Characterization of pulmonary emboli ex vivo by magnetic resonance imaging and ultrasound.

INTRODUCTION

Magnetic resonance imaging (MRI) and transesophageal ultrasound (US) are promising methods for detection and characterization of central pulmonary emboli. Both methods employ different physical principles. We tested how US and MRI characterized pulmonary emboli ex vivo.

METHODS

Thirty six ex vivo pulmonary emboli, obtained during routine autopsies of patients who died of massive pulmonary embolism, were subjected to US imaging (linear vascular probe, 5.7-10 MHz) and to high resolution three-dimensional T1-weighted spin-echo MRI. In another 3 pulmonary thromboemboli and 2 tumor emboli, we compared MRI with immunohistochemistry to platelets, red blood cells and renal carcinoma cells. We also studied model clots in vitro (retracted and non-retracted red whole-blood clots, platelet aggregates and compacted and non compacted fibrin-rich plasma clots) with MRI and US.

RESULTS

T1-weighted MR images of pulmonary thromboemboli consistently showed dark regions that corresponded to red cell-rich regions and bright layers that corresponded to platelet aggregates, but bright signal was obtained also from viable carcinoma cells and necrotic regions in tumor emboli. US images provided less structural detail than MRI, but clot retraction or compaction increased image brightness. The correlation between US and MRI characteristics of pulmonary emboli was poor.

CONCLUSIONS

T1-weighted MRI of pulmonary emboli is capable of non-invasive assessment of the red cell-rich and platelet-rich components of pulmonary thromboemboli. US imaging shows increased brightness with clot retraction or compaction. Thus, both methods detect clot characteristics that influence susceptibility to thrombolytic treatment.