Introduction:
Heart failure with preserved (HFpEF) and mildly reduced ejection fraction (HFmrEF) account for more than half of heart failure decompensations. However, randomized controlled trials (RCTs) conducted in these populations partly reported low numbers of included patients per participating center suggesting that trial populations may only reflect subgroups of real-world HFpEF/HFmrEF populations. We therefore aimed to assess the applicability of eligibility criteria of CHARM-preserved, I-Preserve, TOPCAT, Paragon-HF and EMPEROR-Preserved in a cohort of patients with decompensated heart failure presenting in an emergency unit.

Methods:
We performed a systematic retrospective medical records review of all patients presenting in the internal emergency unit of a university tertiary care center between August 2018 and July 2019 (n=32.028). According to predefined criteria, we identified 1.248 patients with decompensated heart failure. Among 866 patients with decompensated heart failure and documented echocardiography, 575 patients (66%) with a left ventricular EF above 40% were identified. Available medical records of the 575 subjects were systematically investigated for compatibility with the major in- and exclusion criteria of the CHARM-Preserved, I-Preserve, TOPCAT, Paragon-HF and EMPEROR-Preserved trial, respectively. A patient was considered potentially eligible for trial participation if all inclusion criteria were fulfilled and no exclusion criterion was fulfilled.

Results:
The investigated cohort (n=575) included 308 females (53.6%) with a mean age of 80.0 ± 9.5 years. Median (interquartile range) NT-proBNP was 3345 (1640 – 7836) pg/ml and mean eGFR was 48.2 ± 21.5 ml/min/m². HFpEF (EF ≥ 50%) was present in 375 patients (65.2%).

After applying eligibility criteria, 310 patients (53.9%) would have been currently eligible for CHARM-preserved on day of presentation in the Emergency Unit, 74 patients (12.9%) for I-PRESERVE, 31 patients (5.4%) for TOPCAT, 48 patients (8.3%) for PARAGON-HF and 82 patients (14.3%) for EMPEROR-Preserved, respectively. Most common reasons for exclusion were residency in a nursing home, dementia and uncontrolled blood pressure. When classifying modifiable exclusion criteria (such as elevated blood pressure) as potentially reversible, an additional number of 35 patients (6.1%) were potentially eligible for re-screening for CHARM-preserved, 120 patients (20.8%) for I-PRESERVE, 138 patients (24.0%) for TOPCAT, 89 patients (15.5%) for PARAGON-HF and 139 patients (24.1%) for EMPEROR-preserved.

Conclusion:
Our study indicates that particularly the recent HFpEF/HFmrEF outcome trials (TOPCAT, PARAGON-HF, EMPEROR-Preserved) systematically excluded a large proportion of the HFpEF/HFmrEF population. Considering the limitations of our retrospective study design the numbers of patients fulfilling respective exclusion criteria are probably underestimated. Our data suggests that a generalisation of results of TOPCAT, PARAGON-HF and EMPEROR-Preserved should be performed with caution.