Michael A Fischer1,2, Nikolaos Kartalis3,4, Aristeidis Grigoriadis3,4, Louiza Loizou3,4, Per Stål5, Bertil Leidner3,4, Peter Aspelin3,4, Torkel B Brismar3,4. 1. Department of Clinical Science, Intervention and Technology, Division of Medical Imaging and Technology, Karolinska Institute, 14186, Stockholm, Sweden. michaelalexander.fischer@usz.ch. 2. Department of Radiology, Karolinska University Hospital, Huddinge, 14186, Stockholm, Sweden. michaelalexander.fischer@usz.ch. 3. Department of Clinical Science, Intervention and Technology, Division of Medical Imaging and Technology, Karolinska Institute, 14186, Stockholm, Sweden. 4. Department of Radiology, Karolinska University Hospital, Huddinge, 14186, Stockholm, Sweden. 5. Department of Hepatology, Karolinska University Hospital, Huddinge, 14186, Stockholm, Sweden.
Abstract
PURPOSE: To evaluate the diagnostic performance of dynamic perfusion CT (P-CT) for detection of hepatocellular carcinoma (HCC) in the cirrhotic liver. MATERIALS AND METHODS: Twenty-six cirrhotic patients (19 men, aged 69 ± 10 years) with suspicion of HCC prospectively underwent P-CT of the liver using the 4D spiral-mode (100/80 kV; 150/175mAs/rot) of a dual-source system. Two readers assessed: (1) arterial liver-perfusion (ALP), portal-venous liver-perfusion (PLP) and hepatic perfusion-index (HPI) maps alone; and (2) side-by-side with maximum-intensity-projections of arterial time-points (art-MIP) for detection of HCC using histopathology and imaging follow-up as standard of reference. Another reader quantitatively assessed perfusion maps of detected lesions. RESULTS: A total of 48 HCCs in 21/26 (81%) patients with a mean size of 20 ± 10 mm were detected by histopathology (9/48, 19%) or imaging follow-up (39/48, 81%). Detection rates (Reader1/Reader2) of HPI maps and side-by-side analysis of HPI combined with arterial MIP were 92/88% and 98/96%, respectively. Positive-predictive values were 63/63% and 68/71%, respectively. A cut-off value of ≥85% HPI and ≥99% HPI yielded a sensitivity and specificity of 100%, respectively, for detection of HCC. CONCLUSION: P-CT shows a high sensitivity for detection of HCC in the cirrhotic liver. Quantitative assessment has the potential to reduce false-positive findings improving the specificity of HCC diagnosis. KEY POINTS: • Visual analysis of perfusion maps shows good sensitivity for detection of HCC. • Additional assessment of anatomical arterial MIPs further improves detection rates of HCC. • Quantitative perfusion analysis has the potential to reduce false-positive findings. • In cirrhotic livers, a hepatic-perfusion-index ≥ 9 9% might be specific for HCC.
PURPOSE: To evaluate the diagnostic performance of dynamic perfusion CT (P-CT) for detection of hepatocellular carcinoma (HCC) in the cirrhotic liver. MATERIALS AND METHODS: Twenty-six cirrhotic patients (19 men, aged 69 ± 10 years) with suspicion of HCC prospectively underwent P-CT of the liver using the 4D spiral-mode (100/80 kV; 150/175mAs/rot) of a dual-source system. Two readers assessed: (1) arterial liver-perfusion (ALP), portal-venous liver-perfusion (PLP) and hepatic perfusion-index (HPI) maps alone; and (2) side-by-side with maximum-intensity-projections of arterial time-points (art-MIP) for detection of HCC using histopathology and imaging follow-up as standard of reference. Another reader quantitatively assessed perfusion maps of detected lesions. RESULTS: A total of 48 HCCs in 21/26 (81%) patients with a mean size of 20 ± 10 mm were detected by histopathology (9/48, 19%) or imaging follow-up (39/48, 81%). Detection rates (Reader1/Reader2) of HPI maps and side-by-side analysis of HPI combined with arterial MIP were 92/88% and 98/96%, respectively. Positive-predictive values were 63/63% and 68/71%, respectively. A cut-off value of ≥85% HPI and ≥99% HPI yielded a sensitivity and specificity of 100%, respectively, for detection of HCC. CONCLUSION: P-CT shows a high sensitivity for detection of HCC in the cirrhotic liver. Quantitative assessment has the potential to reduce false-positive findings improving the specificity of HCC diagnosis. KEY POINTS: • Visual analysis of perfusion maps shows good sensitivity for detection of HCC. • Additional assessment of anatomical arterial MIPs further improves detection rates of HCC. • Quantitative perfusion analysis has the potential to reduce false-positive findings. • In cirrhotic livers, a hepatic-perfusion-index ≥ 9 9% might be specific for HCC.
Authors: Edgar Bendik; Peter B Noël; Daniela Münzel; Alexander A Fingerle; Martin Henninger; Christian Markus; Alain Vlassenbroek; Ernst J Rummeny; Martin Dobritz Journal: Eur Radiol Date: 2013-09-03 Impact factor: 5.315
Authors: Michael A Fischer; Bertil Leidner; Nikolaos Kartalis; Anders Svensson; Peter Aspelin; Nils Albiin; Torkel B Brismar Journal: Eur Radiol Date: 2013-08-31 Impact factor: 5.315
Authors: Thorsten Sellerer; Peter B Noël; Manuel Patino; Anushri Parakh; Sebastian Ehn; Sascha Zeiter; Jasmin A Holz; Johannes Hammel; Alexander A Fingerle; Franz Pfeiffer; David Maintz; Ernst J Rummeny; Daniela Muenzel; Dushyant V Sahani Journal: Eur Radiol Date: 2018-02-05 Impact factor: 5.315
Authors: Michael A Fischer; Katharina Brehmer; Anders Svensson; Peter Aspelin; Torkel B Brismar Journal: Eur Radiol Date: 2016-02-25 Impact factor: 5.315
Authors: Michael A Fischer; Herman P Marquez; Sonja Gordic; Bertil Leidner; Ernst Klotz; Peter Aspelin; Hatem Alkadhi; Torkel B Brismar Journal: Eur Radiol Date: 2016-07-01 Impact factor: 5.315