Cardiogenic shock (CS) remains the most common cause of mortality in patients with acute myocardial infarction. The SHOCK trial (Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock) demonstrated a survival benefit with early revascularization in patients with CS complicating acute myocardial infarction (AMICS) 20 years ago. After an initial improvement in mortality related to revascularization, mortality rates have plateaued.
A recent Society of Coronary Angiography and Interventions classification scheme was developed to address the wide range of CS presentations. In addition, a recent scientific statement from the American Heart Association recommended the development of CS centers using standardized protocols for diagnosis and management of CS, including mechanical circulatory support devices (MCS).
A number of CS programs have implemented various protocols for treating patients with AMICS, including the use of MCS, and have published promising results using such protocols. Despite this, practice patterns in the cardiac catheterization laboratory vary across health systems, and there are inconsistencies in the use or timing of MCS for AMICS. Furthermore, mortality benefit from MCS devices in AMICS has yet to be established in randomized clinical trials.
In this article, we outline the best practices for the contemporary interventional management of AMICS, including coronary revascularization, the use of MCS, and special considerations such as the treatment of patients with AMICS with cardiac arrest.
Ardiogenic shock (CS) represents an inability of the heart to maintain an effective cardiac out- put commensurate to the metabolic demands of the body attributable to a primary underlying cardiac pathology. Acute myocardial infarction (AMI) is the most common cause of CS.1 Although the incidence of ST-segment–elevation myocardial infarction (STEMI) is decreasing, the incidence of CS complicating AMI (AMICS) remains stable (7%–10%) if not increasing, es- pecially among the elderly.2 An array of acute or acute- on-chronic insults can contribute to its pathogenesis, including exacerbations of ischemic heart disease, val- vular disease, cardiomyopathy, pericardial disease, or arrhythmia. Regardless of cause, CS results in a defi- ciency of end-organ perfusion that is often character- ized by hypotension, tachycardia, peripheral vasocon- striction, pulmonary and systemic venous congestion, decreased urine output, altered sensorium, acute liver or kidney injury, and lactic acidosis.3–5 Although CS re- mains a clinical diagnosis, objective definitions have been established by clinical trials,6,7 and a recent docu- ment has proposed a novel classification system based on clinical characteristics at presentation.8
Mortality associated with AMICS remains high, with 30-day mortality approximating 40% to 45% in con- temporary randomized trials.7,9 After the SHOCK trial (Should We Emergently Revascularize Occluded Coro- naries for CS),6 which demonstrated survival benefit with early revascularization in AMICS at longer follow- up, and with growth in availability of primary percu- taneous coronary intervention (PCI), AMICS-associated mortality declined.10 From 2005 to 2013, this improve- ment appeared to plateau in an analysis of the National Cardiovascular Data Registry despite increasing rates of PCI.11 Mechanical circulatory support (MCS) devices are increasingly used in AMICS, but their effect on mortality has yet to be established in randomized clinical trials.4
The AMIS registry (AMI in Switzerland) of 83 Swiss hospitals documented a decrease in AMI mortality from 8.7% to 7.3% from 1997 to 2017 (P<0.001 for trend)12 and a decrease in development of CS in hospital from 7.8% to 3.5% over the same time period. This was offset, however, by an increase in CS at presentation from 2.5% to 4.6%. Overall, in-hospital mortality of all patients with AMICS decreased from 62.2% in 1997 to 36.3% in 2017 (P<0.001 for temporal trend; Figure 1), likely related to the growth in primary PCI. Of note, patients with AMICS who survive to hospital discharge continue to experience a higher rate of mortality after discharge. In a large series of patients ≥65 years of age surviving to hospital discharge in the ACTION registry (Acute Coronary Treatment and Intervention Outcomes Network), mortality was higher at 60 days (9.6% versus 5.5%) and at 1 year (22.4% versus 16.7%) in patients with AMI with CS compared with patients with AMI without CS.13
In recent years, multiple centers have developed critical pathways and protocols to organize acute invasive care for AMICS with promising results.14,15 Whereas random- ized controlled trials have examined discrete elements of care, including strategies for coronary revascularization,6,9 vasopressor selection,16,17 and MCS,7,18 no contemporary trial has validated a comprehensive algorithm for acute care delivery. In particular, important uncertainties remain in the appropriate use, selection, and management of MCS devices in patients with AMICS. Recognizing these gaps in knowledge, we set out in this scientific statement to critically appraise current evidence, identify areas of consensus and controversy, propose best practices, and highlight necessary areas for future research in the acute invasive management of AMICS.
The shock state, although generally characterized as a lack of end-organ perfusion, has been notoriously diffi- cult to define and classify, largely because the syndrome of shock can be heterogeneous with varying timelines of development. The National Cardiovascular Data Registry’s CathPCI registry, for example, defines shock as >30 min- utes of systolic blood pressure <90 mm Hg, cardiac index <2.2 L·min−1·m−2 determined to be secondary to cardiac dysfunction, or the requirement for inotropic or vaso- pressor agents or MCS.11 Selected statewide databases use different definitions (eg, systolic blood pressure <80 mm Hg despite vasopressors). Heterogeneity of defini-tions propagates uncertainty in comparisons of outcomes across the nation. Furthermore, these definitions may fail to capture patients in preshock or early shock who are at risk for hemodynamic deterioration or mortality.
To address this gap, the Society for Cardiovascular Angiography and Intervention (SCAI) has introduced a classification scheme for a patient’s hemodynamic state.8 Recent publications validated this classifica- tion.19,20 In a series of 10004 patients admitted to the Mayo Clinic cardiac intensive care unit, 43.1% had acute coronary syndromes, 46.1% had heart failure, and 12.1% presented with cardiac arrest.19 After mul- tivariable adjustment, there was a stepwise increase in risk of hospital mortality with increments of SCAI shock stages A to E. In a separate series of 1007 patients pre- senting with CS or large AMI (51% with a preceding cardiac arrest), a stepwise increase in 30-day mortality was again observed in shock stages A to E (Figure 2).20 An important aspect of the SCAI classification is a car- diac arrest modifier. At every stage of SCAI shock, the presence of cardiac arrest significantly increases mor- tality. Hence, this classification appears useful to risk- stratify hospitalized patients, and its gradual universal adoption may reasonably enhance country-wide shock metrics. Future studies are required to prospectively test the clinical utility of this classification scheme and to study the relative predictive value of each element used to define specific SCAI stages.
On AMICS recognition, viable patients with spontane- ous circulation should be brought to the cardiac cath- eterization laboratory of a PCI-capable hospital as soon as possible. Early echocardiography and laboratory examination (arterial blood gas, lactate) are important and can be performed in the cardiac catheterization laboratory with limited delay, taking advantage of the patient transfer in time for preparation.
Classification, stabilization, and diagnostic evalu- ation of AMICS are prerequisites to tailored invasive therapy. Stable patients with risk factors for shock (stage A) or early shock (stage B) can generally pro- ceed directly to coronary angiography and culprit lesion revascularization with continuous reassessment for signs and symptoms of progression of shock. Patients presenting in shock (stages C–E) may first require acute stabilization with attention to blood pressure, end- organ perfusion status, oxygenation, and acid-base status. Especially in cases of STEMI, any necessary sta- bilization efforts must be expedited to minimize delay to reperfusion therapy.21,22 Selected patients with late or extreme forms of shock (stage E) for whom inva- sive management is inconsistent with goals of care and unlikely to provide benefit should instead be evaluat- ed for palliative care. It is important to note that early engagement of palliative care services and aggressive early invasive management are not mutually exclusive. Whereas only 4.5% of patients hospitalized for AMICS between 2000 and 2014 in the National Inpatient Sample received palliative care services,23 it is likely that more patients across the spectrum of AMICS can ben- efit from early engagement in discussion of values and goals of care in parallel with invasive measures.
The minimum necessary dose of vasopressor should be used to maintain mean arterial blood pressure >65 mm Hg, favoring norepinephrine as first-line therapy.16,17 Alternative agents may be preferred in addition to or in- stead of norepinephrine in specific circumstances such as unstable bradycardia, in which case the increased chro- notropic effect of dopamine or epinephrine may be de- sired; dynamic left ventricular (LV) outflow tract obstruc- tion, for which a pure vasopressor such as phenylephrine or vasopressin may be preferred; or refractory hypoxemia or acidosis, in which case efficacy of catecholamine vaso- pressors may be attenuated, favoring the use of vasopres- sin. Of note, the mean arterial blood pressure target of 65 mm Hg is not well established, obligating attentiveness to clinical perfusion status. Caution is required in the pro- gressive escalation of vasopressor and inotrope therapy, noting that higher levels of pharmacological support are associated with higher mortality in observational studies, although this may reflect in part the severity of illness.24 Ongoing studies are evaluating the adjunctive role of mil- rinone, levosimendan, and dobutamine in different shock settings. However, these inotropic agents may be of lim- ited value for initial stabilization in AMICS because of an increased risk for worsening myocardial ischemia.
AMICS predisposes to hypoxemia (resulting from car- diogenic pulmonary edema) and metabolic acidosis (caused by lactic acidosis and acute kidney injury), plac- ing patients at risk for acute respiratory failure. In a se- ries of 439436 admissions for AMICS captured in the National Inpatient Sample, 57% of patients received a diagnosis of acute respiratory failure and 43% under- went mechanical ventilation.25 Worsening hypoxemia and acidosis increase susceptibility to ventricular fibril- lation and may increase risk of death during attempted coronary revascularization. Increased work of breathing to compensate for ventilation-perfusion mismatch and metabolic acidosis may further contribute to progres- sion of AMICS. Hence, strong consideration should be given to early endotracheal intubation and mechanical ventilation. Caution is advised in patients with AMICS and predominant right ventricular failure, including pa- tients with right ventricular myocardial infarction, not- ing that initiation of positive pressure ventilation can abruptly lower systemic arterial pressure. Early intuba- tion and ventilatory support may facilitate revascular- ization because of improved oxygenation, greater seda- tion, and enhanced metabolic profile.
Right-sided heart catheterization provides access to quantitative data to sharpen characterization of indi- vidual patient hemodynamics over time. No randomized trial has been performed to validate the routine use of right-sided heart catheterization in AMICS, the optimal timing of its performance, or specific interventions based on invasive hemodynamic profiles. Key parameters to assess and monitor include central venous pressure, pul-monary capillary wedge pressure, cardiac output, cardi- ac power output, pulmonary artery pulsatility index, and mixed venous oxygen saturation. Cardiac power output (Watts) is calculated as follows: cardiac output×mean ar- terial pressure÷451.28 Pulmonary artery pulsatility index is calculated with the following equation: (pulmonary artery systolic pressure−pulmonary artery diastolic pres- sure)/right atrial pressure.29 Right ventricular stroke work index is calculated as follows: (mean pulmonary artery pressure−central venous pressure)×stroke volume index. Invasive measures, including central venous pressure >10 mm Hg, central venous pressure/pulmonary capil- lary wedge pressure >0.63 mm Hg, pulmonary artery pulsatility index <2.0, and right ventricular stroke work index <450 g·m/m2, may help identify right ventricular dysfunction complicating AMICS, a common phenom- enon identified in 38% and 37% of patients in the SHOCK trial and registry, respectively.30 For patients with early shock, invasive measurements can help to further delineate those patients who are hypotensive but nor- mally perfused and those who are normotensive but hy- poperfused.31 Of note, right-sided heart catheterization is not required to diagnose shock. In cases of AMICS in which performance of right-sided heart catheteriza- tion would cause an undue delay in timely reperfusion therapy, consideration should be given to deferring its performance until completion of PCI.
Patients with AMICS with persistent hemodynamic com- promise despite initial stabilization may benefit from immediate MCS. The rationale for initiation of MCS early in AMICS is to reduce ventricular workload (unloading), increase systemic perfusion, enhance myocardial perfu- sion, and provide hemodynamic support during PCI.
Persistent clinical hypoperfusion, hypotension, vaso- pressor requirement, or cardiac power output <0.6 W despite adequate filling pressures may indicate a role for MCS as an adjunct to stabilization before coronary revascularization. For patients with predominant LV failure, MCS options include intra-aortic balloon coun- terpulsation (IABP), a transvalvular axial flow pump (Impella LP/CP/5.0/5.5), and the TandemHeart percuta- neous LV assist device. Venoarterial (VA) extracorporeal membrane oxygenation (ECMO) may be considered to provide systemic circulatory support, but close monitor- ing for LV distension and worsening pulmonary edema is required. In these cases, VA-ECMO may require an additional LV decompression or venting mechanism, options for which include an IABP, a left-sided Impella device, pulmonary artery cannulation, or surgical LV venting.32,33 For patients with predominant right ven- tricular failure, MCS options include the transvalvular axial flow Impella RP pump and TandemHeart Protek- Duo percutaneous right ventricular assist device. Pa- tients with biventricular failure may be supported with bilateral Impella pumps or VA-ECMO with a concomi- tant LV venting mechanism. Patients with concurrent refractory respiratory failure should be considered for VA-ECMO. In part, the protective mechanisms associ- ated with MCS in AMI are supported by extensive pre- clinical data beginning in the late 1970s.34 Observation- al studies of AMICS systems of care incorporating early MCS have reported improved survival compared with historical controls,15,35,36 but no randomized controlled trial has provided evidence in support of routine use for any short-term MCS platform.
The IABP-Shock II trial (Intra-Aortic Balloon Pump in Cardiogenic Shock II) randomized 600 patients with AMICS to a strategy of routine IABP use or conservative care.7 Among the 277 patients randomized to IABP who received urgent revascularization, 86.6% of patients re- ceived the IABP after revascularization. Compared with the group of patients assigned to conservative care, the use of an IABP was not associated with a reduction in 30- day all-cause mortality (39.7% versus. 41.3%; P=0.69). In addition, key secondary end points, including time to hemodynamic stabilization, intensive care unit length of stay, renal function, and serum lactate levels, did not dif- fer between the 2 groups. Although this trial did not sup- port the IABP as a specific MCS device for the treatment of AMICS, some have argued that the lack of benefits observed in this trial may have been influenced by the timing of device insertion (after revascularization in the majority of patients), variability of shock severity across the study population, or limited hemodynamic effects of IABP relative to other devices. Randomized studies comparing the IABP with other MCS devices have not shown improved survival with any MCS device, although these studies were small and not powered to evaluate hard end points.18,37,38 These findings should not be ex- trapolated to other causes of CS beyond AMI. More pro- spective studies are required to understand the clinical utility of IABP in ischemic and nonischemic forms of CS. Observational studies examining outcomes with MCS devices used for AMICS have reported variable results. The Detroit Cardiogenic Shock Initiative en- couraged an aggressive protocol of early MCS in the management of patients with AMICS.15 Among a co- hort of 41 patients admitted to 4 hospitals in Detroit, MI, with AMICS who were treated with an Impella, 93% of patients were on vasopressors or inotropes before device implantation, and an additional 17% were receiving active cardiopulmonary resuscitation during Impella placement. The majority of patients (66%) received an Impella before revascularization. In this report, 85% of patients survived to device ex- plantation. This number was notably higher than the 51% observed survival to device explantation reported for patients with AMICS in the metro Detroit area be- fore implementation of this protocol. These findings have been further studied in the prospective single- arm National Cardiogenic Shock Registry.36 Systematic exclusion of cardiac arrest and selection bias prompt caution in the interpretation and generalization of the favorable outcomes observed in these studies com-pared with historical controls.
In the RETROSHOCK registry of patients with AMICS admitted to 2 hospitals in Denmark,39 patients treat- ed with early Impella use (n=40) had a significantly lower rate of death compared with a matched group of patients receiving no therapy (40% versus 77.5%; P<0.001). On the other hand, the mortality rate of patients treated with IABP (n=40) was similar to that of a matched group of patients receiving no therapy (27.5% versus 37.5%; P=0.35). These data contrast with those reported in a European multinational reg- istry of patients with AMICS,40 in which 237 patients treated with an Impella were matched to 237 patients enrolled in the IABP-Shock II trial. Among the 237 pa- tients selected from the multinational registry, 38.1% were treated with an Impella before revascularization. Use of an Impella was associated with no difference in 30-day all-cause mortality compared with the matched patients from the IABP-Shock II trial (48.5% versus 46.4%; P=0.64). Severe or life-threatening bleeding was higher in the Impella group (8.5% versus 3.0%; P<0.01), as were vascular complications (9.8% ver- sus 3.8%; P=0.01) and sepsis (35.3% versus 19.4%; P<0.01). Subgroup analysis did not show an interaction between timing of insertion and outcomes. In addition, there were no differences in mortality when analysis was limited to a comparison of registry patients with an
Impella and patients either in the treatment arm (IABP) or in the control arm of the IABP-Shock II trial.
The DanGer trial (Danish-German Cardiogenic Shock) is a prospective open-label multicenter trial that aims to randomize 360 patients with AMICS to the Impella CP or guideline-driven therapy.41 Multiple randomized studies of VA-ECMO in AMICS are also ongoing, including EURO SHOCK (Testing the Value of Novel Strategy and Its Cost Efficacy in Order to Im- prove the Poor Outcomes in Cardiogenic Shock; URL: ClinicalTrials.gov. Unique identifier: NCT03813134), ECLS-SHOCK (Extracorporeal Life Support in Cardio- genic Shock; URL: ClinicalTrials.gov. Unique identifier: NCT03637205), ECMO-CS (Extracorporeal Membrane Oxygenation in the Therapy of Cardiogenic Shock; URL: ClinicalTrials.gov. Unique identifier: NCT02301819), and ANCHOR (Assessment of ECMO in Acute Myocar- dial Infarction Cardiogenic Shock; URL: ClinicalTrials. gov. Unique identifier: NCT04184635). The results of these and other trials will further inform the manage- ment of patients with AMICS. In the meantime, there is cause for caution, with observational data illustrat- ing heterogeneity in safety and outcomes of MCS use in the context of steadily growing use. Indeed, 2 re- cent registry studies demonstrated signals toward in- creased rates of major bleeding and in-hospital death among propensity-matched patients with AMICS treat- ed with an Impella versus IABP.42,43 It is important that we individualize care for our patients, considering the underlying mechanisms of shock, anticipated benefits and risks of MCS, and ideal timing for device insertion. MCS platforms differ substantially with respect to vas- cular access requirements, learning curve, and support provided, and limited data exist to inform allocation of specific MCS devices based on clinical or hemodynamic profile. Specific device selection requires the input of a multidisciplinary team with consideration of patient needs and device availability and familiarity (Figure 3).
Putative benefits of early MCS include support of sys- temic perfusion, reduced cardiac workload, enhanced coronary perfusion and decongestion, and, through these mechanisms, arrest of the progression of shock to end- organ injury and death.44 Offsetting these benefits are variable, device-dependent risks of bleeding, hemolysis, vascular complications, and limb ischemia, as well as the additive complexity of postimplantation management.45 In the context of STEMI, there is a theoretical concern that benefits of MCS may be further offset by increased delay to reperfusion therapy. The Door to Unload-STEMI pilot study, which did not include patients with CS, did not identify harm with a strategy of first unloading the LV for up to 30 minutes before reperfusion but also did not show benefit.46 The STEMI-DTU pivotal trial (Primary Un- loading and Delayed Reperfusion in ST-Elevation Myocar- dial Infarction) will test whether early MCS before reper- fusion limits myocardial damage in patients with anterior
STEMI without CS. It is not certain that findings in STEMI without shock can be extrapolated to AMICS. Among patients with AMICS, emerging observational data sug- gest that early MCS may improve, not worsen, outcomes in select patients.15 Operator technique and judgment in vascular access, management of anticoagulation, surveil- lance and timely management of vascular complications, expertise in device positioning and management, and in- tegration of device use within a global plan of care collec- tively have substantial potential to influence the net ben- efit of early MCS use for an individual patient. An array of trials are attempting to decipher this complex landscape. Until data become available from randomized clinical tri- als sufficiently powered to define risks and benefits of early MCS for patients with different stages of AMICS, we strongly encourage an individualized approach to care and participation in clinical research protocols to test the utility of MCS in AMICS. Early MCS placement before PCI may be considered for patients with AMICS who exhibit refractory hemodynamic instability despite aggressive medical therapy (Figure 4).
PCI of the infarct-related artery is the recommend- ed method of reperfusion for patients with AMICS regardless of time delay.5,47,48 The SHOCK trial estab- lished the clinical benefit of an early invasive strategy with intent for early revascularization in patients with AMICS, demonstrating a significant mortality reduction at 6 months and in long-term follow-up for individuals <75 years of age compared with initial medical stabi- lization.6,49,50 To save 1 life, only 8 patients need to be treated according to this landmark trial. Of note, the SHOCK trial did not identify a difference in its primary end point, all-cause mortality at 30 days, with benefits of early revascularization becoming evident only with longer follow-up. This example underscores challenges in the design and interpretation of randomized trials in AMICS. With the progressive availability of early PCI, multiple registries have since shown a decrease in mortality from prior levels of 70% to 80% to 40% to 50%.12,51–54 Early revascularization has become the most important strategy in the treatment of AMICS, with recent registries highlighting increased risks with revascularization delays.21,22
There is uncertainty about the optimal revascularization approach in AMICS because previous trials assessing the effect of revascularization on outcomes did not specify the mode of reperfusion. PCI is the most widely available and most often performed revascularization therapy in AMICS, whereas coronary artery bypass graft surgery (CABG) is rarely performed. In the IABP-Shock II trial and registry, for example, only 4% of patients had immedi- ate CABG.7 In observational reports comparing PCI with CABG, mode of revascularization has not appeared to influence outcomes of patients with AMICS.55,56 Factors influencing the possible selection of CABG include the suitability of coronary anatomy, including the caliber and quality of prospective distal anastomotic targets for bypass grafts; importance of the infarct-related artery; and surgical availability and experience. Given the very high mortality of patients with unsuccessful PCI, emer- gency CABG should be considered a rescue modality in such cases, as well as in cases in which AMI is complicat- ed by myocardial rupture. A hybrid approach of culprit lesion PCI (with or without stent placement) followed by staged CABG has also been considered, in particular for patients with AMICS and multivessel disease with or without diabetes.
Because of its limited efficacy, fibrinolytic therapy is reserved for patients with ST-segment–elevation AMI when timely PCI is unavailable.47,48 For those patients with AMICS who initially present to a non–PCI-capable facility, the question of safe transfer to a PCI-capable hospital arises.4 Availability of a transfer network and early activation of an established AMICS communica- tion pathway become important early measures. If immediate transfer cannot be arranged safely, evalua- tion for emergency fibrinolytic therapy and subsequent transfer should be considered, granting that many patients may possess contraindications to fibrinolytic therapy such as traumatic resuscitation efforts, cardiac arrest with unclear neurological prognosis, no clear ST- segment elevation on the initial ECG, coagulopathy, and advanced age.
Additional stenoses or occlusions beyond the infarct- related artery can be found in ≈70% to 80% of pa- tients.57 Patients with AMICS with multivessel disease have a higher mortality compared with patients with single-vessel disease.58 Until recently and mainly on the basis of theoretical considerations, multivessel PCI of all critical lesions was encouraged in patients with AM- ICS.47,48,59 This approach was not supported by pooled results of observational studies that demonstrated higher short-term mortality when multivessel PCI was performed in patients with AMICS.60
The CULPRIT-SHOCK trial (Culprit Lesion Only PCI Versus Multivessel PCI in Cardiogenic Shock),9 to date the largest randomized trial in CS, addressed the ques- tion of optimal revascularization therapy in patients with multivessel disease and AMICS. This study of 706 patients with AMI (66% with STEMI) showed a signifi- cant reduction in 30-day mortality or renal replacement therapy (primary end point) with a strategy of culprit lesion–only PCI (with an option for staged revascular- ization of additional lesions) compared with immedi- ate multivessel PCI (45.9% versus 55.4%; relative risk, 0.83 [95% CI, 0.71–0.96]; P=0.01), driven primarily by an absolute 8.2% reduction in mortality.9 These results were further supported by a sustained reduction in the same composite end point at the 1-year follow-up.61 Results were consistent across all predefined subgroups, including patients with out-of-hospital cardiac arrest.
In the vast majority of patients with AMICS, PCI should be limited to the culprit lesion with possible staged revascularization of other lesions. However, the role of multivessel PCI in AMICS remains under active investigation. Notably, few patients in the CULPRIT- SHOCK trial received MCS. Furthermore, recent data from the Korea Acute Myocardial Infarction National Health Registry showed that multivessel PCI was associ- ated with a lower risk of all-cause death than culprit- artery–only PCI at 3 years, suggesting possible benefit of nonculprit lesion revascularization during the index hospitalization on long-term clinical outcomes.62 Select- ed angiographic scenarios such as subtotal nonculprit lesions with reduced TIMI (Thrombolysis in Myocardial Infarction) grade flow or multiple possible culprit lesions may benefit from immediate multivessel PCI. Such deci- sions are complex and not addressed in practice guide- lines, absent a robust clinical trial for each decision step, requiring individualized consideration on a patient and lesion basis.
CS is a potent predictor of stent thrombosis.63 Poten- tial factors contributing to this association may include AMICS-associated abnormalities in coronary perfusion, thrombus burden, microvascular occlusion and dys- function, platelet activation, PCI quality, and limited bioavailability related to absorption (in particular in the setting of morphine or fentanyl) and pharmacodynam- ics of antithrombotic therapies. In this context, a strat- egy of more potent and more consistent antiplatelet therapy may be desired for AMICS,64 although no ade- quately powered randomized trial specific to AMICS has tested this to date. Avenues to increase potency, con- sistency, and rapidity of antiplatelet therapy in AMICS may include preferential use of third-generation oral P2Y12 inhibitors instead of clopidogrel,64,65 administra- tion of crushed ticagrelor via gastric tube,66 and paren- teral administration of cangrelor,67 alone or in combina- tion with ticagrelor.68 Platelet reactivity may be further reduced with adjunctive use of glycoprotein IIb/IIIa in- hibitors,69 but the safety of these agents in the con- text of AMICS is not well established, particularly in the setting of large-caliber access for MCS devices. Rapid reversibility of cangrelor despite bowel, liver, and kidney dysfunction might improve safety.
After completion of PCI, attention turns to preparation for patient transfer to the cardiac intensive care unit. This transition may be aided by a checklist. Critical ele- ments of a general survey of stability include hemosta- sis at all access sites; electric stability, noting evidence of ongoing bradyarrhythmia or tachyarrhythmia; hemo- dynamic stability, including verifying optimal position- ing, securing, performance, and adequate distal limb perfusion with any MCS devices; respiratory stability, in- cluding adequate oxygenation and control of acid-base status; sufficient vascular access; and consideration of an indwelling pulmonary artery catheter.
Comprehensive critical care after acute invasive man- agement comprises prevention, diagnosis, and man- agement of multiorgan system failure complicating AMICS; continuous reassessment of hemodynamics and perfusion status with clinical and invasive measures; ongoing and relentless titration of therapies based on evolving data; anticipation and management of com- plications of acute invasive management; collaboration and shared decision making by a multidisciplinary shock team, including consideration of timing and approach to escalation or de-escalation of MCS; and close com- munication with family to provide regular updates and reassessment of prognosis and goals of care. Areas of consensus, controversy, and uncertainty are considered in detail elsewhere.4
Cardiac arrest is common among patients with AMICS and confers an increased risk of mortality that is inde- pendent of shock stage.8,19 Outcomes are exponentially complicated by a variable degree of hypoxic-ischemic encephalopathy, with a subset of patients at risk for severe neurological disability or brain death regardless of a positive cardiac outcome. Proceeding with a com- plex cardiac evaluation and treatment plan while neu- rological status is unknown for up to several days poses unique difficulties in care delivery and ethics.
In general, patients successfully resuscitated from cardiac arrest with return of spontaneous circulation and neurological function (Glasgow Coma Scale score ≥8) and a diagnosis of AMICS should be triaged to the cardiac catheterization laboratory as soon as pos- sible for complete assessment. Vigil is required during transport and in the laboratory for recurrent arrest. Patients with AMICS and resuscitated cardiac arrest who remain comatose (Glasgow Coma Scale score <8) or unable to follow simple commands should be treated with targeted temperature management as soon as possible.70–72 Early invasive therapy in coma- tose patients with out-of-hospital cardiac arrest should be individualized and based on the absence of unfa- vorable prognostic features, which may include unwit- nessed arrest, an initial nonshockable rhythm, lack of bystander cardiopulmonary resuscitation, >30 minutes to return of spontaneous circulation or ongoing cardio- pulmonary resuscitation, pH <7.2, lactate >7 mmol/L, age >85 years, end-stage renal disease, and noncar- diac cause of arrest.73 In the absence of multiple unfa- vorable prognostic features, patients with AMICS with or without ST-segment elevations should be consid- ered for emergency triage to the cardiac catheteriza-tion laboratory.74 Although a recent randomized study found no penalty in terms of 90-day mortality with a strategy of delayed versus immediate angiography in cardiac arrest without ST-segment elevations, it should be noted that this trial systematically excluded patients with shock.75
Patients with AMICS and ongoing cardiac arrest without return of spontaneous circulation represent the highest-risk group in whom multiorgan failure is uniform and mortality is common. Successful invasive management has been reported with the use of au- tomated cardiopulmonary resuscitation and ECMO in carefully selected patients by experienced multidisci- plinary teams at tertiary centers.76 A majority of such patients stabilized with ECMO who undergo coronary angiography have obstructive coronary artery disease with indication for PCI.77 Further research is required to guide patient selection, define feasibility, and organize delivery of this resource-intensive approach on a broad-er scale. Also requiring further research are the role and optimal modality of LV venting when ECMO is used to support AMICS, noting benefit in meta-analysis of ob- servational studies and variable use and multiple tools in practice.42,43 Results of ongoing trials are anticipated.
A subset of patients with AMICS will die regardless of invasive management. Risk predictive models devel- oped for patients with shock1,78,79 and patients treated with ECMO80 may provide useful adjuncts to clini- cal assessment to identify patients at highest risk for mortality, but it is unclear that such scores should be used to determine eligibility for invasive management because of potential risk-treatment paradox. Assess- ment of the utility of invasive therapy is complex and requires ascertainment of patient and family values and wishes and the clinical judgment of a multidisci- plinary shock team.
AMICS is a complex clinical entity that remains preva- lent and the major cause of death after AMI. Treat- ment decisions made in the early invasive manage- ment of AMICS can have significant ramifications for the progression of shock, patient survival, and out- comes of AMICS at large. Optimization of care re- quires a multidisciplinary team effort to coordinate early assessment and triage (including possible inter- hospital transfer), noninvasive and invasive diagnos- tics, coronary revascularization, and expert ongoing intensive care management, including a sophisticated of adjunctive LV venting for patients on ECMO; and criteria for weaning and discontinuing support;
•Approaches to multivessel coronary artery disease, including criteria for selected application and tim- ing of multivessel PCI and consideration of CABG;
•Choice of pharmacotherapy to support PCI, including antiplatelet therapy and reversal/bridg- ing considerations;
•Management and safety of MCS devices in the cardiac intensive care unit, including strategies to reduce bleeding and vascular complications, anti- coagulation management, parameters for moni- toring device function and necessity, and interplay with vasoactive drug therapies; and Application of targeted temperature management in patients with AMICS and resuscitated cardiac arrest, including optimal target temperature and cooling modality. understanding of the evolving pathophysiology and hemodynamics of AMICS.81 Advances in systematic recognition and classification of AMICS are expect- ed to allow a new wave of clinical investigation into this highly morbid and mortal disease and its invasive management. Essential avenues for future research in invasive management of AMICS include but are not limited to the following:
•System-level approaches to expediting identifi- cation of AMICS and activation of established multidisciplinary shock teams, including at shock centers, non–shock center PCI-capable hospitals, and non–PCI-capable hospitals;
•Applications of SCAI shock classification to thera- peutic critical pathways, including consideration of early MCS, use of invasive hemodynamic monitor- ing, and selection for invasive management;
•Selection of vasoactive drug therapies to support hemodynamics in the presence or absence of MCS devices, including guidance of selection and titration of drug therapies with invasive hemo- dynamic measures and consideration of specific inotrope and vasopressor agents alone and in combination;
•Strategies for allocation of MCS, before or after coronary revascularization, including appropriate initial selection of specific MCS devices; criteria for allocation of secondary MCS devices, including use of adjunctive LV venting for patients on ECMO; and criteria for weaning and discontinuing support;
•Approaches to multivessel coronary artery disease, including criteria for selected application and tim- ing of multivessel PCI and consideration of CABG;
•Choice of pharmacotherapy to support PCI, including antiplatelet therapy and reversal/bridg- ing considerations;
•Management and safety of MCS devices in the cardiac intensive care unit, including strategies to reduce bleeding and vascular complications, anti- coagulation management, parameters for moni- toring device function and necessity, and interplay with vasoactive drug therapies; and
•Application of targeted temperature management in patients with AMICS and resuscitated cardiac arrest, including optimal target temperature and cooling modality.
The American Heart Association makes every effort to avoid any actual or po- tential conflicts of interest that may arise as a result of an outside relationship or a personal, professional, or business interest of a member of the writing panel. Specifically, all members of the writing group are required to complete and submit a Disclosure Questionnaire showing all such relationships that might be perceived as real or potential conflicts of interest.
This statement was approved by the American Heart Association Science Advisory and Coordinating Committee on September 29, 2020, and the Ameri- can Heart Association Executive Committee on October 26, 2020. A copy of the document is available at https://professional.heart.org/statements by using either “Search for Guidelines & Statements” or the “Browse by Topic” area. To purchase additional reprints, call 215-356-2721 or email Meredith.Edelman@ wolterskluwer.com.
Supplemental materials are available with this article at https://www. ahajournals.org/doi/suppl/10.1161/CIR.0000000000000959
The American Heart Association requests that this document be cited as fol- lows: Henry TD, Tomey MI, Tamis-Holland JE, Thiele H, Rao SV, Menon V, Klein DG, Naka Y, Piña IL, Kapur NK, Dangas GD; on behalf of the American Heart Association Interventional Cardiovascular Care Committee of the Council on Clinical Cardiology; Council on Arteriosclerosis, Thrombosis and Vascular Biol- ogy; and Council on Cardiovascular and Stroke Nursing. Invasive management of acute myocardial infarction complicated by cardiogenic shock: a scientific statement from the American Heart Association. Circulation. 2021;143:e815– e829. doi: 10.1161/CIR.0000000000000959
The expert peer review of AHA-commissioned documents (eg, scientific statements, clinical practice guidelines, systematic reviews) is conducted by the AHA Office of Science Operations. For more on AHA statements and guidelines development, visit https://professional.heart.org/statements. Select the “Guide- lines & Statements” drop-down menu, then click “Publication Development.”
Permissions: Multiple copies, modification, alteration, enhancement, and/ or distribution of this document are not permitted without the express permis- sion of the American Heart Association. Instructions for obtaining permission are located at https://www.heart.org/permissions. A link to the “Copyright Per- missions Request Form” appears in the second paragraph (https://www.heart. org/en/about-us/statements-and-policies/copyright-request-form).