47 Percutaneous coronary interventions in acute coronary syndromes

Three different guidelines of the European Society of Cardiology cover the field of percutaneous coronary interventions [1–3]. Their main recommendations are the following:

Percutaneous transluminal balloon coronary angioplasty (PTCA) was first performed by Andreas Grüntzig in 1977. Over the next 15 years, it developed into an effective treatment method for chronic stable CAD. Felix Zijlstra and the Zwolle group (the Netherlands) demonstrated, in 1993, the superiority of PTCA over stand-alone thrombolysis in the treatment of AMI. Almost simultaneously, the combination of aspirin and thienopyridines was shown to considerably reduce the risk of thrombosis (and bleeding) which was previously associated with the use of stents, resulting in the widespread use of stenting, instead of plain balloon angioplasty. As a result of all these achievements, PCI—the new name for the procedure that emerged after stents became a routine part of it—was more and more used for the treatment of ACS, including STEMI. In 2002, the PRAGUE-2 and DANAMI-2 trials proved that PCI (after immediate transport to a PCI-capable hospital) should be used as the primary reperfusion therapy of STEMI also for patients presenting initially to non-PCI hospitals. Many large trials proved that very early coronary angiography, usually followed by immediate PCI or, in some cases, by CABG, was the best management strategy for moderate- to high-risk patients with NSTE-ACS. Thus, modern therapy of all forms of ACS involves PCI as the most effective treatment method, saving many lives and also improving symptoms for most survivors. The more severe the clinical presentation of a patient with ACS, the higher the benefit from an urgent coronary angiography/PCI.

Primary PCI is defined as the intervention in the culprit vessel within the first 12–24 hours after the onset of CP or other symptoms, without prior thrombolytic therapy. In the 1980s, a few case reports of successful primary PCIs were published [4, 5]. In 1990s, several randomized studies were performed, comparing thrombolytic therapy with primary balloon angioplasty [6–11], and a meta-analysis by Weaver et al. [12] clearly demonstrated a reduction in mortality, reinfarction, and stroke with mechanical intervention. This benefit is mainly derived from a very high success rate in the opening of the infarct-related artery by angioplasty (>90%), compared to a 50–60% success rate with thrombolysis which also comes at a cost of a 1–2% rate of intracranial bleeding. Long-term 5-year follow-up demonstrated a highly significant 46% reduction in mortality, in favour of primary balloon angioplasty [13]. The introduction of stenting did not have any significant influence on the 6–12 months’ mortality, compared to balloon angioplasty only, but showed a trend for a reduction in the number of reinfarctions and significantly reduced the incidence of target vessel revascularization by a decrease in restenosis, and thus stenting became routine therapy [14]. Most patients with STEMI present to hospitals not equipped for primary PCI, but an urgent transfer to a tertiary centre for primary PCI was proven to be effective and safe, which extended the availability of a primary PCI strategy [15–17]. These data lead to the question of timing: how much time can be lost in transport before the benefit of primary PCI is lost? Several time points need to be defined first. The time of symptom onset is reported by the patient; first medical contact (FMC) is defined as the first contact with the emergency medical system (EMS), or a general practitioner, or an hospital. A 12-lead ECG is the key to diagnosis and should be obtained and interpreted as soon as possible (on the FMC site, within 10 minutes). Historic terms like crossing of hospital doors are not used anymore. The next important time point is wire passage into the infarct related artery; subsequent balloon dilatation typically results in the restoration of blood flow. Thrombolytic therapy is more effective when the clot is fresh, i.e. the time from the onset of symptoms to FMC is <3 hours. It must be emphasized that, although thrombolytic therapy can be administered faster, its onset of action is not immediate—reperfusion occurs after 30–60 min [18]. Thus, the only situation where thrombolytic therapy might be better is if the patient presents early (ischaemia <3 hours) to a non-PCI hospital, with an estimated time to PCI-mediated reperfusion over 120 minutes (see Figure 47.1) The main reason favouring primary PCI, even in this setting, would be the prevention of intracranial bleeding—stroke is reduced from 2% to 1% by primary PCI [19]. Other benefits of primary PCI over thrombolytic therapy include the fact that early (3 days) discharge of low-risk patients (defined as age <70 years, LVEF >45%, one- or two-vessel disease, successful PCI, no persistent arrhythmias) is safe and cost-effective [20]. These data are summarized in the flow diagram in Figure 47.2.

Primary PCI is the preferred therapeutic option when it can be performed expeditiously by an experienced team, including not only an interventional cardiologist, but also skilled supporting staff. Only hospitals with an established interventional cardiology programme (open 24 hours/7 days) should use primary PCI as a routine treatment option. High-volume operators have better results than low-volume ones, even in the era of routine coronary stenting [21]. PCI in hospitals with off-site cardiac surgery backup can improve the rapid access to primary PCI for a larger section of the population, and the therapy can be delivered with a very favourable safety profile [22–26]. There are significant financial and geographical limitations to the widespread use of primary PCI.

Currently, primary PCI is widely available in rich densely populated countries with well-organized health care systems. Within Europe, the uptake of primary PCI varies, but, within the last years, many countries have implemented primary PCI and established STEMI centres [27]. The Stent for Life initiative has supported the implementation in several countries [28, 29] and continues with this effort. Excellent cooperation between EMS, local hospitals, and PCI centres is vital in avoiding time delays and has justified the creation of local networks for the management of STEMI patients and patients with severe acute cardiac conditions.

Rescue PCI is defined as an urgent PCI performed on a coronary artery which remains occluded, despite thrombolytic therapy. The identification of these patients remains a challenging issue, but <50% of ST-segment resolution in the lead(s) with the highest ST-segment elevations 60–90 min after the start of thrombolysis and persistent symptoms are useful clues. A clinical suspicion is confirmed at angiography by demonstrating a culprit lesion in an epicardial artery with impaired flow (< TIMI 3). A randomized study and a meta-analysis showed that rescue PCI is associated with a significant reduction in heart failure and reinfarction and a trend towards a lower all-cause mortality, when compared with a conservative strategy, at the cost, however, of an increased risk of stroke and bleeding complications [30, 31]. Rescue PCI should be performed when clinical or ECG evidence of a large infarct is present.

Facilitated PCI is defined as a pharmacological reperfusion treatment delivered prior to an emergent planned PCI, in order to optimize antithrombotic therapy during the PCI-related time delay. This concept is intuitive and logical. Full-dose thrombolytic therapy, half-dose thrombolytic therapy with a glycoprotein IIb/IIIa inhibitor (GPI), and GPI alone have all been tested for this indication. There is no evidence of a significant clinical benefit with any of these agents [19, 32–35]. In spite of the fact that pre-PCI patency rates were higher with thrombolytic-based treatments, no mortality benefits, but only more bleeding complications, were observed. The pre-PCI patency rates with upfront abciximab or a bolus dose of tirofiban alone were not higher than with placebo. Facilitated PCI, as it has been tested so far, cannot be recommended.

The so-called pharmaco-invasive approach is defined as a first-line treatment with thrombolytic agents (half- or full-dose thrombolysis), followed by rescue PCI, if needed, or routine non-emergent coronary arteriography with PCI (if needed) within 24 hours of the administration of the thrombolytic treatment. This approach has been found to provide similar benefit on ischaemic endpoints (with a higher incidence of bleeding) in STEMI patients with a presentation within 3 hours.

Some patients present late, >12 hours after the onset of symptoms. If there is clinical or ECG evidence of ongoing ischaemia, then primary PCI should be performed up till 48 hours from symptom onset. Unfortunately, as many as 30% of patients present late with signs of a fully developed infarction. The open artery hypothesis suggested that late patency of an infarct artery would prevent an adverse left ventricular (LV) remodelling, increase the electrical stability, and provide collateral vessels to other coronary beds for protection against future events. The OAT (Occluded Artery Trial) evaluated routine PCI performed at least 24 hours from symptom onset on a totally occluded infarct-related artery with poor or absent antegrade flow. There was no benefit of routine PCI, compared to medical therapy alone, with even a trend towards excess reinfarction, during 4 years of follow-up [36]. The OAT exclusion criteria included NYHA class III or IV of heart failure, rest angina, left main or three-vessel disease, clinical instability, or severe inducible ischaemia on stress testing if the infarct zone was not akinetic or dyskinetic.

This topic is discussed in detail in Chapter 44, and we therefore briefly describe here a practical approach to antithrombotic management, as recommended by current ESC guidelines.

The combination of parenteral anticoagulation and DAPT (aspirin and a P2Y12 inhibitor) is indicated in patients with ACS undergoing PCI. An individual approach for antithrombotic management should be weighted in patients with a high risk of bleeding.

Aspirin 150–300 mg is usually administered as soon as possible, after the confirmation of the diagnosis, either orally or intravenously, and oral therapy with a 75–100mg dose is recommended life long.

The preferred ADP receptor blockers in ACS are now prasugrel and ticagrelor. Both drugs have a more rapid onset of action and a greater antiplatelet efficacy than clopidogrel, and they have proved to be superior to clopidogrel in large outcome trials [37, 38]. Prasugrel is indicated in ACS patients in whom the coronary anatomy is known, when the indication for PCI has been established. Ticagrelor is recommended for all ACS patients, regardless of the initial treatment strategy, including those pre-treated with clopidogrel. Administration of ticagrelor or clopidogrel should be started as early as possible. Prague 18 study compared prasugrel and ticagrelor in primary PCI patients without any difference. Please add NEW reference (4) In NSTEMI patients, pretreatment with prasugrel is not recommended (ACCOAST trial). Cangrelor is an adenosine triphosphate reversible analogue, with a short half-life, administered intravenously; it has been compared to clopidogrel in three CHAMPION trials [39] and might be clinically useful in bridging of DAPT to cardiac, or even non-cardiac, surgery. In STEMI patients, the recent PRAGUE-18 study compared prasugrel with ticagrelor and did not find any difference in ischaemic nor bleeding events and was terminated for futility.

Anticoagulation with UFH is ideally administered in 50–70 IU/kg as an initial dose to allow the safe use of GPIs, if needed, in the catheterization laboratory. Regarding other anticoagulant options, fondaparinux did not show any benefit in the setting of primary PCI and is therefore not recommended in STEMI patients [40] but is highly recommended in NSTEMI patients (OASIS 5). Enoxaparin (0.5 mg/kg IV) showed, in one randomized study, a trend toward a net clinical benefit, without an increase in bleeding, and is therefore an alternative to UFH [41]. Bivalirudin (a direct thrombin inhibitor) was compared to a combination of UFH plus frequent use of GPI, and this resulted in a reduction in major bleeding, but at the expense of a higher rate of stent thrombosis and similar results in terms of early mortality [42]. The large HEAT-PPCI trial compared bivalirudin to UFH (with GPI added only in bailout indications in both groups). UFH reduced the ischaemic endpoints, with no increase in bleeding; this finding confirms the routine use of UFH in primary PCI [43]

The use of GPIs is usually restricted to patients with a large thrombus burden and to severely haemodynamically compromised patients. There is no definitive answer regarding the current role of the routine use of GPIs in patients with ACS undergoing PCI in the era of potent DAPT with prasugrel or ticagrelor.

The ‘no-reflow’ phenomenon in STEMI patients is characterized by an inadequate myocardial reperfusion at the microcirculatory level, after a successful reopening of the epicardial infarct-related artery, without evidence of a persistent mechanical obstruction. The mechanisms of no-reflow are not fully understood—distal thrombotic microembolization, vascular reperfusion injury, adrenergic microvascular constriction, and myocardial oedema may contribute [44]. Depending on the definitions used, 10–40% of patients undergoing reperfusion therapy for STEMI may show evidence of no-reflow [45–47]. A grading system was developed by the TIMI study group for assessing the epicardial flow in infarct-related coronary arteries [48] (see Box 47.1). However, a substantial number of patients with TIMI 3 flow have persistent ST-segment elevation on the post-angioplasty ECG, and the primary objective of reperfusion therapies is not only the restoration of blood flow in the epicardial coronary artery, but also the reperfusion of the infarcted myocardium. This can be judged angiographically by the ‘myocardial blush grade’ (MBG) which uses myocardial contrast density as a measure of the functional integrity of the microvascular bed. [49] No-reflow can cause prolonged myocardial ischaemia, may result in severe arrhythmia and critical haemodynamic deterioration, and is associated with a significantly increased risk of clinical complications [47, 49–51].

The intracoronary administration of vasodilators, such as adenosine, verapamil, nicorandil, papaverine, cyclosporine, and nitroprusside (see Table 47.1), during and after primary PCI has been shown to improve the flow in the infarct-related coronary artery and myocardial perfusion, and/or to reduce the infarct size, but large prospective randomized trials with hard clinical outcomes are lacking [44]. In some cases, the distal injection of vasodilating medication with an aspiration catheter or a microcatheter may improve the flow in an infarct-related artery with the no-flow phenomenon. Small positive study with early use of high-dose N-acetylcysteine before pPCI has been published but awaits comfirmation in larger studies (5). The reduction of microvascular obstruction remains an important unmet need awaiting further research.

The presence of a coronary thrombus creates special challenges in the performance of primary PCI. A large thrombus burden is associated with an increased incidence of distal embolization and no-reflow and may limit reperfusion at the microvascular level. A number of adjunctive strategies has been tried. Unfortunately, the use of sophisticated mechanical thrombectomy systems was associated with a larger infarct size and an unexpected increase in mortality. Consequently, mechanical thrombectomy is now used infrequently with primary PCI [52]. Distal protection devices are represented by filters or proximal balloon occlusion systems. There are nine randomized trials comparing primary PCI with distal protection using filters or balloon occlusion, compared with primary PCI alone; they found that distal protection did not improve myocardial reperfusion or clinical outcomes. Distal protection is not thought to be beneficial with primary PCI for STEMI, except in saphenous vein graft lesions [52].

The simple manual aspiration catheters have two lumens—one for the passage of the catheter over a coronary wire, and the other for the aspiration of the thrombus and atheromatous debris. The TAPAS and EXPIRA trials together randomized over 1200 STEMI patients to aspiration thrombectomy followed by stenting vs stenting alone [53, 54]. Aspiration was successfully performed in 90% of patients (see example in Figure 47.3); a thrombus or an atheroma was retrieved in 72% of patients, and direct stenting (without predilatation) was performed in 59% of patients. The frequency of a high myocardial blush grade (the primary endpoint) and of a complete ST-segment resolution on ECG (the secondary endpoint) was significantly higher with aspiration thrombectomy. These improved results in myocardial reperfusion were associated with a clinical benefit at 1 year, with a lower incidence of cardiac death and cardiac death or MI [53]. Much larger randomized trials TASTE (7244 patients) [55] and TOTAL (10 732 patients) [56–59]—did not show any mortality benefit of the routine use of thrombectomy. Furthermore, the incidence of stroke was increased in TOTAL trial. We suggest that aspiration thrombectomy may be considered only for the treatment of patients with a large thrombus burden.

Stent implantation in the infarct-related artery has several distinguishing features: (a) a large thrombus burden may result in a late stent malapposition, due to thrombus dissolution; (b) a ruptured plaque with a large necrotic core may increase the rate of acute stent thrombosis; (c) the benefit from less neointimal hyperplasia formation, due to DES implantation, may be smaller in the primary PCI setting, due to less viable myocardial tissue present after MI; (d) an emergency situation with a limited patient history presents a challenge in determining the risks of prolonged DAPT, due to non-compliance, non-cardiac operation, etc.

Conflicting data regarding the use of first-generation DES have been reported, and concerns regarding late stent thrombosis and reinfarctions were raised [60, 61].

Newer generations of DES (eluting everolimus, zotarolimus, or biolimus) have been recently tested in patients with STEMI, with encouraging results. A randomized study demonstrated a reduction of the 1-year composite clinical endpoint by implantation of a biolimus-eluting stent, in comparison with a BMS [62]. Implantation of everolimus-eluting stents did not lower the combined clinical patient-oriented endpoint, but it reduced target vessel revascularization and also stent thrombosis rate [63] and clinical benefit was confirmed at 5 years follow-up

In summary, there are enough data to support the routine use of second-generation DES in primary PCI. Optimal stent sizing is of paramount importance. The presence of an infarct-related artery spasm or a thrombus may lead to significant stent undersizing, which is a frequent cause of restenosis and/or stent thrombosis. Bioresorbable vascular scaffolds represent very intuitive idea, initial experience in STEMI setting is available, but there is clear evidence of increased thrombosis rate and the routine clinical use of this technology is not recommended. Deferred stenting as an attempt to preserve distal microcirculation cannot be routinely recommended based on DANAMI 3-DEFER study.

Cardiac catheterization using the radial artery access has become routine in many centres. The radial approach results in quick mobilization of the patient and minimizes the risk of local bleeding complications, which makes it especially attractive in the setting of ACS with aggressive antithrombotic and antiplatelet medications. Recently, several trials have compared radial vs femoral access in a randomized fashion. The large RIVAL trial showed no overall difference between the two strategies, but, in the pre-specified subgroup of patients with STEMI, there was a significant reduction in the combined clinical endpoints and also a lower mortality [69]. This was most prominent in high-volume radial centres. Similar results were observed in RIFLE-STEACS [70] and STEMI-RADIAL [71] which showed a clear reduction in bleeding with the radial approach. The recently published Matrix trial provided further support for routine use of radial access [72]. These data support the ESC guideline recommendations that, for patients with STEMI, the radial access should be considered over the femoral access, if performed by an experienced radial operator. There is a 2–7% conversion to the femoral access, and there seems to be no increase in radiation exposure in experienced centres.

As many as 50% of STEMI patients have multivessel disease. Acute lesions involving bifurcation are also common, and authors would strongly advise to prefer a simple one-stent provisional strategy in this acute setting, thereby minimizing the procedural time and contrast load. Except for patients in cardiogenic shock (CS), only the culprit lesion should be treated in the acute setting; any further intervention is not without risk [73]. This statement is supported by data from the HORIZONS-AMI trial [74] where multivessel PCI in the acute single procedure resulted in significantly higher mortality and stent thrombosis rates. Recently, however, medium sized trials trials [75, 76] showed that multivessel stenting was superior to a strict policy of stenting only the culprit lesion. Primary outcome (different definitions, always composite) was reduced in all of these trials but there was no reduction in mortality or repeat myocardial infarction with most of the benefit derived from reduced repeat revasculariyation. Patients with three vessel coronary disease and critical stenosis (over 90% diameter stenosis) are most likely to benefit from complete revascularization. Thus currently routine revascularization of non-culprit lesions should be considered before hospital discharge. An individual patient assessment is needed.

Emergency PCI in CS may be lifesaving and should be considered at an early stage and performed as soon as possible. There are two important differences to routine primary PCI—the usually recommended time window of 12 hours after the onset of chest pain (CP) is wider, and multivessel PCI on all critical (90% diameter stenosis or more) or angiographically unstable lesions should be considered. The SHOCK trial randomized 300 patients to medical therapy or revascularization (PCI or surgery); 86% of patients received IABP. The 6-month mortality was lower in the revascularization group than in the medical therapy group (50.3% vs 63.1%, P = 0.027). This trial finished enrolment in 1998, when routine stenting for primary PCI was only starting, and therefore revascularization with the current technology might provide even better results [77]. Real-life utilization of IABP during primary PCI for CS is low (20–39%) [78]. A recently published meta-analysis of IABP in CS did not show any efficacy benefit, and, in a large randomized trial, the use of IABP resulted in a significant increase in bleeding complications and stroke [79]. Many studies regarding IABP in the setting of CS are importantly hampered by bias and confounding, and randomized trials were recently performed to clarify the situation. The CRISP-AMI study randomized patients with a large anterior MI to primary PCI, with or without IABP, and did not find any reduction in the infarct size [80]. A large German randomized multicentre trial of patients in CS showed no benefit of IABP at 30 days post-MI (see Figure 47.4) [81]. In summary, routine use of IABP cannot be recommended. Recently available percutaneous LVADs, such as TandemHeart or Impella, are technically feasible and provide superior haemodynamic support but, so far, have not been proven to improve clinical outcomes [82]. Mechanical ventilation with high PEEP should be considered early for patients with hypoxia; it is helpful to stabilize patients prior to PCI [82, 83]. The role of extracorporeal membrane oxygenation is currently studied. In summary, despite significant efforts, CS affects 6–10% of patients with STEMI and remains a leading cause of in-hospital mortality.

Patients presenting with STEMI after CABG are challenging with a larger thrombus load. Angiographic and clinical outcome after primary PCI is, however, similar to that observed in non-post-CABG patients, despite a larger thrombus burden [84]. Information on the number and type of graft is not always available, and this could result in a high contrast and radiation dose. Saphenous vein graft disease behaves quite differently from native coronary atheroma. Thrombolytic therapy has poor efficacy in thrombotic vein graft occlusions. In recent primary PCI trials, the number of patients with prior CABG has been low, so that these trials cannot bring a definitive answer to the question of optimal therapy for such patients [32]. The risk of no-reflow is high, and there is a higher probability of vessel rupture. Embolic protection devices—both proximal balloon occlusion systems and distal filter-based systems—have demonstrated a reduced rate of periprocedural complications. GPIs have not shown any benefit in vein graft PCI. DES have long been disputed in vein graft PCI, but a recent large randomized trial demonstrated a reduction in target lesion revascularization, with no safety concerns [85].

As a result of our collaborative effort to minimize ischaemic time and delays, occasionally a patient undergoes urgent coronary angiography, with no clear culprit lesion identified. We present here a brief differential diagnostic approach to such patients. Coronary artery spasm may result in a temporary coronary artery occlusion, with ST elevation on ECG; serial ECGs are usually helpful, and intracoronary imaging techniques might be helpful by excluding significant coronary atheroma. Stress-induced cardiomyopathy of the Takotsubo type is easily diagnosed with left ventriculography or echocardiography, demonstrating a dyskinetic segment which typically does not correspond to the vascular territory of a single coronary artery. A pulmonary embolism (PE) can result in ST elevation in the anterior leads [86, 87]. Haemodynamically unstable patients with a clinical suspicion of PE might benefit from right heart catheterization (RHC), pulmonary angiography, and possibly local thrombolytic therapy. Aortic dissection may present with ST elevation, and CT aortography or echocardiography should be performed when this diagnosis is suspected. Pericarditis may also be mistaken for STEMI. However, ST elevations are widespread, with no reduction of R wave, and echocardiography is often diagnostic.

Approximately one-third of patients with unstable coronary syndromes have single-vessel disease, and 44–59% have multivessel disease [88, 89]. The role of coronary angiography and revascularization for patients with NSTE-ACS was first studied in 1994 in the randomized TIMI IIIB trial, which hypothesized that early angiography and revascularization would be beneficial in preventing subsequent cardiac events [89]. Since then, numerous trials have addressed this question, with—at first sight—conflicting results. There was significant cross-over; the selection of patients may have been biased towards lower-risk groups. After adjusting for real difference in coronary revascularization, there appears to be a direct relationship between a higher use of revascularization and lower mortality (see Figure 47.5) [90]. There are several meta-analyses comparing a routine invasive with a selective invasive strategy, some of them including studies with out-of-date PCI techniques [91–93]; the most recent analysis (which included well-conducted trials with 5-year follow-up) revealed a significant absolute reduction in cardiovascular death or MI at 5 years [94]. There is a clear consensus that most of the benefit is in biomarker-positive patients. In a sex-specific analysis, biomarker-positive women have a clear benefit from a routine invasive approach, but there was no benefit, and even a trend to more events, in a group of low-risk, troponin-negative female patients.

Thus, the old belief that percutaneous revascularization does not improve mortality in coronary disease, but just improves symptoms, is no longer true. There is a clear mortality benefit for patients with NSTE-ACS.

The issue of timing has been extensively studied and is closely linked to risk stratification. Coronary angiography should be performed urgently, in the same regime as for STEMI, in very high-risk patients with severe ongoing angina, profound or dynamic ECG changes, major arrhythmias, or haemodynamic instability. These patients represent 2–15% of patients admitted with NSTE-ACS; in fact, some of them will have acute occlusion of the left circumflex coronary artery [95–97]. In patients without the above-mentioned life-threatening features, the ideal timing of coronary angiography was debated from two perspectives: (a) an early intervention to prevent ischaemic events that could occur while the patient awaits a delayed procedure; or (b) with an intensive antithrombotic therapy from the start and angiography delayed by a few days, procedure-related complications might be avoided by intervening on a more stable plaque. Results of the ELISA [98], ISAR-COOL [99], TIMACS [100], OPTIMA [101], and ABOARD [102] studies can be summarized as: (1) high-risk patients, defined as having a GRACE risk score [103] >140, derive benefit from early angiography within the first 24 hours; (2) very early (e.g. <2 hours) angiography does not add any incremental benefit but probably does no harm; (3) the timing of angiography is not crucial in low- to intermediate-risk patients but should be performed during the same hospital stay; a strategy may be chosen individually, according to the patient’s, physician’s, or institution’s preference (e.g. efficiency, cost savings, etc.).

There are no randomized data specifically comparing PCI with CABG in the setting of NSTE-ACS. Approximately one-third of patients will have single-vessel disease and are, in most cases, treated with ad hoc PCI. Multivessel disease is present in half of these patients [88]. In very high-risk patients, urgent PCI will usually be preferred, due to a higher surgical risk in emergency situations. However, most patients can be stabilized, and the choice of the revascularization modality can be made in the same manner as for stable disease. These patients should be discussed by the ‘heart team’ (see Chapter 12) and decisions individualized; anatomical criteria (i.e. Syntax score) should describe the complexity of the CAD, and clinical criteria (i.e. EuroScore II) should estimate the operative risk. Fractional flow reserve measurements will help to determine the functional significance of moderate coronary stenoses.

A large observational study of patients with ACS (both STEMI and NSTE-ACS) showed that the 30-day mortality rates were higher in older age groups (65–69 years: 10.9%, 70–74 years: 14.1%, 75–79 years: 18.5%, 80–84 years: 23.2%, ≥85 years: 31.2%, P = 0.001 for trend) [104]. Elderly patients (>75 years) are frequently excluded from RCTs, and therefore evidence-based medicine is lacking. Primary PCI seems to be safe also in elderly patients [105]. GPIs are associated with an increased risk of bleeding. As far as stent selection is concerned, one randomized trial tended to favour DES in the elderly population [106, 122].

Previously, bleeding was considered an inevitable consequence of an effective antithrombotic therapy, and the avoidance of bleeding therefore received little attention. However, recently, it became clear that approximately 5% of patients presenting with ACS experience major bleeding during the next 30 days, and this is associated with a 3.5-fold increased mortality risk which is prolonged and steady throughout 1 year [107–110]. Different classifications of bleeding scales for severity have been developed, some of them based on laboratory values (e.g. TIMI bleeding score) and some clinical such as the GUSTO definition of bleeding: life-threatening (fatal, intracranial, or resulting in haemodynamic compromise), moderate (requiring transfusion), and mild [111, 112]. The Bleeding Academic Research Consortium has been established in 2011, but the resulting classification is rather complicated [113]. The possible mechanisms responsible for the association between bleeding and mortality include: bleeding itself, the discontinuation of antiplatelet or antithrombotic medications, and anaemia with a reduced O₂ delivery. There are also previously unforeseen possible consequences of blood transfusion like NO depletion, resulting in vasoconstriction or decreased O₂ tissue delivery. All data regarding the consequences of bleeding in the setting of ACS are obviously not randomized, but observational, with many potential confounders, and therefore require a cautious interpretation. Older age, female sex, renal insufficiency, baseline anaemia, LMWH administration in last 48 hours, and the use of GPIs and IABP are among the known factors predicting bleeding [98, 114, 118, 119]. Importantly, specific risk scores (such as the CRUSADE score) have been developed in an attempt to assess the haemorrhagic risk in ACS patients. The choice of the arterial approach, an early sheath removal, a very careful arterial puncture site management, and a properly adjusted dosing of antithrombotic medications are critical issues. Persistent hypotension with no obvious explanation after a femoral artery puncture is suspicious of retroperitoneal haemorrhage, and an urgent CT is indicated. Possible indications for a transfusion should be cautious and judged more on clinical, rather than laboratory, parameters. Antiplatelet medications should be restarted as soon as the risk of bleeding allows. Widespread preference of the radial approach will reduce bleeding, but we should keep in mind that not all bleeding is related to an arterial puncture site, with gastrointestinal bleeding being the most common [107]. Advances in stent design may allow a shorter treatment time with DAPT and therefore lower the risk of bleeding.

Contrast medium-induced nephropathy (CIN) is a recognized complication of PCI, defined as an increase in the serum creatinine concentration of ≥25% from the baseline up to 3 days. It may lead to acute renal failure, and it is associated with a significantly increased mortality rate. Patients with STEMI treated with primary PCI are at higher risk of CIN (possibly up to 20% of patients) than those undergoing elective interventions, possibly due to an impaired systemic perfusion, a large volume of contrast medium, and the impossibility of starting renal prophylactic therapies before exposure to contrast medium [115]. So far, the only strategies that are proven effective in preventing CIN are meticulous patient hydration (possibly guided by patient’s haemodynamic status), minimizing the volume of contrast agent, stopping the intake of nephrotoxic drugs, and avoiding short intervals between procedures. N-acetyl cysteine is not effective in CIN prevention and should not be used [116, 117, 121, 123].