Purpose: The present study aimed to determine the predictive value of the heart-liver uptake ratio (H/L ratio) of rectally administered (201)Tl scintigraphy for hepatic decompensation, which was conducted in 107 patients with cirrhosis.
Methods: We retrospectively assessed the predictive value of a noninvasive parameter, H/L ratio, for decompensation during a median follow-up period of 45.4 months using follow-up data from 1996 through 2008 for 107 patients with compensated cirrhosis. Logistic regression analysis and odds ratio estimates were used to estimate independent value of the H/L ratio on the risk of decompensation with 95% confidence intervals.
Results: At first visit, all subjects were confirmed as patients with compensated cirrhosis, 39 by liver biopsy and 68 by standard laboratory and radiological criteria. At end of the evaluation time, 81 patients remained compensated, whereas 26 patients decompensated as evidenced by ascites in 23, hepatic encephalopathy in 8, and variceal bleeding in 1 patient. First-visit parameters except bilirubin level, alanine aminotransferase (ALT), and H/L ratio and last visit parameters except ALT and aspartate aminotransferase-ALT ratio were significantly different between the 2 groups as ascertained by Wilcoxon rank sum test (P < 0.05). Among those parameters, we found that the last visit H/L ratio was a strongly reliable predictor of decompensation with an odds ratio estimates of 14.443, area under the receiver operating characteristic curve of 0.825, cutoff of 0.4, sensitivity of 73.1 %, and specificity of 71.6%.
Conclusions: This evidence indicates that in patients with compensated cirrhosis, an increased H/L ratio at follow-up may be a useful predictive parameter showing a high risk of progression to a decompensated state.