Reduced NT-proBNP threshold for risk prediction in high-risk elderly with subclinical heart failure: support from cardiopulmonary exercise testing

E Potter,M Woessner,T.H Marwick
DOI: https://doi.org/10.1093/ehjci/ehaa946.1139
IF: 39.3
2020-11-01
European Heart Journal
Abstract:Abstract Background Subclinical heart failure (HF) is divided into Stage A and B (SAHF and SBHF) depending on structural or functional cardiac abnormalities. However, risk of HF progression is not solely dependent on echocardiographic abnormalities. Data on oxygen consumption (VO2) and other prognostic cardiopulmonary exercise test (CPET) markers in subclinical HF, particularly in the elderly, is lacking. VO2 may refine risk prediction and provide pathological insight in subclinical HF. Methods Asymptomatic individuals were recruited through primary care if they were ≥65 years with ≥1 non-ischaemic risk factor for HF. Clinical evaluation determined risk profile, biometrics and NT-proBNP. Treadmill CPET was undertaken with a modified Bruce protocol. Low V02 was defined as ≤20th percentile of age and gender specific VO2 in healthy individuals. SBHF defined as systolic (global longitudinal strain, GLS ≤16%), diastolic (E/e' ≥15, E/e' >10 with left atrial enlargement or impaired relaxation with other changes or left ventricular (LV) hypertrophy (LV mass index >95 g/m2 in women or >115 g/m2 in men). Results Of the 91 included individuals (age 71±4 years, 53% female), 46 (51%) had SBHF, average NT-proBNP was 60pg/ml (26–99mg/ml) and did not differ by HF stage (59 [26–85] pg/ml vs. 60 [30–99] pg/ml for Stage A vs. B respectively, p=0.94). Average peak VO2 was 19.8 (16–22.6) ml/kg/min and was low in 71 (78%). VO2 did not differ by HF stage (19.9 [17.7–22.4] ml/kg/min vs. 19.7 [16–22.8] ml/kg/min for SAHF vs. SBHF respectively, p=0.62). NT-proBNP was significantly higher in those with abnormal VO2 (66 [34–110] pg/ml vs 31 [20–70] pg/ml, p=0.016). Within each HF stage, low VO2 was associated with higher NT-proBNP (Figure). Of those with NT-proBNP >100pg/ml, 95% (22/23) had low VO2 compared with 72% of those ≤100pg/ml, p=0.02 (non-signifcant for 125pg/ml cut-off). No associations were found between SBHF or individual echo abnormalities and VO2. In logistic regression analysis NT-proBNP was a significant univariable predictor of low VO2 and remained significant after adjustment for other significant univariables (BMI) (OR 1.02 [95% CI 1.0001–1.03], p=0.048). Conclusion Low VO2 identifies a high-risk cardiac phenotype within subclinical HF stages, but is not necessarily associated with LV dysfunction. Levels of NT-proBNP beneath the cut-off used for HF exclusion may assist risk stratification in this population. NT-proBNP by VO2 in subclinical HF Funding Acknowledgement Type of funding source: Public Institution(s). Main funding source(s): Dr E Potter is supported by a PhD Scholarship from Monash University, Melbourne, Australia
cardiac & cardiovascular systems
What problem does this paper attempt to address?