Defining the Optimal Target of Ultrafiltration Rates in Dialysis

[Hemodoc]

By Peter Laird, MD

I recently attended the Annual California Dialysis Council conference in Palm Springs, CA which was my first visit to that city. As Lori Hartwell said, the weather was great and the conference had many interesting presentations. I was able to catch the last presentation which featured three chief medical officers from FMC, Davita and Innovative Dialysis Systems.  Unfortunately, Dr. Nissenson was not able to join Dr. Lazarus and Dr. Vincent Dennis. The most notable question dealt with ultrafiltration rates and recent studies showing UF rates higher than 13 ml/kg/hour are associated with several adverse reactions.

Dr. Lazarus noted first of all that we cannot escape the central fact that longer duration and more frequent dialysis is the answer to many of these issues that other surrogate targets will never equal in outcomes. However, he couched that answer by stating we don't have a target like Kt/V to guide therapy. Dr. Dennis echoed the sentiments of Dr. Lazarus that the evidence is clear we can no longer ignore  that longer and more frequent dialysis has improved outcomes. The response from the head of Davita in my opinion missed the mark all together talking about how we don't to date truly understand the etiology of interdialytic weight gain and focused his comments on the many different conditions that lead to elevated weight gains.

Yet, is it really true that we do not have a target so to speak for ultrafiltation rates when it is known from more than one study that keeping UF rates at or below 10 ml/kg/hour reduces complication significantly and improves survival.  In addition, granted that interdialytic weight gain is caused by several disparate factors, but the simple fact is that what ever the cause of the fluid gain, we are dealing with a finite 5 liter vascular compartment that must refill during dialysis from the intracellular and extracellular compartments to maintain adequate perfusion pressures and blood pressure. The rate limiting factor is the refill rates from these other tissues. The studies show in simplicity that when the ultrafiltration rates exceed 10 ml/kg/hour, that rate of fluid removal will in the setting of a 3-4 hour dialysis session exceed the refill rates from the other compartments and the patient will become symptomatic. Dr. John Agar in an article on this issue at Home Dialysis Central describes well the factors involved:

Time-limited fluid shifts

How fast can the blood volume be replaced from the interstitium during a treatment? Well, it depends. Lots of factors make a difference:

Your blood protein (albumin) level
How healthy your heart is
How big or small you are
How “leaky” your smallest blood vessels (capillaries) are to fluid
If you assume a maximum rate of fluid removal (ultrafiltration rate or UFR) of 350 to 400ml/hour for an average sized person, you won’t be far off:

If your UFR is less than 400ml/hr, interstitial fluid can refill your blood as fast as dialysis removes it. So, your blood volume won’t drop and your blood pressure will be stable.
If your UFR is higher than 400 ml/hr, your interstitium can’t keep up. Your blood volume must fall and your blood pressure will drop, too. (You’ll feel awful.)
The more your UFR exceeds 400ml/hr, the greater the gap between fluid loss and refill. The greater the gap, the higher the risk of affecting your blood pressure.

Thus, instead of focussing on the many ill defined variables that contribute to interdialytic weight gain alone, the focus needs to be on the simple physiology of fluid removal from the vascular compartment and the refill rates from the intracellular and extracellular compartments. Understanding the sophistication of most nephrologists on minute changes of the differing elements of renal physiology and the terror I felt as a medical student and as a resident in training presenting to these very precision based specialists, I am appalled that one of the easiest to understand principles of fluid diffusion is ignored in its application to dialysis and ultrafiltration rates. The optimal ultrafiltration rate is thus defined by this innate physiology of fluid mechanics in the human body. Perhaps it is time to simply go back to the basics for this complex question. Just as phosphorus and middle molecules are time dependent variables of dialysis, so likewise are ultrafiltration rates.

http://www.hemodoc.com/2011/04/defining-the-optimal-target-of-ultrafiltration-rates-in-dialysis.html

[Bill Peckham]

Interesting Peter. I'd say that the transfer rate is likely a percentage of body weight - something like .5 to .8% and Agar seems to have left one important variable off the list - sodium. Eating a high sodium diet can lower your transfer rate, thus sodium is a double whammy - it makes you thirsty and it makes it hard to draw the fluid off.

[Hemodoc]

Interesting Peter. I'd say that the transfer rate is likely a percentage of body weight - something like .5 to .8% and Agar seems to have left one important variable off the list - sodium. Eating a high sodium diet can lower your transfer rate, thus sodium is a double whammy - it makes you thirsty and it makes it hard to draw the fluid off.

Great point Bill, a simple truth in medicine is that fluids follow the sodium.  Even in dialysis patients, it is not so much the amount of total fluids a person drinks, but how much sodium that they consume. When combined with a high sodium intake and increased fluid intake, the interdialytic weight gain will exceed the targeted levels.

[Bill Peckham]

There is one source of sodium I rarely see discussed - the sodium delivered during dialysis. Saline is salt water, and sodium modeling can give a dialyzor their daily dose of sodium before you even leave the unit. And then it turns out that there is a widespread problem with miscalibrated machines delivering dialysate with widely varying sodium levels. I heard some recent data at ADC and last week I received this abstract from Renal Week 2009:

Edwards C, Singh S, Griffiths M, et al.
Correct Calibration of Delivered Sodium Dialysate Concentration Results in Improved Blood Pressure Control.
ASN Annual Meeting -- San Diego
J Am Soc Nephrol (Nov) 20:204A 2009


Sodium balance is integral to intradialytic cardiovascular stability and good interdialysis blood pressure control. Hypernatraemic dialysis risks positive sodium balance and increases thirst and hypertension.
Hyponatraemic dialysis may lead to negative sodium balance and distressing symptoms on dialysis.

At our centre we prescribe a standard sodium dialysate concentration [Na]d of 140mmol/L measured by ionic conductance between electrodes within the dialysis machine. As part of a quality control procedure prior to a clinical trial we measured the delivered [Na]d using a flame photometry assay. Dialysate was sampled cyclically over a 6-month period from 71dialysis machines. 32/71 [35.7%] machines were delivering a mean [Na]d >144mmol/L. The machines were re-calibrated with fortnightly assessment of delivered [Na]d until they were at target 140mmol/l within the tolerances of the machine [2 mmol/L].

Post calibration there was a significant decrease in mean [Na]d, 143.52.1 v 141.5 1.4 mmol/L [p<0.0001] and only 6.7% dialysis machines continued to deliver [Na]d >144mmol/L.

Delivery of a [Na]d closer to our target of 140mmol/l led to a significant decrease in mean ultrafiltration volume 2.41.1 v 2.10.9L [p<0.01] as a surrogate for interdialytic weight gain. The trend to lower pre-dialysis systolic blood pressure did not reach signficance 144.423 v 140.023 mmHg (p=0.07). However the number of antihypertensive agents used decreased significantly from 1.91.5 to 1.71.5 [p<0.05].

The discrepancy between prescribed and delivered [Na]d can be ascribed to electrode malfunction. With simple measures to correctly calibrate [Na]d we have demonstrated a significant reduction in antihypertensive use with well controlled blood pressure.

Sheesh - one out of three machines were screwing over the dialyzors who had to use them.

If I was incenter before anyone talked to me about my fluid gains between treatments, I'd want to know that their machines are delivering the called for sodium concentration.

EDITED TO ADD: If you really want to dive into the weeds, check out this 2008 paper which gives a bit of history along with some interesting technical analysis: Revisiting the Dialysate Sodium Prescription as a Tool for Better Blood Pressure and Interdialytic Weight Gain Management in Hemodialysis Patients http://cjasn.asnjournals.org/content/3/2/522.full

[Bruno]

I can't say much here because I can see I'm talking with people far more informed than I have, but then, I'm a very reflective user so let me tell you what dialysis and my body's reaction have taught me.

1) Dialysis time is crucial...all sessions should be at least 6 hours...because
2) UF rate must never exceed 400, I consider 350 ideal.

I consider UF rate the crucial protocol in dialysis because longer times mean a lower rate and you should aim at session times and frequency that enable the lowest rate. Removing toxins is not the issue because longer sessions only improve that benefit.

Fluid control is also a key factor, you are always going to be under the hammer if you start a session regularly with over 3 litres to lose. Hearts cannot cope with that sort of loading over a period of time let alone high pump speeds. I use a blood pump speed of 300.

I'm surprised that more attention to uf rate is not paid...as an example, in my clinic we record our uf goal, but not the rate which i find odd as I always plan sessions around the rate with the goal a secondary consideration.

Perhaps that might be a starting point.