(P029) ESTIMATED PULSE WAVE VELOCITY INDEPENDETLY PREDICTS SURVIVAL-TO-DISCHARGE IN PATIENTS REQUIRING EXTRACORPOREAL MEMBRANE OXYGENATION: A SINGLE-CENTRE RETROSPECTIVE COHORT STUDY

Background: Extracorporeal membrane oxygenation (ECMO) is a life-saving device used to support the respiratory and/or cardiovascular function of acutely ill patients. While this supportive device is widely used in intensive care units around the world, appropriate patient selection is difficult given the high associated morbidity and mortality of those requiring this level of intervention. One possible solution to the heterogeneity of ECMO patients is to determine a variable that reflects a chronic marker of health and has independent strength in predicting overall morbidity and mortality. A suitable candidate for this variable is Arterial Stiffness (AS), a chronic marker of vascular compliance, demonstrated to have strong correlation with cardiovascular disease, peripheral vascular disease, cerebrovascular disease, renal disease, and all-cause mortality. Additionally, AS has been shown to be strongly influenced by systemic inflammation, as seen in ECMO patients. In this study, we aim to understand the benefit of estimated pulse-wave velocity, a metric of AS, as an independent predictor of outcomes and survival-to-discharge in our cohort of ECMO patients.

METHODS AND RESULTS: A retrospective cohort study was performed at the London Health Science Centre (LHSC) in London, Ontario, Canada between 1996-2021, totaling 255 patients requiring ECMO. Estimated pulse wave velocity (ePWV) was calculated using an algorithm generated from the Reference Values for Arterial Stiffness Collaboration. Recorded outcomes included: in-hospital death, ischemic stroke, hemorrhagic stroke, renal failure and need for renal replacement therapy (RRT). For adjusted analysis, survival-to-discharge was used. Multivariate logistic regression and propensity-score matching were utilized to control for confounding. On univariate logistic regression, ePWV was found to have a significant protective effect for renal failure (OR 0.88 [0.78-0.99], p=0.034) and RRT (OR 0.87 [0.77-0.98], p=0.027). Higher ePWV was also found to be significantly predictive of ischemic stroke (OR 1.676 [1.31-2.37], p=0.0002) and in-hospital death (OR 1.20 [1.06-1.38], p=0.006), but insignificant for predicting hemorrhagic stroke (OR 1.07 [0.74-1.55], p=0.710). On multivariate analysis and propensity-score matching, 5 of 6 models demonstrated ePWV as a significant independent predictor of survival-to-discharge. (OR 0.70 [0.57-0.84], p=0.00021, OR 0.72 [0.60-0.86], p=0.0002, OR 0.87 [0.75-1.00], p=0.045, OR 0.85 [0.74-0.97], p=0.013)

Conclusion: This study presents ePWV as a promising marker for risk-stratification in ECMO patients. It furthers understanding of the role of arterial health in disease trajectory and strengthens the validity of AS as a marker of interest in medical and surgical management. Further research is needed to validate these findings and develop tangible tools for clinical application.