Many experts would agree that the concept of looking at vessel architecture from the inside using intracoronary probes, instead of simply relying on the angiographic vessel cast, is an elegant way to practise interventional cardiology in order to decide whether to treat a coronary artery and to guide coronary intervention. The historical development of stents neatly exemplifies this notion: the introduction of stents in clinical practice was initially burdened by an unacceptably high incidence of subacute thrombosis. The introduction of intravascular ultrasound (IVUS) opened the way for understanding the reasons for stent failure. IVUS clarified that after optimal angiographic results were obtained, many first-generation stents continued to have a marked underexpansion with irregular eccentric lumen and incomplete apposition of the stent struts to the vessel wall. These findings led to a new strategy for stent deployment based on high-pressure balloon dilatation inside the stent, to be performed with angiographic guidance.1–4 In other words, IVUS taught how to implant a stent, but it then failed to become the technique used for routine guidance.
In the 20 years since the clinical introduction of IVUS, there has been a growth in new imaging modalities, with optical coherence tomography (OCT) being the most promising in terms of improving results of interventional cardiology, which in the current era is characterised by new technical solutions for stents that are now capable of eluting drugs or exerting an antithrombotic action due to specific coverage.
Optical Coherence Tomography Feasibility and Safety
OCT is an optical analogue of IVUS based on infrared light emission. In comparison with IVUS, OCT has improved resolution (10mm) and contrast and a limited penetration that does not exceed 1–2mm, therefore offering a high-resolution superficial picture of coronary arteries. This feature of OCT allows the visualisation of specific components of the atherosclerotic plaques and details the architecture of stented segments, providing similar information to histology.5–8
The first image-acquisition time-domain technique was complex and time-consuming, requiring a soft occlusion balloon and saline injection in the coronary artery. This modality of acquisition limited the widespread application of OCT in the clinical arena, focusing its role in research settings. Vessel imaging by OCT is now much simpler due to the introduction by our group of a non-occlusive modality of image acquisition for the first OCT time-domain technology.9,10 More recently, this technique has been applied for the novel frequency-domain catheters that acquire images at a speed of 20mm/second and make the procedure very user-friendly.
The most comprehensive registry on safety in the use of time-domain OCT included 468 patients studied by OCT in six European centres. Non-occlusive OCT was performed in 45% of cases; ventricular fibrillation occurred in 1.1% of patients, air embolism in 0.6% and vessel dissection in 0.2%.11 There are no data available on the safety of the frequency-domain technology; however, the marked simplification of the acquisition procedures and the reduction in contrast volume demonstrate potential to further reduce the procedural complication rate. Furthermore, the fact that the frequency-domain OCT is so fast, with a 50mm pull-back being obtained in less than five seconds, makes the technology attractive for repeated fine-tuning of stented arteries.
Assessment of Ambiguous Lesions and Deferral of Interventions
OCT provides accurate luminal measurements of lesion severity in cases of ambiguity due to an excellent delineation of the lumen–wall interface. Suboptimal angiographic visualisation may occur in the presence of intermediate lesions of uncertain severity, very short lesions, pre- or post-aneurysmal lesions, ostial or left main stenoses, disease at branching sites, sites with focal spasm or angiographically hazy lesions.8 With these angiographic characteristics, OCT use can change the operator intention-to-treat, avoiding unnecessary and uncertain interventional procedures in some cases. Compared with IVUS, OCT offers more accurate information on the superficial composition of the plaque. This feature further improves the accuracy of the definition of the culprit lesion in uncertain cases and seems particularly worthwhile in the presence of thrombi or calcifications, which create ambiguous images with haziness at angiography. In a recent expert review document on the methodology, terminology and clinical applications of OCT, the assessment of intermediate lesions and the identification of culprit plaques of acute coronary syndromes were considered two important clinical applications of OCT.7
In fact, OCT can detail abrupt plaque ulceration in the lipid pool, which is the site of fibrous and the thrombus. The thrombus is poorly studied by IVUS – a well-known Achilles’ heel of the technique.12 Similar to IVUS, OCT can quantify lesion severity more accurately than quantitative coronary angiography by measuring the minimum lumen area (MLA), with 4mm2 being considered the significant cut-off threshold for a clinically significant flow-limiting stenosis in appropriately sized (>3mm) vessels, excluding the left main coronary artery.13 However, further validation studies between OCT and fractional flow reserve may be needed to corroborate this matter. The only technical drawback of the technique is that plaques located at the very ostium of the left or right coronaries cannot be accurately addressed by OCT; in fact, at the current stage of technology, neither of the two OCT acquisition techniques (occlusive or non-occlusive) appears to be suitable for aorto-ostial assessment.7
Our centre studied the ability of OCT and IVUS to resolve ambiguous angiographic anatomy in stable and unstable clinical scenarios, including identification of culprit complicated plaques in acute coronary syndromes and assessment of lesion severity in stable patients. Sixty lesions were studied, 38 with intermediate severity and 22 exhibiting haziness at angiography. OCT proved angioplasty was unnecessary in 41% of cases. Of note, OCT confirmed the presence of complicated plaques in 61% and avoided intervention in the remaining cases in the 22 cases with haziness. Furthermore, IVUS underdiagnosed complicated plaques, which were identified in only 48% of cases. At a mean follow-up of 28 months, only two patients were admitted to hospital and no myocardial infarctions were reported (unpublished data).
Post-procedural Assessment
A burning question to be addressed is whether OCT will have a role in the future for guidance of routine stenting interventions. Routine IVUS guidance of elective bare-stenting procedures is not supported by the results of previous studies. In fact, meta-analyses on the ability of IVUS to reduce restenosis after stenting have provided conflicting results, with the most important published randomised trial being completely negative.13,14 However, data on IVUS use for the optimisation of drug-eluting stent (DES) deployment were more encouraging. Roy et al.15 recently compared 884 IVUS-guided intracoronary DES implantations with the outcomes of a propensity-score-matched population having angiographic guidance alone.
The rate of definite stent thrombosis at 12 months, which was the primary end-point of the study, was significantly lower in the IVUS-guided group, and a trend in favour of the IVUS group in target lesion revascularisation was observed. OCT offers a superb visualisation of the stent surface and depicts the spatial interactions between stent struts and vessel wall; it is reasonable to think of this technique as superior for guidance of interventions requiring novel devices including DES.
Stent underexpansion is a well-known cause of restenosis also for DES; in a substudy of the Sirolimus-Eluting Stent in De Novo Native Coronary Lesions (SIRIUS) trial, the post-intervention minimal stent area (MSA) that best separated ‘adequate’ from ‘inadequate’ patency was 5mm2, with a 90% positive predictive value for this cut-off point.16 In line with this finding, a second study performed in 670 native coronary artery lesions treated with CYPHER stents, the IVUS cut-offs that best predicted angiographic restenosis were an MSA of 5.5mm2 and a stent length of 40mm.17 Therefore, at least for the CYPHER, an MSA number of between 5 and 5.5mm2 is the range that should identify stents that are prone to restenosis. Interestingly, IVUS data showed that first-generation stents rarely achieve the nominal area, with 66±17% of predicted MSA being the average value.13
These findings support the concept that an imaging modality capable of measuring cross-sectional stent areas can further decrease DES failure. Obviously, with its high resolution OCT can perform this task easily, and the added information can identify even the smallest degree of strut malapposition, which can be misdiagnosed by IVUS. Malapposition is a potential cause of late restenosis due to the fact that the non-adhering struts to the vessel wall cannot appropriately elute the drug. However, this finding was found to have a limited role in the process of restenosis, as shown by IVUS data. Undoubtedly, OCT could become the most useful technique to guide DES positioning if post-deployment findings are critical in preventing thrombosis. Apart from the risk of late thrombosis, which is now a concern for DES, the vast majority of thromboses occur in the acute and subacute phase, affecting both bare-metal and drug-eluting stents.18,19 A final examination after carrying out a DES intervention with OCT should be performed to address underexpansion, which is rather common in the presence of calcified vessels and sometimes difficult to document by angiography. Underexpansion is simply diagnosed by measuring the stent MLA, which is a well-known parameter related to stent thrombosis, and adopting the most used IVUS criterion for optimal stent expansion, which is the comparison between minimal stent lumen areas with reference ones.20
The ability of OCT to directly identify malapposition, uneven stent strut distribution or intrastent small thrombotic formations with high resolution makes the technique an attractive approach to prevent thrombosis. Malapposition still has an uncertain clinical significance, but it is intuitive that a complete apposition of equispaced stent struts is important to reduce thrombogenicity.21,22 Malapposition can be misdiagnosed by IVUS and is easily visualised by OCT, and can contribute to stent thrombosis via different mechanisms. It may favour thrombus formation by reducing blood flow and allowing fibrin and platelet deposition. Also, persistence of acute and late-acquired malapposition is associated with less neointimal hyperplasia and reduced re-endothelialisation, which promote platelet adhesion and subsequent thrombotic stent occlusion.
OCT is also capable of detecting even the smallest amount of thrombus depositions on stent struts, a common finding in patients with acute coronary syndromes after stenting of the culprit lesions.23 The clinical significance of this finding is still unknown, but it is reasonable to think of this feature as a possible cause of acute and subacute stent thrombosis. Other factors such as incomplete coverage of stent struts at follow-up, uneven stent strut distribution and symmetry index may play a role in the genesis of stent thrombosis.22 All of them are easily addressed by OCT, and the last two can be assessed at the post-intervention study, possibly providing operators with adjunctive information on the propensity of stent toward sacute, subacute and possibly late thrombosis. Prospective OCT data are needed to corroborate these concepts.
A Superficial Assessment
The main limitation of OCT resides in the inability to measure plaque burden if the thickness exceeds 1.3–2.0mm. This may have some clinical implications as the most accreditated criterion to identify reference segments by IVUS is a plaque burden <40%. This definition derives from the IVUS finding that a plaque burden >40% at stent margin represents a risk factor for late restenosis and thrombosis.24 This drawback requires new strategies to be applied to OCT-guided interventions, with attention paid to the luminal assessment of treated arteries, according to the concept that optimal lumen enlargement should have an important role to improve outcome after intervention.
This concept is further corroborated by the recent finding that the physiological assessment of lesion severity by means of fractional flow reserve can influence patient clinical outcome.25 Reduction in lumen areas, presence of flaps due to plaque rupture capable of reducing flow or thrombotic formations encroaching the lumen are the only morphological findings that can affect coronary flow; on the other hand plaque burden and vessel remodelling should not impair coronary flow.
The use of OCT for guidance of interventional procedures must be performed according to a new philosophy that requires utmost attention to the luminal findings provided by OCT. OCT is capable of identifying the plaque–lumen contour at a very high resolution and its use should be restricted to the assessment of luminal areas, based on the concept that only luminal reduction and not the increase in plaque dimension can cause flow impairment. Obviously, the same concept can be applied post-intervention, and in particular to post-stenting assessment; the presence or reduction of lumen areas in the stented segments or at stent margins due to flow-limiting dissection, lesion prolapse or large plaque burden can affect the clinical outcome.
The inability of OCT to quantify the plaque burden does not allow an accurate selection of stent size and diameter by this metric. In an OCT-guided approach, this information may be obtained by angiography or will be based on OCT luminal assessment. An immediate review of the most recently performed OCT procedure will have a critical role in guiding the procedure. Future studies will clarify whether this limitation of OCT will be offset by the extraordinary definition of the stent surface provided by the technique. After all, blood flows on the endothelial surface; therefore, it will soon be determined whether ‘high-definition surfing’ of the endothelium or the stent immediately after deployment is of critical importance or just a good exercise.
All OCT data referenced in this article acquired with systems from LightLab® Imaging, Inc.