Ethnic differences in inflammation: Association with obesity and insulin resistance

Inflammation has been linked to cardiovascular disease and diabetes. Minorities including African Americans and Hispanics have greater risk of both cardiovascular disease and diabetes than non-Hispanic whites. Additionally, it is also known that racial/ethnic groups differ in levels of inflammatory markers. STUDY QUESTION: Are ethnic differences in inflammation explained by differences in metabolic traits including adiposity, insulin resistance, glucose tolerance and liver fat? METHODS: This study is a cross-sectional analysis of 1,027 non-diabetic participants in the Insulin Resistance Atherosclerosis Study (IRAS), a large tri-ethnic population (African Americans, non-Hispanic whites, and Hispanics) across different states of glucose tolerance. Markers of inflammation examined were C-reactive protein (CRP) and plasminogen activator inhibitor-1 (PAI-1). Explanatory variables examined include BMI, insulin sensitivity index (SI), 2-h glucose assessed by the oral glucose tolerance test, and liver fat (non-alcoholic fatty liver disease [NAFLD] liver fat score). Racial and ethnic differences in markers of inflammation are analyzed by one-way ANCOVA. Independent relationships between markers of inflammation and demographic and metabolic variables were established using linear regression analysis. RESULTS: BMI, 2-h glucose, SI, and NAFLD score had comparable relationships with CRP concentration across the racial/ethnic groups in the study (p for ethnic interaction \u3e0.11 for all relationships). These metabolic variables also had similar relationships with PAI-1 concentration across racial/ethnic groups (p for ethnic interaction \u3e0.18 for all relationships) except for BMI (p for ethnic interaction = 0.04). Compared with non-Hispanic whites, both African Americans (2.10 ± 0.20 vs. 1.51 ± 0.08 mg/L, p = 0.001) and Hispanics (2.01 ± 0.17 mg/L, p = 0.003) had higher CRP concentration. PAI-1 concentration was also higher in Hispanics (18.9 ± 0.97 vs. 15.5 ± 0.63 ng/ml, p = 0.003), but lower in African Americans (13.3 ± 0.82 ng/ml, p = 0.036). BMI, 2-h glucose, SI, and NAFLD score largely explained the excess concentration of CRP in African Americans and PAI-1 in Hispanics, but not the excess concentration of CRP in Hispanics and the lower levels of PAI-1 in African Americans. CONCLUSIONS: CRP and PAI-l concentrations have a similar relationship with metabolic variables across ethnic groups. Ethnic differences are demonstrated for both CRP and PAI-1 levels. However, adiposity, glucose tolerance, insulin sensitivity, and liver fat did not account for all the ethnic differences in CRP and PAI-l. This research supports the need for improved methods of identifying and stratifying disease risk allowing for better allocation of resources in healthcare. Future research needs to examine other determinants of racial/ethnic differences in CRP and PAI-1 levels (e.g., genetic factors) as well as the impact of race/ethnicity on the relationship between CRP and PAI-l levels and cardiovascular disease.