Atlanta, GA - It could up the cost of treating patients with acute decompensated heart failure, and no one really knows how it might work or whether it's safer or more clinically effective than standard intravenous diuretics. But ultrafiltration therapy could gain ground as a potential alternative acute therapy now that a prospective, randomized trial has suggested it might have important advantages.

Dr Maria Rosa Costanzo

In the Ultrafiltration vs IV Diuretics for Patients Hospitalized for Acute Decompensated CHF (UNLOAD), the treatment apparently reduced the 90-day risk of rehospitalization compared with standard "aggressive" diuresis, without causing any evident renal dysfunction or hypokalemia. Ultrafiltration was also more efficient at removing fluid even though it didn't alleviate dyspnea any better than diuretics.

Although previously announced here at the American College of Cardiology (ACC) 2006 Scientific Sessions and reported by heartwire, the trial's main findings were fully presented for the first time yesterday by principal investigator Dr Maria Rosa Costanzo (Edward Hospital Center, Naperville, IL).

So what's the caveat?

Despite its promising results, UNLOAD isn't the final word on ultrafiltration in acute HF and shouldn't be taken as license to use it on a broad scale in patients who would normally receive standard diuretic therapy, experts cautioned both to ACC attendees after Costanzo's presentation and in conversations with heartwire. Any clinical advantages over diuretics haven't been confirmed in more conclusive trials, they observe, the necessary equipment is more costly, and the mechanisms behind any sustained benefits are unclear.

Dr JoAnn Lindenfeld

"Everybody uses ultrafiltration in some refractory patients even now, but I wouldn't view these data as persuasive enough to use it full-scale in a million patients a year with acute decompensated heart failure," Dr JoAnn Lindenfeld (University of Colorado Health Sciences Center, Denver) told heartwire.

Lindenfeld, who wasn't involved in the trial, questioned whether patients in the IV diuretics arm were consistently diuresed to the fullest possible extent—noting that that seldom occurs even in clinical practice. More convincing to her, she said, would have been superior outcomes with ultrafiltration after both strategies had reduced volume overload to comparable degrees. "What this study tells me is that we need to be more intensive about our diuretic therapy. Then if we see problems, it's time to say that intense diuretics are not adequate. But I don't think this trial told us that."

Dr Peter E Carson

Another limitation of UNLOAD, according to Dr Peter E Carson (Washington VA Medical Center, Washington, DC), also not a trial participant, was its open-label design. Although necessary, it could have biased the future treatment decisions of clinicians who were aware of their patients' randomization assignment, he told heartwire.

But the study may have been enough to support more liberal but still selective use of ultrafiltration in acute HF. UNLOAD investigator Dr John R Teerlink (San Francisco VA Medical Center, CA) said that the prevailing wisdom among the treatment's practitioners is that it's best reserved for patients who need extensive fluid removal or for whom diuretics might be risky. "Where that line is cut will depend on each clinician," he said to heartwire. "The patients who come in and need only one or two kilos taken off are probably not appropriate for this therapy."

Dr John R Teerlink

Dr Barry M Massie (San Francisco VA Medical Center, CA) said his institution contributed eight patients to UNLOAD, and clinicians there were impressed by what ultrafiltration seemed to do for them. But, he told heartwire, the center's closely collaborating cardiology and nephrology staffs decided to hold off on any decision to use the technique more widely until they knew the trial's results.

So how does he feel now? "I think we'll use it. . . . I think it was quite clear-cut that it does reduce readmissions," Massie said. "The nice thing about it is it gives us a chance to treat heart failure without diuretics, if only for the short term." But he noted that as an actual substitute for diuretics, ultrafiltration remains a frontier.

Then there's that issue . . . 

The commercially available ultrafiltration system used in the trial (Aquadex FlexFlow; CHF Solutions, Brooklyn Park, MN) costs about $19 000 per unit, according to the manufacturer. The replaceable filter used at each treatment session goes for about $900. In her formal presentation, Costanzo acknowledged that the procedure is likely to be more expensive than IV diuretics, "but we believe there will be sustained benefits to patients and payers from the significant reduction in rehospitalization and in length of stay."

Carson agreed that up-front equipment expenses would probably be quickly recouped, especially at centers that handle a lot of cases, if the clinical benefits seen in UNLOAD turn out to be real. "I think if the rehospitalization data holds up, this would be a cost-effective therapy."

 . . . and questions about how it works

Mechanisms behind UNLOAD's "signals" of fewer rehospitalizations with ultrafiltration are somewhat of a mystery, according to Teerlink. Despite no advantage over diuretics in alleviating symptoms, the procedure removed more fluid, apparently without aggravating creatinine clearance or promoting hypokalemia. So, he said, ultrafiltration may avoid renal insufficiency and other potential complications of IV diuretic therapy.

There could be other positive effects at work. Greater volume unloading would presumably show observable benefits fairly soon, Carson said. But in UNLOAD, he noted, rehospitalization differences between the ultrafiltration and IV-diuretic groups didn't start to emerge until about a month after acute therapy. "Maybe the neurohormonal milieu is favorably affected by ultrafiltration, and maybe that accounts for the apparent late change in event rates."

Dr Barry M Massie

Massie also pointed out the separation of event-rate curves after 30 days. That timing "may be a fluke, or there's an interesting mechanism that we don't understand." One possible candidate, he said, may relate to the greater salt removal possible with ultrafiltration.

When presenting UNLOAD at the ACC sessions, Costanzo observed that the urine produced by diuresis is hypotonic, whereas fluid removed by ultrafiltration is isotonic—that is, ultrafiltration empties more salt from the body for every liter of removed fluid.

Massie said the impact of that is unknown and it's possible such a mechanism would be involved in any delayed ultrafiltration clinical benefits.

Implications for HF management

Costanzo told heartwire that she hopes "eventually this technology will be used to prevent patients from being admitted with decompensated heart failure."

Massie questioned the practicality of such a strategy. "I'm doubtful that this is going to become a chronic approach, used intermittently by outpatients over long periods of time. I'm not sure the expense would allow it, and I think it would be cumbersome. And, of course, in the outpatient setting you couldn't take off as much fluid in a short period, so it would have to be done frequently."

Moreover, according to Massie, "If you keep on using it, you're going to lose peripheral access over time. I don't think this is something that can be continued forever." Indeed, he said, his nephrologist colleagues have cautioned that repeated ultrafiltrations might ultimately prevent effective venous access in patients who require dialysis later.

On the other hand, he proposed, ultrafiltration could be performed in the emergency room, possibly avoiding admission to an intensive-care unit.

Costanzo said she is on the advisory board of CHF Solutions, has options on future purchase of its stock, and receives honoraria from the company for speaking. Lindenfeld reports receiving research grants from Scios, which markets the acute-HF drug nesiritide.Teerlinksaid he has received research funds from CHF Solutions.