Author + information
- Received September 20, 2018
- Revision received November 19, 2018
- Accepted November 27, 2018
- Published online February 25, 2019.
- Ali Alawieh, PhDa,b,
- Jan Vargas, MDb,
- Kyle M. Fargen, MDc,
- E. Farris Langley, BSa,
- Robert M. Starke, MDd,
- Reade De Leacy, MDe,
- Rano Chatterjee, MDf,
- Ansaar Rai, MDg,
- Travis Dumont, MDh,
- Peter Kan, MDi,
- David McCarthy, MDd,
- Fábio A. Nascimento, MDj,
- Jasmeet Singh, MDc,
- Lukas Vilella, MDc,
- Aquilla Turk, DOb and
- Alejandro M. Spiotta, MDb,∗ (, )@alex_spiotta
- aMedical Scientist Training Program, Medical University of South Carolina, Charleston, South Carolina
- bDepartment of Neurosurgery, Medical University of South Carolina, Charleston, South Carolina
- cDepartment of Neurological Surgery, Wake Forest University, Winston-Salem, North Carolina
- dDepartment of Neurological Surgery and Neuroradiology, University of Miami, Miami, Florida
- eDepartment of Neurosurgery, Mount Sinai Health System, New York, New York
- fDepartment of Radiology and Radiological Sciences, Medical University of South Carolina, Charleston, South Carolina
- gDepartments of Radiology, Neurology & Neurosurgery, West Virginia University, Morgantown, West Virginia
- hDepartment of Neurosurgery, University of Arizona, Tucson, Arizona
- iDepartment of Neurosurgery, Baylor College of Medicine, Houston, Texas
- jDepartment of Neurology, Baylor College of Medicine, Houston, Texas
- ↵∗Address for correspondence:
Dr. Alejandro Spiotta, Medical University of South Carolina, Department of Neurosurgery, 171 Ashley Avenue, Charleston, South Carolina 29425.
Background Endovascular thrombectomy (ET) for acute ischemic stroke is the current standard of care. Although successful ET has high efficacy in improving functional outcomes, the decision to abort a long procedure remains a challenge. Longer procedure time (PT) has been associated with lower rates of functional independence.
Objectives The objective of this study was to evaluate the impact of PT on outcomes and complications after ET using different techniques at a multicenter level and to define the risk of procedure extension in different patient cohorts.
Methods Patients undergoing ET with a stent retriever (SR) or a direct aspiration at first pass technique at 7 U.S. centers between June 2013 and February 2018 were reviewed from prospectively maintained databases that include baseline variables and technical and clinical outcomes. Multivariate analyses were used to assess impact of PT on 90-day modified Rankin scores, successful recanalization, post-procedural symptomatic hemorrhage (sICH), and complications.
Results The study included 1,359 patients and demonstrated a decreased likelihood of good functional outcomes (modified Rankin score 0 to 2) when PT extended beyond 30 min (p < 0.01). Rates of sICH and complications increased exponentially with PT (doubling rates of 26 and 50 min, respectively). The cumulative rate of successful recanalization and good outcomes plateaued after 60 min of PT. In patients with PT >30 min, fewer attempts predicted the success of ET and good outcomes (p < 0.01). Successful recanalization was achieved faster with the direct aspiration at first pass technique than in SR. The direct aspiration technique was more sensitive to PT than SR, and posterior stroke was more sensitive to PT than anterior stroke.
Conclusions Longer ET procedures lead to lower rates of functional independence and higher rates of sICH and complications. Exceeding 60 min or 3 attempts should trigger careful assessment of futility and risks of continuing the procedure.
Endovascular thrombectomy (ET) for acute ischemic stroke (AIS) with large-vessel occlusion (LVO) is the current standard of care (1–6). ET has been shown to significantly impact functional outcomes compared with medical management, and recent data have extended the treatment window to 24 h after last known normal (5,7). Although most of the data support the use of stent retrievers (SRs), aspiration-based thrombectomy (ADAPT) has recently been shown to be a fast and effective method for achieving recanalization (8–11).
Prior studies have shown that for mechanical thrombectomy, the procedural time has a profound impact on patient outcomes. Extended procedures resulted in increased incidence of complications, increased cost, and worse outcomes (10,11). This study represents a large, retrospective, multicenter investigation of the relationships between procedural time, number of recanalization attempts, and outcomes across centers that perform either ADAPT or SR thrombectomy as frontline therapy.
Patients undergoing ET for AIS at 7 U.S.-based comprehensive stroke centers between January 2013 and March 2018 were included (12). Data were reviewed from prospectively maintained databases of consecutive patients treated with ET for AIS with LVO. All patients 18 years of age or older were included in this study. No patients were excluded based on National Institutes of Health Stroke Scale (NIHSS) on presentation, administration of intravenous tissue-type plasminogen activator (t-PA), time from onset to groin, or ASPECT (Alberta Stroke Program Early CT) scores. The choice to administer intravenous t-PA did not affect and was not affected by the decision to undergo ET, and intravenous t-PA was administered based on current guidelines. Both aspiration thrombectomy using ADAPT and SR thrombectomy were acceptable frontline ET approaches. Data from each site were curated and analyzed at the Medical University of South Carolina. The institutional review board at each site approved this study.
Demographic information and admission stroke severity scores were retrieved by review of the patient charts. Pre-operative computed tomography or computed tomography angiography was used to assess location and to determine ASPECT scores for patients with anterior circulation strokes. Procedure notes were reviewed for technical variables, technical outcomes, and complications. Procedure time was defined as the time from groin puncture to first recanalization (TICI [Thrombolysis In Cardiac Ischemia] 2B or more) or to abortion of procedure if TICI 2B was not achieved.
The primary clinical outcome was defined as the 90-day modified Rankin score (mRS) recorded during a follow-up visit to a stroke neurologist at 90 ± 14 days after stroke or via telephone call for patients discharged to nursing homes or hospice or to confirm mortality. At both discharge and the follow-up visit, the NIHSS scores were also collected.
Statistical analyses were performed using SPSS version 24 (IBM Corporation, New York, New York) or GraphPad Prism 6 (GraphPad, La Jolla, California). Univariate analysis was first compared with the Student’s t test or analysis of variance (for parametric variables), Mann-Whitney test or Kruskal-Wallis test (for nonparametric variables), or chi-square test (for categorical variables). Bonferroni test was used for multiple comparisons, and likelihood ratios were used for multiple comparisons in categorical measures. A 2-sided α < 0.05 was considered significant. For baseline variables that were not available in a subset of patients, including race (10% missing), onset-to-groin time (15% missing), sex (7% missing), admission NIHSS score (5% missing), and baseline mRS (12% missing), imputations were performed to avoid bias by excluding these patients from multivariate analysis. Multiple imputation was used for patients missing up to a maximum of 3 baseline variables. Independent multivariate mixed regression models were used for rates of good outcome (logistic), symptomatic intracerebral hemorrhage (sICH) (logistic), TICI score (ordinal), and procedure time (linear), with centers as random variables. In one instance, and as noted in the Results, centers were used as categorical fixed variables, to assess for impact of centers on the reported outcomes. All multivariate models were assessed using the Hosmer-Lemeshow test and C-statistics. Variables were included in the models based on statistical significance on univariate analysis or clinical relevance. Power analysis was performed based on anticipated difference in rates of good outcome in rapid recanalization (<30 min) compared with intermediate (30 to 60 min) and delayed (>60 min) recanalization. We anticipated ≈50% rates of good outcome in the rapid group and a drop of 10% in the intermediate and delayed groups. We also anticipated differential allocation across the 3 groups, with more patients likely to fall in the rapid group (50% of the entire cohort). Using an alpha of 0.017 (correcting for 3 comparisons), a total sample size of ≈1,250 patients would provide 86% power to detect a significant difference. Cumulative distribution curves and curves representing change in rates of good outcome with procedure time were performed using GraphPad Prism 6, and the best-fitting linear and nonlinear model was constructed. Models were assessed using the R2 correlation metric, and the Runs test for deviance of model from actual values.
Of 1,448 patients who underwent ET for AIS, 1,357 were included in this study, and 91 were excluded because of loss to follow-up (n = 72), abortion of thrombectomy before access to intracranial vessels (n = 11), or undocumented procedural variables (n = 25). The mean age was 67 ± 15 years, and 51% were females. Intravenous t-PA was administered in 46% of cases, and 12% of patients presented with posterior circulation stroke.
Procedure time and outcomes
We first assessed the rates of good outcomes, procedural complications, and post-procedural sICH after ET as a function of increasing procedure time, defined as the time from groin puncture to successful recanalization or last failed attempt. As shown in Figures 1A to 1C, increasing procedure time is associated with decreasing rates of good outcomes (mRS 0 to 2) at 90 days. A significant change was observed after 30 min of procedure time, when the rate of good outcomes dropped by 40%. Interpolation of curves representing procedure time versus rates of good or poor outcomes at 10-min or 5-min intervals of procedure time (Figures 1B and 1C, respectively) fit a sigmoidal curve with an inflection point between 30 and 35 min of procedure time (R2 = 0.81, Runs test: p > 0.80). After 60 min of procedure time, the interpolated sigmoidal curve plateaued at an ≈70% risk of poor outcomes. In contrast to procedure time, the association with the number of recanalization attempts and rates of good (or poor) outcomes did not exhibit an inflection point but showed a linear association between 1 and 10 attempts (Figure 1D) (R2 = 0.6; Runs test: p > 0.60).
In parallel to the reduced rates of good outcomes, both complication and sICH rates increased with increasing procedure time (Figure 1A). The rates of sICH and complication across procedure time followed an exponential growth curve, with a doubling time of 26 min for sICH and 50 min for complications (Figures 1E and 1F). This means that for every additional 26 min of procedure time, the risk of post-procedural sICH doubles, whereas complication rates double with 50-min increments of procedure time. On the basis of these findings, we categorized our patient cohort into 3 groups based on procedure time to <30 min (rapid), 30 to 60 min (intermediate), and >60 min (extended).
Technical and clinical outcomes across procedure time groups
When patients were trichotomized based on procedure times, there were no significant differences in major baseline and admission variables between the 3 groups (Table 1) except for a lower rate of prior stroke history in patients in the >60-min group. There was a significant difference in vessel distribution among the 3 groups, with significantly higher rates of M2 or M3 occlusion in the <30-min group compared with the >60-min group and lower rates of tandem occlusions in the <30-min group than in either the 30- to 60-min or >60-min group (Table 1).
The average procedure time in the rapid group was 18 min compared with 44 min and 98 min in the intermediate and extended groups, respectively (Table 2). As shown in Table 2, the extended and intermediate groups had a significantly higher number of attempts, higher rates of intra-arterial t-PA, and lower rates of TICI 2C or 3 scores than the rapid group. A similar pattern, which also reached significance but was of lower magnitude, was observed between the extended and intermediate groups. There were significantly higher rates of using SR as the first-pass technique compared with ADAPT with the group with increasing procedure time (Table 2). Complication rates were higher in the extended versus rapid or intermediate groups and comparable between the intermediate and rapid groups (Table 2).
Clinical outcomes showed significantly improved 90-day mRS scores in the rapid group compared with the intermediate or extended groups and in the intermediate versus extended group. A similar difference was observed in rates of functional independence (mRS 0 to 2) at 90 days (45% vs. 33% vs. 27% in rapid vs. intermediate vs. extended groups, respectively) and rates of mRS 0 or 1 scores. Mortality at 90 days was significantly higher in the intermediate (22%) and extended (39%) groups than in the rapid group (17%) and was also significantly different between the extended and intermediate groups (Table 2). Rates of sICH and parenchymal hemorrhage type 2 (PH2) were significantly higher in the extended and intermediate groups compared with the rapid group (2.3%) but were comparable between the extended and intermediate groups (6%).
Rapid recanalization predicts 90-day outcomes
We used mixed logistic regression models for 90-day good outcomes (mRS 0 to 2), controlling for demographic and procedural variables to show that rapid recanalization (procedure time <30 min) was an independent predictor of good outcome at 90 days (odds ratio [OR]: 1.56; p = 0.004) (Table 3). This effect was present despite controlling for the first-pass effect or the number of attempts, each of which was a predictor of good outcomes in 2 independent models (Table 3). Additional predictors of good outcomes were younger age, male sex, lower pre-stroke mRS and admission NIHSS scores, anterior location of stroke, and absence of complications (Table 3). The effect of rapid recanalization was notably conserved despite adjusting for the presence of tandem occlusion or the identity of the occluded vessel (Table 3).
Because final TICI scores were significantly different across the procedure time groups, we repeated the model, limiting the population to patients with at least TICI 2B or TICI 3, which showed a similar effect of rapid recanalization (<30 min) in predicting favorable outcomes (Table 4). Similarly, the same effect for rapid recanalization was observed when patients were split into groups with anterior circulation or posterior circulation strokes (Table 4). Notably, the impact of rapid recanalization on rates of good outcome was significantly higher in patients with posterior circulation stroke (OR: 5.61; 95% confidence interval: 1.7 to 15.8) compared with anterior circulation stroke (OR: 1.72; 95% confidence interval: 1.28 to 2.30). The effect of procedure time on outcome was maintained despite controlling for ASPECT scores in patients with anterior circulation stroke (Online Table 3). Because there were significantly more patients receiving intra-arterial t-PA therapy in the groups with intermediate and extended procedure time than in the rapid recanalization group (Table 2), we also assessed the effect of procedure time <30 min on outcomes in the subset of patients who did not receive intra-arterial t-PA therapy (Table 4), which showed a similar OR of 1.87 (p < 0.001). When the thrombectomy research center was used as a fixed variable rather than a random effect, the effect of procedure time on outcome was still significant (Table 4).
Predictors of post-procedural hemorrhage
Multivariate logistic regression analysis for predictors of sICH or PH2 after ET demonstrate rapid recanalization (<30 min) was an independent predictor of lower sICH/PH2 rates (OR: 0.47; p = 0.03). Patients with diabetes mellitus and older patients were also at higher risk of sICH (Table 5). Patients with sICH had significantly higher mRS scores at 90 days than patients without sICH (Online Figure 1) (p < 0.0001).
In contrast to the exponential growth of rates of sICH with procedure time, the cumulative rates of good outcome achieved followed a hyperbolic curve, with 67% of the cumulative good outcome rate achieved within 30 min, when the rates of sICH exhibited a steep exponential growth. Between 30 and 60 min, the cumulative rate of good outcome increased to 84% with doubling of the rate of sICH. After 60 min, for every 5% increase in cumulative rate of good outcome, the rate of sICH increased by 200% (Figure 2A). Notably, when individual vessels were analyzed, a similar hyperbolic curve (with 67% of the cumulative good outcome rate achieved within 30 min) occurred in all vessels, except in patients with tandem occlusions, in whom 67% of the good outcome rate was achieved within 60 min of procedure time (Figure 2B).
Procedure time and successful recanalization
The cumulative percentage of successful recanalization (TICI 2B+) over increasing procedure times followed a hyperbolic curve that mimicked that for rates of good outcome (Figure 2C). By 30 min of procedure time, 65% of %TICI2B+max was achieved, and the increase in success rate slowed beyond this point, which indicates diminished returns on success of recanalization. A similar pattern was observed with increasing number of attempts (Figure 2D), wherein 67% of success occurred within the first 3 attempts.
Using a linear regression model for predicting procedure time, procedure times were shorter in patients with lower pre-stroke mRS, anterior circulation occlusion (vs. posterior), in the absence of tandem occlusions, and when aspiration was used as a frontline approach (Table 6). When a similar model was applied only to patients with anterior circulation strokes, while adding ASPECT scores as a covariate, a lower pre-stroke mRS, absence of tandem occlusion, and the use of aspiration as a frontline approach were the only significant predictors of procedure time (Table 6).
Procedure times, attempts, and outcomes
We then split the patients by the procedure time groups as above and performed independent mixed logistic regression analysis for rates of good outcome at 90 days in each group. In patients with procedure time <30 min or 30 to 60 min, fewer attempts but not shorter procedure time were an independent predictor of good outcome (Table 7), which indicates that in addition to fast recanalization, the use of few recanalization steps was an independent predictor of outcomes in the rapid and intermediate recanalization groups. However, in patients with >60 min of procedure time, 1-min increments of procedure time were associated with lower odds of good outcome at 90 days (Table 7).
By selecting patients who had procedure times beyond 30 min, we performed logistic regression analysis for predictors of success of recanalization beyond 30 min (Table 8), and in this analysis, a lower number of attempts was an independent predictor of successful recanalization and good outcome. This indicates that when procedure time exceeds 30 min, patients who have fewer recanalization attempts made have a greater likelihood of successful recanalization and good outcome.
Procedure time and outcomes with aspiration versus SR
Because aspiration at first pass was an independent predictor of shorter procedure times, we performed univariate analysis on patients split by frontline thrombectomy technique (ADAPT vs. SR). Despite similar baseline variables and ASPECT scores, patients with SR had significantly longer procedure times than those treated by ADAPT (59 min vs. 35 min; p < 0.001) (Table 9). Patients with SR also had a lower rate of complete recanalization (TICI 2C/3) and higher mortality rate, although ADAPT was not an independent predictor of mortality on multivariate analysis (Table 9).
Although shorter procedure time was an independent predictor of good outcome in both cohorts, procedure time within 30 min was an independent predictor of good outcome only in the aspiration cohort (Table 10). Consistent with this finding, curves representing the change in rate of good outcome over procedure time demonstrated a sigmoidal pattern in the aspiration and SR cohorts; however, the inflection point was at 30 min with aspiration versus 60 min with SR (Figures 3A and 3B). Mapping the cumulative percentage of recanalization success or good outcome across procedure time showed that SR thrombectomy achieved comparable recanalization and good outcome rates at slower rates in function of procedure times (Figures 3C and 3D).
In recent years, mechanical thrombectomy has been proven as a viable treatment strategy for patients with AIS, with treatment windows extended to 24 hours (1–6,13–15). With the advent of newer techniques and devices, as well as inclusion of more patients, there is a need for optimizing patient selection and delineating the limits of interventions (16), so that patients who will achieve the most benefit can undergo the appropriate treatment with the least possible exposure to harm. This study represents a large, multicenter, retrospective review of modern mechanical thrombectomy techniques, with the goal of defining the impact of procedural variables, procedure time, and number of attempts on the success of treatment, functional outcomes, and risks of complications and hemorrhage (Central Illustration).
The effect of procedure time on patient outcomes has been described previously, with longer times associated with worse outcomes. Previous work has shown that extending the duration of mechanical thrombectomy beyond 60 min, and more recently beyond 35 min, not only decreases the chance of an mRS of 0 to 2 at 90 days but also increases the chance of post-procedural PH2 (10,11). This study supports those findings, demonstrating a significant change in the rate of good outcome between 30 and 35 min of procedure time, with a secondary plateau after 60 min. Importantly, these results were independent of treatment center and were mirrored by an exponential increase in both the complication rate and sICH (defined as PH2), with doubling times of 50 min and 26 min for complication rate and sICH, respectively. Interestingly, we demonstrated that the rate of successful recanalization stabilized after 60 min (or 3 attempts), with a minimal increase in the rate of successful recanalization (<5%) for each doubling of the rate of sICH and complications.
Procedure time and number of attempts
One explanation for the observed effect of procedure time on patient outcomes is that patients with longer procedure times might have occlusions that are difficult to recanalize, and the longer procedure time represents multiple attempts to achieve a good TICI score, with a subsequent increase in risk of intraprocedural and post-procedural complications associated with manipulating the intracranial vasculature. Notably, although tandem occlusions, which are well known to require longer procedure time, were associated with longer procedures in our cohort, this did not explain the significant negative impact of longer procedures on outcomes. Regression models correcting for tandem occlusions still demonstrated significantly lower odds of good outcomes in patients with longer procedure times.
Each thrombectomy pass represents an independent interaction with the vessel with a finite complication rate, so that multiple attempts begin to cumulatively add to the risk profile. Importantly, the number of attempts demonstrated a linear association with rates of good outcome and increased risk of hemorrhage, which is consistent with recent reports in patients undergoing ADAPT or SR thrombectomy (11,17,18). However, there could be a subgroup of occlusions that are relatively easy to recanalize but have challenging proximal anatomy (i.e., type III aortic arch or severe cervical tortuosity), in which gaining access to the level of occlusion can consume valuable time. In this study, the subgroup of patients with longer procedure times (between 30 and 60 min) and fewer attempts at thrombectomy (0 to 2 attempts) most likely represent the latter group, in whom intracranial access proved time consuming but the actual thrombectomy was straightforward and demonstrated a higher likelihood of an mRS of 0 to 2 at 90 days. Thus, the more salient time metric might be the time to recanalization once access is achieved. Importantly, this trend was not observed in patients with a procedure time >60 min, and procedure time, at 1-min increments, was associated with lower odds of good outcome in this group. This suggests that with aspiration, procedures lasting >60 min should be aborted regardless of number of attempts (16).
Aspiration versus SR
Direct aspiration as a first-pass technique was associated with a significantly faster procedural time (35 min for aspiration vs. 59 min for stent retriever; p < 0.001), and although this did not translate into a significant difference in good outcomes at 90 days, there was a significantly higher mortality rate in the SR group (21% for aspiration vs. 28% for SR; p < 0.01). With respect to procedural time, ADAPT seemed to be more sensitive to treatment time, with a higher likelihood of good outcome at 90 days with procedure times <30 min. In fact, the likelihood of an mRS of 0 to 2 at 90 days dropped significantly and then plateaued beyond 30 min in patients who underwent aspiration first. In contrast, although there was a drop-off at 60 min for patients undergoing SR thrombectomy, these patients achieved comparable recanalization rates and good outcomes at slower rates. This again highlights the idea that the time to gain access to the occlusion may not be as critical as the amount of time spent attempting recanalization once access is achieved. Performance of aspiration-based thrombectomy is associated with faster intracranial access times than SR thrombectomy, which could explain why the aspiration cohort was more sensitive to procedure times <30 min.
Posterior circulation thrombectomy
Patients with posterior circulation occlusions tend to present with lower NIHSS scores, with a reported 21% risk of poor outcome and a mortality of 3.6% to 5.9%. There is mounting evidence that these patients might receive similar benefits from thrombectomy as those with LVOs of the anterior circulation (19–25). This study demonstrates that although the procedural time for a posterior fossa mechanical thrombectomy is longer, the impact of rapid recanalization on rates of favorable outcomes in these cases is much higher than for anterior circulation (OR of 5.61 for an mRS of 0 to 2 at 90 days in posterior circulation mechanical thrombectomy versus OR of 1.72 for anterior circulation thrombectomy). This suggests that posterior circulation occlusions, although they can be more challenging, are also more time sensitive.
This study represents a retrospective review of patients undergoing ET for AIS at multiple centers. Different selection criteria (onset to groin threshold, age, admission NIHSS score) may have been used; however, we have included these as variables in the study, and centers were used as random variables or fixed variables on multivariate analyses. In addition, data from different comprehensive stroke centers with different frontline ET techniques (aspiration vs. SR) were included; however, inclusion of this diverse approach allowed for subgroup analyses to compare both techniques and to assess the robustness of the findings across thrombectomy techniques. Because of differences in imaging protocols used after ET across the centers, infarct volume data could not be calculated and was not included in this analysis.
This work supports early evidence on the impact of procedure time on lower rates of good outcomes and higher complication rates and hemorrhage after ET. This effect was present with both SR and ADAPT thrombectomy at various geographically diverse centers. ADAPT showed significantly faster recanalization and higher sensitivity to procedure time. The impact of procedure time is specifically related to the time spent undergoing thrombectomy after establishment of intracranial access. In addition, we also showed that the negative impact of procedure time was greater in posterior circulation strokes than in anterior circulation. Finally, we showed that exceeding 60 min of procedure time or 3 aspiration attempts should trigger a careful assessment of the futility and risks of continuing the procedure.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: Procedure time exceeding 30 to 60 min or >3 aspiration attempts in patients with AIS undergoing ET are associated with unfavorable outcomes, including higher rates of post-procedural intracerebral hemorrhage and other complications.
TRANSLATIONAL OUTLOOK: Future studies should gather data on the relationship between specific thrombectomy devices and techniques and the success of recanalization procedures for patients with AIS.
Dr. Fargen has served as a consultant for Cerebrotech. Dr. Starke has served as a consultant for Medtronic, Penumbra, and Abbott. Dr. De Leacy has served as a consultant for Penumbra and Cerenovus. Dr. Rai has served as a consultant for Stryker Neurovascular and Cerenovus. Dr. Kan has served as a consultant to Medtronic and Stryker Neurovascular. Dr. Turk has been a consultant for, received honoraria from, and served on speaker bureaus for Codman, Covidien, Penumbra, Microvention, Blockade, Pulsar Vascular, and Medtronic; has received research funding from Codman, Penumbra, Microvention, and Pulsar Vascular; and owns stock in Blockade and Pulsar Vascular. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Listen to this manuscript's audio summary by Editor-in-Chief Dr. Valentin Fuster on JACC.org.
- Abbreviations and Acronyms
- a direct aspiration at first pass technique
- acute ischemic stroke
- Alberta Stroke Program Early CT Score
- endovascular thrombectomy
- large-vessel occlusion
- modified Rankin score
- National Institutes of Health Stroke Scale
- parenchymal hemorrhage type 2
- symptomatic intracerebral hemorrhage
- Thrombolysis In Cardiac Ischemia
- tissue-type plasminogen activator
- Received September 20, 2018.
- Revision received November 19, 2018.
- Accepted November 27, 2018.
- Saver J.L.,
- Goyal M.,
- Bonafe A.,
- et al.,
- for the SWIFT PRIME Investigators
- Jovin T.G.,
- Nogueira R.G.,
- DAWN Investigators
- Turk A.S.,
- Frei D.,
- Fiorella D.,
- et al.
- Turk A.S.,
- Spiotta A.,
- Frei D.,
- et al.
- Spiotta A.M.,
- Vargas J.,
- Turner R.,
- Chaudry M.I.,
- Battenhouse H.,
- Turk A.S.
- Alawieh A.,
- Pierce A.K.,
- Vargas J.,
- et al.
- Alawieh A.,
- Starke R.M.,
- Chatterjee A.R.,
- et al.
- Nogueira R.G.,
- Lutsep H.L.,
- Gupta R.,
- et al.,
- for the TREVO 2 Trialists
- Spiotta A.M.,
- Arthur A.S.,
- Fiorella D.,
- et al.
- Alawieh A.,
- Chatterjee A.R.,
- Vargas J.,
- et al.
- Bourcier R.,
- Saleme S.,
- Labreuche J.,
- et al.
- Alawieh A.,
- Vargas J.,
- Turner R.D.,
- et al.
- Gulli G.,
- Marquardt L.,
- Rothwell P.M.,
- Markus H.S.