Objective: Uric acid is known to act as not only a pathogenic inflammatory mediator, but also an antioxidative agent. Several studies showedlevel of uric acid has correlations with incidence, severity, or prognosis of pulmonary diseases. However, it is not well studiedabout the association of uric acid and acute respiratory distress syndrome (ARDS). Methods: Retrospective cohort study with propensity score matching was conducted at a medical intensive care unit of tertiary teaching hospital in South Korea. Medical records of patients diagnosed with ARDS were reviewed who admitted from January 2005 through December 2011. Results: We found total 237 ARDS patients who met the Berlin definition. Patients with serum uric acid level lower than 3.0mg/dL were defined as low uric acid group, and the others were defined as normal to high uric acid group. We selected 40 patients in the each group by propensity score matching. Low uric acid group showed more improvement in ARDS clinically. Otherwise more patients were expired by sepsis in normal to high uric acid group. Adding score of uric acid improved the established scoring system for critically ill patients such as APACHE II and SOFA. In the multivariate analysis, low level of serum uric acid was significantly associated with low in-hospital mortality. Conclusions: In patients with ARDS, low serum level of uric acid can be a good prognostic marker of in-hospital mortality.