Hypervolemia is a common finding in patients with chronic heart failure.1 As a result, loop diuretics have become a cornerstone of chronic heart failure management for the treatment of congestion.2,3 Despite the prevalent use of these agents, there is inconsistent evidence in large trials regarding the optimal choice of loop diuretic and their impact on clinical outcomes.4,5
Current heart failure guidelines recommend loop diuretics in patients with signs of hypervolemia, yet there are no recommendations regarding which should be used as first-line.6,7 Oral furosemide is the most commonly prescribed loop diuretic in patients with heart failure; however, it has inconsistent efficacy due to poor bioavailability, with absorption rates ranging from approximately 10 to 100%.8,9 In contrast, oral torsemide has a consistent bioavailability of up to 90% and is associated with other beneficial effects, including protection against cardiac fibrosis and aldosterone receptor blockade.10–12
There are prior studies that have compared furosemide and torsemide in chronic heart failure patients and their association with clinical outcomes, yet results have been inconsistent.13,14 The Torsemide in Congestive Heart Failure (TORIC) trial, a postmarketing surveillance study, illustrated a reduction in mortality and improvement in symptoms in patients treated with torsemide compared with those on furosemide.13 However, more recent large observational studies have had conflicting results.14,15 Thus, the aim of our study was to systematically review and quantitively analyze the most up-to-date literature comparing furosemide and torsemide in chronic heart failure patients and their effect on outcomes such as mortality, heart failure hospitalizations and improvement in heart failure symptoms.
The meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta Analyses guidelines. Medline and Cochrane databases were systematically reviewed for all relevant studies comparing oral furosemide with oral torsemide in ambulatory heart failure patients until May 2018. The search algorithm used was (furosemide OR Lasix) AND (torsemide OR torasemide). No language restrictions were applied to the search.
Study selection and data extraction
Studies that were considered eligible for the present analysis included randomized clinical trials (RCTs) or observational studies that compared oral furosemide and oral torsemide in chronic heart failure patients and studies that reported intermediate-term (5–12 months) clinical outcomes such as: heart failure readmission status, all-cause mortality or changes in New York Heart Association (NYHA) class. Studies that only included less than 5-month or more than 12-month outcomes, all animal studies, pediatric studies, abstracts, reviews and case reports were excluded. The time interval of the included studies was not prespecified but was chosen because most of the studies on this topic reported outcomes during this time period and we thought that it was most clinically relevant. When duplicate study populations were identified, those with larger cohorts were included. Two investigators (J.A.M. and B.K.H.) independently screened all titles and abstracts that were eligible for inclusion. After screening, full-text versions of the studies were then analyzed by the two independent reviewers. Any uncertainty or disagreement was resolved by a third reviewer (D.G.K.).
Risk-of-bias assessment was performed for all studies. The Cochrane risk-of-bias tool (RoB 2, London, England) was used for randomized studies and the ROBINS-I tool was utilized for nonrandomized studies.16,17 Randomized studies were evaluated for risk of bias in the following categories: sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data and selective outcome reporting (refer to Table, Supplemental Digital Content 1, http://links.lww.com/JCM/A166). Observational studies were assessed for bias: due to confounding, participant selection, departures from intended interventions, measurements of interventions, missing data, selection of reported results and measurement of outcomes (refer to Table, Supplemental Digital Content 2, http://links.lww.com/JCM/A166).
Data that were extracted from each study included: study name, methodology of the study, location, time period, inclusion criteria, exclusion criteria, the number of study groups, diuretic dosages used in the study, number of patients included in each study, age, sex, race, percentage with hypertension, diabetes, prior myocardial infarction, prior stroke, coronary artery disease, chronic kidney disease or end-stage renal disease, a history of coronary artery bypass surgery, percutaneous coronary interventions and medication use including: angiotensin converting enzyme (ACE) inhibitors, beta blockers, spironolactone and prior diuretic use. The main outcomes studied in this meta-analysis included all-cause mortality, heart failure readmissions and change in NYHA functional class.
Odds ratios (ORs) with corresponding 95% confidence intervals (CIs) were used for the outcomes. Unadjusted ORs were used for all studies. A random effects model was used to account for heterogeneity among studies. Heterogeneity was assessed with the Higgins I-square statistic.18I2 greater than 75% indicated significant heterogeneity.18 Forest plots were used to graphically display the effect size in each study and the pooled estimates. Subgroup analyses were conducted for all the outcomes for RCT and observational studies. Funnel plots were not used for assessment of publication bias since there was not a single outcome with more than ten eligible studies. A P value less than 0.05 was considered significant. STATA 14.1 (StataCorp, College Station, Texas, USA) was used as statistical software.
In total, 336 studies were screened and 17 full-text studies were analyzed (Fig. 1). Three studies were excluded even after meeting initial inclusion criteria because of overlapping populations in other included studies. One large observational study by Mentz et al. was excluded as only long-term outcomes, that is 5-year mortality, was analyzed. In the final analysis, 14 studies were included, 10 of which were RCTs and 4 were observational studies. Most of the studies were performed in the United States and Europe, with two studies completed in Japan. All of the studies were written in English.
In total, 8127 patients were included in the study, 5729 were prescribed furosemide and 2398 were given torsemide during the study period. All patients included were ambulatory chronic heart failure patients requiring oral loop diuretic therapy. As shown in Table 1, diuretic regimens in the included studies consisted of either a fixed daily dose of furosemide and torsemide or doses that were modified over the study period based on signs and symptoms of heart failure. Most of the included patients had NYHA class II–III symptoms, although few studies included patients with NYHA class IV symptoms. The mean age of patients ranged from 63 to 75 and the percentage of female patients ranged from 21.6 to 57.5%. Cardiovascular risk factors such as hypertension, diabetes and coronary artery disease were relatively common in the included studies, although a few studies did not report baseline characteristics. Beta blocker use ranged from 9.5 to 100% and ACE inhibitor use from 30.8 to 100% in the included studies (refer to Table, Supplemental Digital Content 3, http://links.lww.com/JCM/A166).
Eight studies were included in the mortality analysis: five RCTs and three observational studies. Follow-up period of the included studies ranged from 5 to 12 months. As shown in Fig. 2, in the overall analysis, there was no significant difference in all-cause mortality among heart failure patients on oral furosemide as compared with oral torsemide (OR 1.01, CI 0.64–1.59, I2 = 65.8%). No difference in all-cause mortality between groups was also seen in subgroup analyses of RCTs (OR 1.12, CI 0.70–1.80) and observational studies (OR 0.97 CI 0.44–2.13). Among RCTs, there was no significant heterogeneity observed (I2 = 0.0%, P = 0.86); however, there was significant heterogeneity among observational studies in the subgroup analysis (I2 = 89.2%, P < 0.005).
Heart failure readmissions
Five studies (four RCTs and one observational study) analyzed the impact of heart failure readmissions in patients with heart failure taking torsemide versus furosemide (Fig. 3). Follow-up periods of the included studies ranged from 6 to 12 months. In the overall analysis, furosemide was associated an increased risk of heart failure readmissions as compared with torsemide (OR 2.16, CI 1.28–2.64, I2 = 0.0%). In the subgroup analysis, among RCTs, those taking furosemide had a higher risk for heart failure readmissions as compared with those taking torsemide (OR 2.04, CI 1.16–3.60, I2 = 0.0%). Only one observational study reported on heart failure readmissions, which illustrated no significant difference in heart failure readmissions between groups (OR 2.91, CI 0.78–10.91).
New York Heart Association class improvement
As illustrated in Fig. 4, nine studies (seven RCTs and two observational studies) reported outcomes on changes in NYHA status in those taking furosemide versus torsemide. Follow-up time ranged from 6 to 12 months. In the overall analysis, heart failure patients taking torsemide were more likely to have improvement in NYHA status as compared with those on furosemide (OR 0.73, CI 0.58–0.93). This finding was consistent in the subgroup analysis among observational studies (OR 0.65, CI 0.50–0.85); however, there was no statistical difference between groups among RCTs (OR 0.91, CI 0.61–1.35). There was no statistically significant heterogeneity among studies in the overall and subgroups analyses (overall I2 19.6, P = 0.268, RCT I2 10.2%, P = 0.352, observational I2 31.6%, P = 0.226).
This was a systematic review and meta-analysis of RCTs and real-world studies comparing torsemide versus furosemide for patients with heart failure. We did not find a difference in mortality but in our analysis, patients treated with torsemide were more likely to have an improved NYHA status and less likely to be readmitted for heart failure when compared with patients on furosemide.
There are only a few prior studies that have directly compared furosemide versus torsemide and their association with mortality in chronic heart failure patients. The TORIC trial, a postmarketing surveillance study, was one of the first landmark trials that illustrated a survival benefit for heart failure patients taking torsemide compared with furosemide. Those prescribed torsemide had more than a 50% reduction in mortality compared with those taking furosemide.13 That being said, the results of the TORIC trial have been questioned due to methodological shortcomings of the trial design.8
Prior to the TORIC trial, studies comparing furosemide and torsemide in patients with chronic heart failure had demonstrated no mortality difference, yet most of these studies were underpowered to detect a mortality difference.20,22,23 More recently, Mentz et al.27 analyzed 4177 patients from the Acute Study of Clinical Effectiveness of Nesiritide in Decompensated Heart Failure trial and similarly demonstrated no mortality difference between patients on furosemide versus torsemide after 6 months. However, in a parallel analysis of the Placebo-controlled Randomized study of the selective A1 adenosine receptor antagonist KW-3902 for patients hospitalized with acute heart failure and Volume Overload to assess Treatment Effect on Congestion and renal function Trial (PROTECT), Mentz et al.14 found an increased mortality after 5 months in those taking torsemide compared with furosemide. Furthermore, in a third study performed by Mentz et al., which was not included in the present analysis as it evaluated long-term outcomes, torsemide was associated with an increased mortality compared with furosemide in an unadjusted analysis but no difference after multivariate adjustment.15 In our current analysis, including the overall analysis and subgroup analyses of RCTs and observational studies, loop diuretic choice did not affect mortality in chronic heart failure patients. Although prior meta-analyses illustrated similar outcomes with respect to mortality, they included studies only until 2003.29,30 Since then, there have been large observational studies that have analyzed mortality outcomes in patients with heart failure on torsemide versus furosemide, which have been included in the present analysis.14,27 A possible explanation to explicate the inconsistent data regarding torsemide and mortality outcomes is that patients with more severe heart failure were prescribed torsemide in these observational studies. This could explain the results, showing either no difference or an increased mortality risk with torsemide, despite the other known benefits of torsemide compared with furosemide. A large RCT comparing furosemide and torsemide in the era of the most updated heart failure therapy is needed to reduce these biases and other confounding factors. The Torsemide Comparison with Furosemide for Management of Heart Failure trial is an on-going RCT on this topic.31
Moreover, our analysis demonstrated a reduction in heart failure readmissions and an improvement in baseline NYHA status with torsemide as compared with furosemide. There are multiple hypotheses that could explain these findings. First, torsemide has better oral bioavailability, a longer half-life and is more potent compared with furosemide, all of which likely leads to a more consistent diuresis.10 Second, torsemide also been shown to have aldosterone receptor blocking activity, leading to less renin-angiotensin-aldosterone system activation.12 In contrast, furosemide increases renin levels, which perhaps exacerbates left ventricular dysfunction and leads to worsening heart failure.32 Third, few studies have illustrated that torsemide has an antifibrotic effect compared with furosemide, although it is unclear what clinical benefit this may have in heart failure patients.11,24
There have been three prior meta-analyses on this topic; nevertheless, our study is unique in that it includes the most up to date literature on the topic and it is the most expansive analysis with respect to included studies and outcomes. DiNicolantonio performed a meta-analysis of two RCTs comparing torsemide versus furosemide and outcomes such as all-cause mortality and heart failure and cardiovascular readmissions. Similarly to our findings, DiNicolantonio illustrated a reduction in heart failure and cardiovascular admissions with torsemide but no statistical difference in all-cause mortality.29 Our analysis added to these findings by including other RCTs and observational studies that were not included in that meta-analysis. In addition, our analysis included more expansive outcomes such as change in NYHA status. Moreover, Bikdeli et al. showed similar findings with respect to all-cause mortality, with no statistical difference between those on torsemide compared with furosemide. Three studies, both RCTs and observational studies, were included in that mortality analysis compared with eight studies in our analysis. Bikdeli et al.30 also analyzed change in NYHA status in five studies and demonstrated no difference between those on torsemide versus torsemide. Our analysis included seven RCTs and two observational studies, which illustrated an improvement in NYHA class with those taking torsemide compared with furosemide. Lastly, a recently published meta-analysis, which included three RCTs, had similar findings; however, of the three RCTs included, two had overlapping study populations.33 If one of these studies would have been excluded, the data would have been identical to DiNicolantonio's prior analysis.
This study has several important limitations. First, many of the studies included were real-world observational studies, including some which were retrospective in nature. As such, the risk of bias is higher with these studies. Second, some of the early studies included were performed at a time where other heart failure therapy, such as beta-blocker and ACE inhibitors, was not widely used. For example, in the TORIC trial, only 9.5% of patients were on a beta-blocker and 30% were taking ACE inhibitors.13 Third, in a few of the included studies, the prescribed doses of furosemide and torsemide were not equivalent. Fourth, most of the studies were not performed using a new-user design and, thus, many of the patients were on diuretics prior to the study period. We did not have access to the number of patients on diuretics prior to the study periods and, therefore, were unable to perform subgroup analyses. Fifth, data regarding postdischarge loop diuretic adjustments and crossover rates between groups were not available for all studies, which could possibly confound our results. Sixth, there was heterogeneity among the individual studies including doses of diuretics used, comorbidities in the individual groups and the severity of heart failure, all of which could have confounded our results. Seventh, lastly, the follow-up of the included studies was not always the same. This could have created additional heterogeneity. Despite the aforementioned limitations, we think that our study benefits from our meticulous study design and strict adherence to systematic review methodology.
Our meta-analysis illustrates that torsemide may be more effective at reducing heart failure readmissions and improving clinical symptoms compared with furosemide in patients with chronic heart failure. Yet, we did not find a difference in intermediate-term mortality outcomes with torsemide versus furosemide. Further studies, including a large RCT in the current era of guideline directed medical therapy for heart failure, are needed to further elucidate whether the choice of loop diuretic in patients with chronic heart failure is important with respect to clinical outcomes.
Conflicts of interest
There are no conflicts of interest.
1. Metra M, Teerlink JR. Heart failure
2. Wu MY, Chang NC, Su CL, et al. Loop diuretic strategies in patients with acute decompensated heart failure
: a meta-analysis
of randomized controlled trials. J Crit Care
3. Fudim M, O’Connor CM, Mulder H, et al. Loop diuretic adjustments in patients with chronic heart failure
: insights from HF-ACTION. Am Heart J
4. Faris RF, Flather M, Purcell H, et al. Diuretics for heart failure
. Cochrane Database Syst Rev
5. Ahmed A, Husain A, Love TE, et al. Heart failure
, chronic diuretic use, and increase in mortality and hospitalization: an observational study using propensity score methods. Eur Heart J
6. Yancy CW, Jessup M, Bozkurt B, et al. 2013 ACCF/AHA guideline for the management of heart failure
: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol
7. Ponikowski P, Voors AA, Anker SD, et al. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure
: the Task Force for the diagnosis and treatment of acute and chronic heart failure
of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure
Association (HFA) of the ESC. Eur J Heart Fail
8. Buggey J, Mentz RJ, Pitt B, et al. A reappraisal of loop diuretic choice in heart failure
patients. Am Heart J
9. Murray MD, Haag KM, Black PK, et al. Variable furosemide
absorption and poor predictability of response in elderly patients. Pharmacotherapy
10. Vargo DL, Kramer WG, Black PK, et al. Bioavailability, pharmacokinetics, and pharmacodynamics of torsemide
in patients with congestive heart failure
. Clin Pharmacol Ther
11. Lopez B, Gonzalez A, Beaumont J, et al. Identification of a potential cardiac antifibrotic mechanism of torasemide in patients with chronic heart failure
. J Am Coll Cardiol
12. Yamato M, Sasaki T, Honda K, et al. Effects of torasemide on left ventricular function and neurohumoral factors in patients with chronic heart failure
. Circ J
13. Cosin J, Diez J, investigators T. Torasemide in chronic heart failure
: results of the TORIC study. Eur J Heart Fail
14. Mentz RJ, Velazquez EJ, Metra M, et al. Comparative effectiveness of torsemide
in heart failure
patients: insights from the PROTECT trial. Future Cardiol
15. Mentz RJ, Buggey J, Fiuzat M, et al. Torsemide
in heart failure
patients: insights from Duke University Hospital. J Cardiovasc Pharmacol
16. Higgins JP, Altman DG, Gotzsche PC, et al. The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ
17. Sterne JA, Hernan MA, Reeves BC, et al. ROBINS-I: a tool for assessing risk of bias in nonrandomised studies of interventions. BMJ
18. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ
19. Düsing R, Piesche L. Second line therapy of congestive heart failure
with torasemide. Prog Pharmacol Clin Pharmacol
20. Noe LL, Vreeland MG, Pezzella SM, et al. A pharmacoeconomic assessment of torsemide
in the treatment of patients with congestive heart failure
. Clin Ther
21. Spannheimer A, Goertz A, Dreckmann-Behrendt B. Comparison of therapies with torasemide or furosemide
in patients with congestive heart failure
from a pharmacoeconomic viewpoint. Int J Clin Pract
22. Murray MD, Deer MM, Ferguson JA, et al. Open-label randomized trial of torsemide
compared with furosemide
therapy for patients with heart failure
. Am J Med
23. Muller K, Gamba G, Jaquet F, et al. Torasemide vs. furosemide
in primary care patients with chronic heart failure
NYHA II to IV – efficacy and quality of life. Eur J Heart Fail
24. Lopez B, Querejeta R, Gonzalez A, et al. Effects of loop diuretics
on myocardial fibrosis and collagen type I turnover in chronic heart failure
. J Am Coll Cardiol
25. Kasama S, Toyama T, Hatori T, et al. Effects of torasemide on cardiac sympathetic nerve activity and left ventricular remodelling in patients with congestive heart failure
26. TORAFIC Investigators Group. Effects of prolonged-release torasemide versus furosemide
on myocardial fibrosis in hypertensive patients with chronic heart failure
: a randomized, blinded-end point, active-controlled study. Clin Ther
27. Mentz RJ, Hasselblad V, DeVore AD, et al. Torsemide
in patients with acute heart failure
(from the ASCEND-HF Trial). Am J Cardiol
28. Trippel TD, Van Linthout S, Westermann D, et al. Investigating a biomarker-driven approach to target collagen turnover in diabetic heart failure
with preserved ejection fraction patients: effect of torasemide versus furosemide
on serum C-terminal propeptide of procollagen type I (DROP-PIP trial). Eur J Heart Fail
29. DiNicolantonio JJ. Should torsemide
be the loop diuretic of choice in systolic heart failure
? Future Cardiol
30. Bikdeli B, Strait KM, Dharmarajan K, et al. Dominance of furosemide
for loop diuretic therapy in heart failure
: time to revisit the alternatives? J Am Coll Cardiol
31. Clinicaltrials.gov. National Library of Medicine. TRANSFORM-HF: ToRsemide
compArisoN With furoSemide
FORManagement of Heart Failure
. Identifier: NCT03296813. Bethesda MD, 2019. https://ClinicalTrials.gov/show/NCT03296813
32. McCurley JM, Hanlon SU, Wei SK, et al. Furosemide
and the progression of left ventricular dysfunction in experimental heart failure
. J Am Coll Cardiol
33. Shah P, Patel H, Mithawala P, et al. Torsemide
in heart failure
patients: a meta-analysis
of randomized controlled trials. Eur J Intern Med