Geographic variation in the performance of the HFpEF-ABA score: Asia versus outside Asia

Background

Identifying heart failure with preserved ejection fraction (HFpEF) can be challenging, and several probability-based scores have been developed to assist diagnosis. Among them, the HFpEF-ABA score is based solely on age, body mass index (BMI), and a history of atrial fibrillation (AF). However, the diagnostic and prognostic performance of the HFpEF-ABA score may vary across regions, given differences in clinical characteristics between Asian and non-Asian populations.

Purpose

To evaluate the diagnostic and prognostic performance of the HFpEF-ABA score in Asia compared with outside Asia.

Methods

We performed a pooled analysis of individual participant-level data from 6 trials of HFmrEF/HFpEF (CHARM-Preserved, DELIVER, I-Preserve, TOPCAT Americas, PARAGON-HF, and FINEARTS-HF). The baseline HFpEF-ABA score was calculated for each patient, and its diagnostic and prognostic performance was evaluated in Asia vs the rest of the world outside Asia. The composite of HF hospitalization or cardiovascular (CV) death was analyzed across baseline score categories (<75%, 75–90%, >90%) using Cox proportional hazards models.

Results

Among 25,905 participants with HFpEF, 3,704 (14%) were enrolled in Asia and 22,212 (86%) outside Asia. Mean age was similar (71±10 vs 71±9 years; P=0.55), whereas BMI was lower (26.9±5.4 vs 30.6±5.9 kg/m²; P<0.001) and AF was more prevalent (52% vs 46%; P<0.001) in Asia.

HFpEF-ABA scores were shifted leftwards in Asia compared with outside Asia (Figure 1A). Overall, 44% of patients in Asia and 38% outside Asia had a score <75% (low probability) and the difference for a very low score (≤60%) was even more striking: 30% (Asia) vs. 21% (outside Asia). The proportions for a score of 75-90% (intermediate probability) were: 23% (Asia) vs. 20% (outside Asia) and for a score >90% (high probability) were: 34% (Asia) vs. 42% (outside Asia).

Risk increased stepwise with increasing HFpEF-ABA score, in both regions: outside Asia, HRs (95% CI) for score 75–90% and >90% versus <75% were 1.42 (1.31–1.54) and 1.86 (1.73–1.99) and in Asia these HRs were 1.07 (0.89–1.29) and 1.43 (1.23–1.67), respectively. When regions were compared directly within each score category, enrollment in Asia was associated with higher risk than enrollemnt outside Asia: HR for a score <75%, 1.41 [1.23–1.61]; for a score 75–90%, 1.13 [0.96–1.35], and for a score >90%, 1.20 [1.07–1.35] (P interaction < 0.001) (Figure 1B).

Conclusions

The HFpEF-ABA score classified a substantial proportion of patients with proven HFpEF as having a low probability of this condition, particularly in Asia. However, in Asia, patients with a low score were at high risk and experienced worse clinical outcomes than patients with similar scores outside Asia. In Asia, the HFpEF-ABA score leads to under-recognition of individuals with high-risk HFpEF who are eligible for evidence-based treatment with SGLT2 inhibitors and finerenone.Figure 1