Regional metabolic liver function measured in patients with cirrhosis by 2-[¹⁸F]fluoro-2-deoxy-D-galactose PET/CT

Abstract

Background & aims: There is a clinical need for methods that can quantify regional hepatic function non-invasively in patients with cirrhosis. Here we validate the use of 2-[(18)F]fluoro-2-deoxy-d-galactose (FDGal) PET/CT for measuring regional metabolic function to this purpose, and apply the method to test the hypothesis of increased intrahepatic metabolic heterogeneity in cirrhosis.

Methods: Nine cirrhotic patients underwent dynamic liver FDGal PET/CT with blood samples from a radial artery and a liver vein. Hepatic blood flow was measured by indocyanine green infusion/Fick's principle. From blood measurements, hepatic systemic clearance (Ksyst, Lblood/min) and hepatic intrinsic clearance (Vmax/Km, Lblood/min) of FDGal were calculated. From PET data, hepatic systemic clearance of FDGal in liver parenchyma (Kmet, mL blood/mL liver tissue/min) was calculated. Intrahepatic metabolic heterogeneity was evaluated in terms of coefficient-of-variation (CoV, %) using parametric images of Kmet.

Results: Mean approximation of Ksyst to Vmax/Km was 86% which validates the use of FDGal as PET tracer of hepatic metabolic function. Mean Kmet was 0.157 mL blood/mL liver tissue/min, which was lower than 0.274 mL blood/mL liver tissue/min, previously found in healthy subjects (p<0.001), in accordance with decreased metabolic function in cirrhotic livers. Mean CoV for Kmet in liver tissue was 24.4% in patients and 14.4% in healthy subjects (p<0.0001). The degree of intrahepatic metabolic heterogeneity correlated positively with HVPG (p<0.05).

Conclusions: A 20-min dynamic FDGal PET/CT with arterial sampling provides an accurate measure of regional hepatic metabolic function in patients with cirrhosis. This is likely to have clinical implications for the assessment of patients with liver disease as well as treatment planning and monitoring.

Fig. 1. Examples of parametric images of the hepatic systemic clearance of FDGal (K_met, ml blood/ml tissue/min) generated from dynamic FDGal PET/CT recordings

Top row is from a cirrhotic patient (panels A and B) and bottom row is from a healthy subject (panels C and D). Left images are transaxial, right images coronal. The patient is a 57 year old male with cirrhosis (alcohol and hepatitis C virus), a global liver function of 45% of expected value as measured by a galactose elimination capacity (GEC) test, and portal hypertension (HVPG 30.5 mmHg).

Fig. 2. Intrahepatic metabolic heterogeneity plotted against global and regional metabolic function and HVPG

The intrahepatic metabolic heterogeneity is expressed in terms of the coefficient of variation (CoV) for K_met (ml blood/ml liver tissue/min) of hepatic FDGal metabolism. Individual values from cirrhotic patients (▼) and healthy subjects (Δ) are plotted against global metabolic liver function (A), regional metabolic liver function (B), and HVPG (C). There was a positive correlation between CoV and HVPG in the patients; the correlation is shown by the straight line with 95% confidence intervals (r =0.69; p <0.05).