HOUSTON — For patients with acute kidney injury (AKI) requiring dialysis, a conservative strategy of administering dialysis only when deemed necessary by metabolic or clinical indications led to a more rapid recovery of kidney function and lower rates of hypotension than conventional dialysis administered routinely three times per week, in the phase 3 LIBERATE-D trial.

This is the only treatment that has been shown “to foster faster recovery” in these patients, said study senior author Chi-yuan Hsu, MD, of the Division of Nephrology, University of California San Francisco (UCSF), who presented the results at Kidney Week 2025: American Society of Nephrology Annual Meeting. The study was simultaneously published in JAMA.

These results suggest that the conservative dialysis approach is safe and that “the benefit can be greater than the risk. We didn’t know that when we started the study,” Hsu told Medscape Medical News.

Problems With Extended Dialysis

With the development of AKI during hospitalization having potentially life-threatening risks, patients are typically placed on hemodialysis after they have gained hemodynamic stability. They generally continue to receive dialysis three times a week until key criteria of kidney function recovery — involving urine output and creatinine clearance — are met.

The problem is that this can take weeks or months, or even longer, and patients can become dependent on dialysis, which is also linked to higher morbidity and mortality.

A key factor perpetuating extended treatment is that dialysis itself can obscure the creatinine and urine‑output trends clinicians rely on to decide when to stop therapy.

“Thrice-weekly hemodialysis can hinder detection of stabilizing serum creatinine levels and obscure recovery because it may take more than 2 or 3 days for serum creatinine level to reach steady state,” Hsu and colleagues noted in the study.

Furthermore, they wrote, “fluid and solute removal during hemodialysis may also reduce urine output, the other main indicator of recovery.”

This study highlights the fact “that kidney replacement therapy with hemodialysis, although life-saving, is nephrotoxic,” wrote Sushrut S. Waikar, MD, MPH, of Evans Biomedical Research Center, Boston, in an editorial published concurrently in JAMA.

Hemodialysis can also induce transient hypoperfusion of the kidneys, resulting in repetitive ischemic insults and delaying or preventing recovery.

And then there are the array of exposures the blood is subjected to in the process, pointedly summarized in Waikar’s editorial.

“Each dialysis session exposes liters upon liters of blood to plastic tubing, an artificial membrane, and rapid solute/volume fluxes that do not resemble anything in normal human physiology,” he wrote.

Ultimately, “dialysis begets dialysis,” said Hsu. Yet, remarkably, guidelines provide no clear exit strategy, he added.

First author Kathleen Liu, MD, PhD, also from UCSF, pointed out that “there is currently no evidence to guide the delivery of dialysis to facilitate recovery in dialysis-requiring AKI.”

Evaluating Effects and Safety of Dialysis Discontinuation

In the first study of its kind to determine if a strategy of less frequent dialysis would effectively and safely accelerate kidney function recovery in AKI, the researchers conducted a phase 3 trial, enrolling 220 hemodynamically stable patients at four sites in the US who had dialysis-requiring AKI.

The patients, who had a mean age of 56 years and were 33% women, had a mean baseline estimated glomerular filtration rate (eGFR) of 64.8 mL/min/1.73 m², and had initiated kidney replacement therapy a median of 9 days prior to randomization.

Study participants were randomized 1:1 to receive either hemodialysis only when key clinical and metabolic indications were met or to the conventional approach of hemodialysis three times per week, with dialysis cessation based on kidney function recovery assessed by urine output or creatinine clearance.

The specific criteria for resuming hemodialysis in the conservative group included the standard indications for AKI dialysis, including development of metabolic acidosis with arterial pH < 7.15; hyperkalemia with potassium level > 6 mmol/L or 5.5 mmol/L, despite medical treatment; a serum urea nitrogen level > 112 mg/dL; or volume overload with clinically significant hypoxemia. Clinician discretion to initiate dialysis was also allowed, beyond the above criteria.

The median number of treatments — including hemodialysis, pure ultrafiltration, and three sevenths continuous kidney replacement therapy days — per week were 1.8 among patients in the conservative dialysis group compared with 3.1 in the conventional dialysis group.

A total of 218 patients completed the study. For the primary outcome, 64% of patients (70 out of 109) in the conservative hemodialysis group achieved renal recovery at hospital discharge compared with 50% (55 out of 109) in the three times per week dialysis group (difference of 13.8%; P = .04).

Those in the conservative dialysis group recovered kidney function earlier, having had 21 vs 5 consecutive dialysis-free days at day 28 of the study.

Notably, there were no significant differences between the groups in terms of adverse events including hyperkalemia or hypoxemia. The total number of dialysis-associated hypotension events was lower with the conservative dialysis approach (69 vs 97 events), even though the mean ultrafiltration rate per hemodialysis or pure ultrafiltration session was higher in the conservative dialysis group (6.4 vs 4.9 mL/kg/h, averaged across all hemodialysis/pure ultrafiltration sessions for each study participant who received at least one session after randomization).

Of note, the lower dialysis-associated hypotension event rate with the conservative regimen suggests “a potential pathophysiologic pathway by which dialysis itself could have delayed recovery from AKI,” the authors reported in the study.

“These observations suggest that thrice weekly hemodialysis at best masks evidence of and at worst directly delays or impedes recovery of kidney function,” they added.

The current findings are consistent with previous research suggesting benefits of reduced hemodialysis, including one study showing that patients with dialysis-requiring AKI were more likely to recover kidney function at day 28 with just 3 days per week of dialysis vs 6 days per week.

Another study found that patients randomized to receive earlier initiation of renal replacement therapy had as much as a 74% higher rate of continued dependence on dialysis at day 90.

Guidelines currently recommend the thrice-weekly dialysis of such patients, but the current results “suggest the amount of dialysis targeted for delivery to patients with dialysis-requiring AKI should be more nuanced” than indicated by those guidelines, the study authors concluded.

Discontinuing Can Be Nerve Wracking

Hsu noted that the process of discontinuing dialysis and likely seeing a temporary rise in serum creatinine and little improvement in urine output in the first few days, can be anxiety-inducing for clinicians.

In time, however, recovery can begin, he said. He described how in one patient in the conservative dialysis group, serum creatinine levels initially soared in the first 5 days from just over 5 mg/dL at the time of cessation to over 11 mg/dL — but then plateaued and dropped back down to about 3 mg/dL by day 12.

“You just have to hang in there and tolerate the patient having very little kidney function, [though] not at the point of it becoming dangerous, of course,” Hsu noted.

“But off dialysis, the serum creatinine levels can plateau after more than the usual 2- to 3-day gap” with conventional dialysis, and “then drop as the kidneys begin to recover on their own.”

Championing Precision Medicine

Commenting on the research, Prabir Roy-Chaudhury MD, PhD, the current president of the ASN and co-director of the University of North Carolina Kidney Center in Chapel Hill, said he agreed that the findings are practice-changing.

“I think this is a really important study,” he told Medscape Medical News. “In general, it would be nice to have a larger study, but even without a larger study, it is important because it champions a much more individualized and precision medicine approach to for stopping dialysis and for the amount of dialysis to give in patients with AKI.”

Roy-Chaudhury expressed hope for a registry to be established “that looks at people with this type of acute kidney injury and this level of severity so that we can observe whether there is an impact of the [conservative dialysis] approach.”

Caveats of the study include that the criteria of dialysis in the conventional group may have been more stringent than is common, he noted. For instance, criteria to continue dialysis in the conventional group included continuing until patients’ urinary output exceeded 1 L/d without diuretics or > 2 L/d with diuretics.

“Personally, if a patient starts putting out 500 mL or 600 mL of urine, then I’m happy to give them diuretics and [take it from there] to see when they need dialysis again,” Roy-Chaudhury noted.

“So I think the standard of care is likely in-between what was used in the conventional arm and the conservative arms of the study.”

In addition, the study included patients who were on the less severe side of the spectrum, he said. “As the authors note, the results of the study would not apply to patients being ventilated,” or requiring more substantial intervention.

“But what this tells us is we really should be individualizing care,” he said. “Yes, there may be a small chance of [a complication] if [patients] don’t receive dialysis three times per week, but now we have a study showing patients did pretty well without it.”

Hsu agreed that in clinical practice, approaches can run the full gamut between the conventional and conservative approach.

He noted, however, that “there may be a dissociation between what opinion leaders believe practice should optimally be vs real life clinical care.”

For example, he cited a study testing a protocol recommending that dialysis be stopped when patients produced 500 cc of urine output which found that clinicians ignored the recommendation two thirds of the time.

“We hope these [new] findings will help nudge nephrologists towards a more conservative approach in the use of dialysis in these patients, which would do many patients a favor,” Hsu said.

This study was funded by the National Institutes of Health/National Institute of Diabetes and Digestive and Kidney Diseases. Hsu had no disclosures to report. Roy-Chaudhury’s disclosures included consulting and/or other relationships with WL Gore, Medtronic, Becton Dickinson, CorMedix, Humacyte, AstraZeneca, Alexion (AstraZeneca Rare Diseases), Bayer, Panoramic Health, Boston Scientific, BioMed Organ Bank, Vera Therapeutics, and Merit Medical.