Do We Need to Abandon High Ultrafiltration Rates in America?

By Peter Laird, MD

Dialysis practices around the world differ significantly from the practice patterns observed in America and many have long believed this is in part the explanation for our higher dialysis mortality.

I recently spoke with a manager of a dialysis unit and his experience with horrified Japanese patients who couldn't believe the blood flow rates used in America compared to Japan. Japan, Europe, Australia and New Zealand have long recognized the survival benefits of longer, slower and gentler dialysis compared to our American style violent sessions.

A recent paper looked at the ultrafiltration rates as a marker of survival. In this retrospective analysis of nearly 2000 patients, the authors categorized the patients into three groups: Ultrafiltration rates 10ml/h/kg, 10-13ml/h/kg, and those over 13ml/h/kg.  The results are striking:

Rapid fluid removal during dialysis is associated with cardiovascular morbidity and mortality

Compared to ultrafiltration rates in the lowest group, rates in the highest were significantly associated with increased all-cause and cardiovascular-related mortality with adjusted hazard ratios of 1.59 and 1.71, respectively. Overall, ultrafiltration rates between 10–13 ml/h/kg were not associated with all-cause or cardiovascular mortality; however, they were significantly associated among participants with congestive heart failure. Cubic spline interpolation suggested that the risk of all-cause and cardiovascular mortality began to increase at ultrafiltration rates over 10 ml/h/kg regardless of the status of congestive heart failure. Hence, higher ultrafiltration rates in hemodialysis patients are associated with a greater risk of all-cause and cardiovascular death.

Many authors have recently noted similar results, but this study, albeit retrospective, carries significant weight due to the numbers of patients analyzed.  The high American dialysis mortality must be viewed in terms of our different dialysis practices first before corrective steps are implemented effectively. Abandoning sodium modeling will likely lead directly to lowered inter-dialytic weight gains that prompt the high ultrafiltration rates.  It is time for American nephrologists to critically evaluate our practices and eliminate those that are associated with higher mortality, and embrace those practices that lower mortality.

Comments

[John Agar]

I have never agreed more. That was a nicely done study but, again, falling on deaf ears.

Peter, if only ... ...

But sadly, profit drives your train. While haemodialysis is linked to, driven by, orchestrated for profit, the concepts of slower, gentler therapy - however much and often it can be shown that such concepts are beneficial for survival, cardiovascular outcomes, symptom-reduction, reduction in hospitalisation and improvement in rehabilitation - these sensible, patient-restoring, life-restoring concepts will never be noticed, never adopted, never 'win'.

And that, I find, is incredibly sad.

John Agar

[Peter Laird, MD]

John, thank you for all of your support over the years. They will only be able to run and hide for so long before the truth catches up with them. It will only change with public outrage. I won't hold my breath, but I do believe that there are factors heading in the right direction. We have the data, it is time to publicize it outside of the daily dialysis crowd who already understands these issues. Even the deaf can read if they can't hear.

[somerville]

Since diabetics make up between 30% and 40% of the dialysis population in the U.S., but constitute a much smaller proportion of dialysis patients in Japan and other countries, high average ultrafiltration rates may be inevitable in the U.S. as long as the dialysis time is held constant. Since diabetes is a condition characterized by intense, hyperglycemia-driven thirst, high intradialytic fluid gains are nearly inevitable for these patients. The higher death rate for diabetic renal patients all over the world may in part be explained by greater interdialytic fluid gains in this patient group, and in part by their more severe comorbidity. The statistical overlap between those with reduced life expectancy because of higher hemofiltration rates and those whose higher hemofiltration rates are a causally insignificant correlate of their lower life expectancy due to their diabetes should be teased out.

[Peter Laird, MD]

Dear Somerville, I could't disagree more with your comments than any that have come before. Those at highest risk of elevated ultrafiltration rates benefit the most from avoiding them. Further, the differences between treatment outcomes in Japan and America is not the patient population, it is the dialysis practices between these two nations that make that difference. Several papers have already addressed this specific issue. I believe that we will hear much more about this issue and perhaps one day people will actually listen.

[Zach]

"There are none so blind as those who will not see."
J. Swift

[Kathy Heffner]

Thx Peter... again great points.... I hope the mindset changes here in the US.... There are so many people who could benifit from longer/slower dialysis.... I still believe that there are many people who would fight against doing dialysis longer...but then we have many people who would like to be given the choice of longer/slower dialysis and yet they are not given this choice...

[John Agar]

In reference to the comment above by 'somerville' and to inform, at the end of the 2008 year ANZDATA shows that 34% of patients in Australia were on dialysis with diabetes as their primary cause of end stage renal failure.

John Agar

[nickp]

"I recently spoke with a manager of a dialysis unit and his experience with horrified Japanese patients who couldn't believe the blood flow rates used in America compared to Japan"

But the article is about ultrafiltration rates. Blood flow rates and ultrafiltration rates arent the same.

[Peter Laird, MD]

Dear nickp, the point of that sentence is in context to the rest of the paragraph as follows:

Japan, Europe, Australia and New Zealand have long recognized the survival benefits of longer, slower and gentler dialysis compared to our American style violent sessions.

Ultrafiltration rates are related to blood flow rates in that America style dialysis operates with lower total duration. Therefore, to achieve the minimal Kt/V goals set forth by Medicare guidelines, the method used is often to elevate the blood flow rates much higher than seen in Japan, Europe, Australia and New Zealand. Since the duration of dialysis in America is shortened, they use much higher ultrafiltration rates as well.

This inevitably leads to excessive fluid loss from the intravascular compartment (Blood in the arteries and veins) while the excess flud (edema) in between the cells takes longer to equilibrate. The only way to overcome this is to slow down the blood flow rate, the ultrafiltration rate and offer dialysis more frequently and of longer duration. So you are correct, blood flow rates and ultrafiltration rates are not the same, but they do interact with each other. Both are related by the time factor. Both are thought by many nephrologists outside of America to be much too high based on the physiology of fluid removal from the interstitial spaces. (fluid in between the cells that we call edema)

Dr. Agar wrote a couple of excellent articles on this issue of dialysis flow rates and ultrafiltration rates for Home Dialysis Central in late 2007 that explains these issues at the physiology level. It is written in a manner that health care professionals and patients alike can gain much information.

http://www.homedialysis.org/resources/tom/200711/

http://www.homedialysis.org/resources/tom/200712/

[John Agar]

Peter

Thanks for referencing those two explanatory notes that I wrote for HDC as (hopefully) 'easy to read' explanations of some of the complexities of dialysis.

I had almost forgotten them but on re-reading, they seem as close to an easy-to-understand mark.

I am grateful your memory is better than mine!

I really do wish people could read - and re-read - them as they do try, as a pair and in lay language, to explain why solutes are different and rates, both blood flow (so-called Qb) and the ultrafiltration rate (UFR) so fundamentally matter.

Not trying to blow my own trumpet here (not too much, anyway) but, I seriously don't think enough dialysis doctors give thought to this stuff. And, it's so fundamental.

Sadly, we should devote more time to thinking about fundamentals, about simple process, about how things happen, and about how this might impact on a body, an organism. Slow change always beats fast. Adaptation was Darwin's way ... and, even all those years ago, his thinking seems still right.

Good dialysis should allow time for adaptation during dialysis ... cellular and organism survival has ever depended upon this principle. Darwin showed us this. Dawkins has reinforced it.

Dialysis must allow time for cellular adaptation ... just like the ads on TV to slow down on the raods, dialysis hurry or speed 'kills. That applies to Qb, UFR ... whatever!

PS: Feel free to read some of the answers to questions at my HDC patient Q&A site too ... see ... http://forums.homedialysis.org/forumdisplay.php/16-Dr.-John-Agar-Nephrologist

John Agar

[nickp]

Thanks for the reply and links Peter.

Regarding blood flow rate what would you say is considered a high blood flow rate?

In my case I do 4hr sessions 3x a week. I use a fistula, 14G needleds and try to have a blood flow rate of 430ml/min so I process just over 100 litres a session.

[Peter Laird, MD]

Dear John, the articles are a bridge between the technically oriented and the patient without that background. You are so correct that they are useful to professional and patient alike. Quite a rare talent to be able to talk to both in the same article.

When I read them the first time a couple of years ago, it put together the entire issue of intradialytic hypotension. The ultrafiltration rate cannot exceed the 3rd space diffusion rates back into the vascular space with out causing hypotension and the symptoms associated with such. It is truly an iatrogenic symptomology in most patients. Throw in some sodium modeling and you have a perfect circle of blame thrown at the patient for being "noncompliant" when it is all the nephrologists fault for not understanding the basic physiology, or worse, not caring.

[Peter Laird, MD]

Dear nickp, giving specific advice is difficult in an internet setting and that specific question is better left to your own medical team.

There are no studies linking blood flow rates to access problems but Dr. Agar has written about this and in his opinion, they may be related. You may want to pose that question directly to Dr. Agar on the HDC forum by the link he gave above and he can educate all of us. Dr. Agar truly has a grasp on these clinical issues of dialysis and I only wish he was here in America to get these folks right, but not even Dr. Scribner could do that.