Plaque heterogeneity and the spatial distributions of its components dictate drug-coated balloon therapy

Abstract

Drug-coated balloon (DCB) angioplasty is one of the potential approaches to alleviating in-stent restenosis and treating peripheral artery disease. An in-silico model has been developed for sirolimus drug eluted from an inflated balloon in a patient-specific arterial cross-section consisting of fibrous tissue, fibrofatty tissue, dense calcium, necrotic core, and healthy tissue. The convection-diffusion-reaction equation represents the transport of drug, while drug binding, both specific and non-specific, can be modelled as a reaction process. The Brinkman equations describe the interstitial flow in porous tissue. An image processing technique is leveraged for reconstructing the computational domain. The Marker and Cell, and Immersed Boundary Methods are used to solve the set of governing equations. The no-flux interface condition and convection do amplify the tissue content, and the regions of dense calcium and necrotic core limited to or extremely close to the interface pose a clinical threat to DCB therapy. Simulations predict the effects of the positioning and clustering of plaque components in the domain. This study demands extensive intravascular ultrasound-derived virtual histology (VH-IVUS) imaging to understand the plaque morphology and determine the relative positions of different plaque compositions about the lumen-tissue interface, which have a significant impact on arterial pharmacokinetics.

Keywords: Drug-coated balloon; MAC and IBM method; Plaque heterogeneity; Spatial distributions; Tissue content.

Figure 1

Geometry and computational domain with heterogeneous tissue compositions: (a) Ref. model; (b) Hard model; (c) Soft model; (d) Circular cross-section.

Figure 2

(a) A typical MAC cell; (b) IBM nodes demarcation and interpolation; (c) Interstitial velocity-vector (not to scale); (d) Sirolimus release profile during balloon angioplasty.

Figure 3

Model validation (Balloon inflation time = 30 s.): (a) Free sirolimus; (b) REC-bound sirolimus; (c) ECM-bound sirolimus.

Figure 4

Effect of convection and interface condition (Ref. model; Balloon inflation time = 30 s): (a, b, c) Averaged drug concentration; (d) Tissue content; (e, f) Fractional Effect.

Figure 5

Spatio-temporal patterns for the Ref. model in the presence of convection and no-flux interface condition (Balloon inflation time = 30 s): (a, b, c) Free sirolimus; (d, e, f) REC-bound sirolimus; (g, h, i) ECM-bound sirolimus.

Figure 6

Effect of varying tissue compositions in the presence of convection and no-flux interface condition ($J_I=0$, Balloon inflation time = 30 s): (a, b, c) Averaged drug concentration; (d) Tissue content; (e, f) Fractional effect.

Figure 7

Spatial patterns for varying tissue compositions in the presence of convection and no-flux interface condition ($J_I=0$) at $t=30$ min. (Balloon inflation time=30 s): (a, b, c, d) Free; (e, f, g, h) REC-bound; (i, j, k, l) ECM-bound sirolimus.

Figure 8

Effect of spatial positions of plaque components appeared in non-clustered manner at $t=4$ h. on (a) Tissue content; (b) Fraction of REC-bound sirolimus; (c) Fraction of ECM-bound sirolimus.

Figure 9

Effect of spatial positions of plaque components appeared in the clustered manner at $t=4$ hrs. on (a) Tissue content; (b) Fraction of REC-bound sirolimus; (c) Fraction of ECM-bound sirolimus.

Figure 10

Large time behaviour of DCB therapy in hard and type-A models in the presence of convection. Balloon inflation time = 30 s): (a, b, c) Hard model; (d, e, f) Type-A model.

Figure 11

Sensitivity analysis at $t=4$ h.: (a, d, g) Tissue content; (b, e, h) Fraction of REC-bound sirolimus; (c, f, i) Fraction of ECM-bound sirolimus.