Transform Hf Torsemide Not Better Than Furosemide For Hospitalized Heart Failure

The head-to-head battle of loop diuretics suggests that proper dosing rather than drug selection should be the focus of clinicians.

CHICAGO, IL—The choice of loop diuretic does not make a difference in how patients hospitalized for heart failure fare, at least in terms of risk of death or readmission, the TRANSFORM-HF trial shows. A decompression strategy using torsemide was no better than using furosemide for the risk of all-cause mortality—the primary endpoint—or a composite of all-cause mortality and total hospitalizations, Robert Mentz, MD (Duke Clinical Research Institute , Durham, NC), one of the trial’s principal investigators, reported here at the 2022 American Heart Association Scientific Sessions. This contrasts with some previous observational research that suggested that torsemide may have an advantage over the more widely used furosemide. “Although we were disappointed that we didn’t show better torsemide, I think we answered the question,” Mentz told TCTMD, noting that issues related to crossovers, treatment interruptions and the COVID-19 pandemic could have affected the results. The takeaway, he said, is that clinicians should focus on getting patients on the right dose of loop diuretics rather than which agent to choose or switch from one to another. And furthermore, he added, “for those other heart failure treatments that have been shown to improve outcomes, we really need to start and titrate them.” The TRANSFORM-HF trial In practice, furosemide is more commonly used than torsemide, probably because it is more widely available and clinicians are used to trying it for decongestion in their heart failure patients. However, previous preclinical and clinical research has shown that torsemide may have some advantages over furosemide, including more stable bioavailability, long-lasting diuretic action, and anti-aldosterone and antifibrotic effects not seen with furosemide. The idea that these differences may translate into better clinical outcomes, including reduced mortality, has been supported by both previous observational studies and clinical experience that leads many physicians to switch patients to torsemide after they fail to respond to furosemide. However, there was a need for a large randomized trial to compare the two head-to-head. TRANSFORM-HF, a pragmatic, open-label trial conducted at 60 US centers, was designed to address this need. The researchers randomized 2,859 patients (mean age 65 years, 37% women) hospitalized for heart failure to a loop diuretic strategy that included either torsemide or furosemide, with dosing left to the discretion of the treating physicians. Approximately two-thirds of patients were receiving a loop diuretic before admission, with approximately 80% of them receiving furosemide. Although we were disappointed that we didn’t show that torsemide was better, I think we answered the question. Robert Mentz There were no limitations based on LVEF. Most patients (64%) had a reduced ejection fraction of 40% or lower, and this group was well treated with other heart failure medications. Most were taking a β-blocker (82%) and an ACE/ARB/ARNI inhibitor (68%), with 44% on a mineralocorticoid receptor antagonist and 8% on a sodium-glucose cotransporter 2 (SGLT2) inhibitor. The trial was designed to streamline procedures for both patients and participating centers, and there were no mandatory in-person follow-up visits. Instead, patients were called at 30 days, 6 months, and 12 months of age. Mortality data were drawn from the National Death Index. Through a median follow-up of 17.4 months, there was no difference in all-cause mortality between the torsemide and furosemide groups (26.1% vs 26.2%; HR 1.02; 95% CI 0.89–1 ,18), with findings consistent across subgroups. Similarly, there was no difference between groups for composite all-cause mortality or all-cause hospitalization at 12 months (47.3% vs 49.3%; HR 0.92; 95% CI 0.83-1, 02) or for total hospitalizations (37.5% vs. 40.4%; rate ratio 0.94; 95% CI 0.84-1.07). The lack of a significant advantage for torsemide in endpoints was consistent in a prespecified on-treatment analysis. Important answers, but questions remain Commenting on TCTMD, Mary Norine Walsh, MD (Ascension St. Vincent Heart Center, Indianapolis, IN), past president of the American College of Cardiology, said that TRANSFORM-HF is valuable because it was done in a realistic way, it answered a lot. -Permanent field question and a different group of patients was enrolled. “It’s important evidence to show that there’s not much of a difference, at least in the hospital population, between these two drugs,” he said. Walsh pointed to some limitations also highlighted by the researchers, including the potential impact of the COVID-19 pandemic and the evolution of guideline-directed medical treatment for heart failure that occurred during the trial, with increasing use of angiotensin receptor- neprilysin inhibitor sacubitril/valsartan (Entresto; Novartis) and SGLT2 inhibitors. Another key factor in interpreting the findings, Walsh added, is that “we cannot extend these data to patients treated outside the hospital.” He said he was not surprised that no difference was seen between factors in this trial because “using all-cause mortality as a primary endpoint is a high bar,” and added that questions about other outcomes important to patients and/or health systems -e.g. decongestion rate, weight loss, and length of stay—were not answered in TRANSFORM-HF. For now, Walsh said, “as best we can tell, it doesn’t matter which diuretic is used to decongest patients hospitalized for acute heart failure or acute chronic heart failure, but we don’t yet understand the differences for outpatients. “ Biykem Bozkurt, MD, PhD (Baylor College of Medicine, Houston, TX), discussing the results at a press conference, agreed that there are some issues that could not be addressed by the trial, noting the possibility that differences between diuretics have arisen for other outcomes such as successful decompression, cardiovascular death, heart failure hospitalizations, and urgent care visits. If torsemide is more potent and effective, there could also be cost implications, he said. In addition, TRANSFORM-HF does not provide information on the use of torsemide in patients routinely treated with the agent, including those unresponsive to furosemide (considered diuretic-resistant) and those with cardiorenal syndrome or chronic kidney disease, who are prone to diuretic resistance, Bozkurt said. But overall, despite some limitations, the trial shows that torsemide and furosemide have similar efficacy, “allowing clinicians to continue what they’re doing, which is treating according to the indications and the specific phenotypes envisioned,” he said. “So I don’t think the study will change our approach to the patients we treat with torsemide, which tends to be diuretic-resistant people where we need bioavailability and efficacy.” Real Trial Lessons The insights gained from the pragmatic design of TRANSFORM-HF will be used to inform future comparative effectiveness studies, Mentz said. First, the trial showed that the use of broad eligibility criteria and improved protocols incorporated into routine care resulted in a more diverse population of participants—overrepresented by women and blacks who had passed many previous heart failure trials—and prompted strong recruitment, which was seen even during the COVID-19 pandemic. The researchers also identified opportunities to enhance patient adherence and retention, he said, citing crossovers and loss to follow-up. “In this context, this real-world comparative effectiveness study provides generalizable results that apply to our patients,” Mentz said during the press conference. In addition, he told TCTMD, there is now a rich database reflecting what happens in routine practice that can be used to conduct future analyzes to answer important questions about diuretics and other drugs. Of note, the TRANSFORM-HF investigators collected data on quality of life and depression, which will be reported later. Addressing the implications for future research, Walsh said: “This type of trial, which enrolled a real population in a very realistic way, is impressive and trials in the future should consider this type of design.”