ABC | Volume 113, Nº2, August 2019

Original Article Coll et al Non-invasive cardiac output measurement Arq Bras Cardiol. 2019; 113(2):231-239 Table 4 – Relative (ICC) and absolute reliability (TE) of the resting, submaximal, and maximal cardiorespiratory and hemodynamic characteristics Variable Relative Reliability Absolute Reliability (SI) ICC ± 90% CI TE ± 90% CI ES ± 90% CI Rest CPO (W) 0.80 ± 0.16 (high) 0.1 ± 0.0 1.1 ± 0.3 (moderate) CO (l/min) 0.83 ± 0.14 (high) 0.33 ± 0.11 1.0 ± 0.3 (moderate) SV (ml) 0.92 ± 0.07 (very high) 5 ± 1 0.6 ± 0.2 (moderate) HR (1/min) 0.91 ± 0.08 (very high) 4 ± 1 0.7 ± 0.2 (moderate) MAP (mmHg) 0.91 ± 0.08 (very high) 6 ± 2 0.7 ± 0.2 (moderate) Submaximal CPO (W) 0.85 ± 0.13 (high) 0.5 ± 0.1 0.9 ± 0.3 (moderate) CO (l/min) 0.59 ± 0.28 (moderate) 1.89 ± 0.60 1.8 ± 0.6 (large) SV (ml) 0.75 ± 0.19 (high) 10 ± 3 1.2 ± 0.4(large) HR (1/min) 0.97 ± 0.03 (very high) 4 ± 1 0.4 ± 0.1 (small) MAP (mmHg) 0.98 ± 0.02 (very high) 2 ± 1 0.3 ± 0.1(small) Maximal CPO (W) 0.82 ± 0.15 (high) 0.5 ± 0.2 1.0 ± 0.3 (moderate) CO (l/min) 0.73 ± 0.20 (moderate) 1.78 ± 0.57 1.3 ± 0.4 (large) SV (ml) 0.75 ± 0.19 (high) 12 ± 4 1.2 ± 0.4 (large) HR (1/min) 0.91 ± 0.08 (very high) 6 ± 2 0.7 ± 0.2 (moderate) MAP (mmHg) 0.82 ± 0.15 (high) 6 ± 2 1.0 ± 0.3 (moderate) P (W) 0.97 ± 0.03 (very high) 11.2 ± 3.6 0.4 ± 0.1 (small) VO 2 (l/min) 0.97 ± 0.03 (very high) 0.13 ± 0.04 0.4 ± 0.1 (small) ICC: intraclass correlation coefficient; CI: confidence interval; TE: typical error; ES: effect size; CV: coefficient of variation; CPO: cardiac power output; CO: cardiac output; SV: stroke volume; HR: heart rate; MAP: mean arterial pressure; P: workload; VO 2: oxygen uptake. small to large test-retest-reliability, whereas the CPO peak demonstrated a superior reliability than its underlying parameters, and (3) CPO was independent of measures of cardiac structure and function as well as traditional cardiopulmonary exercise parameters. Our first finding was that there was no systematic bias during all examination conditions. These outcomes are in line with further studies, investigating hemodynamic and cardiopulmonary exercise parameters. 26,49 Overall, in our study, systematic bias due to learning, subject motivation, and fatiguing effects as well as errors in calibration procedures can be excluded. 45,50 This assumption supports our research design. The second major finding was that all noninvasively measured hemodynamic parameters showed an acceptable test-retest reliability during rest, submaximal, and maximal exertion. Jones et al. 26 first showed a good test-retest reliability of TB in a healthy population at rest as well as during submaximal and maximal exertion. However, acceptable test-retest reliability was impacted by the fact that we have previously excluded a significant number of outliers (n = 8) due to measurement errors. It is further noteworthy that the reliability of our TB measurements was to some degree inferior compared to a previous study in which the reliability of a comparable technology (beat by beat signal morphology impedance cardiography) to evaluate the hemodynamic response was assessed. 20 One possible explanation for the differences may be that we investigated the reliability under less standardized conditions, another one could be related to significant technological differences. Overall, when outliers are excluded, TB can be considered as an appropriate technology to not only assess hemodynamic status in a research setting but also in everyday practice. The central task of the heart is to produce a sufficient CO and maintain an adequateMAP. Therefore, cardiac performance can be best explained by CPO, because it takes both the flow- and pressure-generating capacities of the heart into account. 29 In chronic heart failure, the application of hemodynamic measuring to standard cardiopulmonary exercise testing may help to explain the underlying mechanism of exercise intolerance with impact on clinical decision making, 31 therapy planning, and performance 32 as well as risk stratification. 51 Chomsky et al. 31 showed that CO respond to exercise is a strong predictor of mortality in cardiac transplantation candidates. In addition, Lang et al. 35 demonstrated CPO as the most powerful and independent predictor of survival chronic heart failure outcome in patients with chronic heart failure and that may enhance the prognostic power of traditional cardiopulmonary exercise testing. 235