DENVER—Compared with conventional three-times-a-week hemodialysis, daily in-center hemodialysis led to better outcomes in a composite endpoint of one-year mortality and change in left ventricular heart mass, and in a composite endpoint of one-year mortality and change in physical health, according to results from the randomized Frequent Hemodialysis Network (FHN) Trial. The results were reported during the late-breaking clinical trials session here at the American Society of Nephrology Renal Week 2010 and in the New England Journal of Medicine (2010;363:2287-2300).
More frequent therapy was also associated with better control of hypertension and hyperphosphatemia, but it did lead to more frequent interventions for vascular access.
“It's clear that more frequent dialysis improves these specified outcomes, and that has to be one of the things that a physician and their patient together decide as they consider the choices they have for treatment,” said Alan S. Kliger, MD, Clinical Professor of Medicine at Yale School of Medicine and Chief Medical Officer at the Hospital of Saint Raphael in New Haven, CT, in an interview after he and colleagues presented the trial results.
“I don't think it dictates the best way to do anything. What it tells you is that you have this additional information that we need to share with patients so that patients can decide which form of therapy is better for them.”
The daily in-center hemodialysis trial, which randomized 245 patients to receive in-center dialysis six times a week or conventional three-times-a-week dialysis, was one of two parallel trials conducted by the Frequent Hemodialysis Network.
The other trial looked at nocturnal dialysis, randomizing 87 patients to overnight dialysis six times a week or conventional dialysis three times a week. All of the patients in the nocturnal arm and most of the patients in the conventional arm received dialysis at home. The trial did not show a significant change in either of the two co-primary outcomes, the same outcomes that were analyzed in the daily in-center trial.
“The nocturnal home study showed statistically nonsignificant reduction in the mean left ventricular mass and nonsignificant increase in vascular access intervention,” Dr. Kliger said during a press conference at the ASN meeting. “The fewer number of enrolled patients in the nocturnal study under-powering the study may in part have been responsible for these nonsignificant findings.”
The research was supported by the National Institutes of Health and the Centers for Medicare & Medicaid Services (CMS).
Statistically Significant Difference
The FHN Trials were conducted to address the question of whether more dialysis is better, Dr. Kliger noted during his presentation.
“The HEMO trial showed no benefit of more hemodialysis when delivered three times a week. More frequent hemodialysis delivered daily or overnight can dramatically increase solute removal and more effectively control volume.
“Observational trials showed benefits, but six years ago when we designed these FHN trials, no prospective randomized trial had been done. Since then, one small randomized trial from Alberta, Canada, showed reduced left ventricular mass among patients dialyzed five or six nights a week compared with conventional three-times-a-week treatment. Given the limitations of observational trials, we performed two multicenter, prospective, randomized trials.”
The daily in-center study was conducted between January 2006 and March 2010 at 11 university and 54 community hemodialysis facilities in North America.
“Choosing the outcome measure was a challenge,” Dr. Kliger said. “A mortality study would have needed 3,500 to 5,000 subjects—clearly infeasible. Since reduction in left ventricular mass has been shown to correlate with survival in hemodialysis patients, and patients’ self-reported physical health correlates with survival, we chose two co-primary composite outcomes.
“We stipulated that demonstration of favorable effects on both co-primary outcomes would be required for us to interpret as evidence of overall benefit.”
Patients were followed for one year. Those receiving in-center dialysis six times a week showed significant benefits in both co-primary composite outcomes, with a hazard ratio for death or increase in left ventricular mass of 0.61, and a hazard ratio for death or decrease in the physical health composite score of 0.70, while those receiving nocturnal home hemodialysis six times a week had a statistically nonsignificant reduction in mean left ventricular mass.
Nine domains for secondary analysis also were examined. Both daily in-center and nocturnal home hemodialysis led to statistically significant improvements in pre-dialysis phosphorus and pre-dialysis systolic blood pressure, and daily in-center therapy also had significant benefits for left ventricular mass and the physical health composite. Neither type of more frequent treatment had significant effects on cognitive performance, self-reported depression, serum albumin concentration, dose of erythropoiesis-stimulating agents, or non-access hospitalization or death.
In terms of complications, patients in the frequent-hemodialysis group of the daily in-center trial were more likely to have a vascular access-related intervention, both in terms of time to the first intervention and in the need for multiple interventions.
Among these patients, there were 95 vascular access-related interventions—19 to correct access failure and 76 other procedures—compared with 65 interventions in the conventional-hemodialysis arm of that trial—23 to correct access failure and 42 other procedures. Also, 47% of the patients in the frequent group had at least one procedure, compared with 29% in the conventional-hemodialysis group. In the nocturnal trial, the increase in vascular access interventions in the more frequent group was not significant.
The increase in vascular access interventions observed with more frequent dialysis is reason for concern, said Robert Lockridge, MD, Medical Director of the University of Virginia Lynchburg Home Dialysis Center and Clinical Associate Professor at the University of Virginia, in a phone interview. Dr. Lockridge was involved with the nocturnal home trial but not with the daily in-center trial.
“Clearly there was a trend in the nocturnal group and a difference in the short daily of more interventions, but was it because the nurse was looking at the access more, or was it because the access was being stuck twice as often? I think that's very important to sort out.”
Much of the discussion of the FHN trials has centered on the outcomes measured.
“I think everyone certainly would hope to have a study that examines mortality as an endpoint,” said Christopher Chan, MD, who was part of the Frequent Hemodialysis Network, in a phone interview. Dr. Chan is R. Fraser Elliott Chair in Home Dialysis and Associate Professor of Medicine at the University of Toronto, and Medical Director of Home Hemodialysis for the University Health Network.
“Recruitment is challenging, and it certainly wasn't without significant effort even to recruit the 247 patients that were randomized in the daily trial. I think as an investigator I would certainly welcome an international approach in looking at a trial examining mortality, but there will be feasibility issues and also other challenges.
“In the meantime, I think a surrogate outcome is what we're left with. I think the surrogate marker of left ventricular mass is an important outcome. It does not mean that we have definitive data on survival, and it's unlikely that we're going to have definitive data on survival in the near future, so my interpretation of this trial is that this is a very positive signal. I think this allows us to at least have a strategy for patients with end-stage renal disease, and I think we could really examine how we're delivering dialysis in North America to our patients.”
The big question is how these results should affect clinical care, noted Rajiv Agarwal, MD, Professor of Medicine at Indiana University School of Medicine, who co-moderated the session during which the results were presented.
“Should we be putting all patients on daily dialysis? The answer to that question is a lot more complicated because we don't know how the benefits were achieved. Maybe it is the better volume control in these patients; if they had a better reduction in intravascular volume, maybe they would have a smaller left ventricular mass, and they might do better.
“So bottom line is, yes, it's a positive study. It will probably be a boost in the arm of daily dialysis folks, but whether it's ready for primetime is a much a harder question to answer.
“It didn't save lives; it didn't improve albumin; it did not have an improvement in cognitive function; it did not do a lot of things that daily dialysis has been thought to do; it did not reduce the dose of erythropoietin. So the glass is half full-half empty. I don't think that this is the last word.
“I think we need a true outcome study in this where we don't have a surrogate endpoint of LVH [left ventricular hypertrophy] and mortality combined. We need a large randomized trial. As these studies went, it took a decade to do 200 and some patients—and lots of money to do this study—and I'm not sure if a larger study will ever be done.”
Dr. Lockridge agreed that the size of the trials makes translation of the results to the general population difficult, but it would be very challenging to recruit a larger number of patients into a trial like this.
“In my judgment the best strategy would be that since the federal government is reimbursing for this, they would say, OK, we're going to look at nocturnal, short-daily, and in-center dialysis for the next two years, and we're going to reimburse at four treatments, and we're going to distinguish between short daily and nocturnal, and we're going to look at it for the next two to three years until we have several thousand people that are doing it, and then we're going to compare hospital cost and mortality between age matched, sex matched, and disease matched, and I think that can easily be done if CMS wants to do it,” he said.
While patients in the frequent-hemodialysis arm of the short-daily trial received their treatment in center, home is the more common location for this type of therapy, making the translation of these results to the home setting another question.
“One of the things that was interesting is if you look at the physical health composite score in our nocturnal trial, that improved from baseline in both people who got three-times-a-week dialysis and people who got six-times-a-week dialysis, suggesting that the home setting itself caused, perhaps, the improvement in that physical health composite score,” Dr. Kliger said. “There's no question that the home setting itself, I believe, will offer advantages.”
The results of the Frequent Hemodialysis Network Trials should not be considered in isolation, he added.
“You need to interpret the results of the randomized controlled trials in the light of the other observational trials. You look at the randomized controlled trial, but then you use all of that data together to come up with the best clinical solution.”