To patch or not to patch: is that the real question? The role of hemodynamics in carotid endarterectomy. Illustrative cases

Abstract

BACKGROUND

The utilizationof patches in carotid endarterectomy (CEA) for carotid artery stenosis remains controversial, with conflicting evidence regarding postoperative outcomes. This report accentuates this discourse with two selected representative cases with divergent outcomes.

OBSERVATIONS

Computational fluid dynamics analyses of pre- and post-CEA hemodynamics revealed distinct hemodynamic profiles between the two patients. In the nonpatched internal carotid artery (ICA), the vessel retained a cylindrical shape, exhibiting swirling blood flow and higher wall shear stress (WSS)—patterns typical of healthy vasculature. The patched ICA adopted a bulbous shape, akin to the anatomical carotid bulb, and displayed lower WSS and noncoherent disturbed blood flow, which are features associated with atherosclerosis, endothelial dysfunction, and cellular damage.

LESSONS

This study suggests that the question may not be “To patch or not to patch?” but rather “Is the restoration of the anatomical bulb shape beneficial or deleterious?” It sheds light on the hemodynamic implications of this procedure and provides insight into the ongoing debate surrounding CEA. Using a patch might not necessarily result in improved flow or more favorable outcomes; thus, restoration of the carotid bulb configuration postendarterectomy might not optimize the hemodynamic profile for patients, but rather, a simple tubular shape, without a patch, might offer the best solution.

https://thejns.org/doi/10.3171/CASE24840

Carotid artery stenosis (CAS) is a prevalent vascular condition, predisposing individuals to an increased risk of ischemic stroke with potential debilitating outcomes.¹ Among the various surgical interventions available for managing CAS, carotid endarterectomy (CEA) is a cornerstone procedure.² The use of patches during arterial wall closure after plaque removal remains a topic of debate. Clinical practice varies and is often influenced by individual surgeon preferences: while some surgeons advocate for routine patching, citing evidence of reduced postoperative stroke and restenosis rates, others still prefer primary closure.

From a hemodynamic standpoint, factors such as wall shear stress (WSS) and blood flow patterns are critical in maintaining vascular integrity. Deviations in these parameters can lead to endothelial dysfunction, promote atherosclerosis, and increase the risk of restenosis.³,⁴ Specifically, low WSS (below 4 dyne/cm²) and high oscillatory shear index (OSI) are associated with proatherogenic endothelial phenotypes that contribute to atherosclerosis and restenosis, while WSS levels above 15 dyne/cm² are considered atheroprotective.^5–7 The carotid bulb, a region prone to low WSS and high OSI due to its geometry, plays a pivotal role in the progression of vascular disease.⁸,⁹ Patch closure in CEA, which usually results in restoring the carotid bulb’s original geometry, often results in a more tapered and tortuous shape than primary closure, whereas CEA closure without a patch leads to a simple cylindrical construct.

In this study, we hypothesized that maintaining a more tubular carotid bulb geometry, rather than restoring its natural irregular shape, can reduce the risk of restenosis. We investigated the hemodynamic differences between patched and unpatched CEA procedures using patient-specific flow data and advanced computational fluid dynamics (CFD) to provide insight into how patch use impacts blood flow patterns in the carotid artery.

Illustrative Cases

Study Design

This retrospective study was conducted on two patients with CAS treated at our institution. The two subjects were selected for this pilot study based on the completeness of their anatomical and hemodynamic information, and the divergence of their outcomes: one who underwent CEA with a patch and experienced a stroke 8 years later due to carotid artery restenosis (patient 1), and the other who underwent CEA without a patch and did not develop restenosis on follow-up (patient 2). Specifically, patient 2 underwent annual carotid ultrasound surveillance since the procedure was performed, with a total follow-up duration of 7 years, demonstrating sustained carotid patency without evidence of restenosis. Clinical data, including demographics and atherosclerosis risk factors (e.g., hypertension, diabetes mellitus, and smoking), were recorded. Image quality was assessed by consensus, focusing on vessel wall visualization and venous contamination. Images with poor quality, such as those with ill-defined margins or no separation between the carotid artery and jugular vein, were excluded. Discrepancies in image quality assessment were resolved by consensus. The Strengthening the Reporting of Observational Studies in Epidemiology reporting guidelines were used.

Imaging Reconstruction and CFD Analysis

Contrast-enhanced carotid MR angiography (MRA) was performed in the patients 2 months after the CEA procedure. The carotid arteries were segmented with semiautomatic software (Amira, version 6.2.0 NVIDIA 5370.70, Thermo Fisher Scientific). The reconstructed vascular surface models were smoothed, cleaned, and truncated perpendicularly to their axes using in-house software. These patient-specific models were then meshed with unstructured tetrahedral cells using an advancing-front in-house grid generator.

To calculate the geometric metrics, centerline analysis of patient-specific surface models is used. Each point of the centerline also provides the inner and outer radius of the vessel. The centerline is considered from the base of the bulb at the bifurcation to 5 cm distal into the internal carotid artery (ICA). This allows for the calculation of maximum tortuosity (path distance over Euclidean distance), mean tapering (proximal radius over distal radius), mean eccentricity (outer radius over inner radius), and mean radius (average of inner and outer radii).

To assess the impact of the procedure on local hemodynamics, CFD simulations were performed using a finite element in-house Navier-Stokes solver. Flow rate waveforms for the common carotid artery (CCA) and ICA were obtained post-CEA using the commercially available Non-Invasive Optimal Vessel Analysis software (VasSol, Inc.). The protocol utilized standard 3D time-of-flight MRA of the cerebral vasculature, as described by Zhao et al.¹⁰,¹¹ Assuming rigid walls, the mean flow rate in the external carotid artery (ECA) was calculated and used to scale the ICA waveform, creating a representative ECA flow rate waveform. A Wormsley profile was then applied to the CCA inlet and ECA outlet to accurately represent physiological pulsatility and flow contours. The ICA outlet was modeled as a traction-free boundary condition, and blood was assumed to be incompressible and Newtonian.

Blood flow simulations were conducted for the treated carotid bifurcation in each patient, including the restenosed bulb in the patient with patching (patient 1). Transient simulations were performed with a time step of 0.01 second, covering one cardiac cycle, with data collection during the second cardiac cycle to capture stabilized flow dynamics.

Data Analysis

The analysis focused on comparing area-averaged WSS, OSI, area-averaged vorticity, and swirling flow (assessed by vortex coreline length and streamlines) between the two patients. Vorticity refers to the rate of blood flow rotation, while swirling flow indicates higher WSS and promotes blood mixing, which enhances oxygen and nutrient delivery to the vessel wall.¹² A tyhigh vorticity with a swirling flow pattern suggests healthy circulation. This analysis aimed to explore the hemodynamic impact of patch use in CEA and its potential association with the risk of restenosis. Given the limited sample size, no statistical analysis was performed.

Results

The hemodynamic results and geometric characteristics of the carotid bulbs for both patients are summarized in Table 1, highlighting key differences between the reconstruction techniques. Postoperative angiograms performed directly post-CEA clearly demonstrate that the carotid bulb in patient 1, who underwent CEA with a patch, was wider, more tortuous, and more tapered compared with the bulb in patient 2, who had CEA without a patch (Fig. 1). Additionally, patient 1’s bulb displayed increased eccentricity, resulting in a more irregular shape, while patient 2’s primary closure produced a smoother and more uniform bulb.

TABLE 1.

Geometric and hemodynamic characteristics after CEA of patients 1 and 2

	Parameter	Patient 1 (w/ patch)	Patient 2 (w/o patch)
Geometry	Tortuosity	1.31	1.15
	Tapering	3.24	3.00
	Eccentricity	1.42	1.26
	Radius (mm)	3.4	2.2
Hemodynamics	WSS (dyne/cm2)	5.99	29.5
	OSI	0.045	0.077
	Vorticity (1/sec)	48.8	277
	Coreline length (cm)	2.1	3.2

FIG. 1.

Digital subtraction angiograms showing post-CEA results of patient 1 (A) with a patch and patient 2 (B) without a patch.

In patient 1, who underwent CEA with a patch, the bulb was exposed to low WSS. The area-averaged and time-averaged WSS in the right bulb was 5.99 dyne/cm² (Fig. 2A). In contrast, in patient 2, who did not receive a patch, the WSS in the right bulb was significantly higher, measuring 29.5 dyne/cm² and remaining at physiological levels (Fig. 2B). The OSI for patient 1 was 0.045, whereas patient 2 had a higher OSI value of 0.077.

FIG. 2.

WSS distribution in two carotid artery CFD models. A: Post-CEA model with a patch demonstrating lower WSS across the carotid bifurcation. B: Post-CEA model without a patch showing regions of significantly elevated WSS, indicating altered hemodynamics.

Vorticity, a measure of rotational blood flow, was lower in patient 1’s bulb, with an area-averaged and time-averaged vorticity of 48.8 1/sec, compared with 277 1/sec in patient 2. Similarly, the time-averaged vortex coreline length, representing swirling flow, was shorter in patient 1, measuring 2.1 cm, compared with 3.2 cm in patient 2 (Fig. 3A and B). Not only were patient 1’s corelines shorter, but the surrounding streamlines shown in Fig. 3 also demonstrated less swirling. By contrast, patient 2 exhibited strong vortex pairs throughout the carotid bulb (Fig. 3C and D).

FIG. 3.

Time-averaged velocity streamlines and vortex corelines in two carotid artery CFD models. A and C: Flow patterns in patient 1, showing shorter vortex corelines (2.1 cm in length) with less swirling. B and D: Flow patterns in patient 2, highlighting longer vortex corelines (3.2 cm in length) and more prominent swirling patterns, with strong vortex pairs forming throughout the carotid bulb.

It is worth mentioning that, in patient 1, the region between the carotid bulb formed by the patch and the distal healthy artery demonstrated a high gradient of shear stress, posing a risk of potential cell damage, coinciding with the specific site of restenosis on follow-up (Fig. 4). These findings underscore the significant differences in hemodynamic profiles between the two patients, suggesting that patch use in CEA may lead to altered flow dynamics that could have implications for restenosis risk. The CFD models visualizing these flow patterns are available in Video 1.

FIG. 4.

Patient 1. Digital subtraction angiogram at the 8-year follow-up showing restenosis of the carotid artery.

Video 1. Video showing the CFD models of patient 1 (with patch) and patient 2 (without patch), illustrating the differences in WSS, time-averaged velocity streamlines, and vortex corelines, respectively. Click here to view.

Informed Consent

The necessary informed consent was obtained in this study.

Discussion

Observations

The findings of this study provide valuable insights into the hemodynamic consequences of patch use during CEA and its potential role in carotid artery restenosis. It underscores the importance of understanding the interplay between arterial geometry, WSS, and blood flow patterns in determining surgical outcomes. The European Society for Vascular Surgery guidelines recommend patch angioplasty or eversion endarterectomy over primary closure, with similar recommendations from the Society for Vascular Surgery in the United States.^2,13 However, the literature presents significant variability, with some studies demonstrating that patching reduced the risk of restenosis compared with primary closure, while others indicate that both techniques are equally safe and effective in preventing stroke.^14–17 This discrepancy is mirrored clinically, as the choice of technique varies widely among surgeons, including both neurosurgeons and vascular surgeons, and is frequently left to their discretion, leading to considerable variability in clinical practice. Our pilot study raises critical questions about whether patching always achieves the desired hemodynamic outcomes, particularly in the context of restoring carotid bulb geometry.

The carotid bulb, with its unique anatomical structure, plays a vital role in normal physiological conditions by regulating blood pressure and heart rate through baroreceptor activity.¹⁸ It is also pivotal in local hemodynamics, particularly in creating regions of low WSS, high OSI, and flow recirculation, all of which are contributors to atherogenesis and plaque formation.⁸,^{19, 20} These atherogenic conditions often manifest at the posterior wall of the proximal ICA, where flow disturbance is more pronounced.⁵,^{9,21, 22} However, after CEA, the carotid bulb loses its physiological function due to denervation; therefore, restoration of its anatomical shape postoperatively does not inherently provide a clear benefit.^23–26 Instead, this reinstatement of the bulb can perpetuate the unfavorable flow conditions that initially contributed to plaque formation. Moreover, the geometry of the carotid bifurcation, such as a wider bifurcation angle, a higher ICA/CCA diameter ratio, and increased tortuosity, exacerbates these hemodynamic disturbances, leading to reduced velocities and asymmetrical flow profiles.^9,27 These geometric features, especially at the posterior ICA wall where low WSS persists, invite scrutiny of the practice of restoring the bulb’s original shape during CEA, particularly when this geometry inherently predisposes the region to atherosclerosis.

Our study’s findings align with existing literature that highlights the risks associated with low WSS in the carotid bulb. In the patient treated with a patch (patient 1), the CEA resulted in a bulb geometry that exacerbated this unfavorable hemodynamic condition. The patch broadened the carotid geometry, restoring a shape that, while anatomically correct, was suboptimal from a hemodynamic standpoint. The disturbed flow patterns, characterized by low WSS and the absence of coherent and persistent vortices, likely contributed to the restenosis observed 8 years postprocedure. These findings are consistent with those of Malek et al., who demonstrated that WSS values below 12 dyne/cm² are associated with regions prone to atherogenesis, supporting the hypothesis that the patch’s effect on bulb geometry played a role in patient 1’s restenosis.⁵

In contrast, the patient who underwent CEA without a patch (patient 2) exhibited a hemodynamic profile more indicative of healthy arterial function. The absence of a patch resulted in a more tubular geometry that maintained higher WSS, and a coherent swirling flow, which is generally protective for the endothelium. Higher WSS promotes stable endothelial cell function and reduces the risk of atherosclerosis, despite the slightly higher OSI observed. This outcome underscores the importance of maintaining a hemodynamically favorable geometry rather than simply restoring the original anatomical shape of the bulb, which may no longer serve any physiological purpose. Our concurrent study showed that the flow differences observed between CEA with and without a patch were mainly due to the geometric differences of the reconstructed bulbs and were not strongly affected by flow rates, flow waveforms, or vessel compliance (unpublished data).

These findings prompt a reconsideration of the role of patches in CEA. The current debate, which often centers on whether to use a patch, might be overlooking a more fundamental issue: the impact of restored carotid geometry on postoperative hemodynamics. The real question might not be whether to use a patch but whether to restore the bulb’s original shape or create a cylindrical conduit that promotes optimal flow dynamics. Patches are often necessary to ensure proper closure and prevent hypothetical complications such as vessel narrowing or rupture, but their impact on arterial geometry must be carefully considered. Likewise, and for a similar rationale, eversion CEAs preserve the geometry of the carotid bulb and bifurcation. Given that the carotid bulb is no longer functional post-CEA, its anatomical restoration could perpetuate the same low WSS and turbulent flow conditions that lead to plaque recurrence.^{28, 29}

Therefore, the critical question should not merely be whether to patch or avoid patching, but rather how to optimize the post-CEA geometry to achieve a streamlined conduit that promotes favorable hemodynamics. We advocate for a shift in focus within the surgical community: rather than continuing the debate on patch use, future research and clinical practice should prioritize achieving a cylindrical carotid geometry post-CEA to minimize adverse flow patterns and enhance patient outcomes.

Lessons

Patching, when the CEA repair would otherwise result in stenosis, is inherently logical and intuitively beneficial. However, following CEA, the carotid bulb is denervated and the reconstruction of its anatomical shape does not restore its physiological function. The insights from our study raise the concern that restoring the bulb geometry may not improve blood flow and may create hemodynamic conditions conducive to restenosis. Therefore, the question is not whether to patch or not, but rather, “should the carotid be reconfigured as a cylinder or as a bulb?” This work thereby redirects the ongoing debate about patch utilization in CEA, emphasizing the need for further investigations that consider a broader range of factors, particularly arterial geometry, in surgical decision-making. The ultimate goal should be to refine surgical techniques to optimize hemodynamics, minimize the risk of restenosis, and enhance overall patient outcomes. Larger, prospective studies are needed to test our hypothesis and guide the development of surgical strategies that are tailored to the unique hemodynamic needs of each patient.

This study has several limitations that should be considered when interpreting the results. The sample size is extremely limited, involving only two patients. While the detailed hemodynamic analysis provides valuable insights, the small cohort restricts the generalizability of the findings. Larger, more diverse patient populations would be necessary to extend and validate the results. Additionally, the retrospective nature of the study introduces potential biases, such as selection bias. Indeed, the choice of the two illustrative cases with polar outcomes was deliberate, with the aim of initiating further discussions. Furthermore, our study is based on the availability of high-quality imaging data and specific clinical outcomes, which might not fully represent the broader population of individuals undergoing CEA. This study assumes rigid arterial walls in the CFD simulations, which might not fully capture the dynamic behavior of blood vessels. While this assumption simplifies the modeling process, it might overlook the impact of arterial compliance and pulsatile blood flow on hemodynamics, affecting the accuracy of the results. Finally, the focus on carotid artery geometry and WSS does not account for other factors that might contribute to restenosis, such as genetic predisposition, patient-specific factors (e.g., medication use and compliance, comorbidities), endothelial function, surgical technique used, or type of patch material. While this study offers important insights into the hemodynamic effects of patch use in CEA, its findings should be interpreted with caution given the limitations mentioned above. Future studies involving larger cohorts are needed to confirm these preliminary observations.

Disclosures

Dr. Charbel reported consulting for Transonic Inc. outside the submitted work.

Author Contributions

Conception and design: Charbel, Abou-Mrad, Cebral. Acquisition of data: Charbel, Abou-Mrad. Analysis and interpretation of data: Charbel, Abou-Mrad, Marsh, Cebral. Drafting the article: Charbel, Abou-Mrad, McGuire, Marsh. Critically revising the article: all authors. Reviewed submitted version of manuscript: Charbel, Abou-Mrad, McGuire, Marsh. Approved the final version of the manuscript on behalf of all authors: Charbel. Statistical analysis: Charbel. Administrative/technical/material support: Charbel. Study supervision: Charbel, Cebral.

Supplemental Information

Videos

  Video 1. https://vimeo.com/1057543197.

Previous Presentations

This work was presented as a poster at the Society of NeuroInterventional Surgery (SNIS) 21st Annual Meeting & Fellows Course, Colorado Springs, CO, July 22–26, 2024.

Correspondence

Fady T. Charbel: University of Illinois Chicago, IL. fcharbel@uic.edu.