Patients with chronic kidney disease started chronic dialysis substantially earlier in the course of disease in 2007 than in 1997, according to new research. “These findings are significant given the available evidence that earlier initiation of dialysis does not lead to improved outcomes and given the substantial intensity and cost of chronic dialysis treatment,” explains researcher Ann O'Hare.

The past decade has witnessed a trend toward initiation of chronic dialysis at higher levels of estimated glomerular filtration rate (eGFR). To investigate what this trend means for patients in terms of additional time spent on dialysis, O'Hare and colleagues used model-based estimates to assess differences in the timing of initiation of dialysis between 1997 and 2007. “The trend toward initiation of chronic dialysis at higher levels of eGFR has been well described. In order to estimate how much earlier patients were initiating dialysis in 2007 compared to 1997 we needed to know not only how much eGFR had increased during that time period but also how rapidly patients were losing renal function prior to dialysis initiation.”

Credit: © iStockphoto

To address these questions, the researchers used data from two different sources: the United States Renal Data System (USRDS), a national registry of patients with end-stage renal disease, and Group Health Cooperative, an integrated health-care system in Seattle, WA. “The USRDS provided information on eGFR within 45 days of dialysis initiation during the different time periods, but did not include information on eGFR trajectory prior to dialysis initiation,” explains O'Hare. “We therefore used longitudinal information on predialysis eGFR slope from Group Health data to estimate the eGFR slope for USRDS patients”.

The researchers analyzed USRDS data from a cohort of 75,572 patients who had initiated dialysis in 1997 and 104,711 patients who had initiated dialysis in 2007. Patients who initiated dialysis in 2007 were older and had a higher prevalence of diabetes and vascular disease than those who initiated dialysis in 1997. Mean eGFR at initiation of dialysis was higher in patients who initiated dialysis in 2007 than in those who initiated dialysis in 1997 (10.8 ml/min/1.73 m2 versus 8.1 ml/min/1.73 m2), consistent with known trends.

O'Hare et al. then calculated the eGFR at which each 2007 patient from the USRDS database would have initiated dialysis based on 1997 practice and used a total of 16,302 serum creatinine measurements from 666 patients from Group Health Cooperative to estimate the rate of eGFR loss for each 2007 USRDS patient prior to dialysis initiation. The researchers were then able to estimate the difference in time between when patients from the USRDS initiated dialysis in 2007 and when they would have initiated dialysis in 1997.

Using these model-based estimates, the researchers found that chronic dialysis was initiated an average of 147 days earlier in patients in 2007 than in 1997. “Our main findings were that on average patients who initiated dialysis in 2007 did so approximately 5 months earlier in the course of their kidney disease than would have been the case based on 1997 practice,” explains O'Hare. The researchers also found this difference to be greater in older patients. “Those aged 75 years or older initiated chronic dialysis approximately 8 months earlier in 2007 than in 1997”.

The researchers highlight the impact that this difference in timing of dialysis initiation is likely to have in terms of resource and financial burden. They estimate that an average difference of 147 days would translate into approximately 63 additional hemodialysis treatments involving ≥189 h of treatment, and say that this additional treatment would translate into approximately US$14,490 in additional payments for each dialysis patient.

Of note, the researchers believe it is unlikely that such large increases in treatment intensity and costs are matched by improvements in treatment benefit. “Most observational studies have reported higher rather than lower mortality in patients who initiate chronic dialysis at higher levels of kidney function, and two recent randomized trials failed to show a benefit of earlier dialysis initiation.”

They also believe that the estimated difference in dialysis timing cannot be explained by changes in measured patient characteristics. Rather, they suggest that this difference reflects changes in dialysis initiation practices and call for a careful evaluation of contemporary dialysis initiation practices in the US. However, they point out that their study does not provide information on the circumstances or indications for initiating chronic dialysis among USRDS or Group Health Cooperative patients and therefore does not provide insight into the practice changes that might be driving the observed trends in timing of dialysis initiation. “We are currently conducting a more detailed study that we hope will provide some insights into this question,” concludes O'Hare.