Direct Carotid Puncture for Mechanical Thrombectomy: Technical Advances in Access and Closure

Abstract

BACKGROUND AND OBJECTIVES:

Mechanical thrombectomy (MT) is the standard of care for patients with anterior circulation acute ischemic stroke with large-vessel occlusion and is traditionally performed through transfemoral access (TFA). In cases of difficult vascular anatomy, however, TFA may be prohibitive to achieving reperfusion in a timely manner, and alternative access methods must be used. Direct carotid puncture (DCP) has previously been described only as a bail-out access strategy fraught with complications, with inconsistent access and closure methods used. This study describes our DCP protocol and analyzes patient and procedural outcomes of MT performed through DCP, aiming to determine if expeditious DCP during MT can lead to safe and successful outcomes comparable with those of TFA.

METHODS:

Twenty-two patients who underwent MT through DCP at a single institution from 2017 through 2025 were retrospectively reviewed. Clinical details and outcomes data were collected, including indications for DCP, devices used, technical success of the procedure/revascularization scores, puncture, and revascularization times.

RESULTS:

Of 510 MT performed, 4.3% were via DCP. IV thrombolytic was administered in 45% of DCP cases. Reasons for conversion to DCP were for type 3 and/or tortuous arch (73%) or femoral/iliac artery occlusion/stenosis (27%). Average time from groin to carotid puncture was 30 minutes and from carotid puncture to reperfusion was 28 minutes, compared with our average reperfusion time of 22 minutes for TFA cases. Successful revascularization was achieved in 90.1% of DCP cases, with 0 procedural complications. All patients underwent sheath removal through carotid cutdown in the operating room after the endovascular procedure.

CONCLUSION:

DCP is a safe, feasible, and effective technique for revascularization in acute stroke intervention in patients with difficult anatomy. Given the average time from DCP to reperfusion is comparable with TFA, timely conversion to DCP up-front may improve procedural outcomes in cases of obvious complex anatomy.

ABBREVIATIONS:

DCP

direct carotid puncture

MT

mechanical thrombectomy

TFA

transfemoral access.

Mechanical thrombectomy (MT) is the standard of care for patients with anterior circulation acute ischemic stroke with large-vessel occlusion and is traditionally performed through transfemoral access (TFA). In cases of difficult vascular anatomy, however, femoral access may be prohibitive to performing the intervention in a timely manner, and alternative access methods must be used. Transradial access continues to gain traction in neurointervention by bringing lower access site complications and higher patient satisfaction when compared with TFA; however, some literature raise concern that it has not achieved comparable revascularization times to TFA in stroke intervention.^1-3 An additional access method, albeit less frequently used, is direct carotid puncture (DCP).^4,5

There are limited number of reports describing the utility of DCP in acute stroke intervention.^6-12 This technique has been described only as a bail-out strategy after all other access options have been exhausted, often comparing DCP patient outcomes to those of patients whose procedure was aborted entirely.^6,13 Furthermore, consistent strategies for access, sheath removal, and arteriotomy closure after DCP are lacking, with frequently reported complications related to access and/or closure.¹⁴ With increased awareness of which patients are likely to need DCP, familiarity with the technical nuances involved, and an established protocol for efficient conversion from TFA to DCP during MT can lead to safe and successful patient outcomes comparable with those of transfemoral procedures.

Objective: This study describes the largest single-institution experience of 22 patients who underwent MT via DCP, including the protocol used for efficient access conversion, DCP and arteriotomy closure technique, and procedural outcomes.

METHODS

Study Design

All interventional cases were performed at a Comprehensive Stroke Center. A prospectively maintained database of all MT cases performed between November 2017 and March 2025 was retrospectively reviewed for cases of MT through DCP. All stroke management decisions were made by the attending providers in accordance with current American Heart Association guidelines and published trials. The decision to proceed with DCP was made in all cases in which femoral access was unobtainable or after failure to access the target vessel with a base catheter due to prohibitive anatomy. Clinical details and outcome data were collected from the database. Angiographic imaging was reviewed for each case, assessing the indications for carotid access, devices used, technical success of the procedure/thrombolysis in cerebral infarction score, puncture, and revascularization times. This research was conducted following the principles as outlined in the Declaration of Helsinki and was conducted in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology guidelines. This study was reviewed and approved by our health system Institutional Review Board. All patients and or families were consented for the procedures and for research publication.

DCP Protocol

All thrombectomies at our institution are performed under general anesthesia, which facilitates the transition to DCP when necessary. The patient is positioned with head in neutral position and the neck prepped in the usual sterile fashion. Using ultrasound guidance, the common carotid artery is accessed with an 18G needle, approximately 2 cm above the clavicle. With the 0.035-inch J-wire in place, serial dilations are performed with 4 Fr, 5 Fr, 6 Fr, and 7 Fr dilators, followed by a 7- Fr sheath placement. Under fluoroscopy, the sheath is positioned at the distal common carotid artery and secured using a 2-0 stich in 2 locations. The sheath hub is changed to a Touhy to allow use of intermediate/reperfusion catheters. Craniocervical angiography is performed through the sheath, and the remainder of the procedure continued per usual thrombectomy protocol with the sheath serving as the base catheter. Through the sheath, we advance a reperfusion catheter attempting (A Direct Aspiration, First Pass Technique) and when that fails our second attempt would be with a combination of stent retriever and aspiration. We did not use in any of these cases intracranial angioplasty or intraarterial thrombolytics. In cases of tandem occlusion secondary to atherosclerotic cervical carotid stenosis, we usually perform balloon angioplasty and avoid acute stenting unless flow limiting. None of the patients in this DCP series required acute stent placement. After obtaining reperfusion, repeat cervical angiography is performed to ensure proper sheath positioning, and the sheath is left sutured in place connected to continuous saline drip for later removal.

Sheath Removal

The patient is placed under general anesthesia in the operating room. The head is placed in slight extension and lateral rotation away from the side of the sheath. A 4-cm horizontal incision is made just above the catheter entry point. Two layers of sterile drape are placed—one exposing the entire catheter and the other just the incision. The skin is incised as well as the platysma, the medial border of the sternocleidomastoid is freed and the muscle is lateralized until the common carotid artery is reached, at which point the sheath would be visible. Circumferential dissection around the carotid is performed and vessel loops placed proximal and distal to the sheath entry. A clamp is placed proximal to the sheath entry point into the carotid, and the sheath was removed. A second clamp is placed distal to the arteriotomy site, the vessel lumen inspected and irrigated with heparinized saline, and the puncture site closed with running continuous 6-0 Prolene suture. Before tying the sutures, the distal clamp is removed to back bleed and then the suture tied before removing the proximal clamp. Given the brief clamp time, we do not heparinize the patients. After confirming good Doppler signals on ultrasound, hemostasis is obtained and the incision closed in usual fashion.

RESULTS

Clinical

There were 510 mechanical thrombectomies performed over the study period, of which 22 were performed through DCP (4.3% of all thrombectomies). Patient and procedure details of all DCP interventions are presented in Table 1. The average patient age was 80 years. IV thrombolytic was administered before the procedure in 45% of patients. Average National Institutes of Health Stroke Scale on arrival was 16. Most occlusions were located in the M1 segment (12), with 6 tandem internal carotid artery/middle cerebral artery occlusions, 2 M2 occlusions, and 2 cervical internal carotid artery occlusions.

TABLE 1.

Baseline Characteristics

Characteristic	All patients (n = 22)
Age (y)	80 ± 9
IV thrombolytic administered	10 (45%)
NIHSS	16 ± 6
Vessel involved
Tandem extracranial/intracranial	6 (27%)
Cervical ICA	2 (9%)
M1	12 (56%)
M2	2 (9%)

ICA, internal carotid artery; IV, intravenous; M1, middle cerebral artery—first division, M2, middle cerebral artery—second division; NIHSS, National Institutes of Health Stroke Scale.

Procedural/Outcomes

Reasons for conversion to DCP were as follows: 16 type 3 and/or tortuous arch and 6 femoral/iliac artery occlusion/stenosis (Table 2). The average time from groin puncture to carotid puncture was 30 minutes, and average time from carotid puncture to reperfusion was 28 minutes. Successful revascularization as defined by >/ = thrombolysis in cerebral infarction 2b was achieved in 20 (90.1%) patients. There were 0 procedure-related complications, and there were no complications among the patients who received IV thrombolytic before the procedure. All patients underwent sheath removal through carotid cutdown in the operating room after the endovascular procedure, delaying the sheath removal for 24 hours in cases where IV thrombolytic was administered. Average length of hospital stay was 9 days.

TABLE 2.

Procedural Data

Characteristic	All patients (n = 22)
Groin puncture to carotid puncture time (min)	28 ± 28.4
Groin puncture to revasc time (min)	56 ± 39
Carotid puncture to revasc time (min)	28 ± 16.5
TICI ≥ 2b	20 (90%)
First passes	12 (56%)

Revasc, revascularization; TICI, thrombolysis in cerebral infarction.

DISCUSSION

Many stroke patients suffer from systemic cardiovascular disease, with disease of the main access vessels often hindering access, catheterization, and/or navigation of the arteries once access is obtained. DCP is an underrecognized and underutilized solution that can quickly establish access and facilitate vessel navigability to quickly achieve reperfusion, albeit technical nuances that must be accounted for to perform safely. Numerous previous studies have raised concerns regarding morbidity and mortality associated with DCP. Dumas et al published the largest DCP series to date including a meta-analysis of 227 patients who underwent MT through DCP, citing an overall complication rate as high as 20.3%, and concluding that DCP is technically feasible but should only be reserved as a bailout technique given the high risk profile.¹² The authors challenge this with their institution's experience and highlight a few key factors contributing to their higher procedural success rates and lower complication rates.

The complication rate in this series is notably lower than the majority of current literature, which the authors largely attribute to access and closure methods. In this study, all carotid punctures were performed under ultrasound guidance, minimizing the risk of dissection or pseudoaneurysm formation, and eliminating the risk of inadvertent jugular vein puncture. While neck hematoma formation is known to be the most frequently encountered complication—accounting for over 70% of the complications cited in Dumas et al, the authors experienced no neck hematomas.

Arteriotomy closure is another dilemma, as there are no approved vessel closure devices for the carotid artery to date, yet intravascular closure devices are the most frequently used method of closure in the literature.^14,15 Alternative closure methods described include manual pressure and direct repair through surgical cut down, the latter of which has been described in different clinical scenarios with low complication rate.^8,16 Closing the arteriotomy under direct visualization not only circumvents the uncertainty of closure device applications but also allows flexibility regarding the timing of closure, particularly in cases where IV thrombolytic was administered. In this series, all arteriotomies were closed on postoperative day 1. Combining percutaneous access with open surgical closure allows DCP to be both a fast and safe option.

The rate of conversion to DCP in this study is higher than what is cited in the literature at 4.3%. In addition, after accounting for the steep learning curve involved, increased comfort and experience lead to 16 of the DCP cases being performed in the past 3 years, with a conversion rate of 9.8%. By comparison, the largest DCP meta-analysis reported a conversion rate of 0.4%.¹⁴ Furthermore, familiarizing the entire procedural team with DCP allows for quick and efficient conversion, which was obtained successfully in 100% of cases in this series. The mean time from femoral to carotid puncture in our series was 30 minutes, compared with 75 minutes in the meta-analysis. While this is of course confounded by the timing of the operator's decision to convert to DCP, the factors which contribute to unfavorable vascular anatomy are well known, as is their negative impact on thrombectomy outcomes.¹⁷

By eliminating unnecessary delays to DCP, the overall average reperfusion time in this series is 56 minutes (from the time of TFA). By comparison, the largest series to date cited an average total reperfusion time of 200.2 minutes.¹² When combined with a successful revascularization rate of 90.1%, patients undergoing DCP in this series achieved clinical outcomes comparable with those undergoing MT through femoral access. Cord et al⁶ identified acute ischemic stroke patients with prohibitive vascular access and compared those who underwent DCP with those whose procedure was aborted entirely. There is now an increasing body of evidence supporting how patients who require DCP for MT can have the same benchmarks for success as patients undergoing TFA for acute stroke interventions.

Limitations

There are limitations to this study inherent to its retrospective design. Furthermore, while this is the largest single-institution study of DCP, the relatively small sample size and single-center design limits the study's external validity. The decision to convert to DCP was solely made at the discretion of the interventionalist performing the procedure and was not standardized. While procedure times were compared with those performed through TFA, there was no formal control group for direct comparison.

CONCLUSION

DCP is a safe, feasible and effective technique for revascularization in acute stroke intervention in patients with difficult anatomy. The average time from DCP to reperfusion was 28 minutes in this series, and overall reperfusion time decreased as the threshold to convert from TFA lowered over time. Given its safety, DCP should be considered upfront in cases of obvious complex anatomy. By increasing its adoption into the neurointerventionalists' toolkit, DCP can elevate thrombectomy success rates in select patients and ultimately improve the quality of stroke care provided.

Funding

This study did not receive any funding or financial support.

Disclosures

The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.