Su1016 Risk Factors for Post-Operative Respiratory Failure in Liver Transplant Recipients

Abstract

Rationale: Post-operative respiratory failure (PRF) is a common complication after liver transplant (LT). PRF leads to increased length of intensive care unit (ICU) and hospital stay and predisposes to increased morbidity in the post-operative period. Further investigation is needed to identify risk factors for PRF in LT recipients. Aim: To evaluate pre-operative cardiopulmonary testing, recipient, donor, and surgical variables and their utility in prediction of PRF after LT. Methods: Retrospective case-control study performed on LT recipients from 2010-2012. Post-operative course was reviewed to evaluate for cases of PRF, defined as failure to extubate before or unplanned re-intubation within 48 hours from completion of LT surgery. Pre transplant recipient demographic characteristics were recorded. Donor and surgical characteristics including: simultaneous kidney transplant, cytomegalovirus (CMV) status, gender, donor source (deceased or living), age, warm and cold ischemic times were assessed. Pre-transplant cardiopulmonary testing examined included: the presence or absence of heart failure, diastolic dysfunction, left ventricular ejection fraction, right ventricular systolic pressure, forced expiratory volume in one second to forced vital capacity ratio (FEV1/ FVC), forced vital capacity (FVC), total lung capacity (TLC), forced expiratory volume in one second ( FEV1), and diffusion limit of carbon monoxide (DLCO). Variables of interest were analyzed using univariate logistic regression with an end point of PRF. Factors deemed most significant (p < 0.25) in predicting PRF were included in a multivariate regression analysis to estimate adjusted odds ratios (OR) for each variable. Variables with p < 0.05 were considered significant predictors of PRF in the final model. Results: A total of 170 patients were identified. Of those, 23.7% experienced PRF. Significant predictors including pre-transplant MELD, pre-operative respiratory failure, FVC <80%, FEV1 <80%, TLC <80%, and deceased donor source from univariate logistic regression [Table 1] were entered into multivariate logistic regression. In the final model, pre-operative respiratory failure (OR= 161.48, p = 0.002) was found to be predictive of PRF, in a model that explained 26.3 to 41.7% of variance [Table 2].Conclusions: Pre-operative respiratory failure from any etiology was identified as a significant predictor of PRF after LT. Pre-LT echocardiogram and pulmonary function testing had no predictive value for PRF. These findings support the development of protocols to aggressively wean ventilatory support immediately prior to LT. Univariate Analysis: Post-Operative Respiratory Failure