Introduction

Flozins, also referred to as sodium‑glucose cotransporter 2 inhibitors (SGLT2is), are the only pharmacological treatment that improves prognosis in heart failure (HF) patients regardless of left ventricular ejection fraction (EF). Since the introduction of the 2021 European Society of Cardiology (ESC) guidelines¹ and their subsequent 2023 update,² SGLT2is—along with diuretics—have been included in class I recommendations for all HF phenotypes. Unfortunately, in Poland, flozins are essentially reimbursed only for patients with HF with reduced (HFrEF) and mildly reduced EF (HFmrEF).

The aim of this report is to summarize the use of flozins in Poland based on the HEROES (Heart Failure Observational Study of the Polish Cardiac Society) database (https://heroes.umed.pl).

Patients and methods

HEROES is a multicenter observational study conducted between 2022 and 2024. Detailed study rationale and methodology were published elsewhere,³ and the dataset is available online (http://dx.doi.org/10.60941/JVH1‑5190). Using information from the HEROES database, we summarized trends for flozin treatment across the full spectrum of HF, based on reports of flozin use at discharge from hospital or the last outpatient visit.

As the pace of recruitment was dynamically changing during the study period and depended on the number of active sites, we divided the study cohort into 4 equal quartiles (Q1–Q4) based on consecutive patient enrollment, rather than using even time quartiles. This approach allowed us to assess the implementation of ESC guidelines over time with a similar number of patients in each cohort: Q1, April 28, 2022 to October 3, 2022 (n = 355); Q2, October 4, 2022 to January 22, 2023 (n = 355); Q3, January 23, 2023 to April 20, 2023 (n = 356); Q4, April 21, 2023 to January 31, 2024 (n = 356). We also analyzed the frequency of use of specific SGLT2is and the most commonly reported reasons for not initiating SGLT2i treatment in HF patients.

The study protocol was approved by the Bioethical Committee at the Medical University of Lodz (RNN/316/20/KE, with update KE/762/23).

Results

Patient characteristics and data regarding SGLT2i use are summarized in Table 1. All 1422 patients from the HEROES database were included in the analysis. Most of them were hospitalized (80.4%), and there was a predominance of men (71.4%) in the study cohort. Median (IQR) age of the patients was 69 (60–76) years. The largest group included individuals with HFrEF (49.9%; n = 709), followed by HF with preserved EF (HFpEF; 22.5%; n = 320) and HFmrEF (13.4%; n = 190). In 203 cases (14.3%), EF was not reported.

Overall, just over 60% of the patients received flozins, with the highest usage reported in the HFrEF and the lowest in the HFpEF group. The rate of SGLT2i use increased across the study quartiles: overall, from 60% to 67.4% (P = 0.002); HFrEF, from 80% to 88.2% (P = 0.17); HFmrEF, from 50.5% to 59% (P = 0.02); HFpEF, from 33.1% to 42.9% (P = 0.08).

In the final quartile (Q4; April 2023–January 2024), fewer than half of the HFpEF patients (42.9%) were treated with flozins, and less than 70% of all HF patients received such treatment. The most commonly reported reason for not initiating SGLT2i therapy (aside from the generic response “other”) was the patient’s financial situation. The response “patient cannot afford” was cited in over a quarter of cases, followed by chronic kidney disease (CKD; 20.5%) and urinary tract infections (UTIs; 7.6%).

Discussion

Flozins are a foundational and effective treatment for HF, with a well‑documented safety profile supported by major trials on both empagliflozin and dapagliflozin: DAPA‑HF (Dapagliflozin in Patients with Heart Failure and Reduced Ejection Fraction),⁴ EMPEROR‑Reduced (Cardiovascular and Renal Outcomes with Empagliflozin in Heart Failure),⁵ DELIVER (Dapagliflozin in Heart Failure with Mildly Reduced or Preserved Ejection Fraction),⁶ and EMPEROR‑Preserved (Empagliflozin in Heart Failure with Preserved Ejection Fraction).⁷ Despite an increasing frequency in flozin use during the HEROES study period, uptake remained suboptimal—particularly in the HFpEF group, where fewer than half of the patients received the treatment.

It is notable that the 2021 ESC guidelines¹ recommended flozins primarily for HFrEF, while the 2023 update² extended class I recommendations to all HF phenotypes. However, evidence supporting their efficacy in other HF subtypes was available as early as October 2021 (eg, EMPEROR‑Preserved⁷).

Despite evidence supporting the efficacy of SGLT2is in CKD, it remains one of the most commonly cited barriers to their use in HF patients. UTIs and genital infections (GIs) are rare complications associated with SGLT2i treatment—occurring less frequently with dapagliflozin than with empagliflozin according to the referenced studies (UTI, 0.5%⁴; UTI, 4.9% and GI, 1.7%⁵; UTI, 0.4%⁶; UTI, 9.9% and GI, 2.2%⁷). Nevertheless, in the HEROES study, UTIs were the third most commonly indicated reason for not initiating SGLT2i therapy. UTIs were reported as a reason for treatment inertia in 1.8% of all HEROES patients (n = 26/1422), while the incidence of UTIs reported as adverse events in the pooled data from the aforementioned trials with SGLT2i across all HF phenotypes was 3.96% (n = 410/10 353).^4-7

The main limitation of our study is its observational nature. Due to a lack of study‑specific procedures (ie, current echocardiographic assessment), 14% of the patients with a history of HF included in the HEROES database, who were being treated accordingly, had no reported EF, which may have biased the results. In the presented analysis, these individuals were included only in the “overall” calculations. For specific HF phenotype calculations, only the patients with a confirmed EF (reduced / mildly reduced / preserved) were included in the subgroup analyses. When reporting reasons for inertia of SGLT2i use, the HEROES case report form included the most common causes. However, when investigators selected the “other” option, there was no possibility to specify the reason. This may have biased the results regarding SGLT2i use inertia in our analysis.