It has been well documented that quantifying angiographic left main disease severity especially that which involves the proximal segment is particularly challenging and is subject to significant interobserver variability [36, 37]. Three major anatomical factors which impair left main evaluation include aortic cusp opacification, short length of vessel trunk, and the presence of bifurcation or trifurcation at the distal segment [21]. Contrast streaming in the aortic cusp can obscure the ostium of the left main making angiographic evaluation problematic. On the other hand, the short segment of the left main shaft leaves little reference site for comparison. Also, the bifurcation into sub-branches at the distal end potentially conceals the distal left main. IVUS, in contrast, suffers not from the aforementioned limitation making it an investigation of choice when assessing left main lesion (Fig. 4.3).

In the case of angiographic intermediate left main stenosis, an IVUS minimum lumen diameter (MLD) of <2.8 mm or an MLA <6 mm² suggest a physiologically significant stenosis and thus merits revascularisation [24, 26]. A study of 55 patients with moderate left main disease demonstrates that these cut-off values correlate well with physiologically significant lesion as assessed by FFR with a sensitivity of 93 % and a specificity of 98 % [38]. Additionally, a recent multicentre, prospective study showed that deferral of revascularisation among patients with left main MLA ≥ 6 mm² carried a similar outcome with the revascularised group who has left main MLA < 6mm² during long-term follow-up [39]. Another comparative study used an MLA cut-off value of 7.5 mm² to determine whether a patient should undergo revascularisation or to have a deferral strategy [40]. It concluded that during the mean follow-up of 3.3 ± 2 years, there was no difference in major adverse cardiac event (MACE) between the two cohorts. Altogether, in conjunction with clinical information, it is reasonably safe to defer revascularisation in patients with MLA ≥ 7.5 mm² and to consider revascularisation in patients with MLA < 6mm². Patients with intermediate MLA value (6–7.5 mm²) would require further physiological assessment, for example with FFR [27]. Based on the available evidence, IVUS was given a class IIA indication for the assessment of angiographically indeterminant left main coronary artery disease by the recent joint American College of Cardiology/American Heart Association/Society for Cardiovascular Angiography and Interventions (ACC/AHA/SCAI) guideline [41].

Aside from its ability to provide further stratification of left main disease severity, IVUS also made a major contribution in the area of unprotected left main PCI. Patients with significant left main disease (≥50% stenosis) carry a high-risk mortality and morbidity from cardiovascular events considering the area of myocardium at risk should the flow in this vessel becomes compromised. To date, coronary artery bypass surgery remains the “gold standard” treatment for majority of significant left main disease especially in those with multivessel involvement. Nevertheless, the introduction of drug-eluting stent (DES) has ushered a new era of complex PCI, such as unprotected left main coronary artery disease and bifurcation lesion PCI due to its low rate of restenosis [42]. The use of pre-intervention IVUS in the left main PCI permits a detailed assessment of the vessel anatomy, the size of the vessel, and to determine the extent of ostial involvement at the daughter sub-branches [26, 43]. This information helps the operator to optimise stent selection and provide complete disease coverage. IVUS also assists in ensuring adequate stent expansion and apposition; important risk factors for restenosis and stent thrombosis [44–46]. Finally, IVUS also reveals the extent of calcification. This information will be useful in deciding whether to employ debulking strategy with rotational atherectomy to facilitate the stent placement [47–49].

The role of IVUS-guided strategy for unprotected left main PCI, however, remains a controversial issue given the conflicting data in the literature [46]. Nonetheless, several recent reports provide clinical data favouring its routine use. In a large Korean registry study [50], the use of IVUS-guided left main PCI was associated with a 3 year reduction in mortality when compared with angiographic guidance PCI. This benefit is particularly observed among the 145 matched pairs of patients who receive DES (4.7 % vs. 16 %, p = 0.048). Moreover, a prospective, randomised, non-inferior study comparing PCI vs. CABG for left main disease (PRECOMBAT trial) reported a low event rate in the PCI group and comparable outcome with the CABG group (8.7 % vs. 6.7 %, p = 0.02 for non-inferiority) [51]. In this multicentre trial, IVUS was extensively used. Some of this clinical benefit may be derived from IVUS ability to determine the extent of disease burden and lesion complexity thus assisting the operator with the appropriate procedure strategy. For instance, the MLA of the confluent zone between the distal segment of left main coronary artery and its daughter branches has been shown to be an accurate indicator of bifurcation left main disease severity and a predictor of post-stent underexpansion and clinical outcome [43, 52]. Another study then demonstrated that among patients who undergo bifurcation left main PCI, those with post-stenting underexpansion have significantly lower MACE-free survival rate in 2 years when compared with their counterpart [53]. Taken together, IVUS should be considered to guide unprotected left main PCI especially when it involves the bifurcation point.

IVUS has been instrumental in helping us to understand the arterial responses to different coronary interventional modality [54–57] and has assisted in the improvement of technical details of the devices application and their manufacturing. For instance, IVUS has given a significant insight into the different mechanism of restenosis post-intervention. It discovers that whilst a combination of elastic recoil and concentric remodelling and to a lesser role, local intimal hyperplasia account for the mechanism of restenosis in post-balloon angioplasty [58–60], neointimal hyperplasia is the sole mechanism of late lumen loss in the stented artery receiving bare metal stent (BMS) [61, 62]. These observations were extremely valuable in guiding the development of management strategy for in stent restenosis, which ultimately leads to the development of DES. IVUS has also revolutionised the technique of stent deployment and post-stent medication regime as it is practised today. Following the enthusiasm with the introduction of stents, clinicians began to notice the issue with stent implantation in the form of subacute thrombosis [63]. This issue was further compounded by major bleeding issue due to intensive use of oral anticoagulation regime in order to reduce the incidence of stent thrombosis. IVUS then began to identify the association between subacute stent thrombosis and stent deployment issue, such as inadequate stent expansion and incomplete stent apposition (ISA) [64]. Inadequate stent expansion occurs when part of the stent is insufficiently expanded when compared to the proximal and distal reference site. Incomplete stent apposition, in contrast, refers to a lack of contact between the stent struts and the underlying arterial wall. Based on this observation, Colombo et al. developed post-stent deployment high pressure technique [65]. This strategy not only led to the reduction of stent thrombosis incidence but also allowed oral anticoagulation to be replaced with dual antiplatelet therapy thus reducing the length of hospital stay and vascular complication rate.

The benefit of IVUS in coronary intervention was initially demonstrated in the era of balloon angioplasty. Randomised and non-randomised IVUS trials demonstrated that IVUS-guided strategy resulted in the improvement of final lumen diameter and caused significant reduction in clinically driven reintervention rate when compared with the angiography group [66, 67]. Another randomised trial of 254 patients (BEST trial) then reported a similar clinical outcome between IVUS-guided balloon angioplasty (with provisional stenting) vs. routine angiography-guided stenting [68]. It is however worth noting that in this trial, there is a high crossover rate (44 %) in the balloon angioplasty group to stenting. Taken together, the use of IVUS is effective in optimising final balloon angiographic and clinical outcome. This approach, however, was viewed as time consuming and cost-ineffective especially as the stent profile continued to improve and its price continued to decline. Thus, routine stenting now is the preferred strategy above balloon angioplasty with provisional stenting [63].

The benefit of IVUS-guided strategy in BMS PCI has been examined in both observational [69–71] and randomised trials [72–75]. IVUS can be used to perform pre-interventional evaluation of the extent of lumen obstruction and the mechanism of obstruction (e.g. thrombus or calcification). This in turn allows for each management strategy to be individualised and tailored according to patient’s clinical baseline status. Post-intervention, IVUS may assist with optimising procedural outcome, such as assessment of stent expansion and stent strut apposition, detection of PCI complication (e,g., dissection, plaque shift), and identification of residual stenosis and stent fracture. Of these, inadequate stent expansion has been associated consistently with target vessel failure resulting from either in stent restenosis [71, 76–78] or stent thrombosis [79, 80]. Despite its practical use and demonstrable benefit by case control studies, the data regarding IVUS clinical benefit from randomised trials are conflicting. The TULIP study, a randomised controlled trial involving 144 patients with long disease segment (>20 mm) demonstrated a significant improvement in both angiographic and clinical end points in the IVUS-guided group when compared to the angiographic directed group [74]. Another trial, AVID, however reported a slight different outcome. AVID randomised 800 patients into IVUS vs. angiography-guided intervention with total follow-up of 12 months. Indeed, AVID is the largest randomised trial evaluating the benefit of IVUS in BMS PCI. In this trial, IVUS-guided strategy resulted in the improvement of post-procedural minimum stent area (a marker of adequate stent expansion) but did not result in the reduction of clinical end point (death/myocardial infarction/target lesion revascularisation) except in the subgroup with moderate size vessel (2.5–3.5 mm) and in saphenous vein graft PCI [72]. In this subgroup, the reduction in clinical end point is mainly driven by target lesion revascularisation. Thus the main clinical benefit of IVUS, based on these trials is the reduction of target vessel failure without added mortality benefit. These findings were further replicated in a meta-analysis from seven randomised trials involving 2,193 patients [81]. It concludes that IVUS guidance was associated with a significantly larger post-procedure angiographic MLD, lower rate of 6-month angiographic restenosis, a reduction in the revascularisation rate, and overall MACEs. However, no significant effect was seen for myocardial infarction or mortality. Overall, the efficacy of IVUS in optimising procedural outcome and reducing restenosis rate in BMS PCI is unquestionable, nonetheless its routine use in clinical practice still needs to be weighed against a number of factors, including cost, time, availability, and the operators skill, to accurately acquire and interpret the images.

The arrival of DES opened a new chapter in the world of interventional cardiology. DES had been shown to significantly reduce the risk of restenosis to less than 10 % [44, 82–85] when compared to BMS. It does this by inhibiting neointimal proliferation through its drug coating [86, 87]. As a result, operators began to put less attention on achieving optimal final procedural outcome as in the days of BMS [65]. However, the discovery of increasing incidence of stent thrombosis, especially very late stent thrombosis in the patients receiving DES has resulted in renewed interest in the use of IVUS to direct therapy.

Stent thrombosis, despite its rare occurrence (annual risk ~0.5%), carries a significant mortality and morbidity [88]. The overall prognosis is poor with 10–30% mortality rate. Several precipitating factors have been implicated to cause stent thrombosis and these essentially can be categorised as patient factors, lesion characteristics, device factors, and procedural factors. IVUS has provided significant insights into the morphologic pattern and possible causes of stent thrombosis following DES implantation, specifically stent underexpansion and incomplete stent apposition. Fuji et al. conducted a retrospective analysis on 15 patients who developed stent thrombosis following successful DES implantation [89]. They reported that lesions leading to stent thrombosis had more stent underexpansion, smaller minimum stent area, and residual edge stenosis. The strong association between IVUS detected stent underexpansion and stent thrombosis was also observed and well established in various other DES trials [90–95] and BMS trials [79, 80]. The role of incomplete stent apposition in causing stent thrombosis, on the other hand, is still controversial. Firstly, incomplete stent apposition is common, occurring in 10–20% of DES and can be acute or late [96]. Whilst acute ISA is procedural related, late ISA may be due to a combination of positive vessel remodelling (i.e. vessel expansion), intimal hyperplasia, or dissolved thrombus or plaque which led to a gap between vessel wall and stent [97]. Secondly, there is a discrepancy of observation among various trials regarding the link between ISA and stent thrombosis. In a case control study by Cook et al., a significantly high rate of ISA was noted in the very late stent thrombosis patients as compared to the DES control group [98]. This association was observed in another observational study [99] but not identified in follow-up studies of large randomised DES trials [83, 100–103]. In a recent sub-analysis of a randomised DES trial, Cook et al. reported that very late stent thrombosis and MACE occurs more frequently in patients with ISA than without ISA [104]. In this study, there was no difference in mortality observed.

Based on these IVUS observations, several trials have been conducted to assess the clinical benefit of IVUS-directed therapy in DES PCI. A study by Roy et al. compared 1-year clinical outcomes in 884 patients who underwent IVUS-guided PCI with a propensity-matched cohort of angiographically guided patients [105]. IVUS-directed therapy was found to be associated with lower incidence of stent thrombosis at 30 days (0.5% vs. 1.4%, p = 0.05) and 1 year (0.7% vs. 2.0%, p = 0.01) with no difference in the rates of myocardial infarction or death. Claessen et al. subsequently reported that IVUS guidance strategy resulted in significant reduction of early, medium-term, and long-term clinical outcome in a large registry study [106]. This benefit is mostly driven by reduction in the numbers of myocardial infarction. In another large observational study involving 8,371 patients, IVUS-guided DES PCI was also found to be associated with a 3-year reduction in mortality (HR 0.46; 95% CI 0.33–0.66, p < 0.001) [107]. Despite all these demonstrable benefits, to date there is still a lack of information regarding clinical outcome from IVUS-randomised trials. Indeed, there is only one randomised trial assessing the efficacy of IVUS use in DES implantation. AVIO trial randomised 284 patients with complex lesion (e.g. small and diffuse vessel disease, bifurcation lesion, chronic total occlusion) into IVUS-guided or angiography-guided arms [108]. Optimal stent expansion is defined as achieving ≥70% of the cross sectional area of the post-dilation balloon. The preliminary data from 9-month follow-up indicated that IVUS-guided strategy resulted in larger post-procedural MLD but no difference was observed in the rate of MACE. On balance, there is no strong recommendation for routine IVUS-guided PCI at this stage. However, IVUS use should be considered in patients with high risk of stent thrombosis or in patients whereby the consequence of stent thrombosis is fatal, such as left main coronary artery disease. IVUS should also be considered for evaluation of in stent restenosis or stent thrombosis, especially when it is a recurring event.