Guidelines do not recommend an implantable cardioverter defibrillator (ICD) for prevention of sudden death in patients who develop ventricular arrhythmia (VA) within 48 hours of acute myocardial infarction (AMI) if they are successfully revascularized. We aimed to determine long-term survival in a cohort of early VA survivors treated with percutaneous coronary intervention (PCI) and to determine whether certain high-risk characteristics predicted worse outcomes. This retrospective study included all patients with early VA after AMI treated with PCI at our institution from 2002 to 2012 who survived to hospital discharge. Patients who had an ICD before their index AMI and those who received ICD before hospital discharge were excluded. Overall survival in the early VA survivors was analyzed based on post-MI left ventricular ejection fraction (LVEF) (≥50% vs <50%), MI type (ST-segment elevation myocardial infarction [STEMI] vs non–ST-segment elevation myocardial infarction [NSTEMI]), and single-vessel versus multivessel disease. Baseline presenting clinical and PCI characteristics plus outcomes were compared with matched controls with PCI-treated AMI but no early VA. Of the 79 early VA survivors treated with PCI, there were no significant differences in long-term overall survival between AMI type (STEMI vs NSTEMI), single-vessel versus multivessel disease, and LVEF at time of MI (>50% vs <50%). Despite having lower presenting LVEF (46% vs 55%, p <0.001) and higher rates of cardiogenic shock (28% vs 4%; p <0.001), survivors of early VA had similar overall survival compared with PCI-treated controls whose post-AMI hospital course was not complicated by early VA (p = 0.61). In conclusion, patients with early VA treated with PCI who survive to discharge were more likely to have STEMI, lower LVEF, and cardiogenic shock. Type of AMI or the presence of systolic dysfunction or multivessel disease did not predict long-term mortality. With early PCI, early VA survivors have similar long-term prognosis compared with those without early VA.
Highlights
- •
Presenting characteristics and long-term outcomes of patients with early ventricular arrhythmias after acute myocardial infarction (MI) are analyzed in this study.
- •
Patients with early ventricular arrhythmias were more likely to have ST-segment elevation (vs non–ST-segment elevation myocardial infarction), lower left ventricular ejection fraction, and more frequently presented in cardiogenic shock.
- •
In the era of primary revascularization for acute MI, the presence of early ventricular arrhythmia after MI does not portend a worse long-term prognosis in those who survive to hospital discharge.
The purpose of the present study was to explore the presenting clinical and PCI procedural characteristics in PCI-treated myocardial infarction (MI) survivors with early VA and to examine long-term outcomes compared with PCI-treated patients whose MI is not complicated by early VA. Additionally, we aimed to determine whether certain presenting characteristics, such as MI type (STEMI vs NSTEMI), LVEF at presentation, and presence of single-vessel versus multivessel disease-predicted long-term outcomes in early VA survivors treated with PCI.
Methods
Approval to conduct this retrospective study was granted by the Mayo Clinic’s Institutional Review Board. All patients referred to Mayo Clinic, Rochester, Minnesota, for management of acute MI from January 2002 to January 2012 who developed ≥1 episodes of sustained (>30 seconds) VA within 48 hours of symptom onset (early VA) and survived to hospital discharge were eligible for inclusion. The identification of our early VA cohort has been described previously in detail. For this study, “MI” included STEMI and NSTEMI and was diagnosed in the presence of 2 of the following 3 criteria: (1) typical chest pain lasting >20 minutes, (2) elevation of creatinine kinase (MB fraction) >2 times normal or Troponin T level >0.01 ng/ml (ninety-ninth percentile cutoff) in at least 1 blood sample, when drawn serially in the setting of strong clinical suspicion, and (3) new Q waves on electrocardiogram. Patients with stable, low-grade biomarker elevation after VA without evidence of a culprit lesion on angiography were excluded because biomarker elevation was attributed primarily to the VA itself and cardiac resuscitation rather than secondary to AMI. VA included ventricular fibrillation (VF) and sustained monomorphic or polymorphic ventricular tachycardia with rate >100 beats/min, lasting at least 30 seconds or requiring treatment with cardioversion/defibrillation because of hemodynamic instability. “Early VA” was defined as VA occurring within 48 hours of AMI symptom onset.
Patients with the following attributes were excluded from analysis: (1) VA not attributed to classical acute coronary artery thrombosis (e.g., reversible electrolyte imbalance, spontaneous coronary artery dissection, stress cardiomyopathy, vasospasm, etc.), (2) those who received ICD during the index hospitalization, (3) baseline ischemic or nonischemic cardiomyopathy (LVEF <35%) which would justify ICD for primary prevention of SCD, (4) VA in the setting of stable or unstable coronary artery disease with no elevation of cardiac biomarkers, (5) previous ICD implantation, (6) PCI performed at outside facility before transfer, (7) out-of-hospital arrest without additional documented inhospital early VA, and (8) patients who experienced only intraprocedural VA in the catheterization laboratory during PCI. As one of the primary aims of this study was to examine long-term outcomes in early VA survivors, we also excluded subjects who died during hospitalization or did not have at least 1 follow-up visit recorded.
All patients received standard of care management for AMI and underwent revascularization with PCI at our institution. The control group was selected from our institution’s PCI database. All control patients had AMI and were treated with PCI but did not experience early VA. Each early VA case was matched to 2 controls by PCI date, age (within 5 years), gender, and the presence or absence of multivessel coronary artery disease. Baseline clinical and presenting characteristics and outcomes were compared between groups.
Baseline characteristics and data for the type of MI, method of revascularization, severity of coronary artery disease (single-vessel or multivessel involvement), door-to-balloon time, number of stents placed, PCI success, inhospital medication administration, presentation with cardiogenic shock, and intra-aortic balloon pump use were obtained from the Mayo Clinic electronic medical record and the institutional PCI registry, which has been described previously. Survival was determined using the Mayo Clinic registration database and ACCURINT (LexisNexis, Dayton, OH), an institutionally approved Web-based resource and location service, as previously described. Where available, death certificates and autopsy reports were also reviewed to determine cause of death.
The statistical analysis was performed using SAS, version 9.3, computer software (SAS Institute, Cary, North Carolina). Results are presented as mean (SD) for continuous measures or median (twenty-fifth and seventy-fifth percentiles) for skewed distributions. Categorical measures are presented as number (percentage). Differences between groups are tested using conditional logistic regression to account for the matched nature of the data. Survival was summarized with the Kaplan-Meier curves. Differences in group survival between VA subjects and their matched reference subjects were tested using Cox proportional hazards regression with the baseline hazard stratified by matched case-control groups. For comparisons of survival between subgroups of the VA patients, a log-rank test was used; p <0.05 was considered statistically significant.
Results
A total of 128 patients were identified to have early VA after AMI over the 10-year study period, 26 of whom were excluded because they received ICD before discharge. Of the 102 remaining early VA survivors, 79 were treated with PCI at our institution. These 79 patients were matched to 158 control patients with PCI-treated AMI but no early VA.
Baseline clinical characteristics of both groups are listed in Table 1 . Presenting clinical AMI characteristics, angiographic findings, PCI revascularization characteristics, and inhospital and discharge medications are documented in Table 2 . The patients with early VA were more likely to have STEMI (82% vs 55%, p <0.001) and lower LVEF at presentation (46.3 ± 15.2 vs 54.5 ± 10.9%, p <0.001). Additionally, patients with early VA were more likely to have presented in cardiogenic shock before PCI (28% vs 4%, p <0.001) and be treated with intra-aortic balloon pump (9% vs 1%, p <0.001). The total number and distribution of vessels intervened on and number of stents placed were similar between groups, although the early VA patients less frequently received drug-eluting stents (53% vs 65%, p = 0.048).
| Variable | Controls ( n =158) | Early VA ( n =79) | P value |
|---|---|---|---|
| Age (years) ∗ | 62.9 ± 13.3 | 63.3 ± 13.5 | 0.001 |
| Men ∗ | 114 (72%) | 57 (72%) | NS |
| Body mass index (kg/m 2 ) | 30.4 ± 5.9 | 30.0 ± 5.4 | 0.76 |
| Smoker | 94 (60%) | 57 (73%) | 0.03 |
| Diabetes mellitus | 42 (27%) | 15 (19%) | 0.18 |
| Hypertension | 98 (66%) | 53 (72%) | 0.46 |
| Hyperlipidemia | 117 (74%) | 56 (71%) | 0.61 |
| Chronic kidney disease | 4 (3%) | 2 (3%) | >0.99 |
| Heart failure | 24 (15%) | 11 (14%) | >0.99 |
| Prior PCI | 36 (23%) | 9 (11%) | 0.03 |
| Prior CABG | 14 (9%) | 7 (9%) | >0.99 |
∗ Gender and age were used in the matching algorithm to identify controls.
| Controls ( n =158) | Early VA ( n =79) | P value | |
|---|---|---|---|
| MI Characteristics | |||
| ST-elevation MI | 86 (55%) | 63 (82%) | <0.001 |
| Cardiogenic shock | 6 (4%) | 22 (28%) | <0.001 |
| Intraaortic balloon pump | 1 (1%) | 7 (9%) | <0.001 |
| Left ventricular ejection fraction (%) | 54.5 ± 10.9 | 46.3 ± 15.2 | <0.001 |
| Multivessel coronary artery disease | 70 (44%) | 33 (43%) | ∗ |
| Number of coronary arteries with significant (70%) disease | ∗ | ||
| 1 | 88 (56%) | 43 (57%) | |
| 2 | 50 (32%) | 22 (29%) | |
| 3 | 20 (13%) | 10 (13%) | |
| Left main >70% disease | 2 (1%) | 2 (3%) | 0.24 |
| LAD >70% disease | 94 (59%) | 48 (61%) | 0.66 |
| Circumflex >70% disease | 62 (39%) | 32 (41%) | 0.91 |
| RCA >70% disease | 89 (56%) | 38 (48%) | 0.28 |
| Fibrinolytics given | 12 (8%) | 22 (28%) | <0.001 |
| Door-to-balloon time (minutes) | 206.8 ± 286.2 | 99.5 ± 118.4 | 0.01 |
| PCI native left main | 2 (1%) | 1 (2%) | 0.79 |
| PCI native LAD | 62 (41%) | 31 (42%) | 0.77 |
| PCI native circumflex | 35 (23%) | 18 (25%) | 0.96 |
| PCI native RCA | 62 (41%) | 26 (36%) | 0.66 |
| PCI in graft | 7 (5%) | 2 (3%) | 0.59 |
| Total lesions treated | 1.1 ± 0.3 | 1.1 ± 0.3 | 0.64 |
| Multiple vessels treated | 12 (8%) | 5 (7%) | 0.85 |
| Drug-eluting stents used | 100 (65%) | 39 (53%) | 0.05 |
| Successful PCI † | 142 (90%) | 70 (89%) | 0.75 |
| Emergency CABG | 2 (1%) | 0 (0%) | 0.20 |
| Medications at hospital discharge | |||
| Aspirin | 153 (97%) | 79 (100%) | 0.07 |
| Thienopyridine | 152 (96%) | 76 (96%) | NS |
| Statin | 99 (95%) | 49 (98%) | 0.36 |
| ACE-inhibitor or ARB | 109 (69%) | 71 (90%) | <0.001 |
| Beta-blocker | 148 (94%) | 76 (96%) | 0.40 |
| Antiarrhythmic drug | 1 (1%) | 5 (6%) | 0.01 |
| Digoxin | 6 (4%) | 4 (5%) | 0.66 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
