Estimated Population Prevalence of Heart Failure with Reduced Ejection Fraction in Spain, According to DAPA-HF Study Criteria

Heart failure (HF) is one of the main causes of morbidity, mortality, and high healthcare costs. Dapagliflozin, a sodium-glucose cotransporter-2 (SGLT2) inhibitor, reduced cardiovascular mortality and hospitalization for HF compared to placebo in patients with chronic HF, and reduced ejection fraction (EF) in the Dapagliflozin and Prevention of Adverse Outcomes in Heart Failure (DAPA-HF) study. Our aim was to estimate the number of patients with DAPA-HF characteristics in Spain. Our literature review identified epidemiological studies whose objective was to quantify the prevalence of HF and its comorbidities in Spain. We estimated the prevalence of HF with reduced EF, of New York Heart Association (NYHA) functional class II–IV, and with a glomerular filtration rate (GFR) ≥ 30 mL/min/1.73 m². In this population, we analysed the prevalence of diabetes using data from the REDINSCOR (Spanish Network for Heart Failure) registry. Our estimations indicate there are 594,684 patients ≥45 years old with HF in Spain (2.6% of this population age group), of which 52.4%, 84.0%, and 93.9% have reduced EF, are NYHA II–IV, and have a GFR ≥ 30 mL/min/1.73 m², respectively. By our calculations, approximately 245,789 Spanish patients would meet the DAPA-HF patient profile, and therefore could benefit from the protective cardiovascular effects of dapagliflozin.


Introduction
Heart failure (HF) is a clinical syndrome caused by a structural or functional heart abnormality leading to a reduction of cardiac output or an increase in intracardiac pressure. Characterization of

Study Design and Population
To determine the number of patients with HF with reduced EF, in NYHA functional class II-IV, and with normal or moderate kidney function (≥ 30 mL/min/1.73 m 2 ), we carried out a systematic literature search in PubMed of English or Spanish cross-sectional, cohort, population-based, or other epidemiological studies. To this end, we selected articles that contained the terms "heart failure" in the title; "prevalence" or "burden" in the title or abstract; "ejection fraction", "systole", or "systolic" in the title or abstract; and "general population" or "healthy" in the title or abstract (search date: September 1, 2017). This systematic search was updated on April 15, 2020, and supplemented with articles provided by experts [12,13,[20][21][22][23][24].
Patients' eligibility criteria for the population candidates for dapagliflozin treatment are summarized in Table 1. Full details are provided in the design paper [25]. Table 1. Summary of inclusion and exclusion criteria of the DAPA-HF study [25].

Inclusion Criteria
(1) Provision of signed informed consent prior to any study specific procedures (2) Men or women, aged ≥18 years at the time of consent (3) Diagnosis of HF with left ventricular EF ≤ 40%, which has been present for at least 12 months prior to enrolment (4) Diagnosis of symptomatic HF (NYHA functional class II-IV), within the previous 2 months (5) Optimally treated with pharmacological and/or device therapy for HF (6) NT-proBNP ≥ 600 pg/mL (or if hospitalised for HF within the previous 12 months, NT-proBNP ≥ 400 pg/mL) at enrolment. Patients with atrial fibrillation or atrial flutter must have a level ≥900 pg/mL, irrespective of history of HF hospitalization

Exclusion Criteria
(1) Treatment with SGLT2 inhibitors within 8 weeks prior to enrolment, or previous intolerance of an SGLT2 inhibitor (2) Diagnosis of type 1 diabetes mellitus (3) Symptomatic hypotension or systolic blood pressure <95 mmHg (4) Recent worsening HF or other cardiovascular events or procedures (5) Severe (GFR < 30 mL/min/1.73m 2 by CKD-EPI equation), unstable, or rapidly progressing renal disease at the time of randomization (6) Other conditions likely to prevent patient participation in the trial or greatly limit life expectancy DAPA-HF: Dapagliflozin and Prevention of Adverse Outcomes in Heart Failure; HF: heart failure; EF: ejection fraction; NYHA: New York Heart Association; NT-proBNP: N-terminal proB-type natriuretic peptide; GFR: glomerular filtration rate; CKD-EPI: Chronic Kidney Disease-Epidemiology Collaboration; SGLT2: sodium-glucose cotransporter-2.

Estimation of Population Prevalence and Number of Patients with Heart Failure in Spain
Our analysis applied the prevalence of HF reported for individuals older than 44 years in the largest and most recent representative population study [14] to the entire Spanish population and to each autonomous community's population, according to the 2019 Spanish National Statistics Institute data [26].

Estimation of Ejection Fraction ≤ 40% and New York Heart Association Functional Class II-IV Prevalence in the Heart Failure Population in Spain
The proportion of HF patients with EF ≤ 40% and NYHA functional class II-IV was determined according to the average percentages observed in the most recent Spanish registries of patients with chronic HF [6,7,[27][28][29]. The study designs, in reverse chronological order, are summarized as follows: Prospective cohort studies: The proportion of patients with GFR ≥ 30 mL/min/1.73 m 2 was estimated from the most generalizable data, obtained from patients with chronic HF in 28 Spanish hospitals included in the European Society of Cardiology (ESC) Heart Failure Long-Term Registry [23]. The prevalence of GFR ≥ 30 mL/min/1.73 m 2 was applied to the estimated HF population with EF ≤ 40% and with NYHA functional class II-IV (obtained in Section 2.3).
A sensitivity analysis with the total REDINSCOR prevalence of EF, NYHA functional class II-IV, and GFR ≥ 30 mL/min/1.73 m 2 was done to determine the reliability of our estimates (Tables S1-S3).

Estimation of Type 2 Diabetes Mellitus Prevalence in the Heart Failure Population in Spain
Finally, the prevalence of patients with and without T2D was obtained from a specific analysis of a chronic HF cohort provided by REDINSCOR investigators. The cohort characteristics have been described previously [27,30]. We calculated the prevalence of patients with and without diabetes who met the criteria for HF with EF ≤ 40%, NYHA functional class II-IV, and GFR ≥ 30 mL/min/1.73 m 2 (Tables S1 and S2).

Statistical Analysis
The arithmetic means of reduced EF and NYHA functional class II-IV estimations were obtained. The 95% confidence interval (CI) for the described estimations were obtained by assuming a Poisson distribution of patient counts. Analysis was carried out using R software, version 4.0.0. Table 2 describes the prevalence of HF in Spain and in Europe, as reported in the largest, most recent, and most representative population-based candidate studies. The 2012 study by Farré et al. [14] was considered the most appropriate for estimating prevalence in Spain, since it included a representative population of more than 88,000 people and the results closely resembled those of similar studies in Europe. Applying the prevalence of specific ages used by Farré et al. [14] to current Spanish population in the same age groups yielded a prevalence of approximately 2.6% in the group older than 44 years, or 594,684 (95% CI: 593,175-596,196) patients with HF.

Estimation of Ejection Fraction ≤ 40% and New York Heart Association Functional Class II-IV Prevalence in the Heart Failure Population in Spain
Characteristics of patients with HF included in the selected Spanish studies are shown in Table 3. The average proportion of HF patients with EF ≤ 40% was 52.4%, of which 84.0% had NYHA functional class II-IV. Applying these percentages to the estimated number of patients with HF in Spain, 311,614 (95% CI: 310,522-312,709) patients would have EF ≤ 40%, and 261,756 (95% CI: 260,755-262,760) patients would also have NYHA functional class II-IV. The Spanish population and the number of patients with HF, by age group and by autonomous community, are shown in Table 4. In Table 5, these data are disaggregated by EF ≤ 40%, and by the combination of EF ≤ 40% and NYHA functional class II-IV. Table 4. Total population in Spain according to 2019 data from the National Statistics Institute [26], and the number of patients with heart failure in Spain and its autonomous communities, according to prevalence data published by Farré et al. [14].  Table 5. Total number of patients in Spain and its autonomous communities with heart failure and ejection fraction ≤ 40%, as well as the number of heart failure patients with average ejection fraction ≤ 40% and in New York Heart Association functional class II-IV (based on data summarized in Table 3). According to the prevalence of chronic HF patients with GFR ≥ 30 mL/min/1.73 m 2 described by Crespo-Lerio [23], we estimated the number of HF patients with the combination of EF ≤ 40%, NYHA functional class II-IV, and GFR ≥ 30 mL/min/1.73 m 2 ( Table 6). In Spain, an estimated 245,789 (95% CI: 244,819-246,762) patients would have these characteristics. Table 6. Total number of patients with heart failure ejection fraction ≤ 40%, New York Heart Association functional class II-IV, and glomerular filtration rate ≥ 30 mL/min/1.73 m 2 in Spain and its autonomous communities [23]. A sensitivity analysis with REDINSCOR estimates (Table S3), however, showed that the estimated number of patients who met the clinical characteristics of participants in the DAPA-HF trial were 353,658 (95% CI: 352,494-354,825).

Estimation of Type 2 Diabetes Mellitus Prevalence in the Heart Failure Population in Spain
The number of HF patients with EF ≤ 40%, NYHA functional class II-IV, and GFR ≥ 30 mL/min/1.73 m 2 , with and without T2D, are shown in Table 7 by age group and autonomous community. With these prevalence data, Spain would have 115,473 (95% CI: 114,809-116,140) patients with T2D and 130,316 (95% CI: 129,610-131,025) without T2D who would meet the diagnostic criteria of HF with EF ≤ 40%, NYHA functional class II-IV, and GFR ≥ 30 mL/min/1.73 m 2 . Figure 1 summarizes the sequence of patient selection by the prevalence of patient characteristics or comorbidities analysed, as well as the estimated number of patients with and without each characteristic in Spain. Table 7. Total number of patients with heart failure, ejection fraction ≤ 40%, New York Heart Association functional class II-IV, and glomerular filtration rate ≥ 30 mL/min/1.73 m 2 , with or without type 2 diabetes mellitus, in Spain and its autonomous communities, according to prevalence data from the REDINSCOR registry (Tables S1 and S2).

Main Findings
Among all Spanish patients with HF, the estimated prevalence of patients older than 44 years with reduced EF, NYHA functional class II-IV, and with normal or moderate kidney function (≥ 30 mL/min/1.73 m²) was about 41.3%. Since the population was selected to meet the clinical characteristics of participants in the DAPA-HF trial, it is conceivable that these patients might also benefit from the positive cardiovascular effects attributed to dapagliflozin, in addition to its glucoselowering benefits. This new therapeutic indication would benefit the defined HF patient population, with and without T2D, as shown in the DAPA-HF clinical trial [22]. Regardless of the presence of T2D and the risk of worsening HF, death from cardiovascular causes and hospitalizations for HF

Main Findings
Among all Spanish patients with HF, the estimated prevalence of patients older than 44 years with reduced EF, NYHA functional class II-IV, and with normal or moderate kidney function (≥30 mL/min/1.73 m 2 ) was about 41.3%. Since the population was selected to meet the clinical characteristics of participants in the DAPA-HF trial, it is conceivable that these patients might also benefit from the positive cardiovascular effects attributed to dapagliflozin, in addition to its glucose-lowering benefits. This new therapeutic indication would benefit the defined HF patient population, with and without T2D, as shown in the DAPA-HF clinical trial [22]. Regardless of the presence of T2D and the risk of worsening HF, death from cardiovascular causes and hospitalizations for HF were significantly less frequent among DAPA-HF participants who received dapagliflozin, compared to those who received a placebo. In the DAPA-HF study, dapagliflozin represents the first in a new class of drug for HF with reduced EF. The DAPA-HF study results introduce the opportunity to further study the potential cardiovascular benefits of SGLT2 inhibitors. Prior studies with empagliflozin [34] and canagliflozin [35] showed a reduction in the relative risk of HF hospitalization in T2D patients, suggesting that the observed benefit is not restricted to a particular drug, but is rather a class effect. The Canagliflozin Cardiovascular Assessment Study (CANVAS) showed that canagliflozin reduced the overall risk of HF events in patients with T2D and high cardiovascular risk. No clear difference in effects on HF with reduced versus preserved events was noted [36]. In addition, the Dapagliflozin Effect on Cardiovascular Events-Thrombolysis in Myocardial Infarction 58 (DECLARE-TIMI 58) study with dapagliflozin included a reanalysis of retrospectively-obtained EF. The clinical benefit of dapagliflozin was found to be strong in reduced EF in the subset of patient with available EF. In patients with HF without reduced EF, there was only a reduction of hospitalizations, but not in total or cardiovascular mortality [37]. The former was confirmed in the DAPA-HF study. The dapagliflozin effect on mortality in HF-preserved EF patients remains to be conclusively answered

Reliability of Prevalence Estimates
Based on updated demographic information and a comprehensive literature review to obtain reliable prevalence data, we selected the most recent population-based cross-sectional study, in which Farré et al. analysed data from 88,000 individuals representative of the population of Catalonia, with an estimated a 2.7% HF prevalence in people older than 44 years [14]. Due to slight regional differences in age group population distribution, our estimations yielded 2.6% prevalence when applied to all of Spain. The estimations obtained by Farré et al. are also close to European and North American figures, and are more recent than the larger prevalence estimates reported in the meta-analysis by Hernáez et al. [24] and the 2008 Heart Failure Prevalence Study in Spain (PRICE) by Anguita et al. [15].
Given the diversity of hospital departments and primary care settings participating in the published studies, our estimated prevalence of HF patients with reduced EF and NYHA functional class II-IV (52.4% and 84.0%, respectively) reflects the mean values of the main published Spanish registries [6,7,[27][28][29]. The obtained prevalence values are in good agreement with the most recent European publications [5,38] in the Southern Europe population. Spanish cardiology departments were well-represented in these European studies, carried out in the context of the ESC Heart Failure Long-Term Registry, which had indicated that the prevalence of reduced EF in patients with chronic HF predominantly admitted to cardiology departments would be around 60%. This is slightly higher than our study's estimate, which was based on HF data from cardiology units, internal medicine, and primary care settings. The Linx Registry, one of the most recent HF studies in Spain, also had a large sample of patients and was carried out in cardiology departments in Catalonia. In that registry, de Frutos et al. [39] estimated that the prevalence of NYHA functional class II-IV in patients with HF and EF ≤ 40% was 85.5%, remarkably close to the estimate obtained in the present study.
We used the 6.1% prevalence of GFR < 30 mL/min/1.73 m 2 among chronic HF patients, reported by Crespo-Leiro et al. [23], based on data from 28 Spanish hospitals included in the ESC Heart Failure Long-Term Registry. This is the most recent publication with HF data from Spain, includes the largest series of Spanish patients, and is the only study to provide GFR data generalizable to the HF population. The 6.1% prevalence is among the lowest published in recent decades in Spain, although it is close to the REDINSCOR registry prevalence of 5.5% in HF units within cardiology departments (Tables S1 and S2).

Potential Translational Perspective
The results of the DAPA-HF study demonstrated that the primary composite outcome occurred in 386 of 2373 patients (16.3%) in the dapagliflozin group and in 502 of 2371 patients (21.2%) in the placebo group (hazard ratio = 0.74; 95% CI: 0.65-0.85; p < 0.001). The largest number of events of worsening HF was hospitalizations. Of the patients receiving dapagliflozin, 231 (9.7%) were hospitalized for HF, compared with 318 patients (13.4%) receiving the placebo (hazard ratio = 0.70; 95% CI: 0.59-0.83). Death from cardiovascular causes occurred in 227 patients (9.6%) who received dapagliflozin, and in 273 (11.5%) who received the placebo (hazard ratio = 0.82; 95% CI: 0.69 to 0.98) [22]. Under the placebo group incidence assumptions, among all 245,789 patients of the estimated Spanish target population, the primary outcome would occur in 52,107 patients, 32,936 patients would be hospitalized for HF, and 28,266 patients would die from cardiovascular causes. With dapagliflozin therapy, the expected annual reduction would consist of 5996 hospitalized patients for HF and 3079 deaths from cardiovascular causes.

Strengths and Limitations
The main strength of this study is that it combines the estimates from recently published HF prevalence data, enriched by a specific analysis of the REDINSCOR registry database to estimate HF with reduced EF, NYHA functional class II-IV, and GFR ≥ 30 mL/min/1.73 m 2 , with and without T2D prevalence in the Spanish population. However, we decided not to use the estimates based on REDINSCOR database, because the REDINSCOR registry could be biased toward the profile of patients admitted to HF units in cardiology departments, which could depart from the general HF patient population characteristics. All patients included in the REDINSCOR registry had a NYHA class >I, and the mean estimate of patients with reduced EF was 73%, considerably different from the average of the main Spanish registries that we summarize in Table 3. A sensitivity analysis with REDINSCOR patient characteristics is presented in Table S3, showing rather higher figures than its corresponding Table 6 results.
The study also has several limitations. First, we did not take into account the N-terminal pro B-type natriuretic peptide (NT-proBNP) eligibility criteria prevalence in our estimates, due to the absence of prevalence information in the literature and to the probably small reduction in the number of eligibility patients. Second, among the Spanish studies summarized in Table 3, the definition of reduced EF varied from EF ≤ 40% to EF ≤ 50%. Furthermore, the GALICAP and EPISERVE studies did not differentiate between chronic and acute HF. Third, we had to assume that the prevalence of NYHA functional class II-IV was the same for reduced, mid-range, and preserved EF, owing to the absence of stratified information in the literature. Likewise, we had to assume that the prevalence of GFR ≥ 30 mL/min/1.73 m 2 was the same for each type of EF and NYHA functional class, and for all age groups considered in our study. Fourth, the prevalence of diabetic and non-diabetic patients in the REDINSCOR registry could be biased toward the profile of patients admitted to HF units in cardiology departments. Finally, our study does not report prevalence by sex, as the published studies did not provide this stratified information. We firmly support stratification by sex in all future studies, in order to identify the best treatment guidelines to apply in the whole population.

Conclusions
In this population analysis, we estimated that approximately 245,789 Spanish patients would meet the inclusion criteria of the DAPA-HF: EF ≤ 40%, NYHA functional class II-IV, and GFR ≥ 30 mL/min/1.73 m 2 , as well as 115,473 with T2D. The magnitude of this population highlights the need to introduce effective and safe new drugs to reduce morbidity and mortality in these patients.

Supplementary Materials:
The following are available online at http://www.mdpi.com/2077-0383/9/7/2089/s1, Table S1: Number of patients with the four selection criteria of the DAPA-HF study in the REDINSCOR cohort in Spain. Table S2: Average prevalence of the four main selection criteria of the DAPA-HF study in the REDINSCOR cohort in Spain. Table S3: Total number of patients with heart failure, ejection fraction ≤ 40%, New York Heart Association functional class II-IV, and glomerular filtration rate ≥ 30 mL/min/1.73 m 2 in Spain and its autonomous communities, according to prevalence data from the REDINSCOR registry (Tables S1 and S2).