OBJECTIVE: The goal of our study was to determine the relative value of multiple MR features in predicting clinical progression of disease in patients with compensated cirrhosis. MATERIALS AND METHODS: The MR examinations of 23 patients with compensated cirrhosis (Child A) were retrospectively reviewed independently by two radiologists and correlated with clinical progression after follow-up of all patients for more than 12 months each (12-87 months: average, 39 months) by the same experienced hepatologist. Clinical progression was defined as an increase of the Child grade or the Pugh score by at least two points (5- to 15-point scale). In the initial MR study of each patient, the following MR findings were assessed by each radiologist independently: volume indexes of the spleen and each segment of the liver (based on three-axis measurements), nodular surface, regenerative nodules, ascites, iron or fat deposition, and varices or collaterals. RESULTS: The volume index of the spleen was the most accurate predictor of clinical progression (p = .001), the next most accurate was the number of sites of varices or collaterals (p = .002), and the third most accurate was the ratio of caudate lobe to right lobe volume index (p = .02). Other MR findings failed to correlate with clinical progression. CONCLUSION: As revealed on MR imaging, the volume index of the spleen, the severity of varices, and the volume index ratio of caudate lobe to right lobe can be used to help predict clinical progression of disease in patients with compensated cirrhosis.
OBJECTIVE: The goal of our study was to determine the relative value of multiple MR features in predicting clinical progression of disease in patients with compensated cirrhosis. MATERIALS AND METHODS: The MR examinations of 23 patients with compensated cirrhosis (Child A) were retrospectively reviewed independently by two radiologists and correlated with clinical progression after follow-up of all patients for more than 12 months each (12-87 months: average, 39 months) by the same experienced hepatologist. Clinical progression was defined as an increase of the Child grade or the Pugh score by at least two points (5- to 15-point scale). In the initial MR study of each patient, the following MR findings were assessed by each radiologist independently: volume indexes of the spleen and each segment of the liver (based on three-axis measurements), nodular surface, regenerative nodules, ascites, iron or fat deposition, and varices or collaterals. RESULTS: The volume index of the spleen was the most accurate predictor of clinical progression (p = .001), the next most accurate was the number of sites of varices or collaterals (p = .002), and the third most accurate was the ratio of caudate lobe to right lobe volume index (p = .02). Other MR findings failed to correlate with clinical progression. CONCLUSION: As revealed on MR imaging, the volume index of the spleen, the severity of varices, and the volume index ratio of caudate lobe to right lobe can be used to help predict clinical progression of disease in patients with compensated cirrhosis.