Vascular clips for primary closure in classic carotid endarterectomy: A 5-year exploratory retrospective cohort study

Abstract

Objective

To evaluate the intraoperative efficiency and long-term outcomes of clip-assisted primary closure versus conventional suturing in classic carotid endarterectomy (CEA).

Methods

This single-center retrospective cohort study included 46 classic CEA procedures in 45 patients treated between 2013 and 2019. Procedures were divided into a clip group (n = 15) and a suture group (n = 31). The primary endpoints were carotid clamping time and intraoperative blood loss. Secondary endpoints included operative time, postoperative drainage volume, perioperative stroke, long-term stroke incidence, and all-cause mortality.

Results

The clip group had a significantly shorter carotid clamping time than the suture group (12.67 ± 5.74 min vs. 18.26 ± 7.84 min, P = 0.048) and significantly lower intraoperative blood loss (38.00 ± 18.97 mL vs. 56.77 ± 23.01 mL, P = 0.010). No statistically significant between-group differences were observed in total operative time, postoperative drainage volume, perioperative stroke, long-term stroke incidence, or all-cause mortality.

Conclusions

In this exploratory retrospective cohort study, clip-assisted primary closure during classic CEA was associated with shorter carotid clamping time and lower intraoperative blood loss than conventional suturing. No signal of increased perioperative or long-term adverse outcomes was detected; however, the study was not powered to establish equivalence or non-inferiority for rare safety endpoints. These findings support the technical feasibility of clip-assisted primary closure in selected patients, but larger prospective studies are needed to further evaluate its safety and long-term effectiveness.

Keywords

carotid stenosis surgical anastomosis surgical blood loss postoperative complications vascular surgical procedures

Highlights

1. Vascular clips significantly shortened carotid clamping time during classic carotid endarterectomy.

2. Intraoperative blood loss was significantly lower in the clip group than in the suture group.

3. No signal of increased long-term stroke or loss of patency was detected in the clip group during follow-up.

4. Clip-assisted primary closure was technically feasible in selected patients undergoing classic carotid endarterectomy.

5. This is the first long-term exploratory cohort study evaluating clip-assisted primary closure in carotid endarterectomy.

Introduction

Stroke remains a leading cause of mortality and disability worldwide. In China, stroke rose from the third to the first leading cause of death between 1990 and 2017, with an average mortality rate of 149 per 100,000 people in 2017.¹ Carotid endarterectomy (CEA) has long been established as the gold standard for treating carotid artery stenosis, supported by large-scale randomized trials such as NASCET, ECST, and ACAS.² While the surgical technique is now mature, efforts continue to refine operative efficiency and minimize perioperative complications.

A key step in CEA is vascular closure, traditionally performed with needle-and-thread suturing. However, this method penetrates the vessel wall, potentially damaging the intima and introducing foreign material into the lumen, which may trigger thrombosis, intimal hyperplasia, or restenosis.^3–5 Vascular clips have emerged as a promising alternative, offering advantages such as reduced vessel trauma, elimination of intraluminal foreign bodies, and shorter anastomosis time.^6–12

Currently, these clips are primarily used in dialysis access creation and peripheral vascular bypass procedures. Unlike those contexts, CEA inherently involves intimal disruption, potentially diminishing the theoretical advantage of clip-based closure techniques. Moreover, earlier attempts to use clips in CEA were limited by reports of severe complications,¹³ leading to ongoing controversy regarding their suitability in this setting.^6,7,13

Based on our prior surgical experience with vascular clips in arterial and venous procedures—including thrombectomy reconstruction, AV graft and fistula creation, and peripheral bypass—we cautiously extended their application to selected cases of carotid endarterectomy (CEA).

This exploratory retrospective study analyzed a carefully selected cohort of classic carotid endarterectomy (CEA) patients treated at our institution between 2013 and 2019. All procedures were performed by a single experienced surgeon, with consistent surgical indications and technique selection to minimize bias. We compared vascular clips and traditional sutures in terms of clamping time, intraoperative blood loss, postoperative complications, and long-term outcomes, including 5-year follow-up data. The goal was to assess the feasibility of clip-assisted primary closure in selected cases of classic carotid endarterectomy.

Methods

Research subjects

This retrospective study included all patients who underwent classic carotid endarterectomy (CEA) at the Department of Interventional and Vascular Surgery, Peking University First Hospital between January 2013 and November 2019. This study was conducted in accordance with the ethical principles of the Declaration of Helsinki of 1975, as revised in 2024. All patient details were de-identified and anonymized prior to analysis and reporting. Patients were identified through the hospital’s electronic medical record system.

To ensure procedural consistency and reduce confounding from surgical technique variation, only patients who received conventional classic CEA (cCEA) were included. Because the AnastoClip GC system is intended for direct arteriotomy closure, the control group was intentionally matched using the same closure strategy (primary closure) to isolate the effect of the closure device. Patients undergoing patch angioplasty (pCEA) or eversion CEA (rCEA) were excluded, as our previous clinical experience demonstrated that the GC series of vascular clips was incompatible with the prosthetic patch materials commonly used in our institution, leading to poor mechanical fit. In our pilot experience with the synthetic patches routinely used in our center (e.g., B. Braun), deployment of GC-L clips could cause patch perforation and persistent oozing/bleeding at the closure line; therefore, clip application was not pursued in patch CEA in this cohort. According to the manufacturer’s Instructions for Use, the AnastoClip GC is intended for everting approximation when tissue penetration is desired; therefore, we applied clips only when symmetric eversion and adequate tissue thickness could be achieved. Furthermore, the VCS clip system—used in some external studies—was associated with a risk of anastomotic disruption and bleeding in CEA when used inappropriately, as reported in prior literature.¹³ In light of these concerns and to maximize patient safety, we cautiously limited the clip application to cCEA procedures only in this exploratory study.

Patients were grouped based on the anastomotic technique: those receiving vascular clips (experimental group) and those receiving conventional non-absorbable sutures (control group). Assignment was influenced by preoperative consultation, with medical insurance coverage of the vascular clip system being a key consideration—since uninsured clips posed a significant financial burden, many patients and families opted for sutures. Despite this, baseline characteristics between the two groups showed no statistically significant differences, as verified by comparative analysis.

All surgeries were elective and performed by the same surgical team, with Professor Zhang Xiansheng as the sole operating surgeon, thereby controlling for operator-related variability.

Surgical indications were determined in accordance with the 2011 clinical practice guideline on extracranial carotid artery disease management.¹⁴ Indications included: symptomatic carotid stenosis ≥70% on non-invasive imaging or ≥50% on catheter-based angiography; asymptomatic carotid stenosis ≥70% on non-invasive imaging or ≥60% on angiography; or unstable plaque features despite stenosis <70%. Symptomatic stenosis was defined as ipsilateral transient ischemic attack (TIA) or ischemic stroke occurring within the previous six months. Patients with uncontrolled systemic disease, abnormal routine preoperative assessments, or surgical contraindications were excluded. All procedures were elective; no emergency procedures were included.

Research materials

Nonabsorbable sutures used in this study included the Prolene® series (Ethicon, Johnson & Johnson, UK) and CV-6 sutures (GORE-TEX®, W. L. Gore & Associates, USA). The vascular clips used in this study were all AnastoClip GC® Closure System devices (LeMaitre Vascular, USA), with the GC-L model exclusively used in this study. Clip specifications were selected based on individual vascular anatomy.

Preoperative preparation

All patients underwent standard preoperative evaluations, including complete blood count, liver and renal function tests, serum electrolytes, myocardial biomarkers, glucose and lipid profiles, coagulation parameters, infection screening, urinalysis, and electrocardiogram to exclude surgical contraindications. Additional tests were performed as clinically indicated for comorbid conditions (e.g., thyroid function tests in patients with hypothyroidism). All patients had preoperative carotid imaging completed either at outside institutions or in our outpatient clinic.

For those with chronic antiplatelet therapy, medications were continued perioperatively at the original regimen. Hypertensive and diabetic patients had blood pressure and glucose levels monitored during hospitalization, with pharmacologic adjustments as needed.

Prior to surgery, all patients and/or their families received standardized counseling regarding the disease condition, procedural risks and benefits, postoperative considerations, and the potential differences between traditional sutures and vascular clips in terms of operative time, complication risk, and cost. Final selection of anastomosis method (clip vs. suture) was based on patient preference after informed consent. To minimize bias, only patients undergoing classic carotid endarterectomy by the same surgeon were included in this analysis.

Surgical procedure

All procedures were performed under general anesthesia with patients in the supine position and the head turned 45° to the contralateral side. A longitudinal incision was made along the anterior border of the sternocleidomastoid muscle to expose the common, internal, and external carotid arteries. A carotid sinus block was performed with 0.22% lidocaine, and systemic heparinization (1 mg/kg) was administered 5 minutes prior to arterial clamping. Following standard endarterectomy, the incised anterior walls of the common and internal carotid arteries were reconstructed either by continuous suturing (Prolene or CV-6 suture) or vascular clips, depending on preoperative patient/family choice. Anastomotic integrity was ensured by standard hemostasis techniques. A closed-suction drain was placed before layered closure. Representative intraoperative images of the clip-assisted procedure are provided in Figure 1.

Figure 1.

Intraoperative images of carotid endarterectomy using vascular clips.

No shunts or patches were used in any of the procedures. All patients underwent classic carotid endarterectomy without patch angioplasty or eversion techniques.

Intraoperative time was defined as the interval from skin incision to completion of skin closure. Carotid clamping time was defined as the interval from carotid cross-clamping to unclamping immediately after completion of carotid arteriotomy closure. Intraoperative blood loss was recorded based on the surgical report; if not explicitly documented, anesthesia records were used as the reference.

Postoperative management

Continuous electrocardiographic and blood pressure monitoring was performed postoperatively. For patients with a history of long-term oral antihypertensive therapy, preoperative regimens were resumed to maintain stable hemodynamics, ensuring adequate perfusion of vital organs such as the heart and kidneys. Intravenous mannitol and hydrocortisone were selectively administered to reduce intracranial pressure and provide neuroprotection. Antiplatelet therapy was initiated on the first postoperative day. Patients already on long-term antiplatelet medication resumed their original regimen. Others received 100 mg of aspirin orally once daily. Wound drainage volume and characteristics were closely monitored. The drainage tube was removed when the 24-hour output was less than 10 mL. Neurological assessments—including speech, facial symmetry, and limb motor function—were routinely performed to detect early signs of perioperative stroke.

Post-discharge follow-up

Long - term postoperative information such as the occurrence of postoperative complications, re - examination status, and the patient’s condition at the time of follow-up was obtained through telephone follow-up. The telephone follow-up record form is shown in Table 1.

Table 1.

Telephone follow-up record form.

Basic situation	Follow-up time
	Follow - up personnel
	Patient’s name
	Surgical time
Receiver	Relationship with patients
	Whether to understand the patient’s disease condition
	Do you agree to participate in this study?
Postoperative complications and time	Ipsilateral cerebral infarction
	Intraoperative bleeding on the operative side
	Others
	Did not occur
Postoperative review time and results	Carotid Doppler Ultrasound
	CTA^a
	MRA^b
	DSA^c
	Not reexamined
Current state	Current symptoms
	Deceased	Time and reasons

^acomputed tomographic angiography (CTA),

^bmagnetic resonance angiography (MRA).

^cdigital subtraction angiography (DSA).

Statistical methods

The reporting of this study conforms to the STROBE statement for observational studies.¹⁵ No formal sample size calculation was performed because this was an exploratory retrospective cohort study. The sample size was determined by the total number of consecutive eligible patients who underwent classic carotid endarterectomy during the study period.

The study endpoints were defined as the occurrence of stroke, patient death, or reaching the predefined follow-up cutoff point.

Continuous variables were expressed as mean ± standard deviation (range), and categorical variables as counts and percentages. The Mann–Whitney U test was used for comparisons of continuous variables between groups, while the chi-square test or Fisher’s exact test was used for categorical variables, as appropriate. Long-term stroke-free survival was assessed using Kaplan–Meier analysis and the log-rank test. Because of the limited sample size and small number of outcome events, no multivariable regression model was retained in the revised analysis. A two-tailed P-value of ≤ 0.05 was considered statistically significant.

All statistical analyses were performed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). Figure 3 was generated in R version 4.5.0 using the survival and survminer packages to display the Kaplan–Meier curve together with the number-at-risk table.

Results

Basic information of the patient

A total of 45 patients who underwent carotid endarterectomy (CEA) at the Department of Interventional and Vascular Surgery, Peking University First Hospital, between January 2013 and November 2019 were included, encompassing 46 procedures in total. One patient underwent bilateral CEA 21 months apart, and both procedures were counted separately. All procedures were performed by the same senior vascular surgeon (Dr. Zhang Xiansheng), ensuring procedural consistency. Only patients undergoing classic CEA (without patch angioplasty or eversion technique) were included to reduce technical variability. The selection of vascular clips was based on both clinical indication and patient/family preference. Baseline characteristics, including age, sex, smoking status, comorbidities, imaging findings, and antiplatelet therapy, are summarized in Table 2. Preoperative imaging assessments were primarily based on computed tomographic angiography (CTA), supplemented by MRA, DSA, or duplex ultrasonography when appropriate. One representative case with pre- and post-operative CTA is shown in Figure 2.

Table 2.

Basic information of patients.

Project		Total number (cases)	Control group (cases)	Experimental group (cases)	P value^e
Total number		46	31	15
Operative side	Left	27	19	8
Operative side	Right	19	12	7
Age (years)	Overall average:66.07±8.95 (52∼85)
	The average of the control group:66.19±9.55 (52∼85)				0.972
	The average of the experimental group: 65.80±7.86 (54∼80)
Gender	Male	42	29	13	0.827
Gender	Female	4	2	2	0.827
Smoking^a	Yes	25	17	8	0.923
Smoking^a	No	21	14	7	0.923
Underlying diseases (descriptive subtype distribution only)	Hypertension	35	24	11
	Stroke or other cerebrovascular diseases	20	15	5
	Coronary atherosclerotic heart disease	17	9	8
	Diabetes mellitus	12	5	7
	Malignant tumors	3	3	0
	Arrhythmia	2	2	0
	Lipid metabolism disorder	2	0	2
	Depression	2	1	1
	Bronchiectasis	1	1	0
	Gout	1	1	0
	Nephrotic syndrome	1	0	1
	Hypothyroidism	1	0	1
	Medical history of aplastic anemia	1	1	0
	In total^b	43	30	13
History of previous stroke or other cerebrovascular diseases	Yes	20	15	5	0.334
	No	26	16	10	0.334
Antiplatelet drugs	Aspirin100 mg qd	18	11	7
	Aspirin50 mg qd	1	1	0
	Clopidogrel75 mg qd	1	1	0
	Clopidogrel150 mg qd	1	0	1
	Aspirin100 mg qd + Clopidogrel75 mg qd	2	1	1
	Aspirin100 mg qd + Clopidogrel150 mg qd	1	0	1
	In total	24	14	10	0.171
Imaging examination of the contralateral carotid artery^c	Severe stenosis (stenosis degree ≥ 70%)	8	5	3
Imaging examination of the contralateral carotid artery^c	Mild stenosis (stenosis degree < 70%) or no stenosis	38	26	12	0.745
Types of preoperative imaging examinations^d	DSA	1	1	0
	CTA	41	29	12
	MRA	2	0	2
	Carotid Doppler Ultrasound	2	1	1

^aWhether a person smokes or not is determined by whether there is a record of a long - term smoking history in the medical records, and those who have quit smoking are also included.

^bThe total refers to the total number of patients with surgery-related complications, not the total number of complications. Most patients have more than one complication.

^cThe results of digital subtraction angiography (DSA), magnetic resonance angiography (MRA), computed tomographic angiography (CTA), and Carotid Doppler Ultrasound are all acceptable.

^dOne patient underwent both CTA and cervical vascular color Doppler ultrasound simultaneously, and the CTA results were regarded as the standard.

^eP values are provided only for the prespecified main baseline variables included in the between-group comparison. Detailed subtype distributions are presented descriptively only because of sparse cell counts.

In 31 cases, primary closure was performed using traditional suture during surgery, which served as the control group. In 15 cases, primary closure was carried out using vascular clips during surgery, which served as the experimental group.

The mean age of the 45 patients was 66.07 ± 8.95 years (range, 52–85 years). The average age of the control group was 66.19±9.55 (52 - 85) years, and that of the experimental group was 65.80±7.86 (54 - 80) years. Among them, 25 patients had a smoking habit before surgery. A total of 43 patients had different types of chronic underlying diseases that might affect the surgery before surgery. Among them, 35 patients had hypertension, 20 had stroke or other cerebrovascular diseases, 17 had coronary atherosclerotic heart disease, 12 had diabetes, 3 had malignant tumors, 2 had arrhythmia, 2 had lipid metabolism disorders, 2 had depression, 1 each had bronchiectasis, gout, nephrotic syndrome and hypothyroidism. In addition, 1 patient had a history of aplastic anemia, which had been cured, and the hemoglobin was normal at admission. A total of 24 patients had been taking anti - platelet drugs orally for a long time due to underlying heart or cerebrovascular diseases before surgery, and the drugs included aspirin and clopidogrel. Preoperative imaging examinations indicated that 8 patients had severe stenosis (≥70%) of the contralateral carotid artery to the surgical side.

Surgical-related information of patients

The mean operation time for all patients was 126.02 ± 25.11 minutes (range: 82–197 min), with 125.26 ± 26.70 min in the suture group and 127.60 ± 22.25 min in the clip group.

The mean intraoperative clamping time was 16.43 ± 7.63 minutes (range: 5–37 min), with the clip group showing a shorter clamping time (12.67 ± 5.74 min) compared to the suture group (18.26 ± 7.84 min). Average intraoperative blood loss was 50.65 ± 23.32 mL, with 56.77 ± 23.01 mL in the suture group and 38.00 ± 18.97 mL in the clip group. Systolic blood pressure was recorded on the morning after surgery, with the highest value being 179 mmHg. Wound drainage was monitored routinely, and drainage tubes were generally removed when 24-hour drainage was less than 5 mL. Two patients did not require drainage tube placement due to minimal intraoperative bleeding, and their drainage volume was recorded as 0 mL. The mean total postoperative drainage volume was 14.59 ± 15.54 mL (suture group: 13.52 ± 15.23 mL; clip group: 16.80 ± 15.47 mL). Detailed intraoperative data are presented in Table 3.

Table 3.

Surgical-related information of patients.

Project		Total number (cases)	Control group (cases)	Experimental group (cases)	P value^d
Surgical procedure	Classical carotid endarterectomy	46	31	15
Operation time^a	Overall average:126.02±25.11 (82∼197) min				0.959
	The average of the control group:125.26±26.70 (82∼197) min
	The average of the experimental group:127.60±22.25 (99∼177) min
Clamping time^a	Overall average:16.43±7.63 (5∼37) min				0.048
	The average of the control group:18.26±7.84 (5∼37) min
	The average of the experimental group:12.67±5.74 (7∼27) min
Intraoperative blood loss^b	Overall average:50.65±23.32 (20∼150) mL				0.010
	The average of the control group:56.77±23.01 (20∼150) mL
	The average of the experimental group:38.00±18.97 (20∼80) mL
Measure blood pressure on the morning of the first day after surgery.¹⁶	Systolic blood pressure<120 mmHg^c	10	8	2
	120 mmHg≤Systolic blood pressure<150 mmHg	28	18	10
	Systolic blood pressure≥150 mmHg	8	5	3
Total volume of postoperative wound drainage	Overall average:14.59±15.54 (0∼60) mL				0.898
	The average of the control group:13.52±15.23 (0∼60) mL
	The average of the experimental group:16.80±15.47 (0∼56) mL
Wound drainage tube	No wound drainage tube was indwelled.	2	2	0
Wound drainage tube	One wound drainage tube was indwelled.	44	29	15

^aBoth the operation time and the clamping time shall be based on the records in the anesthesia record.

^bThe intraoperative blood loss shall be based on the records in the surgical record. If the surgical record does not clearly state it, the records in the anesthesia record shall be used as the basis.

^c1 mmHg=133.3 Pa。.

^dP values are provided only for the prespecified main perioperative variables included in the between-group comparison. Other rows are presented descriptively only.

Perioperative complication occurrence, mortality, and follow-up results

A total of 46 carotid endarterectomy procedures were performed in 45 patients, with one patient undergoing contralateral surgery 21 months later. Among these, 41 patients completed follow-up, while 5 were lost to follow-up—all from the suture group.

The median follow-up duration was 88 months (IQR, 74–116; range, 60–143 months), with 114 months (IQR, 83–124; range, 60–143 months) in the suture group and 79 months (IQR, 66–87; range, 60–93 months) in the clip group.

Perioperative and long-term events

During the perioperative period, one patient died. Over the course of the long-term follow-up, a total of seven patients died. Perioperative stroke, defined as stroke occurring within 30 days after surgery, was observed in 4 of 46 procedures (8.7%). One additional stroke occurred beyond the 30-day postoperative window. Throughout the entire study period, no cases of reoperation, wound infection, or major bleeding were observed.

Imaging follow-up

A total of 36 patients underwent postoperative imaging, including CTA, MRA, DSA, or duplex ultrasound. Carotid restenosis or occlusion on the operative side was observed in three patients, none of whom developed new neurological symptoms. Representative CTA images before and after surgery are shown in Figure 2.

Figure 2.

Pre- and postoperative computed tomographic angiography (CTA) of a patient who underwent bilateral carotid endarterectomy. The red arrows indicate the locations of vascular clips used during the procedures.

Detailed complication and follow-up data are provided in Table 4.

Table 4.

Occurrence of perioperative complications, mortality, and follow-up results.

Project			Total number (cases)	Control group (cases)	Experimental group (cases)
Follow-up status	Complete the follow-up		41	26	15
Follow-up status	Lost to follow - up		5	5	0
Follow-up time^a	Overall median (IQR): 88 (74–116) months
	Control group median (IQR): 114 (83–124) months
	Experimental group median (IQR): 79 (66–87) months
Causes of death	Perioperative period^b	Septic shock	1	1	0
	Long-term after surgery^b	Malignant tumors	4	4	0
		Acute myocardial infarction	1	1	0
		Pulmonary embolism	1	1	0
		Unknown	1	1	0
Stroke	Perioperative period^d		4	3	1
Stroke	Long-term after surgery^d		1	0	1
Reexamination results^c	In total		36	24	12
	Carotid Doppler Ultrasound	No stenosis was found.	33	23	10
		Stenosis	1	0	1
		Occlusion	2	1	1
	MRA	No stenosis was found.	1	0	1
		Stenosis	0	0	0
		Occlusion	1	1	0
	CTA	No stenosis was found.	26	17	9
		Stenosis	1	0	1
		Occlusion	0	0	0
	DSA	No stenosis was found.	1	1	0
	DSA	Stenosis	0	0	0

^aFollow-up time was defined as the interval from surgery to the occurrence of death or stroke; for event-free patients, it was calculated up to the study’s follow-up cut-off date, in completed months.

^bPerioperative and long-term deaths were distinguished by the time of hospital discharge.

^cFollow-up results included imaging conducted at both our center and external institutions.

^dStroke occurring within 30 days postoperatively was classified as perioperative; events beyond 30 days were considered long-term postoperative strokes.

Statistical analysis

Variable definition and assignment

Patient prognosis following carotid endarterectomy (CEA) may be influenced by multiple clinical factors, including age, comorbidities, degree of carotid stenosis, contralateral occlusion, and preoperative cerebrovascular history. Given the limited sample size, to avoid reduced statistical power due to overparameterization, this study limited the number of covariates to: age (X1), gender (X2), smoking status (X3),underlying diseases other than cerebrovascular diseases (X4), history of stroke or other cerebrovascular diseases (X5), long - term use of antiplatelet drugs (X6), degree of carotid artery stenosis on the operative side (X7), ipsilateral external carotid artery stenosis or occlusion (X8), and severe stenosis of the contralateral carotid artery (X9).

To evaluate the intraoperative efficiency and safety of vascular anastomosis clips, perioperative parameters—including surgical time (Y1), clamping time (Y2), intraoperative blood loss (Y3), and total volume of postoperative wound drainage (Y4)—were also analyzed. Variable definitions and assignments are listed in Table 5.

Table 5.

Definitions of influencing variables, variable codes, statistical methods, and intergroup comparisons between the clip group and the suture group.

Factor names	Variable naming	Inspection methods and influencing factors		P value between groups
Patient outcomes	Y	All-cause mortality	COX regression analysis	0.966
		Perioperative stroke	Fisher’s exact test	0.606
		Long-term stroke	Kaplan-Meier survival analysis	0.210
		Long-term stroke	Fisher’s exact test	0.326
Surgical time (min)	Y1	Linear regression (X6, X7, X8, group)		0.959
clamping time (min)	Y2	Linear regression (X6, X7, X8, group)		0.048
Intraoperative blood loss (mL)	Y3	Linear regression (X1, X6, group)		0.010
Total volume of postoperative wound drainage (mL)	Y4	Linear regression (X1, X6, group)		0.803
Follow-up time (months)	t	Non-parametric test: Mann-Whitney U test (group)		0.005
Age (years)	X1	Non-parametric test: Mann-Whitney U test (group)		0.972
Gender	X2	Chi-square test (group)		0.827
Smoking status	X3	Chi-square test (group)		0.923
Underlying diseases other than cerebrovascular diseases^a	X4	Chi-square test (group)		1.000
History of previous stroke or other cerebrovascular diseases	X5	Chi-square test (group)		0.334
Long - term use of anti - platelet drugs	X6	Chi-square test (group)		0.171
Degree of carotid artery stenosis on the operative side^b(%)	X7	Non-parametric test: Mann-Whitney U test (group)		0.524
Ipsilateral external carotid artery stenosis or occlusion^c	X8	Chi-square test (group)		1.000
Severe stenosis of the contralateral carotid artery^d	X9	Chi-square test (group)		0.745

^aIncludes any systemic comorbidities such as hypertension, diabetes mellitus, coronary artery disease, atrial fibrillation, chronic kidney disease, and hypothyroidism, all of which were diagnosed by qualified medical institutions and/or managed with long-term pharmacotherapy.

^bPreoperative imaging was performed at different centers using various grading schemes. For standardization: “near-occlusion” was coded as 90%, “severe stenosis” as 70%. If a range was provided (e.g., 80–90%), the median value was recorded (e.g., 85%).

^cIpsilateral external carotid artery stenosis or occlusion was defined as ≥50% stenosis.

^dSevere stenosis of the contralateral carotid artery was defined as ≥70% stenosis.

Intergroup comparison of baseline characteristics

Age (X1) was treated as a continuous variable. The Kolmogorov–Smirnov test indicated that the age data in the experimental group did not follow a normal distribution (P = 0.009). Therefore, the Mann–Whitney U test was applied, revealing no statistically significant difference in age between the two groups (U = 231.000, z = –0.035, P = 0.972).

The degree of carotid artery stenosis on the operative side (X7) was also a continuous variable. As the data in both groups were non-normally distributed (P = 0.000 in the control group; P = 0.038 in the experimental group), the Mann–Whitney U test was used and showed no significant intergroup difference (U = 206.000, z = –0.637, P = 0.524).

Follow-up time (t), treated as a continuous variable, did not follow a normal distribution and was therefore summarized as median (IQR) and compared between groups using the Mann–Whitney U test. A statistically significant difference in follow-up duration was observed between the two groups (U = 93.500, z = –2.751, P = 0.005).

The remaining variables—gender (X2), smoking status (X3), underlying diseases other than cerebrovascular diseases (X4), history of stroke or other cerebrovascular diseases (X5), long-term use of antiplatelet drugs (X6), stenosis or occlusion of the ipsilateral external carotid artery (X8), and severe stenosis of the contralateral carotid artery (X9)—were treated as categorical variables and compared using chi-square tests.

Based on these analyses, there were no statistically significant differences in any of the baseline characteristics between the control and experimental groups, except for follow-up time (t), which was longer in the control group (P = 0.005). See Table 5 for details.

Comparison of perioperative indicators between groups

Perioperative parameters, including operative time, carotid clamping time, intraoperative blood loss, and postoperative wound drainage, were compared directly between the clip and suture groups. The clip group had a significantly shorter carotid clamping time and lower intraoperative blood loss than the suture group. No statistically significant differences were observed in total operative time or postoperative wound drainage volume. Detailed results are presented in Table 3.

Comparison of postoperative stroke incidence and all-cause mortality between groups

Perioperative stroke, defined as stroke occurring within 30 days after surgery, occurred in 4/46 procedures (8.7%). Fisher’s exact test showed no statistically significant difference in perioperative stroke incidence between the two groups.

For long-term stroke incidence, Kaplan–Meier analysis with the log-rank test showed no statistically significant between-group difference (log-rank P = 0.210; Figure 3).

Figure 3.

Kaplan–Meier analysis of long-term postoperative stroke incidence.

All-cause mortality was reported descriptively. Because of the limited number of death events, no multivariable survival regression was retained in the revised analysis.

Discussion

In this single-center exploratory retrospective cohort study of classic carotid endarterectomy (CEA) with primary closure, clip-assisted closure was associated with shorter carotid clamping time and lower intraoperative blood loss than continuous suturing. Total operative time and postoperative drainage were similar between groups. During follow-up, no statistically significant between-group differences were detected in perioperative stroke, long-term stroke, restenosis/occlusion, or all-cause mortality. However, given the small sample size, particularly the 15 procedures in the clip group, and the non-randomized design, these findings should be interpreted as hypothesis-generating rather than evidence of equivalence or non-inferiority.

Influence of vascular clips on clamping time and operative duration

Previous experimental and clinical studies have generally shown that clip-based vascular closure or anastomosis can shorten closure or anastomotic construction time compared with suturing, particularly when compared with interrupted techniques.^13,16–20 Our findings are directionally consistent with this literature. In the present cohort, the clip group had a significantly shorter carotid clamping time than the suture group.

By contrast, total operative time did not differ significantly between groups. A plausible explanation is that carotid arteriotomy closure represents only one component of the overall procedure, whereas exposure, plaque removal, hemostasis, and wound closure were otherwise similar between groups. In addition, the present study may have been underpowered to detect modest differences in total operative duration. A shorter carotid clamping time may still be clinically relevant, particularly in patients with limited collateral reserve or significant contralateral carotid disease.²¹

Influence of vascular clips on intraoperative blood loss and postoperative wound drainage

The lower intraoperative blood loss observed in the clip group is biologically plausible and is consistent with earlier reports suggesting that clip-based closure may reduce bleeding from repeated suture-hole punctures.^3,22,23 In our cohort, however, this difference did not translate into a significant difference in postoperative drainage volume.

One possible explanation is that 24-hour wound drainage reflects not only closure-line bleeding but also soft-tissue exudation from the operative field. Therefore, reducing immediate needle-hole bleeding may not necessarily change total postoperative drainage. In addition, continued antiplatelet therapy did not significantly increase intraoperative blood loss in our regression analysis, which is consistent with contemporary carotid surgery guidance supporting continuation of antiplatelet therapy in appropriate patients.²⁴

Impact of vascular clips on complications, stroke, restenosis, and mortality

Prior studies in other vascular beds have generally not shown an obvious increase in complication rates with clip-assisted closure, although isolated adverse events have been reported, particularly when clip selection or application was suboptimal.^{13,23,25–27} In the present cohort, perioperative stroke occurred in 4/46 procedures (8.7%), with three events in the suture group and one in the clip group; one additional late stroke occurred during follow-up. Imaging follow-up in 36 patients identified three cases of restenosis or occlusion, none of which were associated with new neurologic symptoms.

No statistically significant between-group differences were detected for perioperative stroke, long-term stroke, or all-cause mortality. However, these findings should be interpreted cautiously. Because the total sample was small, particularly the 15 procedures in the clip group, the study had limited power for uncommon but clinically important safety endpoints. Accordingly, the absence of statistically significant differences should be understood as no signal detected, rather than evidence of equivalent safety or non-inferiority.

All observed deaths occurred in the suture group, but this pattern should not be overinterpreted because follow-up duration was longer in that group and the number of events was small. Given the limited number of death events, no robust multivariable analysis of mortality was retained in the revised manuscript.

Potential mechanisms underlying the observed effects of vascular clips

Mechanistically, clip-assisted closure may reduce local tissue trauma, avoid repeated needle-hole bleeding, and lessen foreign-material exposure associated with conventional suturing. Experimental studies of clip-based vascular closure have reported better preservation of endothelial integrity, less compliance mismatch, and less intimal thickening than conventional suturing in some models.^{18,19,22,28–30} These observations provide a plausible explanation for the shorter clamping time and lower blood loss observed in our series and suggest a potential benefit for long-term patency. However, whether such histologic or biomechanical advantages translate into clinically meaningful reductions in restenosis or stroke in CEA remains uncertain and requires larger human studies.

Limitations of the present study

Several limitations merit emphasis

First, this was a non-randomized retrospective study. Treatment allocation was influenced not only by clinical judgment but also by patient/family preference, insurance coverage, and affordability of the clip system. This introduces a substantial risk of selection bias and confounding by indication/affordability. This issue is also clinically relevant in the broader vascular surgery literature, because cost differences between clip- and suture-based closure have been shown to influence the economic attractiveness of clip use even when short-term clinical outcomes appear similar in other settings such as arteriovenous fistula creation.³¹ Although measured baseline variables were broadly similar between groups, unmeasured factors such as socioeconomic status, risk tolerance, health literacy, and decision-making preferences may have influenced both treatment choice and outcomes.

Second, the overall sample size was small, particularly in the clip group (15 procedures). As a result, the study had limited statistical power for rare but clinically important endpoints such as perioperative stroke, long-term stroke, restenosis/occlusion, and all-cause mortality. Therefore, the absence of statistically significant between-group differences should not be interpreted as proof of equivalence or non-inferiority.

Third, clip-assisted procedures were introduced later in the study period, resulting in shorter follow-up in the clip group and potential time-related confounding. Although all operations were performed by the same experienced surgeon using a standardized approach, changes in perioperative management or institutional practice over time cannot be excluded.

Fourth, only classic CEA with primary closure was included. Patch angioplasty and eversion techniques were excluded, even though patch closure has been associated with lower restenosis rates and is recommended in contemporary guidelines.^24,32 This improved internal consistency for a device-focused exploratory comparison but limits generalizability. In addition, we previously observed incompatibility between the GC-L clip and certain synthetic patch materials used in our institution, which precluded safe clip-assisted patch closure in this cohort. Future studies should evaluate clip-assisted closure in conjunction with compatible patch materials and broader CEA techniques.

Fifth, follow-up was not uniform for all patients. Some long-term information was obtained by telephone, imaging was partly performed at outside institutions, and five patients—all in the suture group—were lost to follow-up. These factors may have affected event ascertainment and reduced the precision of long-term comparisons.

In summary, this study provides long-term exploratory data suggesting that clip-assisted primary closure is technically feasible in selected patients undergoing classic CEA and may improve intraoperative efficiency. However, because of the non-randomized design, small sample size, and follow-up limitations, these findings should be regarded as hypothesis-generating. Larger prospective studies are needed before broader conclusions can be drawn regarding the safety and effectiveness of clip-assisted primary closure in carotid surgery.

Conclusion

In this exploratory, non-randomized cohort study, clip-assisted primary closure during classic carotid endarterectomy was associated with shorter carotid clamping time and lower intraoperative blood loss than conventional suturing. No signal of increased perioperative or long-term adverse outcomes was detected; however, the study was not powered to establish equivalence or non-inferiority for rare safety endpoints. These findings suggest that clip-assisted closure is technically feasible in selected patients undergoing classic carotid endarterectomy.

Because of the small sample size, non-randomized single-center design, and potential residual confounding, the results should be interpreted as hypothesis-generating. Larger prospective studies are needed to further evaluate the safety and long-term effectiveness of clip-assisted primary closure in carotid surgery.

Supplemental material

Supplemental material - Vascular clips for primary closure in classic carotid endarterectomy: A 5-year exploratory retrospective cohort study

Supplemental material for Vascular clips for primary closure in classic carotid endarterectomy: A 5-year exploratory retrospective cohort study by Kang She, Xiansheng Zhang, Zhengheng Yu, Jie Yin and Gong Cheng in Science Progress.

Footnotes

Acknowledgements

During the preparation of this manuscript, the authors used ChatGPT-5.3 and ChatGPT-5.4 for language translation, English polishing, grammar checking, and improvement of sentence fluency. After using these tools, the authors carefully reviewed, revised, and verified all text. The authors take full responsibility for the content and accuracy of this publication.

ORCID iDs

Kang She

Xiansheng Zhang

Consent to participate

Informed Consent Statement: Written informed consent was obtained from all patients prior to surgery. For postoperative follow-up, verbal or telephone consent was obtained from the patients and documented in the medical records.

Author contributions

Concept construction: S.K., Z.X.S. Methodology design: S.K., Z.X.S. Resource provision: Z.X.S., Y.J., S.K. Verification and implementation: Z.X.S., Y.J., C.G., S.K. Initial draft writing: S.K. Data analysis: S.K., Y.Z.H. Manuscript review and editing: S.K., Y.Z.H. Follow - up survey research: S.K. Data management and analysis: S.K., Y.Z.H. Project supervision: Z.X.S. Project management: S.K., Z.X.S. Surgical operations: Z.X.S., Y.J., S.K., C.G. All authors have read and agreed to the published version of the manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National High Level Hospital Clinical Research Funding (Scientific Research Seed Fund of Peking University First Hospital) (No. 2023SF03).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data underlying this article are not publicly available because of patient privacy and ethical restrictions. De-identified data may be made available from the corresponding author upon reasonable request and with permission from the relevant ethics committee and institution.*

IRB approval

Institutional Review Board Statement: This study was approved by the Biomedical Research Ethics Committee of Peking University First Hospital, Beijing, China (approval No. 2025R0292-0002; approved on August 12, 2025). All patient details were de-identified and anonymized prior to analysis and reporting.

Supplemental material

Supplemental material for this article is available online.

References

Zhou

Wang

Zeng

, et al. Mortality, morbidity, and risk factors in China and its provinces, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Lond Engl 2019; 394(10204): 1145–1158, doi:10/gf48bv. https://doi.org/10.1016/S0140-6736(19)30427-1

Kolkert

Meerwaldt

Geelkerken

, et al. Endarterectomy or carotid artery stenting: the quest continues part two. Am J Surg 2015; 209(2): 403–412, doi:10/f62tb6. https://doi.org/10.1016/j.amjsurg.2014.06.012

Kirsch

Cavallo

Anton

, et al. An alternative system for cerebrovascular reconstructions: non-penetrating arcuate-legged clips. Cardiovasc Surg 2001; 9(6): 531–539, doi:10/c6tzqp. https://doi.org/10.1016/s0967-2109(01)00089-8

Zeebregts

Kirsch

. Expanding Use of Nonpenetrating Clips in Various Surgical Specialities.

Zeebregts

. Non-penetrating Clips for Vascular Anastomosis. Eur J Vasc Endovasc Surg 2005; 30(3): 288–290, doi:10/fkfq9d. https://doi.org/10.1016/j.ejvs.2005.04.010

Berman

Kirsch

Zhu

, et al. Impact of nonpenetrating clips on intimal hyperplasia of vascular anastomoses. Cardiovasc Surg 2001; 9(6): 540–547, doi:10/dbb2jw. https://doi.org/10.1016/s0967-2109(01)00090-4

Kirsch

Gupta

Zhu

. Sutureless vascular anastomosis: the VCS clip. Cardiovasc Surg Lond Engl 2001; 9(6): 523–525, doi:10/cps5d3. https://doi.org/10.1016/s0967-2109(01)00087-4

Schild

Gillaspie

. Economics of an Interrupted Nonpenetrating Clip Anastomosis for Vascular Access Procedures.

Zeebregts

Kirsch

Van Den Dungen

, et al. Five years’ world experience with nonpenetrating clips for vascular anastomoses. Am J Surg 2004; 187(6): 751–760, doi:10/fp4cqw. https://doi.org/10.1016/j.amjsurg.2003.08.028

10.

Tiwari

Cheng

Salacinski

, et al. Improving the patency of vascular bypass grafts: The role of suture materials and surgical techniques on reducing anastomotic compliance mismatch. Eur J Vasc Endovasc Surg 2003; 25(4): 287–295, doi:10/d8k2m3. https://doi.org/10.1053/ejvs.2002.1810

11.

Zeebregts

Kirsch

van den Dungen

, et al. Evolution of staples and clips for vascular anastomoses. Int Surg 2004; 89(3): 152–160.

12.

Kakkos

Tsolakis

. Vascular Anastomotic Clips Revisited. Eur J Vasc Endovasc Surg 2012; 43(2): 232, 10/bx2p4n. https://doi.org/10.1016/j.ejvs.2011.11.017

13.

Findlay

Joseph

. Carotid Arteriotomy Closure Using a Vascular Clip System. Neurosurgery 1998; 42(3): 550–554, doi:10/fqg3mj. https://doi.org/10.1097/00006123-199803000-00024

14.

Brott

Halperin

Abbara

, et al. ASA/ACCF/AHA/AANN/AANS/ACR/ASNR/CNS/SAIP/SCAI/SIR/SNIS/SVM/SVS guideline on the management of patients with extracranial carotid and vertebral artery disease. J Am Coll Cardiol 2011; 57(8): e16–e94. https://doi.org/10.1016/j.jacc.2010.11.006

15.

von

Altman

Egger

, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ 2007; 335(7624): 806–808. https://doi.org/10.1136/bmj.39335.541782.AD

16.

Sultan

SAH

Madhavan

Evoy

, et al. Arcuate-legged Nonpenetrating Vascular Closure Staples (VCS): Early Experience. Ann Vasc Surg 2001; 15(3): 338–342, doi:10/b8qk4p. https://doi.org/10.1007/s100160010084

17.

Calles-Vázquez

Viguera

Crisóstomo

, et al. Vascular Closure Stapler Clip Anastomosis Decreases Aortic Cross-Clamping Time Compared to Interrupted Nonabsorbable and Running Absorbable Sutures in Growing Pigs. Ann Vasc Surg 2006; 20(1): 35–41, doi:10/bbg2w5. https://doi.org/10.1007/s10016-005-9321-z

18.

Zeebregts

Van Den Dungen

Buikema

, et al. Preservation of endothelial integrity and function in experimental vascular anastomosis with non-penetrating clips. Br J Surg 2002; 88(9): 1201–1208, doi:10/dtsbw3. https://doi.org/10.1046/j.0007-1323.2001.01857.x

19.

Caiati

Madigan

Bhagat

, et al. Vascular Clips Have No Significant Effect on the Cellular Proliferation, Intimal Changes, or Peak Systolic Velocity at Anastomoses in Rabbit Vein Grafts. J Surg Res 2000; 92(1): 29–35, doi:10/bpc3v5. https://doi.org/10.1006/jsre.2000.5843

20.

Zeebregts

Van Den Dungen

JJAM

Van Det

, et al. Randomized clinical trial of continuous sutures or non-penetrating clips for radiocephalic arteriovenous fistula. Br J Surg 2004; 91(11): 1438–1442, doi:10/bxn894. https://doi.org/10.1002/bjs.4719

21.

Yuan

Huo

, et al. A single-center retrospective study with 1-year follow-up after CEA in patients with severe carotid stenosis with contralateral carotid artery occlusion. Front Neurol 2022; 13: 971673. https://doi.org/10.3389/fneur.2022.971673

22.

Dal Ponte

Berman

Patula

, et al. Anastomotic tissue response associated with expanded polytetrafluoroethylene access grafts constructed by using nonpenetrating clips. J Vasc Surg 1999; 30(2): 325–333, doi:10/crfbmg. https://doi.org/10.1016/s0741-5214(99)70144-8

23.

Schild

Raines

. Preliminary prospective randomized experience with vascular clips in the creation of arteriovenous fistulae for hemodialysis. Am J Surg 1999; 178(1): 33–37, doi:10/ck7bp3. https://doi.org/10.1016/s0002-9610(99)00113-0

24.

Naylor

Rantner

Ancetti

, et al. Editor’s Choice – European Society for Vascular Surgery (ESVS) 2023 Clinical Practice Guidelines on the Management of Atherosclerotic Carotid and Vertebral Artery Disease. Eur J Vasc Endovasc Surg 2023; 65(1): 7–111, 10/gr5jk8. https://doi.org/10.1016/j.ejvs.2022.04.011

25.

Donker

JMW

Tijnagel

Van Zeeland

MLP

, et al. Anastomoses in the Common Femoral Artery, Vascular Clips or Sutures? A Feasibility Study. Ann Vasc Surg 2013; 27(2): 194–198, doi:10/f47d6g. https://doi.org/10.1016/j.avsg.2012.02.008

26.

Pikoulis

Koronarchis

Filis

, et al. Study comparing sutures and nonpenetrating titanium clips for arteriotomy closure after embolectomy. Surg Endosc 2001; 15(7): 726–728, doi:10/ckf2q4. https://doi.org/10.1007/s00464-001-0003-9

27.

Cook

Schuman

Standage

, et al. Patency and flow characteristics using stapled vascular anastomoses in dialysis grafts. Am J Surg 2001; 181(1): 24–27, doi:10/ffqjnr. https://doi.org/10.1016/s0002-9610(00)00547-x

28.

Komori

Shoji

Furuyama

, et al. Non-penetrating Vascular Clips Anastomosis Inhibited Intimal Thickening Under Poor Runoff Conditions in Canine Autogenous Vein Grafts. Eur J Vasc Endovasc Surg 2001; 21(3): 241–247, doi:10/ch57xb. https://doi.org/10.1053/ejvs.2001.1297

29.

Baguneid

Goldner

Fulford

, et al. A comparison of para-anastomotic compliance profiles after vascular anastomosis: Nonpenetrating clips versus standard sutures. J Vasc Surg 2001; 33(4): 812–820, doi:10/fjnghj. https://doi.org/10.1067/mva.2001.112806

30.

Lee

Choo

, et al. Anastomosis of Vessels less than 2 mm with the Vascular Clip System Clip Applier. J Korean Med Sci 2001; 16(3): 303–308, 10/gt6jck. https://doi.org/10.3346/jkms.2001.16.3.303

31.

Terranella

Fefferman

Zhang

, et al. Economic evaluation of suture versus clip anastomosis in arteriovenous fistula creation. J Vasc Surg 2021; 73(6): 2098–2104. https://doi.org/10.1016/j.jvs.2020.10.076

32.

Orrapin

Benyakorn

Siribumrungwong

, et al. Patch angioplasty versus primary closure for carotid endarterectomy. Cochrane Database Syst Rev 2022; 2022(8): CD000160. https://doi.org/10.1002/14651858.CD000160.pub4

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.97 MB