Diastolic exercise stress testing in heart failure with preserved ejection fraction: The DEST-HF study.

AIMS

Pulmonary capillary wedge pressure (PAWP) ≥25 mmHg during bicycle ergometry is recommended to uncover occult heart failure with preserved ejection fraction. We hypothesized that PAWP increase would differ in available diastolic stress tests and that the margin of PAWP ≥25 mmHg would only be reliably achieved through ergometry.

METHODS AND RESULTS

We conducted a prospective, single-arm study in patients with an intermediate risk for heart failure with preserved ejection fraction according to the ESC HFA-PEFF score. A total of 19 patients underwent four stress test modalities in randomized order: leg raise, fluid challenge, handgrip, and bicycle ergometry. The primary outcome was the difference (Δ) between resting and exercise PAWP in each modality. Secondary outcomes were differences (Δ) in mean pulmonary artery pressure (mPAP), cardiac output (CO), as well as the ratios between mPAP and PAWP to CO. Compared to resting values, passive leg raise (Δ7.7 ± 8.0 mmHg, p = 0.030), fluid challenge (Δ9.2 ± 6.4 mmHg, p = 0.003), dynamic handgrip (Δ9.6 ± 7.5 mmHg, p = 0.002), and bicycle ergometry (Δ22.3 ± 5.0 mmHg, p < 0.001) uncovered increased PAWP during exercise. Amongst these, bicycle ergometry also demonstrated the highest ΔmPAP (27.2 ± 7.1 mmHg, p < 0.001), ΔCO (3.3 ± 2.6 L/min, p < 0.001), ΔmPAP/CO ratio (2.3 ± 2.0 mmHg/L/min, p < 0.001), and ΔPAWP/CO ratio (2.2 ± 1.4 mmHg/L/min, p < 0.001) compared to other modalities. PAWP ≥25 mmHg was only reliably achieved in bicycle ergometry (31.1 ± 3.9 mmHg). In all other modalities only 10.5% of patients achieved PAWP ≥25 mmHg (handgrip 18.4 ± 6.6 mmHg, fluid 18.1 ± 5.6 mmHg, leg raise 16.5 ± 7.0 mmHg).

CONCLUSIONS

We demonstrate that bicycle ergometry exhibits a distinct haemodynamic response with higher increase of PAWP compared to other modalities. This finding needs to be considered for valid detection of exercise PAWP ≥25 mmHg when non-bicycle tests remain inconclusive.