Article Text

Download PDFPDF

Case series
Direct carotid puncture for endovascular thrombectomy in acute ischemic stroke
  1. Adam Roche1,
  2. Emma Griffin2,
  3. Seamus Looby3,
  4. Paul Brennan2,
  5. Alan O’Hare2,
  6. John Thornton3,
  7. Karl Boyle4,
  8. David Williams4,5,
  9. Barry Moynihan4,5,
  10. Sarah Power2
  1. 1 Department of Radiology, St Vincents University Hospital, Dublin, Ireland
  2. 2 Interventional Neuroradiology Service, Department of Radiology, Beaumont Hospital, Dublin, Ireland
  3. 3 Interventional Neuroradiology Service, Department of Radiology, Beaumont Hospital and Royal College of Surgeons in Ireland, Dublin, Ireland
  4. 4 Department of Geriatric and Stroke Medicine, Beaumont Hospital, Dublin, Ireland
  5. 5 Department of Medicine, Royal College of Surgeons in Ireland, Dublin, Ireland
  1. Correspondence to Dr Sarah Power, Interventional Neuroradiology Service, Department of Radiology, Beaumont Hospital, Dublin, Ireland; sarahpower{at}


Background Mechanical thrombectomy is the standard of care for acute ischemic strokes with proximal intracranial occlusion. Arterial access is commonly achieved with femoral artery puncture, although this is not always possible. In this case series, we describe 11 cases of anterior circulation stroke where direct carotid puncture was used to obtain vascular access.

Methods and materials A review of a prospectively maintained thrombectomy database over a 2-year period (August 2016 – August 2018) was undertaken to identify cases where direct carotid access was performed. CT and angiographic imaging were reviewed. Indications for carotid access, techniques used, technical success of procedure, recanalization rates, procedure-related complications, and patient outcomes were assessed.

Results Eleven patients out of 498 overall thrombectomy procedures (2.2% thrombectomies) underwent direct carotid access. Median National Institutes of Health Stroke Scale was 20. Seventy three percent of patients received intravenous thrombolysis. The direct carotid approach was performed following the failed femoral approach due to unfavorable aortic arch anatomy, vessel tortuosity, and severe atherosclerotic disease. Direct carotid puncture was successful in 10 patients, and unsuccessful in one. Successful recanalization (TICI 2b–3) was achieved in eight patients. One patient had spontaneously recanalized on angiography. There was failed recanalization in one patient with tandem ICA and M1 occlusion. Carotid access complications included one patient with both neck hematoma and asymptomatic ICA dissection, and one of delayed central retinal artery occlusion.

Conclusion This case series highlights direct carotid puncture as a successful alternative when the femoral approach is not possible, allowing thrombectomy in patients who would otherwise be unsuitable.

  • stroke
  • thrombectomy
View Full Text

Statistics from


The importance of rapid diagnosis and appropriate intervention in the setting of acute ischemic stroke with proximal intracranial arterial occlusion has been well established.1–3 Previous studies have demonstrated that shorter procedure times in the endovascular treatment of ischemic stroke are associated with better overall clinical outcomes.4 5 There are many challenges to achieving prompt revascularization in large vessel strokes suitable for mechanical thrombectomy, including rapid clinical assessment, timely interpretation of imaging, and the achievement of adequate arterial access in patients with potentially difficult anatomy or underlying vascular disease.

Successful endovascular access and catheter navigation represents a crucial step in mechanical thrombectomy treatment. Anatomical variants such as unfavorable aortic arch anatomy and arterial tortuosity can present potential barriers to the trans-femoral approach, resulting in prolonged procedure times and potentially prevent recanalization.

Direct carotid puncture is an alternative access technique that can be utilized when the standard trans-femoral approach is difficult or not possible. Prior to the development of femoral approaches to arterial access, all early cerebral angiography procedures were performed using the carotid puncture technique. Carotid puncture has been previously described for the endovascular treatment of intracranial aneurysms and carotid artery stenting. However these procedures are typically performed electively using general anesthetic, in contrast to emergency endovascular thrombectomy which may be performed with none or minimal conscious sedation.6–8

In this study, we describe 11 cases in which direct carotid puncture was performed or attempted for endovascular treatment of acute ischemic stroke.

Materials and methods

This is a single institution study. All interventional cases were performed in Beaumont Hospital, a large university-affiliated teaching hospital with a 24 hours' acute stroke intervention service, which provides emergency endovascular stroke treatment and takes referrals from multiple institutions across Ireland. Patients with acute stroke symptoms are selected for endovascular therapy based on guidelines derived from current best evidence.1–3 9 The criteria for thrombectomy follows national guidelines and includes patients with an anterior circulation stroke up to 24 hours since symptom onset, without evidence of extensive ischemic change on non-contrast CT brain, an Alberta Stroke Program Early CT score (ASPECTS) of ≥5 and proximal intracranial occlusion on CT angiogram with favourable collaterals.

A review of our prospectively maintained thrombectomy database over a 2-year period (August 2016 – August 2018) was undertaken to identify cases where direct carotid access was performed for endovascular thrombectomy in acute ischemic stroke. This ongoing audit of our service is registered with the Institutional Quality and Standards Department. Cross-sectional and angiographic imaging for each case was reviewed. Clinical details and outcome data were obtained from our database. Indications for carotid access, techniques used, technical success of procedure, treatment in cerebral ischemic (TICI) score, procedure-related complications, and patient outcomes were all assessed.

In all cases included in this study, direct carotid puncture was performed with the patient placed in the supine position and the neck extended as much as patient habitus, and comfort allowed. We found that turning the head away from the side of puncture helped with neck extension and patient comfort in some cases. The area was prepped and draped using the usual sterile technique and covering the face of conscious patients was avoided. Local anesthetic (1% lignocaine) was administered to soft tissues. Common carotid puncture site identification and access method was operator-dependent, performed either by palpation or ultrasound guidance. Ultrasound guidance gave the added benefit of allowing visualization of the level of the common carotid bifurcation, and identifying the jugular vein, which could then be avoided. The ideal target puncture site is approximately 3 cm above the clavicle, as a puncture too close to the clavicle results in an almost perpendicular entry angle into the common carotid artery, making access difficult which can result in kinking of dilators and sheath.

The puncture was performed using either a micropuncture kit (n=6) or standard 18G arterial access needle (n=5). For micropuncture access, a 21G micropuncture needle was used to access the carotid at a 45-degree angle, a floppy tip 0.018 inch microwire was inserted, and progress of the wire was assessed with fluoroscopy. Following this, a 5-Fr dilator was advanced into the artery, the microwire removed, and a 0.035 inch J wire inserted with fluoroscopy guidance, over which a 6-Fr sheath could then be placed. Standard access used an 18G access needle and a 0.035 inch J wire, with a similar technique. Once the 6-Fr sheath was placed in the common carotid artery, following test contrast injection, carotid angiography was performed to confirm the site of intracranial occlusion. If required, the sheath could be navigated into the proximal ICA with roadmap and fluoroscopy guidance. In the majority of cases, the puncture site was the common carotid artery. In one case (Case 7) it was necessary to puncture the left internal carotid artery directly as there was a critical stenosis of the ICA origin which could not be crossed from a trans-femoral approach. Direct internal carotid puncture was performed with a standard 18G access needle using palpation and residual contrast in the ICA as a guide. Arterial access was therefore obtained distal to the stenosis, a 0.035 inch J wire inserted with fluoroscopy guidance, over which a 6-Fr sheath was placed. Following test contrast injection in the ICA, angiography was performed to confirm the site of occlusion.


Over the 2-year period, a total of 498 patients underwent endovascular intervention for acute ischemic stroke in our institution with the vast majority of cases using a standard femoral approach. There were 11 cases identified in which direct carotid access was either attempted or performed for mechanical thrombectomy (2.2% of all thrombectomies performed during this period). Patient and procedural details are outlined in table 1. The median NIHSS at presentation was 20, ranging from 12 to 28. Median ASPECTS was 9 (Range 7–10). There was one case of acute large artery occlusive stroke secondary to a thromboembolic event during endovascular treatment of a ruptured aneurysm in a patient with subarachnoid hemorrhage (SAH). In this case, arterial access to the right carotid artery for aneurysm treatment was extremely difficult. This was the second attempt at endovascular treatment, and in view of age and co-morbidities the patient was not considered by neurosurgery to be suitable for clipping. Catheters could not be advanced beyond the proximal right common carotid artery, a thromboembolic complication occurred during attempted access with carotid terminus occlusion. As this occurred during an endovascular procedure, CT was not performed, and NIHSS score was not calculated as the patient was under general anesthetic (Case 11). Direct carotid puncture allowed arterial access for successful thrombectomy which was immediately followed by aneurysm coiling.

Table 1

Patient and Procedural Details 

The location of arterial occlusion included the middle cerebral artery (MCA) M1 segment (n=5), MCA M2 (n=2), one case of tandem MCA M1 and cervical internal carotid artery (ICA) occlusion, one case of MCA M1 segment occlusion and tandem critical ICA stenosis, one case of cervical ICA occlusion, and one case of carotid terminus occlusion. Median time from onset of symptoms to procedure was 3 hours 22 min, with times ranging from 1 hour 28 min to 11 hours 16 min. Thrombolysis with intravenous tissue plasminogen activator was given in eight patients. The median time from thrombolysis to initial groin puncture time was 1 hour 45 min, with times ranging from 6 min to 4 hours 15 min. Contraindications for thrombolysis included one case that was outside the thrombolysis window in our institution (more than 4.5 hours since onset), one case with gastrointestinal bleeding at the time of presentation, and one case of iatrogenic carotid terminus occlusion in a patient with coexisting aneurysmal subarachnoid hemorrhage and unsecured aneurysm.

The trans-femoral approach was initially attempted and was unsuccessful in 10 of the 11 cases in this series. The most common indication for direct carotid puncture was an unfavorable aortic arch or vessel tortuosity resulting in difficulty accessing the internal carotid artery using the standard trans-femoral approach (n=8). One patient had coarctation of the aorta with complex thoracic aneurysmal disease and direct carotid puncture was performed immediately, without attempting femoral access (Patient 1). There was one case where it was not possible to cross a critical ICA stenosis from the trans-femoral approach to the common carotid artery and direct ICA puncture was therefore performed distal to the ICA origin stenosis (Patient 9). There was one case in which the patient had complete occlusion of both iliac arteries and a fem-fem bypass graft, making peripheral access at the groin impossible (Patient 10).

Procedure times are outlined in table 1. Median total length of procedure for cases 1 through 10 was 71 min. Procedure length of 161 min in case 11 includes aneurysm coiling. Where femoral access was attempted first, and subsequent direct carotid access achieved, the median time from femoral to carotid puncture was 39 min. In case 11, the time interval from identification of the carotid terminus thrombus and carotid puncture was 16 min. Median time from carotid puncture to first reperfusion was 26 min.

Mechanical thrombectomy was performed in all cases using a stent retriever device with concomitant aspiration through an intermediate catheter. Devices used are outlined in table 1. The median number of passes was one, with a minimum pass of one and maximum pass of seven. The majority of procedures (n=6) were performed using only local anesthetic (1% lignocaine subcutaneously). There were four cases that used conscious sedation (intravenous fentanyl +/-midazolam) and a single case, with a coexisting SAH (Patient 11), performed under general anesthetic.

Carotid puncture was successfully achieved in 10 of 11 cases. In one case (Patient 8), common carotid access could not be successfully achieved, and a distal fragment of the 0.018 inch microwire sheared off and lodged in the soft tissues of the neck. Difficulty with access in this case was due to the attempted access point being too proximal and close to the clavicle, with a steep, almost perpendicular approach. As a result there was difficulty advancing dilators and sheath over the microwire through the muscular wall of the carotid, with kinking of devices, and fracture of microwire occurred. This resulted in failed recanalization and a 30-day mRS of 4.

Carotid access complications included one case (Patient 6) in which the patient developed a left-sided neck hematoma at the puncture site immediately post-procedure associated with stridor. This patient had received intravenous thrombolysis prior to thrombectomy treatment. Duration from IV tPA to carotid puncture was 120 min, and to insertion of closure device 140 min. CTA of the head and neck was performed to further evaluate. There was some lateral displacement of the airway due to mass effect from the hematoma, but no active extravasation of contrast. Following anesthetic review, no airway support or further intervention was required. The cause of the hematoma was unclear. An Angioseal closure device was inserted at the end of the procedure. There was persistent ooze at the puncture site following the closure device insertion, but no clear failure of the closure device during insertion. While it is possible the hematoma is related to failure of the closure device used, this cannot be certain. Venous bleeding is also a possibility, and this complication did not occur in any of the other cases in which the device was utilized. The CTA also demonstrated an iatrogenic left internal carotid artery dissection in the same patient, which was and remained asymptomatic. The dissection was remote to the common carotid puncture site. It is unclear at what point in the procedure this dissection occurred. There was one case in which delayed ipsilateral retinal artery occlusion occurred 5 days postprocedure (Patient 9). In this case there was a critical stenosis/near occlusion, but not complete occlusion of the ICA on initial CTA, together with M1 occlusion, and direct ICA puncture was performed for M1 thrombectomy as the critical stenosis of the ICA origin could not be crossed from below. On CTA the ICA had filled in a delayed fashion, and minimal contrast passage was seen past the stenosis on angiography. The ICA therefore was not entirely occluded, and was the likely source of this embolus. Repeat CTA obtained at time of onset of the central retinal artery occlusion did not show any new abnormality. Therefore while the mechanism of the delayed central retinal artery occlusion is not entirely clear, it is most likely related to the underlying stroke mechanism, rather than the direct ICA puncture. If the patient had undergone emergent carotid revascularization with endarterectomy early post-procedure it is possible this complication would not have occurred. Overall complication rate for direct carotid puncture in this series was therefore three of 11 patients (27%).

Of the 10 cases with successful carotid access, a TICI score of 2b–3 was achieved in eight patients. In one case of M1 occlusion, angiography following carotid puncture showed the M1 segment occlusion had spontaneously recanalized, without improvement of patient symptoms (Patient 10), angiography demonstrated distal M3 branch occlusions in this patient that were beyond the reach of thrombectomy. In two cases, recanalization was unsuccessful (TICI 0): one case in which common carotid artery access by direct puncture could not be achieved (Patient 8); in the second case common carotid artery direct puncture and access was successful, however, this patient had a tandem occlusion of the ICA and M1 (Patient 7), unfortunately the ICA occlusion could not be crossed, and thrombectomy could not be performed.

A 6Fr AngioSeal closure device (St Jude Medical, MN, USA) was utilized in order to achieve hemostasis at the puncture site in 10 cases. Manual compression of the puncture site was sufficient for hemostasis in the one case of failed common carotid puncture as there was no access to the artery in this case.

On follow-up at 30 days, two of the 11 patients had functional outcome of mRS 0–2 (18%), two patients had mRS of 3 (18%), and three patients mRS of 4 (27%). The mortality rate at 30 days was 36% (n=4), which included the patient with aneurysmal subarachnoid hemorrhage, and cause of death for this group was directly related to their underlying stroke or SAH rather than due to the procedure itself.


This case series included 11 patients that underwent direct carotid puncture for endovascular treatment of acute ischemic stroke over a 2-year period. Anatomical variants associated with technically challenging carotid access in neuro-interventional procedures have been extensively described in the literature, however this has been mostly in conjunction with cases of carotid artery stenting and cerebral aneurysm embolization.10 The most common variant aortic arch occurs when the left common carotid artery shares a common origin with, or arises directly from, the right brachiocephalic artery. The prevalence of this ‘Bovine’ aortic arch is estimated to be between 10% to 20% of the general population. In Type II and Type III aortic arches there is an increased vertical distance from the origin of the brachiocephalic trunk to the top of the aortic arch. In patients with these examples of variant aortic arch anatomy, access to the common carotid artery can be extremely challenging or potentially impossible and can result in prolonged procedure times.11 12 In addition, many patients undergoing endovascular stroke management are elderly, have significant history of hypertension, and can therefore have quite pronounced supra-aortic vessel tortuosity. Kaymaz et al explored the relationship between supra-aortic vessel anatomy and ICA access times in thrombectomy patients.13 ICA access time was found to be significantly increased by anatomy and vessel tortuosity, with take-off angle of the left common carotid artery, take-off angle of the brachiocephalic trunk, and tortuosity of the common carotid artery having the highest impact on ICA access times.13 In another study by Ribo et al, predictors of difficult carotid access were age >75 years, hypertension, dyslipidemia, and left carotid catheterisation.5 In our series, the median age was 78 years, which likely contributed to difficulty with ICA access. CT angiogram of the head and neck is a useful investigation to identify potentially unfavorable aortic arch anatomy or supra-aortic tortuosity prior to attempting endovascular access. While initially attempting femoral access remains the standard approach, there may be a small group of patients that could be identified early as candidates for an alternative access technique and proceed straight to direct carotid puncture, therefore reducing overall procedure time. Our case series demonstrates that this is feasible in the emergency setting.

Underlying vascular disease can also result in severe arterial stenosis and diffuse vessel tortuosity anywhere along the path from the femoral artery to supra-aortic arteries, potentially leading to challenging trans-femoral access and catheter navigation. One case included in this study had a previous fem-fem crossover graft with complete occlusion of the proximal femoral and iliac arteries. In this case, it was not possible to progress the catheter beyond the fem-fem crossover, necessitating the need for a more proximal access point. Diffuse vessel tortuosity can significantly prolong catheter navigation from the distal access point and common carotid access may be difficult or near impossible through a femoral approach. While palpation of the femoral pulses and ultrasound assessment can indicate potential for difficult femoral access, more proximal disease of iliac system or aorta cannot be well assessed. In such cases of vessel tortuosity and severe arterial disease, it is often not possible to predict difficult catheter navigation prior to attempting the procedure.

Direct carotid puncture is by no means a new technique. Early cerebral angiography procedures were performed using direct carotid puncture, before technological advances allowed the development of trans-femoral access for intravascular treatment in the 1960s and 1970s.14 Direct carotid puncture has also been previously reported for endovascular treatment of intracranial aneurysms, vascular malformations, intracranial angioplasty/stenting, and carotid artery stenting as well as in a small number of stroke intervention cases.6–8 14 15 Despite this, it remains a challenging procedure for the neuro-interventional radiologist to perform. There are currently no devices designed specifically for carotid access, which would benefit from shorter length sheaths, catheters, wires, and devices than currently available. Achieving and maintaining carotid access can be extremely difficult with issues including the initial puncture of the muscular common carotid artery with needles specifically designed for more peripheral access. Catheter navigation with inappropriately long catheters and potential kinking of sheaths are also potential issues.

Other alternative access techniques reported in the literature include trans-radial or trans-brachial artery approach for thrombectomy in both anterior circulation as well as posterior circulation large artery occlusive stroke.16–20 A recent study by Chen et al has demonstrated equivalence in efficacy and efficiency between trans-radial and trans-femoral approach for mechanical thrombectomy in anterior circulation stroke,20 and there are reports in the literature of success through the trans-radial approach when a trans-femoral approach has failed.16 Trans-radial is preferred over trans-brachial as the puncture site is easily compressible. In our institution we have successfully utilized brachial or radial artery access for stroke intervention in a small number of posterior circulation cases (n=2) in addition to aneurysm treatments, however as we do not routinely use a trans-radial or trans-brachial approach for angiography or procedures in anterior circulation pathologies, we have little experience in this regard. We do however have an overall large combined experience in acute stroke intervention, and in the vast majority of cases we are successful in the trans-femoral approach to the occluded artery, with only 2.2% of our 498 thrombectomy cases over the 2-year study period requiring a direct carotid puncture approach.

Potential complications of direct carotid puncture include neck hematoma and carotid dissection. Neck hematoma can cause more significant morbidity and mortality than groin hematoma with the potential for rapid airway compromise and possible need for emergent surgical evacuation. Hematoma at access sites in thrombectomy patients is of particular concern, as the majority of patients undergoing treatment for acute ischemic stroke receive thrombolysis with intravenous tissue plasminogen activator (Alteplase) prior to mechanical thrombectomy.9 Hemostasis at the femoral access site is generally successfully achieved using closure devices such as Angioseal (St. Jude Medical, Minnetonka, MN, USA), Exoseal (Cordis Corporation, Bridgewater, NJ, USA) and StarClose (Abbott, Chicago, IL, USA). Options for achieving hemostasis following percutaneous direct carotid puncture include manual compression or off-label use of a closure device.6 7 14 21 In our study, we utilized the Angioseal closure device in all patients in whom a successful carotid puncture was performed, with manual compression alone only performed in the one patient where common carotid access had failed. In our series, intravenous thrombolysis was administered in eight of 11 patients (73%). Neck hematoma occurred following direct carotid puncture and thrombectomy in one patient. While it is possible the hematoma was related to failure of the closure device used, this cannot be certain. Venous bleeding is also a possibility, and this complication did not occur in any of the other cases in which the device was utilized.

A concern with the use of closure devices such as Angioseal following direct carotid puncture is the fear of dislodgement and embolization of the intra-luminal footplate into the intracranial arterial circulation. Other potential complications include inadequate closure with bleeding/hematoma formation, dissection, and injury to the artery wall during closure with pseudoaneurysm formation. There are small numbers of cases already published in the literature using closure devices following carotid puncture. One study already published describes the use of Angioseal as a hemostatic closure device for direct carotid puncture following endovascular procedures of aneurysm treatment and carotid artery stenting.21 Of eight patients included in this study a complication occurred in one patient where neck hematoma developed. Angiography during hematoma formation in this case from a femoral approach did not show contrast extravasation, and it was felt that the hematoma was most probably due to inadvertent injury to the internal jugular vein during the initial puncture and sheath placement. There were no other complications and no evidence of carotid stenosis or vascular wall abnormalities at 6 months' follow-up. In these cases, where carotid puncture was performed for aneurysm treatment or carotid stenting, heparin was administered for the procedure and the effect of heparin was reversed post-procedure with protamine sulfate prior to insertion of the closure device. We typically do not administer heparin for thrombectomy procedures in our department, and the effect of thrombolysis, while short-lived, cannot be readily reversed in such a manner.

Other studies have described the use of the Mynx closure device (CardinalHealth) and the StarClose device in cases of direct carotid puncture for thrombectomy.14 15 The StarClose device does not leave an intravascular foreign body, and instead deploys a nitinol clip onto the surface of the artery around the access site in a purse string fashion, therefore it would be of interest as a device for use in closure of direct carotid puncture. Mokin et al14 described its use in two cases in addition to applying manual pressure. Its use was uneventful in the first case, but a large neck hematoma developed in the second case. Jadhav et al report the use of the Mynx closure device in two cases without good success.15 Hematoma developed in one case, and there was persistent contrast extravasation from the puncture site in the second case on angiography with haemostasis finally achieved by manual pressure.

An alternative technique for direct carotid puncture is following open surgical exposure, ‘carotid cut-down’ technique. Weismann et al have described six cases of endovascular treatment of acute stroke involving combined surgical access to the carotid artery and carotid puncture, with eversion carotid endarterectomy performed if there is coexisting high grade carotid artery stenosis22 In this case series, surgical access to the carotid was performed prior to attempting percutaneous puncture in five cases and one case in which percutaneous carotid puncture was initially performed, but complicated by dissection of the common carotid artery resulting in the need for surgical access. This study advocates for the use of a direct surgical approach over percutaneous carotid puncture, citing easier sheath placement, reduced complications, and safer closure of the access site, using surgical closure with a purse string suture. Neck hematoma did occur, however, in one of the six cases treated in this study, even with suture closure of the artery under direct visual control. Dorfer et al also described surgical cut down with carotid puncture under direct visual control in eight patients where carotid access could not be accessed via percutaneous trans-femoral route.6 This series did not include stroke patients, and was performed for aneurysm or vascular malformation treatment. Hemostasis was achieved with purse string suture with requirement for placement of one or two extra sutures in some cases to achieve complete hemostasis. Patients were then monitored overnight in the intensive care unit for possible neck swelling or respiratory problems. Surgical cut down procedures in this series were performed in a fully equipped neuroendovascular operating suite, by neurosurgeons. In the same paper, Dorfer et al also described direct percutaneous puncture of the common carotid artery in five patients where hemostasis was achieved by manual compression for 20 min.6 No access site complications occurred in either group. As stroke intervention is now performed by interventionists with a variety of different backgrounds, for example, neuroradiology, neurology, stroke, and neurosurgery, not all of whom are familiar with obtaining surgical exposure of the carotid,  in many cases, this carotid cut down technique will require a multidisciplinary approach with a vascular surgeon or a neurosurgeon in the interventional suite, and possibly combined operating room/angio suite. Open surgical exposure may therefore not be feasible in many centers for acute stroke treatment.

This technical challenge of direct carotid puncture can be assisted by using ultrasound to guide the initial carotid puncture. In this study, the choice of using palpation versus ultrasound to guide arterial access was operator-dependent. Ultrasound imaging of the neck prior to carotid puncture has the advantage of allowing the common carotid artery to be rapidly identified and distinguished from the internal jugular vein. The carotid artery bifurcation can be visualized to ensure access is obtained at a suitable point to allow accurate insertion of the arterial access sheath. The use of a micropuncture needle presents another technical challenge to this procedure. In this study, a 21 G micropuncture needle was used in some cases to access the carotid. This technique ensures less hematoma but presents challenges in introducing and advancing dilators and sheaths in the carotid due to the flimsy nature of the 0.018 inch wire, and the muscular nature of the arterial wall. While larger access needles can result in easier initial access, they can increase the risk of hemorrhage and hematoma.

Our mortality rates are high for this cohort (36%). These patients did in fact have successful recanalization. Three out of the four patients had a severe stroke while the fourth patient (Patient 11) had a complicated clinical course with aneurysmal SAH and iatrogenic carotid terminus occlusion. Overall outcomes in this group are not as good as published in trials, however, they were of high NIHSS and once femoral access had failed they were very likely to have poor outcomes if no attempt at carotid access had been made. Two patients (18%) had an excellent outcome (mRS 0). A further two patients (18%) had mRS of 3. Cause of death for this group was directly related to their underlying stroke or SAH rather than to the carotid puncture. Where femoral access was attempted first, and subsequent direct carotid access achieved, the median time from femoral to carotid puncture was 39 min, and this delay to ICA recanalization may well have contributed to the poorer outcomes in the group.

While patients for whom direct carotid access is required represent a small proportion of all stroke patients (representing 11 out of 498 stroke patients suitable for endovascular treatment in our series), we feel that it is important that direct carotid puncture should be considered in situations when trans-femoral access is impossible. Neuro-interventionists should aim to proceed to direct carotid approach early when the femoral approach is proving difficult rather than spending excessive amounts of time persevering with different catheter and wire combinations.


This case series highlights direct carotid puncture as a successful alternative when trans-femoral access is not possible, allowing thrombectomy in patients who would otherwise be unsuitable.


View Abstract


  • Contributors Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work- AR, EG, SL, PB, AO’H, JT, KB, DW, BM, SP; Drafting the work and revising it critically for important intellectual content- AR, EG, SL, PB, AO’H, JT, KB, DW, BM, SP; Final approval of the version to be published- AR, EG, SL, PB, AO’H, JT, KB, DW, BM, SP; Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved- AR, EG, SL, PB, AO’H, JT, KB, DW, BM, SP.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.

  • Competing interests None declared.

  • Ethics approval A prospective database is maintained of all aspects of this process under the remit of ongoing service audit and is therefore excluded from the requirement for ethics approval.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Patient consent for publication Not required.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.