Background: Indocyanine green (ICG) clearance has been proposed as a quantitative liver function test several decades ago. Interest in this method has been renewed following the development of finger pulse densitometry for noninvasive estimation of the ICG plasma disappearance rate (PDR). On the other hand, the model for end-stage liver disease (MELD), which is based on routine laboratory parameters, is widely used for estimation of short-term survival in cirrhosis, but its prognostic value in critically ill cirrhotic patients is unclear.
Aims: The aim of the present study was to compare the diagnostic accuracy of ICG PDR vs. MELD for estimation of short-term prognosis in cirrhotic patients.
Methods: Ninety consecutive cirrhotic patients who were admitted for decompensated disease or were being evaluated for liver transplantation were screened. Patients who underwent liver transplantation within the following 90 days and those with hepatocellular carcinoma were excluded. In the remaining 70 patients, routine laboratory parameters and ICG clearance were analysed. Following an injection of ICG 0.25 mg/kg, PDR was measured by finger pulse densitometry. The diagnostic accuracy of ICG PDR and MELD for prediction of 90-day survival was assessed by receiver-operating characteristic (ROC) curve analysis.
Results: ROC curve analysis revealed superior diagnostic accuracy for MELD as compared with ICG PDR in predicting 90-day survival (area under the ROC curve 0.89 vs. 0.71). A MELD cut-off of 22 provided the best discrimination for prediction of 90-day survival.
Conclusions: MELD is superior to ICG PDR for estimation of short-term survival in patients with decompensated cirrhosis.