggable" weight permits travel, but an even more important benefit is that it permits centralized "swap-out" maintenance via FedEx, etc. I am convinced no home machine will ever be successful in the U.S. if it requires in-home maintenance (which was the Achilles Heel of the otherwise promising Aksys machine). The second attribute is the clean design separation between hydraulic systems (blood, dialysate) and electro-mechanical systems thereby providing what i found to be superb reliability compared to the much more problem-prone Fresenius machine.

    I just hope NxStage finds a future path to getting its patients back to Stage 3...

    Reply to this Comment

    * All fields are required.

    1. Your email will not be displayed publicly
  • Amy Staples

    Apr 21, 1:12 PM

    I'm trying to come to grips with this coupled with why the need for such large filters. When I first started out with nocturnal home hemo I was using an F50 filter (pediatric size) and had phenomenal clearances (this was before both native kidneys were removed). Now however I am using an F160 (after kidney removal) and I'm told by some that regardless of how much *time* of dialysis I receive I should *always* request ne demand the largest filter manufactured for the best clearances. Your thoughts? Blessings

    Reply to this Comment

    * All fields are required.

    1. Your email will not be displayed publicly
    • John Agar

      Apr 21, 8:09 PM

      Dialyser size represents, primarily, membrane surface area. Hollow fibre dialysers are now exclusively the dialyser in use (though patients who dialysed - like the celebrated Nancy Spaeth - back in the 60s to early 70s will well remember the parallel plate sheets of the Kill system, or the later versions in Gambro Lundia series) ... and these consist of thousands of roughly 30cm long hollow tubes bundled together inside a 5-6cm diameter hard plastic casing. Indeed, I have one, sawn in half to show the construction and these tubes, sitting on my office desk as a tool to educate my patients!

      If you were to slit each tube along its length, flatten it out, and join it up with all the other tubes, similarly slit and flattened out, you would end up with a flattened membrane sheet. This will have a total surface area. That surface area determines the total surface across which diffusion occurs within the dialyser. Commonly, this surface area will range from - say - 1.5 square metres up to 2 (or more) square metres, for very big dialysers. Simply put, the bigger the surface area, the 'oomphier' the dialyser, as more 'transport' occurs, per unit of time, over a bigger surface area ... though other factors play a part too: the 'permeability' (i.e. leakiness) of the membrane contstruction, the pressure differences exerted across the membrane, the flow rates of blood (in one direction) and the dialysis fluid (in the other) + a few other factors I won't confuse by adding in.

      In nocturnal patients, we tend to use the smallest surface area dialysers ... e.g. for Fresenius, an FX50, rather than the bigger surface area'ed FX60, or the (bigger again) FX80, or the maxi FX100. In the case of our frequent (4-5 night) long (8-9) hour NHHDs, the bigger surface area dialysers simply remove too much, too fast ... are TOO efficient, and can trigger symptoms suggestive of 'disequilibrium' for some of the smaller, more rapidly removed substances like urea.So, yes Amy, we go deliberately slow with our dialysers, too, to control for excessive removal of osmotically active urea, in particular, to avoid blood-brain barrier disequilibrium!

      Why then long, slow dialysis? Long and slow I directed at slower transport substance removal (e.g. phosphate), and the deep compartmentalise clearance of things like methyl guanidine, guanadinosuccinic acid, and like beasties. And long,slow dialysis slows fluid removal to a rate that is well within the rate that fluid equilibrates across tissue and cell systems, thus ensuring gentle, asymptomatic dialysis.

      You are right! Small surface area dialysers are ideal for long, slow, frequent dialysis. Too big, and you will run into osmotic gradient issues - especially with too rapid a removal rate of urea.

      Reply to this Comment

      * All fields are required.

      1. Your email will not be displayed publicly
  • Talker

    Apr 21, 10:34 AM

    You are a beacon of light in a confusing world dialysis, from a patients point of view.

    Reply to this Comment

    * All fields are required.

    1. Your email will not be displayed publicly
    • John Agar

      Apr 25, 7:45 AM

      Echoing Dori, I do thank you, Talker. That was very kind. I just wish a few more of my colleagues would take the first step into the void and try to bridge the gaps that exist in understanding and knowledge-sharing between professionals and patients.

      Reply to this Comment

      * All fields are required.

      1. Your email will not be displayed publicly
Leave a New Comment

* All fields are required.

  1. Your email will not be displayed publicly