Loop Diuretics Are Heavily Prescribed Outside of HF: That’s a Problem

PRAGUE, Czechia—The number of people taking loop diuretics without a diagnosis of heart failure (HF) may be many times higher than the number of HF patients on these meds, according to an analysis of national health records for the population of Glasgow, Scotland.

The disparity was greater in women, who were 3.7 times more likely to get a loop diuretic without an HF diagnosis, than in men, who were 1.5 times more likely. But for both men and women, prognosis for those prescribed loop diuretics was significantly worse than it was for those with a diagnosis of heart failure. This raises some troubling issues, said John G. F. Cleland, MD, PhD (University of Glasgow, Scotland), who presented the analysis here during a late-breaking session at the European Society of Cardiology Heart Failure 2023 Congress.

“I think whenever you encounter somebody taking a loop diuretic, you need to review this carefully,” he said. “These patients need to be investigated for the possibility that they have underlying cardiovascular dysfunction and heart failure or the possibility that loop diuretics are in themselves, when used inappropriately, lethal. We may be killing these patients by inappropriate use of loop diuretics. That’s also a possibility.”

Other analyses of large randomized controlled trials in atrial fibrillation (AF) and diabetes—populations in which a high prevalence of heart failure might be expected—also have found many more people taking loop diuretics than have formal HF diagnoses. A prior, UK-based study by Cleland and colleagues found a high prevalence of underrecognized heart failure in patients with AF taking loop diuretics, again with a higher preponderance in women.

Whether loop diuretic prescription is a marker for undiagnosed heart failure, or is directly contributing to risk, is unclear.

Commenting for TCTMD, session moderator John J. V. McMurray, MD (University of Glasgow), speculated that the former explanation is more likely, noting that until “fairly recently,” heart failure with preserved ejection fraction (HFpEF) wasn’t a universally agreed upon entity. Many patients may not have got a formal HF diagnosis because their EF was normal.

“I think it is reasonable to assume that many or most patients treated with a loop diuretic could have heart failure,” McMurray said.

Understanding Usage

The aim of the current, epidemiological study, said Cleland, was to look at prescribing patterns and prognosis in a wider swath of patients with cardiovascular disease.

The study drew on electronic health records linked with demographic information, primary care diagnostic codes, as well as prescribing and dispensing information for the city of Glasgow. Of the 1.1 million people in the city sampled between 2009 and 2011, roughly one-quarter had a diagnosis of some form of cardiovascular disease or were taking a medication for CVD, including, for some, a loop diuretic.

Of 160,000 with CVD, 23,963 patients were taking a loop diuretic but had no diagnosis of heart failure; 5,156 had been formally diagnosed with HF but were not taking a loop diuretic, and 7,844 patients had been diagnosed with HF and prescribed a loop diuretic.

We may be killing these patients by inappropriate use of loop diuretics. That’s also a possibility. John G. F. Cleland

Delving into the possible reasons for prescribing the drugs in the absence of confirmed HF, Cleland and colleagues found that patients without a HF diagnosis had little in the way of end-stage renal disease and few were taking three or more antihypertensive agents, which would point to resistant hypertension, both of which might have explained loop diuretic prescribing.

More patients in the loop diuretic group than in the HF group had a diagnosis of chronic obstructive pulmonary disease, “so there may have been some mistaken diagnosis there,” said Cleland. Otherwise investigators found little in the medical records to explain why diuretic prescribing was so high.

One clue might lie in the nearly 50,000 patients who had undergone echocardiography: most of these patients had some degree of left atrial enlargement, but this was more prevalent in the loop diuretics-only patients, potentially pointing to underlying, undiagnosed HFpEF.

Clinical Implications

What’s striking, however, is that patients on loop diuretics had higher rates of hospital admissions per patient year of risk than did HF patients not taking loop diuretics. And while most of those hospitalizations were not for cardiovascular disease, said Cleland, what’s clear is that “the whole picture is dominated by women taking loop diuretics: they contribute more hospitalizations than any of the other groups.”

Mortality numbers are also sobering, Cleland suggested. Among patients taking a loop diuretic without having been diagnosed with heart failure, 40% of men had died within 5 years, as had 30% of women. “And that’s deaths before the diagnosis of heart failure,” he said. “You have to add those deaths that occur after a diagnosis of HF on top of that mortality, so you can see it’s pretty horrendous.”

And while there was also an increased risk of mortality associated with a diagnosis of HF alone, “it really is dwarfed by the impact of loop diuretics,” he continued. “This is by far, numerically and in terms of risk, a much larger problem. Indeed, if you restrict the analysis to people with an ischemic heart disease diagnosis, heart failure no longer has any prognostic significance until you start taking a loop diuretic. In this population, the majority of patients with cardiovascular disease die only after a diagnosis of HF or after they’ve been instituted on regular doses of loop diuretics.”

Takeaways

Following Cleland’s talk, session co-moderator Gerasimos Filippatos, MD (National & Kapodistrian University of Athens Medical School, Greece), asked what in practicable terms physicians should take away from this provocative analysis.

Cleland replied that he and his colleagues are digging deeper by contacting patients who were prescribed loop diuretics and, where warranted, seeking further cardiac evaluation.

“We’re also going through the GPs because it seems to be primary care physicians who are initiating the loop diuretics, although sometimes it’s the respiratory physicians: they’re the second culprit group,” he said. “It’s such an easy audit to just go through a primary care list on a loop diuretic and ask why.”

Speaking with TCTMD, Cleland pointed to a lack of awareness and interest in heart failure among general practitioners as part of the problem, as well as a “knee-jerk reflex” of prescribing a loop diuretic when patients complain of swollen ankles. In many cases, a GP might simply be renewing a prescription made following an orthopedic or respiratory hospitalization without questioning why it was prescribed in the first place, on the perception that the agents are safe.

Asked whether he thinks loop diuretics are truly causing harm or are more likely to be a marker of missed opportunities for diagnosis and optimal medical management of HF, Cleland said he thinks it may be a bit of both.

“I suspect it’s a heterogenous situation,” he said. Loop diuretics are known to increase the risk of osteoporosis, renal dysfunction, and low potassium, all of which can cause morbidity associated with increased mortality. The direct influence of loop diuretics may also differ by sex, he suggested.

In the 30% of the patients prescribed loop diuretics without an HF diagnosis who were men, undiagnosed heart failure would likely to explain their use, Cleland said.

But fully 70% of patients who fit this description were women. In this much larger group, many are likely getting a diuretic inappropriately, said Cleland. He estimated that while two-thirds of the mortality in women prescribed loop diuretics might be explained by undiagnosed heart failure—related to hypertrophic cardiomyopathy, amyloidosis, HFpEF, valve disease, or other pathologies—as much as 30% might be related more directly to harmful effects of the drugs.

As for the possible direct harmful effects of inappropriate prescribing, McMurray didn’t rule it out. “All drugs have adverse effects, and diuretics cause electrolyte disturbances and worsening kidney function, so both of those are bad and could increase mortality,” he said.

An easy first step, Cleland suggested, would be to order a NT-proBNP test for a patient taking a loop diuretic without a clear, persistent indication. If the test is elevated, further cardiac workup would be appropriate. Rather than flood the system, however, “if the natriuretic peptides are low then I would be cautiously withdrawing the loop diuretic, he said.

And for patients with resistant hypertension, Cleland proposed consider switching them to a thiazide diuretic.