Towards Compassionate Dialysis: Thirst and Hemodialysis Duration

This blog post was made by Dr. John Agar on April 7, 2016.
Towards <i>Compassionate</i> Dialysis: Thirst and Hemodialysis Duration

There has been a theme at this blog site over the last year or two, I must confess more than a little fuelled by some of my own contributions [☺] that has consistently encouraged patients (and dialysis professionals) to move to lower hemodialysis ultrafiltration (fluid removal) rates. But, while it’s all very well to say that, the practical problem remains: how?

The simple answer is that there are really only two ways to achieve a lower UF rate:

  1. Encourage/enforce a lower fluid intake in between treatments; a “zip the lips” approach.
  2. Lengthen sessions so fluid accumulated between treatments can be removed more slowly.

The first of these has been “the usual way” for decades. However, interdialytic fluid restriction has manifestly failed to achieve the desired outcome. Further, this failure has led many dialysis professionals to wrongly accuse those unable to control their interdialytic weight (read fluid) gain as “non-compliant problem patients,”a label that is both unfair and unjust. Non-compliance is a term that should never be used against a dialysis patient unless it is being used to describe a severe, active mental health disorder, and even here, compliance is not the right target, rather it is the mental health disorder that demands management. And, compliance has never been the issue.

No, the enemy is not fluid compliance. The enemy is thirst. Thirst is an irresistible urge that we trigger through poorly prescribed dialysis. We, the dialysis professionals, are its manservants and handmaidens. Thirst is a primal, primitive, and irresistible force. Balancing the “internal milieu” is a basic function of every organism, from amoeba to man. While we might appear more solid than an amoeba, we are, in truth, no less dependent on a balanced fluid milieu—a primal soup of water and electrolytes—that fills our cells and the spaces between them. When the balance point of the internal “soup” of an organism is “just right,” the organism is said to be “in homeostasis.”

The purpose of homeostasis is nicely described in the following excerpt:

“The ultimate goal of homeostasis is the maintenance of equilibrium around the set point. While there are normal fluctuations from the set point, the body's systems will usually attempt to revert to it. A change in the internal or external environment (a stimulus) is detected by a receptor; the response of the system is to adjust the deviation parameter toward the set point. [This is a little like why a tennis player will always seek to return to the middle of the baseline to be best placed to react to opponent shots made to either side]. If the body becomes too warm, adjustments are made to cool the animal. If the blood's glucose rises after a meal, adjustments are made to lower the blood glucose level by moving the nutrient into tissues…”1

Blood (a body fluid) is no different. Receptors deep in our brain stems monitor the concentration (osmolality) of our blood in the “thirst centre.” If our blood is too concentrated, our thirst centre will sense this and drive us to take in water to dilute the blood and return its concentration back towards the homeostatic “set point;” the normal range.

Dialysis patients become all too familiar with the names of many of the substances that comprise our internal “soup:” sodium, potassium, bicarbonate, calcium, magnesium, phosphate, and a raft of others; the electrolytes, chemicals, and hormones that dialysis struggles to normalise when the organ that normally does this so effortlessly for us (the kidney) “goes missing.” As dialysis seeks to adjust as many of these substances as possible back towards their mid-point, it must also remove excess water…the fluid in which these chemicals and electrolytes are dissolved.

While we try to get the rates of adjustment and removal right, it is easy to get the differential rates of chemical adjustment and water removal a bit muddled up. If aggressive ultrafiltration removes too much water or removes it too fast, the two twin triggers of thirst result:

  1. The concentration of substances like albumin that are “read by” the osmoreceptors in the thirst centre rise rapidly. The thirst centre will send strong signals to drink water.
  2. A fall in blood pressure from rapid fluid removal reduces brain perfusion (blood flow). This stimulates a reflex need to expand the blood volume by taking on extra water.

So, the poor hapless patient, first ravaged by water depletion during dialysis, is then ravaged by thirst afterward. Once the thirst mechanism is dually activated by (1) and (2), the stimulus to drink is utterly irresistible. Importantly, it isn’t their doing. We have done this. We have activated the primal host defence against dehydration: thirst. And, we have done it by changing their internal milieu so rapidly, so brutally, that instead of ensuring slow post-dialysis trans-compartmental equilibration, we have ensured rapid post-dialysis fluid self-resuscitation as soon as the patient is out the door of the dialysis unit.

Is it any wonder that when s/he returns (reluctantly) for the next treatment, s/he is again awash with accumulated fluid, only to repeat the process? Even worse, there is a heightened risk that this next all-too-rapid drying-down will trigger a “flat,” treated by saline—an infusion of the very salt and water the treatment had been seeking to remove.

Yes, we do have it all wrong. And, the answer is so simple. Longer, slower treatments are symptom-free treatments. Ask any home patient who uses long sessional times: 5, 6, 7 or, even better, 8 hour overnight runs. Ask them about their flats, cramps, hypotensive episodes, or prolonged recovery time. Ask them if their dialysis makes them thirsty. They will look at you as if you had two heads.

I have long talked of “optimum,” not just “adequate” dialysis but, as I grow older, I have come to realise that even “optimum” is the wrong word. We should be aiming for compassionate dialysis. The synonyms for “compassionate” encompass all that we should be striving for in our therapy. Google it, and words pop up like: sympathetic, empathetic, caring, tender, merciful, lenient, tolerant, humane.

Google “thirst,” and words pop up like: crave, want, desire, covet, hunger for, be desperate for, be consumed with need…

Brutal, short dialysis breeds thirst, and thirst breeds desperation.

Gentle, long dialysis breathes compassion, and compassion breathes mercy.

Achieving symptom-free dialysis is relatively simple at home, where patients have full control over the frequency, duration and timing of their treatments:

  • Home HD patients can choose to dialyse longer, and why wouldn’t they?
  • Home HD patients can choose to dialyse more frequently, at least to abolish the long-break (Kjellstrand’s “killer weekend”), and they’d be crazy if they didn’t.
  • Home HD patients can dialyse overnight, while asleep, and thereby access full waking-hour activity: and this is the home HD model I prefer, and recommend.

But facility HD patients are made fully captive by the system, in many ways, and lack these lifestyle (and survival) benefits. So, for the remainder of this post, I intend an important diversion from the usual home dialysis flavour of this blog. It is vital that we do not to ignore the silent majority of dialysis patients who languish in facility-based care, either because they are too frail to manage home care, or are (sadly) unaware of the better outcomes offered by home care.

All too often, facility-based dialysis is run to rigid, inflexible, and staff-structured schedules. Patients of all shapes, sizes, sexes, ages or competing needs are shoe-horned into a one-size fits-all model. They must also stretch the interdialytic break to 68 hours over weekends, so staff can have Sundays off and to save the cost of two extra treatments each month—conveniently glossing over the doubling in the death rate associated with the long break. This is both (a) unthinking and (b) stupid…yet it is what is done. Not only should we ask why, but we should be trying to fix it.

Setting aside the factors of weight, age, and co-morbidity, let’s focus on three metrics:

  1. Solute (waste) clearance, currently measured by Kt/Vurea, which incidentally only measures small solute clearance. Kt/Vurea is now fairly reliably achieved for most dialysis patients, unless there are access problems.
  2. Volume (fluid) management, best measured by the UFRmax. NOTE: See and play with our UFRmax calculator. A ball-park UFRmax can be calculated pre-dialysis to determine whether a safe rate of fluid removal is (or is not) likely for the upcoming session.
  3. Dialysis-related symptoms

If, for now, we assume that an adequate Kt/Vurea is achieved, the problem devolves to providing flexible schedules that dovetail variable length dialysis runs within the total operating hours of a clinic.

How can this be achieved, especially in the short-hour treatment programs of the US?

Firstly, from some data received from a very small US patient survey, it seems that some US centres open around 7.00 am, fit in 2 shifts, and have patients finishing at 4.00pm. While many units may open and close at different times to these, assuming this limited model and with the (assumed) 1 hour turnaround it suggests, this means:

  • An 8.00 – 11.30 am morning session
  • A 12.00 – 3.30 pm session
  • A mean maximum sessional duration of around 3.5 hours.

Contrast this with most Australian and New Zealand units. These also offer 2 shifts but open between 8.00-8.30am and close at 8.30-9.00pm (or a little later) with 2 x 5-hour shifts from 8.30-9.00am to 1.30-2.00pm, then from 3.00-3.30pm to 8.00-8.30pm. This schedule also allows the capacity to run a few longer sessions if or when needed.

From the US ESRD Networks [1-18] Annual Report (2013) just 1,122 facilities (18%) offer evening shifts. This is perhaps a reason why only 13% of US facility patients are in any form of work. Despite this, it should be possible, short of opening longer, and especially in the evenings (= the obvious answer) to either:

  • Dovetail longer and shorter sessions
  • Sort all shorter sessions into one shift and all longer sessions to another

As the latter would mean some patients might need to shift-change, it may be unpopular. The best solution may be to dovetail. So … how can this be worked out?


  1. First, determine the mean weekly Kt/Vurea (or PRU) for each patient, ensuring the Kt/Vurea is >1.3/run (or the mean PRU is >70%). For those who fail to achieve those goals, dialysis must lengthen anyway.
  2. Next, for the likely majority who are achieving the required Kt/Vurea, calculate the UFR for all sessions for a 3 month run-in period. Group the data by 1st session of the week (alone) and then by 2nd and 3rd sessions of the week (combined).
  3. Then, determine what sessional duration would be required (if implemented) to extend all sessions to achieve a UFR <10 ml/kg/hr. (This might mean that, in particular, session 1 of each week would need up-resourcing to allow sessional time extension). If so, so be it! Do the right thing and sacrifice a few $’s of profit for the good of the patients.

Option 1

The best option is, of course, to lengthen clinic hours to accommodate both a Kt/Vurea >1.3 and a UFRmax <10 ml/kg/hr … for all patients! I can guarantee that this would result in a smoother, less symptomatic dialysis program, better volume management, happier hearts, and improved survival.

Yes, you would have to persuade your patients that longer dialysis = less symptomatic, smoother dialysis with a shorter recovery time, but all available data confirms that would be so. Yes, you would take a modest financial hit from the longer hours and increased wages this would entail. But, for your patients—and for your conscience—you should do this.

In a stroke, you would turn your dialysis service into a European/ANZ/Canadian and other western country look-a-like, a change that would be to the eternal benefit of US dialysis. This would be the best option, by far. Imagine the satisfaction your team would gain by being the first US service to self-civilize itself!

Option 2

This is by far the lesser option, but is one path towards the ultimate goal of Option 1.

Once you have measured Kt/Vurea and UFRmax (as above), begin to plan “inter-digitation:”

Group shorter morning sessions with longer afternoon sessions (and vice versa) such that your dialysis schedule converts from looking like the top half (see diagram below) to the bottom half (as below).

Dialysis Staggering Chart

By sharing the nurse/technician loads in the change-over period, it should be possible to significantly shorten the change-over, as not all patients will come off and go on competitively. But, do the maths and the schedule map first.

Then, seek a willing set of patients to set up a trial run—say, for 4 weeks. Survey for “flats” (hypotensive episodes), cramps, and recovery time both during a 2 month run-in period, and during the trial run.

Finally, if the results suggest easier, less symptomatic dialysis—and they will—there is then a discussion to be had with your patients! If their dialysis is smoother, that discussion should be easy.

At the end of the day, wouldn’t it be super-cool if your unit was the first to report a successful “lengthening program” to the annual scientific meeting of the ASN!

  1. Source: Boundless. “Homeostatic Process.” Boundless Biology. Boundless, 08 Jan. 2016. Retrieved 22 Mar. 2016



    Sep 25, 2016 11:10 AM

    Dear John and Vic

    Great post ,

    UFR is a practice that needs to be addressed urgently - many lives can be prolonged and suffering reduced on dialysis. Patients today cannot wait for the next randomised trial !

    The best driver of change is patients/users themselves, helping them understand the trade off, (as in Home dialysis)and I think that is a good place to start . Successful models tested and published will help the providers too.
    Reply to a Comment
    *All fields are required.
    Your email will not be displayed publicly
  • Gale Schulke

    Apr 12, 2016 9:56 PM

    Fortunately the Dialysis Company I am associated with is moving towards that exact model you described for incenter dialysis. More and more patient-clients are receiving longer sessions requiring a change in what has been the status quo for years. The idea of "shifts" is evolving into a different structure to allow for longer sessions and to allow for "nocturnal" dialysis, as well.
    Reply to a Comment
    *All fields are required.
    Your email will not be displayed publicly
  • John agar

    Apr 11, 2016 9:30 PM

    Thanks, Vic.
    Much appreciated - as ever.
    Sadly, though, there have been no responses (yet) to my suggestion(s) from any dialysis professionals - perhaps especially from those in the US - as to the feasibility (or otherwise) of my proposed way(s) forward.
    Ah well ...
    Reply to a Comment
    *All fields are required.
    Your email will not be displayed publicly
  • Vic Johnson

    Apr 11, 2016 1:09 PM

    Dr Agar, this seems so simple, so intuitive, so basic - all forms of medical treatment ought to be based on 'compassion.' That haemodialysis in some sectors ignores this is a blight on the profession and industry.

    Going off into left field for a moment, By way of example, dental surgery and tooth removal is for many people a horrendous and painful procedure which many, (I was one,) ignore until it can no longer be ignored and submit to with trepidation, girding our loins for the inevitable pain.

    I once had an amazing experience when a locum dentist, a Sikh gentleman, performed a wisdom tooth extraction for me. I expressed my dislike of needles in the gums and he assured me he could make the process almost pain free. He explained that he would apply a little anaesthetic to my gums, then deaden them further with a series of small pinpricks, followed by slightly deeper injections etc. He proceeded and after three or four applications, he asked how my gums felt. I said that they felt sufficiently numb that I was happy for him to insert the needle fully and give me the full nerve deadening injections. He smiled and said "I already have" and proceeded to perform a painless extraction. It's something I have never forgotten, - if one dentist can perform such a painless procedure, why don't they all use similar techniques?

    .....and so too, for Haemodialysis. I am the fortunate recipient of such treatment here in Australia, - why then are so many people in what is regarded by many as the worlds most advanced nation subject to such barbaric and outdated protocols that seemingly are devoid of compassion?
    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