Despite the robust evidence supporting the prognostic implications of complete revascularization in patients admitted with a STEMI, studies examining the value of PCI in improving outcomes in patients with a chronic coronary syndrome show mixed results. A retrospective analysis including 16 029 patients who had positron emission computed tomography myocardial perfusion imaging demonstrated that an early surgical or percutaneous revascularization was associated with improved prognosis in patients with an ischaemic burden >5–10%. (3) These findings, however, were not confirmed in a post hoc analysis of the Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial that included 1379 patients who had stress perfusion imaging and quantitative coronary angiography. (4) At 7.9 years of follow-up, the extent of CAD—defined by the number of the diseased vessels—and not the severity of ischaemia was a predictor of survival. Percutaneous coronary intervention in this cohort did not improve outcomes over optimal medical therapy; more importantly, there was no interaction between the extent of ischaemia or CAD and the treatment strategy (i.e. conservative vs. PCI).

In line with these findings, the International Study Of Comparative Health Effectiveness With Medical And Invasive Approaches (ISCHEMIA study) that included 5179 patients, with moderate or severe ischaemia in non-invasive imaging, who were randomized to optimal medical therapy or optimal medical therapy plus PCI demonstrated no differences in outcomes between groups at 3.3 years of follow-up for the composite endpoint of cardiovascular death, MI, admission for unstable angina, heart failure symptoms, or resuscitated cardiac arrest (15.5% vs. 13.8%, P = 0.34). (5) In this study, PCI was associated with an improvement in the quality of life, a reduction in the angina symptoms and a lower incidence of spontaneous MI [hazard ratio (HR) 0.67, 95% confidence interval (CI) 0.53–0.83; P < 0.01]. An important limitation of the ISCHEMIA study is the high (28%) crossover rate from the conservative to the invasive arm which may have affected the reported results; the as-treated analysis has not been reported yet.

The association between the presence of viable myocardium, surgical revascularization, and clinical outcomes was recently evaluated by a post hoc analysis of the Surgical Treatment for Ischaemic Heart Failure (STICH) study. (6) This analysis, that included 601 patients who had a left ventricular ejection fraction ≤35% and viability assessment, failed to demonstrate an impact of the presence or absence of myocardial viability on the survival benefit noted in patients undergoing surgical revascularization at 10.4-year follow-up. The REVascularisation for Ischaemic VEntricular Dysfunction (REVIVED) study (NCT01920048) is currently examining the safety and efficacy of PCI in improving prognosis in patients with heart failure.

The optimal revascularization strategy in patients with advanced CAD [i.e. three-vessel disease or left main stem (LMS) disease] and in diabetic patients has been discussed in the 2018 European Society of Cardiology (ESC) Guidelines on myocardial revascularization: surgical revascularization is currently the recommended treatment strategy in diabetic patients with multivessel CAD, while PCI has a IIB indication in patients with a SYNTAX score ≤22 and is not recommended in patients with SYNTAX score >22. (7)

These recommendations are in line with the findings of the Future Revascularization Evaluation in Patients with Diabetes Mellitus: Optimal Management of Multivessel Disease (FREEDOM) Follow-On study that included 1900 diabetic patients with multivessel disease that were randomized to surgical or percutaneous revascularization and reported a higher mortality rate at 8 years of follow-up in the PCI arm compared to the surgical revascularization group (24.3% vs. 18.3%, P = 0.010). (8) Conversely, the Synergy between PCI with Taxus and Cardiac Surgery (SYNTAX) Extended Survival study that included 1689 patients with LMS or three-vessel disease did not demonstrate differences in the all-cause mortality between patients allocated to PCI and those treated surgically at 10 years of follow-up (27% vs. 24%, P = 0.092). There was, however, a treatment effect by subgroup interaction according to the presence or absence of three-vessel disease; mortality was increased in the PCI group compared to the coronary artery bypass graft (CABG) arm (HR 1.41, 95% CI 1.10–1.80), while there was no differences between the two groups in patients with LMS disease (HR 0.90, 95% CI 0.68–1.20); conversely, there was no difference in outcomes for the two treatment strategies in diabetic and non-diabetic patients (P-for interaction 0.660). (9) A limitation of both studies is the fact that the patients in the PCI arm were treated with a 1st generation drug-eluting stent (DES) that is not currently used in contemporary practice, and the fact that they both reported only all-cause mortality instead of patient-orientated cardiovascular endpoints.

The Evaluation of XIENCE vs. Coronary Artery Bypass Surgery for Effectiveness of Left Main Revascularization (EXCEL) study overcame these limitations; in this study, 1905 patients with LMS disease and SYNTAX score ≤32 were randomized to PCI with a 2nd generation DES or CABG. (10) In the PCI arm, intravascular ultrasound (IVUS) imaging was used in 77.2% of the cases. (11) At 5-year follow-up, there were no differences between groups for the combined endpoint of all-cause death, MI, or stroke (22.0% in the PCI arm vs. 19.2% in the CABG group; P = 0.13). The event rate at 30-day follow-up was lower in the PCI arm (4.9% vs. 8.0%), there was no difference between groups for the period 30 days to 1 year (4.1% vs. 3.8%), while for the period 1–5 years of follow-up a higher event rate was reported in patients undergoing PCI (15.1% vs. 9.7%). Patients randomized to CABG were more likely to suffer a cerebrovascular event (5.2% vs. 3.3%), while those treated with PCI had increased all-cause mortality (13.0% vs. 9.9%) and more often ischaemia driven revascularization (16.9% vs. 10.0%). Similarly to what it has been reported in the SYNTAX study, there was no difference in the outcomes between the two treatment strategies in diabetic and non-diabetic patients at 3- and 5-year follow-up. (10, 12)

In 2019, the 3-year follow-up data of the DKCRUSH V study were published; similar to what has been reported at 1-year follow-up, double kiss-crush technique was associated with a lower incidence of target lesion revascularization (TLR, 5.0% vs. 10.3%, P = 0.029) target vessel MI (1.7% vs. 5.8%, P = 0.017), and definite or probable stent thrombosis (0.4% vs. 4.1%, P = 0.006) compared to provisional T-stenting. (13) Double kiss-crush technique, however, is a challenging procedure and requires skills and expertise; therefore, considering that the findings of the DKCRUSH V study may not be reproduced by centres with less experienced operators, the recently published 14th consensus document from the European Bifurcation Club advocates the use of provisional T-stenting technique for the treatment of bifurcations lesions and proposes a two stent strategy only in lesions with a complex anatomy, when access to the side branch is challenging, or when there is ostial disease in the side branches extending >5 mm form the carina and/or increased calcification. (14) In the case of a two stent strategy, the European Bifurcation Club recommends the use of culotte or TAP technique and when the crush technique is considered it proposes the use of the double kiss-crush.

In 2019, the EuroCTO Club published a consensus document that summarizes the current evidence (Figure 1), discusses the indications for chronic total occlusion (CTO) revascularization, presents the advances in CTO equipment, and provides recommendations about training in CTO PCI. (15) In line with the ESC guidelines on myocardial revascularization and taking into account the findings of randomized controlled studies, the EuroCTO Club recommends CTO recanalization in the presence of symptoms despite optimal medical therapy; in asymptomatic patients, ischaemic burden assessment is recommended and CTO revascularization is advised if there is evidence of increased ischaemic burden (≥10% of the left ventricular mass). These recommendation are in line with the findings of the recently reported Drug-Eluting Stent Implantation vs. Optimal Medical Treatment in Patients With Chronic Total Occlusion (DECISION-CTO) trial. (16) In this study, 815 patients with a CTO were randomized in 1:1 ratio to complete revascularization or to the treatment of the obstructive non-CTO lesions whenever these were present. Only one-fourth of the patients included in the two groups had a single-vessel disease. At 4-year follow-up, there was no difference between the two groups for the combined endpoint of death, MI, stroke, or revascularization (22.4% vs. 22.3%, P = 0.86) or patients’ quality of life. These findings indicate that in case of multivessel disease revascularization of the non-CTO lesion and re-evaluation of the extent of ischaemia and patient symptoms should be considered before advocating recanalization of a CTO. Limitations of the study—the largest of its kind—included the high crossover rate (19.6%) from the non-CTO PCI group to the CTO-PCI group within the first days from randomization as well the fact that it was underpowered for the primary endpoint as patient recruitment was early terminated because of a slow enrolment rate.

Percutaneous coronary intervention in small vessels has been associated with a higher incidence of major adverse cardiovascular events (MACE) and TLR due to in-stent restenosis. In 2019, a pre-specified sub-analysis of the Biodegradable Polymer and Durable Polymer Drug-eluting Stents in an All Comers Population (BIO-RESORT) study was published that compared outcomes following PCI in small vessels (<2.5 mm) using ultrathin-strut cobalt chromium biodegradable polymer sirolimus-eluting stents (strut thickness 71 μm) or very thin-strut platinum chromium biodegradable polymer everolimus-eluting stents (strut thickness 78 μm) or previous-generation thin strut cobalt-chromium durable polymer zotarolimus-eluting stents (strut thickness 102 μm). A higher incidence of TLR was noted in the thicker strut zotarolimus-eluting stent than the ultrathin-strut sirolimus-eluting stent group (5.3% vs. 2.1%, P = 0.006), while there was no difference in the TLR rate between the everolimus and zotarolimus-eluting stent groups (4.0% vs. 5.1%, P = 0.31). (17) These findings convincingly highlight the prognostic implications of strut thickness in small vessels in the DES era and are in line with previous studies reporting outcomes in bare-metal stents. (18)

In-stent restenosis represents the most common cause of stent failure; its treatment is challenging and is associated with poor prognosis and a high TLR rate. (19) The two most effective treatment strategies today are drug-coated balloon angioplasty or DES implantation. In 2019, the Difference in Anti-restenotic Effectiveness of Drug-eluting stent and drug-coated balloon AngiopLasty for the occUrrence of coronary in-Stent restenosis (DAEDALUS) patient-level meta-analysis was published that included 1976 patients treated with a paclitaxel-coated balloon or a DES. (20) At 3-year follow-up, paclitaxel-coated balloon angioplasty was associated with a higher incidence of TLR comparing to DES implantation (HR 1.32, 95% CI 1.02–1.70; P = 0.035); however, there was no difference between groups for the combined endpoint of death, MI, or target lesion thrombosis (Figure 2).

The ESC Guidelines on myocardial revascularization recommends the use of 2nd generation DES in daily clinical practice. (7) The extended follow-up of the Comparison of Biolimus Eluted From an Erodible Stent Coating With Bare Metal Stents in Acute ST-Elevation Myocardial Infarction (COMFORTABLE-AMI) study and the nested intravascular imaging analysis published this year has provided further evidence about the superiority of DES over bare-metal stents in patients admitted with a STEMI. At 5-year follow-up, Biolimus stent implantation was associated with a lower incidence of target vessel MI (2.2% vs. 5.0, P = 0.02) and ischaemia driven TLR (4.4% vs. 10.4%, P < 0.001) than treatment with a bare-metal stent. (21)

The BIOSTEMI study also focused on the treatment of patients with STEMI and randomized 1300 subjects to ultrathin cobalt chromium sirolimus-eluting stent vs. durable polymer everolimus-eluting stent implantation. At 12-month follow-up, treatment with ultrathin sirolimus-eluting stents was associated with a lower incidence of target lesion failure (TLF) than everolimus-eluting stents (4% vs. 6%; rate ratio: 0.59, 95% Bayesian credibility interval: 0.37–0.94; posterior probability of superiority 0.986). (22) Conversely, the TALENT study that compared outcomes in all-comer patients randomized to ultrathin cobalt chromium sirolimus-eluting stent and durable polymer everolimus-eluting stent failed to show a difference for the incidence of the composite endpoint of cardiac death, target-vessel MI, or clinically indicated TLR between groups (4.9% vs. 5.3%, P_{for non-inferiority} < 0.0001). (23)

Bioresorbable scaffolds have been introduced to overcome the limitations of DES and improve long-term outcomes. However, the increased event rate reported in these devices at short- and intermediate-term follow-up raised concerns about their safety and today are not recommended for routine clinical use. A recent meta-analysis of randomized studies comparing the Absorb bioresorbable vascular scaffold (BVS) and the everolimus-eluting stent showed a higher incidence of TLF in the Absorb BVS at 5-year follow-up (14.9% vs. 11.6%, P = 0.030) that was attributed to a higher incidence of target vessel MI and ischaemia driven TLR. (24) Landmark analysis demonstrated a higher event rate in the Absorb BVS group for the period 0–3 years of follow-up; however, for the period 3–5 years of follow-up, the incidence of cardiac death, target vessel MI, ischaemia driven TLR, and device thrombosis was similar between groups in patients who had not experienced an event in the first 3 years. These findings for the first time provide unique insights about the timing of the events in bioresobable scaffolds and indicate a low event rate at long term after their full resorption.

Intravascular lithotripsy (IVL) has emerged over the last years as an effective alternative for the treatment of calcified lesions that are associated with an increased risk of complications and worse prognosis. (25) It involves the advancement of a catheter with a balloon on its tip that contains multiple emitters which generate sonic pressure waves that selectively fracture vascular calcium without affecting the integrity of the fibroelastic tissue of the plaque. (26) The Shockwave Coronary Rx Lithoplasty Study (DISRUPT CAD) was the first study that systematically examined the safety and efficacy of IVL in 60 patients with heavily calcified lesions and length ≤32 mm; the procedure was successful in all the lesions resulting in an acute gain of 1.7 mm and a post-procedural percent diameter stenosis of 12.2%. The overall MACE rate at 6 months of follow-up was 8.3%; three peri-procedural MI and two cardiac deaths were reported. (27) Similar were the findings of the DISRUPT CAD II study that included 120 patients; in that study, the in-hospital MACE rate was 5.8% (7 non-Q wave MI), while at 30-day follow-up, the MACE rate was 7.6%. Optical coherence tomography (OCT) imaging was performed in 48 patients before and in 47 after stenting and demonstrated that IVL caused 3.4 ± 2.6 fractures per lesion resulting in an acute gain of 4.79 ± 2.45 mm² and an excellent stent expansion of 102.8 ± 30.6%. (28) Recently, Wilson et al. (29) showed that IVL therapy is associated with ventricular ectopics and asynchronous pacing. In this study, no malignant arrhythmias were reported; the ongoing DISRUPT CAD III study is expected to provide further evidence about the safety and efficacy of IVL in the treatment of calcified lesions (NCT03595176).

Recent studies have shown that the fractional flow reserve (FFR) and the resting indices including the instantaneous wave free ratio (iwFR) have a value not only in guiding revascularization but also in assessing the final results post-PCI and predicting prognosis. (35, 36) There are however occasional discordances between hyperaemic FFR and resting indices. Several studies this year attempted to examine the physiological characteristics of lesions with discordant FFR and iwFR and identify lesion types and subgroup of patients where FFR should be preferred to iwFR and vice versa. (37, 38) A recent sub-analysis of the Functional Lesion Assessment of Intermediate Stenosis to Guide Revascularization (DEFINE-FLAIR) study comparing outcomes in patients with a lesion in the left anterior descending coronary artery deferred from revascularization based on the FFR or iwFR estimations showed a lower event rate in the iwFR group at 1-year follow-up that was attributed to a lower incidence of unplanned revascularizations (2.22% vs. 4.99%, P = 0.03). (39) Conversely, a post hoc analysis of the same study in diabetic patients showed no differences in outcomes between the FFR and iwFR groups (7.2% vs. 10.0%; P = 0.30); however, the incidence of non-fatal MI was higher in the iwFR group (4.7% vs. 1.9%; P = 0.05) with a significant interaction for the presence of diabetes (P = 0.04). (40)

In parallel with the introduction of the resting indices for the assessment of the functional severity of intermediate lesions, efforts have been made to design computerized-based methodologies that are able to post-process coronary angiography or invasive imaging data to derive FFR. In 2019, two new solutions have been presented for computational-derived FFR: the first relies on three-dimensional quantitative coronary angiography to derive vessel geometry and estimate the pressure drop across a lesion, while the second on the processing of OCT imaging data; the latter enables combined morphological and physiological assessment of atherosclerotic lesions and of the procedural results post-PCI. (41, 42) Preliminary validation of these solutions showed promising results; however, further evaluation of their efficacy in a large number of patients is required before their broad application in the clinical arena.

Cumulative evidence has highlighted the value of IVUS in guiding PCI. A meta-analysis of randomized controlled trials published this year including 4724 patients underscored the prognostic benefit of IVUS guidance, demonstrating a lower incidence in MACE (5.4% vs. 9.0%; P < 0.001), cardiac death (0.6% vs. 1.2%, P = 0.03), TLR (3.1% vs. 5.2%, P = 0.001), and definite/probable stent thrombosis (0.5% vs. 1.1%, P = 0.02) rates in the IVUS-guided compared to the angiography-guided group. (43) In line with the above findings, the 5-year follow-up analysis of the Impact of Intravascular Ultrasound Guidance on Outcomes of XIENCE PRIME Stents in Long Lesions (IVUS-XPL) study that included 1400 patients with long lesions ≥28 mm randomized to IVUS- and angiography-guided PCI, reported a lower incidence of MACE (5.6% vs. 10.7%, P = 0.001) in the IVUS-guided group attributed to a lower incidence of TLR (4.8% vs. 8.4%, P = 0.007). A landmark analysis for the follow-up period 1–5 years indicated that IVUS guidance was associated with clinical benefit at long-term follow-up (HR 0.53, 95% CI 0.29–0.95; P = 0.031). (44) These findings highlight the prognostic implications of IVUS in guiding revascularization and support its routine use to optimize procedural results and improve the short- and long-term outcomes post-PCI.

Fractional flow reserve is currently recommended to guide revascularization in patients with a chronic coronary syndrome and intermediate lesions. The FORZA study examined the value of OCT in deferring PCI; the study included 350 patients with intermediate lesions who were randomized to OCT- and FFR-guided PCI. (45) Revascularization in the OCT group was performed based on area stenosis and minimum lumen area cut-off values, while in the FFR group PCI was performed if the FFR was ≤0.80. OCT and FFR were repeated in the two groups and used to optimize stent deployment. At 13 months of follow-up, OCT-guided PCI was associated with a higher incidence of revascularization and increased cost while there was no difference in the incidence of MACE—defined as the composite endpoint of all-cause death, MI, target vessel revascularization—between the FFR- and OCT-guided groups (8.0% vs. 3.4%, P = 0.064). For the primary endpoint of the study, i.e. the incidence of MACE and significant angina at 13 months of follow-up, OCT-guided PCI was marginally superior to FFR-guidance (14.8% vs. 8.0%, P = 0.048). The FROZA study is the first that compared in a randomized fashion intravascular imaging vs. physiology guided PCI revealing limitations of both approaches in guiding revascularization (i.e. increased procedural cost and number of vessels treated in the OCT-guided group and a higher incidence of MACE and angina symptoms in the FFR-guided group). Combined physiology and imaging-guided revascularization is likely to overcome the limitations of both modalities and optimize procedural results and the clinical outcomes of patients with obstructive CAD.

In 2019, the European Association of Percutaneous Cardiovascular Interventions published an expert consensus document about the value of intravascular imaging in guiding treatment in ACS and in ambiguous coronary angiography findings. (46) This report highlights the value of intravascular imaging and in particular of OCT in identifying the culprit lesion when this cannot be detected by coronary angiography and in tailoring therapy in patients admitted with an ACS (Figure 3). It also underscores the value of intravascular imaging in assessing ambiguous coronary angiographic findings, in detecting embolic events and intramural haematomas, in assessing lesions caused by an external compression of the vessel by other organs and it summarizes the evidence that supports its role in identifying vulnerable plaques and high-risk patients (Figure 4).

Non-invasive functional imaging has an established role in the diagnosis of obstructive CAD in symptomatic patients. (47) In the Myocardial Perfusion CMR vs. Angiography and FFR to Guide the Management of Patients with Stable Coronary Artery Disease (MR-INFORM) study, non-invasive imaging and in particular cardiac magnetic resonance (CMR) imaging was found to be not only useful for the diagnosis of CAD but also for guiding revascularization. (48) In this study, 918 patients were randomized to CMR- or FFR-guided revascularization. CMR-guided PCI was associated with a lower incidence of coronary angiography and PCI (35.7% vs. 45.0%, P = 0.005). At 1-year follow-up, there was no difference between groups for the primary endpoint of all-cause mortality, MI, or target vessel revascularization (3.6% vs. 3.7%, P = 0.91). This report is among the few that compared the role of non-invasive imaging vs. invasive guidance for PCI. A limitation of this study is the fact that the event rate was lower than the 10% event rate assumed in the power calculation and thus it may have been underpowered in detecting differences in outcomes between the two study groups.

Similar were the findings of the Complete Revascularization or Stress Echocardiography in Patients With Multivessel Disease and ST-Segment Elevation Acute Myocardial Infarction (CROSS-AMI) study that compared angiography vs. stress echocardiography-guided revascularization in patients admitted with a STEMI that had non-culprit lesions with a diameter stenosis >50% on quantitative coronary angiography. (49) The study was prematurely stopped after enrolling 77% of the patients because of a slow recruitment (n = 306). The authors reported a higher incidence of non-culprit lesion revascularization in the angiography group (88% vs. 22%). At 1-year follow-up, there were no differences between groups for the primary endpoint of cardiac death, MI, coronary revascularization, or re-admission because of heart failure (14% vs. 14%, P = 0.85). A limitation of the CROSS-AMI study was the fact that it was underpowered to assess differences between groups. Therefore, further research is needed to examine the value of non-invasive imaging in guiding revascularization in patients with an ACS.