Published January 1997 | Version v1
Journal article

Invasion of pulmonary venous trunk by lung cancer. Correlation of CT and MRI appearances with pathologic findings

  • 1. West China Univ. of Medical Sciences, Chengdu (China). First Univ. Hospital

Description

The authors study the spread pathway of neoplastic invasion of pulmonary vein trunk (PVT) and the correlation of CT and MR imaging features with pathologic findings. The CT and MRI findings in 31 lung cancer patients with proved PVT involvement were blindly analysed and compared with the corresponding macroscopic and microscopic appearances. Pathologic appearances of neoplastic spread along PVT were classified into three types: The cancer abutted the PVT wall side by side and mostly invaded the adventitia in 5 cases; The cancer circularly encased and infiltrated PVT wall, causing remarkable narrowing of PVT lumen and occlusion in 16 cases; Longitudinal and creeping infiltration along the PVT wall was observed in 10 cases. Correspondingly, CT and MRI findings were also classified into three groups: The mass was closely attached to one side of PVT wall but the signal of blood flow was normal; The PVT was encased by mass becoming either narrowed or obliterated; The mass was attached to or encased the PVT outside pericardium, in addition, the pericardial segment of PVT became widened on CT and the signal of blood flow was abnormal on MRI. The Kappa values between pathologic appearance and imaging findings on ct and MRI were 0.37 and 1.00 respectively. conclusion: lung cancer invaded PVT by means of attaching to or infiltrating the vessel wall either circumferentially or longitudinal in a creeping pattern. MRI findings correlated better than CT with the pathological patterns. The authors suggested that MRI study is a useful imaging modality for preoperative assessment of PVT status in patients with lung cancer

Additional details

Publishing Information

Journal Title
Chinese Journal of Radiology
Journal Volume
31
Journal Issue
1
Journal Page Range
p. 20-24.
ISSN
1005-1201
CODEN
CHFSAG