Background New device technology has changed the techniques used for revascularization of emergent large vessel occlusion in acute stroke. We report technical results using stent retrievers (SRs) for thrombectomy alone versus SRs used in conjunction with a new group of devices, intracranial aspiration catheters (IACs). Our aim is to demonstrate differences in procedural time and thrombectomy attempts between these two groups.
Methods A retrospective evaluation was performed of a prospectively maintained database of 97 patients treated at a single institution for anterior circulation stroke with SRs. Patients were divided into two groups, a combination group defined as the SR/IAC group and the SR alone group defined as the SR group.
Results Patients in the SR/IAC group had a mean age of 66 years vs 59 years in the SR group (p=0.008). Mean presenting National Institutes of Health Stroke Scale (NIHSS) scores in the SR/IAC and control groups were 18.7 and 18.2, respectively (p=0.50). Recanalization rates (Thrombolysis In Cerebral Infarction (TICI) 2b or 3) in the SR/IAC and SR groups were 85% (58/68) and 90% (26/29), respectively (p=0.41). Mean time from groin arteriotomy to recanalization was 50±3.6 min (range 19–136) in the SR/IAC group (n=59) and 61±6.6 min (range 28–140) in the SR group (n=27) (p=0.049). The total number of thrombectomy attempts in the SR/IAC and SR groups were 1.9±0.1 (range 1–4) and 2.5±0.6 (range 1–6), respectively (p=0.009). Post-procedural subarachnoid hemorrhage was seen in 15% (10/68) and 10% (3/29) of cases in the SR/IAC and SR groups, respectively (p=0.41).
Conclusion When using SRs for intracranial stroke thrombectomy, the concurrent use of IACs is associated with a decrease in procedural time and thrombectomy attempts compared with use of SRs alone.
Statistics from Altmetric.com
New device technology has altered the techniques used for intracranial arterial thrombectomy in the setting of an ischemic stroke with emergent large vessel occlusion (ELVO). These advancements have resulted in faster recanalization and higher recanalization rates, yielding better clinical outcomes compared with earlier devices.1–5 However, with the increasing number of choices in device technology, the optimal approach to quick, safe thrombectomy remains to be definitively characterized. This carries value in ELVO, as time to recanalization is of critical importance. Faster times to recanalization have been associated with improved clinical outcomes.6–8 For example, Khatri and colleagues recently demonstrated a 10% decline in the likelihood of a good outcome for a 45 min delay to reperfusion.7
Key advancements in device technology include stent retrievers (SRs) and large-bore intracranial aspiration catheters (IACs). There are numerous technical reports on these and other devices. For example, a recent report compared IAC usage alone versus the SR/IAC combination.5 There are also single-arm series on the use of an IAC alone without an SR which suggest that this technique may be as efficacious or even more efficacious than using an SR.3 ,4 ,9–11 There are also single-arm series on the SR/IAC combination approach.12 Still others have compared use of balloon guide catheters versus non-balloon guide catheters with SRs.13 The majority of recent trials have the common element of SRs, which have become standard of care at most institutions. However, it has yet to be shown in a single report what differences are seen when comparing the use of SRs alone to SRs used concurrently with IACs. Our purpose is to compare SR thrombectomy alone versus a combination approach using both SRs and IACs.
Since SRs became available at our institution in April 2012, we have used these devices in all intracranial thrombectomy cases. IACs became available at our institution in July 2013, and since that time we have used these catheters in conjunction with SRs in all intracranial thrombectomies because of their potentially complementary role in thrombectomy. Here we report the technical results from all consecutive anterior cerebral circulation SR thrombectomy cases, comparing the SR thrombectomy alone group with an SR/IAC combination group. Our goal is to investigate and highlight any differences in recanalization rates, procedural times, number of thrombectomy attempts, and rates of post-procedural subarachnoid hemorrhage (SAH) between these two cohorts of patients.
Our study was approved by our hospital's institutional review board and conducted in compliance with the Health Insurance Portability and Accountability Act. A prospective multidisciplinary database of intracranial thrombectomy cases that captures baseline demographic variables, procedural data, and patient outcomes is maintained at our comprehensive stroke center. A retrospective analysis of this database was performed to assess procedural and peri-procedural variables related to all intracranial anterior cerebral circulation thrombectomy cases performed at our institution from April 2012 until April 2016.
Medical record review
We recorded baseline patient and imaging characteristics as well as procedural and peri-procedural variables. Data gathered were as follows: patient age and gender, administration of IV tissue plasminogen activator (t-PA), National Institutes of Health Stroke Scale (NIHSS) score at presentation, guide catheter used, location of guide catheter tip, location of arterial occlusion, rate of recanalization defined as Thrombolysis In Cerebral Infarction (TICI) grade 2b or 3, time from groin arteriotomy to vessel recanalization, time required for carotid bulb stent and/or angioplasty if applicable, number of thrombectomy attempts needed to achieve recanalization, rates of post-procedural SAH, and whether SAH resulted in new symptomatic neurological deficits. Patients were divided into two groups: (1) patients who underwent mechanical thrombectomy using SRs alone without an IAC (SR group); and (2) patients who underwent mechanical thrombectomy using SRs with concurrent use of an IAC (SR/IAC group). Patients without recanalization (defined as TICI 0 or 1) were excluded from the analysis of time to recanalization and number of thrombectomy attempts needed to achieve recanalization. Time required for cervical artery stent and/or angioplasty was subtracted from the total time from groin arteriotomy to recanalization.
Two experienced neurointerventionalists, unblinded to the mechanical thrombectomy technique employed, reviewed the treatment angiograms and follow-up non-contrast head CT (NCCT) scans to determine thrombus location, post-thrombectomy TICI score, and the presence of post procedural SAH. Differences in image interpretation were resolved by consensus.
Mechanical thrombectomy exclusion criteria
Institutional exclusion criteria for mechanical thrombectomy during the study period were: (1) mild stroke symptoms, defined as an admission NIHSS score <6; (2) the presence of a large completed territorial infarction by NCCT, defined as an Alberta Stroke Program Early CT Score (ASPECTS) <6, or by MRI, defined as an infarction volume of >70 mL of brain; (3) functional dependence prior to stroke onset, defined as a pre-stroke modified Rankin Scale (mRS) score of ≥3; or (4) any intracranial hemorrhage. There was no strict time from last known well cut-off for exclusion.
The mechanical thrombectomy procedure was performed in a biplane neuroangiography suite (Artis Zee; Siemens, Munich, Germany) by one of three attending neurointerventional radiologists with 4–13 years of experience in neurointervention. The procedure was performed under either general anesthesia or monitored moderate conscious sedation with IV fentanyl and midazolam administered by a sedation nurse.
Statistical analysis was performed using the GraphPad Prism 7 software package for Mac (GraphPad Software, La Jolla, California, USA). The Student t-test was used to analyze continuous variables and the Fisher exact test was used to analyze categorical variables. Multiple regression analysis was also performed to determine variables predictive of the dependent variables, time to recanalization, and number of thrombectomy attempts. This was performed using the following variables: gender, age, NIHSS, administration of IV-tPA, and presence or absence of an IAC. A p value of ≤0.05 was considered statistically significant. Statistical analysis was performed using SPSS V.19.0 (IBM Corp, Armonk, New York, USA).
A total of 99 patients underwent intracranial anterior cerebral circulation thrombectomy at our institution between April 2012 and April 2016. Two patients were excluded from this cohort due to the presence of a large volume of thrombus extending from and throughout the cervical segment of the internal carotid artery (ICA) distally into the middle cerebral artery (MCA). The remaining 97 patients had an occlusion at the ICA terminus and/or MCA. SRs were used in all cases as the first line of thrombectomy. SRs were used in every thrombectomy attempt except for the following attempts used if an SR was unsuccessful: three attempts with balloon angioplasty, two attempts with a separator wire, and two attempts with aspiration alone with an IAC. A total of 69 Trevo devices (Stryker Neurovascular, Fremont, California, USA) and 68 Solitaire FR devices (Medtronic Neurovascular, Irvine, California, USA) were used. Eleven of 97 patients underwent stent and/or angioplasty at the ICA bulb, requiring an additional mean time of 33 min (range 10–50 min) prior to intracranial mechanical thrombectomy.
Twenty-nine cases (SR group) were performed consecutively prior to use of IACs, followed by 68 consecutive cases (SR/IAC group) using these catheters. There were no crossovers from one group to the other group. A 6 Fr 80 cm sheath was used for triaxial access in all cases. A Navien guide catheter (Medtronic Neurovascular, Irvine, California, USA) was used in all cases in the SR group and was positioned in the cervical or petrous segment of the ICA. IACs used in the SR/IAC group were as follows: 5 Max ACE (n=55) or ACE 64 (n=8) (Penumbra, Alameda, California, USA), ARC (n=3) (Medtronic Neurovascular, Irvine, California, USA), and Catalyst 6 (n=2) (Stryker Neurovascular, Fremont, California, USA). IACs were positioned at the site of the occlusion in 65 of 68 cases and at the level of the ophthalmic artery proximal to the occlusion in the remaining three cases.
Females comprised 47% (32/68) and 55% (16/29) of cases in the SR/IAC and SR groups, respectively. Mean age was 66±1.7 years (range 27–91) and 58±2.7 years (range 36–81) in the SR/IAC and SR groups, respectively (p=0.008), mean NIHSS scores at presentation were 18.7 and 18.2 in the SR/IAC and SR groups, respectively (p=0.50), and mean time from last known well to arteriotomy in the SR/IAC and SR groups was 278 min and 301 min, respectively (p=0.30). These data, in addition to sites of arterial occlusions, are shown in table 1.
Recanalization rates were 85% and 90% in the SR/IAC and SR groups, respectively (p=0.41). After exclusion of a total of 10 patients without recanalization, mean time from groin arteriotomy to recanalization was 50±3.6 min (range 19–136) in the SR/IAC group (n=59) and 61±6.6 min (range 28–140) in the SR group (n=28) (p=0.049). The number of thrombectomy attempts needed to achieve recanalization was 1.9±0.1 (range 1–4) in the SR/IAC group and 2.5±0.6 (range 1–6) in the SR group (p=0.009). Any post-procedural SAH was seen on imaging in 15% (10/68) and 10% (3/29) of cases in the SR/IAC and SR groups, respectively (p=0.41). No cases of SAH resulted in new neurological deficits. These data are shown in table 2.
Multiple regression analysis demonstrated that gender, age, NIHSS, and IV-tPA were not independent predictors of time to recanalization and number of thrombectomy attempts. The presence of an IAC significantly predicted a lower mean number of thrombectomy attempts (p=0.038), while the presence of an IAC did not reach statistical significance in the prediction of lower time to recanalization (p=0.091).
Early mechanical arterial revascularization in ELVO-related stroke has been correlated with improved clinical outcomes compared with treatment with IV thrombolysis alone.2 ,8 ,14–16 While correct selection of patients with ischemic stroke due to confirmed ELVO was important in recent trials, the improved efficacy of SRs in rapid and effective reperfusion was critical for demonstrating benefit of mechanical revascularization.1 ,2 ,6 ,14–16 Techniques and technical advancements such as these, leading to greater ease of recanalization in stroke thrombectomy, continue to evolve.3–5 ,9–13
To our knowledge, this is the first report comparing patients in whom SRs were used with or without the additional use of an IAC. Addition of an IAC to SR thrombectomy was associated with reductions in time to recanalization and number of thrombectomy attempts with no statistically significant increase in post-procedure SAH, although overall recanalization rates were similar in both groups. These findings suggest that the addition of IACs facilitates rapid clot extraction. This was despite the fact that the SR/IAC group was significantly older than the SR group, noting that the arteries of older patients can be more challenging and time consuming to traverse than those in younger patients. While a difference in mean time to recanalization of 11 min between our groups is relatively small in comparison to time of symptom onset to recanalization in most cases, a recent prior report demonstrated that for every 15 min acceleration to recanalization, 34 per 1000 treated patients had improved disability outcome.17
Previous technical reports have compared time to recanalization, recanalization rates, and hemorrhage rates between SRs and older retrieval devices,1 ,2 SRs with and without a balloon guide catheter,13 and IAC alone for first-line thrombectomy versus an SR/IAC combination.5 There are additionally several single-arm series describing these techniques.3 ,4 ,9–11 These reports offer a benchmark from which we can compare our results. The time to recanalization and number of thrombectomy attempts in our groups were similar to these recent prior reports.5 ,13 Similar to our SR group, Velasco et al13 demonstrated a mean procedure time of 54.8 min and a mean of 2.4 thrombectomy attempts when using SR with a non-IAC, Envoy guide catheter. Similar to our SR/IAC group, Delgado Almandoz et al5 demonstrated a mean time to recanalization of 51 min and a mean of 2.2 thrombectomy attempts when using the SR/IAC combination. Neither of these reports, however, directly compared the use of SRs with and without an IAC. One recent report did compare the SR/IAC approach to SR alone in a total of 22 in vitro models; the authors from this report found better recanalization rates for the SR/IAC approach, although the catheters used in the SR/IAC approach were positioned in the paraclinoid or terminal segment of the ICA model18 rather than directly at the clot surface itself, as in nearly all of our SR/IAC cases.
Our overall rate of reperfusion (TICI 2b/3) (87%) and the reperfusion rates within each of our groups are within the range of the recent landmark randomized controlled trials using SRs.8 ,14–16 Our reperfusion rates in each group are also within the range of several previous technical reports evaluating the approaches we describe.2–5 ,9–11 It is noteworthy that recanalization rates did not change significantly after the introduction of IACs which, we believe, speaks to the already high degree of efficacy of SR thrombectomy in acute stroke without an IAC.
The incidence of any post-procedural SAH of 15% in our SR/IAC group is similar to a recent report on using the SR and IAC combination,5 although that report described mostly symptomatic SAH attributable to the concurrent use of potent antiplatelet/antithrombotic medications or IA-tPA. No cases of SAH in our groups were symptomatic. While not statistically significant, we saw a higher percentage of SAH in the SR/IAC group than in the SR group, while the rate of SAH in an IAC alone group described by Delgado Almandoz et al5 was just 2.2%. These findings suggest validity to the already described notion that manipulation of the intracranial vasculature during mechanical thrombectomy may incite intimal damage resulting in SAH19 and, therefore, great care must be taken when considering use of any of these techniques in patients with stroke. Further investigation is warranted in this regard. Most patients in our SR/IAC group were not intubated or paralyzed for the procedure; patient motion-related vascular injury during thrombectomy is possible and also warrants further investigation.
We did not use aspiration alone for first-line thrombectomy nor did we use balloon guide catheters in the cervical ICA. Recent reports comparing IACs alone for first-line thrombectomy versus the SR/IAC approach found that the IAC alone groups had faster times to recanalization, higher rates of recanalization, lower rates of post-procedural SAH, and even better clinical outcomes at 90 days.3 ,5 A recent report on SR thrombectomy with use of balloon guide catheters found improved revascularization rates, faster times to recanalization, and higher rates of one-attempt thrombectomy compared with cases in which a balloon guide catheter was not used.13 An evolving combination of current techniques warrants further ongoing investigation.
There is currently an additional cost associated with IACs used in our SR/IAC group, and it remains uncertain if the additional cost is worthwhile at the present time. Benefits of faster times to recanalization include a greater likelihood of improved patient outcomes with less associated long-term disability.6–8 Additionally, a decrease in procedural time can help diminish operator fatigue, might result in less radiation and contrast material exposure, and can lower angiography suite costs which are calculated in minutes at our institution. These potential benefits were not evaluated in our study. Although not the topic of this paper, IAC thrombectomy alone without an SR may result in a significant cost savings and warrants further investigation.
The limitations of our study are the modest sample size, retrospective design, lack of external adjudication of angiographic findings, single-center experience, and somewhat heterogeneous patient populations in the two groups with regard to patient age. A single-center experience, however, provides some degree of uniformity with regard to procedural techniques. An additional limitation of our study is that the two groups we investigated were gathered sequentially; skill of interventionalists is known to evolve with greater experience, and this may have influenced the technical factors we report. It is possible that the operators became more facile with the devices with greater experience, thus impacting time to recanalization. A subgroup analysis of the data, however, demonstrated that mean time to recanalization and mean number of thrombectomy attempts were both greater in the more recent half of the SR/IAC group than in the older half. Both halves of the SR/IAC group independently had shorter mean time to recanalization and mean number of thrombectomy attempts compared with the SR alone group. Intraluminal dwell time for SRs after stent deployment is another factor that affects overall procedural time and was not studied here, although we believe anecdotally that our dwell times of approximately 5 min did not differ between the SR/IAC group and the SR alone group.
Our results demonstrate that use of an IAC in SR stroke thrombectomy for ELVO improved the time to recanalization and reduced the number of thrombectomy attempts needed to achieve recanalization compared with use of a SR alone. Although our study suggests that IACs should be considered for anterior cerebral circulation thrombectomy when using a SR, there are, nevertheless, further questions to be investigated, such as the risk of intracranial vascular injury with the use of these devices. Further prospective studies at high-volume centers with experienced external adjudication will help provide answers to questions regarding the most effective and safest techniques for stroke thrombectomy.
Contributors All authors provided substantial contributions to the conception and design of the work and in drafting the work or revising it and in preparing the final manuscript for submission. All authors are agreeable to be accountable for all aspects of the work, including its accuracy or integrity. GJ, YS, EL, VG, and DG were involved in the acquisition and analysis of the data for the work. YS and GJ were involved with the statistical analysis of the data.
Competing interests GJ has research related grants funded by Stryker Neurovascular, Medtronic, Microvention, and Codman Neurovascular. TM has research related grants funded by Stryker Neurovascular and Medtronic. DG has research related grants funded by Stryker Neurovascular and Medtronic.
Ethics approval Ethics approval was obtained from the University of Maryland Medical Center Institutional Review Board.
Provenance and peer review Not commissioned; externally peer reviewed.
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.