Non-invasive hemodynamic monitoring techniques for detecting Preload responsiveness in critically ill patients

Volume expansion is the first-line treatment for acute circulatory failure in almost all cases. However, its inconsistent effectiveness and its side effects make it necessary to predict the effects before undertaking it. Several tests have been developed to detect this state of preload dependence, and we are interested in refining some of them and improve their use. The end expiratory occlusion (EEXPO) test consists in transiently stopping mechanical ventilation at end expiration to increase venous return and, thus, cardiac preload. In preload responder patients it is expected to increase cardiac output. We have gathered an amount of information from the existing literature, showing that its ability of detecting fluid responsiveness is very high, through a systematic review and meta-analysis. The effects of this test must be assessed on cardiac output, and several methods have already been described for this. Bioreactance is a completely non invasive technique to measure cardiac output, which has been demonstrated to be reliable in detecting preload responsiveness through a passive leg raising (PLR) maneuver. Nevertheless, its role in detecting a positive EEXPO test was never investigated. In the main study of this PhD project, we show that the current available commercial version of the bioreactance device is not suitable for this purpose, due to its very long averaging and refreshing times (24 and 4 seconds, respectively). However, when we used a research version of the device, which allowed us to reduce both intervals (8 and 1 second for averaging and refreshing times, respectively), bioreactance proved to be a reliable method to detect preload responsiveness through the EEXPO test. We have also described an original and reliable method for measuring the effects of both PLR and EEXPO tests, which consists of measuring the perfusion index (PI), the ratio between the pulsatile and the non-pulsatile portion of the pulse oxygen saturation signal. Although the signal could not be collected stably in all patients, changes in this index during passive leg raising were able to measure the effects of the PLR test and predict the response to volume expansion. Similarly, Pl measurements could identify a positive EEXPO test, even though their changes were of lower amplitude compared to those of a PLR test. Our results open up the possibility of measuring the effects of these tests of preload responsiveness using non-invasive tools. This could prompt their use outside the intensive care unit, such as in the emergency department, where tracking changes of cardiac output during the first hours of the management of acute circulatory failure could lead to a more rational use of fluid administration.