Stages of Ventricular Arrhythmogenesis Following Coronary Artery Occlusion

Arrhythmia mechanisms and prognostic implications vary, depending on the time of occurrence after the onset of coronary occlusion. The timeline has been divided into an acute phase (first 30 minutes) and a delayed phase (or subacute phase, 6 to 48 to 72 hours).

Two distinct phases of arrhythmogenesis (Table 57-1) occur during the initial 30 minutes (“acute phase”) of ischemia after experimental coronary artery ligation.⁸ Phase 1a arrhythmias occur 2 to 10 minutes following coronary artery occlusion (peak 5 to 6 minutes) and are caused by re-entry within the ischemic myocardium resulting from the inhomogeneity of refractory periods in normal and ischemic tissue. Mapping studies have revealed the presence of low-amplitude fractionated electrograms.^7,8

Phase 1b arrhythmias occur 10 to 30 minutes following coronary artery occlusion (peak 15 to 20 minutes). The precise mechanism of type 1b ventricular arrhythmias is unclear. By this stage, the inhomogeneities in subepicardial refractoriness and conduction have improved to near-normal values.⁸ Because of the important role of catecholamines in arrhythmogenesis, it has been postulated that abnormal automaticity is the underlying mechanism.²⁹ Myocardial stretch mechanisms have also been implicated in the generation of abnormal automaticity.³⁰ Studies in canine hearts during the first 30 minutes following coronary artery occlusion have suggested that up to 60% of ventricular tachycardias (VT) are focal in origin, arising from Purkinje fibers.³¹ Generally, during these two phases, spanning the first 30 minutes following occlusion, no permanent structural damage occurs. On reperfusion, ischemic cells survive and generally recover function. However, toward the end of phase 1b, changes in the internal axial resistance of cardiac tissue are first noted, indicating the onset of irreversible cellular and gap junction damage.³²

The subacute or delayed phase occurs 6 to 72 hours following coronary artery occlusion (peak 12 to 24 hours).³³ It coincides with the onset of cell death; reperfusion at this stage does not reduce the amount of cell damage. Although substantial myocardial cell death occurs in the infarcted region, subendocardial Purkinje fibers survive with altered electrophysiological properties predisposing to arrhythmia generation.³⁴ A reduced resting membrane potential and spontaneous membrane depolarizations lead to abnormal automaticity. Delayed after-depolarizations resulting in triggered activity have also been demonstrated.³⁵ In addition, heterogeneity of conduction and refractoriness at the border zone, which is the interface between the dead myocardium and the still-viable myocardium, may lead to re-entrant arrhythmias.

Reperfusion Arrhythmias

Reperfusion arrhythmias are more common after short ischemic episodes than after long ischemic periods.³⁶ In the canine model, reperfusion arrhythmias have been shown to occur in two stages. Immediately following restoration of perfusion after coronary artery occlusion, VF may occur due to multiple wavelet re-entry. This occurs as a result of a rapid but inhomogeneous return of APs to previously unexcitable cells within the ischemic zone and a shortening of refractory periods in the border zone brought about by the washout of K⁺ and metabolites from the extracellular space.³⁷ In addition, premature depolarizations may be induced by triggered activity. Although overall electrical function can return to normal at this stage, gap junction injury may persist with a corresponding inhomogeneous delay in conduction properties.

Accelerated idioventricular rhythms are commonly seen following reperfusion in the canine model. This arrhythmia may be due to the increased adrenergic stimulation of Purkinje fibers near the ischemic region causing enhanced automaticity or triggered activity.³⁸ As accumulation of catecholamines is required, these arrhythmias typically occur after 20 to 30 minutes of occlusion. Compared with the canine model, the incidence of early reperfusion arrhythmias in the human population is significantly lower. This probably reflects the longer occlusion times and less rapid or incomplete reperfusion typically seen in patients presenting with AMI.

Premature Ventricular Contractions

PVCs are seen in the majority of cases of acute MI. Early PVCs (within the first 48 hours) do not appear to affect the prognosis, but frequent or complex PVCs occurring beyond 48 hours after AMI may be associated with increased arrhythmic risk. In the human heart, R-on-T PVCs are rarely observed, accounting for only 1.8% of PVCs during the first 24 hours of admission, and most PVCs do not trigger severe ventricular tachyarrhythmias.^46–48 However, in a canine model, 24% of PVCs occurring between 12 and 30 minutes (phase 1b) resulted in R-on-T and were responsible for the initiation of 34% of spontaneous episodes of VT and fibrillation.

Several studies from the prethrombolytic era have suggested that frequent PVCs (>10 PVCs per hour), complex PVCs (ventricular bigeminy, couplets, or multiform ventricular premature beats), or both are a risk factor independent of the degree of myocardial damage and left ventricular systolic dysfunction,^20,49 but in another trial, PVC frequency had no independent predictive value in multivariate analysis.⁵⁰ Antiarrhythmic suppressive therapy (lidocaine) has not been shown to improve outcomes, and class Ic antiarrhythmics may increase mortality. Electrophysiology study for risk stratification is currently not recommended for either early or late post-MI PVCs.

Accelerated Idioventricular Rhythm

Accelerated idioventricular rhythm (AIVR) is commonly witnessed in the first 12 hours after admission for AMI. Although more common in patients with successful reperfusion therapy, it is not a specific marker, with 63% of patients with occluded arteries still demonstrating the arrhythmia.⁵¹ The presence of AIVR does not affect the prognosis.

Nonsustained Ventricular Tachycardia

The presence of NSVT identifies patients at risk of in-hospital cardiac arrest. NSVT that occurs within the first 2 to 3 hours does not carry an adverse prognosis, whereas NSVT that occurs beyond several hours after admission does, particularly in patients with prior MI. NSVT in the setting of AMI occurs in 1% to 7% and possibly in as many as 75% of patients (Figure 57-1).⁵² NSVT occurring 24 hours after AMI carries a worse prognosis than NSVT occurring within the first 24 hours following AMI (Figure 57-2). This is contrary to the commonly held belief that arrhythmias occurring within the first 48 hours following MI do not carry an adverse long-term prognosis. NSVT in the setting of healing MI (7 to 10 days following MI) is also associated with a poorer prognosis. Aside from β-blockers, antiarrhythmic therapy is not currently recommended for either early or late post-MI asymptomatic NSVT. Electrophysiology testing is not currently recommended for risk stratification of NSVT in the first several weeks after AMI but is considered “reasonable” for risk stratification in patients with remote MI, NSVT, and left ventricular ejection fraction (LVEF) 40% or less.⁵³

FIGURE 57-1 Kaplan-Meier survival curves for control and case patients stratified by time to occurrence of nonsustained ventricular tachycardia (NSVT) ≤24 hours from presentation (from time of admission). Patients with NSVT >24 hours after presentation had poorer survival rates (P < .0001) than the other two groups.

(From Cheema A, Sheu K, Parker M, et al: Nonsustained ventricular tachycardia in the setting of acute myocardial infarction: Tachycardia characteristics and their prognostic implications, Circulation 98[19]:2030–2036, 1998.)

FIGURE 57-2 Plot of relative risk of nonsustained ventricular tachycardia using various time cutoffs from presentation. All time cutoffs <13 hours were significant.

(From Cheema A, Sheu K, Parker M, et al: Nonsustained ventricular tachycardia in the setting of acute myocardial infarction: Tachycardia characteristics and their prognostic implications, Circulation 98[19]:2030–2036, 1998.)

Ventricular Tachycardia, Polymorphic Ventricular Tachycardia, and Ventricular Fibrillation

The incidence of documented “early” sustained monomorphic VT (SMVT) within the first 48 hours of AMI is in the range of 2% to 3% in STEMI and less than 0.9% in NSTEMI.^39,45,54 It may indicate extensive myocardial damage and serve as an independent predictor of mortality.^54,55 As discussed above, arrhythmia mechanisms undergo dynamic changes in the early minutes and hours after onset of ischemia and may involve both re-entry (within ischemic areas of slowed conduction and increased refractoriness) or non–re-entrant mechanisms (triggered activity or increased automaticity). SMVT, however, implies stability of the ventricular depolarization pattern, which is most readily provided by a stable re-entry circuit. Thus, it is likely that SMVT, even in the early hours following MI, occurs in the presence of an already established permanent substrate (developing necrosis or pre-existing scar). Electrolyte abnormalities or ischemia that can cause the events that initiate re-entry (PVCs, NSVT) should be corrected; however, SMVT should be addressed as it would be even in the absence of these factors. From the currently available data, it is unclear that SMVT can be lumped together with other early post-MI arrhythmias in terms of its effect on the long-term prognosis—as most studies have not analyzed it separately from VF, polymorphic VT, or NSVT. In the GUSTO-I (Global Utilization of Streptokinase and t-PA for Occluded Coronary Arteries) study, patients with early (<48 hours) sustained VT had a 7.1% 1-year mortality among 30-day survivors, compared with 6.1% in patients with “late” (>48 hours) VF, and 2.6% in patients without any early or late VT or VF.³⁹ Therefore, SMVT, occurring even “early” after MI, is generally considered by many experts to be an indicator of high risk for future arrhythmia and SCD, warranting further investigation and intervention.^53,56

Polymorphic VT, which occurs in 0.3% to 2% of patients, may be a marker of ongoing ischemia; therefore, it can often be effectively managed by anti-ischemic interventions. It is more often seen in patients who also develop VF.⁵⁷ In general, efforts are made to correct potential triggering factors such as hypokalemia, hypomagnesemia, abnormal serum calcium, or bradycardia (in those with bradycardia or pause-dependent onset).⁵³ In a case series of 11 patients with polymorphic VT, none had sinus bradycardia, but 3 of 11 had a sinus pause preceding the onset.⁵⁸ None had prolonged Q-T interval, hypokalemia, or abnormal serum magnesium or calcium. Nine of eleven had signs of recurrent ischemia immediately before arrhythmia onset. VF occurs in 3.7% of all acute STEMIs in the first 48 hours, and this is likely an underestimation, as prehospital events are not included.^39,41 Of these, most VF episodes occur within the first 4 hours (3.1%).⁴¹ When all VF events, before and after 48 hours, were included, VF was found to occur in 6.7% of STEMI patients and in 1.3% of NSTEMI patients.^39,45 In the first 4 hours of admission, VF was more likely to occur in the setting of hypokalemia, low blood pressure, larger infarct size, current smoking, and a younger age. VF was more common in inferoposterior infarcts, possibly because of greater autonomic upset. The association of initial bradycardia with early fibrillatory risk fits with the observation that vagal overactivity may precede VF. VF at all stages of infarct evolution is more common in patients with larger infarcts as determined by serial cardiac enzyme measurements.⁵⁹

Traditionally, primary VF has referred to VF that occurs during the first 48 hours of an uncomplicated MI (without recurrent ischemia or heart failure), and in the GISSI (Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Miocardico)-2 trial, it was associated with increased in-hospital mortality; however, no statistically significant association with post-discharge mortality (1-year mortality of those who survived to hospital discharge) was observed. These effects of early ventricular arrhythmias on early mortality were confirmed in the percutaneous coronary intervention era.^60,61 In contrast, nonprimary VF (VF occurring in the setting of recurrent ischemia or heart failure or beyond the first 48 hours following MI) was associated with marked increases in both 30-day mortality and 6-month mortality.⁴¹ In the GUSTO trial, all “late” (>48 hours following MI) sustained VT, VF episodes, or both were associated with markedly increased long-term mortality at 1 year among 30-day survivors.³⁹ More recently published data from the GUSTO-V trial found that “early” VT or VF (<48 hours) was associated with increased in-hospital mortality but not with 1-year mortality among 30-day survivors⁴²; however, all arrhythmias (VF, all VT) were pooled in this analysis. The temporal cutoff between “early” and “late” arrhythmias at 48 hours following MI is arbitrary to some extent; data to suggest that this should be at 24 hours or even earlier exist⁵²; clearly, decisions should be individualized and based on expert evaluation and judgment. Moreover, additional tools for risk stratification are needed, and this is an area of active investigation; in the future, these may include a combination of electrophysiological testing, genetic evaluation, scar imaging, autonomic evaluation, and so on.

Preliminary data from a post hoc analysis suggest that ranolazine, an antianginal that inhibits the late inward Na⁺ current, may decrease the incidence of VT or VF (as well as SVT or AF), but this requires further study.⁶² A large pooled analysis has suggested that early administration of intravenous β-blockers in acute myocardial infarction may decrease the mortality and incidence rates of VF, but the Clopidogrel and Metoprolol in Myocardial Infarction Trial (COMMIT) performed on 46,000 patients with AMI did not show the mortality benefit of this intervention.^63,64

In summary, the available data on the outcomes of ventricular arrhythmias in AMI have limitations. These data come mostly from thrombolytic trials, retrospective analyses, limited numbers of events, and analyses of multiple types of arrhythmia, rather than specific arrhythmias, pooled together. Mainly on the basis of large thrombolytic trials, “early” sustained VF or polymorphic VT occurring within the first 24 to 48 hours of an uncomplicated AMI is associated with increased in-hospital and 30-day mortality but appears to have little effect on long-term mortality in patients surviving hospital discharge; however, this may be because high-risk patients die during their initial hospital stay. Conversely, all sustained ventricular arrhythmias that are “late” (>24 to 48 hours following MI) or in the context of complicated MI and any sustained monomorphic VT are considered indicators of high risk of arrhythmia and SCD, and patients are considered survivors of cardiac arrest. The temporal cutoff between “early” and “late” arrhythmias is unclear and may be close to 24 hours or even earlier. Occasionally, complete revascularization may be achievable with sufficient treatment—in the absence of prior MI, residual scar, SMVT, or systolic dysfunction—but most of these patients should be considered for defibrillator implantation, and expert, individualized decisions should be made. Detailed acute and chronic management guidelines for ventricular arrhythmias associated with AMI have been published.^2,3,53,56

Incidence

Holter monitoring has revealed that 15% of patients recovering from MI have supraventricular tachycardia, ranging from single atrial premature beats (APBs) to sustained AF during their hospital stay.⁶⁵ The prevalence of SVT increases during the first month after MI. Overall, the incidence of AF in AMI in the modern era is 7.8% to 28%.^66–71

Mechanism of Atrial Fibrillation During Acute Myocardial Ischemia

The pathophysiology of AF that occurs in the course of AMI has many components. Inflammation (pericarditis), changes in hemodynamics (atrial stretch and dilation), and atrial ischemia may all play a role.^72–75 Following significant ventricular damage, end-diastolic volume and pressure rise, causing an increase in atrial pressure and wall tension. This predisposes to AF and also explains the close relationship between heart failure and AF in the setting of MI. In an angiographic study, AF that occurred during inferior MI was shown to more likely occur in the setting of an occluded proximal left circumflex artery, with or without right CAD, if it was combined with impaired perfusion of the AV nodal artery.⁷⁶ AF did not occur in patients with right coronary artery occlusions if the circumflex artery was unobstructed. In a series of 266 patients, all 12 who developed atrial arrhythmias had inferior infarction. In the vast majority of these patients, the sinus node artery was distal to the site of right coronary occlusion, which suggests that sinus node ischemia may also play a role.⁷⁷ Evidence of atrial infarction in the 12-lead electrocardiogram (ECG; manifesting as PR-segment displacement) may also predict the onset of AF during AMI.⁷⁸ Other risk factors include advanced age, the presence of congestive heart failure, three-vessel coronary disease, right coronary artery (RCA) occlusion, female gender, anterior Q-wave MI, previous MI, and previous coronary artery bypass graft (CABG).^66,79

Consequences of Atrial Fibrillation During Acute Myocardial Infarction

The development of AF results in the loss of atrial contraction and rapid, irregular heart rates, which, in turn, will cause impaired diastolic filling and increased myocardial oxygen demand. Atrial contraction is an important component of ventricular filling, particularly in failing hearts. In the ischemic canine heart, induced AF was shown to cause a reduction in cardiac output, a fall in mean aortic pressure, and a fall in mean myocardial blood flow.⁸⁰ This may precipitate a vicious downward spiral, with AF exacerbating heart failure, which, in turn, promotes AF. Both will increase the ischemic burden and the likelihood of ventricular arrhythmias.

In patients who sustain an AMI, hospital mortality is significantly higher in those with AF than in those without it (Figure 57-3).^{66,68,69,71,72} It has been suggested that AF may be a risk factor for VF.⁸¹ AF occurs in patients with signs of heart failure and larger infarctions. In large-scale trials, the negative impact of AF has been shown to be independent of other variables.^70,79 However, it is possible that the increase in in-hospital mortality is restricted to those patients with new-onset AF (after admission) rather than pre-existing AF (see Figure 57-3).^70,82

FIGURE 57-3 Kaplan-Meier analysis of cumulative survival rates for patients with myocardial infarction during 8-year follow-up. Survival rates are stratified by presence or absence of atrial fibrillation (AF). These patients were divided into AF group 1 (45 patients who developed AF within 24 hours of onset of acute myocardial ischemia [AMI]; 5 ± 2 hours, range 1 to 15), AF group 2 (41 patients who developed AF 24 hours after onset of AMI; 6 ± 4 days, range 2 to 14), and AF group 3 (14 who developed AF before the onset of AMI). The remaining 939 patients were classified as sinus patients.

(From Sakata K, Kurihara H, Iwamori K, et al: Clinical and prognostic significance of atrial fibrillation in acute myocardial infarction, Am J Cardiol 80:1522–1527, 1997.)