Introduction

Hemorrhagic myocardial infarction (hMI) patients are predisposed to adverse outcomes in the chronic stage of MI, yet physiological underpinnings contributing to this observation are not well understood. We investigated the connection between resting perfusion, vascular characteristics and iron deposition in infarct territories with cardiac MRI and histology. Given that iron is known to scavenge nitric oxide (NO), we hypothesized that hMI areas containing iron deposits would negatively impact endothelial function in these same regions. We tested our hypothesis by evaluating perfusion defects in patients and dogs with a history of hMI; and through histological staining for iron, endothelial cells and nitric oxide synthase (NOS) in excised myocardial sections of dogs with chronic hMI.

Methods

Patient Studies: Reperfused infarction (MI) patients (n=14) were studied with MRI at 6 months post reperfusion. T2*-w, first-pass perfusion (FPP) at rest and LGE images were acquired in a clinical 1.5T system. Rest perfusion scans matched to T2* and LGE scans were acquired and analyzed in CVI42. Imaging slices positive for MI were categorized as hemorrhagic (hMI+) or non-hemorrhagic (hMI-) based on T2* images. MI and iron-rich regions were segmented using mean+5SD and mean-2SD criteria, respectively. For each imaging slice, Relative Perfusion Index (RPI) was calculated as a ratio of signal intensity upslope in MI zone to remote zone. T-test was used to study the perfusion difference between hMI+ and hMI- territories. Statistical significance was set at p<0.05.
Animal Studies: Dogs (n=22) were subjected to ischemia reperfusion of the LAD and followed for 8 weeks. At 8 weeks post MI, T2*-w, FPP and LGE MRI scans were performed in a clinical 3T system. Data was analyzed as described in the patient study. Excised hearts were fixed in formalin and embedded in paraffin for histological sectioning and staining. Apical, MI-containing sections were stained for Prussian Blue (iron), CD31 (endothelial cells) and eNOS/iNOS (endothelial/inducible NOS). Stained slides were digitized and processed in Definiens software. An in-house Matlab script was used to register the Prussian blue slides with CD31 counterparts to facilitate regional analyses.

Results

Patients with hemorrhagic infarctions (hMI+, characterized by iron deposits), had decreased RPI in infarct zones relative to patients with non-hemorrhagic (hMI-) infarcts (0.72±0.11 vs. 0.83±0.08 respectively, p=0.003), see Fig. 1. A corroboratory reduction in RPI was also observed in the canine cohort (hMI+ (0.62 ± 0.16) vs. hMI- (0.85 ± 0.14)) at 8 weeks post-MI; see Fig. 2. Preliminary evidence shows that endothelial NOS (eNOS) is upregulated in hMI+ animals compared to hMI- animals (Fig. 3D, E); and that the number of vessels within the scarred myocardium and iron deposits are positively correlated (Fig. 3A, B, C). Similar findings in iNOS-stained slides were found in that hMI+ animals had increased iNOS activity compared to hMI- animals.

Discussion

Major adverse cardiac events (MACE) are a major hurdle in the prognosis of hemorrhagic myocardial infarctions and elucidating the pathological relationships alongside MRI markers of hemorrhagic MI may help to uncover unique therapeutic targets for the treatment of these severe types of infarctions. This study provides a significant step supporting how hemorrhagic MIs could promote adverse outcomes. Accordingly, it lays the groundwork for further studies in this field.

Conclusion

MRI findings show that hMI results in rest perfusion defects within the scarred myocardium. Histological findings from animals support the notion that iron deposition within the MI zones may impact nitric oxide synthase activity, which is a key mediator of vascular flow. These findings provide early insight into the causal relationships between chronic iron deposits and negative outcomes observed in patients with hMI.

Acknowledgements

No acknowledgement found.

References

No reference found.