Patient-Specific Topological and Hemodynamic Analysis of Coronary Vasculature for Assessing the Risk of Intramyocardial Hemorrhage After Reperfusion Therapy Using Coronary Angiography and CMR (RF_TH_183)

Background: Primary percutaneous coronary intervention (PCI) has significantly reduced immediate mortality in STEMI patients. However, reperfusion therapy can lead to intramyocardial hemorrhage (IMH) in ~40% of these patients and is a strong predictor of adverse CV events. The mechanisms contributing to the development of IMH are unclear. We hypothesized that hemodynamic forces, influenced by the coronary vasculature's topological organization, play a key role in IMH development. We used invasive coronary angiography and CMR, along with a computational model to determine patient-specific biomarkers related to coronary network topology and hemodynamics associated with IMH.

Methods:

Pre- and post-PCI coronary angiograms (e.g., Fig. 1A) acquired from STEMI patients (n=17) were analyzed and stratified into 4 groups based on the presence of IMH and coronary collaterals (CL): IMH-/CL+, IMH+/CL-, IMH-/CL-, and IMH+/CL+, where ‘+’ denotes presence and ‘-’ denotes absence. The topology of coronary arteries and collaterals was assessed. A Navier-Stokes computational fluid dynamics model with input pressures of 100mmHg (with a 40mmHg threshold for rupture of microvessels) was used to predict IMH (Fig. 1B). CMR was performed 3-days post PCI (LGE (Fig. 1C) was used to identify MI and T2* CMR was used to identify IMH (Fig. 1D)).





Results:

There were 85±32 branches in the coronary angiogram (diameter: 1.6±1.2mm; and length: 8.9±6.7mm). More arterioles were visible in the IMH-/CL+ vs. IMH+/CL- groups (95±12 vs. 76±18), while IMH-/CL+ and IMH+/CL+ groups were closer to the average (84±12 and 81±20). Of these, 31±11 vessels of 0.9±0.2mm diameter were terminal arterioles delivering blood to the myocardium. The Junctional Exponent (JE), a biomarker of coronary network topology, was lower in the IMH-/CL+ group compared to others, both pre-PCI (1.48±0.1 vs. 1.74±0.02 IMH+/CL-, p=0.021; 1.79±0.12 IMH-/CL-, p=0.021; 1.6±0.12 IMH+/CL+, p=0.008) and post-PCI (1.44±0.11 vs. 1.87±0.19 IMH+/CL-, p=0.02; 1.64±0.07 IMH-/CL-, p=0.04), except for IMH+/CL+ (1.58±0.11, p=0.46), indicating an efficient branching structure in IMH-/CL+. Simulations suggested that prior to PCI, in patients without collaterals, the mean pressure can exceed the boundary pressure for IMH (Fig. 1B, cyan vessel segments). Post PCI, the pressure is restored but continues to exceed the IMH boundary pressure. Alternatively, if present, collaterals can redistribute blood flow both pre and post PCI therefore lowering overall blood pressure and preventing IMH. Overall, mean segment pressures were lower in IMH-/CL+ and IMH+/CL+ groups (57±2 and 59±3mmHg) vs. IMH+/CL- and IMH-/CL- groups (63±2 and 64±1mmHg, p=0.1), with lower median terminal arterial flows in CL+ patients (p=0.04).

 



Conclusion:

Topological and hemodynamic factors play a key role in the development of IMH, which is mitigated in the presence of a robust collateral circulation.