Hepatic vitamin A stores should be the best early indicator of vitamin A status because more than 90% of total body vitamin A is stored in the liver. The objective of the present study was to evaluate the hepatic vitamin A stores in all stages of chronic liver disease (CLD), including chronic hepatitis, liver cirrhosis, and hepatocellular carcinoma (HCC). One hundred forty-four patients (age 55.34 ± 9.38 years) were evaluated in a cross-sectional study. Vitamin A nutrition status was analyzed by serum retinol levels and relative dose-response (RDR) method. Patients with cholestasis were excluded from the sample group. Biochemical, clinical, and anthropometric evaluations were performed. Vitamin A deficiency (VAD) was detected in 51.4% of all patients. Patients with adequate levels of serum retinol presented adequate liver vitamin A reserves; in contrast, nearly half the patients with low serum retinol levels presented adequate levels of retinol in the liver, although none of the patients with hepatitis had this condition. Therefore, the effectiveness of the RDR method for evaluating vitamin A nutrition status was limited in patients with cirrhosis and HCC, perhaps due to the advanced age of these patients, since those in the chronic hepatitis group, who were younger, responded adequately to the test. Thus, the RDR method should be modified when applied to later stages of CLD, considering the time and dose of retinyl palmitate supplementation, as VAD may be a risk factor for the progression of the disease.