The present study demonstrated a relationship between %MAP changes with the %Eaeff, %EaNet, and %RaNet, but not with %EaDyn. We also found that the %Eaeff provided the highest predictability of %MAP in the multiple linear regression model compared with the others. Furthermore, the %Eaeff had the highest performance for discriminating participants who developed an increase or decrease in MAP after VE. As a result, our study suggests that the %Eaeff has a strong relationship with %MAP and could be the best surrogate arterial load parameter.
The Eaeff has been shown to strongly correlate with the arterial elastance derived from the pressure-volume curve of the heart [Ea(PV)], which is known as a gold standard. (7) Eaeff changes in response to arterial tone and vasoactive infusion. (5) Also, it reduces after fluid administration (4), which is consistent with the physiologic response of arterial tone to fluid therapy.
Interestingly, our results showed that the baseline EaDyn was a poor predictor of increasing or decreasing MAP after VE. In the previous study, the EaDyn has been defined as an arterial load parameter (4) because it is calculated from pressure and volume variation changed along with the cyclic alteration of lung volume. Therefore, the EaDyn depends not only on the arterial system but also on the lung volume change. Theoretically, the actual arterial elastance given arterial property is believed to be constant during the cardiac cycle. (10) Also, it should be assumed to be stable during the breathing cycle. Therefore, we are concerned about being an arterial load parameter of the EaDyn, which may be affected by the lung volume, and might not represent the actual arterial elastance.
Even though the EaDyn was proved, by Garcia et al., to be an excellent predictor of MAP responsiveness compared to the arterial elastance and SVR in patients with circulatory shock receiving the fluid challenge. (3, 4) However, the study by Khwannimit et al. (11) demonstrated no significantly different EaDyn between MAP responders and non-responders. In addition, the Eadyn differently responded to vasodilator and vasoconstrictor from the arterial elastance and resistance variables. (5) It was also found to have an inverse correlation with MAP in the study, as mentioned earlier.
Regarding the EaNet and RaNet, both variables share similar variables for calculation, namely the PP. However, the PP is not constant along with the mechanical ventilation. The cyclic mechanical ventilation affects the PP, making the EaNet and RaNet vary, despite unchanged arterial properties. This reason might contribute to the results of poorer predictors in both parameters. However, the study by Chemla et al. (6) validated net compliance (reciprocal of EaNet) to estimate total arterial compliance by area method, used as a gold standard. They showed a strong correlation (r = 0.98, P < 0.001) between both parameters. In this regard, Chemla and the team enrolled different populations from our study; therefore, the interpretation should not be made similarly.
Our study's limitation was that we did not measure the Ea(PV) at the bedside, which is considered the gold standard, as performed in the study by Monge García et al. (2). However, the Eaeff has been validated as a surrogate parameter of Ea(PV) (7) and is feasible at the bedside. A second limitation was that we did not use the thermodilution method for CO measurement. Pulse contour analysis by the Vigileo system is not the gold standard of CO measurement. Finally, we could not control some potential confounding factors, such as the artery's vasomotor tone in participants with sepsis during the VE period, which may have affected the arterial load. Future human studies are required to understand better the physiologic effect of medications on the change in the Eaeff.