P4382O2-pulse measure obtained by gas exchange analysis is an accurate esteeme of stroke volume?

Background It is axiomatic that exercise performances depends on cardiac output increase. Among several methods for noninvasively estimating stroke volume, O2 pulse (i.e. the ratio of VO2/HR) is generally used during gas exchange analysis, but its accuracy has never been well investigated. Purpose To test the validity and the accuracy of O2 pulse as a measure of stroke volume (SV) in comparison with the echocardiographic methods. Methods From 22nd of August 2016 to 30th of April 2018 our laboratory performed 259 cardiopulmonary exercise tests (CPET) combined with echo. The group of patients was characterized by a heterogeneous spectrum of diseases: 70 coronary artery disease, 10 heart failure with preserved ejection fraction, 13 heart failure with middle ejection fraction, 26 heart failure with reduced ejection fraction, 29 Fabry disease, 7 hypertrophic cardiomyopathy, 10 dilated cardiomyopathy, 30 aortic stenosis, 16 aortic regurgitation, 8 mitral stenosis, 76 mitral regurgitation, 6 tricuspid regurgitation, 7 congenital heart diseases. The median age was 65 yr (52,5; 74,5) and 61,5% of patients was male. In our group are present 20 healthy subjects with normal heart. O2 pulse was measured at rest, at 2 minutes, 4 minutes, 6 minutes and at maximum of stress. In some cases the maximum of stress correspond to 4- or 6-minutes of stress. All the echocardiographic images were taken with Epiq7C, Philips. The left ventricle SV was measured as SV = LVOT area x LVOT VTI, all the measures and the images where acquired by the same operator to reduce at minimum the external error introduce by the operator. Results As first step we analysed the absolute value of O2-pulse and SV at rest and during the different steps of exercise to see if and how they behave. In the figure attached is possible to see that both the values increased during the exercise with a small reduction at the maximum that may be explained by a premature interruption of the exercise. Secondly we analysed the correlation between O2 pulse and LVOT-SV at rest and during 2-, 4-, 6- minutes of exercise and at maximum which was respectively: r=0,23 (p<0,0001), r=0,36 (p<0,0001), r=0,30 (p<0,0001), r=0,25 (p<0,0001) and r=0,51 (p<0,0001) (see figure 2). The analysis has been possible in 259 patients at rest, in 246 at 2-minutes of stress, in 221 at 4-minutes of stress and in 149 patients at 6-minutes of stress. At maximum of exercise, the analysis has been carried out in 229 patients, but we need to keep in mind that maximum of stress includes both 4- and 6-minutes of exercises according to the capability of each patients. Conclusion In our analysis we show an incremental trend during exercise of O2-Pulse value and of LVOT-SV. At rest and during exercise it is present a good correlation between O2-pulse and LVOT-SV. These data reinforce the concept that O2 pulse reflects an estimation of SV and may explain its reported prognostic validity in prospective studies.