From Medscape Medical News
Daniel M. Keller, PhD
July 2, 2010 (Munich, Germany) — In a large randomized, multicenter, controlled study of patients with stage 5 chronic kidney disease (CKD), patients who were assigned to begin dialysis at a lower estimated glomerular filtration rate (eGFR) did just as well as patients who initiated dialysis earlier, researchers announced here at the XLVII European Renal Association-European Dialysis and Transplant Association Congress and the 2nd German Association for Nephrology conference.
Results were also published online June 27 in the New England Journal of Medicine.
During a median follow-up of 3.59 years, there were no significant differences in mortality or the frequency of adverse events between the 2 groups in the Initiating Dialysis Early and Late trial.
Speaking here, lead investigator Bruce Cooper, MB, BS, PhD, from the Department of Renal Medicine at the Royal North Shore Hospital in Sydney, Australia, said practice variations and the worldwide trend toward earlier initiation of dialysis prompted this study.
Physicians have also been concerned that delayed referral and initiation of therapy could adversely affect prognosis. Therefore, the aim of the study was to see whether initiating dialysis early in people with stage 5 CKD reduced the rate of deaths from any cause.
Between July 2000 and November 2008, 828 patients from Australia and New Zealand who were 18 years of age or older with progressive CKD and eGFR levels of 10.0 to 15.0 mL/minute/1.73 m2 were randomly assigned to begin dialysis when their eGFR levels were 10 to 14 mL/minute/1.73 m2 (early-start group; n = 404) or 5 to 7 mL/minute/1.73 m2 (late-start group; n = 424). Patients could receive continuous peritoneal dialysis or hemodialysis as the planned initial treatment, according to physician preference. Late-start patients could initiate dialysis with an eGFR level higher than 7.0 mL/minute/1.73 m2 if the treating physician recommended it; for example, because of symptoms.
All baseline characteristics were similar for the 2 groups, including the causes of end-stage renal disease, coexisting conditions, and classes of medications, and the groups did not differ significantly in the pharmacologic interventions during the trial period.
At baseline, eGFR level, as measured by the Cockcroft-Gault equation (C-G eGFR), for the groups was a mean of 13 ± 1.4 mL/minute/1.73 m2, patient weights averaged 82 kg, and both groups had an average age of 60 years.
Patients had been under the care of a nephrologist for about 30 months before randomization. A total of 59 patients who were randomized had not started dialysis by the end of the trial mainly because their eGFR levels had not fallen to the target level or because they had died. During the study period, 78 early-start and 74 late-start patients received transplants, and their data were censored for the study analysis.
At the start of dialysis, the C-G eGFR level was 12.0 mL/minute/1.73 m2 for the early-start group and 9.8 mL/minute/1.73 m2 for the late-start group, for a mean difference of 2.2 mL/minute/1.73 m2 (95% confidence interval, 1.8 - 2.6 mL/minute/1.732; P < .001). In the early-start group, 18.6% started dialysis with C-G eGFR levels of less than 10 mL/minute/1.73 m2. However, 75.9% of the late-start group initiated dialysis with a C-G eGFR level higher than 7 mL/minute/1.73 m2, mainly because of uremia (n = 234), fluid overload (n = 28), or physician discretion (n = 25).
Dr. Cooper reported that the early-start group began dialysis in a mean of 1.8 months vs 7.4 months for the late-start group, giving about a 6-month difference in start time. Nonetheless, there was no difference in the time to death between the 2 groups. Showing a cumulative hazard plot of mortality vs time, he said: "These 2 lines are linear, and they overlap each other completely." The hazard ratio was 1.04 for the early-start group (95% confidence interval, 0.83 - 1.30; P = .75). The median time to death was about 7 years.
The researchers also looked at other outcomes, including composite or individual fatal or nonfatal cardiovascular events, composite or individual infectious events, and individual infectious complications. No outcome showed a statistically significant difference between the groups.
Dr. Cooper emphasized that there also was no increase in complications as a result of the treatments. "There was no significant difference between the patient groups in regards to need for access revision, access site infection, serious fluid or electrolyte disorder requiring hospitalization, or placement of a temporary dialysis access catheter," he said. In a subgroup analysis, the outcomes did not differ according to baseline eGFR level, age, sex, diabetes, body mass index, history of cardiovascular disease, or albumin.
"So, from these trial results we know that planned early initiation of dialysis in patients with stage 5 kidney disease in this study was not associated with improved survival or clinical outcome, " he concluded. "We also know that from the subsequent analyses of the defined subgroups there was demonstrated consistency of these results, and despite clinically important separation in both eGFR and time to initiation of dialysis, there was no difference. Our results, though, do demonstrate that in a closely managed setting, dialysis can be delayed until GFR drops to below 7 mL/minute or until more traditional clinical indicators for dialysis should arise...and that dialysis should not be started based on an estimate of GFR alone."
According to Dr. Cooper, the study provides level A evidence for guideline formulation, and he noted that some of its strengths are that it was performed in a variety of clinical practice settings and that it had long and complete patient follow-up.
Norbert Lameire, MD, professor at the University of Ghent in Belgium and coauthor of an editorial on the study in the New England Journal of Medicine, said at the conference: "It's obviously a fantastic study. There's no question about it." But he noted that one concern raised about the results is that governments (or other payers) may now try to impose rules to begin dialysis later. "Actually, the study confirms for the first time in a nice trial that clinical symptoms should dictate the start of dialysis, and not the numbers of a lab test," he said.
He wrote in his editorial that 76% of patients who were supposed to start late actually started dialysis earlier for clinical reasons, "meaning that when you have symptoms, irrespective of your kidney function, if you want to do something for your patient's health, you should start dialysis," he advised.
Dick De Zeeuw, MD, PhD, a clinical pharmacologist and clinical trialist at the University Medical Center in Groningen, the Netherlands, said the study results are very clear, but one possible concern is that among the late-start group "the major proportion of the patients...before they hit the [7 mL/minute/1.73 m2 GFR] target were instigated by the doctor to start earlier than that, mostly because of uremic symptoms," he told Medscape Medical News.
The question remains: What would have happened if most of this group really had started dialysis at a lower eGFR level? "But really, to give [the investigators] credit, that isn't that important because they sort of mimicked practice, and they answered a question that was really important: Don't be driven by a single number, which we do in guidelines," Dr. De Zeeuw said.
Therapy must be based on individual patient need, he emphasized.
XLVII European Renal Association-European Dialysis and Transplant Association (ERA-EDTA) Congress: Late-Breaking Trials 2. Presented June 27, 2010.
N Engl J Med. Published online June 27, 2010. Abstract Abstract
No comments:
Post a Comment