Optical coherence tomography findings of balloon angioplasty/stenting for in-stent restenosis after carotid artery stenting

Abstract

Objectives

The optimal therapeutic methods for in-stent restenosis (ISR) after carotid artery stenting (CAS) remains controversial. This study aimed to use optical coherence tomography (OCT) to evaluate the in-stent architectures during endovascular angioplasty/stenting for ISR.

Materials and Methods

Six lesions of ISR after CAS were evaluated by OCT during endovascular angioplasty/stenting.

Results

In one lesion, the OCT system could not be crossed because of elongation distal to the ISR lesion. In five lesions, pre-procedural OCT clearly revealed neointimal hyperplasia or neoatherosclerosis. The mean in-stent area stenosis was 84%. After regular balloon angioplasty, tissue compression and dissection of various sizes and layers were detected. After balloon angioplasty (with a mean balloon size of 5.4 mm), the minimum lumen area (from 1.7 ± 0.6 to 11.4 ± 5.3 mm², p < 0.01) and the minimum in-stent area (12.7 ± 2.6 to 21.8 ± 5.0 mm², p < 0.01) showed a significant increase. Additional stent was placed in one lesion that developed into a flap by dissection after balloon angioplasty. In another lesion in which sufficient dilatation was not achieved by balloon angioplasty, a major stroke occurred by acute occlusion of the ISR lesion 10 months later.

Conclusions

OCT can detect the in-stent architecture of ISR lesions after balloon angioplasty and additional stent placement. However, which dissection should be treated by additional stent remain problematic.

Introduction

Carotid artery stenosis is a major cause of atheroma-stroke. Carotid artery stenting (CAS) is a recognized safe treatment option for preventing stroke recurrence.^1,2 However, in-stent restenosis (ISR) after CAS remains a problem during the chronic phase and stroke may recur. ³ In endovascular treatment for ISR after CAS, balloon dilatation, cutting and scoring balloon, drug-eluting balloon dilatation, and stent placement have been performed.^4–6 As yet, the optimal therapy remains unsolved. ³ Conventional angiography and intra-vascular ultrasound (IVUS) have been used during balloon angioplasty/stent for ISR after CAS. However, the resolution of these is inferior to that of optical coherence tomography (OCT). OCT is an intravascular modality used for coronary interventions to detect detailed microstructural information regarding atherosclerotic plaques and placed stents, and has a high resolution capacity of 10 mm.^7,8 It provides more detailed information on the presence of tissue prolapse between the stent struts, thrombus formation and dissection IVUS. ⁹ It has used in carotid arteries, and its effectiveness and limitations have been elucidated. ¹⁰ In this study, five lesions of restenosis after CAS were evaluated using OCT during balloon angioplasty/stenting.

Materials and methods

Patients

Between January 2015 and December 2020 (6 years), 321 CAS procedures were performed at our hospital. A total of 252 lesions were followed up by imaging (ultrasound examination, CTA) for at least over 1 year. Of these, seven (2.8%) ISR lesions required retreatment by endovascular procedures. Three patients presented with minor strokes, while four patients were asymptomatic. Endovascular retreatment was performed for all seven lesions. Written informed consent was obtained from all patients. This study was approved by the Ethics Committee of our hospital.

CAS procedure and OCT technique

Dual oral antiplatelet agents were administered for at least five days before CAS. CAS was performed under local anaesthesia. Systemic anticoagulation was achieved by administering heparin to maintain an activated clotting time of at least 250 s An 8F Optimo balloon guiding catheter (BGC) (Tokai Medical Products, Aichi, Japan) was navigated to the common carotid artery (CCA) from the femoral artery, and a distal filter protection device (FilterWire EZ; Stryker, Fremont, CA, USA), or Spider FX; Medtronic, Minneapolis, MN, USA) was used to cross the ISR lesions. Dragonfly (St Jude Medical, St Paul, MN, USA) optical fibre used to investigate the frequency domain of the OCT system was encapsulated within the guidewire of the distal filter. Pullbacks were started while the CCA was occluded by BGC injecting 20 ml of 50% saline diluted contrast medium from BGC to completely replace the blood from the artery. Balloon dilatation was performed using a 4.5–6.0 mm Shiden balloon catheter (Kaneka Medics Corp., Osaka, Japan). Slow inflation (1 atm/30 s) to nominal pressure (8 atm) was maintained for 1 min, and deflated slowly (1 atm/30 s). After balloon dilatation, the lesions were evaluated by OCT. In one lesion, a stent was deployed because of flap formation due to dissection. Carotid ultrasonography or CT angiography was used to evaluate restenosis. Dural antiplatelets therapy was continued for at least three months. Cutting/scoring balloons and drug-eluting balloons were not used for any lesions.

Assessment of OCT images

OCT data were analyzed using the available OCT systems (Illumien Optis Imaging System, Review Software, V.C. 0.2; St Jude Medical). Definitions of lesion morphology such as lipid, ulceration, thrombus, or neovascularization were determined based on previous coronary OCT studies.^7,8 Homogeneous neointima has uniform optical properties and does not show focal variations in backscattering pattern. Heterogeneous neointima has focally changing optical properties and shows various backscattering patterns. ¹¹ Minimum lumen area and in-stent lumen area were calculated by two-dimensional cross-section images. Lumen area was automatically traced and supplemented by manual correction. ¹²

Statistical analysis

Lumen areas are expressed as mean ± standard deviation. Comparisons were made using the t-test for paired samples, and a p-value < 0.05 was considered statistically significant.

Results

OCT procedures and clinical results

Characteristics of the lesions and OCT findings are summarized in Table 1. OCT was used in six lesions. However, in one lesion, the OCT system could not be crossed because of elongation distal to the ISR lesion. Instead, IVUS was used. OCT was not performed in one patient with renal dysfunction. Hence, OCT examination was performed for five lesions. No complications occurred during the OCT and balloon angioplasty/stent procedures. No clinical complications were observed in any patient.

Table 1.

Patient characteristics, OCT findings and clinical course.

Case number	Risk of ISR	Initial placed stent	Initial maximum balloon diameter (mm)	Time of retreatment after initial CAS (months)	OCT findings before balloon dilatation			Balloon diameter (mm)	OCT findings after balloon dilatation		Additional stent	Follow-up (months)
					Min. area (mm3)	In-stent area stenosis (%)			Minimum area (mm3)
1	HT	Precise	4	21	1.3	94	Homogeneous	5	12.8	Surface dissection, deep dissection, tissue compression	None	34
2	HT, DM	Precise	3.5	60	0.8	92	Heterogeneous, ulceration, neovascularization	4.5	10	Flap formation by dissection, tissue compression	Precise	58
3	DM, CAS after CEA-restenosis	Precise	5.5	81	1.2	57	Heterogeneous, irregular surface	5.5	2.2	Insufficient dilatation, surface dissection	None	10, major stroke
4	HT, HL	Caroridwall	4.5	14	2.5	86	Homogeneous	6	16	Tissue compression	None	36
5	HT, DM, Smoking	Caroridwall	4.5	33	1.8	89	Homogeneous	6	15.4	Tissue compression	None	60

OCT: optical coherence tomography; ISR: in-stent restenosis; HT: hypertension; DM: diabetes mellitus; HL: hyperlipidemia.

OCT findings

Representative lesions are shown in Figures 1 to 3. OCT examination before balloon angioplasty revealed homogeneous high-intensity tissue in three lesions (Figure 1(C)), and heterogeneous tissue with signal attenuation in two lesions (Figures 2(C) and 3(C)).

Figure 1.

(A) Pre-preprocedural angiography shows in-stent restenosis after CAS. The arrow indicates the corresponding location of panel C. (B) Longitudinal image of OCT. (C) Axial image of OCT revealed uniform signal-rich band without signal attenuation. (D) Angiography after balloon angioplasty shows good dilatation. The arrow indicates the corresponding location of panel F. (E) Longitudinal image of OCT. (F) Axial image of OCT after balloon angioplasty, compression of the tissue, dilatation of the previous stent, surface dissection (*), and dissection in the inner-stent neointimal layer (**) were detected.

Figure 3.

(A) Pre-preprocedural angiography shows in-stent restenosis (ISR) after CAS. The arrow indicates the corresponding location of panel C. (B) Longitudinal image of OCT. (C) Axial image of OCT revealed irregular surface covered by signal-rich band with backscattering, and heterogeneous region. Stent struts are partly invisible. (D) X-ray of the balloon angioplasty shows insufficient dilatation. (E) Angiography after balloon angioplasty showed insufficient dilatation. The arrow indicates the corresponding location of panel G. (F) Longitudinal image of OCT. (G) Axial image of OCT shows slight compression of the tissue, surface dissections, and insufficient lumen gain with a minimum diameter of 1.23 mm. (H) Emergent angiography 10 months after the procedure shows severe ISR (arrow).

Figure 2.

(A) Pre-preprocedural angiography shows in-stent restenosis after CAS. The arrow indicates the corresponding location of panel C. (B) Longitudinal image of OCT. (C) Axial image of OCT reveals high intensity band with ulceration and neovascularizations in the neointimal layer (*). Signal-intensity poor region also exists in the deep layer. Stent struts are partly invisible. (D) Angiography after balloon angioplasty showed good dilatation. The arrow indicates the corresponding location of panel F. (E) Longitudinal image of OCT. (F) Axial image of OCT, tissues are compressed and flap formation with dissection between the neointima and stent (*). (G) Angiography after stent placement. The arrow indicates the corresponding location of panel I. (H) Longitudinal image of OCT. (I) Axial image of OCT after stent placement shows good recanalization with small tissue prolapses between the stent struts (*).

By measurement of minimum lumen area and in-stent lumen area (Figure 4), the mean in-stent area stenosis was 84%. After balloon angioplasty (with a mean balloon size of 5.4 mm), the minimum lumen area (from 1.7 ± 0.6 to 11.4 ± 5.3 mm², p < 0.01) and the minimum in-stent area (12.7 ± 2.6 to 21.8 ± 5.0 mm², p < 0.01) showed a significant increase (Figure 5). After balloon angioplasty, a minimum luminal area of over 10 mm² was obtained in four lesions; however, in one lesion (case 3), only a 2.2 mm² luminal area was obtained, regardless of the 5.5-mm balloon dilatation (Figure 3(G)). Tissue compression was detected in all lesions (Figures 1(F), 2(F), and 3(G)). In-stent plaque volume is decreased from 11 ± 2.4 to 10.4 ± 2.6 mm² (p = 0.30, statistically, not significant). Surface dissection was detected in three lesions. Interestingly, in case 1, dissections were detected in the layer just under the stent lumen, without dissection of the neointimal surface (Figure 1(F)). In case 2, an additional stent was placed due to flap formation by surface dissection (Figure 2(F) and (I)).

Figure 4.

Assessment method of lumen area on optical coherence tomography (OCT) images in case 1. The actual measurements of flow area (green dotted circles) and in-stent lumen area (white dotted circles) using the dedicated software in the sample cross-section image of the OCT pullback. In-stent plaque volume was calculated as the in-stent area minus the flow area. (A), (B) Pre-procedural OCT image. In-stent plaque volume was 13.6 mm². (C), (D) OCT image after balloon-angioplasty. In-stent plaque volume was 14.9 mm².

Figure 5.

Changes in lumen area and plaque volume before and after balloon angioplasty.

Follow-up

At 34–60 months, four lesions did not present with restenosis. However, in case 3, major cerebral infarction occurred due to severe ISR 10 months after the procedure (Figure 3(H)).

Discussion

In this study, OCT findings of ISR after CAS were analyzed during retreatment in five lesions. In-stent neointimal hyperplasia or neoatherosclerosis before balloon angioplasty, and tissue compression and surface/deep dissection after balloon angioplasty were clearly detected. Reports of the OCT findings of ISR after CAS were rare; Matsumoto et al. reported in a case report that ISR 10-years after CAS were observed by OCT, that demonstrated heterogeneous intima and thin fibrous cap with lipid-rich component, and plaque prolapse after additional stent. ¹³ Liu et al. also reported in a case report that ISR with a fibrotic neointimal growth, intimal disruption and artery dissection after balloon angioplasty, and well apposed stent with tissue protrusions after stent placement. ¹⁴

In-stent neointimal hyperplasia is commonly seen in coronary ISR caused by biological mechanism. ¹⁵ On the other hand, OCT findings after balloon angioplasty in coronary ISR were rare; Alfonso et al. reported that in-stent tissue compression and deep dissections were clearly visualized by OCT after balloon angioplasty after ISR after coronary drug-eluting stent (DES). ¹⁶ Treatment algorithms for ISR after coronary arterial DES placement are proposed based on OCT findings. Homogeneous pattern reflects neointimal hyperplasia, and balloon angioplasty including high-pressure, scoring, and drug-coated balloons, repeated DES placement, and vascular brachy-therapy are recommended. Heterogeneous pattern reflects neoatherosclerosis, and drug-coated balloon and repeated DES placement are recommended ¹⁵ (Shlofmitz).

Angioplasty/stenting for ISR after CAS differs from coronary artery interventions in the following ways. A self-expandable stent is used for carotid lesions, whereas a balloon-expandable stent is used for coronary lesions, a bare stent is used for carotid lesions, a drug-eluted stent is used for coronary lesions, and, a larger balloon size is used in carotid lesions than in coronary lesions. After bare metal stenting, the neointimal tissue was comprised of vascular smooth muscle cells and an extracellular matrix with a diffuse pattern.^15,17 In the field of coronary DES, Waksman In-Stent Restenosis Classification is commonly used for coronary DESs as follows. Type I: stent-related mechanical causes including underexpansion (Type IA) and stent fracture (Type IB), and Type II: biologic causes of neointimal hyperplasia (Type IIA), neoatherosclerosis, non-calcified (Type IIB), and neoatherosclerosis calcified (Type IIC). ¹⁵ In this study, three lesions were Type IA (Figure 1(C)), and two lesions were Type 2B (Figures 2(C) and 3(C)). Therapeutic guidance for coronary ISR is recommended depending on the type such as excimer laser atherectomy, rotational atherectomy, and vascular brachytherapy compared with carotid ISR.^15,18 Drug-coated and cutting/scoring balloons have been used for carotid ISR.^3–6 Regular balloons were used in this study.

OCT can detect the vascular architectures at a high resolution. Conventional methods, such as angiography, 3D angiography, and IVUS, cannot detect smaller dissections than OCT. In this study, the dissection of the superficial and deep layers was clearly detected. In case 1, the dissection was isolated in the inner-stent neointimal layer (Figure 1(C)). In case 2, the dissection spread from the surface to the inner-stent layer (Figure 2(C)), whereas in case 3, the dissection spread only to the superficial layer (Figure 3(C)). An additional stent was placed for one lesion (case 2) observing flap formation with dissection because it caused thrombosis, migration of the dissected flap, and progression of the dissection. We believe that an additional stent is not required for small dissections localized in the superficial layer without flap formation. The optimal treatment for ISR following CAS remains unclear. ³ OCT findings may help to determine which lesions should be placed in an additional stent.

However, OCT for carotid lesions has some limitations. ¹⁹ First, OCT evaluation was difficult in elongated lesions because OCT system cannot reach the elongated ICA distal to the stenotic lesion. This occurred in one lesion. Second, the penetration depth for OCT is 5–6 mm. Therefore, if the guidewire is biased to one side, deep layers outside the previous stent would not be observed (Figure 3(C)). In this series, three CCA lesions were excluded as a result. Third, previous stent struts were difficult to visualize for ISR after CAS due to OCT-specific signal attenuation. In this series, the previous stent struts were not visible in two lesions (Figures 2(C) and 3(C)). Fourth, the real-time motion of plaque cannot be detected by OCT. Therefore, the mobility of the large dissection in case 3 was not identified (Figure 3(F)).

Conclusion

OCT can detect ISR in-stent architecture during CAS. OCT findings may contribute to optimizing therapeutic options for ISR after CAS.