DataSheet_1.docx

Abstract

Objective: Nonalcoholic fatty liver disease (NAFLD) is becoming a global public health challenge. A convenient NAFLD indicator will greatly facilitate risk appraisal and prevention. As a readily available and inexpensive hematological index in routine clinical examinations, red blood cells (RBCs) are gaining increasing attention in many diseases, such as metabolic syndrome, but their association with NAFLD is unknown. Methods: This health management cohort study included 27,112 subjects (17,383 non-NAFLD and 9,729 NAFLD) with up to 5 years of follow-up (median 2.8 years). NAFLD was diagnosed by ultrasonography. NAFLD severity was categorized as mild, moderate, or severe. The generalized estimation equation (GEE), an extension of generalized linear models that allows for analysis of repeated measurements, was used to analyze the association between RBC count and NAFLD. Results: Overall, 4,332 of 17,383 (24.9%) subjects without NAFLD at baseline developed NAFLD. Incident NAFLD risk was positively associated with RBC count. After adjustment for hemoglobin and other confounders, the risk of incident NAFLD was 21%, 32%, and 51% higher in the second, third, and fourth RBC count quartiles, respectively, than in the lowest quartile. In 1,798 of 9,476 (19.0%) subjects with NAFLD at baseline, the severity of NAFLD increased. NAFLD progression risk increased progressively as RBC count increased (P for trend < 0.001). Every one-unit (10^12 cells/L) increase in RBC count was associated with a 53% (OR 1.53 [95% CI 1.32-1.77]) increased risk for NAFLD progression. Conclusions: Elevated RBC count was independently associated with a high risk of NAFLD incidence and progression. This finding revealed a convenient NAFLD risk indicator.