Efficacy and safety of anticoagulant and antiplatelet therapies in the medical management of carotid free-floating thrombus: A systematic review

Abstract

Background

The optimal treatment for carotid free-floating thrombus (CFFT) remains uncertain due to limited evidence, with no randomized clinical trials and scarce guidelines, such as ESVS 2023, favoring conservative management. Anticoagulation (ACT) and antiplatelet (APT) therapies are emerging as promising alternatives to high-risk surgical interventions. This systematic review aimed to evaluate the safety and efficacy of ACT and APT therapies for CFFT.

Methods

A systematic search was performed across PubMed, Embase, Web of Science, and Cochrane databases. Safety and efficacy endpoints were assessed. A two-sample t-test compared baseline characteristics between groups, and a Chi-square test evaluated differences in categorical variables. Statistical significance was set at p < 0.05. Data were analyzed using R 4.3.0 with the meta package v.7.0-0.

Results

Four studies met the inclusion criteria, involving 170 patients diagnosed with CFFT. The APT group included 96 patients (mean age 55.35 ± 13.52 years; 56.25% male), and the ACT group included 74 patients (mean age 58.57 ± 14.28 years; 51.35% male). Thrombus regression was slightly lower in APT (42%) compared to ACT (48%). Both groups showed similar rates of residual stenosis. Antiplatelet had fewer ischemic events within 30 days (none vs. 4% in ACT) and lower intracranial hemorrhage rates (3.3% vs. 5.4% in ACT) but higher mortality (6.3% vs. none in ACT).

Conclusion

Both ACT and APT are effective for managing CFFT, each with distinct efficacy and safety profiles. However, randomized trials are necessary to better assess these therapies in CFFT management.

Introduction

Carotid free-floating thrombus (CFFT) consists of an elongated, wall-attached, nonocclusive thrombus with circumferential blood flow at its distal section, moving in accordance with the cardiac contractions and commonly named by its doughnut-like appearance in imaging.^1,2 This sort of thrombus is more often caused by atherosclerotic disease, with its prognosis worsening as the atherosclerosis severity increases. ³ It is also caused by prothrombotic states (e.g., trauma, inflammation, and pregnancy), vasculitis, and the currently discovered Coronavirus 2019 (COVID-19).^1,4–6

The rise in CFFT reporting cases is largely due to the increased accessibility of advanced imaging tools such as digital subtraction angiography, carotid ultrasound, and computed tomography angiography (CTA).^7,8 Associated with that comes the possibility of earlier diagnosis, which contributes to minimizing the silent but severe complications related to the pathology in question, such as watershed area ischemia and transient ischemic attacks (TIAs), which can lead to subsequent death by stroke.^7,9 An early diagnosis, in turn, allows for the earlier management of the condition.

Nowadays, evidence pointing to the best treatment modality for CFFT remains scarce, with no randomized clinical trials (RCTs) and few guidelines, such as ESVS 2023, which strongly indicates the conservative treatment, proposing management.^10,11 However, medical care with anticoagulation (ACT) and antiplatelet (APT) therapies are showing promising outcomes as an alternative for the high-morbidity risked surgical approaches (e.g., endarterectomy, stenting, and revascularization), which are usually indicated for severe artery stenosis cases.^10,12,13

However, the safety and efficacy of the ACT and APT therapies remain uncertain for CFFT management. Thus, in this systematic review, we aimed to assess the safety and efficacy of both ACT and APT for treating CFFT by synthesizing the existing data. Through this study, we intend to provide insights and contribute to filling gaps while guiding further and stricter research.

Methods

This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis statement guideline.

Eligibility criteria

In this systematic review, eligible studies met the following criteria: (1) original studies enrolling patients with CFFT treated with APT and ACT; (2) reporting on safety and/or efficacy outcomes; and (3) comprising four or more patients. We excluded studies with (1) population overlap (e.g., overlapping institutions and recruitment periods); (2) conference abstracts; and (3) studies that reported patients receiving simultaneous ACT and APT therapies.

Search strategy, study selection, and data extraction

We conducted a comprehensive search on Medline, Embase, Web of Science, and Cochrane from inception to 17 June 2024 with the following search strategy: (anticoagulant* OR heparin OR warfarin OR VKA OR vitamin K antagonist* OR DOAC OR direct oral anticoagulant* OR “thrombin inhibitor” OR apixaban OR rivaroxaban OR antiplatelet* OR aspirin OR ASA OR “Dual antiplatelet” OR “glycoprotein IIb/IIIa” OR ADP OR clopidogrel OR tirofiban OR P2Y12 OR prasugrel OR ticagrelor OR cangrelor OR management OR treatment OR intervention) AND (carotid OR cervico-cephalic arteries OR ICA OR internal carotid arter*) AND (floating thrombus OR free-floating thrombus OR intraluminal thrombus OR intraluminal thrombus OR non-occlusive thrombus OR intraluminal carotid artery thrombi). Two authors (R.C. and S.D.) selected and assessed the studies for inclusion, and two authors (M.V. and F.V.R.) independently extracted the prespecified baseline characteristics in the data extraction process. Any conflict was resolved by a third senior author (M.Y.F.).

The baseline characteristics extracted include (1) authors and year; (2) study design; (3) number of patients in treatment with ACT or APT therapy; (4) regimen details of the intervention; and (5) main relevant patient characteristics, as clinical presentation.

Quality assessment

Two independent authors (C.F. and M.V.) performed the quality assessment, and any conflict was solved by a third and more experienced investigator (M.Y.F.). The Risk Of Bias In Non-randomized Studies of Interventions ¹⁴ tool was used to evaluate the risk of bias across seven domains. Articles were categorized as low, moderate, or serious risks according to their reporting follow-up and outcomes.

Endpoints and definitions

Endpoints were: (1) complete thrombus regression, defined as total resolution of CFFT on the first repeat vascular imaging; (2) time to complete resolution; (3) residual stenosis, defined as remaining narrowing of the blood vessel after treatment; (4) clinical outcome, expressed by final modified Rankin Scale or classified as improved or worsened in comparison to baseline status; (5) type of surgical intervention; (6) 30-day recurrent stroke or TIA; (7) symptomatic intracranial hemorrhage (ICH); (8) any ICH; and (9) 90-day intervention-related mortality. These efficacy and safety outcomes were applied to both APT and ACT groups in the included studies.

Statistical analysis

Categorical and numerical data were summarized as counts and measures of central tendency, respectively. A two-sample t-test was used to assess differences in mean values of baseline features of patients between groups. A Chi-square test was used to test the differences between categorical variables. Statistical significance was set at p values <0.05. Statistical analysis and data visualization were performed using R 4.3.0 (R Core Team, 2023) with the meta package v.7.0-0. ¹⁵

Results

Search results

A total of 1155 records were identified from Medline (n = 426), Embase (n = 465), Cochrane (n = 7), Web of Science (n = 247), and additional records from snowballing (n = 10). ¹⁶ After removing 536 duplicates, 619 records remained for title and abstract screening. Among these, 73 articles were selected for full-text evaluation. Ultimately, four studies satisfied our inclusion criteria. This search process is illustrated in Figure 1.

Figure 1.

PRISMA flow diagram.

Patient baseline characteristics

A total of 179 patients diagnosed with CFFT under CTA or CT were included. Two distinct subgroups were identified depending on the medical management after the diagnosis of CFFT (Table 1), characterizing APT and ACT therapy subgroups. Patients in APT regimens were under either dual antiplatelet therapy with aspirin or clopidogrel/ticagrelor, while patients in ACT regimens were under unfractionated heparin, low-molecular-weight heparin, warfarin, or direct oral anticoagulants.

Table 1.

Baseline clinical features of patients.

Baseline clinical features	Antiplatelet therapy (n = 96)	Anticoagulant Therapy (n = 74)	p-value
Sex—n (%)
Male	54 (56.25)	38 (51.35)	0.992
Female	42 (43.75)	36 (46.65)	0.127
Age (mean ± SD)—years			0.795
Ethnicity—n (%)	55.35 ± 13.52	58.57 ± 14.28	0.001
African American	42 (43.75)	13 (17.57)
White	30 (31.25)	7 (9.46)
Hispanic and other	10 (10.42)	2 (2.70)
NR	14 (14.58)	52 (70.27)
Comorbidities—n (%)*
Hypertension	53 (55.21)	40 (54.05)
Stroke/TIA	18 (18.75)	2 (2.70)
Diabetes mellitus	27 (28.12)	24 (32.43)
Hyperlipidemia	16 (16.67)	4 (5.41)
Smoking	27 (28.12)	7 (9.46)
Cancer	8 (8.33)	3 (4.05)
Atrial fibrillation	5 (5.21)	10 (13.51)
Recent COVID infection	5 (5.21)	NR
Prior medication use—n (%)
None	NR	42 (56.76)
Antiaggregation therapy	NR	8 (10.81)
Anticoagulation	NR	2 (2.70)

N: number; NR: not reported; TIA: transient ischemic attack.

*Percentages exceed 100% due to the concurrent presence of comorbidities.

The group receiving APT therapy encompassed 96 patients (56.25% male) with a mean age of 55.35 ± 13.52 years. In this subset, patients were predominantly of African American ethnicity (43.75%) and had hypertension (55.21%), diabetes mellitus (28.12%), smoking history (28.12%), past history of stroke or TIA (18.75%), and hyperlipidemia (16.67%).

In the group of 74 patients who received ACT therapy, 51.35% were male, with a mean age of 58.57 ± 14.28 years. Among the reported ethnicities, 17.57% of patients were African American. Medical comorbidities included hypertension (54.05%), diabetes mellitus (32.43%), atrial fibrillation (13.51%), past history of stroke or TIA (2.7%), and hyperlipidemia (5.41%). A comprehensive list of medical comorbidities is shown in Table 1. In this group, prior medication use included ACT (2.7%) and APT therapy (10.81%), whereas 56.76% of patients reported no use of blood thinners.

The results from the chi-square test of independence found no significant statistical difference between proportions for the variables of sex, age, or ethnicity between the two groups. On the other hand, medical comorbidities differed among groups (χ² (df: 6, N: 244) = 22.46, p = 0.001).

Clinical presentation

With respect to initial presentation, all patients from both subgroups presented with an acute symptomatic stroke or TIA. APT admission median NIHSS ranged from 8 to 10, and admission ASPECTS from 8 to 9, while ACT admission NIHSS ranged from 2 to 7 and ASPECTS of 9. Details on the admission NIHSS and ASPECTS are presented in Table 2.

Table 2.

Clinical presentation and lesion characteristics.

Study, year	Drug regimen (n)	Clinical presentation (n)	Admission NIHSS [median (IQR)]	Admission ASPECTS [median (IQR)]	Thrombotic etiology (n)	Occlusion site (n)	Stenosis degree (n)	Treatment upon discharge (n)	Length of hospital stay [median (IQR)]
Antiplatelet regimen
Aboul-Nour, et al. 2024	(Initial treatment assignment) Dual or single APT—aspirin, clopidogrel and ticagrelor (37)	Acute ischemic stroke or TIA (37)	10 (3.5–17)	8 (6–10)	Atherosclerosis (23) Web (3) Hypercoagulability (3) Embolic stroke of undetermined source (6) Left ventricular thrombus (1)	Carotid artery (37) - Intracranial (12)	0–49% (12) ≥ 50% (20) Occluded (5)	APT (32) ACT (5)	11 (4–19.5)
Aboul-Nour et al., 2024	(Discharge regimen) Dual or single APT—aspirin, clopidogrel and ticagrelor (45)	Acute ischemic stroke or TIA (45)	9 (3–17.5)	9 (6–10)	Atherosclerosis (30) Web (4) hypercoagulability (4) Embolic stroke of undetermined source (6)	Carotid artery (45) - Intracranial (13)	0–49% (15) ≥ 50% (25 Occluded (5)	APT (45)	9 (5–15.5)
El Harake et al., 2023	Single APT—aspirin or clopidogrel (9)	Symptomatic ischemic stroke or TIA (9)	NA	NA	NA	Carotid artery (9)	NA	Single APT (9)	NA
Vassileva et al., 2014	Single APT—clopidogrel 75 mg (5)	Acute ischemic stroke (5)	8 (7–22)	NA	NA	Internal carotid artery (5)	NA	NA	NA
Anticoagulant regimen
Aboul-Nour et al., 2024	Direct oral anticoagulants, warfarin, heparin or low molecular weith heparin (22)	Acute ischemic stroke or TIA (22)	7 (1.75–15.25)	9 (8–10)	Atherosclerosis (11) Web (1) Dissection (1) Hypercoagulability (4) Embolic stroke of undetermined source (4) Left ventricular thrombus (1)	Carotid artery (22) - Intracranial (2)	0–49% (10) ≥ 50% (12)	NA	7 (4–13.5)
Onalan et al., 2024	UFH (23) LMWH (29)	Stroke (45) TIA (7)	2 (0–22)	NA	Atherosclerosis (17) Atrial Fibrillation (17) Embolic stroke of undetermined origin (8) Dissection (7) Hypercoagulability (2)	-Extracranial carotid artery (26) -Intracranial carotid artery (10) -Basilar artery (8) -Middle cerebral artery (7)	NA	APT, ACT, or stenting antiaggregant treatment according to the etiology. If idiopathic, patients were followed up with antiaggregant treatment.	NA

ACT: anticoagulation therapy; APT: antiplatelet therapy; TIA: transient ischemic attack.

Considering all patients, the etiology of thrombus included atherosclerosis (50.48%), embolic stroke of undetermined source (11.43%), hypercoagulable state (6.67%), and carotid web (6.67%). Amongst subgroups, patients currently receiving APT had a higher proportion of atherosclerosis as etiology than current ACT (66% vs. 50%) and more reports of carotid webs (9% vs. 1%). However, the subgroup on ACT had 17 strokes related to atrial fibrillation and eight dissections as culprits. Hypercoagulability was relatively equal amongst APT and ACT groups (9% vs. 10%).

When looking at the occlusion site, possible sites included the carotid artery or intracranial vessels. Patients on APT had the obstruction reported at the carotid level in 79% of patients versus 64% for patients on ACT. For patients with poststroke management discharge available data, 94% of patients initially prescribed APT were adherent to APT. In comparison, the remaining 6% were initially prescribed ACT at discharge but switched to the current APT. The degree of stenosis was graded into three tiers depending on the percentage of occlusion: 0–49% (25.71%), >50% (42.86%), and 100% (9.52%).

Efficacy and safety outcomes

Regarding thrombus regression at the latest follow-up available, patients receiving APT exhibited slightly lower rates when compared to ACT (42% vs. 48%). However, the timing of thrombus regression evaluation was not available for some studies (Onalan et al. 2024). Across all studies, follow-up over time showed an increase in rates of proportional thrombus regression ([patients with thrombus regression at last follow-up]/[patients with thrombus regression at initial follow-up]), ranging from 62% to 600% more patients exhibiting complete regression. Similarly, time to complete resolution showed that some thrombi may take up to 115 days after therapy to resolve. Residual stenosis was noted in both groups, with similar proportions.

The incidence of complications was relatively low in both therapy groups. In the APT group, no ischemic event was reported within 30 days, while in the ACT group, the incidence of stroke or TIA within 30 days was 4%. Neither group reported any events of symptomatic ICH. The occurrence of any ICH was 3.3% in the APT group, while 5.4% in the ACT group. Finally, the in-hospital 90-day intervention-related mortality rate was 6.3% in patients receiving APT, compared with none in patients receiving ACT. Further details on clinical outcomes and length of follow-up are reported in Table 3.

Table 3.

Efficacy and safety outcomes of antiplatelet and anticoagulation therapy.

Study, year	Drug regimen (n)	Complete thrombus regression (n)	Time to complete resolution (days)	Residual stenosis (n)	Clinical outcomes (n)	Surgical intervention (n)	30-day recurrent stroke/ TIA	Symptomatic ICH	Any ICH	90-day intervention-related mortality	Length of follow-up (months)
Antiplatelet regimen
Aboul-Nour, et al. 2024	(Initial treatment assignment) Dual or single AP—aspirin, clopidogrel and ticagrelor (37)	At 30 days (6/19) At latest follow-up (14/28)	At 30 days 5 (4–11) At latest follow-up 47 (9–85.75)	0–49% (13) ≥ 50% (10) Occlusion (3) Normal (2)	NA	Stent (1) and TCAR (1)	None	None	1/37	In-hospital (3/37)	NA
Aboul-Nour et al., 2024	(Discharge regimen) Dual or single AP—aspirin, clopidogrel and ticagrelor (45)	At 30 days (9) At latest follow-up (18)	At 30 days 5 (2–11.5) At latest follow-up 39 (9–115.5)	0–49% (18) ≥ 50% (11) Occlusion (3) Normal (3)	NA	NA	None	None	2/45	In-hospital (3/45)	NA
El Harake et al., 2023	Single AP—aspirin or clopidogrel (9)	At 7 days (4)	NA	Stenosis > 60% (1) Carotid occlusion (3) Downstream occlusion (2)	Neurological recurrence (1) Loss of follow up (1)	None	None	None	None	None (out of eight, one loss of follow-up)	21.6 ± 12.9*
Vassileva et al., 2014	Single AP—clopidogrel 75 mg (5)	At 30 days (1) At 3 months (3) At 6 months (5)	At 30 days, complete regression in one case and significantly diminished in the other four. At 6 months, ICA appeared normal in all cases.	None	At discharge 4-Improved 1-No change 1 year mRS 0 (3) mRS 2 (1) mRS 3 (1)	NA	None	None	NA	None	28.8 (12–48)
Anticoagulant regimen
Aboul-Nour et al., 2024	Direct oral anticoagulants, warfarin, heparin or low molecular weith heparin (22)	At 30 days (8) At latest follow-up (13)	At 30 days 10 (6.5–18.5) At latest follow-up 19 (9–165.5)	0–49% (13) > 50% (4) Occlusion (0) Normal (3)	NA	NA	1/22	NA	4/22	None (in-hospital)	NA
Onalan et al., 2024	UFH (23) LMWH (29)	Timing NA (23)	Thrombus resolved completely after 7 days in 24 patients, after 14 in 26 patients, and after 21 days in 2 patients.	None	At 3 months Good functional outcome (mRS 0–2) (49) Median mRS score 0 (0–5)	NA	2/52	None	None	None	3

ICH: intracranial hemorrhage; mRS: modified Rankin Scale; TCAR: transcarotid artery revascularization; TIA: transient ischemic attack.

*Mean ± SD.

Risk of bias assessment

The risk of bias was judged moderate across all four included studies. Figure 2 represents the individual assessments of the included studies according to each domain. The summary plot in Figure 3 displays the overall distribution of the different types of bias.

Figure 2.

Traffic light plot.

Figure 3.

Summary plot.

Discussion

There is no agreement in the literature regarding the best management of CFFT among ACTs, single or dual APT agents, or a combination of APT and ACT therapies. The optimal initiation of any combined or ACT treatment also remains uncertain, and the duration of the treatment is highly variable. ¹⁷ This systematic review evaluated the safety and efficacy of APT and ACT therapies for treating CFFT. Patients receiving APT demonstrated marginally lower rates of thrombus regression compared to those on ACT, 42% versus 48%. Both groups exhibited a low complication rate. No ischemic events occurred within the first 30 days in the APT group, whereas the ACT group had a 4% incidence of stroke or TIA during the same period. The incidence of ICH was 3.3% in the APT group and 5.4% in the ACT group. Additionally, the in-hospital mortality rate related to the intervention at 90 days was 6.3% for the APT group, with no deaths reported in the ACT group.

Both medical and surgical treatment options have been employed, with neither notably outperforming the other in terms of efficacy. ¹ The data in the literature does not show significant differences in clinical and radiological results between these strategies. ¹⁰ However, the evolution of CFFT management may offer advantages over time. The traditional practice of immediate surgical intervention in symptomatic CFFT cases, as originally described, may need to be reevaluated. Emergency carotid endarterectomy and thrombectomy for stroke patients due to acute internal carotid artery thrombosis present a 40% risk of perioperative cerebral events. ¹⁸ In cases where the thrombus extends distally, urgent open revascularization carries an elevated risk of clot displacement, potentially leading to cerebral embolization during exposure of the carotid bifurcation. ¹⁹

The study by Papadoulas et al. ¹⁹ demonstrates that carotid endarterectomy should be performed when severe atherosclerotic stenosis persists after resolution with medical treatment. The study by Tolaymat et al. ²⁰ demonstrated that thromboendarterectomy can be performed, regardless of the degree of carotid stenosis, when certain factors are present, such as reduced thrombus in the carotid bulb or the emergence of the underlying plaque. In fact, complete resolution of CFFT without neurological progression was observed in more than 80% of patients treated medically. ¹ Therefore, pharmacological therapies, such as anticoagulants and APT agents, warrant further investigation, which is the very aim of the present study.

In our review, the included studies showed efficacy in clinical results. Only the study by El Harake et al. ¹⁰ reported neurological recurrence in one patient using the APT regimen. Considering the safety of these medications, this systematic review reveals positive results, with no deaths reported, either with the use of APT agents or ACTs. Overall, both therapeutic approaches exhibited safety and efficacy in managing CFFTs. Furthermore, the administration of these therapies should be tailored based on the patient's individual risk factors.

Regarding the guidelines, the American Heart Association and the European Stroke Organization have not established a specific strategy for the treatment of CFFT.^21–23 Therapeutic ACT is recommended only by the European Society for Vascular Surgery guidelines for patients with recent carotid territory symptoms and recurrent symptoms. Thrombolytic therapy is not recommended for patients with CFFT.^11,23,24 Antithrombotic treatment is generally the preferred modality, with a success rate that can reach up to 86%.^17,25

For acute phase treatment, ACT is widely used due to its focus on the thrombotic component of the disease.^7,21 A study indicated the safety of ACT in patients with intraluminal thrombi, observing no recurrent events in the 14 patients treated exclusively with this method. ²¹ However, Nour et al. ²⁶ did not identify significant differences between patients treated with ACT and those treated with APT. Additionally, patients allocated to APT were more likely to be discharged with the designated treatment compared to those allocated to ACT.

Regarding APT, they have been primarily used to address the most frequently implicated etiology, such as atherosclerosis. ⁷ Previous studies have demonstrated that underlying atherosclerosis and ulcerated plaques, as well as the presence or absence of intracranial stenosis, are considered factors that favor treatment with APT. Other factors included the presence or absence of microbleeds detected by MRI. ²⁷

Although the literature presents reviews on this topic, as the studies by Bhatti et al. ¹ and Fridman et al., ⁷ this systematic review is the first to include exclusively studies that administered APT and ACT in separate groups, without mixed treatment to provide a clear visualization and comparison of both approaches. This resulted in more strict inclusion criteria and a careful exclusion of treatment groups that included ACT and APT simultaneously, such as El Harake et al., ¹⁰ Schartz et al., ²⁸ and Singh et al., ²⁹ which did not evaluate ACT exclusively.

This study contributes significantly to the emerging literature on this topic by describing the applicability of APT and ACT, providing clear insights that can inform clinical practice based on individual patient risk factors. Antiplatelet demonstrated fewer ischemic (none vs. 4% in ACT) and hemorrhagic events (3.3% vs. 5.4% in ACT), though a higher mortality rate (6.3% vs. none in ACT). Therefore, based on our results and previous studies in the literature, ACT therapy should be used for acute phase treatment due to its focus on the thrombotic component of the disease and later replaced by APT therapy. Additionally, this systematic review highlights the limited number of studies assessing the outcomes of these therapies separately, underscoring the need for further research to fill this critical gap based on a more careful patient selection.

This study has limitations. The included studies’ observational and retrospective nature limits our findings’ generalization. Furthermore, our results and interpretation should be analyzed without available long-term outcomes of these therapies, particularly regarding the durability of treatment effects and late complications.

Conclusion

This systematic review identified that the available literature comprising only observational evidence demonstrated that APT and ACT therapies are effective options for managing CFFT, each with distinct efficacy and safety profiles. Although the available data suggest that these therapeutic approaches are efficacious and reliable, the lack of RCTs limits the ability to establish a clear cause and effect. In this context, our study highlights the need for RCTs to rigorously evaluate the efficacy and safety of APT versus ACT therapies in managing CFFT.