Rate and Predictors of Successful Cardiopulmonary Resuscitation in End-Stage Liver Disease

Background Cardiopulmonary resuscitation (CPR) after cardiac arrest in terminally ill patients is controversial. End-stage liver disease (ESLD) patients are unique from other terminally ill as they are generally younger and may be candidates for curative liver transplantation, but multiple studies have suggested poor outcomes when these patients require CPR. Predictors of success of CPR in ESLD have not been fully investigated, limiting end-of-life discussions. Aim The aim of this study was to quantify the rate and predictors of successful CPR in ESLD. Methods We performed a retrospective chart review of patients with ESLD who received CPR from 2/2002 to 12/2013 at a single institution. Pre-cardiac arrest variables were collected for analysis as predictors of survival. Our primary outcome was survival to hospital discharge. Results Of the 38 patients who underwent CPR, six survived to hospital discharge. When comparing those who survived to discharge with those who did not, we found no significant difference in age ( p = 0.34), gender ( p = 0.85), presence of ascites ( p = 0.67), location at time of arrest ( p = 0.39), concurrent GI bleeding ( p = 0.48), and multiple individual lab values. Significant predictors of not surviving to hospital discharge were a model for end-stage liver disease (MELD) ≥20 (OR 6.0, p = 0.044) and presentation with a non-shockable rhythm (PEA/asystole) (OR 29, p < 0.001). Conclusion ESLD patients requiring CPR have worse outcomes as their MELD score increases. CPR in ESLD when MELD is <20 or with a shockable rhythm has a greater likelihood of success.