Article Text

Download PDFPDF

Original research
Initial experience with SOFIA as an intermediate catheter in mechanical thrombectomy for acute ischemic stroke
  1. Johnny H Y Wong1,
  2. Huy M Do2,
  3. Nicholas A Telischak2,
  4. Adrienne M Moraff1,
  5. Robert L Dodd1,
  6. Michael P Marks2,
  7. Shreya M Ingle2,
  8. Jeremy J Heit2
  1. 1 Department of Neurosurgery, Stanford University Hospital, Stanford, California, USA
  2. 2 Department of Radiology, Interventional Neuroradiology Division, Stanford University Hospital, Stanford, California, USA
  1. Correspondence to Dr Jeremy J Heit, Department of Radiology, Interventional Neuroradiology Division, Stanford University Hospital, 300 Pasteur Drive, S047, Stanford, CA 94305, USA; jheit{at}


Background The benefits of mechanical thrombectomy for emergent large vessel occlusion (ELVO) have been established. Combined mechanical/aspiration (Solumbra) and a direct aspiration as a first pass technique (ADAPT) are valid procedures requiring an intermediate catheter for clot suction. Recently, SOFIA (Soft torqueable catheter Optimized For Intracranial Access) was developed as a single lumen flexible catheter with coil and braid reinforcement, but its suitability for mechanical thrombectomy had not been evaluated.

Objective To describe our initial experience with SOFIA in acute stroke intervention and evaluate its efficacy and safety.

Methods All patients with ELVO undergoing endovascular stroke intervention with SOFIA were identified. Demographic, presentation, treatment, and complication data were recorded. Primary outcome was Thrombolysis in Cerebral Infarction (TICI) 2b/3 revascularization rate and the number of passes required. Secondary outcomes included complication rates and discharge National Institute of Health Stroke Scale (NIHSS) score.

Results 33 patients with a mean age of 72 years were treated for ELVO with SOFIA and IV tissue plasminogen activator was administered in 67%. Vessel occlusion involved the internal carotid artery (15.2%), M1 (48.5%), and M2 (24.2%) segments, and posterior circulation (12.1%). Median presentation NIHSS score was 14 (IQR 11–19) and discharge NIHSS 4 (IQR 2–14). The Solumbra technique represented 94% of treatments and ADAPT 3%. The TICI 2b/3 revascularization rate was 94%, including 48.5% TICI 3 with an average of 1.6 passes. The symptomatic reperfusion hemorrhage rate was 6%. Procedural complications occurred in four patients, but were unrelated to SOFIA. Mortality was 21%, secondary to failed revascularization, hemorrhagic transformation, and baseline medical condition.

Conclusions Mechanical and aspiration thrombectomy with SOFIA is safe and effective with high revascularization rates. Its trackability, stability, and luminal size make SOFIA suitable for stroke intervention.

  • Thrombectomy
  • Technology
  • Stroke
  • Intervention
  • Catheter

Statistics from

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.


Stroke affects approximately 795 000 people in the USA annually, has a prevalence of 2.6%, and accounts for 1 in 20 deaths.1 Ischemic stroke comprises 87% of strokes, and large vessel occlusion accounts for 29–46% of all ischemic stroke presentations.1–3

Several recent randomized trials have established the superiority of mechanical thrombectomy for emergent large vessel occlusion (ELVO) in acute stroke.4–8 Compared with maximal medical therapy with IV tissue plasminogen activator (tPA), revascularization rates and functional outcomes were significantly improved using mechanical thrombectomy with stent retrievers. Consistent with prior studies using IV tPA, timely revascularization was beneficial in reducing ischemic burden and final infarct volumes.9–13

Variable techniques are used for performing mechanical thrombectomy, and there is little standardization among them between centers. The application of flow arrest proximal to the clot using a balloon guiding catheter has been used to reduce the likelihood of clot fragmentation and distal embolization.14 ,15 Similarly, more distal flow arrest may be achieved through the combined use of stent retrievers and suction with a distal access catheter at the clot interface (the ‘Solumbra’ technique) or suction through a distal access catheter alone may be used (the ‘ADAPT’ technique (a direct aspiration as a first pass technique).16–18 Both the Solumbra and ADAPT techniques require an intermediate or distal access catheter to apply suction at the clot interface.

An ideal intermediate catheter for stroke intervention would negotiate tortuosity in cerebral vessels, such as the siphon at the cavernous internal carotid artery (ICA) or distal cervical segments of the vertebral artery, without difficulty, provide adequate stability for microcatheter vessel selection into distal ICA, M1 or M2 segments for large vessel occlusion, and have an adequate inner diameter to generate strong suction at the clot interface for clot retrieval.19 Recently, the SOFIA (Soft torqueable catheter Optimized For Intracranial Access; MicroVention, Tustin, California, USA) catheter was introduced with a coil and braid design that is specifically intended as an intermediate or distal access catheter for intracranial use. SOFIA has been used as an intermediate catheter for the first-line aspirational technique, as well as combined stent retriever and aspiration technique with proximal balloon occlusion in Europe.19–21 To our knowledge, the application of SOFIA for mechanical thrombectomy using predominantly the Solumbra technique has not yet been evaluated in North America. The purpose of this study was to describe our initial experience with the 5-French SOFIA catheter in combination with the ‘Solumbra’ technique for acute stroke and to evaluate its safety and efficacy in this setting.

Materials and methods

This retrospective review was approved by the institutional review board at Stanford University and was compliant with the Health Insurance Portability and Accountability Act. All patients with acute ischemic stroke from large vessel occlusion treated with mechanical thrombectomy using the 5-French SOFIA access catheter (MicroVention) were identified from a prospectively collected stroke intervention database. Patient data were obtained from the electronic medical record.

At our institution, suitability for mechanical thrombectomy was assessed on the basis of CT with angiogram and perfusion, or MRI diffusion-weighted imaging, gradient echo, MR angiography, and MR perfusion imaging. The size of the infarct core, and territory at risk were calculated based on RAPID software, an automated imaging processing software package.22 Patients were considered appropriate candidates for acute stroke intervention if there was a confirmed large vessel occlusion (ICA, M1 or M2) with infarct core <70 mL and mismatch volume >15 mL or mismatch ratio >1.8. Patients with basilar artery occlusion and salvageable tissue were also included.

All procedures were performed on a Siemens Artis Zee biplane system in a dedicated neuroangiography suite. In our institution, mechanical thrombectomy was performed preferentially with monitored anesthesia care, but general anesthesia was administered if clinically indicated. Equipment selection for each procedure was based on the preference of the treating neurointerventionalist; there are four attending neurointerventionalists at our institution. In most instances, a triaxial access system was used: arterial access was via the common femoral artery with an 8-French 45 cm sheath (Cordis Corporation, Miami Lakes, Florida, USA), a 6-French 90 cm shuttle sheath (Cook Medical, Bloomington, Illinois, USA) over a 5-French Berenstein catheter (Cordis Corporation) was positioned into the proximal C1 segment of the internal carotid artery or origin of the vertebral artery, and the 5-French 115 or 125 cm SOFIA (MicroVention) was advanced into the internal carotid or vertebral artery (see online supplementary video in appendix). Alternative distal access catheters used during this period were at the treating physicians' discretion, and included: 5-MAX ACE, ACE-60, and ACE-64 (Penumbra, Alameda, California, USA). IV heparin was administered after groin access was obtained.

Solumbra was our preferred first-line thrombectomy technique. The vessel occlusion was crossed with a 0.014″ microwire, and the Traxcess microwire (MicroVention) was most commonly used. Next, a 0.025″ Velocity microcatheter (Penumbra) or 0.021″ Trevo Pro 18 microcatheter (Concentric Medical, Mountain View, California, USA) was advanced distal to the occlusion, and the microwire was removed. A Solitaire 2 (ev3 Neurovascular, Irvine, California, USA) or Trevo XP Provue Retriever (Stryker, Alameda, California, USA) stent retriever was deployed through the microcatheter such that the device bridged the thrombus. The 5-French SOFIA was then advanced up to the clot interface and suction was applied with the Penumbra aspiration system through the 5-French SOFIA for 2 min to achieve flow arrest. The SOFIA and the stent retriever could then be removed as a single unit while maintaining flow arrest through the SOFIA (figure 1). In one patient, the MindFrame Capture LP (MindFrame Inc, Irvine, California, USA) device was used for a M2 occlusion with the equivalent Solumbra technique. The ADAPT technique was used in a selected group of patients, particularly in the posterior circulation.

Figure 1

Combined mechanical and aspiration thrombectomy for a right M1 occlusion. (A) DSA anteroposterior (AP) projection demonstrates a right proximal M1 occlusion (arrow); (B) single roadmap image demonstrates Solitaire stent deployed in the proximal M2 segment distal to the clot (arrow), and SOFIA (Soft torqueable catheter Optimized For Intracranial Access) positioned at the cavernous internal carotid artery (dotted arrow); (C) single radiograph demonstrating Solitaire stent deployed in the proximal M2 segment (arrow) and the SOFIA positioned at the M1 occlusion, applying suction at the clot interface for the Solumbra technique; (D) post-mechanical thrombectomy DSA AP projection demonstrates a Thrombolysis in Cerebral Infarction 3 reperfusion.

Patient demographic data, risk factors, clot location, and IV tPA administration were recorded. Functional outcome data (National Institute of Health Stroke Scale (NIHSS) and modified Rankin Scale scores) were included from presentation, discharge, and at the 3-month follow-up. Technical details, such as equipment used, number of passes to achieve recanalization, final Thrombolysis in Cerebral Infarction (TICI) reperfusion score, and complications were recorded. The primary outcome was the rate of successful revascularization, defined as TICI 2b or TICI 3 revascularization and the number of passes required to achieve recanalization. Secondary outcome measures were complication rates and NIHSS score on discharge.


A total of 33 consecutive patients (20 female, 13 male) were treated for large vessel occlusion using the SOFIA catheter. One patient was treated in August 2014, with the remainder of the cohort being treated between May 2015 and March 2016. The SOFIA cohort represented 57% of all ELVO strokes treated at our institution during this time period. The mean patient age of the cohort was 72 years. Patient demographic details and vascular risk factors are summarized in table 1.

Table 1

Baseline patient demographics, risk factors and clot characteristics

Median NIHSS was 14 (IQR 11–19) on presentation, and the mean time from symptom onset to presentation was 322 min. IV tPA was administered in 22 patients (67%). Vessel occlusion predominantly affected the anterior circulation in 29 patients (88%), while four patients (12%) had posterior circulation strokes. Within the anterior circulation strokes, clot locations were distributed in the ICA terminus (5 patients, 15.2% of entire cohort), M1 segment (16 patients, 48.5%) and M2 segment (8 patients, 24.2%). Two patients had tandem lesions in the proximal cervical ICA, which required balloon angioplasty before intracranial access was obtained. These did not require carotid artery stenting.

The Solumbra technique was used in 31 patients (94%). This included 25 patients using the Solitaire 2 stent retriever delivered through 0.025″ Velocity microcatheter, 5 patients using the Trevo XP combination stent retriever and microcatheter, and 1 patient who had a M2 clot treated with a MindFrame capture device. The ADAPT technique was used in one patient (3%), who had a basilar artery thrombosis. IA thrombolysis was delivered by microcatheter in one patient (3%) with a distal M2 occlusion not deemed favorable for mechanical thrombectomy (table 2).

Table 2

TICI 2b, number of passes, and time to revascularization by mechanical thrombectomy techniques

Overall, successful revascularization was achieved in 31 patients (94%), including TICI 3 in 16 patients (48.5%). The average number of passes to reperfusion was 1.6 (95% CI 1.3 to 2.0), and revascularization with single pass occurred in 20 patients (61%). Mean time from access to reperfusion was 46 min. No significance differences in TICI 2b/3 revascularization rates or time to revascularization were noted between the Solitaire or Trevo stent retrievers, or other techniques (table 3).

Table 3

Treatment outcomes

Median NIHSS score on discharge was 4 (IQR 2–14). Mortality was 21.2%, secondary to failed revascularization (6%), hemorrhagic transformation (6%), and severe baseline medical disease (9%). No complications were directly related to the SOFIA catheter. One patient who had embolization into a new territory received IA thrombolysis only for a distal M2 inferior division occlusion. At conclusion of the procedure, a distal superior M2 division embolus was identified, possibly due to prolonged duration of the process. Symptomatic intraparenchymal hemorrhage following reperfusion occurred in two patients (6%). Procedural complications occurred in four patients (12%). One patient had a vertebral artery dissection related to placement of the shuttle sheath for a posterior circulation thrombus treated with a Trevo stent retriever, but made an excellent recovery. Two patients had subarachnoid hemorrhage. One of these patients had subarachnoid hemorrhage secondary to perforation during vessel selection with a microwire, and despite a TICI 3 recanalization, died several days later. The second patient, who had moderate residual deficits from established infarction, had subarachnoid hemorrhage during clot retrieval with the MindFrame stent retriever. One patient had hemorrhage into a contralateral pre-existing intracranial tumor, and shortly died from the symptomatic intracerebral hemorrhage.


Safe and timely revascularization is the goal of mechanical thrombectomy for acute ischemic stroke. Factors that may be optimized to achieve these goals include ease and reduced time for equipment preparation, excellent trackability and maneuverability of catheters to access the clot interface, and reduced number of attempts at clot retrieval to achieve successful revascularization (TICI 2b or 3). Recent experience with the SOFIA catheter in mechanical thrombectomy has been applied to the ADAPT technique, and more recently, in association with balloon-guide catheters in Europe.19–21 Here we describe our experience with the SOFIA catheter in acute stroke intervention, predominantly with the Solumbra technique.

In this study, Solumbra was applied in 31 patients (Solitaire 27 patients, Trevo 3 patients, and MindFrame 1 patient). At our institution, we prefer a triaxial system, with initial access obtained through an 8-French groin sheath, and selection into the internal carotid artery for an anterior circulation stroke, or proximal vertebral artery for posterior circulation stroke, with a 6-French 80 or 90 cm shuttle sheath over a 5-French Berenstein catheter. Additionally, before groin puncture, the 5-French 125 cm SOFIA over a 0.025″ Velocity microcatheter and 0.014″ Traxcess microwire is prepared separately, and introduced once selection into the ICA or vertebral artery is achieved. We found this to be the most time efficient method to access and treat the occluded vessel.

Before the introduction of SOFIA, alternative intermediate catheters were used, which occasionally were limited by an inability to track around tortuosity at the carotid siphon for anterior circulation strokes, or extradural vertebral artery for posterior circulation strokes. The choice of intermediate catheters was at the physician's discretion, and during this time, the 5-French SOFIA was used in about 50% of all ELVO cases. We found that the 5-French SOFIA catheter easily negotiated the petrous and cavernous segments of the ICA, even without tracking over a glide wire or microcatheter (see online supplementary video). However, it is our preferred practice to advance the 5-French SOFIA to the clot interface once the microcatheter and microwire have traversed the site of occlusion. The 5-French SOFIA reached the clot interface in 100% of our procedures without difficulty and was highly trackable over only a microcatheter and microwire. In this series, only the 5-French SOFIA was used. Experience with the 6-French SOFIA for mechanical thrombectomy is limited.

Compared with other trials,4–7 our patient population was more elderly (mean age 72 years and 33% aged >80), which often requires navigation of more tortuous arterial anatomy to reach the occluded vessel. In such situations, we found that the 5-French SOFIA 125 cm responded well and provided adequate length and support for mechanical thrombectomy with the shuttle guide positioned in either the mid-cervical ICA or distal common carotid artery. The slightly shorter length compared with other competing intermediate catheters may theoretically be associated with improved trackability and reduced loss of microcatheter length on the operating table. The 5-French SOFIA's luminal size with an internal diameter of 0.055″ permitted delivery of the stent retriever through a 0.025″ microcatheter and the ability to apply suction at the clot interface simultaneously.

In this series, the successful revascularization rate (TICI 2b or 3) was 94% with 49% achieving TICI 3 revascularization. These results are comparable to those obtained in the recent randomized trials.4–7 In other stroke intervention series with SOFIA and the ADAPT technique, TICI 2b reperfusion was achieved in 87% of patients.19 In this series, SOFIA could reach the clot interface in 96% of cases. Similarly, Kabbasch et al 20 used a first-line lesional aspiration technique and achieved a TICI 2b or above revascularization rate of 90%, of which 67% were successful with first pass aspiration alone. In contrast, 61% of the patients in our series had successful revascularization with one attempt with the ‘Solumbra’ technique. Of these patients, 55% (11 patients) had TICI 3 reperfusion.

The ADAPT technique was used in one case for a distal-basilar occlusion in the current series, in which the SOFIA catheter passed directly to the clot interface without a microcatheter or microwire. Successful revascularization was achieved with a single attempt.

Although our complication rates were higher than expected in comparison with other interventional stroke series, no complications were directly related to the SOFIA catheter. One patient had a subarachnoid hemorrhage after thrombectomy with the MindFrame device for a distal M2 vessel occlusion, which caused a vessel laceration. Another patient had an intraparenchymal hemorrhage into a coincidental contralateral intracranial tumor, which might have been caused by intraprocedural heparinization.

Limitations of this study include the small sample size of the cohort, the retrospective nature of the study, self-reported results and outcomes, and the inclusion of patients from a single neurovascular center. Our study may include selection bias because intermediate catheter selection in each case was based upon the treating neurointerventionalist's discretion. A large prospective study would be required to accurately compare differences between intermediate access catheters.


Combined mechanical and aspiration thrombectomy with the SOFIA catheter is safe and effective in this series, and no complications were directly attributable to the use of the 5-French SOFIA catheter. Its ease of preparation, ability to track around tortuous vessels without causing dissections, stability for delivery of the stent retriever, and its luminal size permitting suction at the clot interface make it a suitable intermediate catheter for the Solumbra technique in stroke intervention. These factors might have contributed to a successful revascularization (TICI 2b/3) rate of 94%, with a low number of passes necessary for revascularization.


JHYW acknowledges the support of the Royal Australasian College of Surgeons for the Stuart-Morson Travelling Scholarship to enable this research to be conducted.



  • Contributors JHYW, HMD and JJH conceived and designed the study. JHYW and JJH acquired the data and performed the statistical analyses. All authors participated in the interpretation of the data, the preparation and editing of the manuscript. The final version was approved by all authors.

  • Competing interests JJH and HMD received consulting fees from MicroVention, Inc.

  • Ethics approval This study was performed in accordance with the Stanford University's internal review board policies and is in compliance with the Health Insurance Portability and Accountability Act.

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

  • Data sharing statement Published and unpublished data may be shared for additional publications pending the authors’ agreement to a research collaboration when appropriate.