Exertional and non-exertional variables of cardiopulmonary exercise testing in heart failure with and without cardiac amyloidosis

Aims:

Cardiac Amyloidosis (CA) constitutes an important etiology of heart failure with preserved (HFpEF) or mildly reduced (HFmrEF) ejection fraction. Amyloid infiltration to the heart leads to a different exercise response compared to non-CA HFpEF/HFmrEF patients. Since patients with CA show early exhaustion, we aimed to investigate if non-exertional variables of cardiopulmonary exercise testing (CPET) provide additional information in comparison to traditional peak oxygen consumption (VO_2peak).

Methods:

We retrospectively compared CPET variables of patients with HFpEF and HFmrEF with (n=21) and without (n=21, HF) CA reporting to our outpatient clinic for follow-up visits. Patients had to be in NYHA classes II-IV and indications for CPET were based on clinical suspicion of deterioration of heart failure and exercise capacity. Exertional and non-exertional CPET variables as well as laboratory and echocardiographic markers were analyzed. The primary outcome was the difference in CPET variables between groups. The secondary outcome was rehospitalization in patients with CA due to decompensated heart failure during a follow-up of 24 months. Correlations between CPET, NTproBNP and echocardiographic variables were calculated to detect patterns of discrimination between the groups.

Results:

CA patients were inferior to HF controls in exertional CPET variables, such as VO_2peak (Q1-Q3: 13.0-16.0 ml/kg/min vs. 17.1-21.7ml/kg/min, p<.001), peak oxygen pulse (Q1-Q3, 11.0-16.0ml/beat/kg*100 vs. 13.6-16.7ml/beat/kg*100, p=0.046), circulatory (CP, Q1-Q3:1543.0-2298.0ml/kg/min x mmHg vs. 3227.9-4206.5ml/kg/min x mmHg, p<.001) and ventilatory power (VP, Q1-Q3: 3.5±1.3mmHg vs. 5.7±1.4mmHg, p<.001). Similarly, CA patients were inferior in non-exertional CPET variables, such as VE/VCO₂ (Q1-Q3: 33.0-46.0 vs. 28.8-35.8, p=0.004), and the oxygen equivalent at the first ventilatory threshold (EqO₂ at VT1, Q1-Q3: 25.0-33.0 vs. 21.0-25.4, p=0.004). Patients with CA were hospitalized more often than HF controls (n=11 vs. n=2, p=0.002). Rehospitalization was associated with VE/VCO₂ (p=0.019), peak oxygen pulse (p=0.042), EqO₂ at VT1 (p=0.003), CP (p=0.024), VP (p<.001), but not VO_2peak (p=0.127) in CA patients. Higher performance was correlated with lower E/e’ and NTproBNP as well as higher resting heart rate and stroke volume in CA, but to a lesser extent in HF patients.

Conclusions:

Patients with CA displayed worse performance in exertional and non-exertional CPET variables compared to non-CA HF patients. Rehospitalization in patients with CA was associated with several exertional and non-exertional CPET variables, which may bear prognostic utility. As correlations between resting echocardiographic markers and CPET variables differed between CA and HF patients, this may serve to facilitate earlier diagnosis of CA.