Should We Care That Being on Dialysis Has a 1-Year Mortality of 30% or Higher?

For those patients with end-stage renal disease (ESRD) that has reached the point where their kidneys no longer function, dialysis offers a form of chronic life support. Previous studies have suggested that about 30% of those initiating dialysis die within the year, a new study believes the percentage is more significant, around 50%. Are they right? More importantly, does it matter?

The Study

The authors made use of the Health and Retirement Study, a longitudinal study of older Americans and identified all the participants who initiated the dialysis, were over 65 and participated in traditional Medicare over 16 years. This data set also provided information on co-morbidities, the functional status of the patients as determined by their activities of daily living (ADL) and where dialysis was started, as outpatient or inpatient. 

The database yielded 391 patients so that the sample size is so low that making generalizations requires much more than a grain of salt. 

• 22% of patients died within 30-days of initiating dialysis, an additional 22% in the next five months, and another 10% over the remaining six months – a total of 54.5% mortality in the first year.
• 73% of patients with impaired ADLs, 70% of those over 85, and 62% initiating care as an inpatient died in that first year.

They concluded that in for physicians discussing dialysis these findings would “help to frame prognostic expectations and support more informative discussions.” And while this is not in any way a “death panel” type of recommendation, I suspect that the researchers believe that this will keep people from pursuing what might be seen as futile care – where the patient cannot recover or that the treatment will not alleviate suffering. 

Will knowing this additional information make a difference?

Let’s reframe the choice facing patients and their families. Do you want to die in the next two weeks or do you want to initiate dialysis that will prolong your life but may reduce its quality? Tough call, especially when you have not considered what to do previously. 

Of all the risk factors the authors looked at, it was the location where dialysis began, specifically as an inpatient, that had the most substantial impact of early mortality. There are two inter-related reasons for this to be the case. Dialysis initiated in the hospital is not elective; it is either an emergency or urgent; the two differ only in the time to disaster, about a 24 to 48 hours difference. The fact that more of these patients need urgent or emergency dialysis suggests that the opportunity to have an informed discussion is limited — these patients face a genuine life or death choice. 

My experience suggests that a 50 or 70% chance of living on dialysis is a far more persuasive argument than dying now. And even for those who might have considered not starting dialysis previously, here and now decisions with immediate consequences are different than those kitchen-table conversations of hypotheticals. Furthermore, it is the rare patient and family that does not bring emotional considerations to the decision, how can it be otherwise? Dialysis does not provide a cure; it provides successful palliation and results in less suffering. Is it not surprising that patients and families choose palliative intervention over dying? 

Informed consent is more than statistics and framing “prognostic expectations,” it is a decision that comes with emotional and family considerations, and often those issues take more than an informed discussion to percolate and resolve. They take time. We may have to accept this higher rate of what might be considered futile care by recognizing that it is for some patients and their families, a comfort measure and not a treatment. 

Source: One-Year Mortality After Dialysis Initiation Among Older Adults JAMA Internal Medicine DOI: 10.1001/jamainternmed.2019.0125