Diagnostic efficacy of hepatic computed tomography in the detection of body iron overload.

To determine whether or not hepatic computed tomography density is an alternative to liver biopsy for the diagnosis of body iron overload, hepatic computed tomography density was determined in healthy controls, patients with idiopathic hemochromatosis, and patients with liver disease. Hepatic computed tomography density was obtained from a single computed tomography slice taken through the liver on a General Electric CT/T 8800 scanner. In 69 controls, hepatic computed tomography density ranged from 11 to 36 units (mean = 30). Given an upper limit of normal of 36 computed tomography units, 4 of 6 untreated patients with idiopathic hemochromatosis had elevated hepatic computed tomography density compared with 0 of 8 patients after venesection (X2 = 7.5, p = 0.01). Hepatic computed tomography density correlated directly with serum ferritin (r = 0.72, p = 0.01). In 58 consecutive patients with clinical or biochemical evidence, or both, of liver disease who underwent liver biopsy for diagnostic purposes, 0 of 52 patients with normal hepatic iron had an elevated hepatic computed tomography density as compared with 4 of 6 patients with excess iron (X2 = 35, p less than 0.001). An elevated hepatic computed tomography density associated with an elevated serum ferritin indicates iron overload; however, a normal hepatic computed tomography density does not exclude iron overload. When computerized tomography is applied to patients with an unexplained elevation of the serum ferritin, it provides a noninvasive alternative to liver biopsy for the detection of excess hepatocellular iron.