Effect of Hinge Motion on Stent Edge-Related Restenosis After Right Coronary Artery Treatment in the Current Drug-Eluting Stent Era

Abstract

Background: Stent edge-related restenosis (SER) remains a potential limitation of drug-eluting stent (DES). Hinge motion at the stent edge could lead to mechanical stress and contribute to incidents of SER. We investigated the effect of hinge motion on SER after implantation of current-generation DES in the right coronary artery (RCA), where excessive vessel movement is commonly observed.

Methods and Results: Of 647 consecutive lesions in the RCA treated with second-generation or later DESs, 426 with follow-up angiography were included in this study. Intravascular imaging analysis was performed for 584 stent edges and reference segments. Binary restenosis occurred in 42 lesions (9.9%), and 55% were SERs. The hinge angle was significantly larger in the SER group than in the other restenosis or the no-restenosis group (17.9° vs. 11.6° and 10.6°, respectively; P<0.001). Lesions with an excessive hinge angle (>11.5°) had an increased rate of target lesion revascularization (19.1% vs. 7.2%; P<0.001) during the median follow-up period of 1,578 days. In per-edge analysis, hinge angle and residual plaque burden were independent predictors of SER. The coexistence of excessive hinge motion and residual plaque burden had a synergistic effect on stenotic progression in quantitative angiographic analysis (P_interaction<0.001) at follow-up angiography.

Conclusions: Substantial stress determined by angulation at a stent edge and its interaction with residual plaque can be considered as one plausible mechanism for SER.

Drug-eluting stents (DESs) significantly reduce neointimal progression in stented segments. However, restenosis observed at a stent edge and/or its marginal reference, generally called stent edge-related restenosis (SER), has been a potential limitation of coronary stents.¹ An excessive edge vascular response, provoked by the mechanical interaction between the stent platform and marginal segment, could contribute to the occurrence of SER.² Stent placement changes the natural geometry of a vessel, especially in a tortuous coronary artery. In a previous study, the longitudinal straightening effect of a metallic stent was reported to be related to restenosis and adverse clinical outcome.³ A similar relationship between hinge motion and restenosis was investigated for first-generation DES.⁴

Editorial p 1969

More recent DESs have superior flexibility and conformability as a result of various modifications of their stent design.⁵ These improvements could translate into better clinical outcomes.^5,6 However, even in the current DES era, SER is a common restenosis pattern. In terms of vessel factors around a stent edge, residual plaque burden and lipid arc have been reported to be predictors for SER, either with bare-metal stents or DESs.^7,8 The vascular response to the mechanical stress imposed on a stent edge was assumed to differ according to local vascular properties.

The aim of the present study was to investigate the association between hinge motion at the stent edge and the incidence of SER in the right coronary artery (RCA) where vessel movement is commonly excessive. In addition, we performed intravascular imaging analysis to demonstrate the differential effects of hinge motion according to the local vascular properties and the consequent morphological changes of the implanted stent and vessel.

Methods

Study Population

From March 2009 to March 2019, 594 patients with 647 lesions in the RCA were treated with second- or newer-generation DESs at NTT Medical Center Tokyo. Follow-up coronary angiographies (CAGs) were performed for 426 lesions 6–18 months after the index procedure (Figure 1). Binary restenosis was defined by diameter stenosis >50% and classified as SER if it involved the 5-mm reference segments or the 5-mm in-stent segments adjacent to the stent borders, both distally and proximally. In addition, all available post-stent intravascular ultrasound (IVUS) and optical coherence tomography (OCT) data were analyzed. Consequently, 303 post-stent records (IVUS, n=249; OCT, n=54) were obtained. There were 22 inadequate records at stent edge segments (7 proximal edge segments, 15 distal edge segments) because of poor imaging quality. Therefore, 584 records at stent edge segments (IVUS, n=488; OCT n=96 records; 296 proximal edge segments, 288 distal edge segments) were included in the analysis.

Figure 1.

Flow of patients and lesions in the right coronary artery (RCA) treated with a drug-eluting stent (DES). CAG, coronary angiography.

To examine temporal changes, 39 paired records at follow-up (19 proximal edge segments, 20 distal edge segments) were analyzed. The clinical endpoints after the procedure were obtained from outpatient records for all 386 patients except for 5 cases where the hinge angles could not be obtained at both edges. The median follow-up period was 1,541 days (interquartile range [IQR] 636–2,446 days). The clinical endpoints included any target lesion revascularization (TLR), clinically driven TLR, and target-vessel myocardial infarction (MI).⁹ Detailed definitions of these clinical endpoints are provided in the Supplementary Methods.

This study complied with the Declaration of Helsinki regarding investigations in humans and was approved by the Ethics Committee of NTT Medical Center Tokyo. All patients provided written informed consent for percutaneous coronary intervention; informed consent for the present study was in the form of an opt-out option on the NTT Medical Center Tokyo website.

Quantitative CAG of Stent Edge Segments and Hinge Motion Measurement

Quantitative CAG was performed for the 5-mm reference segments and the 5-mm in-stent segments after stent placement and at follow-up using dedicated software (Q-angio; Medis, Leiden, Netherlands). The main angiographic parameters were reference vessel diameter, minimum lumen diameter, and percentage diameter stenosis (DS%).

Hinge motion was quantified by the difference in the angle between diastole and systole at each proximal or distal stent edge after stent placement (i.e., ∆angle=β−α (degrees); Figure 2).⁴ Two independent observers (T.J. and M.M.) measured the hinge angle using multiple projections. In the case of a disagreement between observers, a consensus reading was obtained. Angiographic views in which the hinge angle was maximal were selected for analysis. Hinge angle was also defined per each stent as the stronger hinge angle between the proximal and distal edge.

Figure 2.

Quantification of hinge motion. Representative case of stent edge-related restenosis (SER). Hinge movement was observed during (A) systole and (B) diastole in the post-stent angiogram. SER developed within the marginal reference of the distal stent edge (C, arrowhead). Arrows represent both the stent edges. The hinge angle was calculated as the difference in the vessel angle at the stent edge between that at end-systole (α) and that at end-diastole (β) (hinge angle=24° in this case).

Intravascular Imaging Analysis of Stent Edge Segments

All intravascular images were analyzed by 2 trained physicians (M.I. and M.Y.), who were blinded to information such as clinical presentation and lesion characteristics during analysis. Using a validated quantitative IVUS analysis system (echoPlaque; Indec Systems, Santa Clara, CA, USA), vessel, lumen, and stent area were manually traced at the leading edge of boundaries from the 5-mm reference segments and the 5-mm in-stent segments adjacent to both sides of the stent borders. Lesion characteristics, including plaque type, degree of calcification, edge dissection, and strut malapposition, were determined for each stent edge segment.¹⁰ Details of these imaging definitions are provided in the Supplementary Methods. Follow-up intravascular imaging records were also analyzed to evaluate temporal changes. For sensitivity analysis, the measurement data in OCT were multiplied by 1.04 (corrected OCT cohort) to adjust the values measured in OCT to IVUS, as per previous studies.^11,12

Evaluation of Pre-Stenting Vessel Factors Related to Subsequent Hinge Motion

To examine the association between pre-stenting vessel factors and post-stent hinge motion, 2 parameters were measured: (1) pre-stenting hinge angle, namely the ∆angle formed at the stent edge immediately before stent placement in an angiogram; and (2) vascular curvature index, which was the maximum angle formed at the stent edge position during a cardiac cycle. The pre-stenting hinge angle was considered to be an index of vessel hyperkinesis, and the vascular curvature index was considered an index of static vessel tortuosity. Of a total of 426 lesions, pre-stenting angiography was available for 327 to detect the exact location of the stent edge and to assess these parameters.

Statistical Analysis

Continuous values are expressed as the mean±SD or median with IQR and were compared using paired or unpaired Student’s t-tests or non-parametric tests for variables with a skewed deviation (Mann-Whitney or Kruskal-Wallis tests for multiple-group comparisons). Categorical variables are expressed as numbers and percentages and were compared using χ² statistics or Fisher’s exact test. Receiver operating characteristic (ROC) curve analysis was used to determine the best cut-off value of hinge angle for predicting binary restenosis, and “excessive hinge motion” was defined as any hinge angle over this value. Linear regression analysis was used to investigate the association between hinge angle and ∆DS% obtained by quantitative angiographic analysis, as well as associations between the pre-stenting vessel factors and the post-stent hinge angle. Multiple logistic regression was used to determine independent predictors of SER events. The following variables, considered as classic predictors of binary restenosis, were tested: hinge angle, vessel area, stent area, minimum lumen area (MLA) within the reference segment, residual plaque burden, calcification angle, and the presence of malapposition and edge dissection. The Kaplan-Meier method was used to estimate the effect of hinge motion on clinical endpoints, including TLR, clinically driven TLR, and target vessel MI. The rate of those clinical endpoints was compared between lesions with and without excessive hinge motion using the log-rank test.

Statistical analyses were performed using SPSS ver. 25 (IBM Corp., Armonk, NY, USA). Two-sided P<0.05 was considered significant.

Results

Study Population and the Incidence of Overall Restenosis

Angiographic binary restenosis was detected in 42 lesions (9.9%), with 55% occurring at stent edges (15 proximal edges, 8 distal edges). Baseline patient and lesion characteristics are provided in Table 1. The hinge angle was significantly larger in the SER group than in the group with other types of restenosis and the no-restenosis group (17.9±7.4° vs. 11.6±5.0° and 10.6±5.8°, respectively; P<0.001). The prevalence of known risk factors, including diabetes, hemodialysis, and hyperlipidemia, was higher in the SER than no-restenosis group.

Table 1. Comparison of Patient Backgrounds and Lesion Characteristics

	SER (n=23)	Other restenosis (n=19)	No restenosis (n=384)	P value*	P value†
Patient background
Male sex	19 (79)	17 (90)	316 (82)	0.33	0.63
Age (years)	69±9	68±6	68±10	0.94	0.90
Body mass index (kg/m2)	24.1±3.2	24.0±4.0	24.4±4.3	0.99	0.77
Diabetes	14 (61)	6 (32)	110 (29)	0.030	<0.001
Hemodialysis	3 (13)	5 (26)	7 (2)	0.28	0.001
Dyslipidemia	14 (61)	7 (37)	142 (37)	0.12	0.022
Hypertension	14 (61)	14 (74)	181 (47)	0.25	0.38
Previous MI	0 (0)	0 (0)	19 (5)	–	0.28
Previous CABG	0 (0)	0 (0)	8 (2)	–	>0.99
Current smoking	9 (39)	7 (37)	112 (29)	0.75	0.15
ACS	8 (35)	5 (26)	131 (34)	0.77	0.72
Post-procedural angiographic findings
Hinge angle (°)	17.9±7.4	11.6±5.0	10.6±5.8	0.009	<0.001
Lesion segment				0.23	0.009
Proximal	14 (61)	7 (37)	120 (31)
Mid	4 (17)	6 (32)	141 (37)
Distal	5 (22)	6 (32)	123 (32)
CTO	3 (13)	3 (16)	19 (5)	0.80	0.095
Ostium	4 (17)	0 (0)	21 (5)	0.056	0.044
Stent overlap	0 (0)	1 (5)	13 (3)
Lesion type				0.19	0.038
A	0 (0)	1 (5.3)	26 (7)
B1	2 (9)	0 (0)	123 (32)
B2	14 (61)	8 (42)	181 (47)
C	7 (30)	10 (53)	54 (14)
Calcification				0.70	<0.001
None-mild	8 (35)	5 (26)	262 (68)
Moderate	7 (3)	5 (26)	95 (25)
Severe	8 (35)	9 (47)	27 (7)
No. stents per lesion				0.093	0.92
1	19 (83)	10 (53)	326 (85)
2	3 (13)	5 (26)	40 (10)
3	1 (4)	4 (21)	18 (5)
Stent diameter (mm)	3.23±0.46	3.17±0.38	3.26±0.43	0.40	0.60
Total stent length (mm)	28±19	46.6±28	28.3±18.3	0.017	0.92
Stent to vessel ratio	1.11±0.15	1.07±0.18	1.11±0.17	0.39	0.98
Reference vessel diameter (mm)	3.14±0.49	3.20±0.51	3.18±0.48	0.44	0.55
MLD (mm)	2.82±0.53	2.86±0.50	2.87±0.49	0.65	0.64
DS% (%)	9.6±4.1	10.9±5.5	9.9±4.8	0.58	0.96
Follow-up angiographic findings
Reference vessel diameter (mm)	2.77±0.73	3.11±0.59	3.25±0.56	0.11	0.002
MLD (mm)	1.82±1.03	1.30±0.79	2.84±0.57	0.21	0.003
DS% (%)	34.4±28.1	58.7±22.8	12.6±7.2	0.009	0.001
Stent fracture	1 (4)	1 (5)	0 (0)	0.89	0.057

Unless indicated otherwise, data are given as the mean±SD or n (%). *P values for comparisons of the stent edge-related restenosis (SER) vs. other restenosis groups. ^†P values for comparisons of the SER vs. no-restenosis groups. ACS, acute coronary syndrome; CABG, coronary artery bypass grafting; CTO, chronic total occlusion; DS%, percentage diameter stenosis; MI, myocardial infarction; MLD, minimum lumen diameter.

Angiographic findings at the time of the procedure and at follow-up are summarized in Table 1. Severe calcification and complex lesions including chronic total occlusion were rarely observed in the no-restenosis group. The SER group had a higher prevalence of ostium lesions. The other restenosis group was characterized by longer total stent length with multiple stent implantation.

The background characteristics of lesions without follow-up CAG are summarized in Supplementary Table 1. Patients with these lesions were older (74±11 years) and had a lower prevalence of hyperlipidemia (29%).

Distribution of Hinge Angles

Hinge angle was quantified at each stent edge, and the distribution of hinge angles within RCA segments is shown in Figure 3. The mean hinge angle was 8.6±5.7°. Larger hinge angles were observed in proximal segments in the RCA than in mid and distal segments (9.9±6.2° vs. 8.7±5.7° and 7.1±4.9°, respectively; P<0.001). In addition, proximal edges had a larger hinge angle than distal edges in each stent (9.4±5.8° vs. 7.8±5.5°, respectively; P<0.001). No significant differences were seen among commercial stent designs (P=0.21) or the number of stents per lesion (P=0.18).

Figure 3.

Distribution of hinge angles at the stent edge according to (A) the position within each stent, (B) the segment in the right coronary artery (RCA), (C) the number of stents per lesion, and (D) the commercial design of the stent (“Other” stent types included Biofreedom [n=13] and Orsiro [n=2] stents). The hinge angle was significantly (P<0.001) larger at proximal than distal edges (A). In addition, hinge angle differed significantly (ANOVA, P<0.001) among edge positions in RCA segments (B). There was no significant difference in hinge angle for different numbers of stents (C; ANOVA, P=0.18) or between different types of commercially available stents (D; ANOVA, P=0.16).

Hinge Angles and Intravascular/Angiographic Profiles for SER and Non-SER Edges

The hinge angle and vascular profiles on post-stent intravascular imaging or angiography were compared between SER and non-SER edges (Table 2). The hinge angle was significantly larger in the SER than non-SER group (15.7±7.0° vs. 8.4±5.7°, respectively; P<0.001). There were no significant differences in reference vessel diameter, MLA, and DS% in the post-stent quantitative angiographic analysis.

Table 2. (A) Analysis of Lesion Characteristics at Post-Stent in Lesions With or Without SER: Comparison of Hinge Angle and Edge Vascular Variables, (B) Multivariate Logistic Regression Analysis for Factors Predicting Stent Edge-Related Restenosis

(A)	SER+ (n=15)	SER− (n=569)	P value
Angiographic findings
Hinge angle (°)	15.7±7.0	8.4±5.7	<0.001
Reference vessel diameter (mm)	3.48±0.64	3.16±1.01	0.30
MLD (mm)	3.07±0.73	2.88±0.99	0.58
DS% (%)	12.2±11.3	9.1±6.4	0.13
Intravascular imaging findings
Stent edge
Stent area (mm2)	9.3±2.8	8.9±2.8	0.66
Stent long diameter (mm)	3.58±0.57	3.55±0.59	0.82
Stent short diameter (mm)	3.19±0.53	3.14±0.53	0.73
Reference segment
Reference vessel area (mm2)	20.5±6.6	16.7±6.1	0.03
Reference minimum lumen area (mm2)	8.9±3.8	8.5±3.7	0.97
Residual plaque burden (%)	58±12	48±13	0.007
Calcification			0.54
0–90°	0 (0)	59 (10)
90–180°	2 (14)	53 (9)
180–270°	1 (7)	15 (3)
270–360°	0 (0)	11 (2)
Plaque type			0.19
Lipid	2 (13)	63 (11)
Fibrolipid	13 (87)	406 (71)
Fibrous	0 (0)	98 (17)
Malapposition	1 (6)	52 (9)	>0.99
Dissection	2 (12)	31 (5)	0.14
(B)	Odds ratio (95% CI)	P value
Hinge angle	1.17 (1.08–1.26)	<0.001
Residual plaque burden	1.08 (1.02–1.14)	0.005

Unless indicated otherwise, data are given as the mean±SD or n (%). CI, confidence interval; DS%, percentage diameter stenosis; SER, stent edge-related restenosis. Other abbreviations as in Table 1.

On post-stent intravascular images, the residual plaque burden was larger in SER than non-SER edges (58±125 vs. 48±13%, respectively; P=0.007), whereas there were no significant differences in stent area (9.3±2.8 vs. 8.9±2.8 mm²; P=0.66) or MLA (8.9±3.8 vs. 8.5±3.7 mm²; P=0.97). There were no significant differences in the prevalence of edge dissection (12% vs. 5%; P=0.14) or malapposition (6% vs. 9%; P>0.99) between SER and non-SER edges. Of note, no SER event occurred at edges where the residual plaque had been characterized as fibrous.

In multivariate logistic regression analysis, hinge angle (odds ratio [OR] 1.17 per degree; 95% confidence interval [CI] 1.08–1.26 per degree) and residual plaque burden (OR 1.08 per percentage point; 95% CI 1.02–1.14 per percentage point) were independent predictors of SER events. On ROC curve analysis for predicting SER events, the area under the curve for hinge angle was 0.80 (95% CI 0.69–0.92) with an optimal cut-off value of 11.5° (sensitivity 0.79, specificity 0.77; Figure 4).

Figure 4.

Receiver operating characteristic curve analysis for predicting stent edge restenosis. (A) A hinge angle at the stent edge of 11.5° and (B) a residual plaque burden of 54.2% were determined to be cut-off values for predicting stent edge restenosis. AUC, area under the curve; CI, confidence interval.

The same analyses were performed in the IVUS cohort (n=488) excluding the data obtained from OCT (n=96), and also in the IVUS plus corrected OCT cohort. These analyses demonstrated consistent results: post-stent hinge angles were considerably higher for SER than non-SER edges and the hinge angle and residual plaque burden were independent predictors for subsequent SER development (Supplementary Table 2). The intravascular imaging variables obtained by IVUS and OCT are shown separately in Supplementary Table 3.

Differential Effects of Hinge Motion According to Residual Plaque Burden

The effect of hinge angle on the incidence of SER and stenotic progression in angiograms differed according to the residual plaque burden (Figure 5). There was a substantial risk of restenosis and stenotic progression caused by the hinge angle when the stent edge was positioned with a plaque burden >50%, but not <50% (OR of edge restenosis 1.21 [95% CI 1.11–1.33] vs. 1.00 [95% CI 0.82–1.23] per degree; and ∆DS% OR 1.09 [95% CI 0.86–1.33] vs. 0.41 [95% CI 0.23–0.60] %/degree).

Figure 5.

Interaction effects between hinge motion and plaque characteristics. Forest plots represent the effects of the hinge angle on (A) the odds ratio of stent edge restenosis and (B) the percentage change in diameter stenosis between immediately after stent placement (post-stent) and follow-up. Vertical solid lines correspond to equal changes. Horizontal solid lines represent 95% confidence intervals (CIs).

Relationship Between Pre-Stenting Vessel Factors and Post-Stent Hinge Motion and Subsequent SER

To assess the effect of pre-stent vessel factors on post-stent hinge motion and the incidence of SER, the pre-stenting hinge angle and vascular curvature index were calculated in the angiogram immediately before stent placement. The pre-stenting hinge angle and vascular curvature index were positively associated with the post-stent hinge angle (R²=0.28 [P<0.001] and R²=0.13 [P<0.001], respectively; Supplementary Figure 1). Of note, in the ROC analysis for predicting the incidence of SER, the post-stent hinge angle had a larger area under curve (0.79; 95% CI 0.65–0.92) than the pre-stenting hinge angle (0.73; 95% CI 0.61–0.86) or vascular curvature index (0.69; 95% CI 0.56–0.83; Supplementary Figure 2).

Chronic Stent Recoil at Edges With Restenosis

Post-stent and follow-up intravascular images were compared for 38 paired records. Stent area decreased significantly at SER edges (from 9.2±2.6 to 8.1±2.6 mm²; P=0.039), with a decrease in the asymmetry index (from 0.88±0.09 to 0.79±0.16; P=0.007; Supplementary Figure 3), whereas no significant change was observed at the non-SER edges in stent area (from 8.3±3.2 to 8.5±3.2 mm²; P=0.67) or the asymmetry index (from 0.84±0.11 to 0.83±0.18; P=0.67). There was a significant decrease in reference vessel area in SER edges (from 19.4±5.7 vs. 16.8±5.2 mm²; P=0.028), but not in non-SER edges (from 20.6±6.9 vs. 19.2±6.9 mm²; P=0.23). In addition, hinge angle was associated with a significant decrease in stent area at edges (−0.10 mm²/degree; P<0.001) and with a trend for a decrease in reference vessel area (−0.12 mm²/degree; P=0.22).

Clinical Consequences of Hinge Motion

Excessive hinge motion was defined as a hinge angle >11.5°, corresponding to the optimal cut-off value in the ROC analysis. All lesions were classified as lesions with (n=157) or without (n=264) excessive hinge motion, except for 5 lesions in which the hinge angle could not be obtained at both edges. The median follow-up period for the lesions was 1,578 days (IQR 661–2,465 days).

The rate of any TLR (19.1% vs. 7.2%; P<0.001) and clinically driven TLR (7.6% vs. 3.4%; P=0.044) was significantly higher in lesions with excessive hinge motion (Figure 6). In addition, the rate of target vessel MI was higher in lesions with excessive hinge motion, although this difference was not statistically significant (2.5% vs. 0.8%; P=0.078). The same analysis was performed for all treated lesions, including those without follow-up CAG (n=639; median follow-up period 1,333 days [IQR 462–2,204 days]), which demonstrated consistent results (any TLR, 14.3% vs. 5.3% [P<0.001]; clinically driven TLR, 5.7% vs. 2.3% [P=0.030]; target vessel MI. 2.4% vs. 1.3% [P=0.33]).

Figure 6.

Kaplan-Meier analysis of the incidence of (A) any target lesion revascularization (TLR), (B) clinically driven TLR, and (C) target vessel myocardial infarction (MI) in lesions with follow-up coronary angiography (CAG) and with or without excessive hinge motion. There were significant increases in any TLR (19.1% vs. 7.2%; P<0.001) and clinically driven TLR (7.6% vs. 3.4%; P=0.044) in lesions with excessive hinge motion. In addition, the rate of target vessel MI was higher in lesions with excessive hinge motion, although the difference did not reach statistical significance (2.5% vs. 0.8%; P=0.078).

Discussion

The main findings of this study are that: (1) under the current clinical setting where all percutaneous coronary intervention procedures were performed using second-generation or newer DESs, SER accounted for more than 50% of the overall restenosis in the RCA; (2) the hinge angle was associated with the incidence of SER as well as stenotic progression in quantitative angiographic analysis; (3) the adverse effect of hinge angle was evident at edges with a >50% residual plaque burden after stent placement; and (4) the group with excessive hinge motion, defined as a hinge angle >11.5°, had a significantly higher rate of clinically driven TLR and any TLR.

Nature of Hinge Motion and Its Association With SER

Hinge motion after stent implantation could be caused by vessel movement combined with the difference in compliance between the stented and unstented segments. We focused on edge-related stenosis in the RCA because of its relatively excessive vessel motion, which has been reported to provoke a mechanical burden that manifests as stent fracture.¹³ The proximal segment of the RCA, fixed by the aortic root, shows a conical pendulum movement,¹⁴ and the mid-segment has also been demonstrated to exhibit larger displacement during a contraction.¹⁵ Accordingly, the calculated hinge angle was larger in the proximal segment, followed by the mid-segment, and was larger at the proximal than distal edges within a stent. In addition, the present study demonstrated that there was a positive association between the pre-stenting hinge angle or vessel curvature index and the post-stent hinge angle. Native vessel movement and/or tortuosity may contribute to the post-stent hinge motion at the edge. The adverse effects of hinge motion have been studied, with major adverse cardiovascular events reported for bare-metal stents,³ TLR reported for sirolimus-eluting stents (C-SES; Cypher^®; Cordis J&J, Miami, FL, USA),⁴ and SER reported in a combined analysis of C-SES, paclitaxel-eluting stents (Taxus^®; Boston Scientific, Natick, MA, USA), and zotarolimus-eluting stents (Endeavor^®; Medtronic, Fridley, MN, USA).¹⁶

In the present study, SER accounted for 55% (23/42) of overall restenoses, and hinge angle was a dominant predictor of SER. Along with hinge angle, known predictors for restenosis, including diabetes and hyperlipidemia, were highly prevalent in SER lesions, whereas these factors were not associated with the other restenosis lesions. Ostium stenting, which was reported to increase the incidence of stent recoil, was also highly prevalent in the SER group.¹⁷ Total stent length was significantly longer only in the other restenosis lesions group, whereas the diameter of the stent and reference vessel appeared to be equivalent among the SER, other restenosis and no-restenosis groups. It is possible that the advances in DESs and/or medical treatment have worked favorably to decrease the other types of restenosis, whereas factors reflecting local inflammation or the mechanical burden still have an impact on SER. In terms of commercial stent-design, no significant differences were observed in hinge angle or the incidence of SER among 8 different DES platforms. Hinge motion at the stent edge could still participate in the development of SER, even in the current DES era.

Edge Vascular Environment and Effect of Hinge Motion

Past studies have pointed out the importance of residual plaque burden in the reference vessel in the occurrence of SER.^2,4,8 The findings of the present study agree with these previous results. Interestingly, no SER developed among the edge segments with pure fibrous plaques. This could reflect the potential role of the lipid component in plaque progression, as suggested by previous studies.^7,8,18 In the present study, more than 75% of SERs (10/13) were detected at edge segments with a >50% residual plaque burden, and all except one (i.e., 9/10) had a hinge angle >11° (upper 75th percentile). The coexistence of excessive hinge motion and high residual plaque burden appeared to be a key mechanism for SER. It is theoretically assumed that a substantial plaque deposit provides conditions for atherosclerotic progression,^18,19 and this phenomenon may be mechanically provoked by hinge motion with local inflammation after stent placement.²⁰ In addition, the changes in geometry caused by stent implantation in the curved coronary artery have been reported to alter the distribution of wall shear stress at stented segments and in the transition zones.²¹ Changes in wall shear stress at the edges with excessive hinge motion may result in local plaque progression and potentially restenosis.²

The pathophysiology of the edge vascular response and consequent SER is based on multiple mechanisms, including chronic stent recoil and constrictive vessel remodeling.² We demonstrated that the hinge angle was associated with a reduction in stent area at follow-up. In addition, axial stent deformation was detected in 10 of 13 SER cases (Supplementary Figure 3), implying a continuous mechanical stress at the hinge position.²² In contrast, there were few incidents of stent fracture even at edges with excessive hinge motion, which was in contrast with previous reports.²³ The flexible design of newer-generation DES may prevent fracture and instead cause deformation.¹³ With regard to vessel remodeling, there was a trend for a decrease in reference vessel area with a larger hinge angle. In previous reports, constrictive remodeling was observed within reference segments of DES to various degrees.² The endothelial injury and flow turbulence that accompany hinge motion could contribute to constrictive remodeling.^2,24 It was also notable that cleaved calcification and its protrusion were visualized by follow-up IVUS in 2 SER lesions (Supplementary Figure 3). When mechanical stress was imposed on calcified plaques, the calcification could crack and result in stenotic progression.

Hinge Motion and Clinical Consequences

Finally, the hinge angle was associated with both clinically driven and any TLR. The disproportion of physical stress on both sides of the edges could have substantial effect on clinical events, even when using second-generation or newer DESs. To minimize the adverse effects of hinge motion, it may be beneficial to select a segment without excessive vessel movement as the optimal stent landing zone. Consideration of plaque characteristics around the stent edge should be also important, especially in the presence of excessive vessel movement. However, we often encounter diffuse disease where the reference vessels are rarely normal and it is difficult to select a plaque-free position for stent placement. Even then, the stent landing zone should be determined to minimize the angle, and/or a stent with better flexibility should be chosen. In addition, a “no stent strategy” using drug-coated balloons or the biodegradable vascular scaffolding may reduce changes in vessel geometry and suppress the adverse effect of mechanical stress at the stent edge.^25,26

Study Limitations

The present study has several limitations. First, this was a retrospective analysis from a single center; follow-up CAGs were conducted over a relatively broad period, and follow-up intravascular imaging analysis was performed for lesions mostly requiring retreatment, which could cause selection bias. Second, the sample size was relatively small. Some analyses may be underpowered to detect statistically significant differences. Third, this study included both IVUS and OCT. Discrepancies between IVUS and OCT measurements have been reported to varying degrees.^11,12 The difference between the quantitative measurement of IVUS and OCT by the observers in the present study was assessed using a swine model under stent implantation. The stent area and lumen area were 1.2% and 1.3% larger in IVUS than OCT with a strong correlation (stent area 19.4 vs. 19.2 mm², R=0.98; lumen area 11.2 vs. 11.0 mm², R=0.98), which implies that the quantitative discrepancy would likely have had little effect on the results of the present study. For the sensitivity analysis, the OCT measurements were multiplied by 1.04 to adjust them to IVUS measurements based on previous studies.^11,12 The analyses using the IVUS cohort and the IVUS plus corrected OCT cohort demonstrated consistent results with the present study (Supplementary Table 2). Although differences in the quantitative and qualitative imaging modalities could affect the interpretation of the results, these analyses may underscore the robustness of the results of the present study. Fourth, the treatment strategy was decided by each interventionist. Selection bias regarding treatment strategy may be present. Finally, attention may be required to translate some analyses, because both stent edges were not independent and more than one lesion per patient was included in the analysis.

Conclusions

The present study demonstrated that hinge motion at the stent edge remained an independent predictor of SER and was associated with an increased risk of TLR, even in the current DES era. Substantial stress, determined by angulation at the stent edge, and its interaction with residual plaque can be considered as one of the plausible mechanisms for edge restenosis. For RCA lesions with excessive vessel movement, it is important to determine the stent landing zone to minimize hinge motion. Further, optimization of future stent design is needed to reduce mechanical stress, a potential limitation of DESs.

Acknowledgments

The authors are indebted to all who helped with this study.

Sources of Funding

This study did not receive any specific funding.

Disclosures

None declared.

IRB Information

This study was approved by the Ethics Committee of NTT Medical Center Tokyo (No. 20-77).

Data Availability

Data will not be shared.

Supplementary Files

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-21-0196

References

• 1.   Dangas GD, Claessen BE, Caixeta A, Sanidas EA, Mintz GS, Mehran R. In-stent restenosis in the drug-eluting stent era. J Am Coll Cardiol 2010; 56: 1897–1907.
• 2.   Gogas BD, Garcia-Garcia HM, Onuma Y, Muramatsu T, Farooq V, Bourantas CV, et al. Edge vascular response after percutaneous coronary intervention: An intracoronary ultrasound and optical coherence tomography appraisal: From radioactive platforms to first- and second-generation drug-eluting stents and bioresorbable scaffolds. JACC Cardiovascular interventions 2013; 6: 211–221.
• 3.   Gyöngyösi M, Yang P, Khorsand A, Glogar D. Longitudinal straightening effect of stents is an additional predictor for major adverse cardiac events. J Am Coll Cardiol 2000; 35: 1580–1589.
• 4.   Ino Y, Kubo T, Kitabata H, Shimamura K, Shiono Y, Orii M, et al. Impact of hinge motion on in-stent restenosis after sirolimus-eluting stent implantation. Circ J 2011; 75: 1878–1884.
• 5.   Kedhi E, Joesoef KS, McFadden E, Wassing J, van Mieghem C, Goedhart D, et al. Second-generation everolimus-eluting and paclitaxel-eluting stents in real-life practice (COMPARE): A randomised trial. Lancet 2010; 375: 201–209.
• 6.   Yamasaki M, Tsujino I, Lima-Filho MO, Ako J, Shimohama T, Hasegawa T, et al. Comparison of vascular response to the everolimus-eluting stent versus the paclitaxel-eluting stent: Intravascular ultrasound results from the SPIRIT III trial. EuroIntervention 2012; 8: 724–731.
• 7.   Ino Y, Kubo T, Matsuo Y, Yamaguchi T, Shiono Y, Shimamura K, et al. Optical coherence tomography predictors for edge restenosis after everolimus-eluting stent implantation. Circ Cardiovasc Interv 2016; 9: e004231.
• 8.   Kang SJ, Cho YR, Park GM, Ahn JM, Kim WJ, Lee JY, et al. Intravascular ultrasound predictors for edge restenosis after newer generation drug-eluting stent implantation. Am J Cardiol 2013; 111: 1408–1414.
• 9.   Thygesen K, Alpert JS, Jaffe AS, Chaitman BR, Bax JJ, Morrow DA, et al. Fourth universal definition of myocardial infarction (2018). J Am Coll Cardiol 2018; 72: 2231–2264.
• 10.   Ikutomi M, Honda Y, Fitzgerald PJ, Yock PG. Intravascular imaging. In: Moscucci M, Cohen MG, Chetcuti SJ, editors. Atlas of cardiac catheterization and interventional cardiology, 1st edn. Philadelphia: Wolters Kluwer, 2018; 297–362.
• 11.   Kubo T, Akasaka T, Shite J, Suzuki T, Uemura S, Yu B, et al. OCT compared with IVUS in a coronary lesion assessment: The OPUS-CLASS study. JACC Cardiovasc Imaging 2013; 6: 1095–1104.
• 12.   Yamaguchi T, Terashima M, Akasaka T, Hayashi T, Mizuno K, Muramatsu T, et al. Safety and feasibility of an intravascular optical coherence tomography image wire system in the clinical setting. Am J Cardiol 2008; 101: 562–567.
• 13.   Kan J, Ge Z, Zhang JJ, Liu ZZ, Tian NL, Ye F, et al. Incidence and clinical outcomes of stent fractures on the basis of 6,555 patients and 16,482 drug-eluting stents from 4 centers. JACC Cardiovasc Interv 2016; 9: 1115–1123.
• 14.   Nakagawa Y, Hayashi H, Izumi C, Kondo H, Tamura T, Enomoto S, et al. Four-dimensional computed tomography-based finite element modeling of the behavior of the right coronary artery. Circ J 2017; 81: 1059–1061.
• 15.   Shechter G, Resar JR, McVeigh ER. Displacement and velocity of the coronary arteries: Cardiac and respiratory motion. IEEE Trans Med Imaging 2006; 25: 369–375.
• 16.   Kim YG, Oh IY, Kwon YW, Han JK, Yang HM, Park KW, et al. Mechanism of edge restenosis after drug-eluting stent implantation: Angulation at the edge and mechanical properties of the stent. Circ J 2013; 77: 2928–2935.
• 17.   Tsunoda T, Nakamura M, Wada M, Ito N, Kitagawa Y, Shiba M, et al. Chronic stent recoil plays an important role in restenosis of the right coronary ostium. Coron Artery Dis 2004; 15: 39–44.
• 18.   Stone GW, Maehara A, Lansky AJ, De Bruyne B, Cristea E, Mintz GS, et al. A prospective natural-history study of coronary atherosclerosis. N Engl J Med 2011; 364: 226–235.
• 19.   Virmani R, Kolodgie FD, Burke AP, Farb A, Schwartz SM. Lessons from sudden coronary death: A comprehensive morphological classification scheme for atherosclerotic lesions. Arterioscler Thromb Vasc Biol 2000; 20: 1262–1275.
• 20.   Nakano M, Otsuka F, Yahagi K, Sakakura K, Kutys R, Ladich ER, et al. Human autopsy study of drug-eluting stents restenosis: Histomorphological predictors and neointimal characteristics. Eur Heart J 2013; 34: 3304–3313.
• 21.   LaDisa JF Jr, Olson LE, Douglas HA, Warltier DC, Kersten JR, Pagel PS. Alterations in regional vascular geometry produced by theoretical stent implantation influence distributions of wall shear stress: Analysis of a curved coronary artery using 3D computational fluid dynamics modeling. Biomed Eng Online 2006; 5: 40.
• 22.   Ishida K, Minami Y, Fujiyoshi K, Ishida M, Ako J, Inomata T. Recurrent in-stent restenosis with eccentric neointimal hyperplasia at a lesion with hingelike motion. JACC Cardiovasc Interv 2020; 13: e103–e104.
• 23.   Lee MS, Jurewitz D, Aragon J, Forrester J, Makkar RR, Kar S. Stent fracture associated with drug-eluting stents: Clinical characteristics and implications. Catheter Cardiovasc Interv 2007; 69: 387–394.
• 24.   Ward MR, Pasterkamp G, Yeung AC, Borst C. Arterial remodeling: Mechanisms and clinical implications. Circulation 2000; 102: 1186–1191.
• 25.   Jeger RV, Farah A, Ohlow MA, Mangner N, Möbius-Winkler S, Leibundgut G, et al. Drug-coated balloons for small coronary artery disease (BASKET-SMALL 2): An open-label randomised non-inferiority trial. Lancet 2018; 392: 849–856.
• 26.   Brugaletta S, Gogas BD, Garcia-Garcia HM, Farooq V, Girasis C, Heo JH, et al. Vascular compliance changes of the coronary vessel wall after bioresorbable vascular scaffold implantation in the treated and adjacent segments. Circ J 2012; 76: 1616–1623.