Aortic Remodeling Kinetics in Response to Coarctation-Induced Mechanical Perturbations

Figures

• 

  Figure 1. Experimental protocol to characterize hypertension precursors in response to CoA in rabbits. Male New Zealand White rabbits were randomly selected for a control group or designated to undergo descending thoracic CoA (nine study groups: three severities and three durations) via left thoracotomy in the third intercostal space. Weekly Doppler ultrasound imaging characterized local hemodynamic and morphology adaptation to CoA before PC-MRI prior to the final week. At the end of the protocol, invasive characterization was conducted for blood pressure measurement, and the aorta was dissected to quantify arterial stiffening and dysfunction through uniaxial extension testing and myograph analysis, respectively.

• 

  Figure 2. Quantification of hemodynamics (A) and morphology (B) in a rabbit model of CoA. Doppler ultrasound follow-up was performed weekly to quantify the temporal evolution of hemodynamic changes (A, spectral Doppler) and morphology (B, Doppler b-mode) in response to the CoA.

• 

  Figure 3. Representative thickness and body weight evolution over time for a CoA rabbit. Thickness (squares) was quantified as a function of age and through weekly ultrasound imaging at follow-up where 2D sagittal sections of the aorta were identified in b-mode images, and triplicate thickness measurements were performed with mean values shown at each timepoint. Body weight evolution was extrapolated using a sigmoid fit (dashed line) to the measured body weights (circles).

• 

  Figure 4. High-fidelity BP measurements proximal to CoA via catheterization. Rabbits (n = 55) were anesthetized for measurement of the BP above the coarctation prior to tissue harvest using high-fidelity pressure transducers (Harvard Apparatus, Holliston, MA, USA) attached to 0.86 mm ID noncompliant fluid-filled catheters inserted into the right common carotid artery and advanced to the aortic arch. Systolic blood pressure (SBP: black circles), mean arterial pressure (MAP), and diastolic blood pressure (DBP: open circles) were then quantified from the BP waveforms. Linear regression was then used to mathematically model MAP (dashed line) as a function of percent area obstruction (formula).

• 

  Figure 5. Percentage area obstruction vs. days after CoA surgery. CoA morphology was evaluated on a weekly basis noninvasively using ultrasound imaging. Proximal aorta and CoA diameters were quantified in triplicate, and the average value was plotted (scatter) at each date. The percentage area obstruction (%AO = 100 × (Dprox2 − DCoA2)/Dprox2) was, therefore, quantified and plotted vs. days after surgery. For the dissolving sutures (i.e., short and long duration CoA groups), a plateau followed by exponential decay function was fitted to the %AO, while, for the permanent suture (i.e., prolonged CoA group), an exponential function was used to fit the data.

• 

  Figure 6. Thickness evolution for CoA rabbits (scatter) and G&R model predictions (lines). Coarctation-induced arterial thickening (shades of gray) was quantified from Doppler ultrasound b-mode images. To compare with control group, thicknesses were normalized to body weight (black). CoA groups included all combinations of severity (mild, intermediate, and severe) and duration (short, long, and prolonged) studied. G&R model constitutive parameters were tuned to fit empirical measurements (n = 2/group) with adjusted R2 characterizing goodness of fit.

• 

  Figure 7. Empirical stress–stretch curves for CoA (gray) vs. control (black) groups and G&R model prediction (red). Error bars representing the SEM were quantified and averaged at each strain energy level discretized over the range of 0 to 3 MPa. The constitutive material parameters identified are listed in Table 2.

• 

  Figure 8. Comparative representation of active contraction response to PE in CoA vs. control. Dose response curves (myography) by increasing half-log dose of phenylephrine (PE: administered every 5 min). The measured contractile force was normalized to maximum K+ contraction achieved for each segment. The nine CoA groups consist of all combinations of severity (mild, intermediate, and severe: different gray levels) and duration (short, long, and prolonged: different line styles) studied, compared to the control group (black) in each plot (A–C,E–G,I–K). Bar plots (D,H,L–O) represent the area under each dose–response curve (AUC). Error bars represent the SEM. * Significant difference compared to the control group in a two-tailed Student’s t-test with a 5% significance level, assuming similar variation among the groups.

• 

  Figure 9. Comparative representation of CoA vs. control active relaxation response to ACh. Dose–response curves (myography) by increasing the half-log dose of acetylcholine (ACh: administered every 5 min). Relaxation was measured relative to the initial contraction (EC50 concentration). The nine CoA groups consisted of all combinations of severity (mild, intermediate, and severe: different gray levels) and duration (short, long, and prolonged: different line styles) studied and compared to the control group (black) in each plot (A–C,E–G,I–K). Bar plots (D,H,L–O) represent area under each dose–response curve (AUC). Error bars represent SEM. * Significant difference compared to the control group in a two-tailed Student’s t-test with 5% significance level assuming similar variation among the groups.

• 

  Figure 10. Coarctation-induced active vascular response normalized to the control group. Area under the active dose–response curves (AUC) for each CoA group normalized to that of the control. Dose–response curves were quantified via wire myography and a half-log increase in phenylephrine (PE, gray) and acetylcholine (Ach, pink) to characterize NO active response in SM contraction and relaxation, respectively. A similar index was also quantified by taking the ratio of active stress per unit relaxed SM for CoA relative to control groups using the G&R framework (green) for two rabbits per group. G&R constitutive parameters were tuned to achieve minimum error to empirically quantified active response, i.e., PE and Ach. No significant differences were observed between results from empirical and G&R through two-tailed Student’s t-test with a 5% significance level.

• 

  Figure 11. G&R model prediction (A) and prediction error (B) relative to the measured impaired vascular response. (A) Contour plot showing impairment index quantified by taking the ratio of active stress per unit relaxed SM for CoA relative to control using the G&R framework. Contour lines indicate regions of constant impairment indices. CoA severity, shown on the vertical axis, is characterized through peak-to-peak trans-coarctation BP gradient (BPG), whereas duration is represented on a logarithmic scale of age in days. (B) Model error relative to measured impairment index, i.e., area under the active dose–response curves (AUC) for each CoA group normalized to that of the control. Dose–response curves were quantified via wire myography and half-log increase of phenylephrine (PE, gray) and acetylcholine (Ach, pink) to characterize NO active response in SM contraction and relaxation, respectively. Model error was then quantified as a percentage relative to the empirical PE and Ach impairment indices. Rabbits used in the tuning process were excluded for model prediction error.