Journal of the American College of Cardiology Vol. 63, No. 14, 2014� 2014 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jacc.2014.01.019

Mechanisms, Pathophysiology, andClinical Aspects of Incomplete Stent Apposition

Guilherme F. Attizzani, MD,*yz Davide Capodanno, MD, PHD,*x Yohei Ohno, MD,*

Corrado Tamburino, MD, PHD*xCatania, Italy; Jundiaí, Brazil; and Cleveland, Ohio

In

From the *Division

Italy; yDivision o

Brazil; zHarringto

Center, Cleveland

Foundation, Catan

Medical, Inc. All o

to the contents of

Manuscript rece

2013, accepted Jan

complete stent apposition (ISA) is characterized by the lack of contact of at least 1 stent strut with the vessel wall ina segment not overlying a side branch; it is more commonly found in drug-eluting stents than bare-metal stents. Theaccurate diagnosis of ISA, initially only possible with intravascular ultrasound, can currently be performed with higheraccuracy by optical coherence tomography, which also enables strut-level assessment due to its higher axialresolution. Different circumstances related both to the index procedure and to vascular healing might influence ISAoccurrence. Although several histopathology and clinical studies linked ISA to stent thrombosis, potential selectionbias precluded definitive conclusions. Initial studies usually performed single time point assessments comparingoverall ISA percentage and magnitude in different groups (i.e., stent type), thus hampering a comprehensiveunderstanding of its relationship with vascular healing. Serial intravascular imaging studies that evaluated vascularresponse heterogeneity recently helped fill this gap. Some particular clinical scenarios such as acute coronarysyndromes, bifurcations, tapered vessels, overlapping stents, and chronic total occlusions might predispose to ISA.Interventional cardiologists should be committed to optimal stent choices and techniques of implantation and useintravascular imaging guidance when appropriate to aim at minimizing acute ISA. In addition, the active searchfor new stent platforms that could accommodate vessel remodeling over time (i.e., self-expandable stents) andfor new polymers and/or eluting drugs that could induce less inflammation (hence, less positive remodeling)could ultimately reduce the occurrence of ISA and its potentially harmful consequences. (J Am Coll Cardiol2014;63:1355–67) ª 2014 by the American College of Cardiology Foundation

Incomplete stent apposition (ISA), also described as stentmalapposition, is a morphological characteristic defined bythe lack of contact between at least 1 stent strut and theunderlying intimal surface of the vessel wall in a segment notoverlying a side branch. Due to the lack of sufficient resolu-tion and the inability of providing cross-sectional images,angiography does not allow the diagnosis of ISA, whereashigh-resolution intravascular imaging modalities, such asintravascular ultrasound (IVUS) (axial resolution of approx-imately 150 mm) and optical coherence tomography (OCT)(axial resolution of approximately 10 to 15 mm) do allowdiagnosis (1). According to the time that ISA is diagnosedrelative to the index procedure, ISA is labeled as acute(i.e., diagnosed post-procedure), persistent (i.e., diagnosedpost-procedure and persistent at follow-up assessment), and

of Cardiology, Ferrarotto Hospital, University of Catania, Catania,

f Interventional Cardiology, Pitangueiras Hospital, Jundiaí, SP,

n Heart and Vascular Institute, University Hospitals, Case Medical

, Ohio; and the xExcellence Through Newest Advances (ETNA)

ia, Italy. Dr. Attizzani has received consultant fees from St. Jude

ther authors have reported that they have no relationships relevant

this paper to disclose.

ived October 21, 2013; revised manuscript received December 12,

uary 7, 2014.

acquired (i.e., not present post-procedure, but identified atfollow-up assessment). In the absence of baseline intravas-cular imaging assessment, it is described simply as late ISA(2,3). Several factors are responsible for this pathologicalphenomenon and its different presentations, as follows:1) inadequate stent implantationdin this setting, there arebasically 2 possibilities: a) marked mismatch between stentsize selection and luminal dimensions (i.e., stent diametersmaller than reference lumen diameter), a situation in which,regardless of optimal stent expansion, ISA will occur (4);or b) stent underexpansion despite an adequate stent–arteryratio due to several different factors, such as inadequatepressure of implantation and/or plaque-related factors(Fig 1); 2) chronic stent recoil (5); 3) thrombus dissolutionafter primary percutaneous coronary intervention (PCI) (6);4) positive vessel remodeling (6–11); and 5) inadequate(i.e., insufficient and/or delayed) neointimal hyperplasia(12–15).

Incomplete Stent Apposition and PotentialRelationship With Adverse Clinical Outcomes

A clear relationship between ISA and adverse cardiovascularevents is still controversial. Although some studies have

Figure 1 ISA in Vessels W

(A) A perfectly rounded vessel (red

(B) The gray-dashed lines correspo

implanted with high pressure, reach

stent is very well and evenly expan

occurred because there is a clear s

3.0-mm stent also implanted with h

diameter reaches 3.1 mm, but the

consequence of calcified plaques (

expansion, therefore, leading to ISA

Craig Skaggs.

Abbreviationsand Acronyms

BMS = bare-metal stent(s)

DES = drug-eluting stent(s)

EES = everolimus-eluting

stent(s)

ISA = incomplete stent

apposition

IVUS = intravascular

ultrasound

OCT = optical coherence

tomography

OLP = overlapping

PCI = percutaneous coronary

intervention

PES = paclitaxel-eluting

stent(s)

ST = stent thrombosis

SES = sirolimus-eluting

stent(s)

ZES = zotarolimus-eluting

stent(s)

Attizzani et al. JACC Vol. 63, No. 14, 2014Incomplete Stent Apposition April 15, 2014:1355–67

1356

failed to demonstrate such con-nections (7,16–18), others havedone so (12,19–24) (Table 1).Lee et al. (25) demonstrated thatamong 30 patients with very latestent thrombosis (ST), 73.9% ofthe drug-eluting stent (DES) groupexhibited ISA (among whom64.7% had late acquired ISA),whereas none of the bare-metalstent (BMS) group had ISA.Although the results showed ahigher incidence of ISA in pa-tients with DESs compared withBMSs, confirming data fromprevious reports (20), they alsohighlighted that ISA is not anisolated factor responsible forDES thrombosis (i.e., 26.1% ofDES patients had stent throm-bosis without evident ISA) (25).Guagliumi et al. (26) demon-strated by combined IVUS and

OCT assessments that the presence and magnitude (i.e.,area and distance) of ISA were significantly higher in pa-tients with DES thrombosis compared with those withoutDES thrombosis. In addition, these investigators implicated1 mechanism that led to ISA (i.e., positive remodeling) andanother mechanism that could be a consequence of ISA(i.e., uncovered stent struts) as independent predictors ofDES thrombosis (Fig. 2). Eosinophilic-rich inflammatoryinfiltrates associated with positive remodeling were also

ith Similar Dimensions

circle) with 3-mm diameter is depicted.

nd to a representation of a 2.5-mm stent

ing 2.7 mm. In this situation, although the

ded, incomplete stent apposition (ISA)

tent–vessel mismatch. (C) Conversely, a

igh pressure is seen, in which the horizontal

vertical diameter is only 2.5 mm as the

yellow with red dots) that impair proper

in some regions of the stent. Figure is by

demonstrated (27). Alfonso et al. (27) also identified ISAin patients with ST using IVUS and OCT (ISA was iden-tified in 40% and 47% of patients, respectively). Kang et al.(28) performed OCT imaging in 33 patients with very lateST and found that in the DES group, 52% and 64% hadISA and thrombi, respectively, whereas no patients pre-sented with ISA in the BMS group. ISA was also morefrequently found in 124 patients with very late definite DESversus BMS thrombosis (52% vs. 16%, respectively,p ¼ 0.005) in a multicenter registry. Notably, greater ISAarea was also demonstrated in the DES group comparedwith the BMS group. Moreover, among the DES group, ahigher prevalence of ISA was identified in sirolimus-elutingstents (SES) compared with paclitaxel-eluting stents (PES)(58% vs. 37%, respectively, p ¼ 0.02) (29). More recently,higher rates of ISA were revealed in 34 patients with DES(56%) and BMS (11%) thrombosis in Italy (30). Althoughextremely insightful, cautious interpretation of these resultsis crucial due to potential selection bias.

Understanding the underlying mechanisms involved inISA is of paramount importance to the interventional cardi-ologist because it enables the identification of which featuresare potentially modifiable and which resources should beutilized during the index intervention to optimize implanta-tion quality andminimize the likelihood of having ISA and itspotentially harmful consequences (31–33). Although physi-ological vascular healing after stent implantation leads toprogressive reduction of ISA over time (12–14,34,35), recentserial OCT assessments elucidated this complex mechanismmore clearly, showing that the greater the acute ISA (i.e., afterthe index procedure), the higher the possibility of its persis-tence at follow-up (14,15). Furthermore, these studiesdemonstrated that acute ISA affects the overall vascularresponse negatively, ultimately leading to delayed stent strutcoverage (12–15) and even to ST (12). The comprehension ofthis complex scenario, moreover, is key for investigators andbiomedical engineers so that they can actively search forconstant improvements in mechanical (i.e., stent platforms)(36) and biological (i.e., antirestenotic drugs, polymers) (37)components of coronary artery scaffolds.

Histopathology Investigations

Histopathology studies have been instrumental in enabling abetter understanding of mechanisms that lead to ISA. Thepivotal demonstration by Virmani et al. (37) that Cypher(Cordis, Johnson & Johnson, New Brunswick, New Jersey)SES thrombosis was linked to hypersensitivity reactions andto a complex proinflammatory milieu raised the initialconcerns regarding the safety issues of DESs (37). The in-vestigators identified the presence of ISA with thick layersof fibrin thrombus separating stent struts from the vesselwall. In a larger series of patients, Joner et al. (38) ratifiedthe complexity of the multifactorial scenario (i.e., proceduraland clinical related features) associated with DES failure,describing the presence of ISA coupled with delayed arterial

Table 1 Impact of ISA on Cardiovascular Events

First Author (Ref. #) Design NClinical

Presentation Type of StentAssessed

byFollow-Up(Months) Assessment of Malapposition

Positive Association BetweenISA and Cardiovascular Events

Guagliumi et al. (22) PP, SC 21 Stable/ACS ZES OCT 6 Malapposed struts per stent: 2.0 � 4.2% (OLP),3.8 � 7.1% (non-OLP)

No

Kubo et al. (23) PP, SC 45 Stable/ACS SES OCT 9 Malapposed struts per stent: 0.2 � 1.2% (stable),4.0 � 6.7% (ACS)

No

Guagliumi et al. (24) PP, SC 77 Stable/ACS SES/PES/ZES/BMS OCT 6 Malapposed struts per stent: 2.9 � 7.0% (SES; OLP),1.9 � 4.2% (SES; non-OLP),malapposed struts per stent: 5.5 � 15.6%(PES; OLP), 0.7 � 2.1% (PES; non-OLP),malapposed struts per stent: 0.1 � 0.2%(ZES; OLP), 0.1 � 0.1% (ZES; non-OLP),malapposed struts per stent: 0.1 � 0.1%(BMS; OLP), 0.2 � 0.9% (BMS; non-OLP)

No

Guagliumi et al. (22) PP, SC 42 Stable/ACS EES CoCr/EES PtCr OCT 6 Malapposed struts per stent: 0.5%, malappositionarea 0.02 mm2 (EES CoCr), malapposed strutsper stent: 0.0%, malapposition area 0.01 mm2

(EES CoCr)

No

Hong et al. (16) RP, SC 557 Stable/ACS SES/PES IVUS 6 LSM CSA: 3.0 � 1.9 mm2 (SES), 3.2 � 1.9 mm2

(PES) (LSM occurred in 12.1% of lesions)No

Steinberg et al. (17) PP, MC 1,580 Stable/ACS PES/BMS IVUS 9 Late-acquired ISA was identified in 2.7% of BMS,3.1% of PES-SR, and 15.4% of PES-MR

No

Tanabe et al. (18) PP, MC 469 Stable/ACS PES/BMS IVUS 6 Mean ISA area: 3.6 � 1.2 mm2 (PES-SR),2.1 � 1.4 mm2 (PES-MR), 3.0 � 2.1 mm2 (BMS)

No

Cook et al. (19) PP, MC 194 Stable/ACS SES/PES IVUS 8 Max ISA CSA: 4.5 � 5.0 mm2 (SES), 5.0 � 5.1 mm2

(PES), total ISA length: 2.3 � 1.4 mm (SES),1.0 � 0.9 mm (PES)

Yes (the presence of ISA afterDES implantation was associatedwith a higher rate of MI and VLST)

Cook et al. (21) PP, SC 188 Stable/ACS SES/PES IVUS 8 Max ISA CSA: 8.3 � 7.5 mm2 (VLST), 4.0 � 3.8 mm2

(control), max ISA length: 6.3 � 6.3 mm2 (VLST),1.5 � 1.0 mm2 (control)

Yes (ISA is highly prevalent in patientswith VLST after DES implantation)

ACS ¼ acute coronary syndrome(s); BMS¼ bare-metal stent(s); CoCr ¼ cobalt-chromium; CSA ¼ cross-sectional area; EES¼ everolimus-eluting stent(s); ISA¼ incomplete stent apposition; IVUS¼ intravascular ultrasound; LSM ¼ late stent malapposition; MACE¼major adversecardiac event(s); MC ¼ multicenter; MI ¼ myocardial infarction; MR ¼ moderate-release; OCT ¼ optical coherence tomography; OLP ¼ overlapping; PCI ¼ percutaneous coronary intervention; PES ¼ paclitaxel-eluting stent; PP ¼ prospective; PtCr ¼ platinum-chromium;RP ¼ retrospective; SC ¼ single center; SES ¼ sirolimus-eluting stent(s); SR ¼ slow-release; VLST ¼ very late stent thrombosis; ZES ¼ zotarolimus-eluting stent(s).

JACCVol.63,No.14,2014

Attizzaniet

al.April15,2014:1355–67

Incomplete

StentApposition

1357

Figure 2 Late ISA, Thrombosis, and Neoatherosclerosis

This 60-year-old patient had multiple drug-eluting stents implanted in overlapping fashion in the right coronary artery (RCA) 6 years before the current hospitalization due to

non–ST-segment elevation myocardial infarction. (A) Angiography demonstrates “haziness” in the mid-RCA (white arrow). The white dashed lines in the longitudinal optical

coherence tomographic (OCT) image (bottom) correspond to the respective locations of the cross sections identified by the same letter. (B) A normal pattern of thin neointimal

proliferation is depicted without incomplete stent apposition (ISA) or uncovered struts. (C, D, E). However, ISA with some attached thrombus is depicted (white arrows). (F) A

huge intraluminal thrombus burden is depicted (white arrow), whereas in (G) another normal pattern (thin, bright tissue covering the struts) of vascular response is depicted.

(H) A transition to an abnormal pattern of stent coverage is shown as low-signal intensity tissue, identified as peri-struts (i.e., acellular material, such as fibrin or proteoglycan).

(I) Although normal tissue covers the struts from 1 to 7 o’clock, the white dashed square in J demonstrates lipidic tissue (marked light attenuation with poorly defined borders,

white asterisk) covering the stent (i.e., neoatherosclerosis), which almost “hides” the stent struts (white arrows), and a thin cap overlying the lipid plaque (white arrowheads).

This case has important teaching points, as follows: 1) In a long stented segment, there is minimal ISA, but it is linked to stent thrombosis; therefore, it is important to perform

lesion-level assessments rather than display only an overall percentage of the ISA and the assessment of ISA clusters in clinical trials; 2) Although a clear relationship between

the acellular tissue peri-strut (i.e., potentially secondary to inflammation drug and/or polymer toxicity) identified in H and the ISA (i.e., eventually secondary to positive

remodeling due to drug and/or polymer toxicity) cannot be clearly derived from this case, one can speculate that these pathological phenomena could be linked by a proin-

flammatory milieu (35); and 3) Similar rationale of item 2 can be derived from the lipidic neointimal tissue covered by thin fibrous cap identified more proximally (I and J).

Attizzani et al. JACC Vol. 63, No. 14, 2014Incomplete Stent Apposition April 15, 2014:1355–67

1358

healing and the presence of fibrin surrounding the stentstruts as 1 of the mechanisms associated with ST (35)(Fig. 2). More recently, Cook et al. (39) correlated thehistopathology of thrombus aspirates with IVUS findingsin patients with very late DES thrombosis. Interestingly,they were able to establish a correlation between the am-ount of eosinophil infiltrates with the extent of positivevessel remodeling, suggesting hypersensitivity reactions ofall 3 layers of the vessel wall in patients with ISA; in pa-tients without positive remodeling, however, only a few orno eosinophils were detected. Despite the outstandingcontribution of histopathology research toward a bettercomprehension of the mechanisms that lead to ISA and itspotential consequences, the data presented by these studiesshould be carefully interpreted because they represent ahighly selected population (i.e., patients who had a DESimplanted and died). Therefore, although ISA is suggested

as 1 of the potential mechanisms linked with adverse car-diovascular events in these studies, a linear cause and effectrelationship cannot be derived from these findings.

Another significant role played by histopathology in thisscenario is the validation of intravascular imaging findings,which is of utmost importance because they enable accuratein vivo qualitative (rather than merely quantitative) assess-ments of vascular responses in a prospective fashion, whichcould ultimately facilitate the understanding of the patho-physiology of ISA. The feasibility and accuracy of tissuecharacterization utilizing OCT has been previously demon-strated (40–42). This could be particularly important to helpelucidate, for example, the underlying mechanisms that leadto ISA in different scaffolds (Fig. 3) (43). Radu et al. (44)suggested the term “pseudoapposed” to define stent strutsthat were “apposed” to low-signal intensity regions, whichwere likely composed of acellular material such as fibrin or

JACC Vol. 63, No. 14, 2014 Attizzani et al.April 15, 2014:1355–67 Incomplete Stent Apposition

1359

proteoglycan, and similar to what has been described byhistopathology (38). Further investigations are required toassess the relevance of these findings.

Fundamental Differences BetweenIVUS and OCT in ISA Interpretation

Due to their high resolution, intravascular imaging modal-ities are essential to properly diagnose ISA in vivo. Althougha complete description of IVUS and OCT imaging is beyondthe scope of this paper, it is important to be cognizant of thefundamental differences and potential additive values of theseimaging systems in the diagnosis of ISA, hence, facilitatingdata interpretation.

IVUS is a sound-based imaging system with an axialresolution of approximately 150 mm and a penetration of 4to 8 mm. It enables detailed assessment of vessel size becausethe external elastic membrane is clearly visualized, thereforeallowing accurate quantification of positive vascular remod-eling (45,46). The diagnosis of ISA by IVUS is determinedby the presence of blood speckle (or also by color Dopplerimaging in the case of lower frequency IVUS) behind thestent struts not overlying a side branch. Data are providedin cross-sectional (i.e., area of ISA) and longitudinal (i.e.,volume of ISA) levels (6). Conversely, OCT is a near-infrared, light-based system with an approximately 10-foldhigher resolution (approximately 10- to 15-mm axial reso-lution) compared with IVUS, but with less tissue penetra-tion (1 to 3 mm). This lack of deep tissue penetrationprecludes accurate quantification of positive vessel remod-eling, whereas the blood-free environment obtained duringOCT image acquisition, coupled with its higher axial reso-lution, provides a sharper delineation of the stent–lumeninterface compared with IVUS, thus allowing easier imageinterpretation (1,47) (Fig. 4). Several authors demonstratedthat OCT is superior to IVUS in detecting ISA, includingin the left main coronary artery (1,12,27,48). Anotherimportant piece of information added by OCT on top ofcross-sectional and longitudinal levels of assessment is thestrut-level analysis, which enables the diagnosis and quan-tification of ISA in each individual strut. It is also importantto highlight that the initial diagnosis of ISA by OCT,including that performed in the catheterization laboratory inour daily practice, is qualitative, whereas the final determi-nation of whether or not 1 strut exhibits ISA is quantitative(i.e., considering blooming, strut, and polymer thicknesses)(Fig. 5) (49,50). Taken together, these characteristicsindicate that IVUS and OCT complement each other in thediagnosis and elucidation of ISA mechanisms (26,27).

Study Designs and Endpoints

Several studies have evaluated ISA by single time point (i.e.,follow-up) assessments (51–55). Although the informationprovided by these studies is extremely valuable in the deter-mination of ISA rates using different stents in vivo, it is

impossible to completely elucidate this morphological featurein a single “snapshot.” Healing after stent implantation is acontinuum influenced by several different factors, related toboth baseline characteristics (i.e., clinical and procedurerelated) and to alterations that occur along this complexbiological process (i.e., inflammation, delayed healing, andpositive remodeling); therefore, serial intravascular imagingstudies using IVUS and/or OCT (11,12,14,15,35,56) havehelped to fill the gap left by single time point assessments.Despite the importance of serial invasive imaging assess-ments to comprehensively characterize the natural historyof ISA, increased costs, coupled with ethical issues of sub-mitting patients to the inherent risks of repeated invasiveinterventions, x-ray exposure, and contrast media injections,have hampered their more extensive utilization (57,58).

Another consideration is that data presentation acrossstudies evaluating ISA lacks standardization. The majorityof trials provide overall percentages, areas, and volumes ofISA associated with the groups (i.e., different stents) beingevaluated (59,60), rather than identifying and quantifyingsegments in which the abnormality is present (i.e., lesion-level assessment) (61), therefore restricting the assessmentof intra- and interindividual heterogeneity of vascularresponses, and consequently, of ISA. Räber et al. (62)demonstrated that, despite an overall equivalent percentageof ISA late after SES and PES implantation, significantclustering of ISA was depicted in the former, suggestingdifferent patterns of vascular responses between the scaffoldsbeing evaluated. Gutiérrez-Chico et al. (15) described het-erogeneous morphological healing patterns in patients withacute ISA using serial OCT imaging assessments. Anothertool to assess whether heterogeneity exists in vascular re-sponses is the coefficient of variation, which is obtainedby dividing each lesion into subsegments (i.e., dividingthe standard deviation by the mean percentage of ISA)(35,63–65). These analytical approaches could ultimatelyexpand our knowledge of assessing pathological vascularreactions to stents, particularly ISA, and avoiding, forinstance, the “diluting” of the results of clustered ISAs thatmay occur in localized segments of certain individuals,and in the entire population of the study. Whether thesefindings significantly affect clinical outcomes remains to bedemonstrated.

Incomplete Stent Apposition inDifferent Clinical Scenarios

Acute coronary syndromes. Nakazawa et al. (66) usedhistopathology to demonstrate that vascular healing wasdelayed and ST rates were increased in patients who under-went DES implantation for myocardial infarction comparedwith stable angina; culprit sites in the former had greaterfibrin deposition and inflammation, together with less neo-intimal proliferation. In addition, higher rates of strutspenetrating the necrotic core and greater external elasticmembrane areas were found. Hong et al. (67) showed in vivo

Figure 3 Representative Images of Late Stent Thrombosis in First-Generation SES and ISA in PES

(A) Histological sections from sirolimus-eluting stents (SES). A 40-year-old woman who received 2 SES in the left anterior descending artery (LAD) and right coronary artery (RCA)

17 months ante mortem died suddenly 4 days after surgical removal of a melanoma (wide excision). Antiplatelet therapy (aspirin and clopidogrel) was discontinued 5 days before

the surgery. Histological sections of the SES in the LAD showed total thrombotic occlusion and diffuse inflammation (a). Numerous inflammatory cells were observed within

the neointimal area (b). Inflammatory reaction predominantly consists of T lymphocytes (c) (CD45RO) and eosinophils (d) (Luna stain). Note that the same reaction was

observed in the SES in the RCA (e), and severe inflammation resulted in incomplete stent apposition (ISA) of stent struts (f). (B) Histological sections from a PES showing ISA. A

69-year-old man who received a paclitaxel-eluting stent (PES) in a saphenous vein graft died suddenly 3 months after stent placement. Histological sections showed thrombotic

occlusion in the PES (a, b); note the ISA was secondary to severe fibrin deposition (c). A 48-year-old man with a PES implant in the proximal LAD died suddenly at 40 months.

Histological sections showed thrombotic occlusion of the PES (d). Most struts show ISA with fibrin deposition underneath the stent struts (e, f). Thr ¼ thrombus. Adapted and

reprinted with permission from Nakazawa et al. (43).

Attizzani et al. JACC Vol. 63, No. 14, 2014Incomplete Stent Apposition April 15, 2014:1355–67

1360

that BMS implantation during primary PCI was an inde-pendent predictor of late ISA, but it was not linked toincreased rates of adverse events at 3-year follow-up. In aserial assessment by IVUS, BMS and PES rates of acute ISAwere equivalent, whereas PES were more frequently associ-ated with late acquired ISA at follow-up, mostly becauseof thrombus dissolution and positive remodeling. The in-vestigators also found that an area of acute ISA >1.2 mm2

determined resolved ISA from persistent ISA (6). Likewise,no differences were observed between SESs and BMSs

regarding acute ISA rates after primary PCI (38.5% and33.8%, p ¼ 0.51), whereas late acquired ISA was more oftenidentified in SES, mainly secondary to positive vesselremodeling (84% of the cases) (68). Data obtained from theHORIZONS-AMI (Harmonizing Outcomes With Revas-cularization and Stents in AcuteMyocardial Infarction) studydid not link early stent thrombosis with the presence of acuteISA (69); in addition, the 13-month follow-up of theHORIZONS-OCT study included 188 stents (PES andBMS) and showed higher rates of late ISA with PES

Figure 4 ISA by IVUS and OCT

In A-I, an intravascular ultrasound (IVUS) cross-sectional image shows stent struts (red arrows) with ISA (yellow asterisks). The same co-registered region of interest is

demonstrated in an OCT image (A-II), in which one can also identify the stent struts (red arrows) with ISA (yellow asterisks). In B-I, however, the IVUS image does not enable a

clear diagnosis of ISA, whereas in the same co-registered region (B-II) imaged by OCT, the presence of stent struts (red arrow) with ISA (yellow asterisks) is clearly depicted.

The blue lines in the longitudinal views located in the bottom of each cross section demonstrate the region of the pullback from which the cross sections were obtained.

Abbreviations as in Figures 2 and 3.

JACC Vol. 63, No. 14, 2014 Attizzani et al.April 15, 2014:1355–67 Incomplete Stent Apposition

1361

compared with BMS (70). Gonzalo et al. (71) used OCTimaging to demonstrate, in vivo, that DESs implanted duringprimary PCI compared with stable and/or unstable anginamore often had ISA at 9-month follow-up (75% vs. 25.8%,respectively, p ¼ 0.001). The investigators also identifiedDES implantation during primary PCI as the only inde-pendent predictor of ISA (odds ratio: 9.8, 95% confidenceinterval: 2.4 to 40.4; p¼ 0.002). Davlouros et al. (72) showedin a small study that ISA was correlated with a lack ofcoverage 6 months after PES implantation for acute coronarysyndromes. Kubo et al. (23) compared unstable versus stableangina patients who underwent DES implantation and

Figure 5 Quantification of ISA by OCT

In daily clinical practice, the initial assessment of ISA by OCT is qualitative, as demonstrate

several factors such as polymer and strut thicknesses (different for each stent) and one-h

malapposed distance. ISA quantification of some struts are demonstrated by the purple sq

strut (blue arrow), which although seemed to be ISA by qualitative assessment, was actu

whereas another strut was demonstrated to be malapposed (green square, malapposed d

demonstrated significantly higher rates of both acute (67%vs. 32%; p ¼ 0.038) and late ISA (33% vs. 4%, respectively;p ¼ 0.012) in the former setting; no adverse events werereported in the 55 patients in this study at 9-month follow-up. Adrenergic vasoconstriction, which leads to underesti-mation of the reference vessel size, which then leads to anequivocal stent choice (i.e., smaller than vessel dimensions), isanother potential cause of ISA after acute myocardialinfarction (73). The European Guidelines on ST-segmentelevation myocardial infarction recommend intracoronaryadministration of nitrates for more accurate stent sizingduring primary PCI (74).

d in A; however, the accurate determination of ISA is ultimately quantitative, because

alf of the blooming should be taken into account for the determination of the final

uares in B, whereas the region highlighted by the white square in C demonstrates a

ally properly apposed (i.e., in contact with the lumen boundary, blue dashed line),

istance calculation on the right side of the figure). Abbreviations as in Figure 2.

Attizzani et al. JACC Vol. 63, No. 14, 2014Incomplete Stent Apposition April 15, 2014:1355–67

1362

Overlapping stents. In overlapping (OLP) stent regions, agreater amount of metal and increased concentrations ofeluting drugs or polymers might increase the risk of flowdisruption, inflammation, fibrin deposition, and thromboticevents. Finn et al. (75) performed histopathological evaluationof OLP stents in animal models and showed that althoughboth first-generation SES and PES promoted delayed arterialhealing and inflammation at OLP sites compared with BMS,PES had greater fibrin deposition and medial cell loss, as wellas a greater external elastic membrane area compared withboth SES and BMS. In concordance, immature neointimawas found by Shinke et al. (76) in OLP PES regions;furthermore, these investigators used angioscopy to identifymore thrombus in PES compared with BMS OLP regions.Recently, an in vitro assessment linked BMS deployed in ISAor OLP fashions with higher thrombogenicity compared withwell-apposed, length-matched controls (1.58-fold, p< 0.001;and 2.32-fold, p < 0.001). The investigators also demon-strated that thin struts were less prone to thrombosiscompared with thick-strut stents (77). The concerns raised byhistopathology and in vitro data were amplified by in vivoOCT evaluation performed by Tanigawa et al. (78), whoshowed significantly higher rates of acute ISA in OLP versusnon-OLP regions despite aggressive stent optimization,and by Guagliumi et al. (24), who demonstrated higher ratesof late ISA inOLP segments of SES and PES compared withzotarolimus-eluting stents (ZES) and BMS. In the samestudy, numerically higher, although not statistically signifi-cant, rates of ISA were found in OLP compared withnon-OLP PES, whereas no differences were shown in theSES, ZES, and BMS groups (24). Aoki et al. (79) did nothighlight concerns regarding ISA in OLP PES segmentsat 9-month follow-up. Meanwhile, Otake et al. (80) de-monstrated that despite the vessel size increase after OLPPES, no significant ISA was demonstrated; likewise, nosignificant ISA was demonstrated in the everolimus-elutingstent (EES) group in the same retrospective IVUS study at9-month follow-up. In addition, Dawkins et al. (81) did notshow increased ISA in OLP segments of PES. An interestinginsight in this setting was recently demonstrated byGuagliumi et al. (22); similar stent platforms and anti-restenotic drugs (i.e., zotarolimus), but different polymers andrelease kinetics, led to different rates of ISA in OLP regions(i.e., slow-release ZES showed higher rates of ISA comparedwith fast-release ZES). Different stent platforms (i.e., cobalt-chromium vs. platinum-chromium), but same drug-releasekinetics, in EES did not reveal differences in terms ofISA (82). The importance of the underlying plaque inthe results of OLP stenting was investigated by Tahara et al.(83), who demonstrated that increased rates of ISA anduncovered struts were present when OLP stenting was per-formed in normal-appearing arterial segments, whereasrestenosis occurred exclusively in segments with previous high-grade stenosis.Chronic total occlusions. Several different mechanismsmight be implicated in ISA after chronic total occlusion

PCI. Stent–vessel mismatch, which occurs because theoccluded artery can appear smaller than its actual size dueto chronic hypoperfusion, could be responsible for acuteISA, whereas injury to the adventitial layer caused by theguidewire, creation of a false lumen, or subintimal stentingcould contribute to late acquired or persistent ISA. Honget al. (16) identified chronic total occlusion PCI as an in-dependent predictor of DES ISA. In the same study, theincidence of ISA in this scenario was 27.5%.Bifurcations. Coronary artery bifurcations are complexstructures in which stent–vessel interactions are dependent onseveral variables, such as stent strut location (i.e., flow divideror lateral wall) (84), side branch angulation, plaque distri-bution, or the technique used for stent implantation (i.e., 1 or2 stents) or post-dilation (i.e., kissing-balloon or separatemain vessel and side branch dilations). The difficulty ofachieving adequate stent apposition utilizing 2 stent tech-niques in coronary artery bifurcations was demonstratedin vitro by Murasato et al. (85) and Ormiston et al. (86), andtheir results were consistent in an in vivo IVUS study thatrevealed that more than 60% of the ISA was proximal tothe carina in lesions treated with the “crush” technique (87).In line with these results, Foin et al. (88) showed, in vitro, that“crush” stenting led to higher rates of ISA compared with theCulotte and T with protrusion technique. Tyczynski et al.(89) used OCT imaging to demonstrate higher rates of ISAin a side branch compared with non-side branch segmentsafter bifurcation stenting. Interestingly, no differences be-tween simple and complex techniques were revealed (89).

Several interventions have been proposed aimed at opti-mizing stent implantation in bifurcations. Distal cell guide-wire re-crossing to the side branch led to a reduction of ISAboth in vitro and in vivo (90,91). In the latter, OCT-orientedguidewire re-crossing to the side branch significantly reducedrates of ISA compared with angiography-oriented pro-cedures (9.5% vs. 42.3%, respectively; p < 0.0001). Foinet al. (92) compared the use of a final kissing balloon in vitrowith a 2-step (main branch and side branch) final balloondilation technique in bifurcations. Although the investigatorsdemonstrated equivalent rates of ISA in the side branchostium region for both techniques, they depicted higherrates of ISA at the proximal stent edge with the kissingballoon (30.7� 26.4% vs. 2.8� 9.6% for 2 step, p¼ 0.0016)(92), which could be reduced by performing an additionalballoon dilation in the proximal stent segment (33.4% vs.0.6%, respectively, for nonproximal post-dilation and prox-imal post-dilation; p ¼ 0.02) (93). More recently, Rahmanet al. (94) demonstrated that although kissing-balloon in-flations led to some amount of proximal main branch stentasymmetry, the improved stent expansion secondary to thatmaneuver was sufficient to adequately appose stent struts inthe majority of cases. Di Mario et al. (95) proposed thatstent optimization in bifurcations could be maximized byhigh-resolution intravascular imaging guidance. A large areaof ISA, which was not suspected after angiography-guidedstent optimization with a kissing balloon, was seen by

JACC Vol. 63, No. 14, 2014 Attizzani et al.April 15, 2014:1355–67 Incomplete Stent Apposition

1363

3-dimensional OCT imaging in a case of left main coronaryartery distal bifurcation intervention recently described byFujino et al (96). Among 403 patients with IVUS-guidedSES implantation in the left main coronary artery(including distal bifurcation), 28 presented with acute ISA;however, no correlations between this pathological phe-nomenon and adverse cardiovascular events at 2-year clinicalfollow-up were established (97). So far, however, no clinicalbenefits have been found with the use of routine intravascularimaging guidance during bifurcation stenting; therefore,further research in this field is mandatory (98).Vessel tapering. Vessel tapering is frequently observed inlong coronary lesions, which might lead to ISA due tomarked differences in vessel geometry in the same regionof interest. Although multiple post-dilations with differentballoon sizes is the usual strategy used with the aim ofobtaining adequate apposition when stenting distinctvessel dimensions, aggressive post-dilation with balloonsdisproportionately larger than the nominal stent diameter canlead to stent strut deformation (99) and eventual polymerdamage (100,101). This could negatively affect vascular re-sponses. Carrozza et al. (102) described increased stent recoilwith larger acute expansion diameters in which excessiveDES overstretching left important gaps between the strutcells, which could induce suboptimal antirestenotic drugdelivery (103). Importantly, Russo et al. (104) used an animalto demonstrate that vessel overexpansion might be counter-productive because it leads to more intense neointimal pro-liferation. Therefore, reaching equipoise between adequate(but not excessive) stent expansion and absence of ISA while

Figure 6 STEMI in a Tapered Vessel Treated With Self-Expandable S

(A) An ostial occlusion in the left anterior descending artery (LAD) during ST-segment my

observed in the mid LAD (white arrows). Final result after implantation of 2 STENTYS nitino

in (C). White arrows identified by letters have their OCT assessments represented with t

whereas distal and proximal references are shown in D and G.

avoiding vessel injury is a difficult goal to accomplish intapered coronary arteries.

Potential Alternatives to ReduceISA in Daily Clinical Practice

As previously described, ISA is multifactorial; features directlyrelated to the index intervention and others associated withvascular response to stents are implicated in its pathophysi-ology. Therefore, it is likely that combined, rather than isolatedapproaches, are necessary to successfully reduce ISA rates.Although a clear relationship between ISA and adverse car-diovascular events is still controversial, interventional cardiol-ogists must be committed to minimizing acute ISA byoptimizing stent choice and techniques of implantation. Somenovel devices could be helpful in addition to conventionalstrategies in this setting. Verheye et al. (105) demonstratedpromising results with the use of self-expandable nitinol stents(STENTYS, STENTYS SA, Paris, France) in coronarybifurcations. Interestingly, these devices (either BMS or PES)have small interconnections that enable easy provisionalstenting and access to the side branch. Progressive stentexpansionwas demonstrated by increasing stent area over time,which could potentially reduce rates of persistent and late ac-quired ISA because the stent conforms to vascular remodeling.Minimal ISA was revealed at 6-month IVUS follow-up. Thissame device was tested during primary PCI and demonstratedan early reduction of ISA compared with balloon-expandablestents (106). Self-expandable stents could also help reduceISA in tapered lesions because the same stent could properly

tents

ocardial infarction (STEMI). After flow restoration (B), marked vessel tapering is

l self-expandable stents in the mid and proximal LAD in overlapping fashion is shown

he corresponding letters in E and F; a marked discrepancy in stent areas is shown,

Attizzani et al. JACC Vol. 63, No. 14, 2014Incomplete Stent Apposition April 15, 2014:1355–67

1364

accommodate different vessel dimensions along the lesion(Fig. 6), thereby avoiding complications (e.g., dissections andvessel perforation) in smaller segments while allowing optimalstent apposition in larger segments. Further investigations arenecessary to establish whether self-expandable stents couldreduce adverse clinical outcomes compared with balloon-expandable stents in all of these clinical settings.

More recently, everolimus-eluting bioresorbable vascularscaffolds (Absorb, Abbott Vascular, Santa Clara, California)emerged as a new tool in the armamentarium against coronaryartery disease. These novel devices demonstrated promisingresults in randomized studies (107) and are being progressivelyincorporated into routine clinical practice in Europe, theMiddle East, and Asia (108–110). Bioresorbable vascularscaffolds are more conformable than metallic stents (111), andtheir struts, including those associated with ISA, arecompletely resorbed 2 to 3 years after implantation. Thisfeature could potentially eliminate concerns related to verylong-term metallic exposure to the coronary circulation incases of delayed healing (i.e., ISA and noncoverage) (112,113).Whether the implantation of these novel devices can reduceadverse events related to late ISA compared with metallicstents in real-world populations warrants further investigation.

Another critical step in reducing late ISA, which is notdirectly procedure related, is the active search for polymersand antirestenotic drugs that continue to inhibit neointimalhyperplasia and induce less toxicity and inflammatory re-actions in the vascular wall, resulting in less positive vascularremodeling. This “perfect equilibrium” between drug and/orpolymer combinations has been pursued thoroughly by in-vestigators over recent years in both pre-clinical and clinicaltrials with promising results in terms of reducing ISA(32,37,114–116).

Table 2 Summary of Conclusions

ISA is a multifactorial, morphological feature that is affected byprocedure-related elements and by stent-vessel interactions over time.

ISA is more commonly identified in DES than in BMS.

OCT is superior compared with IVUS in diagnosing ISA.

Although an association between ISA and adverse clinical events is stronglysuggested by histopathology and in vivo studies, highly selected populationsmay introduce bias to these findings; therefore, ISA is the most likely1 feature among several others that might predispose to ST.

The magnitude of acute ISA may predict its resolution and impact onvascular healing.

Standardization in data reporting and lesion-level assessment with the evaluationof clustering and heterogeneity could ultimately lead to a better understandingof ISA and its impact on clinical outcomes.

Particular clinical scenarios such as acute coronary syndromes, OLP stents,bifurcations, vessel tapering, and CTO are more prone to exhibit ISA.

Adequate stent sizing, optimal techniques of implantation, andintravascular imaging guidance can effectively reduce ISA rates.

Novel scaffolds such as self-expandable stents could play a role in reducing ISAbecause they can accommodate eventual positive vessel remodeling over time.

Finding antirestenotic drugs and/or polymers that cause less inflammation,and therefore, less positive remodeling, is another important aspect aimedat reducing ISA.

CTO ¼ chronic total occlusion, DES ¼ drug-eluting stent(s); other abbreviations as in Table 1.

Conclusions

The main highlights of this comprehensive review aresummarized in Table 2. ISA is a multifactorial, pathologicalfeature, and although a clear cause-effect relationship withST is still unclear, a body of data suggests that it mightbe one of the contributors in the prothrombotic vascularmilieu. Optimal stent sizing and techniques of implantation,together with the use of new technologies, could minimizethe occurrence of ISA.

AcknowledgmentsThe authors thank the following experts for their valuablecontribution to our paper: Gaku Nakazawa, MD, andTsutomu Saiki, BMET, from the Tokai University Schoolof Medicine, Kanagawa, Japan, and Yusuke Fujino, MD,from the New Tokyo Hospital, Chiba, Japan.

Reprints and correspondence: Dr. Davide Capodanno, Depart-ment of Cardiology, Ferrarotto Hospital, University of Catania, ViaCitelli 1, 95100 Catania, Italy. E-mail: dcapodanno@gmail.com.

REFERENCES

1. Bezerra HG, Attizzani GF, Sirbu V, et al. Optical coherence to-mography versus intravascular ultrasound to evaluate coronary arterydisease and percutaneous coronary intervention. J Am Coll CardiolIntv 2013;6:228–36.

2. Takano M, Inami S, Jang IK, et al. Evaluation by optical coherencetomography of neointimal coverage of sirolimus-eluting stent threemonths after implantation. Am J Cardiol 2007;99:1033–8.

3. Fujino Y, Attizzani GF, Nakamura S, Costa MA, Bezerra HG.Coronary artery aneurysms after sirolimus-eluting stent implantation:multimodality imaging evaluation. J Am Coll Cardiol Intv 2013;6:423–4.

4. Kume T, Waseda K, Ako J, et al. Intravascular ultrasound assessmentof postprocedural incomplete stent apposition. J Invasive Cardiol2012;24:13–6.

5. Fujino Y, Attizzani GF, Nakamura S, Costa MA, Bezerra HG.Frequency-domain optical coherence tomography assessment ofstent constriction 9 months after sirolimus-eluting stent implanta-tion in a highly calcified plaque. J Am Coll Cardiol Intv 2013;6:204–5.

6. Guo N, Maehara A, Mintz GS, et al. Incidence, mechanisms, pre-dictors, and clinical impact of acute and late stent malapposition afterprimary intervention in patients with acute myocardial infarction: anintravascular ultrasound substudy of the Harmonizing Outcomes withRevascularization and Stents in Acute Myocardial Infarction (HO-RIZONS-AMI) trial. Circulation 2010;122:1077–84.

7. Ako J, Morino Y, Honda Y, et al. Late incomplete stent appositionafter sirolimus-eluting stent implantation: a serial intravascular ultra-sound analysis. J Am Coll Cardiol 2005;46:1002–5.

8. Shah VM, Mintz GS, Apple S, Weissman NJ. Background incidenceof late malapposition after bare-metal stent implantation. Circulation2002;106:1753–5.

9. Feres F, Costa JR Jr., Abizaid A. Very late thrombosis after drug-eluting stents. Catheter Cardiovasc Interv 2006;68:83–8.

10. Mintz GS, Shah VM, Weissman NJ. Regional remodeling as thecause of late stent malapposition. Circulation 2003;107:2660–3.

11. Siqueira DA, Abizaid AA, Costa J de R, et al. Late incompleteapposition after drug-eluting stent implantation: incidence and po-tential for adverse clinical outcomes. Eur Heart J 2007;28:1304–9.

12. Ozaki Y, Okumura M, Ismail TF, et al. The fate of incomplete stentapposition with drug-eluting stents: an optical coherence tomography-based natural history study. Eur Heart J 2010;31:1470–6.

JACC Vol. 63, No. 14, 2014 Attizzani et al.April 15, 2014:1355–67 Incomplete Stent Apposition

1365

13. Gutiérrez-Chico JL, Regar E, Nüesch E, et al. Delayed coverage inmalapposed and side-branch struts with respect to well-apposed strutsin drug-eluting stents: in vivo assessment with optical coherence to-mography. Circulation 2011;124:612–23.

14. Fujino Y, Attizzani GF, Bezerra HG, et al. Serial assessment of vesselinteractions after drug-eluting stent implantation in unprotected distalleft main coronary artery disease using frequency-domain opticalcoherence tomography. J Am Coll Cardiol Intv 2013;6:1035–45.

15. Gutiérrez-Chico JL, Wykrzykowska J, Nüesch E, et al. Vascular tissuereaction to acute malapposition in human coronary arteries: sequentialassessment with optical coherence tomography. Circ CardiovascInterv 2012;5:20–9.

16. Hong MK, Mintz GS, Lee CW, et al. Late stent malapposition afterdrug-eluting stent implantation: an intravascular ultrasound analysiswith long-term follow-up. Circulation 2006;113:414–9.

17. Steinberg DH, Mintz GS, Mandinov L, et al. Long-term impact ofroutinely detected early and late incomplete stent apposition: an in-tegrated intravascular ultrasound analysis of the TAXUS IV, V, andVI and TAXUS ATLAS workhorse, long lesion, and direct stentstudies. J Am Coll Cardiol Intv 2010;3:486–94.

18. Tanabe K, Serruys PW, Degertekin M, et al. Incomplete stentapposition after implantation of paclitaxel-eluting stents or bare metalstents: insights from the randomized TAXUS II trial. Circulation2005;111:900–5.

19. Cook S, Eshtehardi P, Kalesan B, et al. Impact of incomplete stentapposition on long-term clinical outcome after drug-eluting stentimplantation. Eur Heart J 2012;33:1334–43.

20. Hassan AK, Bergheanu SC, Stijnen T, et al. Late stent malappositionrisk is higher after drug-eluting stent compared with bare-metal stentimplantation and associates with late stent thrombosis. Eur Heart J2010;31:1172–80.

21. Cook S, Wenaweser P, Togni M, et al. Incomplete stent appositionand very late stent thrombosis after drug-eluting stent implantation.Circulation 2007;115:2426–34.

22. Guagliumi G, Ikejima H, Sirbu V, et al. Impact of drug releasekinetics on vascular response to different zotarolimus-eluting stentsimplanted in patients with long coronary stenoses: the LongOCTstudy (Optical Coherence Tomography in Long Lesions). J Am CollCardiol Intv 2011;4:778–85.

23. Kubo T, Imanishi T, Kitabata H, et al. Comparison of vascularresponse after sirolimus-eluting stent implantation between patientswith unstable and stable angina pectoris: a serial optical coherencetomography study. J Am Coll Cardiol Img 2008;1:475–84.

24. Guagliumi G, Musumeci G, Sirbu V, et al. Optical coherencetomography assessment of in vivo vascular response after implantationof overlapping bare-metal and drug-eluting stents. J Am Coll CardiolIntv 2010;3:531–9.

25. Lee CW, Kang SJ, Park DW, et al. Intravascular ultrasound findingsin patients with very late stent thrombosis after either drug-eluting orbare-metal stent implantation. J Am Coll Cardiol 2010;55:1936–42.

26. Guagliumi G, Sirbu V, Musumeci G, et al. Examination of the in vivomechanisms of late drug-eluting stent thrombosis: findings fromoptical coherence tomography and intravascular ultrasound imaging.J Am Coll Cardiol Intv 2012;5:12–20.

27. Alfonso F, Dutary J, Paulo M, et al. Combined use of opticalcoherence tomography and intravascular ultrasound imaging in pa-tients undergoing coronary interventions for stent thrombosis. Heart2012;98:1213–20.

28. Kang SJ, Lee CW, Song H, et al. OCT analysis in patients with verylate stent thrombosis. J Am Coll Cardiol Img 2013;6:695–703.

29. Kosonen P, Vikman S, Jensen LO, et al. Intravascular ultrasoundassessed incomplete stent apposition and stent fracture in stentthrombosis after bare metal versus drug-eluting stent treatment: theNordic Intravascular Ultrasound Study (NIVUS). Int J Cardiol 2013;168:1010–6.

30. Pesarini G, Dandale R, Rigamonti A, et al. Late and very late coronarystent thrombosis: intravascular ultrasound findings and associationswith antiplatelet therapy.CatheterCardiovasc Interv 2013;82:1056–65.

31. Radu M, Jørgensen E, Kelbæk H, Helqvist S, Skovgaard L,Saunamäki K. Optical coherence tomography at follow-up afterpercutaneous coronary intervention: relationship between proceduraldissections, stent strut malapposition and stent healing. Euro-Intervention 2011;7:353–61.

32. Uren NG, Schwarzacher SP, Metz JA, et al. Predictors and outcomesof stent thrombosis: an intravascular ultrasound registry. Eur Heart J2002;23:124–32.

33. Alfonso F, Suárez A, Angiolillo DJ, et al. Findings of intravascularultrasound during acute stent thrombosis. Heart 2004;90:1455–9.

34. Gutiérrez-Chico JL, Jüni P, García-García HM, et al. Long termtissue coverage of a biodegradable polylactide polymer-coated bio-limus-eluting stent: comparative sequential assessment with opticalcoherence tomography until complete resorption of the polymer. AmHeart J 2011;162:922–31.

35. Attizzani GF, Bezerra HG, Ormiston J, et al. Serial assessment byoptical coherence tomography of early and late vascular responses afterimplantation of an absorbable-coating sirolimus-eluting stent (fromthe first-in-human DESSOLVE I trial). Am J Cardiol 2013;112:1557–6.

36. Tanigawa J, Barlis P, Dimopoulos K, Dalby M, Moore P, DiMario C. The influence of strut thickness and cell design on imme-diate apposition of drug-eluting stents assessed by optical coherencetomography. Int J Cardiol 2009;134:180–8.

37. Virmani R, Guagliumi G, Farb A, et al. Localized hypersensitivityand late coronary thrombosis secondary to a sirolimus-eluting stent:should we be cautious? Circulation 2004;109:701–5.

38. Joner M, Finn AV, Farb A, et al. Pathology of drug-eluting stents inhumans: delayed healing and late thrombotic risk. J Am Coll Cardiol2006;48:193–202.

39. Cook S, Ladich E, Nakazawa G, et al. Correlation of intravascularultrasound findings with histopathological analysis of thrombusaspirates in patients with very late drug-eluting stent thrombosis.Circulation 2009;120:391–9.

40. Attizzani GF, Bezerra HG, Chamié D, et al. Serial evaluation ofvascular response after implantation of a new sirolimus-eluting stent withbioabsorbable polymer (MISTENT): an optical coherence tomographyand histopathological study. J Invasive Cardiol 2012;24:560–8.

41. Templin C, Meyer M, Müller MF, et al. Coronary optical frequencydomain imaging (OFDI) for in vivo evaluation of stent healing:comparison with light and electron microscopy. Eur Heart J 2010;31:1792–801.

42. Malle C, Tada T, Steigerwald K, et al. Tissue characterization afterdrug-eluting stent implantation using optical coherence tomography.Arterioscler Thromb Vasc Biol 2013;33:1376–83.

43. Nakazawa G, Finn AV, Vorpahl M, Ladich ER, Kolodgie FD,Virmani R. Coronary responses and differential mechanisms of latestent thrombosis attributed to first-generation sirolimus- andpaclitaxel-eluting stents. J Am Coll Cardiol 2011;57:390–8.

44. Radu M, Jørgensen E, Kelbaek H, Helqvist S, Skovgaard L,Saunamäki K. Strut apposition after coronary stent implantationvisualised with optical coherence tomography. EuroIntervention 2010;6:86–93.

45. Maehara A, Mintz G, Weissman N. Advances in intravascular im-aging. Circ Cardiovasc Interv 2009;2:482–90.

46. Mintz GS. What to do about late incomplete stent apposition? Cir-culation 2007;115:2379–81.

47. Bezerra HG, Costa MA, Guagliumi G, Rollins AM, Simon DI.Intracoronary optical coherence tomography: a comprehensive review.J Am Coll Cardiol Intv 2009;2:1035–46.

48. Fujino Y, Bezerra HG, Attizzani GF, et al. Frequency-domain opticalcoherence tomography assessment of unprotected left main coronaryartery disease–a comparison with intravascular ultrasound. CatheterCardiovasc Interv 2013;82:E173–83.

49. Mehanna EA, Attizzani GF, Kyono H, Hake M, Bezerra HG.Assessment of coronary stent by optical coherence tomography, meth-odology and definitions. Int J Cardiovasc Imaging 2011;27:259–69.

50. Prati F, Guagliumi G, Mintz GS, et al. Expert review document part2: methodology, terminology and clinical applications of opticalcoherence tomography for the assessment of interventional pro-cedures. Eur Heart J 2012;33:2513–20.

51. Moore P, Barlis P, Spiro J, et al. A randomized optical coherencetomography study of coronary stent strut coverage and luminal pro-trusion with rapamycin-eluting stents. J Am Coll Cardiol Intv 2009;2:437–44.

52. Chen BX, Ma FY, Luo W, et al. Neointimal coverage of bare-metaland sirolimus-eluting stents evaluated with optical coherence to-mography. Heart 2008;94:566–70.

Attizzani et al. JACC Vol. 63, No. 14, 2014Incomplete Stent Apposition April 15, 2014:1355–67

1366

53. Kim JS, Jang IK, Kim JS, et al. Optical coherence tomography eval-uation of zotarolimus-eluting stents at 9-month follow-up: compari-son with sirolimus-eluting stents. Heart 2009;95:1907–12.

54. Matsumoto D, Shite J, Shinke T, et al. Neointimal coverage ofsirolimus-eluting stents at 6-month follow-up: evaluated by opticalcoherence tomography. Eur Heart J 2007;28:961–7.

55. Kim BK, Shin DH, Kim JS, et al. Optical coherence tomography-based evaluation of malapposed strut coverage after drug-elutingstent implantation. Int J Cardiovasc Imaging 2012;28:1887–94.

56. Kawamori H, Shite J, Shinke T, et al. Natural consequence of post-intervention stent malapposition, thrombus, tissue prolapse, anddissection assessed by optical coherence tomography at mid-termfollow-up. Eur Heart J Cardiovasc Imaging 2013;14:865–75.

57. Durán A, Hian SK, Miller DL, Le Heron J, Padovani R, Vano E.A summary of recommendations for occupational radiation protection ininterventional cardiology. Catheter Cardiovasc Interv 2013;81:562–7.

58. ACT Investigators. Acetylcysteine for prevention of renal outcomes inpatients undergoing coronary and peripheral vascular angiography:main results from the randomized Acetylcysteine for Contrast-induced nephropathy Trial (ACT). Circulation 2011;124:1250–9.

59. Kim JS, Jang IK, Fan C, et al. Evaluation in 3 months duration ofneointimal coverage after zotarolimus-eluting stent implantation byoptical coherence tomography: the ENDEAVOR OCT trial. J AmColl Cardiol Intv 2009;2:1240–7.

60. Li S, Wang Y, Gai L, et al. Evaluation of neointimal coverage andapposition with various drug-eluting stents over 12 months afterimplantation by optical coherence tomography. Int J Cardiol 2013;162:166–71.

61. Finn AV, Nakazawa G, Joner M, et al. Vascular responses to drugeluting stents: importance of delayed healing. Arterioscler ThrombVasc Biol 2007;27:1500–10.

62. Räber L, Baumgartner S, Garcia-Garcia HM, et al. Long-termvascular healing in response to sirolimus- and paclitaxel-eluting stents:an optical coherence tomography study. J Am Coll Cardiol Intv 2012;5:946–57.

63. Attizzani GF, Bezerra HG. Contemporary assessment of stent strutcoverage by OCT. Int J Cardiovasc Imaging 2013;29:23–7.

64. Morino Y, Limpijankit T, Honda Y, et al. Late vascular response torepeat stenting for in-stent restenosis with and without radiation: anintravascular ultrasound volumetric analysis. Circulation 2002;105:2465–8.

65. Guagliumi G, Bezerra HG, Sirbu V, et al. Serial assessment of cor-onary artery response to paclitaxel-eluting stents using optical coher-ence tomography. Circ Cardiovasc Interv 2012;5:30–8.

66. Nakazawa G, Finn AV, Joner M, et al. Delayed arterial healing andincreased late stent thrombosis at culprit sites after drug-eluting stentplacement for acute myocardial infarction patients: an autopsy study.Circulation 2008;118:1138–45.

67. Hong MK, Mintz GS, Lee CW, et al. Incidence, mechanism, pre-dictors, and long-term prognosis of late stent malapposition afterbare-metal stent implantation. Circulation 2004;109:881–6.

68. van der Hoeven BL, Liem SS, Dijkstra J, et al. Stent malappositionafter sirolimus-eluting and bare-metal stent implantation in patientswith ST-segment elevation myocardial infarction: acute and 9-monthintravascular ultrasound results of the MISSION! Intervention study.J Am Coll Cardiol Intv 2008;1:192–201.

69. Choi SY, Witzenbichler B, Maehara A, et al. Intravascular ultrasoundfindings of early stent thrombosis after primary percutaneous inter-vention in acute myocardial infarction: a Harmonizing Outcomes withRevascularization and Stents in Acute Myocardial Infarction (HO-RIZONS-AMI) substudy. Circ Cardiovasc Interv 2011;4:239–47.

70. Guagliumi G, Costa MA, Sirbu V, et al. Strut coverage and latemalapposition with paclitaxel-eluting stents compared with bare metalstents in acute myocardial infarction: optical coherence tomographysubstudy of the Harmonizing Outcomes with Revascularization andStents in Acute Myocardial Infarction (HORIZONS-AMI) trial.Circulation 2011;123:274–81.

71. Gonzalo N, Barlis P, Serruys PW, et al. Incomplete stent apposi-tion and delayed tissue coverage are more frequent in drug-elutingstents implanted during primary percutaneous coronary interventionfor ST-segment elevation myocardial infarction than in drug-eluting stents implanted for stable/unstable angina: insights fromoptical coherence tomography. J Am Coll Cardiol Intv 2009;2:445–52.

72. Davlouros PA, Nikokiris G, Karantalis V, et al. Neointimal coverageand stent strut apposition six months after implantation of a paclitaxeleluting stent in acute coronary syndromes: an optical coherence to-mography study. Int J Cardiol 2011;151:155–9.

73. van Werkum JW, Heestermans AA, Zomer AC, et al. Predictors ofcoronary stent thrombosis: the Dutch Stent Thrombosis Registry.J Am Coll Cardiol 2009;53:1399–409.

74. Steg P, James SK, Atar D, et al. ESC Guidelines for the managementof acute myocardial infarction in patients presenting with ST-segmentelevation. Eur Heart J 2012;33:2569–619.

75. Finn AV, Kolodgie FD, Harnek J, et al. Differential responseof delayed healing and persistent inflammation at sites of over-lapping sirolimus- or paclitaxel-eluting stents. Circulation 2005;112:270–8.

76. Shinke T, Li J, Chen JP, et al. High incidence of intramural thrombusafter overlapping paclitaxel-eluting stent implantation: angioscopicand histopathologic analysis in porcine coronary arteries. Circ Car-diovasc Interv 2008;1:28–35.

77. Kolandaivelu K, Swaminathan R, Gibson WJ, et al. Stent thrombo-genicity early in high-risk interventional settings is driven by stentdesign and deployment and protected by polymer-drug coatings.Circulation 2011;123:1400–9.

78. Tanigawa J, Barlis P, Dimopoulos K, Di Mario C. Optical coherencetomography to assess malapposition in overlapping drug-elutingstents. EuroIntervention 2008;3:580–3.

79. Aoki J, Mintz GS, Weissman NJ, et al. Chronic arterial responsesto overlapping paclitaxel-eluting stents: insights from serial intravas-cular ultrasound analyses in the TAXUS-V and -VI trials. J Am CollCardiol Intv 2008;1:161–7.

80. Otake H, Honda Y, Yamasaki M, et al. Vascular response to over-lapping everolimus-eluting stents: comparison with paclitaxel-elutingstents. Circ J 2010;74:1023–5.

81. Dawkins KD, Grube E, Guagliumi G, et al. Safety and efficacy ofmultiple, overlapping polymer-based paclitaxel-eluting stents. Euro-Intervention 2007;3:213–21.

82. Guagliumi G, Capodanno D, Ikejima H, et al. Impact of differentstent alloys on human vascular response to everolimus-eluting stent: anoptical coherence tomography study: the OCTEVEREST. CatheterCardiovasc Interv 2013;81:510–8.

83. Tahara S, Bezerra HG, Sirbu V, et al. Angiographic, IVUS and OCTevaluation of the long-term impact of coronary disease severity at thesite of overlapping drug-eluting and bare metal stents: a substudy ofthe ODESSA trial. Heart 2010;96:1574–8.

84. Nakazawa G, Yazdani SK, Finn AV, Vorpahl M, Kolodgie FD,Virmani R. Pathological findings at bifurcation lesions: the impact offlow distribution on atherosclerosis and arterial healing after stentimplantation. J Am Coll Cardiol 2010;55:1679–87.

85. Murasato Y, Horiuchi M, Otsuji Y. Three-dimensional modeling ofdouble-stent techniques at the left main coronary artery bifurcationusing micro-focus X-ray computed tomography. Catheter CardiovascInterv 2007;70:211–20.

86. Ormiston JA, Currie E, Webster MW, et al. Drug-eluting stents forcoronary bifurcations: insights into the crush technique. CatheterCardiovasc Interv 2004;63:332–6.

87. Costa RA, Mintz GS, Carlier SG, et al. Bifurcation coronary lesionstreated with the “crush” technique: an intravascular ultrasound anal-ysis. J Am Coll Cardiol 2005;46:599–605.

88. Foin N, Alegria-Barrero E, Torii R, et al. Crush, culotte, T andprotrusion: which 2-stent technique for treatment of true bifurcationlesions? Insights from in vitro experiments and micro-computed to-mography. Circ J 2013;77:73–80.

89. Tyczynski P, Ferrante G, Moreno-Ambroj C, et al. Simple versuscomplex approaches to treating coronary bifurcation lesions: directassessment of stent strut apposition by optical coherence tomography.Rev Esp Cardiol 2010;63:904–14.

90. Foin N, Torii R, Alegria E, et al. Location of side branch accesscritically affects results in bifurcation stenting: insights from benchmodeling and computational flow simulation. Int J Cardiol 2013;168:3623–8.

91. Alegría-Barrero E, Foin N, Chan PH, et al. Optical coherence to-mography for guidance of distal cell recrossing in bifurcation stenting:choosing the right cell matters. EuroIntervention 2012;8:205–13.

92. Foin N, Torii R, Mortier P, et al. Kissing balloon or sequentialdilation of the side branch and main vessel for provisional stenting of

JACC Vol. 63, No. 14, 2014 Attizzani et al.April 15, 2014:1355–67 Incomplete Stent Apposition

1367

bifurcations: lessons from micro-computed tomography and compu-tational simulations. J Am Coll Cardiol Intv 2012;5:47–56.

93. Foin N, Secco GG, Ghilencea L, Krams R, Di Mario C. Finalproximal post-dilatation is necessary after kissing balloon in bifurca-tion stenting. EuroIntervention 2011;7:597–604.

94. Rahman S, Leesar T, Cilingiroglu M, et al. Impact of kissing ballooninflation on the main vessel stent volume, area, and symmetry afterside-branch dilation in patients with coronary bifurcation lesions: aserial volumetric intravascular ultrasound study. J Am Coll CardiolIntv 2013;6:923–31.

95. Di Mario C, Iakovou I, van der Giessen WJ, et al. Optical coherencetomography for guidance in bifurcation lesion treatment. Euro-Intervention 2010;6 Suppl J:J99–106.

96. Fujino Y, Attizzani GF, Tahara S, et al. Frequency-domain opticalcoherence tomography assessment of unfavorable kissing-balloon resultin unprotected left main intervention. J Am Coll Cardiol Intv 2013;6:1324–5.

97. Kang SJ, Ahn JM, Song H, et al. Comprehensive intravascular ul-trasound assessment of stent area and its impact on restenosis andadverse cardiac events in 403 patients with unprotected left maindisease. Circ Cardiovasc Interv 2011;4:562–9.

98. Biondi-Zoccai G, Sheiban I, Romagnoli E, et al. Is intravascular ul-trasound beneficial for percutaneous coronary intervention of bifurca-tion lesions? Evidence from a 4,314-patient registry. Clin Res Cardiol2011;100:1021–8.

99. Basalus MW, van Houwelingen KG, Ankone MJ, Feijen J, vonBirgelen C. Micro-computed tomographic assessment followingextremely oversized partial postdilatation of drug-eluting stents.EuroIntervention 2010;6:141–8.

100. Watanabe T, Fujita M, Awata M, et al. Integrity of stent polymerlayer after drug-eluting stent implantation: in vivo comparison ofsirolimus-, paclitaxel-, zotarolimus- and everolimus-eluting stents.Cardiovasc Interv Ther 2014;29:4–10.

101. Guérin P, Pilet P, Finet G, et al. Drug-eluting stents in bifurcations:bench study of strut deformation and coating lesions. Circ CardiovascInterv 2010;3:120–6.

102. Carrozza JP Jr., Hosley SE, Cohen DJ, Baim DS. In vivo assessmentof stent expansion and recoil in normal porcine coronary arteries:differential outcome by stent design. Circulation 1999;100:756–60.

103. Foin N, Sen S, Allegria E, et al. Maximal expansion capacity withcurrent DES platforms: a critical factor for stent selection in thetreatment of left main bifurcations? EuroIntervention 2013;8:1315–25.

104. Russo RJ, Silva PD, Yeager M. Coronary artery overexpansion in-creases neointimal hyperplasia after stent placement in a porcinemodel. Heart 2007;93:1609–15.

105. Verheye S, Ramcharitar S, Grube E, et al. Six-month clinical andangiographic results of the STENTYS� self-apposing stent inbifurcation lesions. EuroIntervention 2011;7:580–7.

106. van Geuns RJ, Tamburino C, Fajadet J, et al. Self-expanding versusballoon-expandable stents in acute myocardial infarction: results fromthe APPOSITION II study: self-expanding stents in ST-segment

elevation myocardial infarction. J Am Coll Cardiol Intv 2012;5:1209–19.

107. Serruys P, Ormiston JA, Onuma Y, et al. A bioabsorbable everolimus-eluting coronary stent system (ABSORB): 2-year outcomes and re-sults from multiple imaging methods. Lancet 2009;373:897–910.

108. Manna AL, Ohno Y, Attizzani GF, Tamburino C. Successfulretrograde recanalization of chronic total coronary occlusion withmultiple bioresorbable vascular scaffolds (‘full polymer jacket’): initialexperience and rationale. Eur Heart J 2013;34:2925.

109. Grasso C, Attizzani GF, Patané M, Ohno Y, Capodanno D,Tamburino C. First-in-human description of everolimus-elutingbioabsorbable vascular scaffold implantation for the treatment ofdrug-eluting stent failure: insights from optical coherence tomogra-phy. Int J Cardiol 2013;168:4490–1.

110. Kajiya T, Liang M, Sharma RK, et al. Everolimus-eluting bio-resorbable vascular scaffold (BVS) implantation in patients with ST-segment elevation myocardial infarction (STEMI). EuroIntervention2013;9:501–4.

111. Gomez-Lara J, Garcia-Garcia HM, Onuma Y, et al. A comparison ofthe conformability of everolimus-eluting bioresorbable vascular scaf-folds to metal platform coronary stents. J Am Coll Cardiol Intv 2010;3:1190–8.

112. Serruys PW, Onuma Y, Dudek D, et al. Evaluation of the secondgeneration of a bioresorbable everolimus-eluting vascular scaffold forthe treatment of de novo coronary artery stenosis: 12-month clinicaland imaging outcomes. J Am Coll Cardiol 2011;58:1578–88.

113. Onuma Y, Serruys PW, Perkins LE, et al. Intracoronary opticalcoherence tomography and histology at 1 month and 2, 3, and 4 yearsafter implantation of everolimus-eluting bioresorbable vascular scaf-folds in a porcine coronary artery model: an attempt to decipher thehuman optical coherence tomography images in the ABSORB trial.Circulation 2010;122:2288–300.

114. Tada N, Virmani R, Grant G, et al. Polymer-free biolimus a9-coatedstent demonstrates more sustained intimal inhibition, improvedhealing, and reduced inflammation compared with a polymer-coatedsirolimus-eluting Cypher stent in a porcine model. Circ CardiovascInterv 2010;3:174–83.

115. Barlis P, Regar E, Serruys PW, et al. An optical coherence tomog-raphy study of a biodegradable vs. durable polymer-coated limus-eluting stent: a LEADERS trial sub-study. Eur Heart J 2010;31:165–76.

116. Guagliumi G, Sirbu V, Musumeci G, et al. Strut coverage and vesselwall response to a new-generation paclitaxel-eluting stent with anultrathin biodegradable abluminal polymer: Optical CoherenceTomography Drug-Eluting Stent Investigation (OCTDESI). CircCardiovasc Interv 2010;3:367–75.

Key Words: drug-eluting stent(s) - incomplete stent apposition -

intravascular ultrasound - optical coherence tomography - percutaneouscoronary intervention - stent malapposition.