Author + information
- Ify R. Mordi, MD∗ ( and )
- Dana K. Dawson, DM, DPhil
- ↵∗Division of Molecular and Clinical Medicine, University of Dundee, Ninewells Hospital and Medical School, Dundee DD1 9S7, United Kingdom
We thank Dr. Pellicori and colleagues for their interest in our paper (1).
We acknowledge their points regarding the heterogeneity of patients included in trials of heart failure with preserved ejection fraction (HFpEF) and that this may be the reason why many of these trials have produced neutral results, underscoring the need for studies such as our own to further our understanding of the HFpEF phenotype.
Regarding levels of brain natriuretic peptide (BNP) within our study, we accept that some of the patients had BNP levels lower than the diagnostic threshold of 35 ng/l at the time of study inclusion (which we reported in our paper); however, as we state in the Methods section of our paper (1), patients were required to have BNP levels >35 ng/l at the time of HFpEF diagnosis. Natriuretic peptide levels do of course fluctuate in HFpEF, and once patients are given loop diuretic agents (or other medications) at the time of HFpEF diagnosis, BNP levels would likely be reduced, even if these medications were later discontinued (2). Indeed, extrapolating from heart failure with reduced ejection fraction, following treatment, BNP can even be reduced to normal levels despite persisting echocardiographic abnormalities (3). Of note, our study does, however, show that even in a stable outpatient setting, patients with HFpEF have higher BNP levels than those with hypertension.
Dr. Pellicori and colleagues also raise points that in fact underline the importance of our study: first, the fact that some patients with hypertension were prescribed loop diuretic agents without any other features diagnostic of HFpEF and also that some of the healthy patients had BNP levels >35 ng/l. Both of these underscore the need for more accurate phenotyping to diagnose HFpEF, for example speckle-tracking echocardiography and cardiac magnetic resonance. Even in a large community cohort of subjects free of cardiovascular disease, 12.8% of patients had natriuretic peptide levels above the 80th percentile (4). Although, as Dr. Pellicori and colleagues state, a high BNP level might qualify for a diagnosis of HFpEF if a patient had reported symptoms, both control subjects and patients with hypertension in our study were asymptomatic. We recognize that the diagnosis of HFpEF is a complex matter, which is why a multimodality approach will almost certainly provide more clarity rather than a single parameter. Speckle-tracking echocardiography and T1 mapping may potentially play an important role in providing this, especially in combination with other parameters such as natriuretic peptides.
Please note: Dr. Mordi is supported by a National Health Service Education for Scotland/Chief Scientist Office Post-Doctoral Clinical Lectureship (PCL/17/07). Dr. Dawson has reported that she has no relationships relevant to the contents of this paper to disclose.
- 2018 American College of Cardiology Foundation
- Mordi I.R.,
- Singh S.,
- Rudd A.,
- et al.
- Desai A.S.
- McKie P.M.,
- Cataliotti A.,
- Lahr B.D.,
- et al.