OBJECTIVES: To examine racial and ethnic disparities in receiving total knee replacement surgery among eligible patients. METHODS: Using a nationally-representative inpatient care dataset, the HCUP National Inpatient Sample, from 1998 to 2007, we had an available sample of 809,750 Caucasians, 74,869 African- Americans, 48,010 Latinos, 8,976 Asians, and 2,554 Native Americans who were eligible for  total knee replacement surgery based on their ICD 9 codes. Multivariable logistic regressions were used to estimate the racial and ethnic disparities in the rates of actually receiving knee replacement surgery among these eligible subjects. Blinder-Oaxaca decomposition techniques were employed to determine the extent to which racial and ethnic disparities reflect differences in observable population characteristics versus unobserved heterogeneity across racial and ethnic groups.  RESULTS: Approximately 90% of Caucasians, 70% of African-Americans, 75% of Latinos and Asians, and 83% of Native Americans who were eligible to receive total knee replacement surgery actually did receive it. These disparities persisted in the multivariable analyses. Compared to Caucasians, the odds of having the total knee surgery was 0.46 for African Americans (p<0.0001), 0.49 for Latinos (p<0.0001), 0.45 for Asians (p<0.0001), and 0.67 for Native Americans (p<0.0001). The Blinder-Oaxaca decomposition showed that 39%-59% of the observed differences in the racial and ethnic disparities in receiving the total knee replacement surgery can be explained by observed population characteristics, most important of which are differences in age and health insurance coverage by race and ethnicity. CONCLUSIONS: Substantive racial and ethnic disparities exist in total knee replacement surgery among the eligibles. Observable population characteristics account for most of the differences, with age and health insurance key factors driving racial and ethnic disparities in receiving this surgery. Unobserved heterogeneity, which may include physician-patient relationships, mistrust, and cultural factors, are also related to these disparities.