Abstract
Background
The role of warfarin in anterior ST‐segment elevation myocardial infarction (STEMI) complicated by left ventricular (LV) dysfunction in patients treated with primary percutaneous coronary intervention (PCI) and dual antiplatelet therapy is unclear. Warfarin may prevent cardioembolic events but significantly increases bleeding in the setting of dual antiplatelet therapy.
Hypothesis
The incidence of LV thrombus in anterior STEMI patients treated with PCI is low, and clinical predictors might be valuable in determining patients at risk.
Methods
We performed a retrospective, single‐center study of 687 consecutive patients with anterior STEMI treated with PCI from 2006 to 2013. Baseline variables were evaluated in 310 patients at high risk for LV thrombus based on echocardiographic criteria. Patients with definite, probable, and no LV thrombus were compared by ANOVA, χ2, or t test where appropriate. Logistic regression analysis was performed.
Results
The incidence of LV thrombus was 15% (n = 47 probable/definite thrombus). Cardiac arrest was the only independent characteristic associated with increased risk of LV thrombus (odds ratio [OR]: 4.06, 95% confidence interval [CI]: 1.3‐12.7). Trends were observed for a lower risk in cardiogenic shock (OR: 0.33, 95% CI: 0.10‐1.05) and aspirin use at baseline (OR: 0.43, 95% CI: 0.17‐1.1). Treatment variables associated with LV thrombus included unfractionated heparin use post‐PCI (OR: 2.43, 95% CI: 1.16‐5.1) and use of balloon angioplasty without stent.
Conclusions
In contemporary practice with primary PCI, definite LV thrombus following anterior STEMI with LV dysfunction is challenging to predict. Further investigation is needed to determine if there is a subset of patients that should be treated with prophylactic warfarin.
Introduction
Left ventricular (LV) thrombus formation leading to cerebrovascular accident is a potentially catastrophic complication following anterior ST‐segment elevation myocardial infarction (STEMI). The incidence of LV thrombus has decreased in the primary percutaneous coronary intervention (PCI) era compared with the thrombolytic era, but it still occurs in the range of 2.9% to 15%.1 Systemic anticoagulation with warfarin in the presence of documented LV thrombus reduces the risk of systemic embolization.2 There is, however, limited data supporting the routine prophylactic use of warfarin following anterior myocardial infarction (MI) complicated by LV dysfunction. Furthermore, concomitant warfarin and dual antiplatelet therapy (DAPT) is associated with a high risk of major bleeding.3, 4 The 2013 American College of Cardiology/American Heart Association Guidelines, therefore, assigns a class IIb recommendation for the prophylactic use of warfarin in patients with STEMI and anterior apical akinesis or dyskinesis.5 However, there is limited data on identification of subsets of patients who may experience a net benefit from warfarin following anterior MI.
The goal of this study was to determine the incidence of LV thrombus in a subset of anterior STEMI patients deemed at highest risk for thrombotic complications: those with significant LV dysfunction or apical aneurysm. We then sought to identify clinical, angiographic, and therapeutic variables associated with LV thrombus formation.
Methods
Study Population
We performed a retrospective, single‐center observational study at Rhode Island Hospital, a tertiary referral hospital with a 24/7 primary PCI service. Included in the initial screening were 687 consecutive patients with anterior STEMI treated with primary PCI from January 2006 to May 2013. Patients were excluded if they did not have a transthoracic echocardiogram within 14 days of admission (n = 148). In addition, patients were excluded if they had preserved left ventricular ejection fraction (LVEF) or no anterior wall motion akinesis or dyskinesis (n = 229), as there is no guideline recommended indications for oral anticoagulation in this subset. Our analysis patients met criteria for LV dysfunction defined as an LVEF <40%, or <50% with apical hypokinesis or akinesis. Of the 687 patients screened, 310 (45%) patients met our prespecified inclusion criteria. Institutional review board approval was obtained through Lifespan/Brown University.
Data Collection
Demographic, clinical, and angiographic data were obtained from the electronic medical record. Cardiac catheterization procedure logs and reports were reviewed for procedure date, intraprocedural medications, procedural findings and outcomes, LV function, and regional wall motion abnormalities by left ventriculogram. Left ventricular function was assessed by a transthoracic echocardiogram in multiplanar views using a 16‐segment model for scoring the severity of wall motion abnormality. Left ventricular wall motion was created according to the American Society of Echocardiography on a scale of 1 to 5.6 The presence of LV thrombus was evaluated in apical 2‐ and 4‐chamber views.
A LV thrombus was noted to be definite when there was a discrete echodense mass in the LV that was distinct from the endocardium and adjacent to an area of hypokinetic or akinetic myocardium. Our interpreting cardiologist had the option of classifying a probable thrombus, if the images were technically challenging and there was an echodense mass that could not be excluded. In many cases the patients were critically ill and contrast echo was not performed. The echocardiograms of patients with probable thrombus were independently reviewed by 2 cardiologists board certified in echocardiography. Due to poor intra‐observed correlation, echocardiograms could not be reclassified as definite or no thrombus; therefore, probable thrombus was combined with definite thrombus and compared with patients with no LV thrombus. All clinical decisions were at the discretion of the treating cardiologists.
Statistical Analysis
Baseline and in‐hospital characteristics were compared in 3 groups according to the echocardiographic findings (no LV thrombus, probable LV thrombus, and definite LV thrombus) using a 1‐way ANOVA. A P value <0.05 was considered significant. Because of the small number of patients with definite LV thrombus and the lack of a gold standard for diagnosis, a dichotomous comparison was also performed grouping patients with definite and probable thrombus. The definite/probable and no LV thrombus groups were compared by using Pearson χ2 or a 2‐way t test where appropriate, with P < 0.05 considered statistically significant. Logistic regression analysis was performed using a stepwise multivariable approach on the dichotomous groups. Variables with P values <0.1 were retained in the multivariable model. All statistical analyses were performed using Stata/SE 13.0 (College Station, TX).
Results
A total of 310 consecutive patients met the inclusion criteria and were included in the analysis. Definite or probable LV thrombus was detected in 47 patients (15.2%). Only 9 patients (2.9%) had definitive thrombi, and the remaining 38 patients had probable thrombi (12.3%). The mean age of the study population was 61.8 ± 14 years, and 71.5% were male. Although the rate of LV thrombus detection varied each year, no statistical trend was observed over time.
Baseline characteristics in patients with definite LV thrombus, probable LV thrombus, and no LV thrombus are shown in Table 1. Analysis of these 3 groups was limited by the small number of patients with definite LV thrombus. There were no prehospital clinical characteristics that differed among the groups. Differences were observed in post‐PCI Thrombolysis In Myocardial Infarction (TIMI) flow, no stent use, clopidogrel use after admission, use of low‐molecular‐weight heparin in the emergency room, and unfractionated heparin use post‐catheterization. The definite LV thrombus group had a significantly lower mean LVEF at 29.7%, compared with 33.5% in the probable LV thrombus group and 36.8% in the no LV thrombus population (P = 0.01). The prevalence of anterior Q waves also differed (73% no thrombus, 84% probable, and 78% definite thrombus; P = 0.04). These variables, however, were not significant in the logistic regression model.
Table 1.
Baseline Characteristics in Patients With Definite Thrombus, Probable Thrombus, and No LV Thrombus Post–Anterior STEMI
| No Thrombus, N = 263 | Possible Thrombus, N = 38 | Thrombus, N = 9 | P Value | |
|---|---|---|---|---|
| Demographics | ||||
| Mean age, y | 62.4 | 59.5 | 57.4 | 0.06 |
| Male sex | 186 (70.7) | 27 (71.1) | 8 (88.9) | 0.5 |
| Smoker | 141 (56.6) | 23 (60.5) | 3 (33.3) | 0.5 |
| DM | 44 (16.7) | 5 (13.2) | 2 (22.2) | 0.77 |
| HTN | 153 (58.2) | 20 (52.6) | 3 (33.3) | 0.29 |
| Hyperlipidemia | 136 (51.9) | 12 (31.6) | 3 (33.3) | 0.16 |
| Family history of CAD | 125 (53.2) | 17 (44.7) | 5 (55.6) | 0.81 |
| Prior MI | 57 (21.7) | 5 (13.2) | 2 (22.2) | 0.47 |
| Acute MI characteristics | ||||
| Mean time from symptom onset to ER, h | 19.1 | 19.3 | 7.5 | 0.74 |
| Mean D2B time, min | 95.6 | 75.68 | 40.33 | 0.2 |
| Persistently elevated Tna | 142 (69.6) | 24 (63.2) | 6 (66.6) | 0.47 |
| Cardiogenic shock | 45 (17.1) | 6 (15.8) | 3 (33.3) | 0.43 |
| Cardiac arrest | 32 (12.2) | 9 (23.7) | 1 (11.1) | 0.15 |
| Angiographic findings | ||||
| Proximal LAD involvement | 151 (57.4) | 17 (44.7) | 3 (33.3) | 0.82 |
| Mean % stenosis | 96.6 | 97.7 | 98.2 | 0.59 |
| Pre‐TIMI flow of 0–1 | 204 (77.6) | 31 (81.6) | 8 (88.9) | 0.86 |
| Pre‐TIMI flow of 2 | 33 (12.5) | 4 (10.5) | 1 (11.1) | |
| Pre‐TIMI flow of 3 | 26 (9.9) | 3 (7.9) | 0 (0.0) | |
| Post‐TIMI flow of 0–1 | 3 (1.1) | 3 (7.9) | 0 (0.0) | 0.04 |
| Post‐TIMI flow of 2 | 8 (3.0) | 2 (5.3) | 1 (11.1) | |
| Post‐TIMI flow of 3 | 252 (95.8) | 33 (86.8) | 8 (88.9) | |
| No stent use | 13 (4.9) | 7 (18.4) | 1 (11.1) | 0.01 |
| BMS use | 102 (38.8) | 8 (21.1) | 5 (55.6) | |
| DES use | 148 (56.3) | 23 (60.5) | 3 (33.3) | |
| Regional wall motion abnormality | 222 (95.3) | 32 (84.2) | 9 (100.0) | 0.74 |
| Dyskinesis noted | 30 (12.9) | 5 (13.2) | 0 (0.0) | 0.1 |
| Medication use | ||||
| Chronic aspirin use | 88 (34.0) | 8 (21.1) | 1 (11.1) | 0.27 |
| ASA in ER | 247 (93.9) | 35 (92.1) | 9 (100.0) | 0.1 |
| GP IIb/IIIa inhibitor | 83 (31.8) | 7 (18.4) | 4 (44.4) | 0.4 |
| Chronic clopidogrel | 16 (6.2) | 1 (2.6) | 1 (11.1) | 0.71 |
| Clopidogrel ER or post‐cath | 243 (92.4) | 37 (97.4) | 9 (100.0) | 0.03 |
| Prasugrel post‐cath | 5 (1.9) | 1 (2.6) | 0 (0.0) | 0.96 |
| Ticagrelor post‐cath | 1 (0.4) | 0 (0.0) | 0 (0.0) | 0.97 |
| Chronic warfarin | 12 (4.7) | 4 (10.5) | 0 (0.0) | 0.42 |
| Warfarin post‐cath | 87 (33.6) | 13 (34.2) | 2 (22.2) | 0.91 |
| LMWH in ER | 3 (1.2) | 1 (2.6) | 0 (0.0) | <0.01 |
| LMWH during cath | 2 (0.7) | 1 (2.6) | 0 (0.0) | 0.83 |
| LMWH post‐cath | 24 (9.3) | 2 (5.3) | 1 (11.1) | 0.25 |
| UFH in ER or cath | 257 (98.5) | 37 (97.4) | 9 (100.0) | 0.94 |
| UFH post‐cath | 144 (57.6) | 29 (76.3) | 8 (88.9) | 0.04 |
| ECG | ||||
| Anterior Q waves | 192 (73.0) | 32 (84.2) | 7 (77.8) | 0.04 |
| Persistent ST elevation | 145 (55.1) | 23 (60.5) | 6 (66.6) | 0.08 |
| Atrial fibrillation/flutter | 9 (3.4) | 3 (7.9) | 0 (0.0) | 0.3 |
| Echo | ||||
| Mean time to first echo, h | 45.1 | 58.8 | 64.1 | 0.1 |
| Mean LVEF, % | 36.8 | 33.50 | 29.70 | 0.01 |
| Anterior hypo/akinesis | 261 (99.2) | 38 (100.0) | 9 (100.0) | 0.84 |
| Dyskinesis echo | 38 (14.5) | 7 (18.4) | 3 (33.3) | 0.27 |
Abbreviations: ASA, aspirin; BMS, bare‐metal stent; CAD, coronary artery disease; cath, catheterization; D2B, door to balloon; DES, drug‐eluting stent; DM, diabetes mellitus; ECG, electrocardiography; echo, echocardiography; ER, emergency room; GP, glycoprotein; HTN, hypertension; LAD, left anterior descending artery; LMWH, low‐molecular‐weight heparin; LV, left ventricular; LVEF, left ventricular ejection fraction; MI, myocardial infarction; STEMI, ST‐segment elevation myocardial infarction; TIMI, Thrombolysis In Myocardial Infarction; Tn, troponin; UFH, unfractionated heparin.
Data are presented as n (%) unless otherwise noted.
≥2 elevated troponins.
Categorization and analysis in 2 groups, definite or probable LV thrombus compared with no LV thrombus, is shown in Table 2. Patients with LV thrombus were less likely to have hyperlipidemia (31.9% vs 51.9%, P < 0.05) and more likely to have postprocedure TIMI flow grade 0–1 (6.4% vs 1.1%, P = 0.03) and no stent placed (17.0% vs 4.9%, P ≤ 0.01). Patients found to have LV thrombus had a lower mean LVEF (32.8% vs 36.8%, P ≤ 0.01). With respect to postprocedural medications, administration of unfractionated heparin postcatheterization was common, but more so in patients with LV thrombus (80.4% vs 57.6%, P = 0.01). Baseline characteristics were otherwise similar, including time of symptom onset to hospital arrival and door‐to‐balloon time.
Table 2.
Baseline Characteristics in Patients With (Definite/Probable) and Without LV Thrombus Post–Anterior STEMI
| No Thrombus, N = 263 | Thrombus, N = 47 | P Value | |
|---|---|---|---|
| Demographics | |||
| Mean age, y | 62.4 | 59.1 | 0.07 |
| Male sex | 186 (70.7) | 35 (74.5) | 0.6 |
| Smoker | 141 (56.6) | 26 (57.8) | 0.95 |
| DM | 44 (16.7) | 7 (14.9) | 0.75 |
| HTN | 153 (58.2) | 23 (48.9) | 0.24 |
| Hyperlipidemia | 136 (51.9) | 15 (31.9) | 0.04 |
| Prior MI | 57 (21.7) | 7 (14.9) | 0.29 |
| Acute MI characteristics | |||
| Mean time from symptom onset to ER, h | 19.1 | 20.2 | 0.56 |
| Mean D2B time, min | 95.6 | 99.1 | 0.39 |
| Persistently elevated Tna | 142 (69.6) | 30 (81.1) | 0.26 |
| Cardiogenic shock | 45 (17.1) | 9 (19.1) | 0.73 |
| Cardiac arrest | 32 (12.2) | 10 (21.3) | 0.09 |
| Angiographic findings | |||
| Proximal LAD involvement | 151 (57.4) | 27 (57.4) | 0.99 |
| Mean % stenosis | 96.6 | 97.8 | 0.85 |
| Pre‐TIMI flow of 0–1 | 204 (77.6) | 39 (83.0) | 0.67 |
| Pre‐TIMI flow of 2 | 33 (12.5) | 5 (10.6) | |
| Pre‐TIMI flow of 3 | 26 (9.9) | 3 (6.4) | |
| Post‐TIMI flow of 0–1 | 3 (1.1) | 3 (6.4) | 0.03 |
| Post‐TIMI flow of 2 | 8 (3.0) | 3 (6.4) | |
| Post‐TIMI flow of 3 | 252 (95.8) | 41 (87.2) | |
| No stent use | 13 (4.9) | 8 (17.0) | <0.01 |
| BMS use | 102 (38.8) | 13 (27.7) | |
| DES use | 148 (56.3) | 26 (55.3) | |
| Regional wall motion abnormality | 222 (84.4) | 41 (87.2) | 0.78 |
| Apical dyskinesis | 203 (11.4) | 5 (10.6) | 0.97 |
| Medication use | |||
| Chronic ASA | 88 (31.8) | 9 (19.1) | 0.09 |
| ASA in ER | 247 (93.9) | 44 (95.7) | 0.05 |
| GP IIb/IIIa inhibitor | 83 (31.6) | 11 (23.4) | 0.43 |
| Chronic clopidogrel | 16 (6.0) | 2 (4.3) | 0.55 |
| Clopidogrel ER or post‐cath | 249 (95.4) | 46 (97.9) | 0.62 |
| Prasugrel post‐cath | 5 (1.9) | 1 (2.1) | 0.83 |
| Ticagrelor post‐cath | 1 (0.4) | 0 (0.0) | 0.76 |
| Chronic warfarin | 12 (4.7) | 4 (8.5) | 0.32 |
| Warfarin post‐cath | 87 (33.6) | 15 (32.6) | 0.95 |
| LMWH in ER | 3 (1.2) | 1 (2.3) | 0.05 |
| LMWH during cath | 2 (0.8) | 1 (2.1) | 0.62 |
| LMWH post‐cath | 24 (9.3) | 3 (6.7) | 0.84 |
| UFH in ER or cath | 257 (98.4) | 46 (97.9) | 0.8 |
| UFH post‐cath | 144 (57.6) | 37 (80.4) | 0.01 |
| ECG | |||
| Anterior Q waves | 192 (74.8) | 39 (83.0) | 0.02 |
| Persistent ST elevation | 145 (55.3) | 29 (61.7) | 0.04 |
| Atrial fibrillation/flutter | 9 (3.4) | 3 (6.5) | 0.24 |
| Echo | |||
| Mean time to first echo, h | 45.1 | 59.8 | 0.98 |
| Mean LVEF, % | 36.8 | 32.8 | <0.01 |
| Anterior hypo/akinesis | 261 (95.6) | 47 (100.0) | 0.55 |
| Apical dyskinesis | 38 (13.9) | 10 (21.3) | 0.23 |
Abbreviations: ASA, aspirin; BMS, bare‐metal stent; CAD, coronary artery disease; cath, catheterization; D2B, door to balloon; DES, drug‐eluting stent; DM, diabetes mellitus; ECG, electrocardiography; echo, echocardiography; ER, emergency room; GP, glycoprotein; HTN, hypertension; LAD, left anterior descending artery; LMWH, low‐molecular‐weight heparin; LV, left ventricular; LVEF, left ventricular ejection fraction; MI, myocardial infarction; STEMI, ST‐segment elevation myocardial infarction; TIMI, Thrombolysis In Myocardial Infarction; Tn, troponin; UFH, unfractionated heparin.
Data are presented as n (%) unless otherwise noted.
≥2 elevated troponins.
A multivariate logistical regression model was performed to assess independent predictors of LV thrombus (Table 3). The only clinical predictor of an increased risk of LV thrombus was cardiac arrest (odds ratio [OR]: 4.1, 95% confidence interval [CI]: 1.3‐12.7). Trends were observed for a lower risk of LV thrombus in cardiogenic shock (OR: 0.33, 95% CI: 0.10‐1.05) and aspirin use at baseline (OR: 0.43, 95% CI: 0.17‐1.1). Treatment variables associated with LV thrombus included unfractionated heparin use post‐PCI (OR: 2.43, 95% CI: 1.16‐5.1) and use of balloon angioplasty without stent (drug‐eluting stent OR: 0.22, 95% CI: 0.06‐0.77; bare‐metal stent OR: 0.23, 95% CI: 0.07‐0.75, compared with balloon angioplasty).
Table 3.
Independent Predictors of LV Thrombus Following PCI for Anterior STEMI in Patients With LV Dysfunction
| Variables | OR | 95% CI | P Value |
|---|---|---|---|
| Baseline clinical characteristics | |||
| Cardiogenic shock | 0.33 | 0.10–1.05 | 0.06 |
| Cardiac arrest | 4.06 | 1.3–12.7 | 0.02 |
| ASA at baseline | 0.43 | 0.17–1.1 | 0.07 |
| Treatment characteristics | |||
| DESa | 0.22 | 0.06–0.77 | 0.02 |
| BMSa | 0.23 | 0.07–0.75 | 0.02 |
| UFH post‐cath | 2.43 | 1.16–5.1 | 0.02 |
Abbreviations: ASA, aspirin; BMS, bare‐metal stent; cath, catheterization; CI, confidence interval; DES, drug‐eluting stent; LV, left ventricular; OR, odds ratio; PCI, percutaneous coronary intervention; STEMI, ST‐segment elevation myocardial infarction; UFH, unfractionated heparin.
Compared with balloon angioplasty.
Discussion
In this study of a large number of patients with anterior STEMI treated with PCI that had significant LV dysfunction, we observed an incidence of early LV thrombus of 15.2%. In the prethrombolytic and thrombolytic eras, LV thrombus was very common in anterior MI, occurring in up to 40% and 28% of patients, respectively.7, 8 The advent of primary PCI, with earlier reperfusion and shorter door‐to‐balloon times, and use of aggressive antiplatelet and antithrombin therapy, has led to a decrease in the incidence of LV thrombus. Several studies in the primary PCI era have evaluated LV thrombus in patients with STEMI by transthoracic echocardiogram. Studies that include all patients with anterior STEMI, independent of LVEF, have observed lower rates of thrombus (2.9% to 7.1%) compared with our study.9, 10, 11 In a higher‐risk group of anterior‐wall MI (AWMI) patients treated with PCI enrolled in the Autologous Stem Cell Transplantation in Acute Myocardial Infarction (ASTAMI) trial, 15% of patients had LV thrombus detected by serial echocardiography, and 10% were identified within the first week.12 The rate of definite or probable LV thrombus we observed suggests that patients with anterior STEMI and an LV of <40% or <50% with an apical aneurysm are a group that remains at risk for cardioembolic complications in the primary PCI era.
Consistently identified risk factors for LV thrombus formation in STEMI include anterior location, large infarct size, and LV aneurysm.1 In our study, LV thrombus was more common in patients with suboptimal revascularization, such as those with TIMI flow grade 0–1. The use of either drug‐eluting or bare‐metal stents, compared with balloon angioplasty, was independently associated with a lower risk of LV thrombus. These findings are both likely related to infarct size. Several studies have shown an association between lower TIMI flow grade and larger infarcts. Stents decrease the risk of reinfarction and target‐vessel revascularization when compared with angioplasty; however, stents may not be placed in patients with no reflow or slow flow.
In our study, which included only anterior STEMI patients with LV dysfunction, we identified cardiac arrest as the only clinical characteristic independently associated with development of LV thrombus. Patients with cardiac arrest may have larger infarcts due to proximal vessel location, lack of preconditioning, or potentiation of transmural injury due to hypotension. In addition, the transient stasis of blood associated with cardiac arrest may also predispose to thrombosis in the arterial system. A novel swine model of cardiac arrest has consistent shown the development of LV thrombosis with resolution with resuscitation.13 The lack of additional clinical predictors of LV thrombus emphasizes the difficulty clinicians have in determining “at risk” patients and weighing risks and benefits of use of oral anticoagulation.
The utilization of unfractionated heparin (UFH) postprocedure was associated with an increased risk of LV thrombus in our study. We hypothesize that cardiologists at our facility used UFH in patients suspected to have thrombus based on left ventriculogram or in whom they believed were at high risk for thrombus formation, rather than suggesting a direct causality between UFH and thrombus. At present, there are no pharmacological agents or therapeutic strategies that have definitely been shown to impact the risk of LV thrombus formation. Prior studies have produced conflicting results regarding the role of UFH. A randomized controlled trial by Turpie et al found that MI survivors treated with high‐dose heparin had a significantly lower incidence of LV thrombus compared with patients receiving low‐dose heparin.14 On the other hand, the Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto (GISSI)‐2–connected study found that high‐dose heparin did not prevent thrombus formation.8 A small randomized trial examined 2 anticoagulation strategies to prevent LV thrombus in patients with acute anterior MI with Q waves and an LVEF <40%. Patients were assigned to low‐molecular‐weight heparin (LMWH) for 30 days or to UFH bridge to warfarin with oral anticoagulation for 3 months. At 3 months, more patients in the LMWH group had probable thrombus (15% vs 4%), although the difference did not meet statistical significance.15 We did not collect information on the dosing of UFH used, and LMWH was used in <10% of patients in our study. Overall, we observed a high rate of UFH use post‐PCI in this patient cohort, which reflects the practice prior to the 2013 STEMI guidelines that reclassified anticoagulation in these patients from class I to class IIb.5
Identification of patients at high risk for cardioembolic events post‐MI remains an important question. There is unclear benefit and potential harm to prophylactically starting antithrombin therapy in AWMI patients receiving DAPT post stenting. Le May et al performed a retrospective study comparing outcomes in AWMI patients without LV thrombus treated empirically with or without warfarin. At 180 days, all‐cause mortality, stroke, reinfarction, and major bleeding were significantly higher in the warfarin group (OR: 4.0, 95% CI: 2.1‐7.5).9 The outcomes may be explained in part by the risk of DAPT and antithrombin therapy, termed triple therapy. Compared with DAPT alone or warfarin with no or a single antiplatelet agent, triple therapy increases the risk of bleeding 2‐ to 3‐fold.3, 16 In a small study examining outcomes of PCI patients treated with triple therapy for atrial fibrillation or other indication, the rate of major bleeding was 4.7% at approximately 6 months, two‐thirds of the events occurred within the first month, and 50% of these major bleeding events were fatal.17 Other strategies to reduce bleeding in PCI patients requiring anticoagulation have been examined. The use of clopidogrel as a single antiplatelet agent with warfarin after coronary stenting was compared with triple therapy, and although bleeding was reduced and no difference in ischemic outcomes was observed, the safety of this strategy is still debatable.18 A recent meta‐analysis from Gao et al4 reaffirms the risk of triple therapy and that alternative therapeutic regimens in PCI patients requiring oral anticoagulation may reduce bleeding without compromising safety.15
Study Limitations
Our findings are limited by the retrospective single‐center nature of our study, which may introduce detection and treatment bias and limit generalizability. In addition, echocardiograms were performed early and not at a routine time point post‐PCI, and therefore the incidence of thrombus may be underestimated based only on a single echocardiogram. Echo contrast was not routinely used, leading to a high rate of probable thrombus reading, which may overestimate the incidence of true thrombus. Due to the nature of the regression analysis, unmeasured variables may exist that could be significant predictors of LV thrombus. The power to detect significant changes may be limited by sample size. The majority of patients were treated with DAPT with aspirin and clopidogrel; therefore, no comment can be made about the more potent P2Y12 inhibitors prasugrel and ticagrelor.
Conclusion
In summary, our study confirmed that the incidence of definitive LV thrombus following anterior STEMI with LV dysfunction is low, but a significant number of additional patients have probable thrombus. A greater use of echo contrast, serial echocardiography, or imaging with magnetic resonance imaging may be useful for reclassification. Patients with cardiac arrest and those who do not receive stents are subgroups with greatest risk. In the absence of studies confirming a benefit of prophylactic warfarin, high‐risk patients such as those in our study should be followed closely by noninvasive imaging in the first month following AWMI, and in those identified to have LV thrombus the risks and benefits of anticoagulation can be individualized.
This research was funded by an NIH/NHLBI, T35 HL094308‐03 training grant.
The authors have no other funding, financial relationships, or conflicts of interest to disclose.
References
- 1. Delewi R, Zijlstra F, Piek JJ. Left ventricular thrombus formation after acute myocardial infarction. Heart. 2012;98:1743–1749. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Vaitkus PT, Barnathan ES. Embolic potential, prevention and management of mural thrombus complicating anterior myocardial infarction: a meta‐analysis. J Am Coll Cardiol. 1993;22:1004–1009. [DOI] [PubMed] [Google Scholar]
- 3. Gao F, Zhou YJ, Wang ZJ, et al. Meta‐analysis of the combination of warfarin and dual antiplatelet therapy after coronary stenting in patients with indications for chronic oral anticoagulation. Int J Cardiol. 2011;148:96–101. [DOI] [PubMed] [Google Scholar]
- 4. Gao XF, Chen Y, Fan ZG, et al. Antithrombotic regimens for patients taking oral anticoagulation after coronary intervention: a meta‐analysis of 16 clinical trials and 9185 patients. Clin Cardiol. 2015. doi: 10.1002/clc.22411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. O'Gara PT, Kushner FG, Ascheim DD, et al. 2013 ACCF/AHA guideline for the management of ST‐elevation myocardial infarction: executive summary: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines: developed in collaboration with the American College of Emergency Physicians and Society for Cardiovascular Angiography and Interventions. Catheter Cardiovasc Interv. 2013;82:E1–E27. [DOI] [PubMed] [Google Scholar]
- 6. Lang RM, Bierig M, Devereux RB, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography's Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr. 2005;18:1440–1463. [DOI] [PubMed] [Google Scholar]
- 7. Nihoyannopoulos P, Smith GC, Maseri A, et al. The natural history of left ventricular thrombus in myocardial infarction: a rationale in support of masterly inactivity. J Am Coll Cardiol. 1989;14:903–911. [DOI] [PubMed] [Google Scholar]
- 8. Vecchio C, Chiarella F, Lupi G, et al. Left ventricular thrombus in anterior acute myocardial infarction after thrombolysis: a GISSI‐2 connected study. Circulation. 1991;84:512–519. [DOI] [PubMed] [Google Scholar]
- 9. Le May MR, Acharya S, Wells GA, et al. Prophylactic warfarin therapy after primary percutaneous coronary intervention for anterior ST‐segment elevation myocardial infarction. JACC Cardiovasc Interv. 2015;8(1 part B):155–162. [DOI] [PubMed] [Google Scholar]
- 10. Osherov AB, Borovik‐Raz M, Aronson D, et al. Incidence of early left ventricular thrombus after acute anterior wall myocardial infarction in the primary coronary intervention era. Am Heart J. 2009;157:1074–1080. [DOI] [PubMed] [Google Scholar]
- 11. Shacham Y, Leshem‐Rubinow E, Ben Assa E, et al. Frequency and correlates of early left ventricular thrombus formation following anterior wall acute myocardial infarction treated with primary percutaneous coronary intervention. Am J Cardiol. 2013;111:667–670. [DOI] [PubMed] [Google Scholar]
- 12. Solheim S, Seljeflot I, Lunde K, et al. Frequency of left ventricular thrombus in patients with anterior wall acute myocardial infarction treated with percutaneous coronary intervention and dual antiplatelet therapy. Am J Cardiol. 2010;106:1197–1200. [DOI] [PubMed] [Google Scholar]
- 13. Budhram GR, Mader TJ, Lutfy L, et al. Left ventricular thrombus development during ventricular fibrillation and resolution during resuscitation in a swine model of sudden cardiac arrest. Resuscitation. 2014;85:689–693. [DOI] [PubMed] [Google Scholar]
- 14. Turpie AG, Robinson JG, Doyle DJ, et al. Comparison of high‐dose with low‐dose subcutaneous heparin to prevent left ventricular mural thrombosis in patients with acute transmural anterior myocardial infarction. N Engl J Med. 1989;320:352–357. [DOI] [PubMed] [Google Scholar]
- 15. White DC, Grines CL, Grines LL, et al. Comparison of the usefulness of enoxaparin versus warfarin for prevention of left ventricular mural thrombus after anterior wall acute myocardial infarction. Am J Cardiol. 2015;115:1200–1203. [DOI] [PubMed] [Google Scholar]
- 16. Nikolsky E, Mehran R, Dangas GD, et al. Outcomes of patients treated with triple antithrombotic therapy after primary percutaneous coronary intervention for ST‐elevation myocardial infarction (from the Harmonizing Outcomes With Revascularization and Stents in Acute Myocardial Infarction [HORIZONS‐AMI] trial). Am J Cardiol. 2012;109:831–838. [DOI] [PubMed] [Google Scholar]
- 17. Rogacka R, Chieffo A, Michev I, et al. Dual antiplatelet therapy after percutaneous coronary intervention with stent implantation in patients taking chronic oral anticoagulation. JACC Cardiovasc Interv. 2008;1:56–61. [DOI] [PubMed] [Google Scholar]
- 18. Dewilde WJ, Oirbans T, Verheugt FW, et al. WOEST Study Investigators . Use of clopidogrel with or without aspirin in patients taking oral anticoagulant therapy and undergoing percutaneous coronary intervention: an open‐label, randomised, controlled trial. Lancet. 2013;381:1107–1115. [DOI] [PubMed] [Google Scholar]