Starting Dialysis after 75: New Outcomes Data for Shared Decision-Making


An analysis of outcomes in patients initiating dialysis at 75 years or older could provide important knowledge for physicians, elderly individuals, and their caregivers in the shared decision-making process of whether to start renal replacement therapy. In their review of 5 years of data from a cohort in this age group Bjorg Thorsteinsdottir, MD, and her coworkers from the Mayo Clinic uncovered sobering results, including the frequent loss of independent living and a high mortality rate after starting dialysis (1). Presented at Kidney Week 2013, their work fills an evidence gap and could aid physicians in delivering patient-centered care to individuals in this fast-growing segment of the U.S. population.

Thorsteinsdottir reviewed patient records for individuals 75 years or older who started any form of dialysis at the Mayo Clinic in Rochester, MN, between 2007 and 2011. The 379 patients studied were mainly male (66 percent) with a mean age of 80.8 years at time of initiation. The majority (76 percent) of patients started hemodialysis in the hospital after admission for an acute medical event or surgery. Of note, more than half (60 percent) of the patients included in the analysis began dialysis in the intensive care unit (ICU).

A precipitous early mortality rate was observed in the cohort. Of those initiating dialysis in the ICU, the 6-month mortality rate was 73 percent, with only 23 percent surviving at 1 year. The non-ICU hospital starters had a 22 percent and 41 percent mortality rate at 6 and 12 months, respectively. However, patients starting hemodialysis as outpatients did surprisingly well with only a 4 percent mortality rate at 6 months and 89 percent survival rate after 1 year. This confirms that age alone was not a good predictor of survival, and that comorbid burden and context are more important predictors of morbidity as previously shown (3), said Thorsteinsdottir. They also observed a frequent loss of independent living, with only 37 percent of patients able to return home after hospitalization.

Vanita Jassal, MD, of the University Health Network in Toronto, ON, Canada, and who was not associated with the study, noted it was important to recognize the cohort of patients the authors followed are probably a mixture of two different types of patients: those who initiated dialysis having a chronic renal injury, and those who had no previous or minimal renal disease who had an acute event requiring intensive or other supportive care and an acute reason for starting dialysis.

“I think that the data are exceedingly valid and important, but I think that some of the data they’re presenting represent this very acute, unwell population referred for care to a tertiary care center who we know fare quite poorly, particularly in the earlier period,” Jassal said.

“The results are striking, and suggest that existing estimates of life expectancy for older adults initiating dialysis based on registry data are quite optimistic,” said Ann O’Hare, MD, of the VA Puget Sound Health Care System in Seattle, WA. “This study shines a spotlight on the subgroup of older patients who initiate dialysis in the hospital and never make it into the USRDS registry, in most instances because they do not survive to discharge. While this study has important lessons for all of us, it is important to keep in mind that Mayo clinic is a tertiary referral center so it is not clear how generalizable these practices are to other centers,” she said.

Informing the clinical approach

Thorsteinsdottir pointed to Sharon Kaufman who, in her book And a Time to Die: How American Hospitals Shape the End of Life (4) highlighted “how the technological imperative to treat contributes to overtreatment near the end of life and illustrates how patients and their loved ones are often reduced to largely passive actors in the aggressive quest for life extension. Several recent studies on the ESRD population demonstrate this well, documenting that patients perceive a lack of choice and many experience regret for having started dialysis.”

By contributing to increased awareness about the context and likely outcomes for very elderly patients facing the need to start dialysis, we hope to challenge the moral and technological imperative to treat everyone irrespective of their health status or prognosis, Thorsteinsdottir said. Because of the high treatment burden and poor outcomes for certain subgroups of patients, dialysis should only be started after shared decision making guided by well-informed patient’s goals and values.

“Our study also emphasizes the importance of not being too nihilistic as evidenced by the excellent survival of the outpatient starters,” Thorsteinsdottir said. “Also several patients recovered renal function despite advanced age and serious illness.”

“I think that the data are quite sobering,” said Jassal. “These data should be used to help engage families in the shared decision-making process that the Renal Physicians Association clinical practice guidelines suggest we use (2).” An emphasis on shared decision-making before chronic dialysis initiation was one of ASN’s Five Things Physicians and Patients Should Question contributed to the ABIM Foundation’s Choosing Wisely initiative.

“Overall these findings demonstrate that many older adults initiate dialysis in the setting of an acute illness, and that in these situations, survival is extremely poor,” said O’Hare. “Physicians should warn their patients with chronic kidney disease about the different circumstances in which they may be faced with decisions about dialysis initiation, including acute illness.”

A call for more research

Although this work provided important data, each physician interviewed for this article emphasized the need for more research into elderly individuals and dialysis initiation.

Jassal pointed to more research for patient-centered care. “In older individuals it’s important to ask about their goals of care, but it’s also important to ask if there are other treatments we should be giving them to address the other barriers to them feeling well,” he said. Both Jassal and Thorsteinsdottir also called for research to help identify those patients who could do well on dialysis.

“It is imperative that we get better at preventing acute kidney injury in the elderly,” said Thorsteinsdottir. “Most importantly we need early identification and early goal-directed treatment of infections and the systemic inflammatory response to avoid sepsis. We also need to develop kidney protective measures in the perioperative period and avoid nephrotoxic drugs in this population.”

O’Hare emphasized more needs to be done to help patients and families anticipate the course of their kidney disease, to educate them about what to expect in the future, and to ensure that they are aware of all of their treatment options. “In particular, we should provide patients with realistic expectations about different treatment options and about positive alternatives to dialysis, such as hospice and palliative care so that they do not feel that they have no choice but to initiate dialysis should the need arise,” she said.



Thorsteinsdottir B, et al. Starting Dialysis at Age 75 Years or Older—Outcomes Data to Help in Shared Decision Making. J Am Soc Nephrol 2013; 24 (Suppl):46A (Abstract)


Renal Physicians Association. Shared Decision-Making in the Appropriate Initiation of and Withdrawal from Dialysis. Clinical Practice Guideline (ed 2). 2010: Rockville, MD.


Schoonover KL, et al. Risk factors for hospitalization among older, incident haemodialysis patients. Nephrology 2013; 18:712–717.


Kaufman SR. And a Time to Die: How American Hospitals Shape the End of Life, New York, Scribner, 2005.