Liver function quantification of patients with portal vein embolization using dynamic contrast-enhanced MRI for assessment of hepatocyte uptake and elimination. (August 2020)

• Record Type: Journal Article
• Authors: Hindel, Stefan; Geisel, Dominik; Alerić, Ivana; Theilig, Dorothea; Denecke, Timm; Lüdemann, Lutz
• Is Part Of: Physica medica. Volume 76(2020)
• Page Start: 207
• Page End: 220
• Publication Date: 2020-08
• Subjects: MR functional liver imaging -- DCE-MRI -- Gd-EOB-DTPA -- Tracer kinetic modeling; Medical physics -- Periodicals; Biophysics -- Periodicals; Biophysics -- Periodicals; Imagerie médicale -- Périodiques; Radiothérapie -- Périodiques; Rayons X -- Sécurité -- Mesures -- Périodiques; Physique -- Périodiques; Médecine -- Périodiques; 610.153
• DOI: 10.1016/j.ejmp.2020.07.002
• Languages: English
• ISSNs: 1120-1797

Abstract

Abstract: Purpose: We evaluated pharmacokinetic models which quantify liver function including biliary elimination based on a dynamic Gd-EOB-DTPA-enhanced magnetic resonance imaging (MRI) technique with sparse data collection feasible in clinical routine. Methods: Twelve patients with embolized liver segments following interventional treatment of primary liver cancer or hepatic metastasis underwent MRI. During Gd-EOB-DTPA bolus administration, a 3D dynamic gradient-echo (GRE) MRI examination was performed over approx. 28 min. Interrupted data sampling was started approx. 5 min after contrast agent administration. Different implementations of dual-inlet models were tested, namely the Euler method (DE) and convolution with residue functions (C). A simple uptake model (U) and an uptake- elimination model (UE) extended by incorporating the biliary contrast agent elimination rate ( K e ) were evaluated. Results: The uptake-elimination model, calculated via the simple Euler method (UE- DE) and by convolution (UE-C), yielded similar overall estimates in terms of fitting quality and agreement with published values. The Euler method was approx. 50 times faster and yielded a mean elimination rate of K e = 1.8 ± 1.2 mL/(min · 100 mL) in nonembolized liver tissue, which was significantly higher ( p = 8.8 · 10 - 4 ) than in embolized tissue K e = 0.4 ± 0.4 mL/(min · 100 mL). Fractional hepatocyte volume v h was not significantly higher in nonembolized tissue (52.4 ± 13.4 mL/100 mL) compared to embolized tissue (44.4 ± 26.1 mL/100 mL). Conclusions: Interrupted late enhancement MRI data sampling in conjunction with the uptake-elimination model, deconvolved by integration of the differential rate equation and combined with the simple uptake model implemented with the Euler method (U-DE), turned out to be a stable and practical method for reliable noninvasive assessment of liver function.