Sunday, September 25, 2011

How bad is the long break for dialysis patients?

In this week’s NEJM, Foley and colleagues published a retrospective analysis of the End-Stage Renal Disease Clinical Performance Measures Project (CPM) cohort examining the association of the long interdialytic break (i.e. Friday to Monday for MWF patients and Saturday to Tuesday for TTS patients) and clinical outcomes. The study is already receiving plenty of press (, LA Times) and should be read by all.

Before diving into the study mechanics and results, a review of existing data on the “long break” is warranted. Bleyer and colleagues published 2 prior studies examining the timing of hemodialysis (HD) and patient deaths. The first demonstrated a higher rate of cardiac death on Mondays and Tuesdays. In a subsequent analysis, they found that HD patients had a 3-fold increased risk of sudden cardiac death (SCD) in the 12 hours prior to the 1st HD session of the week. In a later study, Karnik and colleagues examined patient and HD-specific factors associated with a higher risk of cardiac arrest and SCD during dialysis. They corroborated Bleyer’s finding of increased deaths on Mondays (but, interestingly, not on Tuesdays) and identified low dialysate potassium, older age, diabetes, catheter use, and intradialytic hypotension as factors associated with sudden death. Similar findings have also been reported in Europe.

With growing evidence that alternative dialysis schedules (daily, nocturnal, quotidian, etc) improve outcomes (specifically LV mass, QOL), the authors of this new study hypothesized that the long interdialytic interval is associated with increased morbidity and mortality.

Study Basics:

-Design: retrospective cohort

-Population: 32,065 thrice-weekly HD patients from the ESRD CPM from 2004-2007

-Aim: compare rates of death and hospital admission on the day following the “long break” to rates on other days

-Outcomes: mortality (all-cause and cause-specific: cardiac, vascular, infectious, other) and hospital admission rates

Study Findings:

Mean age 62.2y, 45.1% women, 36.3% black, 43.7% w/ DM as ESRD primary dx, 27.7% w/ catheter use, mean vintage of 3.8y (24.2% had been on HD <1y), and mean DSL of 217min

Mean follow-up time: 2.2

Mortality rate: 41.1%

o Cardiac cause: 17.4%

o Vascular cause: 2.7%

o Infectious cause: 4.8%

Event (selected)

Event on Day after Long Break (Rate per 100 Person-Yr)




All-cause Mortality




Cardiac Mortality




Infectious Mortality




Cardiac Arrest




Myocardial Infarction








Hospitalization- MI




Hospitalization- CHF




Hospitalization- Dysrhythmia





· Retrospective nature (potential residual confounding)

· Misclassification due to limitations in outcome adjudication (ICD-9 codes, CMS death forms)

· Study suggests association but not causation and provides little data to suggest the mechanism


The long 72 hour interdialytic interval is associated with higher all-cause, cardiovascular, and infectious mortality as well as with higher rates of cardiovascular related hospitalizations

Potential Mechanisms: (almost all of which deserve further study)

· Elevated potassium levels and associated membrane destabilization

· Potassium shifts (high serum K+ against a low, and often unadjusted dialysate K+)

· High fluid burden and associated cardiac myofiber stretch and stress to the conduction system

· High ultrafiltration rates and associated cardiac stunning

· Catecholamine and cortisol surges and enhanced sympathetic nervous activation

Now What?

There is no doubt that a randomized trial of session timing, length, and schedule is needed. This new study provides further justification (and “clinical equipoise” as noted by the authors) for such. In the meantime, we are mired in a delivery system driven by strict schedules and tight budgets. What do we tell our patients? How do we alter practice now within the confines of the current system? Here are some thoughts:

· Potassium-directed (for patients w high pre-dialysis K+)

o More frequent monitoring of pre-HD K+ and subsequent tailoring of dialysate K+

o Avoidance of 0-1meq K+ dialysate baths

o K+ profiling to maintain a constant K+ gradient

o Target colonic excretion (bisacodyl for ex) of K+ over the long break

· Fluid-directed (for patients with high inter-dialytic weigh gain)

o Strict fluid and salt restrictions

o Extended dialysis session length to minimize ultrafiltration rates

· Alternative schedules

o Additional weekly session

§ Justified under current Medicare policy for patients w large weight gain, intolerance to ultrafiltration, and intradialytic hypotension

o Encourage home HD and PD for appropriate patients

What's next?: Further research is NEEDED before wide-spread practice alteration is warranted particularly given the profound policy implications, but our patients deserve this now. Specifically, we must determine:

o Optimal dialysis scheduling

o Patient preferences regarding frequency and duration (i.e.... will they come??)

o Cost-effectivness analyses (will more frequent HD reduce morbidity and thus cost?)

Posted by Jenny Flythe


MooseMom said...

Thank you for asking, "What now?" What ARE you going to tell your patients who already know the hazards of this weekend-lengthened interdialytic break? And thank you for offering some potential solutions within the current despotic regime.

Are MORE RCTs really necessary to show that just about any alternative schedule (daily, nocturnal, etc) is better than what we have now? Really? Really?

Anonymous said...

"Further research is NEEDED before wide-spread practice alteration is warranted."


Surveys CLEARLY show what American nephrologists actually BELIEVE reagrding dialysis modalities -- when it is THEY who are the ones unfortunate enough to be sitting in the chair. This belief includes many if not *most* of the bloggers represented on this site. Pretend it was you or your loved one. THEN what would you recommend?

Yet to *patients* we portray it as if things are still "undecided" or "lacking sufficient evidence."

This inconsistency between BELIEF and actual PRACTICE is at best dishonest.

. . .but our patients deserve this now."

Something else that patients "deserve" right now is honesty.

Denise Eilers, RN, BSN said...

I find it highly interesting that when I speak at conferences I routinely ask the audience (professionals) how many would choose 3x/wk in center minimal dialysis. The response varies from stunned silence to muffled laughter. I even once heard an attendee whisper "Is she kidding?"

Optimal treatment should include the option of more frequent dialysis, nocturnal treatment,and home based therapy. Most of all, it needs to eliminate the "killer weekend."

Ask patients how they feel after 2 days off dialysis.

Although I honestly understand the need for scientific evidence, in this case, do we need any more RCTs? It's a sad state of affairs when anecdotal evidence is considered suspect and patient experience is disregarded.

Jenny Flythe said...

I appreciate the thoughts and fully understand the comments. Alterations to the American dialysis system require extremely high levels of evidence because of the government-funded dialysis provision system that we have. Dialysis care is very expensive and prior to making policy recommendations, we must have clear data from randomized trials. I agree that the evidence is strong enough to say that the long break is bad, but we do not have trial-guided evidence as to what the best treatment structure is. We have some ideas but do not have clear answers.

In the meantime, we must practice in our current system: adding additional treatments for appropriate patients (likely not needed by all), working to minimize weight gain between sessions, and using alternative means to manage electrolyte abnormalities and shifts. There are feasible practice alterations available to us in the current system that may improve outcomes. While we await further randomized data, we must be more vigilant in selecting appropriate patients for such strategies. We must also educate our patients as to why additional sessions and measures are needed. Unlike many of the patients who read this blog, not all patients are amenable to such practices and it is our job to better educate them and to re-assess patient preferences.

Anonymous said...

Just a cautionary note - everyone, patients, nephrologists etc all BELIEVED Epo was a wonderful treatment. It is only now that we have appropriately carried out trials that we know the true risks involved. While the current system of dialysis therapy is certainly flawed, we do need proper data on which to base a new system and avoid repeating our past mistakes.
Lets be a bit more scientific this time around and not just hope for the best.

Anonymous said...

When thrice a week dialysis for three or four hours was started ( i think in the 1970s), there was no RCT to support its use. Thrice a week dialysis provided just barely enough dialysis for a patient to survive and not get healthy. WE are essentially in that same practice pattern for more than a quater of century inspite of great advances in this field of medicine.

So now we are in a situation where we are trying to do more research to increase dialysis times when initially there was no evidence to base the exisiting practice on.

I think one of the greatest dis service to the dialysis as a whole is the concept of Urea kinetic modelling where adequacy of dialysis is based solely on a number - URR or kt/v, not on how patients feels. I have seen a great no. of examples where pts are dialyzed only for three hours because they meet the URR or kt/v criteria but their BP, volume control, phosphorous etc are essentially down the drain.

For dialysis to change we need buyin from all the stakeholders- from the patients , the doctors, the providers and of course the government.

REMEMBER - DIALYSIS is not part of the Problem, it is part of the solution.

MooseMom said...

Yes, dialysis care is very expensive, but even more expensive is the hospitalizations required when the long week-end break has engendered complications. So, pick your poison. Spend more money now and get people properly dialyzed even on a Sunday (I'd bet God would approve), and spend less on hospital bills.

It depends on what your goal is...patients who are healthy or patients who are merely alive.

And I agree with anonymous who reminds us that the current thrice weekly regime was just a random choice; where are the RCTs to back THAT up?

Ms Flythe, if you were to guess what the best treatment structure would be, based upon what you know and what you see in dialysis patients, what would that guess be? And if you found yourself diagnosed with renal disease and might face dialysis, what treatment structure in a perfect world would you choose for yourself?

How long are you all going to make your current patients wait while you put together more and more RCTs?

Let me ask you this...if you had a patient who decided, after much investigation, that he wanted to do NHHD, would you support him if, in your clinical judgment, he'd be a good candidate?

And let me ask this just as one human being to another. Do you ever worry about your patients on a Sunday? Do you ever think twice about essentially refusing a patient life-saving treatment on a Sunday? How does that make you feel?

Anonymous said...

RE: policy and costs. If every-other-day-dialysis prevents even JUST ONE DAY of hospitalization per patient per year, it would more than pay for itself. Several RCT's have already surpassed this bar many times over. If we need a changed incentive structure, then directly pass all costs resulting from excess hospitalizations to the dialysis units and nephrologists who caused these hospitalizations, and pass all savings from reduced hospitalizations the same way. However, setting up fair accountable care rules which prevent cherry picking healthier patients to manipulate outcomes is an unsolved dilemma.

RE: EPO - a red herring and not at all a legitimate analogy. (Based on a statistical sample of ONE mis-step, all other proposed changes to nephrology practice are now suspect.) The body of evidence in support of more frequent and longer dialysis is orders of magnitude greater than was the case for the use of EPO, now spanning hundreds of studies over decades of time.

RE: RCT's - FHN *was* our RCT. Headlines mis-represented the inadquate sample for the nocturnal arm as a "negative" trial, but the actual results were nothing of the sort. This part of the study was was under-powered, but the results were rather clear.

For future trials, anything other than single-blinding is both unethical and impractical, and informed patients now refuse the inferior treatments up front, which undermines the sample. So, instead of true randomization, you get a sample only of those patients even *willing* to participate a particular way, or samples biased by how well ill-informed the subjects are.

We need to get past this fixation on RCT, and be willing to accept propensity-matched techniques.

This is not some head-to-head comparison of two statins with 99% similar effectiveness. The differences are much more significant (unless you don't believe that doubling or tripling survival is a meaningful signal in your data.)

Strong signals and careful attention to detail still generate MEANINGFUL data.

American nephrology's shameful secret is that most nephrologists derive the major portion their income from their dialysis units, and most practices now owned outright or in direct financial partnerships with the LDOs. This inherent conflict of interest generates an incredible amount of rationalization and denial pervading the entire profession.
Instead of fighting on behalf of their patients, many nephrologists are pressured to continue to act as apologists for maintaining the status quo, even though what they ACTUALLY BELIEVE is quite clear.

The REAL question is how do we improve dialysis without hurting the bottom line so drastically that the practice cannot succeed financially. Further, how do we get nephrologists to accept the fact that it is their moral responsibility to HELP make this happen rather than for all practical purposes working to PREVENT it from happening.
This is not a "scientific" problem, so demanding yet one more, or 100 more, or 1000 more RCT's will not help in the slightest.

Clinical decisions need to be guided by the best *available* CURRENT evidence, not some hypothetical future "optimal" evidence.

But, as a delay tactic, demanding completely unrealistic and unfunded multi-year RCT's before doing anything whatsoever has been an incredibly successful strategy.

It is time for the profession to actually own up to this. We are on the front lines of a major paradigm shift, which won't happen without a fight.

Enough excuses. Patients are becoming more informed, and more disgusted, by this continued failure of will and of courage every single day.

MooseMom said...

@PanicAttacks, could you define "further research"? And Anonymous brings up a good are you going to ethically construct an RCT where half of the patients get the current crap dialysis (with none on Sundays) and the other half get optimal dialysis?

I had thought that this blog represented the youngest, the brightest, the most innovative and the most tech-savvy of the lot, so I am shocked and disturbed when I hear the same old intellectual timidity..."We need more research, time, cultural change, RCTs, whatever other delay tactic we can think of before changes can be made."

I'll tell you a secret. More and more patients are becoming educated, too. We have information at our fingertips, and we use it. We find blogs such as this one, and we form our own communities and exchange information. We viscerally understand that the problem is not a lack of research, rather, it is a matter of money. More and more of us are asking questions and demanding answers. If patients still choose standard clinic dialysis, they will now KNOW that it is dangerous in and of itself and WHY. The result of that will be that you will have more and more patients who see dialysis for what it really is, and the non-compliance rate will rise. Knowing that your dialysis treatment is actually harming you gives rise to depression and anger and non-compliance. At the very least, make every modality easily available at your clinic so that people can give truly informed consent.