Morphological and Stress Vulnerability Indices for Human Coronary Plaques and Their Correlations with Cap Thickness and Lipid Percent: An IVUS-Based Fluid-Structure Interaction Multi-patient Study

Liang Wang, Jie Zheng, Akiko Maehara, Chun Yang, Kristen Billiar, Zheyang Wu, Richard G. Bach, David Muccigrosso, Gary S. Mintz, Dalin TangPLOS Computational Biology 2015 · DOI

TL;DR

This study introduces novel morphological and stress-based indices to quantify coronary plaque vulnerability, addressing the challenge of predicting rupture without direct rupture data. By integrating in vivo IVUS imaging with computational fluid-structure interaction (FSI) models, the researchers found strong correlations between these new indices and key plaque features, suggesting their potential for improved patient screening and risk assessment.

What Problem Does This Paper Address?

Atherosclerotic plaque rupture is a primary cause of cardiovascular events like heart attacks. However, accurately predicting which plaques are most likely to rupture (i.e., identifying "vulnerable plaques") is a significant clinical challenge. This difficulty stems from the lack of direct in vivo data on plaque rupture events, making it hard to validate predictive models. Current methods often rely on morphological characteristics, but mechanical stress within the plaque is also believed to play a crucial role. This paper aims to bridge this gap by developing and validating quantitative indices that combine both morphological and mechanical stress information to better assess plaque vulnerability.

Methods

The study employed a multi-faceted approach:

  • Data Acquisition: In vivo intravascular ultrasound (IVUS) data were collected from 14 patients. This data provided detailed cross-sectional images of coronary plaques. X-ray angiography was also used to capture vessel curvature and cyclic bending due to cardiac motion.
  • Patient-Specific 3D Model Construction: Using the IVUS data, 3D computational models of coronary plaques were built. These models incorporated fluid-structure interaction (FSI) to simulate blood flow dynamics and mechanical stresses within the plaque and vessel wall. Cyclic bending of the vessel, derived from angiography, was also included to mimic physiological conditions.
  • Material Modeling: A modified Mooney-Rivlin material model was used to describe the hyperelastic, anisotropic, and nearly-incompressible properties of the human coronary artery. Plaque components (calcification and lipid) were modeled as hyperelastic, isotropic, and nearly-incompressible.
  • Computational Simulation: The FSI models were solved using ADINA software, simulating cardiac cycles to obtain plaque wall stress/strain and flow shear stress.
  • Index Development:
    • Morphological Indices: Three indices—cap index, lipid index, and morphological index—were developed based on American Heart Association (AHA) plaque classification schemes. These indices quantify features like cap thickness and lipid content. The morphological index was defined as the maximum of the cap index and lipid index.
  • Stress Index: A stress index was introduced, derived from critical plaque wall stress (CPWS) values, categorized into five stress intervals.
  • Correlation Analysis: Linear Mixed-Effects (LME) models were used to analyze the correlations between the mechanical indices (CPWS, CPWSn, CFSS) and the morphological factors (min cap thickness, cap index, lipid percentage, lipid index, morphological index). The agreement between the stress index and the morphological indices was also assessed.

Datasets

  • IVUS Data: In vivo intravascular ultrasound (IVUS) data from 14 patients were used to construct patient-specific 3D plaque models.
  • X-ray Angiography Data: Used to capture vessel curvature and cyclic bending.

The study generated a dataset of 617 IVUS slices from these 14 patients, for which morphological and mechanical data were extracted and analyzed.

Key Insights and Ideas

  • Bridging Morphology and Mechanics: The core idea is that plaque vulnerability is a complex interplay of its physical structure (morphology) and the mechanical forces it experiences. This study provides a framework to quantify both aspects simultaneously.
  • Novel Indices for Quantifiable Assessment: Developing specific indices (cap index, lipid index, morphological index, and stress index) allows for a more quantitative assessment of vulnerability, moving beyond qualitative descriptions.
  • Stress as a Key Indicator: Critical plaque wall stress (CPWS) emerged as a significant mechanical factor strongly correlated with morphological features associated with rupture.
  • Complementary Nature of Indices: The study highlights that morphological and stress indices, while correlated, also have differences. This suggests that using both types of indices together could lead to more robust plaque assessment strategies.

Key Results

  • Strong Correlations:
    • CPWS showed significant correlations with morphological index (r = 0.7411) and cap index (r = 0.7852).
  • The stress index demonstrated a strong agreement with the morphological index (66.77%) and a high correlation with cap index (r = 0.8185).
  • Min cap thickness was strongly correlated with CPWS (r = -0.6414), with thinner caps leading to higher stress.
  • Weaker Correlations: Lipid percentage and lipid index showed weaker correlations with CPWS compared to cap-related features.
  • Agreement: The stress index and morphological index showed a substantial agreement rate of 66.77%, indicating their potential to complement each other.

Limitations and Future Work

  • Patient-Specific Material Properties: The study acknowledges that patient-specific material properties were not available and were approximated. Future work may involve developing methods to measure these in vivo.
  • IVUS Resolution: The resolution of IVUS (100-150 microns) is insufficient to clearly identify very thin caps (<100 microns). The authors suggest combining IVUS with Optical Coherence Tomography (OCT) for improved spatial resolution and cap thickness quantification.
  • Single Angiography View: Only one view angle was used for angiography to reconstruct 3D curvature. Using multiple views would provide a more accurate 3D curvature.
  • Adventitial Layer Omission: The adventitial layer of the vessel was not included in the models due to lack of IVUS data, which could affect stress/strain calculations.
  • Microcalcifications: Microcalcifications were not included in the morphological index due to imaging limitations.
  • Heart-Vessel Interaction: The interaction between heart motion and coronary bending was simplified. Future models could incorporate this for more comprehensive analysis.

The authors suggest that further investigation into plaque rupture and clinical event data is needed to validate these indices and that combining morphological and stress-based assessments holds promise for improved plaque vulnerability assessment and patient screening.

Summary updated

BibTeX
@article{wang2015morphological,
  author = {Liang Wang and Jie Zheng and Akiko Maehara and Chun Yang and Kristen Billiar and Zheyang Wu and Richard G. Bach and David Muccigrosso and Gary S. Mintz and Dalin Tang},
  title = {{Morphological and Stress Vulnerability Indices for Human Coronary Plaques and Their Correlations with Cap Thickness and Lipid Percent: An IVUS-Based Fluid-Structure Interaction Multi-patient Study}},
  journal = {PLOS Computational Biology},
  year = {2015},
  doi = {10.1371/journal.pcbi.1004652},
}
Figure