SCAI Expert Consensus Statement on Management of In-Stent Restenosis and Stent Thrombosis

Abstract

Stent failure remains the major drawback to the use of coronary stents as a revascularization strategy. Recent advances in imaging have substantially improved our understanding of the mechanisms underlying these occurrences, which have in common numerous clinical risk factors and mechanical elements at the time of stent implantation. In-stent restenosis remains a common clinical problem despite numerous improvements in-stent design and polymer coatings over the past 2 decades. It generates significant health care cost and is associated with an increased risk of death and rehospitalization. Stent thrombosis causes abrupt closure of the stented artery and therefore carries a high risk of myocardial infarction and death. This Society for Cardiovascular Angiography & Interventions (SCAI) Expert Consensus Statement suggests updated practical algorithmic approaches to in-stent restenosis and stent thrombosis. A pragmatic outline of assessment and management of patients presenting with stent failure is presented. A new SCAI classification that is time-sensitive with mechanistic implications of in-stent restenosis is proposed. Emphasis is placed on frequent use of intracoronary imaging and assessment of timing to determine the precise etiology because that information is crucial to guide selection of the best treatment option. SCAI recommends image-guided coronary stenting at the time of initial implantation to minimize the occurrence of stent failure. When in-stent restenosis and stent thrombosis are encountered, imaging should be strongly considered to optimize the subsequent approach.

Introduction

Coronary stenting has transformed revascularization strategy, producing excellent procedural and clinical outcomes in myriad clinical settings. Despite proven short and long-term benefits, in-stent restenosis (ISR) and stent thrombosis (ST) continue to be limitations. There remains no definitive management approach for either condition despite a greater understanding of the underlying mechanisms ensuing from advances in intracoronary imaging.

In this Society for Cardiovascular Angiography & Interventions (SCAI) Expert Consensus Statement, practical algorithmic approaches to ISR and ST are offered. A pragmatic outline of assessment and management of patients presenting with stent failure is presented. A new SCAI classification that is time-sensitive with mechanistic implications of ISR is proposed. Emphasis is placed on frequent use of intracoronary imaging and assessment of timing to determine the precise etiology, as that information is crucial to guide selection of the best treatment option.

Methodology

This statement has been developed according to SCAI Publications Committee policies for writing group composition, disclosure and management of relationships with industry, internal and external review, and organizational approval. Detailed author disclosures are included as Supplemental Table 1. The work of the writing committee was supported exclusively by SCAI, a nonprofit medical specialty society, without commercial support. Writing group members contributed to this effort on a volunteer basis and did not receive payment from SCAI. Group members in each section performed literature searches, and the section leads in collaboration authored initial section drafts with other members of the writing group. The draft manuscript was peer reviewed in February 2023 and the document was revised to address pertinent comments. The writing group unanimously approved the final version of the document. The SCAI Publications Committee and Executive Committee endorsed the document as official society guidance in March 2023. SCAI statements are primarily intended to help clinicians make decisions about treatment alternatives. Clinicians also must consider the clinical presentation, setting, and preferences of individual patients to make judgments about the optimal approach.

In-stent restenosis

ISR remains a common clinical problem despite numerous improvements in-stent design and polymer coatings over the past 2 decades. ISR generates significant health care cost and is associated with an increased risk of death and rehospitalization. The incidence of ISR is 10%; 25% of ISR cases present with acute myocardial infarction (MI) with a 30-day mortality rate of 10% to 25%.1, 2, 3, 4

Risk factors

Clinical. The incidence of ISR varies depending on individual patient, angiographic and procedural characteristics as listed in Table 1.1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 Second-generation drug-eluting stents (DES) have a 5.7% ISR rate in patients without diabetes, and 8.7% rate in those with diabetes.⁵ Beyond 1 year, there is a gradual increase in major adverse cardiovascular events (MACE); the 5-year ISR rate is 9% to 12% in noncomplex lesions.⁶

Table 1.

In-stent restenosis risk factors1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15

Patient factors	Angiographic factors	Procedural factors
•Diabetes mellitus •Renal insufficiency •ACS presentation •Female •Recurrent ISR	•Lesion length >20 mm •Diameter <3 mm •Chronic total occlusion •Ostial location •Bifurcation •Saphenous vein graft •Severe Calcification •Multivessel CAD	•Underexpansion •Stent fracture •Bare metal stent •Stenoses proximal and distal to stent •Major arterial dissection involving media or >3 mm length •Multiple stent layers

ACS, acute coronary syndrome; CAD, coronary artery disease; ISR, in-stent restenosis.

Recurrent ISR is not unusual in contemporary practice. The failure to appreciate and address the original mechanism of ISR underlies refractory cases of recurrence. As in first ISR, the use of intracoronary imaging may provide insights into the underlying mechanisms. Recurrent ISR occurs in approximately 20% of all ISR cases.^7,8 Recurrence is independently predicted by the number of stents placed at the location.^9,10 The 1-year MACE (43.1%) and target lesion revascularization (41.2%) rates were significantly higher in the ≥3 stent layer group than in the 1-stent-layer and 2-stent-layer groups. Importantly, on multivariable analysis, the number of metallic layers and hemodialysis requirement were identified as independent predictors of MACE. A third layer of metal is almost always associated with underexpansion and should be avoided.

Pathogenesis and contributory factors

The preferred treatment strategy depends on a precise diagnosis and understanding of the cause. Consequently, identifying the mechanism in each case using intracoronary imaging and optimizing the interventional result are critical steps (Table 1).

Biologic factors

The primary biologic mechanism of ISR is neointimal tissue proliferation or hyperplasia, an exaggerated homeostatic healing response to arterial wall damage sustained during stent implantation.¹ The distribution of neointimal tissue proliferation may be focal or diffuse along the length of the stent. Causative factors are local inflammation resulting from mechanical disruption of the intima/media leading to aggressive neointimal hyperplasia/proliferation that consists of smooth muscle cells and extracellular matrix. Hypersensitivity reactions to the metal and/or the polymer of early-generation DES are also recognized mechanisms of neointimal hyperplasia.¹

Neoatherosclerosis is an increasingly recognized mechanism of stent failure seen with current generation DES. It is characterized by accumulation of lipid-laden foamy macrophages sometimes with necrotic core formation within stented segments.¹⁶ Injury to the vessel by balloon inflation and stent deployment stimulates neointima formation. The subsequent intimal and medial damage leads to proliferation and migration of vascular smooth muscle cells, macrophages, and extracellular matrix formation. These activate the coagulation cascade and an inflammatory response. This combination of events, along with elution of antiproliferative drug, inhibits endothelialization. The lack of endothelium allows incorporation of low-density lipoprotein into the artery wall early after DES implantation. At later stages, the healed in-stent neointima is prone to atherosclerosis development.

Additional mechanisms of ISR include elastic recoil and relocation/subluxation of axially transmitted plaque (tissue intrusion) (especially early) and reorganization of thrombus, neointima formation, and remodeling (especially late).6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16

Mechanical factors

The primary mechanical cause of ISR is underexpansion. This may result from stent undersizing, low deployment pressures, or underlying calcified lesions. Other mechanical causes include stent recoil, longitudinal stent deformation, stent fracture, crushed stents, dislocated stents, and geographic miss. Geographic miss results from incorrect placement of the stent so that it does not fully cover the diseased segment. Stent fracture may be seen at hinge points in the coronary artery and after stenting a calcified nodule. Other findings, such as early-stent malapposition, tissue prolapse, and asymmetry/eccentricity have little or no prognostic value. Stent underexpansion may occur as a result of undersizing, low deployment pressures, or heavily calcified lesions.9, 10, 11, 12, 13

Some interventional cardiologists favor routine poststent placement dilation with high-pressure balloons; while this can be an effective strategy, it can also lead to edge dissections. Instead, postprocedural imaging might be a more effective use of time and effort.

Definition and classification

In-stent restenosis is established angiographically as a binary event, defined as recurrent diameter stenosis at the stent segment >50% of the vessel diameter.¹⁶ Additional criteria for clinically relevant ISR include: recurrent angina, objective signs of ischemia, or abnormal fractional flow reserve.17, 18, 19

Morphologic patterns

Coronary angiography remains the standard diagnostic method to determine ISR severity and morphologic pattern:

Mehran

The Mehran System¹⁹ classifies restenotic lesions on the basis of morphology and extent of disease, with 4 subclasses based on location within the stented segment. Lesions were classified as focal (class I), diffuse intrastent (class II), diffuse proliferative (class III), and total occlusion (class IV). This schema was highly relevant to bare metal stenting (BMS), but its applicability to DES ISR is uncertain.

Waksman

The Waksman ISR Classification²⁰ is based on mechanistic considerations informed by intracoronary imaging. There are 5 groups of DES ISR identified: mechanical (type I; underexpansion I A, stent fracture I B), biologic (type II; intimal hyperplasia II A, neoatherosclerosis noncalcified II B, neoatherosclerosis calcified II C), mixed pattern (type III), chronic total occlusions (type IV), and lesions previously treated with >2 stents (type V).

Intravascular ultrasound- and optical coherence tomography-based classifications

Kang et al²¹ has proposed an intravascular ultrasound (IVUS)-based classification that incorporates length of the restenosis as well as minimal luminal area. Gonzalo et al²² and Ali et al²³ have proposed optical coherence tomography (OCT) classifications that rely on both quantitative and qualitative parameters.

Timing

Table 2 is the proposed new SCAI classification incorporating the cause of ISR based on time from implantation. SCAI recommends that early (<30 days), late (30 days to 1 year), and very late (>1 year) timing categories be adopted for all future diagnostic and therapeutic studies. By integrating mechanistic etiology with timing, this classification will be useful to determine best treatment options.

Table 2.

SCAI classification of in-stent restenosis: a system based on time interval and causative factor

Classification	Time interval	Morphologic substrates
Early	<30 d	•Undersizing •Underexpansion •Stent fracture
Late	30 d to 1 y	•Delayed healing (including drug induced) •Uncovered stent struts •Intimal hyperplasia (especially in BMS ISR)
Very late	>1 y	•Neoatherosclerosis •Intimal hyperplasia •Stent fracture

BMS, bare metal stent; ISR, in-stent restenosis.

Imaging adjuncts to diagnosis

SCAI strongly recommends routine evaluation by intravascular imaging to determine the cause of ISR, to inform therapeutic strategy, and to confirm effective treatment after percutaneous coronary intervention (PCI).24, 25, 26, 27, 28, 29 Identifying the mechanism of stent failure is paramount because the causative factors will influence the selection of treatment and devices to manage the ISR, ultimately impacting the durability of the repeat revascularization. Despite being the primary means of assessing ISR in clinical practice, angiography alone is usually inadequate because of limited resolution and inherent deficiency in quantifying vessel size, stent size, stent expansion, number of stent layers, in-stent calcific neoatherosclerosis, and extrastent calcific disease. Identifying the mechanism of ISR depends on visualizing the stent and its relation to the arterial wall, rather than the lumen itself.

In contrast to angiography, IVUS and OCT provide detailed assessment of the native artery and stented segment (Figure 1A, B). Recent intravascular imaging studies demonstrate that suboptimal stent deployment is common—occurring in 31% to 58% of patients—and that suboptimal stent deployment confers an increased risk of adverse events.30, 31, 32, 33 The relative advantages of IVUS and OCT are summarized in Table 3.^34,35

Figure 1.

Mechanisms of in-stent restenosis evaluated by intravascular imaging. A′-D′ are optical coherence tomography (OCT) or intravascular ultrasound (IVUS) images corresponding to the in-stent restenosis seen in the angiographic images (A-D, white arrows). A″-D″ are representative diagrams provided to clarify the intracoronary images, A′-D′. Top. Mechanisms of in-stent restenosis evaluated by OCT. (A) A patient experienced recurrent in-stent restenosis (ISR), and OCT visualized a severely underexpanded stent because of circumferential thick calcium behind stent with only a minimum amount of neointimal hyperplasia. (B) This patient was treated with a single drug-eluting stent. At the time of ISR, the OCT image showed a lack of stent struts over half of the arterial circumference (double headed arrow) while stents struts were overlapped at 7 to 9 o’clock. These are typical features of stent fracture. (C) Excess amount of neointimal hyperplasia within a well-expanded stent. (D) Lipidic neointima (strong signal attenuation) within the stent struts indicating neoatherosclerosis. Bottom. Mechanisms of in-stent restenosis evaluated by IVUS (A) IVUS visualized an underexpanded stent with a minimum amount of neointimal hyperplasia. By looking at the adjacent segment, the cause of underexpansion was a small vessel with a myocardial bridge. (B) IVUS delineates overlapped struts within a single stent at 7 to 10 o’clock indicating stent fracture. (C) Excess amount of neointimal hyperplasia within a well-expanded old stent. (D) Calcified plaque (superficial hyperintensity with acoustic shadow from 8 to 12 o’clock) within the stent indicates neoatherosclerosis.

Table 3.

Applications of OCT vs IVUS^34,35

	IVUS	OCT
Assessing lesion severity in left main disease	+++	+
Assessing de novo lesion characteristics
Thin cap fibroatheroma	−	+++
Thrombus	+	+++
Plaque rupture	++	+++
Calcified nodule	+	+++
Dissection	++	+++
Positive remodeling	+++	+
Plaque burden	+++	+
Aorto-ostial disease	+++	−
Stent optimization
Expansion	++	+++
Apposition	++	+++
Stent failure
Neointimal hyperplasia	+	++
Underexpansion	++	+++
Malapposition	++	+++
Renal impairment	+++	+

+++ Excellent; ++ Good; + Poor; − Not advised

IVUS, intravascular ultrasound; OCT, optical coherence tomography.

Suboptimal minimal stent area (MSA) is a major predictor of stent failure, and an IVUS optimized MSA of >5.0 mm² or OCT optimized MSA of >4.5 mm² are optimal goals. Another useful criterion is to achieve a target MSA >90% of the closest proximal or distal reference segment. In addition, intraluminal diagnostic imaging should be performed to ensure that there are no inflow or outflow obstructions within 5 mm of the proximal or distal stent edge. In particular, any major edge dissections (defined as >60°, >3 mm in length, or penetrating the media) should be stented.33, 34, 35

Physiologic assessment

Patients with ISR of intermediate range severity on coronary angiography present a clinical challenge because of potential short and long-term complications, and it is recommended that objective evidence of myocardial ischemia is demonstrated prior to proceeding with repeat intervention. Even though there are no randomized clinical trials (RCTs) assessing coronary physiology to guide management of ISR, there are several retrospective observational trials that suggest that it may assist in clinical decision-making.^36,37 Deferral of coronary revascularization in patients with ISR and fractional flow reserve >0.80 was associated with similar outcomes over 36 months to patients with de novo coronary stenosis.³⁷ Further studies may define the value of coronary physiology assessment in developing decision strategy before and after intervention.

Proposed treatment strategies

A summary of existing RCTs and registries,^20,38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51 including clinical situations in which particular treatment modalities have been shown to be advantageous, are presented in Table 4.52, 53, 54, 55 The most common treatment approach for the first episode of ISR is to implant a second DES, based on the rationale that DES therapy has superior efficacy over balloon angioplasty alone. However, this is not always necessary and may not be the best solution, particularly when the reference vessel and the resultant minimal lumen area are small.²⁰ If the underlying etiology is not directly addressed and corrected, there is a high likelihood of recurrent ISR, and the rate of ISR in second layer DES is high: 12% to 16% at 12 months and 33% at 3 to 5 years.56, 57, 58

Table 4.

Summary of in-stent restenosis and stent thrombosis management strategies

Modalities of treatment	When to consider	Other considerations
In-stent restenosis1,2,20,38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51
Balloon angioplasty	•Focal, discrete lesions •Stent underexpansion •Need for short DAPT	•Risk of recurrence and edge dissection high •Use of DCB associated with lower risk of TLR and binary restenosis
Repeat DES	•If only one prior layer, may consider over balloon angioplasty alonea •Focal edge restenosis, stent gap, stent fracture	•Reduction in need for target revascularization compared with angioplasty alone •No definitive consensus for change in-stent type but RIBS III trial showed reduction of restenosis rate and improved event-free survival in cohort receiving a different DES platform
Cutting and scoring balloons	•May modify neointimal growth •May help to avoid additional stent layer	•Scoring balloon angioplasty superior to PTA alone at 6-8 mo (improved angiographic outcomes and reduced stenosis)
Atheroablation	•Should be considered if mode of stent underexpansion calcification and resistant to high-pressure balloons •Should be considered when significant neointimal hyperplasia present	•Clinical trials for use of atheroablation for ISR negative
DCB	•May help to avoid additional stent layer	•Treatment with DCB non-inferior to DES in terms of 6-mo MLD
Vascular brachytherapy	•Refractory ISR •Limited availability	•Due to delay in endothelialization, patients may need lifelong DAPT •Treatment can be repeated every 12 mo
Intravascular lithotripsy	•May be considered when there is highly calcified neoatherosclerosis	•Limited data to suggest proper case selection

Stent thrombosis52, 53, 54, 55
Balloon angioplasty	•May be needed in addition to repeat DES and/or aspiration thrombectomy to restore coronary blood flow
Repeat DES	•May be needed in addition to PTA and/or aspiration thrombectomy to restore coronary blood flow •Should normally limit to significant residual dissections after PTA	•No stent type associated with reduction in ST
Aspiration thrombectomy	•Consider when heavy thrombus burden present •Consider adjunctive glycoprotein IIb/IIIa inhibitors if persistent heavy thrombus burden after aspiration	•Associated with improved microvascular perfusion during STEMI because of ST •Majority of patients undergoing aspiration thrombectomy had successful recanalization
Pharmacologic therapies	•Consider glycoprotein IIb/IIIa inhibitor infusion •Assess compliance and consider switch to higher potency antiplatelet therapy if the patient was compliant and still taking DAPT •May consider drug resistance testing and prolonged DAPT duration	•Consider patient’s renal function and bleeding risk when continuing glycoprotein IIb/IIIa inhibitor after PCI •Prolonged anticoagulation and antiplatelet therapy may be beneficial when residual thrombus is detected following intervention

DAPT, dual antiplatelet therapy; DCB, drug-coated balloon; DES, drug-eluting stent; ISR, in-stent restenosis; MLD, minimal lumen diameter; PCI, percutaneous coronary intervention; PTA, percutaneous transluminal angioplasty; STEMI, ST-elevation myocardial infarction; TLR, target lesion revascularization.

^aAvoid when there are already 2 layers of stent

General strategic approach

The critical principle is to obtain the largest acute lumen gain as possible by maximizing the immediate postprocedural minimal luminal area. To operationalize this concept, a complete diagnostic evaluation of the cause of ISR must be pursued.⁵⁹ An algorithmic approach is provided in Figure 2. Repeat PCI should be routinely performed following intracoronary imaging assessment. The mechanism of the initial ISR should be determined, with correction of any underlying mechanical factors with image guidance to ensure optimal sizing and expansion. A second stent should be image-guided to ensure correct stent expansion to ensure appropriate stent expansion.

Figure 2.

SCAI algorithmic approach to in-stent restenosis. DES, drug-eluting stents; ISR, in-stent restenosis; PCI, percutaneous coronary intervention.

Besides repeat DES, a number of adjunct treatments exist that may be highly effective.58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69 If there is significant underexpansion, it is critical to increase expansion by applying high-pressure balloons. If there is additional hyperplasia, perhaps preparation with scoring/cutting balloons, rotational atherectomy (RA), orbital atherectomy (OAS), drug-coated balloons (DCBs), vascular brachytherapy (VBT), excimer laser coronary angioplasty (ELCA), or intravascular lithotripsy (IVL) may be useful.

When ISR is predominantly because of neointimal hyperplasia, treatment is dependent on the pattern of ISR. For focal ISR, a high-pressure or scoring/cutting balloon may be sufficient; ELCA or atherectomy may be beneficial in selected cases. For diffuse ISR, atherectomy or scoring/cutting balloon angioplasty followed by repeat DES implantation is typically advised.

If stent underexpansion is not because of calcification, atheroablation should be used only if significant neointimal hyperplasia is also present. RA, OAS, and ELCA may debulk neointima hyperplasia, although mechanistic evaluations fail to demonstrate this effect. In cases where intravascular imaging identifies an arc of calcium >270° or >0.67 mm in thickness, atherectomy vessel preparation should be considered to optimize lesion and stent expansion.60, 61, 62, 63, 64, 65, 66, 67, 68, 69

If stent underexpansion is due to significant peri-stent calcium (>90°), RA, OAS, and ELCA may be employed to improve stent underexpansion by disrupting the calcified plaque behind the stent. IVL may also be useful. These techniques are associated with calcium modification and/or fracture, and when followed by high-pressure inflations may reduce stent underexpansion. However, if unsuccessful, coronary artery bypass grafting may be necessary (see Figure 3).

Figure 3.

Stepwise, imaging-guided treatment of a calcified ISR lesion. In-stent restenosis of a first-generation drug-eluting stent (Taxus, Boston Scientific) in the mid LAD, implanted 18 years prior and confirmed patent several years afterward. Intrastent (type II) restenosis (A, inset with orthogonal view) was noted on angiography with multiple unsuccessful attempts made at dilatation (B, C) using noncompliant balloons (24-26 atm). Subsequent IVUS imaging (D, distal to proximal) reveals both calcified and noncalcified tissue within an appropriately sized stent, likely representing neoatherosclerosis. Despite successful rotational atherectomy (E: 1.5 mm burr, 160-170,000 rpm × 8 passes), non-compliant balloon dilation still revealed a focal waist (F). Complete expansion of a 2.5 × 12 mm Shockwave C2 IVL balloon was achieved after 80 pulses at 6 atm (G) and confirmed with IVUS. A 2.75 × 26 mm drug-eluting stent was implanted at 20 atm with excellent angiographic (G) and IVUS results. IVL, intravascular lithotripsy; IVUS, intravascular ultrasound; LAD, left anterior descending artery.

Balloon angioplasty

Balloon angioplasty should be the initial step in focal lesions or if short dual antiplatelet therapy (DAPT) duration is required. In the setting of stent underexpansion, high-pressure noncompliant balloon inflations are the preferred strategy.

Super high-pressure balloons

Double layer, noncompliant coronary balloons (OPN NC, SIS Medical) capable of inflation pressures ranging from 35 to 55 atm have recently become available in the United States. This class of percutaneous transluminal coronary angioplasty balloon has performed favorably in severely calcified de novo lesions and may be a consideration in ISR secondary to an underexpanded stent.

Repeat DES

In general, repeat DES implantation has historically shown superior results compared with balloon angioplasty alone. However, this approach should only be undertaken once appropriate sizing and expansion of the original stent has been assured using intravascular imaging. A second stent may not be necessary if the original stent was underdeployed and can be corrected.

If focal edge restenosis, stent gap, or stent fracture is identified, conventional or high-pressure balloon dilation at the site of the mechanical complication should be the initial treatment. This should then be followed by repeat DES implantation when the ISR is focal. Repeat DES to cover the entire diseased segment can be performed when the ISR is diffuse or proliferative, but care should be taken to minimize the stent coverage as much as possible.44, 45, 46, 47, 48, 49, 50, 51, 52 There is no definitive evidence regarding which type of DES should be used to treat ISR of a previously implanted DES, and there is no consensus on whether a different stent type or drug should be used when an additional DES is implanted. However, the RIBS III trial³⁸ assessed the impact of selecting a different DES for treatment of ISR and demonstrated better angiographic and clinical outcomes at 9-month follow-up in the cohort that received a different DES than the first implanted stent.

Cutting and scoring balloons

The use of balloons incorporating cutting or scoring elements has been shown in very small series to result in better acute angiographic outcomes in ISR compared with BA.^58,59 Small IVUS-guided studies suggest that the use of cutting compared with traditional balloons is associated with larger lumen gain, lower lumen loss, and preserved angiographic result at follow-up.^45,69 However, a randomized study comparing standard balloons vs cutting balloons for the treatment of ISR failed to demonstrate superiority of the cutting balloon in terms of recurrent ISR and MACE.⁵⁸

Atheroablation

Despite the appealing concept of atheroablation, clinical trials for the treatment of ISR have been negative.62, 63, 64 The utilization of RA when IVUS or OCT confirms the presence of calcium within neoatherosclerotic plaques is reasonable, but no controlled trials exist. RA might also have value in lesions refractory to high-pressure balloon angioplasty. However, the use of mechanical atheroablative technologies within stents poses risks of device entrapment, and some suggest reserving mechanical atherectomy for bailout use. Excimer laser has similarly shown no special benefits.^61,64,67

DCBs

Preliminary clinical studies using sirolimus-based DCBs showed promising preliminary results; however, randomized trials are needed to demonstrate efficacy with DES in ISR. Most data are with paclitaxel DCBs. Several trials and meta-analyses have demonstrated similar outcomes of DCB compared with DES in the management of ISR^44,45,70,71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83 and are summarized in Table 5. Although coronary DCBs are not available in the United States, the European Society of Cardiology/European Association for Cardio-Thoracic Surgery Guidelines⁶⁵ give DCBs a class I indication for the treatment of ISR. DCB outcomes may be optimized with inflation time >60 seconds and a balloon: artery ratio >0.91. Neointimal modification with RA improves acute luminal improvement over DCB therapy alone, with lower late lumen loss; however, there are similar clinical outcomes at 6 months.⁵⁰

Table 5.

Selected Trials Evaluating Management Strategies for In-stent Restenosis

Trial, publication year	Investigation Time	No. of lesions centers, region	Design	Drug-coated balloon, carrier agent, commercial name	Control device	Restenotic stent	Endpoint(s)	Follow-up (mo)	Principal findings	P-value
PCB vs Uncoated balloon
PACCOCATH ISR I&II, 201275	Dec 2003 - Dec 2005	54/54 Multicenter, Germany	Core lab, CEC	Paclitaxel, iopromide 3 μg/mm2, PACCOCATH	Uncoated balloon	BMS, DES	LLL (mm)	6	0.11 ± 0.44 vs 0.80 ± 0.79	.001
							TLR (%)	12/60	4 vs 37 / 9 vs 39	.001/.004
							MACE (%)	12/60	9 vs 44 / 28 vs 59	.001/.009
Habara et al, 201186	Sep 2008 - Nov 2009	25/25 1, Japan	—	Paclitaxel, iopromide 3 μg/mm2, SeQuent Please	Uncoated balloon	SES	LLL (mm)	6	0.18 ± 0.45 vs 0.72 ± 0.55	<.01
							TLR (%)	6	4 vs 42	<.01
							MACE (%)	6	4 vs 40	<.01
PEPCAD-DES, 201272,76	Nov 2009 - Apr 2011	72/38 Multicenter, Germany	Core lab	Paclitaxel, iopromide 3 μg/mm2, SeQuent Please	Uncoated balloon	DES	LLL (mm)	6	0.43 ± 0.61 vs. 1.03 ± 0.77	<.01
							TLR (%)	6/36	15 vs 37 / 19 vs 37	<.01/<.01
							MACE (%)	6/36	17 vs 50.0 / 21 vs 53	<.01/<.01
PCB vs DES
PEPCAD II, 200970,77	Jan 2006 - Dec 2006	66/65 10, Germany	Core lab, CEC	Paclitaxel, iopromide 3 μg/mm2, SeQuent Please	PES, durable polymer, stainless steel (132 μm)	BMS	LLL (mm)	6	0.17 ± 0.42 vs 0.38 ± 0.6	.03
							TLR (%)	12	6 vs 15	.15
							MACE (%)	12/36	9 vs 22 / 35 vs 42	.08/–
SEDUCE, 201474	Jun 2009 - Oct 2011	24/25 2, Belgium	Core lab, CEC	Paclitaxel, iopromide 3 μg/mm2, SeQuent Please	EES, durable polymer, CoCr (81 μm)	BMS	LLL (mm)	9	0.28 vs 0.07	.1
							TLR (%)	12	4.2 vs 8	.576
RIBS V, 201473	Jan 2010 - Jan 2012	95/94 25, Spain	Core lab, CEC	Paclitaxel, iopromide 3 μg/mm2, SeQuent Please	EES, durable polymer, CoCr (81 μm)	BMS	LLL (mm)	6 to 9	0.14 ± 0.5 vs 0.04 ± 0.5	.14
							TLR (%)	12/36	6 vs 1 / 8 vs. 2	.09/.04
							MACE (%)	12/36	8 vs 6 / 12 vs 10	.60/.64
TIS, 201678	Jan 2012 - Aug 2014	74/74 1, Czech Rep.	Core lab, CEC	Paclitaxel, iopromide 3 μg/mm2, SeQuent Please	EES, durable polymer, CoCr (81 μm)	BMS	LLL (mm)	12	0.02 vs 0.19	<.01
							TVR (%)	12	7.4 vs 16.2	.110
							MACE (%)	12	10.3 vs 19.1	.213
ISAR-DESIRE3, 201371,79	Aug 2009 - Oct 2011	137/131/134 3, Germany	Core lab, CEC	Paclitaxel, iopromide 3 μg/mm2, SeQuent Please	1)PES, durable polymer, stainless steel (132 μm) 2)Common balloon	DES	ISR diameter (%)	6 to 8	38% vs 37.4% vs 54.1%	<.01a
							TLR (%)	12/36	22 vs 14 vs 44 / 33 vs 24 vs 51	.09/.11b
							MACE (%)	12/36	24 vs 19 vs 46 / 38 vs 38 vs 56	.5/.91b
PEPCAD China ISR, 201480	Mar 2011 - Apr 2012	113/108 17, China	Core lab, CEC	Paclitaxel, iopromide 3 μg/mm2, SeQuent Please	PES, durable polymer, stainless steel (132 μm)	DES	LLL (mm)	9	0.46 ± 0.51 vs 0.55 ± 0.61	.0005a
							TLR (%)	12/24	15.6 vs 12.3 / 15.9 vs 13.7	.48/.66
							TLF (%)	12/24	16.5 vs 16 / 16.8 vs 18.6	.92/.73
RIBS IV, 201844	Jan 2010 - Aug 2013	154/155 23, Spain	Core lab, CEC	Paclitaxel, iopromide 3 μg/mm2, SeQuent Please	EES, durable polymer, CoCr (81 μm)	DES	Binary restenosis	6 to 9	19% vs 11%	.27
							TLR (%)	12	16.2 vs 21.8	.26
							MACE (%)	12	18.4 vs 23.3	.35
RESTORE, 201881	Apr 2013 - Oct 2016	86/86 10, South Korea	Core lab, CEC	Paclitaxel, iopromide 3 μg/mm2	EES, durable polymer, CoCr (81 μm)	DES	LLL (mm)	9	0.15 ± 0.49 vs 0.19 ± 0.41	.54
							TLR (%)	12	7 vs 5	.51
							MACE (%)	12	6 vs 1	.10
DARE, 201847	May 2010 - Jun 2015	137/141 8, Netherlands	Core lab, CEC	Paclitaxel, iopromide 3 μg/mm2, SeQuent Please	EES, durable polymer, CoCr (81 μm)	BMS, DES	MLD (mm)	6	1.71 ± 0.51 vs 1.74 ± 0.61	<.01a
							TVR (%)	12	7.1 vs 8.8	.65
							MACE (%)	12	10.9 vs 9.2	.66
BIOLUX-RCT, 201882	Aug 2012 - Jan 2015	163/80 14, Germany, Latvia	Core lab, CEC	Paclitaxel, BTHC 3 μg/mm2, Pantera Lux	DES, bioresorbable polymer, CoCr (60–80 μm)	BMS, DES	LLL (mm)	6	0.03 ± 0.40 vs 0.20 ± 0.70	.40
							TLR (%)	12	12.5 vs 10.1	.82
							TLF (%)	12	16.9 vs 14.2	.65
DAEDALUS, 202083		Pooled analysis of 10 RCTc	Core lab, CEC	Paclitaxel, iopromide/BTHC 3 μg/mm2	DES	BMS, DES	TLR (%)	36	16 vs 12, HR 1.27 (0.90-1.79)	.17
							Safety endpointd	36	9 vs 11, HR 0.79 (0.58-1.10)	.16
SCB vs PCB
FIM LIMUS DCB, 201984	Dec 2015 - Jan 2017	25/25 5, Malaysia	Core lab, CEC	Sirolimus, crystalline coating 4 μg/mm2, SeQuent SCB	Paclitaxel, iopromide 3 μg/mm2, SeQuent Please	DES	LLL (mm)	6	0.21 ± 0.54 vs 0.17 ± 0.55	.794
							TLR (%)	12	16 vs 12	>.99
							MACE (%)	12	16 vs 12	>.99
Scheller et al. 202285	Dec 2015 - Feb 2020	50/51 10, Malaysia, Germany, Switzerland	Core lab, CEC	Sirolimus, crystalline coating 4 μg/mm2, SeQuent SCB	Paclitaxel, iopromide 3 μg/mm2, SeQuent Please	DES	LLL (mm)	6	0.25 ± 0.57 vs 0.26 ± 0.60	<.35a
							TLR (%)	12	16 vs 10	.39
							MACE (%)	12	18 vs 14	.60

BMS, bare metal stent; BTHC, butyryl-tri-hexyl citrate; CEC, clinical events committee; CoCr, cobalt-chromium; DES, drug-eluting stent; EES, everolimus-eluting stent; ISR, in-stent restenosis; LLL, late lumen loss; MACE, major adverse cardiovascular events; MLD, minimal lumen diameter; PCB, paclitaxel-cboated balloon; PES, paclitaxel-eluting stent; SCB, sirolimus-coated balloon; TLF, target lesion failure; TLR, target lesion revascularization; TVR, target vessel revascularization;

^aNon-inferiority.

^bPCB vs PES.

^cPEPCAD II, ISAR-DESIRE 3, PEPCAD China ISR, RIBS V, SEDUCE, RIBS IV, TIS, DARE, RESTORE, BIOLUX-RCT.

^dAll-cause death, myocardial infarction, or target lesion thrombosis.

VBT

Vascular brachytherapy inhibits neointimal formation within the stent by delivering radioactive Strontium-90 β-radiation locally, decreasing proliferation. This treatment modality was commonly employed for bare metal ISR 2 decades ago, with minimal long-term benefit demonstrated. The use of VBT has undergone resurgence in use for DES ISR.86, 84, 87 The more overlapping stent layers there are, the less effective brachytherapy is for ISR. With 3 or more stent layers, the 3-year target lesion failure rate exceeds 50%. Accordingly, many centers consider recurrent ISR after failure of 2 DES layers to be an indication for intraventricular block.

IVL

Intravascular lithotripsy is a promising new approach in calcified lesions and has been evaluated in nonrandomized series as a treatment for highly calcified neoatherosclerosis causing ISR.^45,51,88 The mechanism of action is emission of an electrical charge from a pair of lithotripters resulting in generation and collapse of vapor bubbles within a pressurized, fluid-filled semicompliant balloon. This results in acoustic shockwaves that exert an instantaneous field force of up to 50 atm creating fractures within intimal and medial calcium, facilitating subsequent stent expansion as assessed by IVUS and OCT. Whether this device provides a long-term benefit, alone or in combination, in this difficult morphologic subset will require RCTs. IVL has also been combined with VBT.⁵⁰

Management of recurrent ISR

Patients with recurrent ISR are refractory to usual treatment modalities.⁶⁰ The rates of repeat revascularization have been reported to exceed 50% within 2 years. For this reason, new approaches are often tried in these cases, but typically published results are anecdotal reports and uncontrolled series, rather than controlled trials.

Drug-coated balloons may have a beneficial effect in recurrent ISR, but more investigation is needed. Adverse events are significantly higher in patients treated with DCBs with >2 stent layers versus no significant differences in patients with 1 or 2 prior stents. Atheroablative modalities are often employed, but published reports are anecdotal. Multilayer (>2 stents) underexpanded ISR is particularly recalcitrant to standard treatment. VBT is often reserved for refractory ISR and might be considered when multiple layers of stent are present.^50,84,87

Surgical revascularization should be considered in discussion with the patient after several interventional procedures have failed. Although no studies have supported a specific approach as to when repeat procedures should be avoided, Table 6 lists key considerations to assist in deciding which patients might be better treated with coronary artery bypass graft surgery or optimal medical therapy. A heart team approach may be advisable to consider all of the relevant factors in an individual case.

Table 6.

Considerations for CABG in refractory/recurrent in-stent restenosis

•Multivessel CAD especially LM or proximal LAD involvement

•Prior CABG

•Suitability of distal vessel for grafting (including diffuseness of CAD, extent of “metal jacket,” and size of vessel)

•Global and regional LV function including viability (especially the segment subtended by the involved vessel)

•Comorbid conditions (including age, frailty, life expectancy, and activity level)

•Anticipated completeness of revascularization

•Response to optimal medical therapy

CABG, coronary artery bypass grafting; CAD, coronary artery disease; LAD, left anterior descending coronary artery; LM, left main coronary artery; LV, left ventricular.

Stent thrombosis

ST is an acute or subacute thrombotic occlusion that usually presents as an acute MI or acute coronary syndrome and is associated with high rates of morbidity and mortality. These thrombi can be notoriously difficult to treat with traditional interventional techniques because they tend to be large, friable, and adherent.⁸⁹

Incidence and clinical presentation

The overall incidence of ST is 0.5% to 1.0% in the first year and 0.2% to 0.6% in every subsequent year.90, 91, 92 The rate is lower for elective stent placement (0.3%-0.5%) but higher in acute coronary syndrome (3.4%) and MI. In contemporary practice, the observed mortality rate (∼30%) is high, although recent clinical trials and studies requiring autopsy confirmation suggest a better survival, with an average rate of <10%.90, 91, 92 The ST rate is higher in ST-elevation myocardial infarction presentations treated with primary stenting.93, 94, 95

Stent thrombosis causes abrupt closure of the stented artery, and therefore carries a high risk of MI and death. As with any acute vessel closure, the clinical ramifications of ST are influenced by the amount of threatened myocardium, the degree of myocardial viability, the presence and adequacy of collaterals, and the speed and success of revascularization. Approximately 20% of patients with a first ST experience a recurrent ST episode within 2 years.⁹²

Classification

Stent thrombosis is classified by the Academic Research Consortium criteria⁹⁶ based on the presenting clinical scenario and timing after initial stent placement. Timing is classified as acute, subacute, early, late, and very late. ST occurring within 24 hours is acute; 24 hours to 30 days is subacute; from 30 days to 1 year is late; and >1 year is defined as very late. Early ST is defined as occurring within 30 days (ie, acute plus subacute). The clinical presentation defines whether the likelihood of ST is definite, probable, or possible. This categorization accurately portrays ST and is important in investigating its pathophysiologic associations.⁹⁷

Pathogenesis

Numerous clinical and technical risk factors have been associated with ST, as summarized in Table 7.98, 99, 100, 101, 102, 103, 104, 105 Prevention of ST is dependent on optimal stent implantation and the duration and compliance with DAPT. Premature or patient-initiated termination of DAPT, sometimes because of bleeding or perioperative concerns, is responsible for most cases of ST.^100,101,106, 107, 108, 109, 110, 111, 112 Congenital or acquired hyporesponder DAPT status seen with clopidogrel is uncommon with prasugrel or ticagrelor.^107,108 Prior generation stents were susceptible to ST with discontinuation of DAPT out to 5 years and longer in anecdotal cases. With the newest generation of stents, the duration of treatment can be decreased safely to 3 months¹¹³ or 1 month.¹¹⁴

Table 7.

Risk factors for stent thrombosis98, 99, 100, 101, 102, 103, 104, 105

Clinical risk factors	Procedure related	Lesion related	Stent related	Antiplatelet related
•ACS (STEMI/NSTEMI) •Left ventricular dysfunction •Chronic kidney disease •Diabetes mellitus •COVID-19	•Stent length •Stent underexpansion •No reflow •Residual stenosis •Dissection •Multiple stents •Bifurcation stenting	•Necrotic core •Bifurcation lesions •Prior brachytherapy •Multivessel disease •Inflow and outflow obstruction	•Bio-compatible polymers •Polymer/stent thickness •Drug dosage	•Adherence •CYP2C19 polymorphisms •High on-treatment platelet reactivity •Antiplatelet type •Dual antiplatelet therapy duration

ACS, acute coronary syndrome; NSTEMI, non-ST-elevation myocardial infarction; STEMI, ST-elevation myocardial infarction

Circumstances that abet stasis and turbulence, such as underexpanded stents, stents in small vessels, or long lesions, are associated with ST.115, 116, 117, 118, 119, 120, 121 These common mechanical etiologies of ST are associated with stent underexpansion in both early and late ST. A second concern is injury or endothelial disruption caused by edge dissection. The delayed healing and endothelial in-growth observed with early-generation DES are now less common. Early, late, or very late ST can be associated with a calcified nodule, perhaps because of consequent underexpansion.

Neointima containing smooth muscle cells develops beginning 2 weeks after BMS.^106,121 However, DES, depending on the antiproliferative drug and generation, demonstrate delayed endothelial maturation as compared with BMS.¹⁰⁷ When there is an intense neointimal growth reaction to the stent, which is an exaggerated response to arterial healing that peaks at 30 days to 6 months after implantation, vascular narrowing, which can precipitate ST, may occur. Neointimal thickness at stent strut sites is increased when there has been damage to the tunica media. Drug elution and coating polymers may delay this time frame.^{98,103,105,121,122}

Correlates of timing of ST and mechanism

The incidence of Academic Research Consortium definite or probable ST within 2 years is 4.4% and is distributed among the acute, subacute, late, and very late time periods. Each time frame is associated with somewhat different mechanistic circumstances.99, 100, 101, 102, 103^,105,121, 122, 123, 124

Early ST

The strongest predictor of early ST is premature discontinuation of DAPT in the first 30 days following stent implantation.¹²² Mechanical predictors of early ST are similar to those associated with ISR—underexpansion and inflow-outflow problems, including geographic miss, significant residual dissections, and especially occult intramural hematomas—especially at the distal stent edge. Early-stent occlusion is usually the consequence of platelet-rich thrombi forming on an inadequate procedural result. The adequacy of stent expansion and the presence of an occult intimal dissection should be specifically evaluated when the etiology is unclear.

Acute ST

The incidence is 0.2% to 0.6%.

Subacute ST

The incidence is 1.0% to 1.3%.

Late ST

The incidence is 0.4% to 0.6%. The unifying morphologic finding is impaired neointimal healing, defined as delayed development of an endothelialized layer of smooth muscle cells and extracellular matrix that completely covers the stent.¹²³

Several morphologic substrates have been associated with late ST. Stenting across major arterial side branches is well recognized. Overlapping and bifurcation stenting¹²⁵ are prone to underexpansion and malapposition; careful technique and close attention to adjunct imaging are essential. Stent struts that are not apposed to the vessel wall, which can be because of malapposition or late remodeling, produce increased blood flow turbulence and low-flow foci at the margins of the stent. Low-flow velocity increases fibrin and platelet deposition. Intimal dissection and plaque disruption in the arterial segments adjacent to stents can progress and cause flow obstruction. Stenting of lipid-rich plaques with plaque prolapse also increases risk of ST. Lipid-rich plaques with significant necrosis are prone to stent struts penetrating deeply into the lipid core, thus losing contact with the vessel wall. Also, stents placed in a stenosis with large lipid core might delay the development of a fully endothelialized neointima because of scarcity of migrating and proliferating smooth muscle cells in proximity to the stent. Finally, diffuse ISR with thrombosis can occur because of overproduction of intimal hyperplasia and neoatherosclerosis.^103,118, 119, 120, 121

Residual stent edge dissection has been associated with target lesion revascularization.^101,123,124 Edge dissections are defined as being major by OCT when they extend in an arc of >60° and are >3 mm in length. Intimal dissection and plaque disruption in the arterial segments adjacent to stents can progress and cause flow obstruction.

Very late ST

The incidence has been reported between 0.4% to 0.8%. Very late ST (VLST) occurs 1 to 5 years after stent implantation at a rate of 2% per year. The most commonly identified causes of VLST are malapposition, uncovered struts, neoatherosclerosis, and stent underexpansion. The degree and extent of malapposition and uncovered stent struts are the most important correlates of thrombus formation in VLST.^101,124 Intravascular imaging can identify suboptimal stent deployment.

Diagnostic imaging modalities

Intravascular imaging is crucial to developing rational individual case strategy because the causative mechanism is highly influential in selecting treatment strategy to manage the thrombus and prevent its recurrence. Although no imaging modality is rated in the class I category by existing guidelines because of an absence of randomized trials, angiography alone is clearly inadequate because of its intrinsic inadequacy to assess stent expansion, neoatherosclerosis, calcification, and remodeling. IVUS and OCT provide detailed assessment of the stented segment and accurately identify the underlying mechanisms. For these reasons, SCAI recommends that imaging be strongly considered when the etiology of ST is uncertain from clinical and angiographic information.

Angiographic factors

The extent of coronary artery disease is an independent predictor of ST.^105,122 The Dutch Stent Thrombosis Registry included 21,009 consecutive patients and showed that multivessel disease, long lesions, and multiple lesions (total stent length) are independent determinants of ST.¹¹⁷ The volume of thrombus burden is important in appraising the risk of distal embolization, as is the degree of residual flow, presence of collaterals, and status of the microvasculature.53, 54, 55 The size of the jeopardized myocardial segment and the residual left ventricular function are also critical factors.

Intravascular imaging

There is substantial data suggesting that intravascular imaging improves outcomes (see Figures 4A, B and 5). IVUS has been considered the standard imaging modality for ST.^101,102,122 OCT, which has 10-fold higher axial resolution, has more recently shown great promise.^{115,119,126,127} Although the absence of a multicenter, controlled trial limits a formal level of indication,²⁴ these modalities appear to be useful in optimal lesion preparation strategies.

Figure 4.

Mechanisms of stent thrombosis evaluated by intravascular imaging. A′-C′ are optical coherence tomography (OCT) or intravascular ultrasound (IVUS) images corresponding to stent thrombosis seen in the angiographic images A-C (white arrows). A″-C″ are representative diagrams provided to clarify the intracoronary images A′-C′. Top. Mechanisms of stent thrombosis evaluated by OCT (A) Subacute stent thrombosis in which OCT showed a severely underexpanded stent occupied by white thrombus in the mid left anterior descending artery. (B) Very late stent thrombosis occurred 2 hours after noncardiac surgery and after discontinuation of antiplatelet therapy. OCT showed uncovered stent struts occupied by white thrombus. (C) Lipidic plaque (strong signal attenuation) within stent struts indicating neoatherosclerosis resulting in plaque rupture with thrombus. Bottom. Mechanisms of stent thrombosis evaluated by IVUS (A) Subacute stent thrombosis in which IVUS showed a severely underexpanded stent and thrombus in the mid left circumflex artery. (B) Very late stent thrombosis in which IVUS showed a well-expanded stent occupied by thrombus. Because there is no neointimal hyperplasia, it was speculated that uncovered stent struts were the cause of stent thrombosis. (C) Lipidic plaque (strong signal attenuation) appeared within the stent struts indicating neoatherosclerosis resulting in plaque rupture with thrombus.

Figure 5.

Early restenosis due to re-protrusion of a calcified nodule. This patient underwent percutaneous coronary intervention to treat lesions in the distal and mid right coronary artery. Optical coherence tomography (OCT) showed an eruptive calcified nodule (white arrows) in both lesions. A calcified nodule is characterized by an accumulation of small calcium fragments typically with strong signal attenuation due to accompanying and overlying fibrin. The patient came back for staged procedure of LAD (left anterior descending artery) 6 weeks later. OCT showed reprotruding calcified nodules within the stent.

There are several well-defined imaging criteria to optimize stenting (Table 8).115, 116, 117, 118, 119, 120 IVUS can be of great value during stent placement to assess stent sizing, expansion, and apposition. Several studies suggest that IVUS-guided stent placement reduces ST, restenosis, and repeat revascularization.^{101,102,116,118,}128, 129, 130 Stent undersizing occurs frequently when visual estimation alone is used for size selection, and this is a major contributor to the risk of ST (and ISR). The incidence of ST is significantly reduced when IVUS is used to guide stent placement and optimize expansion.¹²⁸ In the Assessment of Dual Antiplatelet Therapy With Drug-Eluting Stents (ADAPT-DES) study,¹²⁹ at 1 year, there was a significant reduction in definite/probable ST (0.52% vs 1.04%, P =.01) and MI (2.5% vs 3.7%, P =.002)¹²⁸ when IVUS guidance was employed.

Table 8.

Intravascular imaging correlates of stent thrombosis^103,105,115, 116, 117, 118, 119, 120

•Small cross-sectional area of <5 mm

•Underexpansion

•Malapposition or incomplete stent apposition

•Correct stent sizing

•Late positive remodeling or aneurysm formation

•Inflow/outflow lesions proximal or distal to the stented segment

•Plaque prolapse or protrusion

•Edge dissection

•Significant residual stenosis

•Stent overlap

•Plaque characteristics: lipid content; delayed or absent endothelialization of stent struts

•Hypersensitivity or inflammatory reactions

•Strut fractures

•Neoatherosclerosis (very late stent thrombosis)

Optical coherence tomography may be superior to IVUS in assessing the cause of ST. In particular, uncovered struts and underexpansion are seen in acute and subacute ST, and neoatherosclerosis, uncovered struts, and malapposition are seen in late and very late ST.^126,131,132 OCT has been shown to be highly effective in determining the underlying cause of VLST in >98% of cases, and multiple mechanisms are usually present (55%). OCT demonstrates malapposition, neoatherosclerosis, uncovered struts, and stent underexpansion underlying ST.¹²⁷ OCT identified adverse features requiring further intervention in 35% of cases, with significantly lower risk of death and MI at 1 year in the CLI-OPCI trial.⁹⁷

In ULTIMATE,¹³⁰ IVUS-guided PCI was associated with important reductions in target vessel failure (cardiac death, MI, or target vessel revascularization). At 3 years, target vessel failure occurred in 47 patients (6.6%) in the IVUS-guided group and in 76 patients (10.7%) in the angiography-guided group (P = .01), driven mainly by the decrease in clinically driven target vessel revascularization (4.5% vs 6.9%; P = .05). Both OPINION¹³³ and ILUMIEN III¹³¹ compared IVUS and OCT and concluded that these methods have similar treatment outcomes.

Mechanical and pharmacologic treatment of ST

Balloon inflation, aspiration thrombectomy and pharmacologic treatment with anticoagulants and antiplatelet drugs are the mainstays of treatment (Table 4).52, 53, 54, 55 Emergent PCI is indicated when the presentation is acute, although optimal reperfusion is achieved in only 2/3.⁸⁹ The SCAI recommended algorithm to organize these approaches is provided in Figure 6.

Figure 6.

SCAI algorithmic approach to stent thrombosis.

Following diagnostic angiography, most ST occlusions can be treated initially with balloon angioplasty alone, sometimes with adjunctive thrombus aspiration when the clot burden is large. Additional stent implantation should ordinarily be limited to significant residual dissections, especially if recent DAPT has been discontinued. High-pressure inflations with noncompliant balloons to assure stent apposition may be necessary in some cases. At this time, no particular stent design or polymer coating has been shown to prevent ST more than others.

Glycoprotein IIb/IIIa antagonists should be considered to improve microvascular reperfusion because of distal embolization, and prolonged infusions up to 72 hours have been successful in anecdotal cases. Prolonged anticoagulation and antiplatelet therapy may be beneficial when residual thrombus is detected following intervention. Compliance and drug resistance should be evaluated in detail. More potent antiplatelet therapy should be considered, including aspirin, prasugrel, or ticagrelor.^107,108,134 If platelet aggregation studies are available and reveal insufficient (<50%) inhibition of platelet aggregation with standard DAPT, the sustained administration of 150 mg/d clopidogrel may be considered. Long-term non-vitamin K oral anticoagulants or warfarin are rarely necessary but may be considered for selected cases of recurrent ST. Although mechanical and extraction thrombectomy have been employed anecdotally, there are no large-scale studies to evaluate their benefit or subgroups of most/least promise.¹³⁵

After flow is re-established and the thrombus is no longer angiographically visible, angiographic and clinical circumstances that reflect distal embolization should be sought and treated. Once the patient is stabilized, the etiology of the stent closure should be pursued. If DAPT has not been interrupted, the stent should be assessed with either IVUS or OCT to determine the adequacy of stent apposition, expansion, and the presence of edge dissections or intramural hematomas. Optimization of stent deployment with appropriate high-pressure balloon expansion should be performed. Additional stent implantation should be avoided if possible because the probability of recurrent ST increases proportionately to the stent length; however, treatment of edge dissections and progression of disease with additional stents are imperative to prevent repeat ST.

Conclusion

Stent failure remains the major drawback to the use of coronary stents as a revascularization strategy. Recent advances in imaging have substantially improved our understanding of the mechanisms underlying both ISR and ST, which have in common numerous clinical risk factors and mechanical elements at the time of stent implantation. SCAI recommends image-guided PCI at the time of initial stent implantation to minimize the occurrence of ISR and ST. When ISR or ST is encountered, imaging should be strongly considered to optimize the subsequent approach to these challenging cases.

Peer review statement

Given his role as associate editor, Sandeep Nathan had no involvement in the peer review of this article and has no access to information regarding its peer review.

Declaration of competing interest

Ziad Ali served on the advisory council and received honoraria and consulting fees from Boston Scientific and Abbott Laboratories. Roxana Mehran is a primary investigator for Abbott. Gary Mintz received honoraria from Boston Scientific. Amir Lotfi, Lloyd Klein, John Messenger, Sunil Rao, Karim Al-Azizi, Yader Sandoval, Sandeep Nathan, Jennifer Rymer, and Akiko Maehara reported no financial interests.

Funding sources

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.