Background Over half of patients who receive intravenous tissue plasminogen activator for middle cerebral artery division (MCA-M2) occlusion do not recanalize, leaving a large percentage of patients who may need mechanical thrombectomy (MT). However, the outcomes of MT for M2 occlusion have not been well characterized.
Objective To determine if MT of M2 occlusion is as safe and efficacious as current standard-of-care MT for M1 occlusions.
Methods With institutional review board approval, we retrospectively reviewed records of 212 patients undergoing MT for isolated MCA M1 or M2 occlusions during a 36-month period (Sept 2013 to Sept 2016) at two centres. Treatment variables, clinical outcomes, and complications in each group were recorded.
Results There were 153 M1 MCA occlusions and 59 M2 MCA occlusions. No statistically significant difference was found in the rate of mortality (20% in M1 vs 13.6% in M2, p=0.32), excellent (34.5% vs 37.3%, p=0.75) or good (51% vs 55.9%, p=0.54) clinical outcomes between the two groups. Infarct volumes (48.4 mL vs 46.2 mL, p=0.62) were comparable between the two groups, as were the rates of hemorrhagic (3.3% vs 3.4%, p=1.0) and procedural complications (3.3% vs 5.1%, p=0.69).
Conclusion Our data on MT targeting M2 occlusions demonstrates reasonable safety and functional outcomes. Further randomized clinical trials are needed to clarify which patients may benefit from MT for M2 occlusions.
- device thrombolysis
Statistics from Altmetric.com
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.
Mechanical thrombectomy (MT) is the standard-of-care treatment for patients with ischemic stroke (IS) with proximal large intracranial vessel occlusion within 6 hours from onset of symptoms.1–5 Most patients who were enrolled in the recent randomized MT trials had proximal large vessel occlusions, with some studies excluding patients with distal circulation occlusion such as M2 divisions of the middle cerebral artery (MCA). Smaller distal vessels can be more challenging to access when sharp angles and tortuosity are encountered. This can affect recanalization efficiency and outcome, and may theoretically result in a higher risk of complications, including vessel dissection and perforation. However, the availability of newer generation MT devices and techniques has allowed operators to navigate more technically difficult procedures.
M2 occlusions have a higher rate of revascularization with IV tissue plasminogen activator (tPA) alone as compared with more proximal occlusions.6–8 However, more than half of those who receive tPA for a M2 occlusion still do not achieve adequate recanalization.9 Furthermore, a substantial number of patients remain ineligible for tPA therapy and are left with few treatment options. These circumstances create a subgroup of patients who may benefit from endovascular therapy when tPA is either ineffective or not recommended10. M2 IS may have a low growth rate of infarct volume and smaller final infarct volumes than found with main MCA trunk infarctions, 11 but can be just as disabling, particularly if language centres are involved. The ideal selection criteria for MT in patients with M2 occlusions remain controversial, which leads to heterogeneous practice patterns among stroke specialists. The purpose of this study was to compare treatment variables, clinical and technical outcomes, and complications of patients undergoing thrombectomy using second-generation stentrievers, manual aspiration, or a combination of both for MCA M1 vs M2 occlusions.
With institutional review board approval, we retrospectively collected data on all patients who underwent MT between September 2013 and September 2016 at two tertiary care centres in Ohio, USA. Baseline demographics (age, sex, glycohemoglobin, smoking status, history of stroke or transient ischemic attack, atrial fibrillation, hyperlipidemia), treatment times (onset to door, door to groin, duration of procedure, onset to recanalization), recanalization grade and rates, procedural complications, National Institute of Health Stroke Scale (NIHSS) at admission and discharge, and modified Rankin Scale (mRS) at baseline and at 3 months were recorded. Good recanalization was defined as Thrombolysis in Cerebral Infarction score ≥2b. NIHSS scores were assessed by certified practitioners at admission and discharge. Baseline mRS score was obtained on admission. Ninety-day mRS scores were assessed by stroke neurologists or nurse practitioners during a clinic visit; good clinical outcome was defined as mRS <2 and excellent clinical outcome as mRS <1.
Improvement in NIHSS (≥4 points) and dramatic improvement in NIHSS (≥8 points) were calculated based on admission and discharge NIHSS. Change in mRS score was calculated based on the difference between 3-month mRS scores and mRS score before stroke. Patients who underwent MT with first-generation devices or wire manipulation were excluded.
Hospital protocol and inclusion criteria
Patients over 18 years of age were included in the study if they presented with symptoms of acute IS attributed to an isolated M1 or M2 MCA occlusion confirmed on digital subtraction angiography. The M2 MCA artery was defined as occlusion beyond the bifurcation of the M1 artery and extending to the apex of the circular sulcus.
Patients who presented within 6 hours from last seen normal were selected for MT if they had an ASPECT (Alberta Stroke Program Early CT) score ≥7, confirmed occlusion on CT angiography, and NIHSS score ≥5. Age, comorbidity status, baseline functional status, and poor vascular anatomy were not considered exclusion criteria for MT. Patients with unclear time of onset or time of onset >6 hours were selected for MT if a favourable CT perfusion pattern (large perfusion deficit with a small core infarct determined visually) was identified. Patients with large infarct burden on presentation or rapidly resolving symptoms with or without IV tPA did not undergo MT and were managed conservatively. Patients with concurrent internal carotid artery (ICA) occlusion who underwent MT were excluded from this analysis.
Routine clinical care included a CT brain scan and CT angiography. After MT, patients were monitored in the neurointensive care unit for complications and had a CT brain scan at 24 hours to evaluate for intracerebral hemorrhage. Infarct volumes were calculated using diffusion weighted imaging performed within 48 hours of MT using computer-assisted volumetric analysis of infarct regions (iPlan Net, version 3.6; Brainlab AG, Germany). Boundaries of the infarct zone were determined visually by one of the authors (DES).
Angiographic images were reviewed for confirmation of accurate classification of occlusions by the performing neurointerventionalist (MJ or SFZ) and a neuroradiologist (MB). CT brain images were reviewed for hemorrhage and confirmation of ASPECT score by an author (HS) and compared with reports from the neuroradiologist and neurointerventionalist, respectively. Symptomatic intracranial hemorrhage was defined as hemorrhagic conversion with an increase of ≥4 points on the NIHSS.
Mechanical thrombectomy was performed using a second-generation stentriever (TREVO or Solitaire) with focal aspiration or direct aspiration with a Penumbra catheter. The MT technique was left to the discretion of the neurointerventionalist. Procedures were performed with conscious sedation or under general anesthesia when patients did not follow instructions or had severe respiratory distress.
Data was retrospectively identified in a prospectively collected MT database in a Microsoft Excel file and exported to a statistical analysis software "R: a language and environment for statistical computing; EZR version 1.32." Continuous variables were compared using the Student t test with Welch test or the Mann–Whitney test, and categorical data with the Fisher exact test. Patients lost to follow-up were excluded from the final analysis. A multivariate logistic regression was performed to evaluate clinically significant predictors of favourable outcome. A p value of ≤0.05 was considered statistically significant.
We analysed 212 patients during a 36-month period (Sept 2013 to Sept 2016) at one primary stroke centre and one comprehensive stroke centre. There were 153 isolated M1 MCA occlusions and 59 M2 MCA occlusions. Baseline characteristics and selection methods are summarized in table 1.
Patients with M2 occlusions were more likely to receive IV tPA (37.9% in M1 group versus 54.2% in the M2 group, p=0.043). Within the M2 group, superior division occlusions were more common (62.7% vs 37.3%) (see table 2). Time metrics between the two groups were comparable.
Of 212 patients, eight (3.8%) patients did not return to the stroke clinic and were lost to follow-up. There was no statistically significant difference in the rate of mortality, excellent or good clinical outcomes between the two groups (see table 3 and figure 1).
Mean discharge NIHSS (M1: 6.6±6.6 vs M2: 6.5±6.8, p=0.947), improvement in NIHSS of ≥4 points (M1: 73.9% vs M2: 76.3%, p=0.722), and dramatic improvement in NIHSS (M1: 60.1% vs M2: 50.8%, p=0.275) were similar between the two groups. There was a significant difference between the two groups in change from baseline mRS (median change 2.9 in M1 vs 2 in M2; p=0.003).
Infarct volumes were comparable between the two groups (48.4 mL vs 46.2 mL, p=0.62), as was the rate of hemorrhagic complications (3.3% vs 3.4%, p=1.0). There were no instances of periprocedural vessel dissection in the M2 group, and two (1.3%) cases of flow-limiting dissection in the M1 group. There were three cases of subarachnoid hemorrhage in the M1 (2.0%) and M2 group (5.1%) each (see table 3).
We performed a multivariate logistic analysis and found that ASPECT score (OR=1.920, CI 1.32 to 2.81, p=0.007), NIHSS (OR=0.904, CI 0.85 to 0.96, p=0.001), age (OR=0.965, CI 0.94 to 0.99, p=0.007), and procedure time (OR=0.987, CI 0.976 to 0.999, p=0.039) were the best predictors of a good clinical outcome (mRS <2) in the combined cohort. Predictors of a good clinical outcome in patients with M1 occlusions who recanalized included ASPECT score (OR=2.55, CI 1.31 to 4.95, p=0.006), smoking status (OR=4.92 CI 1.34 to 18.1, p=0.017), NIHSS (OR=0.899, CI 0.825 to 0.980, p=0.015), and use of intravenous tPA (OR=0.277, CI 0.084 to 0.913, p=0.035). In patients with M2 occlusions and successful recanalization, ASPECT score (OR=2.11, CI 0.93 to 4.75, p=0.073) and NIHSS (OR=0.833, CI 0.73 to 0.95, p=0.006) were also predictors of a good clinical outcome.
Our study demonstrates that clinical outcomes, radiographic outcomes, and rates of procedural and hemorrhagic complications are comparable between patients with M1 and M2 occlusions who were selected by similar criteria and treated with the same endovascular techniques with second-generation MT devices. Similar clinical outcomes have been shown in multiple other studies12–16 and a post hoc analysis17 of the STAR, SWIFT, and SWIFT PRIME studies (See online supplementary table 1).
Supplementary data 1
Recanalization rates for both groups in our study were comparable and similar to those of other studies.12 13 18 A larger number of patients had superior division MCA branch occlusions than inferior branch occlusions (63% vs 37%), although a specific division of the M2 was not a predictor of good clinical outcome. The rate of IV tPA administration was higher in the M2 group, although this probably did not affect final clinical outcomes given that a recent post hoc analysis of the SWIFT and STAR studies found no benefit of IV tPA in patients undergoing MT.19 The increased use of direct manual aspiration in the M2 group was unlikely to have affected clinical outcomes, given that preliminary results of the ASTER trial (Interest of Direct Aspiration First Pass Technique (ADAPT) for Thrombectomy Revascularization of Large Vessel Occlusion in Acute Ischemic Stroke) showed comparable recanalization rates to MT with stentrievers.20
Patients with M2 occlusions had a significantly smaller change from baseline mRS score than those with M1 occlusions, although this did not translate into a significant difference in excellent or good clinical outcomes between the two groups. The higher rate of dramatic improvement in NIHSS score in the M1 group may indicate that the treatment effect for M2 occlusions is inferior to that for M1 occlusions.
The rate of hemorrhagic complications in our study was similar to that found in previously published studies.12 18 Shi et al 15 described a lower rate of intracranial hemorrhage in M2 occlusions than in M1 occlusions, which was not reproduced in our cohort. We had no incidences of vessel dissection in patients with M2 occlusions as seen in the study performed by Sarraj et al.18 Data from our cohort are consistent with data published by Dorn et al,13 who noted that MT for M2 occlusions does not pose an increased risk of complications as compared with MT for M1 occlusions.
Large cohorts describing clinical outcomes of patients with IS due to M2 occlusion are lacking. Untreated M2 occlusions can lead to moderate to severe disability in up to 50% of patients at discharge,11 and failure to treat can lead to similar rates of mortality at 6 months as compared with M1 occlusions.21 The site of occlusion was also not a predictor of clinical outcome in our cohort, consistent with other studies.12 14 16 17 This could be attributed to the comparable infarct volumes or involvement of eloquent areas supplied by M2 vessels.11
Revascularization may become futile at a certain distance in the anterior circulation. Prior reports showed minimal infarct growth and no association between recanalization and functional outcome in patients with diffusion–perfusion mismatches due to distal M3 and M4 occlusions.22 However, a post hoc analysis of the small sample of patients with M2 occlusion from the recent five randomized clinical trials concluded that MT may be a reasonable treatment option for this group.14
Our study had multiple limitations, including the retrospective design and the biases inherent to observational studies. As a non-randomized study, there may be an inherent selection bias of treating patients with M2 occlusions with only severe initial presentations, left hemispheric syndromes, or favourable anatomy. Furthermore, we did not compare our cohort of patients with M2 occlusions with patients who were treated conservatively. Lastly, we did not account for a difference in stroke etiologies resulting in M2 occlusions, which may have an effect in the ease of revascularization.
MT for M2 occlusions demonstrates reasonable safety, recanalization rates, complications, and functional outcomes. Further randomized clinical trials will clarify which patients may benefit from MT for M2 occlusions.
Contributors HS made substantial contributions to the conception and design of the work, the acquisition, analysis, and interpretation of data, and drafted the research. GR, DES, JS made substantial contributions to the acquisition, analysis, and interpretation of data. MB made substantial contributions to the acquisition, analysis, and interpretation of data, and revised it critically for important intellectual content. SFZ made substantial contributions by revising the research critically for important intellectual content. MJ made substantial contributions to the conception and design of the work and revised it critically for important intellectual content.
Funding This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Ethics approval Hospital institutional review board.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Unpublished data are available upon request from the corresponding author.