Relationship between hemoglobin levels and risk for suspected non-alcoholic fatty liver in Taiwanese adults.

Body iron levels have recently been shown to be a strong predictor for non-alcoholic fatty liver disease (NAFLD). The aims of this study were to investigate the prevalence of NAFLD in a general adult population, and to investigate the relationship between body iron levels, NAFLD and the metabolic syndrome (MetS). 2186 adults participated in the third National Nutrition and Health Survey in Taiwan (NAHSIT, 2005-2008). The participants underwent anthropometry measurements and phlebotomy after an overnight fast, and those with excessive alcohol intake, iron overload of serum ferritin > 600 ng/ml, hepatitis viral infection and hepatocellular carcinoma were excluded. Suspected NAFLD was diagnosed by three alanine transaminase (ALT) cut-points: cut-point 1: serum ALT > 40 U/l; cut-point 2: ALT ≥ 25 U/l for male and ALT ≥ 17 U/l for female; and cut-point 3: ALT ≥ 35 U/l for male and ALT ≥ 26 U/l for female. The prevalence proportion of suspected NAFLD among Taiwanese adults was 6.6% (cut-point 1), 36% (cut-point 2); and 14.3% (cut-point 3). Body iron levels were significantly higher in individuals with suspected NAFLD compared with those without. Distribution of hemoglobin levels, but not serum ferritin levels, by decade of age showed strong correlation with the prevalence of suspected NAFLD in individuals with MetS. Multivariate adjusted odds ratio (OR) showed that the best predictors for suspected NAFLD with the MetS were hemoglobin [OR 1.43 (1.21-1.68); P < 0.0001] and hyperlipidemia [OR 1.52 (1.19-1.94); P = 0.0007]. In individuals without MetS, the adjusted OR of suspected NAFLD was markedly higher for hemoglobin [OR 1.25 (1.12-1.41); P < 0.0001]. In conclusion, adults with high hemoglobin levels (14.4 μg/dl for male and 13.2 μg/dl for female) are at the greatest risk for developing abnormal liver function. Hemoglobin test should be considered as a part of clinical evaluation for patients with NAFLD.