Translation of a-t 2023; 54: 87
 

BEWARE MISINFORMATION

TORASEMIDE: FLAWED META-ANALYSIS CLAIMS ADVANTAGES

Your assessment of torasemide (TOREM, generics) as a variant without particular significance continues to be valid even after the TRANSFORM-HF study (a-t 2023; 54: 14). Now, though, there is a new meta-analysis of 10 RCTs by SINGH et al.¹ that finds significantly and considerably fewer hospitalisations in patients treated with torasemide than those treated with furosemide (LASIX, generics). Although the mortality does not differ, do you think the results are a reason to change your assessment?

N.N. (name etc. stated in a-t 11/2023)
Conflict of interest: none

The results of the meta-analysis by SINGH, S. et al.,¹ which in addition to the large TRANSFORM-HF study² with 2,859 patients exclusively evaluated small and very small studies with a maximum of 234 participants, does not change anything about our assessment of torasemide (TOREM, generics; a-t 2023; 54: 14). In addition to all-cause mortality, the clinical endpoints of the meta-analysis, in which the loop diuretics furosemide (LASIX, generics) and torasemide were compared in patients with heart failure, were hospital admissions for any cause and cardiovascular and heart failure-related hospitalisations. While the mortality did not differ between patients treated with furosemide and those treated with torasemide (Risk Ratio [RR] 1.02; 95% Confidence Interval [CI] 0.91-1.15), a significant advantage for torasemide was calculated for the three hospitalisation endpoints. The difference in the endpoint hospital admissions for any cause, in which the TRANSFORM-HF study was also evaluated, was only minimal (RR 1.06; 95% CI 1.01-1.11). Cardiovascular (RR 1.36; 95% CI 1.13-1.65) and heart failure-related hospitalisation (RR 1.65; 95% CI 1.21-2.24) should, however, increase significantly in patients receiving furosemide according to these evaluations.¹

After the TRANSFORM-HF study, two papers^3,4 have the largest weight in the analyses when it comes to the endpoints of cardiovascular and heart failure-related hospital admissions, as these endpoints were not recorded in TRANSFORM-HF and the remaining very small studies evaluated alone would at best provide a neutral result, estimated from the graphical representations. These two papers, though, are two separate evaluations of the same study.⁵ The open-label, one-year study was conducted in the 1990s at two American sites and included a total of 234 patients with heart failure.^3,5

Before the publication of the overall study in 2014, a separate evaluation³ of just the 193 patients in the larger of the two study sites was published that focused on the costs of health care, but in which the results of the actual endpoints of the study (namely hospital admissions) were reported for this subgroup.³ Cardiovascular and heart failure-related hospitalisations were decreased significantly in patients receiving torasemide, both in the overall study⁴ and in the evaluation of the larger site alone.³ The authors rightly rated the results as not definitive.⁴ However, a confirmation with more robust data is still lacking to this day.

The meta-analysis thus evaluates one study twice, thus distorting the data. It should be withdrawn. We have therefore contacted the journal but had not yet received a response at the time of going to press, –Ed.

©  arznei-telegramm (Berlin/Germany), November 2023, protected by copyright laws.

Autor: Redaktion arznei-telegramm - Wer wir sind und wie wir arbeiten

Diese Publikation ist urheberrechtlich geschützt. Vervielfältigung sowie Einspeicherung und Verarbeitung in elektronischen Systemen ist nur mit Genehmigung des arznei-telegramm^® gestattet.