Is 3–6 months anticoagulation with warfarin necessary after left ventricular thrombectomy with left ventricular aneurysm surgery?

Abstract

Objectives

No recommendation exists on the optimal antithrombotic therapy after left ventricular thrombus (LVT) resection in the current guidelines. The study aimed to investigate the role of prophylactic anticoagulation with warfarin for 3‐6 months in LVT recurrence and other clinical outcomes after LVT resection and left ventricular aneurysm (LVA) surgery.

Methods

All consecutive patients undergoing LVT resection together with LVA surgery in our institution between 2010.1.1 and 2021.4.1 were included in the study. Individuals included were divided into two groups based on whether warfarin was administered at discharge. Patients with warfarin were matched to their counterparts without warfarin based on the baseline characteristics via propensity score matching (PSM) at the ratio of 1:1. The primary outcome was LVT recurrence. The secondary outcomes were major adverse cardiac and cerebrovascular events (MACCEs) and the composite endpoint of LVT recurrence and MACCEs.

Results

After PSM, a total of 118 patients were included in the study, among whom 59 received warfarin therapy at discharge and 59 didn't. During the median follow‐up of 56.5 months, 21 out of 118 patients had LVT recurrence and the recurrence rate was 17.8% There was no systemic embolism resulting from the recurrent LVT. Kaplan–Meir analysis showed no significant difference in LVT recurrence (p = .86), MACCEs (p = .48) and the composite endpoint of LVT recurrence, and MACCEs (p = .89). Cox proportional hazards regression model showed that history of PCI (hazard ratio [HR] 2.778, 95% confidence interval [CI] 1.087–7.100, p = .033) and LVA surgical strategy of linear suture (HR 8.768, 95% CI 1.139–67.517, p = .037) were independent risk factors of LVT recurrence.

Conclusions

Prophylactic anticoagulation with warfarin for 3–6 months may be unnecessary with no benefit in terms of LVT recurrence and other clinical outcomes.

Abbreviations

ACCF/AHA

American College of Cardiology Foundation/American Heart Association

ACEI/ARB

angiotensin‐converting enzyme inhibitor/angiotensin receptor blocker

CABG

coronary artery bypass grafting

CRRT

continuous renal replacement therapy

CT

cardiac computed tomography

CMR

cardiac magnetic resonance imaging

DAPT

dual antiplatelet therapy

ESC

European Society of Cardiology

LVA

left ventricular aneurysm

LVEDD

left ventricular end diastolic diameter

LVEF

left ventricular ejection fraction

LVT

left ventricular thrombus

MACCEs

major adverse cardiac and cerebrovascular events

NYHA

New York Heart Association

PCI

percutaneous coronary intervention

SAPT

single antiplatelet therapy

STEMI

ST‐elevation myocardial infarction

1. INTRODUCTION

The prevalence of left ventricular thrombus (LVT) is low in the general population but relatively high in patients with ischemic myocardial disease, especially in patients with myocardial infarction. With the advent of percutaneous coronary intervention (PCI), the incidence of LVT has dramatically declined from about 33%–10%. ¹ In terms of LVT therapy, American and European guidelines for the management of myocardial infarction recommended anticoagulation with warfarin for 3‐6 months and discontinuation of warfarin if repeated imaging of left ventricle showed no evidence of LVT after 3–6 months anticoagulation. ² , ³ However, these recommendations were provided mostly based on the studies conducted by physicians where LVT patients were managed with drugs and percutaneous intervention instead of surgical methods. Despite the scarcity of available data, surgical removal of LVT was not proposed because the risk of embolism caused by LVT may not outweigh the surgical risks. ⁴ Nearly all the studies of surgically removing LVT were case reports ⁵ , ⁶ and lacked high‐quality research or evidence to instruct clinical practice of managing LVT patients who underwent left ventricular thrombectomy.

Left ventricular aneurysm (LVA) is commonly recognized as an independent risk factor of LVT. ⁷ Surgical repair is indicated when LVA is associated with refractory angina, heart failure, ventricular arrhythmia, or thromboembolism. ⁸ LVT, if any, is usually resected simultaneously when there is an indication for LVA surgery. It is still controversial whether short‐term prophylactic anticoagulation with warfarin for 3–6 months after LVT resection is necessary. Some consider short‐term prophylactic anticoagulation with warfarin after LVT resection could reduce the risk of LVT recurrence. ⁹ But no relevant clinical data or evidence exists to confirm the effect of short‐term prophylactic anticoagulation with warfarin in preventing LVT recurrence after LVT resection. Moreover, warfarin could increase the risk of gastrointestinal bleeding ¹⁰ as well as patients' burden due to regular monitoring international normalized ratio (INR). Thus, it's necessary to conduct research to assess the effect and necessity of short‐term prophylactic anticoagulation with warfarin for 3–6 months after LVT resection in reducing the risk of LVT recurrence after surgery.

In the present study, two different anti‐thrombotic regimens after LVA surgery and LVT resection simultaneously were retrospectively compared with the aim of determining whether short‐term anticoagulation with warfarin for 3–6 months is necessary after LVT resection.

2. PATIENTS AND METHODS

2.1. Data source and ethical statement

The present study was conducted on the basis of a retrospective analysis of data extracted from the Information Center of Fuwai Hospital, Chinese Academy of Medical Sciences, and Peking Union Medical College, Beijing, China. The study design was approved by the ethics committee of Fuwai Hospital, Beijing, China (Approval No.: 2021‐1644) on March 3, 2022 and the written informed consent was waived by the EC because of the retrospective design.

2.2. Study population

A total of 948 consecutive patients undergoing LVA surgery at Fuwai Hospital, Beijing, China between 2010.1.1 and 2021.4.1 were screened. The inclusion criteria were those who underwent LVA surgery and LVT resection simultaneously and were discharged alive. The exclusion criteria were those who had a clear indication for anticoagulation with warfarin after surgery according to the current clinical guidelines, including undergoing concomitant valvular surgery, pulmonary endarterectomy, complicated with persistent arial fibrillation or postoperative atrial fibrillation, etc. Those with congenital left ventricular diverticulum, left ventricular pseudoaneurysm and hypertrophic obstructive myopathy were also excluded. As patients were divided into two groups based on whether 3–6 months warfarin were administered postoperatively, those receiving new oral anticoagulants after surgery were also excluded. The flow chart of patient enrollment is demonstrated in Figure 1.

Figure 1.

Study flowchart. AF, atrial fibrillation; HOCM, hypertrophic obstructive cardiomyopathy; NOAC, new oral anticoagulant.

2.3. Assessment, surgical management of LVT

Transthoracic echocardiography was applied routinely to screen LVT in all patients before surgery and the suspected LVT diagnosis in the echocardiography was further confirmed via intraoperative exploration. If there was a discrepancy in the diagnosis between preoperative transthoracic echocardiography and intraoperative exploration, the intraoperative exploration should prevail. All the surgical procedures were performed under general anesthesia and cardiopulmonary bypass. After general anesthesia, median sternotomy was performed and cardiopulmonary bypass was established with cannulation of right atrium and ascending aorta. Cold blood cardioplegic solution was administered through the aortic root to arrest the heart. A vertical incision was made at the site of LVA 1‐2 cm parallel to the left anterior descending artery or posterior descending artery. LVT together with the fibrous endocardium was completely resected with scissors and scalpel and intraoperative exploration confirmed no residual LVT. LVA was managed via linear repair (close the left ventricle with mattress suture and running suture of 2‐0 polypropylene) if the size of LVA was relatively small. Otherwise, left ventricle reconstruction might be required via an endoventricular purse‐string suture or patchy plasty. For left ventricle reconstruction, an endoventricular purse‐string was placed and tied to shrink the opening to about 2 cm. And the following suture is similar to linear repair.

2.4. Antithrombotic strategy

In consideration of patients' severe coronary heart disease and the risk of saphenous vein graft occlusion, antiplatelet drugs were routinely prescribed on the first day following surgery when there was no evidence of active bleeding. Antiplatelet drugs in the present study included aspirin, clopidogrel, and ticagrelor. Single antiplatelet therapy (SAPT) denoted antiplatelet therapy with only one antiplatelet drug listed above. Dual antiplatelet therapy (DAPT) denoted the combination of two of these antiplatelet drugs. If DAPT was adopted, it would be transferred to SAPT 1 year after surgery. As there was no clear recommendation in the current guidelines, clinicians in our center held different views on whether short‐term anticoagulation with Vitamin K antagonist warfarin for 3‐6 months should be administered prophylactically after LVA surgery accompanied by LVT resection. Some clinicians tended to prescribe prophylactic warfarin for these patients after surgical LVT removal and others tended not. If warfarin was prescribed prophylactically after surgery, it would be discontinued 3–6 months after surgery and then followed by DAPT before 1 year and SAPT afterward. Considering potential bleeding risks, a relatively lower INR of 2.0–2.5 was targeted in the warfarin therapy according to recommendations provided in the guideline. ²

2.5. Follow‐up protocol

According to the follow‐up protocol in our center, all patients who were discharged from hospital alive were instructed to visit the outpatient clinic 3 months, 6 months, 1 year after surgery and then once every year. Transthoracic echocardiography was routinely performed every time patients visited outpatient clinic. If LVT was suspected by echocardiography, cardiac computed tomography (CT) or cardiac magnetic resonance imaging (CMR) would be applied to confirm the LVT recurrence. As for those who didn't return to Fuwai hospital's outpatient clinic, follow‐up was completed by telephone or Wechat, a popular social communication platform in China to acquire information about patients' clinical condition and the latest echocardiographic results.

2.6. Study endpoints and data analysis

The primary endpoint was the LVT recurrence which was assessed by transthoracic echocardiography and further confirmed by cardiac CT or CMR. The secondary endpoints were major adverse cardiac and cerebrovascular events (MACCEs) and the composite endpoint of MACCEs and LVT recurrence. MACCEs were a composite endpoint comprising all‐cause mortality, nonfatal myocardial infarction, revascularization, and stroke.

Continuous variables were presented as mean ± standard error (if normally distributed) or median (interquartile range, IQR) (if not normally distributed). Categorical variables were presented as numbers and percentage. Continuous variables were compared with the Student's t‐test (if normally distributed) or Mann‐Whitney‐U test (if not normally distributed). Categorical variables were compared with $X^2$ χ ² test or Fisher's exact test. The cumulative incidence of each endpoint was compared with Kaplan–Meier method and the log‐rank test was applied to assess the difference. Cox proportional hazards regression model was utilized to identify the independent risk factors of LVT recurrence. Variables in the multivariable analysis were selected based on the results of univariable analysis and stepwise regression analysis. Variables with p < .05 in the univariable analysis and stepwise regression analysis were entered into the multivariable analysis. p < .05 was considered a statistically significant difference. All analyses were performed with R 4.0.4 (R Foundation for Statistical Computing).

3. RESULTS

A total of 186 patients met the eligibility criteria. Of these 186 patients, 127 patients received antiplatelet therapy only and 59 patients received anticoagulation with warfarin in addition to antiplatelet therapy postoperatively. There were 2 out of 186 patients (1.08%) who had a history of systemic arterial embolization. After PSM, a total of 118 patients were included in the final analysis, among whom 59 received warfarin therapy at discharge and 59 didn't. The baseline characteristics of the two groups were summarized in Table 1. Before PSM, patients with anticoagulation therapy (median, 53 y; IQR, 45.5‐60 y) were younger than their counterparts without anticoagulation therapy (median, 59 years; IQR, 51–63.5 y) (p = .017). After PSM, there was no significant difference between the two groups in baseline characteristics including age, sex, height, weight, smoking history, coronary heart disease family history, hypertension, diabetes mellitus, stroke history, hyperlipidemia, chronic renal dysfunction, ventricular arrhythmia, history of PCI, New York Heart Association (NYHA) grade, time between last myocardial infarction and surgery, preoperative left ventricular ejection fraction (LVEF), and preoperative left ventricular end‐diastolic diameter (LVEDD). As is shown in Table 2, intraoperative data balanced between the two groups, including LVA surgical strategy, concomitant coronary artery bypass grafting (CABG), bypass time, cross‐clamping time, and IABP. No significant difference was recognized in such postoperative data as re‐exploration for bleeding, ventilation duration, in‐hospital stroke, continuous renal replacement therapy (CRRT), hospital stay length, postoperative LVEF, and postoperative LVEDD as is recorded in the last cardiac echocardiography before discharge. Compared with the anticoagulation group, more prescription of DAPT was observed (49.2% vs. 13.6%, p < .001) in the group without anticoagulation therapy. There was no significant difference in other medications at discharge including angiotensin‐converting enzyme inhibitor/angiotensin receptor blocker (ACEI/ARB), diuretics, statins, β blockers, antiarrhythmic drugs, and digitalis.

Table 1.

Baseline characteristics

	Before PSM			After PSM
	Without warfarin (N = 127)	With warfarin (N = 59)	p‐Value	Without warfarin (N = 59)			With warfarin (N = 59)	p‐Value
Age (y)	59 (51,63.5)	53 (45.5,60)	.017	54 (48,60)			53 (45,60)	.603
Sex (male)	110 (86.6%)	56 (94.9%)	.126	57 (96.6%)			56 (94.9%)	1.000
Height (cm)	171 (165,176)	172 (169,176)	.091	172 (169,177)			172 (169,176)	.376
Weight (kg)	72.74 ± 1.02	76.08 ± 1.31	.056	76.68 ± 1.54			76.08 ± 1.31	.770
History of smoking	79 (62.2%)	46 (78.0%)	.050	42 (71.2%)			46 (78.0%)	.526
Family history of CHD	12 (9.5%)	10 (16.9%)	.150	7 (11.9%)			10 (16.9%)	.600
HTN	58 (45.7%)	31 (52.5%)	.474	29 (49.2%)			31 (52.5%)	.854
DM	30 (23.6%)	14 (23.7%)	.999	16 (27.1%)			14 (23.7%)	.833
History of stroke	21 (16.5%)	12 (20.3%)	.670	10 (16.9%)			12 (20.3%)	.813
Hyperlipidemia	75 (59.1%)	37 (62.7%)	.754	36 (61.0%)			37 (62.7%)	1.000
CRD	13 (10.2%)	4 (6.8%)	.626	5 (8.5%)			4 (6.8%)	1.000
Ventricular arrythmia	11 (8.7%)	4 (6.8%)	.884	6 (10.2%)			4 (6.8%)	.903
History of PCI	41 (32.2%)	16 (27.1%)	.589	19 (32.2%)			16 (27.1%)	.687
NYHA classification			.942					.950
I	6 (4.7%)	2 (3.4%)		3 (5.1%)	2 (3.4%)
II	64 (50.4%)	31 (52.5%)		30 (50.8%)	31 (52.5%)
III	46 (36.2%)	22 (37.3%)		21 (35.6%)	22 (37.3%)
IV	11 (8.7%)	4 (6.8%)		5 (8.5%)	4 (6.8%)
Time between last MI and surgery			0.244					.827
< 21 d	7 (5.5%)	5 (8.5%)		5 (8.5%)	5 (8.5%)
21 d–3 m	29 (22.8%)	19 (32.2%)		16 (27.1%)	19 (32.2%)
>3 m	91 (71.7%)	35 (59.3%)		38 (64.4%)	35 (59.3%)
Pre‐op LVEF (%)	41 (37.5,47)	42 (38,49.5)	0.574	41 (38,45)	42 (38,50)			.321
Pre‐op LVEDD (mm)	57 (53,61.5)	55 (51,60)	0.333	57 (54,64)	55 (51,60)			.169

Abbreviations: CHD, coronary heart disease; CRD, chronic renal dysfunction; DM, mellitus diabetes; HTN, hypertension.

Table 2.

Perioperative data and medication at discharge

	Without warfarin (N = 59)	With warfarin (N = 59)	p‐Value
LVA surgical strategy			.455
Linear repair	37 (62.7%)	32 (54.2%)
Nonlinear repair	22 (37.3%)	27 (45.8%)
Concomitant CABG	51 (86.4%)	48 (81.4%)	.616
Ventilation time (h)	17 (15,23.5)	17 (15.5,23.5)	.477
Bypass time(min)	107 (82,143)	99 (68,130)	.330
Cross‐clamping time(min)	71 (55,103)	72 (50,99)	.766
Re‐exploration	3 (5.1%)	1 (1.7%)	.611
Perioperative stroke	1 (1.7%)	0 (0.0%)	1.000
IABP	2 (3.4%)	4 (6.8%)	.675
CRRT	0 (0.0%)	1 (1.7%)	1.000
Hospital stay (d)	21 (15,28)	20 (14,30)	.493
Medication at discharge
DAPT	29 (49.2%)	8 (13.6%)	<.001
ACEI/ARB	4 (6.8%)	2 (3.4%)	.675
Diuretics	54 (91.5%)	49 (83.1%)	.269
Statins	51 (96.4%)	45 (76.3%)	.237
β blockers	52 (88.1%)	52 (88.1%)	1.000
Antiarrhythmic drugs	4 (6.8%)	6 (10.2%)	.741
Digitalis	15 (25.4%)	13 (22.0%)	.829
Post‐op LVEF	45.03 ± 0.82	46.97 ± 1.02	.142
Post‐op LVEDD	5.73 ± 0.72	52.49 ± 0.86	.833

Abbreviations: ACEI/ARB, angiotensin‐converting enzyme inhibitor/angiotensin receptor blocker; CABG, coronary artery bypass grafting; CRRT, continuous renal replacement therapy; DAPT, dual antiplatelet therapy; IABP, intra‐aortic balloon pump; LVEF, left ventricular ejection fraction; LVEDD, left ventricular end‐diastolic diameter.

We confirmed all the patients in the warfarin group received warfarin therapy after discharge by follow‐up. During the median follow‐up of 56.5 months (interquartile range: 32–89 months), 21 out of 118 patients had LVT recurrence and the recurrence rate was 17.8%. There was no systemic embolism resulting from the recurrent LVT during the follow‐up. No serious bleeding events requiring intervention was reported in the two groups. As is illustrated in Figure 2, Kaplan–Meir analysis showed no significant difference in LVT recurrence (log‐rank test, p = .86). During the median follow‐up of 56.5 months, no significant difference was observed in MACCEs (log‐rank test, p = .48) and the composite endpoint of MACCEs and LVT recurrence (log‐rank test, p = .89). (Figure 3A,B)

Figure 2.

Kaplan–Meier analysis of the incidence of LVT recurrence. LVT, left ventricular thrombus.

Figure 3.

Kaplan–Meier analysis of the incidence of the secondary endpoints.

By univariable COX regression analysis, history of PCI (hazard ratio [HR] 3.110, 95% confidence interval [CI] 1.281–7.551, p = .012), LVA surgical strategy of linear repair (HR 12.463, 95%CI 1.665–93.307, p = .014) and postoperative LVEDD (HR 1.060, 95%CI 1.002–1.121, p = .042) were risk factors of LVT recurrence whereas warfarin wasn't (HR 1.086, 95%CI 0.452–2.613, p = .853). On multivariable analysis, history of PCI (HR 2.778, 95%CI 1.087–7.100, p = .033) and LVA surgical strategy of linear repair (HR 8.768, 95%CI 1.139–67.517, p = .037) were independent risk factors of LVT recurrence while warfarin couldn't reduce the risk of LVT recurrence (HR 1.302, 95%CI 0.533–3.184, p = .563) (Table 3).

Table 3.

Univariable and multivariable analyses of predictors of LVT recurrence after surgery

	Univariable analysis			Multivariable analysis
	HR	95%CI	p	HR	95%CI	p
Warfarin	1.086	0.452–2.613	.853	1.302	0.533–3.184	.563
Age	1.003	0.960–1.047	.903
Sex	0.394	0.054–2.905	.361
Weight	1.026	0.982–1.071	.247
History of smoking	0.899	0.345–2.342	.827
History of stroke	0.301	0.040–2.263	.243
Hyperlipidemia	1.486	0.590–3.743	.400
HTN	0.808	0.329–1.986	.642
DM	0.982	0.326–2.958	.974
CRD	2.086	0.607–7.175	.243
History of PCI	3.110	1.281–7.551	.012	2.778	1.087–7.100	.033
Linear repair	12.463	1.665–93.307	.014	8.769	1.139–67.517	.037
Post‐op LVEDD	1.060	1.002–1.121	.042	1.054	0.991–1.121	.095
Post‐op LVEF	0.998	0.941–1.059	.953
DAPT	0.739	0.268–2.036	.558

Abbreviations: CRD, chronic renal dysfunction; DAPT, dual antiplatelet therapy; PCI, percutaneous coronary intervention; LVEF, left ventricular ejection fraction; LVEDD, left ventricular end‐diastolic diameter.

4. DISCUSSION

The major findings of the study are: (a) The rate of LVT recurrence after LVT resection with LVA surgery was 17.8% during the median follow‐up of 56.5 months. (b) The prevalence of systemic arterial embolization before LVT resection was low and systemic arterial embolization after LVT resection was rare. (c) Short‐term prophylactic anticoagulation with warfarin for 3–6 months after LVT resection didn't reduce risks of LVT recurrence and MACCEs. (d) LVA surgical strategy of linear repair and history of PCI were independent risk factors of LVT recurrence after LVT resection.

To our knowledge, this is the largest cohort of patients undergoing LVA surgery accompanied by LVT resection to date. Our research demonstrated similar clinical outcomes including LVT recurrence and MACCEs between the two groups with and without 3‐6 months prophylactic anticoagulation with warfarin after surgery. Simpson also reported a case series of 38 patients undergoing LVT resection together with LVA surgery and found that chronic anticoagulation before surgery didn't affect the frequency of systemic arterial embolization before surgery and the prevalence of LVT at surgery. ¹¹ However, the relatively small sample size weakened the statistical power and the lack of follow‐up data couldn't shed light on the effect of anticoagulation on the prognosis in their study. Lee et. al. also found that anticoagulation with warfarin couldn't reduce the risk of adverse cardiac and cerebral events in patients with LVA, ¹² which is consistent with our results. LVA is a well‐established and main risk factor of LVT for patients after myocardial infarction ⁷ and LVA together with LVT is routinely resected during surgery. In other words, the main risk factor, LVA, is eliminated after surgery. In addition, the scenario of LVT resection is similar to that of LVT resolution by anticoagulation therapy in terms of results. As was illustrated in previous studies, if the patient was at low risk after LVT resolution by anticoagulation therapy, anticoagulation was unnecessary. ¹³ 2012 ESC guideline and 2013 ACCF/AHA guideline for STEMI management provided similar recommendations that anticoagulation could be discontinued if evidence of thrombus is no longer present. ² , ³ Thus, short‐term prophylactic anticoagulation with warfarin for 3‐6 months after LVT resection and LVA surgery may have little impact on LVT occurrence.

The rate of LVT recurrence after LVT resection in our study was higher than that in patients whose LVT was resolved by anticoagulation reported by Leow et al. (17.8% vs. 8.7%). ¹⁴ According to Leow et al., younger age and thrombus protrusion were independent risk factors of LVT recurrence. Similarly, Leow et al. also found that more intense anti‐thrombotic therapy with anticoagulation plus DAPT didn't reduce the risk of LVT recurrence after LVT resolution by anticoagulation. ¹⁴ Another interesting finding in our study was that there was no systemic arterial embolization resulting from recurrent LVT. There were two possible reasons. First, the small absolute number of LVT recurrence from the limited studied sample together with the relatively low risk of recurred LVT falling off resulted in no systemic arterial embolization detected during the follow‐up. Second, according to the current guideline, ³ once LVT recurrence was detected, anticoagulation would be administered, which could lower the risk of systemic arterial embolization. ¹⁵

Different LVA surgical strategies in our institution have been described previously in the literature. ¹⁶ In the present study, we found that compared with left ventricular reconstruction, LVA surgery of linear repair was an independent risk factor of LVT recurrence. As was demonstrated by Zheng, linear repair was associated with less improvement of LVEF than left ventricular reconstruction. ¹⁶ And lower LVEF has been well established as a risk factor of LVT by other studies. ⁷ , ¹⁷ , ¹⁸ In addition, linear repair could reduce the ventricular wall curvature which might affect the local hemodynamics and increase the risk of LVT recurrence. ¹⁹ History of PCI could lead to more distal anastomosis site with compromised run‐off, local inflammation, competitive flow, side‐branch occlusion and distal microembolization, all of which might decrease graft patency and worsen cardiac function afterward. ²⁰ As previously mentioned, lower LVEF is a risk factor of LVT. ⁷ , ¹⁷ , ¹⁸ Thus, a history of PCI might increase LVT recurrence risk by decreasing graft patency and worsening cardiac function characterized by lower LVEF after CABG.

Previous research established dilated LVEDD and reduced LVEF as independent risk factors of LVT following ST‐elevation myocardial infarction (STEMI). ¹⁷ Stasis of blood may explain why enlarged left ventricle or LVA are more prone to ventricular thrombosis. However, postoperative LVEDD and LVEF didn't affect the LVT recurrence after surgery in the multivariable analysis of our study. One possible explanation is that left ventricular remodeling and persisting alteration in cardiac function after LVA reconstruction have been established by previous studies ²¹ , ²² and the postoperative LVEDD and LVEF measured before discharge couldn't predict the long‐term status after discharge exactly.

Several limitations should be recognized in our study. First, our results may be biased by the retrospective, nonrandomized design of the study. Second, the relatively small sample size may prevent us from making a strong conclusion to affect daily clinical practice. However, the amount of surgery for LVA with LVT in itself is extremely small, which makes it difficult to perform a prospective randomized controlled trial to clarify the present issue. And to our knowledge, this is the largest cohort of patients undergoing LVT resection to date. Third, despite we confirmed all the patients in the warfarin group received warfarin therapy after discharge by follow‐up, it was hard for us to obtain detailed INR profile of every patient due to the retrospective design. Lastly, LVT recurrence was assessed by transthoracic echocardiography and further confirmed by CT or CMR. All of these modalities couldn't confirm LVT with 100% sensitivity and specificity as intraoperative exploration. ²³ In addition, the accurate time of LVT recurrence couldn't be acquired. Instead, we used the time that LVT recurrence was detected for the first time in the follow‐up imaging as the LVT recurrence time, which might bias the result.

5. CONCLUSION

In conclusion, short‐term prophylactic anticoagulation for 3–6 months with warfarin may be unnecessary with no benefit in terms of LVT recurrence and MACCEs. Patients with a history of PCI and LVA surgical strategy of linear repair should be closely followed up for LVT recurrence after discharge. In addition, as new oral anticoagulation (NOAC) has presented a non‐inferior effect on the treatment of LVT after myocardial infarction, ²⁴ further research may be warranted to investigate the role of NOAC in patients undergoing LVT resection.

AUTHOR CONTRIBUTIONS

Wei Feng: Conceptualization, methodology, and supervision. Shicheng Zhang: Data curation, data analysis, and writing original draft. Siyuan Huang: Follow‐up patients and data collection. Xieraili Tiemuerniyazi: Follow‐up patients and Data collection. Yangwu Song: Follow‐up patients and data collection.

CONFLICTS OF INTEREST

The authors declare no conflicts of interest.

INFORMED CONSENT STATEMENT

The written informed consent was waived by the local ethic committee of Fuwai Hospital, Beijing, China because of the retrospective design.

INSTITUTIONAL REVIEW BOARD NUMBER

2021‐1644 (date: Mar 3rd, 2022)

Zhang S, Huang S, Tiemuerniyazi X, Song Y, Feng W. Is 3–6 months anticoagulation with warfarin necessary after left ventricular thrombectomy with left ventricular aneurysm surgery? J Card Surg. 2022;37:5103‐5110. 10.1111/jocs.17215