48 Coronary artery bypass graft surgery

The surgical management of acute coronary syndrome still remains a challenge for the cardiac surgeon. Although most patients can be managed by percutaneous coronary intervention, for patients with complex multivessel or left main coronary artery disease (high SYNTAX score), in whom percutaneous coronary intervention is not possible or is unsuccessful, urgent or emergent coronary artery bypass graft surgery is the only available option. It is very important for surgeons to determine the optimum timing of surgical intervention, which is usually based on the clinical presentation, coronary anatomy, and biomarkers. Surgeons should be conversant with the different operative techniques, whether off- or on-pump coronary artery bypass graft surgery, that would help in achieving the best possible outcomes in such situations. Early and late survival of patients depends not only on an efficiently executed operation, but also on the competency of the post-operative care delivered. Modern perioperative management is reinforced by the availability of a variety of mechanical cardiopulmonary assist devices, like the intra-aortic balloon pump, the extracorporeal membrane oxygenation, and an array of ventricular assist devices, which aid us in managing very sick patients presenting with cardiogenic shock.

The results of coronary artery bypass graft surgery for acute coronary syndrome, as published in the literature, vary significantly, because of the heterogeneity of patient populations, operative timing, and haemodynamic status, making a comparison of surgical outcomes almost impossible. Only one randomized trial has been conducted to that effect, to date. A heart team approach, involving an interventional cardiologist and a cardiac surgeon, is mandatory to determine the best treatment strategy and achieve the best possible outcomes in patients with acute coronary syndrome.

An expeditious diagnosis and the earliest possible implementation of treatment in patients with unstable angina (UA)/non-ST-segment elevation myocardial infarction (NSTEMI) (see Chapter 46), and ST-segment elevation myocardial infarction (STEMI) (see Chapter 43), can lead to the confinement of further irreversible myocardial damage, with beneficial effects on early and long-term outcomes. In order to achieve an early and successful revascularization, the treatment of acute coronary syndrome (ACS) requires an interdisciplinary integrated approach between family physicians, intensivists, non-invasive and invasive cardiologists, and cardiac surgeons. When an urgent or emergent revascularization is deemed appropriate, percutaneous coronary intervention (PCI) (see Chapter 47) is most commonly used as the first-line therapy in patients with acute coronary syndrome (ACS), due to its immediate and widespread availability, its lesser degree of invasiveness, and the major advances in stent technology [1, 2]. However, in certain situations, such as complex coronary artery disease (CAD) (i.e. left main stem or bifurcation stenosis, in-stent restenosis, or stenosis of the proximal LAD coronary artery), chronic total coronary artery occlusion, or unsuccessful or complicated PCIs, urgent or emergent coronary artery bypass grafting (CABG) becomes necessary in patients with STEMI [3]. For patients with UA/NSTEMI, the selection of PCI or CABG, as the means of revascularization, should generally be based on the same considerations as those without ACS [4]. Due to this difference in the urgency and type of therapy required in patients with or without ST-segment elevation, the ESC [1, 5, 6] and AHA guidelines [2–4] have also been structured separately for the two different manifestations of ACS.

CABG can either be carried out conventionally on an arrested heart, with cardiopulmonary bypass (CPB) support, or with the maintenance of the native coronary perfusion on a beating heart, with or without CPB support. Mechanical assist device systems for short- to mid-term circulatory support offer an excellent option to bridge the crisis period in patients with impending or existing cardiogenic shock (CS). This chapter discusses the indications, surgical techniques, results, and future prospects of CABG in patients with ACS.

For patients with UA/NSTEMI, the indications that favour a CABG procedure over a PCI are the same as those for patients without ACS [4, 7]. In the wake of a lack of RCTs comparing the outcomes of PCI and CABG in patients with left main and/or multi-vessel disease suffering from UA/NSTEMI, Chang et al analysed the 5-year results of a pooled database of patients with UA/NSTEMI enrolled in the SYNTAX (Synergy between PCI with Taxus and Cardiac Surgery), PRECOMBAT (the Premier of Randomized Comparison of Bypass Surgery versus Angioplasty Using Sirolimus-Eluting Stent in Patients with Left Main Coronary Artery Disease) and BEST (the Randomized Comparison of Coronary Artery Bypass Surgery and Everolimus- Eluting Stent Implantation in the Treatment of Patients with Multivessel Coronary Artery Disease) trials, who underwent PCI and CABG. The primary outcome of death, MI and stroke occurred significantly more often following PCI than CABG (18% vs. 13.4%; P = 0.036), and was driven chiefly by a significantly higher rate of MI (7.5% vs. 3.8%; P = 0.0006). In addition, the rate of repeat revascularization was significantly lower in the CABG group than in the PCI group (HR 0.56; P <0.001) (Chang M et al, 2017). Once a consensus has been reached by the heart team (see Chapter 12) that CABG is the choice of revascularization, the timing of surgery should be primarily determined, based on the symptom complex, the coronary pathology and haemodynamic stability of the patient, and, to a lesser extent, on the need for routine and/or additional diagnostic tests (see Table 48.1). There are no randomized trials conducted to date, comparing early with a delayed CABG strategy for patients with NSTEMI. One of the largest cohort studies analysing the effect of timing of CABG in patients with NSTEMI found surgery <6 hours from NSTEMI as an independent predictor of mortality [8]. Contrarily, evaluation of the CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes with Early Implementation of the American College of Cardiology/American Heart Association Guidelines) and ACTION Registry–GWTG (Acute Coronary Treatment and Intervention Outcomes Network Registry—Get With The Guidelines) databases regarding the timing of CABG for NSTEMI and in-hospital outcomes revealed that, although patients operated on >48 hours after the acute event had a higher risk profile, hospital mortality, stroke, MI, and congestive heart failure were similar to those operated on within 48 hours [9]. Braxton and colleagues similarly showed that non-Q wave MI patients may receive CABG surgery at any time, with similar outcomes to non-MI patients [10]. Our group recently evaluated 758 consecutive patients with NSTEMI undergoing CABG and showed no difference in in-hospital mortality and long-term outcomes in patients operated within 24 hours of infarct, compared to those operated after 3 days, despite a higher risk profile. In contrast, CABG performed between 24 and 72 hours showed a non-significant trend towards poorer long-term outcomes [11]

In the Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease study (APPROACH), the time from admission to CABG, in patients operated on beyond 48 hours (non-emergent) after the index admission, was not associated with an increased risk of short-term mortality [12].

Hence, we believe that, for patients with NSTEMI, decision making about the timing of surgery is dependent more on the clinical presentation and status of the patient, rather than the time elapsed since the acute event. Patients with persistent or recurrent chest pain (CP)/angina equivalent, electrocardiogram (ECG) changes and/or rising cardiac biomarker levels refractory to medical therapy, ventricular arrhythmias, subtotal non-collateralized or severe distal left main coronary artery stenoses, inducible ischaemia or inadequate flow reserve measurements, and/or haemodynamic instability with impending or manifested CS should be operated on an emergent basis, i.e. before the beginning of the next workday (see Figure 48.1). Our study demonstrated that delaying surgery in such patients, with the intent of stabilizing them, may worsen their clinical status with time, resulting in poorer outcomes, which could probably explain the non-significant trend towards inferior results in patients undergoing CABG between 24 and 72 hours after NSTEMI [11]. Additionally, patients who have undergone a failed or complicated PCI should undergo emergent CABG. Furthermore, CABG should be considered on an urgent basis (within the same hospital stay) in patients with left main or three-vessel disease involving the proximal LAD artery and those with a large area of the myocardium at risk. Conversely, patients with long-standing well-collateralized coronary lesions, with stable haemodynamic parameters, usually respond favourably to medical therapy and do not require immediate surgery. Such patients can undergo an early elective operation (within 3–4 weeks). In hospitals without the availability of cardiac surgery, transfer to a tertiary care centre with a cardiac surgical facility after completion of the necessary routine diagnostic procedures is essentially the preferred strategy.

Previous studies have shown an increase in the risk of mortality after emergency CABG early after STEMI. A review of the New York State Cardiac Surgery Registry [13], which investigated the effect of the timing of CABG after a transmural AMI, reported an overall mortality of 3.3%, which decreased as the timing of CABG increased after a transmural AMI. Multivariate analysis showed that CABG within 3 days of a transmural AMI was an independent predictor of mortality. Thielmann and colleagues [14], in a retrospective review of 138 patients with STEMI undergoing CABG, identified the time to operation as a major predictor of morbidity and mortality on unadjusted univariable and risk-adjusted multivariable logistic regression analysis. Similarly, Weiss and colleagues [15], in a retrospective review of the California Discharge Data, identified 9746 patients with STEMI undergoing CABG. Patients undergoing early CABG (0–2 days post-AMI) had a significantly higher mortality than the late CABG group (day 3 or later) (5.6% vs 3.8%, P < 0.001). They suggested that CABG may best be deferred for at least 3 days after admission for an AMI. Hence, emergency surgical revascularization in STEMI is confined to a very select group of patients. These include patients with symptoms and signs of ongoing ischaemia, who have a coronary anatomy unsuitable for PCI, or who have already undergone an unsuccessful PCI or developed severe heart failure or CS [3]. Patients developing complications of STEMI, such as acute MR, VSD, or papillary muscle or free ventricular wall rupture, also need emergent surgery [3].

Not uncommonly, the infarct-related artery can be successfully acutely reopened, but it unmasks an accompanying multivessel CAD during angiography, which essentially requires CABG. In this situation, an urgent CABG is recommended after a few days of cooling off, following the acute infarct, or earlier in the case of recurrent ischaemia, haemodynamic instability, or critical coronary anatomy.

Nevertheless, no randomized controlled trials assessing the optimal timing of CABG after STEMI exist in the literature, and the above-mentioned studies are also almost a decade old. Advancements in operative and anaesthesia techniques and the ever-improving post-operative ICU care could definitely have a positive impact on the outcomes of emergent CABG following STEMI. Khan et al. recently demonstrated no difference in the 1-month and 1-year mortality in patients undergoing CABG within and after 24 hours following STEMI [16]. Grothusen and colleagues recently reported excellent outcomes in patients undergoing early surgery (mean 5 hours) following STEMI. The significantly lower 30-day, 5− and 10-year mortality in STEMI than in NSTEMI patients (2.7% vs 6.6%, P = 0.04; 87% vs 73%, P<0.001; 74% vs 57%, P<0.001) undergoing CABG was probably due to the younger age, lower risk profiles and earlier timing of surgery in the STEMI patients [156]. A more recent study revealed that the haemodynamic profile of the patients is more important in determining the postoperative outcomes than the timing of surgery itself [18].

A randomized controlled trial would be justified in the present era to answer this question.

It is well-known that women are less likely to receive diagnostic coronary angiography and subsequent revascularization than men, even in patients with ACS [17]. A recently published Danish study including 52 565 patients, which comprised 36% women, found that women underwent DCA less frequently than men, both within 1 day (31% vs 42%; P <0.001) and within 60 days (67% vs 80%; P <0.001) after hospital admission [18]. Additionally, women were less likely to undergo PCI (58% vs 72%; P <0.001) and CABG (6% vs 11%; P <0.001) within 60 days of hospitalization than men. The differences, which remained constant over the period of this study from 2005 to 2011, could probably be explained by several factors such as physicians’ tendency to underestimate the risk of ACS [19] in women, lack of benefit from an early invasive strategy in women [20], the greater possibility of non-obstructive CAD in women (22% of women; more than twice as frequent as in men) [18], caused by coronary artery spasms, rupture or erosion of eccentric plaques, microvascular disease including Takotsubo syndrome observed more often in women and the higher likelihood of unfavorable periprocedural factors such as small vessel size, tortuosity, and potential complications in women. The impact of this sex-related discrepancy in management of ACS on clinical outcomes should be the focus of future research.

Although a ‘heart team’, first introduced by the SYNTAX trial [21], may not be required for STEMI or unstable NSTEMI patients, who primarily undergo culprit-lesion PCI [22], the never-ending debate about the choice of revascularization technique, whether CABG or PCI, for haemodynamically stable patients with UA/NSTEMI still exists [1, 23, 24] and warrants a heart team approach. This is especially true in patients with isolated proximal LAD disease, which is involved in as many as 15% of patients presenting with ACS [25] and is associated with worse prognosis [26]. An analysis of 842 patients undergoing revascularization of the proximal LAD culprit lesion in the ACUITY trial (PCI: n=562, 66.7%; CABG n=280, 33.3%) revealed no differences in rates of death, MI, MACE, and stroke for the two revascularization strategies, but PCI patients had significantly higher revascularization rates at 1 month and 1 year, which was mainly driven by target lesion failure. In contrast, CABG was associated with higher peri-procedural major bleeding (8.1% vs 54.4%; P < 0.001) [27]. The same applies to patients with left main disease, with low- to intermediate-risk SYNTAX scores (available at: <http://ir-nwr.ru/calculators/syntaxscore/frameset.htm>) [28]. A definite algorithm, based on the SYNTAX score [29], should be drafted by the heart team at each institution, delineating the criteria for selection of the appropriate therapy. In fact, the development of definite algorithms, based on the SYNTAX score [29], drafted by heart teams, or risk stratification scores, by combining the SYNTAX score with relevant preoperative clinical parameters (logistic Clinical SX Score, SYNTAX II score, Global Risk Approach), to aid in determining the ideal revascularization strategy in haemodynamically stable patients with NSTEMI could be the foreseeable future [30].

The operative risks of patients can be fairly assessed by scoring systems, which render as valuable prognostic tools to predict early and late outcomes of the procedure for a particular patient. The two most commonly used risk stratification systems in patients undergoing CABG are the STS score and the European System for Cardiac Operative Risk Evaluation score (EuroSCORE), which allow online and offline operative risk calculations of an individual patient (available at: <http://riskcalc.sts.org/STSWebRiskCalc273/de.aspx> and <http://www.euroscore.org>) [31–33]. Several preoperative parameters are recorded into a computer-based system before the operation to calculate the actual percentage of the predicted risk. These include the timing of surgery, age, gender, prior heart surgery, race (only for STS score), left ventricle ejection fraction (LVEF), haemodynamic status, percentage of stenosis of the left main coronary artery, and the number of major epicardial coronary arteries with >70% stenosis (only for STS score). Long-standing comorbidities, like diabetes (only for STS score), peripheral vascular disease, chronic renal insufficiency, and chronic obstructive pulmonary disease (COPD), are also included in the calculation. Whereas the STS score reliably allows for an estimation of the procedural risk for mortality and morbidity, it is widely accepted that the EuroSCORE overestimates the procedural risk by a factor of approximately 2.5 [34]. Hence, the EuroSCORE investigators recently developed a modified version of this risk scoring method called EuroSCORE II (available at: <http://www.euroscore.org/calc.html>) [35], which performs better in predicting the operative morbidity and mortality after an isolated CABG [36] (see Appendices I and II).

Another score recently introduced is the age, creatine and ejection fraction (ACEF) score, which uses just the age, LVEF, and serum creatinine and appears to be as good as more complex scores in predicting mortality in patients undergoing an elective CABG [37].

With the expanding use of PCI to treat CAD, new risk stratification scores have been recently developed to predict the long-term adverse event rates, according to the type of therapy used. The SYNTAX score, which is a lesion-based scoring system used to quantify the coronary anatomical complexity [28, 38], is used for the short- and long-term risk stratification of patients undergoing PCI, but not so much for CABG [21]. Secondly, the SYNTAX score alone has been found to be inadequate in predicting mortality, when compared with clinical-based scoring systems. Clinical variables correlate well with clinical endpoints such as death or MI [39]. As a result, a number of scoring systems that combine the SYNTAX score with preoperative clinical characteristics have been developed. Two such scores are the Global Risk Classification (GRC), which is a combination of the EuroSCORE and SYNTAX score strata, and the Clinical SYNTAX score (CSS), which is a combination of the SYNTAX score and the ACEF score [40, 41] (see Appendices I and II).

The GRC did substantially enhance the identification of low-risk patients who could safely and efficaciously be treated with CABG or PCI [42]. However, the most recently developed SYNTAX score II, which contained eight predictors, i.e. the anatomical SYNTAX score, age, creatinine clearance, LVEF, the presence of an unprotected left main CAD, peripheral vascular disease, female sex, and COPD, was found to be a better guide for decision making between CABG and PCI than the original anatomical SYNTAX score, with regard to long-term mortality in patients with complex CAD [43] (see Appendix III).

Surgeons should be aware that these risk stratification scores have not been validated in patients undergoing emergent procedures for ACS. However, these scores are the best guides available at the present time to predict early and long-term outcomes in patients with CAD undergoing PCI or CABG.

A median sternotomy continues to be the standard approach used to perform CABG in patients with multivessel CAD. This is particularly true in patients with borderline haemodynamic stability undergoing urgent or emergent surgery for ACS. It not only enables the surgeon to have an excellent approach to, and vision of, all epicardial coronary arteries, but it also facilitates a quick and easy access to all cardiac structures, in case an emergent implementation of CPB is required. In addition, haemodynamic situation permitting, the internal thoracic arteries (ITAs) can also be harvested in a short time through this approach. As these patients are operated on an emergency basis, routine antiseptic preparation of these patients may frequently not be possible and thus represents a potential risk for the development of post-operative wound infections, not only at the sternotomy site, but also, more frequently, at the site of graft harvest. Thus, harvesting the ITAs in a skeletonized fashion and other conduits using minimally invasive techniques, with the preservation of multiple skin bridges, or through endoscopic techniques, particularly in diabetic patients, when time and haemodynamic stability permit, leads to a lower incidence of post-operative wound infections [44, 45].

In haemodynamically stable patients with ACS due to proximal or mid-LAD lesions not amenable to PCI or failed PCI, a minimally invasive direct coronary artery bypass (MIDCAB) [46] could be considered a potential option, as long as the indication is semi-elective or at the most urgent. This procedure is performed through a small left anterior thoracotomy, through the fourth or fifth intercostal spaces. The left ITA is harvested and anastomosed to the LAD. Performance of this operation in emergency situations, like an iatrogenic dissection of the LAD in the catheterization laboratory, in ACS patients with haemodynamic instability, or in those in CS, should be strictly avoided. Furthermore, this operation should be performed by an experienced surgeon, who can expedite the procedure efficiently.

With the ever growing expertise and experience in performing off-pump CABG (OPCABG), it has also become a valuable option in patients undergoing CABG in the acute phase of MI [47, 48]. Although OPCABG is often criticized for the inferior quality of anastomoses, hence lower graft patency and higher incomplete revascularization rates, when compared to conventional CABG performed on CPB with an arrested heart [49], there is enough evidence to show that there is a uniform reduction in myocardial trauma and the release of markers of myocardial necrosis in patients undergoing OPCABG [50, 51], through the avoidance of ischaemic cardiac arrest and reperfusion injury. In addition, earlier revascularization of the culprit lesion, attenuation of the no-reflow phenomenon, and a reduction of myocardial oedema curtail the extent of myocardial necrosis. Even though these subtle benefits of OPCABG are not evident in patients undergoing elective CABG, it can be assumed that patients with ACS undergoing OPCABG will specifically profit from the preservation of the native coronary perfusion and a correspondingly lesser ischaemia–reperfusion injury. To date, only a few retrospective studies [52–58] and one RCT, comparing the results of OPCABG and conventional CABG in ACS patients, have been published. ‘Time is myocardium’ is a terminology common to interventional cardiologists. OPCABG achieves quick revascularization of the target vessel, thus keeping the ischaemic time of the jeopardized myocardium as short as possible [55]. In this context, the coronary artery supplying the ischaemic territory, which is commonly the LAD territory, is grafted first and reperfused. As a result, not only is the infarct-associated ischaemic time shortened, but it also improves the tolerance of the heart to ensuing manipulations required to graft the lateral and inferior walls. Another advantage of OPCABG is that, in the era of DAPT, surgery can be executed with minimal blood loss and transfusion requirements, even in patients on DAPT close to surgery [59, 60].

In haemodynamically unstable patients, the heart may not tolerate manipulations to expose the coronary arteries. In such situations, preoperative implantation of an intra-aortic balloon pump (IABP) followed by OPCABG enables achievement of acceptable perioperative outcomes and excellent mid- to long-term survival [61]. If haemodynamic stability is not accomplished following IABP implantation, CABG can be performed on-pump, without arresting the heart (beating heart). In a study comparing CPB-supported beating-heart surgery to conventional CABG with cardioplegic arrest in ACS patients, not only was the mortality in the CPB-supported beating-heart group lower (6.5% vs 8.0%), but the occurrence of perioperative MI, excessive post-operative bleeding, and confusion states were also significantly lower [55].

The left ITA (LITA) is the best conduit available to graft the LAD artery [7]. The endothelium of the ITA releases NO and prostacyclin, which are potent vasodilators and inhibitors of platelet function [62–64], and, as a result, has a good response to vasodilators and is less spastic in the presence of vasopressors used in the post-operative period [65–67]. The ITA flow characteristics are comparable to normal coronary arteries. Elami et al. have proven that the non-use of the ITA is one of the most significant independent predictors of a low cardiac output after CABG for an AMI [68]. It is speculated that the ITA could be more resistant to low-flow situations, like the no-reflow phenomenon, than other arterial conduits, which represents a major reason why the LITA remains a very important option for these patients. In addition, LITA bypass to the LAD artery not only improves the survival rate, but it also reduces the incidence of late MI, hospitalization for cardiac events, the need for reoperation, and recurrence of angina [69, 70]. Caceres and co-workers recently proved that ITA grafting was independently associated with a lower risk of mortality and did not seem to compromise outcomes in patients with ACS [71]. The only possible drawback of using the LITA in patients with ACS is that its preparation requires an additional 10–20 min. Although this should not pose a problem in most patients with ACS, surgery in patients in CS should be expedited as quickly as possible. In such patients, the surgeon must either forego the use of the LITA or harvest the LITA after the establishment of CPB. Under emergency conditions, it would also be faster to harvest the LITA as a pedicle, rather than in a skeletonized fashion.

Due to the excellent short- and long-term outcomes of CABG performed with the LITA, one would expect that the use of the right ITA (RITA), in addition, would further improve the results. This fact has been confirmed by a number of studies published in the literature [72–74]. However, the use of bilateral ITAs (BITAs) is recommended only in those patients with ACS, who are haemodynamically stable with a good LV function and in whom one does not expect the use of high-dose inotropes post-operatively. ITAs also require a higher perfusion pressure to maintain an adequate myocardial blood flow. Hence, BITAs would be more commonly harvested in stable patients with UA/NSTEMI, as opposed to those with STEMI.

The third most popular arterial graft used in CABG is the radial artery. It is a muscular artery and is therefore highly susceptible to spasm, especially in the presence of high-dose vasopressor support [75, 76]. Secondly, the chances of graft failure are higher if it is used to graft a coronary artery that is not severely stenosed, i.e. <70% for left-sided and <90% for the right coronary artery [77]. The overall long-term patency of the radial artery is excellent but is highly influenced by the site of the proximal anastomosis and competitive flow [78]. The indications for its use in patients with ACS are similar to the ones for the use of a second ITA graft, although there are reports about the safety of its use in patients with moderate to severe LV dysfunction [79].

The saphenous vein is one of the most commonly used conduits in CABG, especially in emergency situations, because it is easy and quick to harvest and graft and is resistant to vasospasm, thus potentially being a safe option in patients on high doses of vasopressors. It can be harvested easily by using open, semi-open (bridged), or endoscopic techniques [80].

ACS can result in the development of some deadly complications, which are life-threatening in nature. This section will focus primarily on CS and acute mechanical complications like ventricular septal rupture, papillary muscle rupture causing severe MR, and FWR of the LV.

CS (see Figure 48.2), which has been discussed in detail in Chapter 49, is one of the commonest causes of death in patients presenting with ACS [81], with hospital case fatality rates being almost up to 60% [82]. Acute coronary revascularization positively influences short- and long-term outcomes in patients with ischaemia-related CS [83, 84]. The window period between FMC and PCI has been liberalized to 12 hours, even for multivessel PCI [1, 85]. Emergency CABG in patients with a coronary anatomy not suitable for PCI, or those undergoing an unsuccessful or a complicated PCI, has to be, most often, performed on CPB to ensure the maintenance of an adequate end-organ perfusion. However, our strategy, for the past decade, to perform bypass grafts on CPB, without clamping the aorta, i.e. on a beating heart [53, 86], was associated with a lower perioperative morbidity and mortality, with similar long-term results to conventional CABG with ischaemic arrest [55]. In contrast to PCI, CABG on CPB offers: (1) acute volume unloading of the LV until revascularization is achieved, (2) a higher possibility of complete revascularization, and (3) the option of implanting an assist device for hearts which require a longer recovery period for the hibernating myocardium and reperfusion injury.

Most mechanical complications, described in Chapter 45, occur in the first 24 hours after an AMI but may ensue even 1 week thereafter. They include free left ventricular wall rupture (LVWR), post-infarct ventricle septum defect (VSD), and acute MR and have been discussed in greater detail in Chapter 45. Prompt repair (with or without CABG) is indicated in most cases.

FWR leads to acute pump failure within minutes, electromechanical dissociation, and death [87]. A lifesaving operation is possible only in selected cases, as a very small number of patients are capable of reaching the hospital, with mortality rates approaching 60% [88].

Post-infarct VSD is seen in approximately 1% of AMI patients and has a 1-year mortality of >90%, if not operated upon. Definitive treatment consists of an urgent/emergent closure of the VSD with a patch (Dacron or bovine pericardium), which should be performed, even in haemodynamically stable patients, because the rupture site can expand abruptly, resulting in sudden haemodynamic collapse in previously stable patients [89]. Although the appropriate timing of surgical repair is elusive, patients in CS, due to a large left-to-right shunt volume, should undergo emergent surgery. In haemodynamically stable patients, surgery can be delayed for 3–4 weeks, during which the patient can be optimized with inotropic and mechanical support [90]. Despite surgery, hospital mortality still ranges from 20% [91] up to around the 50% mark [92, 93], although it did reduce in the last decade of the 20th century [94]. Alternatively, there are reports of transcatheter VSD closure using interventional occluder devices, but with equally bad outcomes [95].

Acute MR develops due to: (1) papillary muscle rupture or dysfunction and (2) dilatation of the mitral valve annulus due to infarct-related LV dysfunction. Treatment of patients with acute MR with associated pulmonary oedema and/or CS incorporates emergent mitral valve surgery. Surgical treatment most commonly involves mitral valve replacement to keep the aortic clamp time as short as possible. Alternatively, mitral valve repair can be attempted in more suitable valve pathologies and stable haemodynamic conditions, using chordal transfer or replacement techniques, in combination with a ring annuloplasty. However, valve repairs in these patients always bear a higher than average risk of failure. Overall 30-day mortality still remains higher than 25% [96, 97]. There have been recent case reports describing successful percutaneous mitral valve repair in such patients with the MitraClip system [98].

The perioperative care of patients undergoing emergency CABG is virtually similar to that of any other patient undergoing an elective CABG or heart operation (as outlined in Chapter 77), with a few points that need to be alluded to. Patients with ACS (especially STEMI) need to be aggressively monitored and stabilized preoperatively in the ICCU, so that the haemodynamics are adequately optimized at the time of surgery. Bedside routine invasive monitoring with peripheral and PACs, mechanical support with IABP, and/or extra-corporeal membrane oxygenation (ECMO) (see Chapter 30) for patients in frank CS are some of the ardent measures that may have to be taken for preoperative optimization in this patient subset (see Chapter 30). Ischaemia–reperfusion injury is a well-known phenomenon in patients undergoing on-pump CABG, especially as emergency procedures for ACS. Even in OPCABG, short periods of regional myocardial ischaemia are produced during target vessel anastomosis, resulting in myocardial injury, the severity of which varies, according to the area of distribution of the vessel and the extent of collateralization. Obviously, increasing the number of grafts performed for multivessel CAD could theoretically put greater myocardium at risk of injury, particularly in patients with evolving MI, UA, or both. The beneficial effect of glucose–insulin–potassium (GIK) solution, as a cardioprotective agent after cardiac surgery, has been controversial [99, 100]. However, a recent meta-analysis addressing the effects of GIK solution in adult cardiac surgical patients [101] and a RCT in patients undergoing OPCABG for ACS [102] revealed a significant reduction in myocardial injury and an improvement in haemodynamic performance, when the GIK solution was initiated during surgery and continued for 6 hours after reperfusion. The majority of patients with ACS also receive aggressive antiplatelet therapy, which has been detailed in Chapter 44. The use of aspirin following CABG is associated with a significant decline in non-fatal MI, non-fatal stroke, or vascular death in patients with unstable angina and acute MI [103] and a 40 % reduction in bypass graft occlusions [104]. As per the ESC guidelines, aspirin should be given to all patients without contraindications, at an initial loading dose of 150–300 mg and at a maintenance dose of 75–100 mg daily long-term, regardless of the treatment strategy (see Chapter 44). However, these effects are subject to the patients’ platelet inhibitory responses to antiplatelet therapy (APT), which not only demonstrate diversity between subjects, but also within the same subjects at different time points that are often influenced by different clinical conditions [105]. Additionally, prevalence of aspirin resistance, which hinders platelet inhibition, appears to increase significantly in the early post-operative period [106], probably as a result of CPB-induced inflammation and platelet hyperactivity leading to an increase in potential for occurrence of adverse ischaemic events. DAPT, with addition of a P2Y₁₂ inhibitor such as clopidogrel in patients with aspirin resistance, has not been found to reduce the incidence of adverse ischaemic events or death, at least after elective primary CABG [107]. Moreover, DAPT has demonstrated no impact on saphenous vein graft intimal hyperplasia or patency at 1-year follow-up [108]. A recently published review reported that only 12 studies concerning the efficacy of the use of DAPT after CABG existed in the literature [109], of which only one was a RCT [108]. Furthermore, the outcomes revealed by them were variable, most likely because of inadequate cohort sizes, poor study design, and heterogeneous surgical population, which lead to conflicting conclusions. The authors recommended that DAPT can be commenced in the postoperative period once the possibility of major bleeding is unlikely. Once implemented, a P2Y12 inhibitor should preferably be maintained over 12 months, unless there are contraindications such as an excessive risk of bleeding [6]. This is obviously a matter of concern in patients requiring emergency CABG.

The use of mechanical circulatory support and perioperative coagulation management, which have been comprehensively reviewed in Chapters 30 and 70, respectively, warrant a brief mention here.

IABP counterpulsation (see Figure 48.3), a class IC recommendation for use in CS in the ESC guidelines [6], can also be used for haemodynamic support during catheterization and/or angioplasty, before high-risk surgery, for mechanical complications of MI or for refractory post-MI UA. In many critical situations, it also enables the surgeon to perform the CABG off-pump. However, the scientific evidence for the benefits of IABP in STEMI and CS still remains controversial [110–112]. The recently published large multicentre prospectively randomized IABP-SHOCK II trial reported that the use of IABP (see Chapter 49) did not reduce 30-day mortality in patients with CS complicating AMI, for whom an early revascularization strategy was planned [113]. Nevertheless, IABP implantation, at least in the context of CABG, remains a class I indication [12]. With regard to the timing of implantation, some reports and meta-analyses [114–117] demonstrated a significantly lower mortality rate in high-risk patients treated with preoperative IABP. We usually implant an IABP preoperatively in patients with STEMI and CS and at the first sign of cardiac failure in NSTEMI patients in our institution. This is supported by a study by Ranucci and co-workers, who proved that postponing the use of IABP may be deleterious in patients with drug-refractory heart failure [118].

Patients in CS with severe LV dysfunction commonly cannot be weaned from CPB after the completion of revascularization, despite maximum inotropic therapy and IABP support. In such cases, a temporary mechanical circulatory support device may be considered as a bridge to recovery or as a bridge to other procedures like LVADs or heart transplant (HTx). Use of these devices is a class IIa recommendation, according to the ESC guidelines [6]. One option is the implantation of a veno-arterial ECMO or an isolated blood pump to support the left and/or right heart function. The main advantages of these systems are the ease of implantation and maintenance of an adequate end-organ function [119], which allows the heart team to decide about further lines of therapy, especially in patients with an unknown medical history, comorbidities, and neurological status. When warranted, the implantation of a permanent assist device (see Chapter 31) may serve as a bridge to HTx or as destination therapy (see Figure 48.3). In general, non-pulsatile assist devices (see Figure 48.4) are associated with acceptable long-term survival rates [120].

Emergent CABG procedures in patients with ACS require a more aggressive coagulation management protocol, because the overwhelming majority are acutely treated with platelet aggregation inhibitors. Acetylsalicylic acid (see Chapter 44), not being associated with an increased risk of bleeding [121], should not be withheld before urgent CABG [3, 122]. It is common knowledge that thienopyridine derivatives, like clopidogrel, ticagrelor, or prasugrel, should be withheld, ideally for 5 days preceding surgery [123] and at least 3 days when the benefits of an urgent revascularization outweigh the risks associated with excessive post-operative bleeding [124, 125]. Among 20,304 NSTEMI patients in the ACTION (Acute Coronary Treatment and Intervention Outcomes Network) Registry (2009–2014) who underwent catheterization within 24 hours of admission and CABG during the index hospitalization, 32.9% received a pre-catheterization P2Y₁₂ inhibitor and the time from catheterization to CABG was longer among patients who received pre-catheterization P2Y₁₂ inhibitor than those who did not (median 69.9 hours vs 43.5 hours, P<0.0001), longer for patients treated with prasugrel (114.4 hours) or ticagrelor (90.4 hours]) compared with clopidogrel (69.3 hours, P<0.0001). P2Y₁₂ inhibitor use was associated with higher risks of post-CABG bleeding (OR 1.33, 95% CI 1.22−1.45), post-CABG blood transfusion (OR 1.51, 95% CI 1.411.62) and the need for surgical or procedural intervention for treatment of bleeding (34% versus 23%, P<0.001). Furthermore, the rates of post-CABG bleeding and transfusion were higher in those undergoing CABG within 5 days of the pre-catheterization dose of P2Y₁₂ inhibitor than those undergoing CABG >5 days afterwards (Badri et al, 2017).Similarly, a Swedish observational study including 2244 ACS patients on acetylsalicylic acid and either ticagrelor (n = 1266) or clopidogrel (n = 978) within the last 14 days before acute or urgent CABG (99.3% with CPB) demonstrated that the difference in the incidence of major bleeding complications between the two P2Y₁₂ inhibitors was mainly driven by a significant reduction in major bleeding complications in the ticagrelor group when clopidogrel/ticagrelor was discontinued 3–5 days before surgery (unadjusted OR 0.39, P = 0.006). Discontinuation of ticagrelor at 3–5 or >5 days before surgery did not affect perioperative major bleeding rates (unadjusted OR 0.93, P = 0.80), whereas discontinuation at 0–3 days was associated with a significantly higher rate of major bleeding compared with both 3–5 days (unadjusted OR 5.17, P < 0.0001) and >5 days (unadjusted OR 4.81, P < 0.0001) [126]. In contrast, clopidogrel-treated patients had a higher incidence of major bleeding complications when discontinued 3–5 compared with >5 days before surgery (unadjusted OR 1.71, P = 0.033). However, if the P2Y₁₂ inhibitors could not be discontinued before surgery, ticagrelor was associated with a higher risk for severe bleeding than with clopidogrel, probably due to its stronger antiplatelet effect [127]. Nevertheless, OPCABG, being associated with a lower risk of bleeding [128, 129], can be performed even within 24 hours of stopping the thienopyridines [3]. Our study also demonstrated a long-term beneficial effect of continuation of P2Y₁₂ inhibitors till the day of surgery [11]. Patients receiving abciximab, an IV GP IIb/IIIa receptor antagonist with biological effects up to 48 hours, has to be discontinued at least 12 hours before CABG or can only be antagonized by platelet transfusion in emergent situations [130]. In contrast, short-acting agents, like tirofiban and eptifibatide, should be discontinued only 2–4 hours [131, 132] before surgery. The performance of extended coagulation profile tests or TE is recommended immediately after protamine antagonization in such patients. This allows for the specific substitution of coagulation factors, platelets, protamine, or antifibrinolytic agents, as deemed necessary [133]. The use of cell-savers cannot be overemphasized in these patients.

Real-life data from large databases are not consistent with respect to the percentage of AMI patients undergoing CABG. The analysis of data from the Research Data Centers of the Federal and State Statistical Offices of Germany for the years 2005, 2007, and 2009 demonstrated an increase in the number of CABGs performed in AMI patients (2005: 9402 [4.6%]; 2007: 10 296 [4.9%]; 2009: 10 501 [5.2%]). The rate of CABG in STEMI and NSTEMI patients in 2009 was 4.3% and 5.7%, respectively [134]. Contrarily, a serial cross-sectional analysis of hospitalizations from 2001 through to 2011, using the Healthcare Cost and Utilization Project Nationwide Inpatient Sample (NIS) in the United States (US) revealed a reduction in CABG use for both STEMI (39% decrease; P_trend <0.001) and NSTEMI (14% decrease; P_trend = 0.005) over the study period [135].

The perioperative mortality in patients undergoing CABG during the acute phase of MI is much higher than that in patients with stable angina, in whom 30-day mortality rates of approximately 1% have been reported [23]. The in-hospital mortality reported from Germany for the year 2009 was 11.6% in STEMI and 7.5% in NSTEMI patients [134]. Despite this fact, early CABG after ACS is justified, if the benefits of an emergent operation outweigh the risks. The timing of emergent CABG after the onset of ACS, the type of MI, and the presence of CS have a major impact on mortality and morbidity.

Comparing various studies is difficult, because patient populations vary enormously, with respect to age, the timing of surgery, haemodynamic stability, and most importantly the type of ACS, whether STEMI or NSTEMI. Analyses portraying excellent results often exclude patient subsets at the highest risk. As a result, the reported operative mortality in patients with ACS varies between 1.6 and 32% [8, 9, 13–15, 118, 119, 136–139] (see Table 48.2).

Although several studies have shown that the adoption of an early invasive strategy for the treatment of patients with UA/NSTEMI, especially those with higher risk scores, is more beneficial [140–143], very few studies, depicted in Table 48.2, have analysed the benefits of early CABG in patients with NSTEMI [144]. Second, the exact timing of early CABG in patients with NSTEMI has also not been adequately investigated. This could probably be due to the fact that, unlike STEMI patients, the timing of early CABG in patients with NSTEMI does not really impact the risk of surgery. Dayan and co-workers have written a best evidence topic, according to a structured protocol [145]. They could find only seven articles in the literature that were of any relevance. A recent publication reported an acceptable 30-day mortality of 2% in patients with NSTEMI undergoing immediate CABG [146]. (See also Chapter 46) The US study by Sugiyama and colleagues showed a significant reduction in in-hospital mortality after CABG for NSTEMI, from almost 5% in 2001 to 2.9% in 2011 (P_trend <0.001) [135].

PCI being the first option for revascularization in patients with STEMI, emergency CABG is performed only in very specific situations (already discussed earlier) and is therefore relatively uncommon. This is chiefly due to the higher in-hospital mortality rate associated with emergent CABG [8, 15, 147], which seems to vary with the timing of surgery after STEMI (see Table 48.2). Contrary to this, DeWood et al. reported an overall in-hospital mortality rate of 5.2% for early CABG after STEMI and showed that CABG performed within 6 hours after symptom onset had a lower in-hospital and long-term (10 years) mortality rate than those undergoing CABG beyond 6 hours after symptom onset [148]. The recently published US study by Sugiyama et al. demonstrated no change in in-hospital mortality for CABG in patients with STEMI. It ranged between 4.5% and 6.2% [135]. ]. Rohn et al, in a series of 135 patients undergoing CABG within 24 hours of STEMI, reported that acute CABG in patients with STEMI can be performed with good results. They demonstrated no significant difference in 30-day mortality between patients operated on within 6 hours and those between 6 and 24 hours after STEMI (5.7% vs 9%, P = 0.45). They, however, found poor haemodynamic status of patients such as LVEF (HR 1.103, P = 0.0024), preoperative ventilation (HR 13, P = 0.00076), preoperative inotropic support (HR 8.6, P = 0.0074), cardiogenic shock (HR 28.7, P = 5.8 × 106) and Killip class at admission (HR 6.9, P = 0.00038) to be the independent predictors of 30-day mortality [18].

For patients in CS, the mortality ranges between 21.3% and 46.7% [83, 128, 139]. However, the SHOCK trial showed that early revascularization was beneficial and that patients undergoing CABG had similar survival rates to those undergoing PCI, despite the former more likely to be diabetic and to have a more complex coronary anatomy [149, 150]. Additionally, mortality after CABG in such patients has significantly declined over the first decade of this century. A retrospective analysis of 508 patients in CS undergoing CABG demonstrated a reduction in in-hospital mortality from 42.2% to 24.6% over a 15-year period. Factors that most commonly impacted mortality were serum lactate >4 mmol/L (OR 4.78, P < 0.0001), STEMI (OR 2.10, P=0.001), age >75 years (OR 2.01, P = 0.03), and LVEF <30% (OR 1.83, P = 0.01). Moreover, hospital survivors had good long-term outcomes (5- and 10-year survival was 64.3±3.0% and 49.8±3.0%, respectively), which justifies the use of surgical revascularization in patients not suitable for PCI [151]. Furthermore, high-risk STEMI patients with CS appear to have better outcomes after CABG, compared to PCI, when the latter cannot achieve complete revascularization [85, 152]. (See also Chapter 49.)

Some patients require cardiopulmonary resuscitation (CPR) en route to the operating theatre or prior to induction of anaesthesia [28]. Such CABG procedures called as salvage operations are relatively rare [153] and are associated with a high mortality [136, 154]. A multicenter study from the Nordic countries evaluated 38 patients undergoing salvage CABG procedures. Hospital mortality was 41% and the 1- and 5-year survival was 50% and 46%, respectively [155]. Of the nine patients who received cardiac massage during sternotomy, only one survived. Although the salvage CABG patients have a high perioperative mortality, mid-term survival of patients who can be salvaged is encouraging.

However, patients with AMI who have been successfully resuscitated following a cardiac arrest may have a better prognosis. A very recent study reported on 129 patients, the majority of them with 3-vessel disease following a STEMI, who underwent CABG within a median period of 4 hours after cardiac arrest. The 30-day mortality rate was 23%. Hypoxic brain injury occurred in 12% of patients, with patients resuscitated for more than 20 minutes being most vulnerable. Most patients (79%) discharged alive showed good neurological outcome according to the cerebral performance category scale [156].

To date, there are only a few non-randomized retrospective studies that compare beating-heart (with or without CPB) and cardioplegic arrest techniques in this patient subset [54, 55, 57, 156–159]. In most studies, one encounters comparable hospital mortality between both groups, although a trend favouring OPCABG procedures is often evident [54]. OPCABG was found to be beneficial with respect to perioperative MI [65], IABP implantation [63, 66], the reoperation rate [55, 159], inotropic requirement [55, 150], acute kidney injury (AKI) [148, 152], stroke rate [53], and the duration of intensive care and hospital stays [55, 156–159]. The ACUITY trial showed that patients undergoing OPCABG had fewer events of bleeding MI, but higher reintervention rates at 30 days, than those undergoing an on-pump procedure. However, at 1 year, there was no difference between the groups in death, total MIs, reinterventions, strokes, or major adverse cardiac events, but there was a lower rate of non-Q wave MI in the OPCABG group [160]. In haemodynamically unstable patients or those in frank CS, our group revealed the beneficial effects of CPB-supported beating-heart surgery, in terms of lower hospital mortality (19.3% vs 33.3%), less bleeding, transfusion requirement, inotropic support, shorter ventilation time, lower stroke rate, and shorter ICU stay than CABG with cardioplegic arrest [55].

Long-term outcomes in patients undergoing emergency CABG are determined not only by the quality of surgery and occurrence of post-operative complications, but also by the severity of coronary atherosclerosis, LV function, age, gender, overall health status, and the presence and severity of associated comorbidities [53]. The occurrence of ischaemic clinical events, stroke, or renal failure after CABG has a significant negative impact on long-term survival [161].

Adherence to treatment guidelines is associated with a gradual lowering of both short- and long-term mortality [162]. Based on current guidelines, the indications for performing an urgent/emergent CABG in patients with NSTEMI remain the same as for patients with stable angina pectoris. On the other hand, CABG should be performed on an emergency basis in patients with STEMI whenever PCI has failed or is not amenable, as in some chronic total occlusive or complex bifurcation or trifurcation lesions. Patients with multivessel disease in CS should also undergo emergent CABG if complete revascularization is not possible with PCI. Although it would be ideal to follow the guidelines strictly for every case, there are, however, many factors that influence the decision making in the ‘real world scenario’. The indications for coronary revascularization ultimately depend on individual patient factors such as the general condition, life expectancy, age, and associated comorbidities.

One of the most important prerequisites for a successful outcome in patients needing emergency surgery is an expeditious diagnosis and immediate referral to a cardiac surgery-capable centre. The cardiac surgeon should be included in the decision-making process (heart team approach) in cases of patients directly referred or transferred to a tertiary care centre and who are potential surgical candidates. It is now a class I recommendation [7]. This also holds true for patients undergoing a high-risk PCI procedure. Surgical backup and collaboration between the cardiologist and cardiac surgeon is mandatory to reduce delay in management and to obtain the best outcome possible. The technique of CABG should ideally be left to the discretion of the operating surgeon, although it should be noted that outcomes are probably more favourable in patients undergoing beating-heart CABG, with or without CPB. Several studies have demonstrated that early mortality still continues to remain the Achilles heel for emergency CABG, especially in STEMI patients, even though long-term outcomes are comparable to patients undergoing elective surgery (see Table 48.2).