Determination of cardiac output from pulse pressure contour during intra-aortic balloon pumping in patients with low ejection fraction

Fig. 1. Aortic flow simulated from aortic pressure with WKflow during a normal (A) and an augmented (B) heartbeat. In panel A1 a recording of a single beat of aortic pressure measured with a high fidelity fiber optic pressure transducer on the IAB catheter, black line. In panel A2 the resulting flow is plotted. With this simulated flow an aortic pressure beat is simulated, backwards, see A1 dotted red line which falls together with the measured pressure line. In an optimization, peripheral resistance (Rp) of WKflow is adjusted to create an optimal fit between measure and simulated pressure at end of diastole. The vertical narrow line indicates the moment of aortic valve closure which was derived from the first local minimum in aortic flow. In panels B1 and B2 the same simulation is shown during IABP. Here the red dotted line represents the simulated aortic pressure curve without augmentation. From this curve systolic, diastolic and mean arterial blood pressure can be calculated

Fig. 2. Example of the effect of a vena cava occlusion during IABP in a 1:2 mode. In upper panel the aortic pressure (Pa) is plotted. The narrow vertical upwards plotted lines indicated start of ejection (begin systole) and the downwards lines the moment of aortic valve closure (end systole). In the middle panel the left ventricular volume (Vlv) signal is plotted. The volume variation during systole is equal to stoke volume. Begin and end systole are indicated. In the lower panel beat-to-beat stroke volume (SV) derived from Vlv (open circles) and from WKflow (plus sign) are plotted at begin systole of each beat

Fig. 3. Regression between left ventricular stroke volume (SV) measured by the conductance method (SVlv) and WKflow method (SVwf) during vena cava occlusion during during IABP in 1:2 mode (upper panel) and with IABP switched off (lower panel). Each line is the result of a linear fit through the beat-to-beat SVlv and SVwf data points obtained during each single vena cava occlusion procedure

Fig. 4. Bland-Altman plot of stroke volume (SV) by conductance method (SVlv) and once-calibrated WKflow (SVwf). Data is obtained during vena cava occlusion procedures to diminish SV during IABP in 1:2 mode as well as with IABP switched off. Limits of agreement in SV in ml (dashed line) and percentage (dotted line) are indicated. SV ranged from 10 to 75 ml

Fig. 5. Change in stroke volume (SV) by conductance method (∆SVlv) and WKflow method (∆SVwf) obtained during vena cava occlusion. For explanation see text

Fig. 6. Example of irregular heartbeats. In upper panel the aortic pressure (Pa) is plotted. The narrow vertical upwards plotted lines indicate the start of ejection (begin systole) and the downwards lines the moment of aortic valve closure (end systole). In the lower panel beat-to-beat stroke volume (SV) derived from conductance method (open circles) and from WKflow (plus sign) are plotted at begin systole of each beat. A remarkable concordance between SVlv and SVwf is observable with beat-to-beat variations up to 65 ml and heart rate variating between 52 and 108 bpm

Fig. 7. Bland-Altman analysis for agreement between the two methods after one calibration by thermodilution per patient showed a bias or accuracy of 0.828 ± 3.287 ml, precision 7.6% and LOA from − 5.75 to 7.40 ml. SV ranges from 10 up to 95 ml