No Mortality Difference for ICU Loop Diuretics in Meta-Analysis

Thomas R. Collins

August 01, 2022

In a meta-analysis, researchers found no difference in all-cause mortality for intensive-care patients with fluid overload who were treated with loop diuretics, compared with those given a placebo or who did not undergo an intervention.

The study, published recently in Annals of Intensive Care, also found a lower rate of serious adverse events (SAEs) among those given diuretics, but the authors say the certainty of the evidence is low and that there is a high risk of bias in the studies.

"The level of evidence is low, especially due to few and small randomized trials in the field," said Sine Wichmann, MD, anesthesiologist at Copenhagen University Hospital, who is the primary author of the study.

After a review of the available literature, researchers were able to include only six trials in which loop diuretics were compared to placebo or no intervention, two trials comparing a loop diuretic with another loop diuretic, and two trials comparing loop diuretics with another kind of diuretic. All the trials were small, with 12 to 249 participants.

Only four trials, totaling 359 participants, reported all-cause mortality. There was no difference between loop diuretics and placebo or no intervention, with a follow-up between 29 and 90 days. Researchers performed a trial sequential analysis (TSA), a test used in meta-analyses to account for the subject size relative to the outcome being measured. In TSAs, the bar for significance can be increased compared to the standard threshold in trials in which the number of participants is small. In this meta-analysis, for mortality, there was no significant difference before or after TSA.

None of the trials reported on the proportion of patients with one or more SAEs, but six reported events the researchers considered to be SAEs. They determined the SAE with the highest event rate in each trial. There were fewer SAEs in the group treated with loop diuretics, but that finding was not statistically significant after TSA adjustment (CI, 0.55 – 1.20).

Wichmann said the evidence does not tell clinicians anything definitive about subgroups of patients who might benefit from diuretics. But she said that on the basis of this analysis, care should be taken in the use of loop diuretics, given the lack of an overall indication of mortality reduction.

"I think this review should make the ICU [intensive care unit] physicians look at their patients with fluid overload and assess if they might benefit from diuretics," Wichmann noted. "If an ICU patient has organ dysfunctions due to fluid overload, diuretics might be the best choice, but if the patient does not have any organ dysfunctions due to fluid overload, maybe the patient will be better off without."

She said that a problem in the field is that there is no clear definition of fluid overload, nor is it clear when fluid overload becomes a relevant risk factor — whether it's 5% overload, 10%, or higher.

She added there is "no gold standard" for assessing fluid status among ICU patients.

"Several surrogate measurements are used, but none are precise," Wichmann said. "The medicine must be individually titrated to effect. All this makes trial design more difficult."

Ian McCoy, MD, assistant professor of medicine at the University of California, San Francisco, who has researched diuretics outcomes, said it can be surprising that no large, randomized controlled trials have established the safety and efficacy of diuretics, given how often they are used. He pointed out that furosemide, one of the most commonly used loop diuretics, was developed in the 1960s at about the same time that the authority of the US Food and Drug Administration was expanded to require assessment of safety and efficacy before therapies hit the market.

"Now 60-plus years later, diuretics are so established in clinical care, and some of their effects are so readily apparent ― one can observe urine output increase and watch a patient with pulmonary edema feel less short of breath in less than 30 minutes ― that most clinicians will not deem there to be clinical equipoise for randomization in most clinical situations," he said.

He agreed that the evidence remains weak and that practice-changing recommendations can't be made until larger randomized, controlled trials are conducted. But he said diuretic-vs-placebo trials are unlikely.

He said that it is more likely that there will be trials such as FACTT and CHAMPION "comparing different fluid management strategies that incorporate diuretics in specific clinical contexts."

Wichmann has received research funding from the Novo Nordisk Foundation and other foundations. McCoy has disclosed no relevant financial relationships.

Ann Intensive Care. Published online Jun 13, 2022. Full text

Tom Collins is a freelance writer in South Florida who has written about medical topics from nasty infections to ethical dilemmas, runaway tumors to tornado-chasing doctors. He travels the globe gathering conference health news and lives in West Palm Beach.

For more news, follow Medscape on Facebook, Twitter, Instagram, and YouTube.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.