The shortage of donor organs highlights the need to better identify patients most likely to benefit from hepatic transplantation. Reduced aerobic capacity (decreased peak oxygen consumption [VO(2)] during symptom-limited cardiopulmonary exercise testing) is frequently present in cirrhosis. Peak VO(2) during cardiopulmonary exercise testing may predict short-term outcome after hepatic transplantation. Symptom-limited testing was performed on a cycle ergometer (continuous ramp protocol) and VO(2) determined using a metabolic cart. One hundred fifty-six patients were tested; 59 subsequently underwent hepatic transplantation. Results showed that survivors and nonsurvivors were similar in age, duration of liver disease, Child-Pugh score, MELD score, resting cardiovascular function, pulmonary function, and gas exchange. The 6 (10.2%) patients dying within 100 days of transplantation were more likely to have reduced aerobic capacity (peak VO(2) <60% predicted and VO(2) at anaerobic threshold [VO(2)-AT] <50% predicted peak VO(2)) compared to survivors (4/6 vs. 7/53, P <.01). Using a multiple logistic regression model controlling for duration and severity of liver disease and time to transplantation, reduced aerobic capacity was independently associated with 100-day mortality. In conclusion, reduced aerobic capacity during cardiopulmonary exercise testing is associated with decreased short-term survival after hepatic transplantation. Further study is needed to determine if cardiopulmonary exercise testing can be used to improve allocation of donor organs. To ensure optimum allocation of donor organs, it is important to identify patients most likely to benefit from transplantation. Investigators have identified a number of preoperative, intraoperative, and postoperative factors that predict increased risk for postoperative mortality. Unfortunately, predictive accuracy has not been high, and the timing of factor identification does not optimize organ utilization. Identification of predictors of survival at the time of listing for transplantation might lead to better resource allocation.