Serum ferritin and transferrin saturation in patients with chronic alcoholic and non-alcoholic liver diseases

To compare serum ferritin concentration and transferrin saturation in patients with alcoholic and non-alcoholic chronic liver diseases.

Consecutive patients with liver diseases.

The department of internal medicine in a teaching hospital.

Three hundred and twelve patients with different liver diseases consecutively admitted between 1987 and 1992.

None.

Fasting serum iron, transferrin and ferritin.

Serum ferritin was increased above 200 micrograms L-1 in all 18 patients with haemochromatosis (range 310-6500 micrograms L-1), in 64 of 111 alcoholics (58%) and in 30 of 137 (22%) with chronic non-alcoholic liver diseases (P < 0.01). Twelve of 111 alcoholics (11%) had serum ferritin above 1000 micrograms L-1 compared with one of 137 (0.7%) with chronic non-alcoholic liver diseases. In 13 alcoholics who abstained after admission, serum ferritin decreased from 1483 micrograms L1 +/- 1134 to 388 micrograms L-1 +/- 237 (P < 0.001) after 1 1/2 to 6 weeks. The transferrin saturation was increased above 62% in 13 of 18 patients (72%) with haemochromatosis, in 16 of 105 alcoholics (15.2%) and in three of 132 (2.3%) with chronic non-alcoholic liver disease (P < 0.01).

Serum ferritin is more frequently elevated in abusing patients with alcoholic liver disease than in patients with other chronic liver diseases such as autoimmune liver diseases and hepatitis C. Because serum ferritin decreases rapidly during abstinence, the measurement of ferritin for the detection of haemochromatosis in patients abusing alcohol should be postponed until the patients are abstaining. Most of the patients with increased serum ferritin have normal transferrin saturation values which can be used to separate them from haemochromatosis.