Open Access
Hagenah, Kristen E
Graduate Program:
Master of Science
Document Type:
Master Thesis
Date of Defense:
March 21, 2017
Committee Members:
  • Dr. Keefe Manning, Thesis Advisor
  • Dr. William Hancock, Committee Member
  • Dr. Graham Thomas, Committee Member
  • Ghost Cells
  • Quantitative Susceptibility Mapping
  • Left Ventricular Thrombus
Every year about 735,000 people in the United States have a myocardial infarction (MI) and about 20% of post myocardial infarction patients develop left ventricular thrombus (LVT).1 A previous study by Niemann et al. found that older clots do not respond to certain anti-thrombotic therapy, illustrating the importance of knowing the thrombus age for doctors making treatment decisions. Despite advances being made in diagnostic imaging, current detection techniques are still lacking to determine the different characteristics of LVT. The present study aims to improve upon distinguishing these characteristics in magnetic resonance imaging (MRI) by studying thrombi with varying levels of hemoglobin, because it is thought that aged clots have less hemoglobin in them. Ghost cells were used in this study because they are red blood cells with little to no hemoglobin. Ghost cells were made by incubating packed O- red blood cells and lysing solution overnight in a cold room, then placed in a water bath to reform the cells without hemoglobin. These ghost cells were separated by centrifugation, and then combined with plasma from a blood donor to create ghost cell blood. The thrombi in this study were created using a backwards-facing step (BFS) model to mimic the regions of disrupted flow in the left ventricle after a MI. The BFS was integrated into a closed flow loop that circulated human blood at 0.76 L/min (Re =490). Thrombi were formed using either ghost cell blood or whole blood, and were then imaged by MRI. Images of each thrombus were then post processed using quantitative susceptibility mapping (QSM) to determine their magnetic properties. A positive QSM indicates a paramagnetic substance, such as hemoglobin. After the thrombi were imaged, they were sectioned to conduct histology and assess cellular distribution. The results showed a significant positive correlation for red blood cell (RBC) content and QSM value for the whole blood thrombi, whereas for the ghost cell thrombi the correlation was not significant. This lack of correlation for ghost cells depicts how most of the hemoglobin was removed from the RBCs. The average QSM value of the ghost cell thrombi was then compared to the average QSM value of the whole blood thrombi, and the values were considered significantly different using a two-tailed T-test with 90% confidence. These results show that the variation in hemoglobin content between the two groups was able to be detected in MRI.