RESUMO
BACKGROUND: Recent evidence suggested a potential role of complement fraction C3 as a biomarker of nonalcoholic fatty liver disease (NAFLD) in the general population. Aim of this study was to evaluate the performance of C3 for prediction of NAFLD in RA patients. MATERIALS AND METHODS: For the present study, consecutive RA patients were recruited. NAFLD was diagnosed according to predefined ultrasonographic (US) criteria. For comparison, the hepatic steatosis index (HSI) was calculated. RESULTS: Of 164 consecutive RA patients, 41 (25%) were diagnosed with NAFLD. The NAFLD group had a significant lower proportion of females (P = 0·04), higher BMI (P < 0·0001), C-reactive protein (P = 0·04), complement C3 (P = 0·001) and HSI (P = 0·003). In a logistic regression model, only male sex (OR 2·65, 95% CI: 1·08-6·50, P = 0·03), increasing BMI (OR 1·22, 95% CI: 1·02-1·46, P = 0·03) and complement C3 (OR 5·05, 95% CI: 1·06-23·93, P = 0·04) were associated with higher likelihood of being diagnosed with NAFLD. Finally, we built ROC curves for BMI, complement C3 and their combination for prediction of having NAFLD. The best cut-off for BMI was 28·5 kg/m2 and yielded a sensitivity of 66% and a specificity of 71%; the best cut-off for complement C3 was 1·23 g/L and yielded a sensitivity of 76% and a specificity of 64% for classification of NAFLD cases. CONCLUSIONS: Our results provide preliminary evidence for a potential role of complement C3 as a surrogate biomarker of NAFLD in RA patients.