Initiating Dialysis in the Elderly: When Is Right?

By Ajay K. Singh, MBBS, FRCP, MBA

An elderly patient of mine with stage 3 CKD recently had an episode of acute kidney injury while visiting family in Puerto Rico. His bout of gastroenteritis tipped his eGFR into the 10 to 12 mL/min/1.73m2 range and he was initiated on maintenance hemodialysis. In speaking with his nephrologist it seemed quite obvious to me that as soon as the patient had reached a threshold eGFR in the 10-15 mL/min/1.73m2 range the time had come to initiate dialysis.  The patient related to me that at no time was an alternative to dialysis discussed. Simply waiting so that he could think about his preferences was not considered.

Choosing the right time to initiate dialysis on patients with incident kidney failure can be quite confusing. This is especially true when it comes to elderly patients. Should one wait until the patient is symptomatic? What level of GFR represents the right time? To what extent should co-morbid conditions, such as dementia or functional status, be taken into account? Is not doing dialysis an option? Should there be a discussion with the patient and family about conservative options?

For nephrologists, the decision making regarding when to initiate dialysis in elderly patients is often difficult and confusing. We have not completely stepped out of the shadow of “healthy start” era propagated by guideline groups such as NKF KDOQI in 1997 and others1, who recommended initiation when the arithmetic mean of the urea and creatinine clearances fell below approximately 10.5 mL/min per 1.73 m2, (except in well nourished patients). The IDEAL study2 from Australia and New Zealand, showing that early start of dialysis was not beneficial, questioned the validity of the healthy start proponents. Steve Rosansky and colleagues3 backed up the idea of taking a more holistic approach, based on symptoms and patient preference.

For patients, initiating dialysis is often heart wrenching. The profound change in quality of life, the lost of autonomy, and the potential complications and consequences of the therapy itself are frequently overwhelming.  There is also the financial burden, especially coverage for prescriptions and other ancillary services.

One question that has not been definitively answered is whether structured pre-dialysis care makes a difference to whether patients get initiated on dialysis and how well they do.

A recent paper published in JAMA Internal Medicine by Manjula Kurella Tamura and colleagues4 raises several key issues about initiating dialysis in elderly patients,. The paper represents a natural experiment of a large Veterans Administration (VA) data set. It is also worth noting that one in five patients initiated on dialysis in the US originates through the VA system.

In a nutshell, Kurella et al retrospectively evaluated a cohort of 11,215 VA patients age ≥67 years, who were overwhelmingly male and white, and developed incident kidney failure that required initiation of dialysis. They compared two groups: those who, prior to initiation, received Medicare pre–end-stage renal disease (ESRD) care, and those who had VA pre-ESRD care. The difference was that pre-ESRD care for VA patients was more likely to be integrated with an electronic medical record (EMR), incorporated palliative care input, and lacked financial incentive for initiation of dialysis. In contrast, those patients who underwent Medicare pre-ESRD care, the authors suggest, had less integrated and/or coordinated care, that may have lacked palliative consultative input, and was vulnerable to a financial incentive for dialysis initiation.

The findings of the study were striking and supported the idea of taking a more structured, integrated, and perhaps holistic approach to pre-ESRD care.

Comparing dialysis initiation between Medicare versus VA-cared patients, there was a markedly higher rate of initiation of dialysis among patients compared with those in the VA pre-ESRD system: the unadjusted frequency of dialysis treatment was 81.9% with Medicare versus 52.7% for pre-ESRD VA-care patients. This difference did not change much even after adjustment or by using propensity matching.

Further, rather than initiation of dialysis conferring a survival advantage, the higher rate of dialysis initiation among Medicare pre-ESRD patients was associated with a 10% higher mortality compared with those who received pre-ESRD nephrology care in the VA.

While the study conclusions might question whether providing choice to VA patients, in terms of receiving care from Medicare rather than the VA, is beneficial, some caution in interpreting the findings is essential. This was a retrospective study of an administrative data set. The authors made several assumptions about the care these patients received in the VA versus with Medicare. In addition, residual confounding could easily be at play both for the models and the propensity matching approach. Lastly, ascertainment bias may also explain some of the differences.

Still, the observations are thought provoking. For all the grief around inefficiency and waiting lists that the VA receives, its inherent organization, coordination, and integration may be important as a factor in optimizing pre-ESRD care [described in reference 5]. A darker conclusion is that receiving care by Medicare providers as compared with the VA may generate perverse incentives for providers but not make much of a difference or perhaps be even harmful to elderly patients.


  1. Golper TA. The rationale for Healthy Start dialysis. Blood Purif.1999;17(1):1-9.
  2. Cooper BA, Branley P, Bulfone L, et al. IDEAL Study. A randomized, controlled trial of early versus late initiation of dialysis. N Engl J Med. 2010;363(7):609-619.
  3. Rosansky S, Glassock RJ. ‘Early’ dialysis start based on eGFR is no longer appropriate. Nat Rev Nephrol. 2010;(12):693-694.
  4. Kurella Tamura M, Thomas IC, Montez-Rath ME, et al. Dialysis initiation and mortality among older veterans with kidney failure treated in Medicare vs the Department of Veterans Affairs. JAMA Intern Med. 2018;178(5):657-664.
  5. Fischer MJ, Stroupe KT, Kaufman JS, et al. Predialysis nephrology care among older veterans using Department of Veterans Affairs or Medicare-covered services. Am J Manag Care. 2010;16(2):e57-66. PubMed PMID: 20148611.