Abstract WP111: Who Needs Neuroprotection With Endovascular Stroke Therapy? Findings From the Trevo Retriever Registry

Background: Combined neuroprotection with endovascular therapy may improve clinical outcomes of only a subset of individuals treated for acute ischemic stroke. The risk/benefit profile of adjunctive treatment may not warrant neuroprotection with successful reperfusion, yet many other individuals have poor outcomes despite revascularization. We hypothesized that optimal candidates for adjunctive neuroprotection may be defined by analyses of subjects with poor clinical outcome despite successful reperfusion in the Trevo Retriever Registry.

Methods: The Trevo Retriever Registry dataset was analyzed to define the subset of cases with poor clinical outcome (day 90 mRS 4-6) after successful reperfusion (eTICI 2b50, 2b67, 2c, 3). Multivariate analyses were used to identify predictors of poor outcome using these distinct definitions of successful reperfusion. The influence of covariates, including TLSW, baseline clinical and imaging variables (e.g. ASPECTS, ASITN collateral grade), on defining such optimal neuroprotective candidates was delineated.

Results: Successful reperfusion adjudicated by core lab, defined as eTICI ≥ 2b50 included 1,162 subjects, with eTICI ≥ 2b67 in 920, eTICI ≥ 2c in 652 and eTICI 3 in 209. Poor outcome (day 90 mRS 4-6) occurred in 316/1162 (27%) with eTICI ≥ 2b50, 243/920 (26%) with eTICI ≥ 2b67, 172/652 (26%) with eTICI ≥ 2c and 61/209 (29%) with eTICI 3. Across all subsets, multivariate analyses to predict poor outcome after successful reperfusion identified increased age (per year, OR 1.04-1.05, all p=<0.02) as a factor, adjusting for withdrawal of care. Expectedly, greater baseline NIHSS severity predicted greater day 90 disability (OR 1.07-1.08, all p<0.001). TLSW was a predictor only with eTICI ≥ 2b50 (per hour, OR 1.02, p=0.039). History of diabetes was a factor only with eTICI ≥ 2b50 and eTICI ≥ 2b67 (OR 2.05-2.19, p<0.001). Worse collateral grade (ASITN 0-1) was the most potent predictor (OR 2.27-2.71 versus ASITN 2, p=0.027-0.052; OR 3.85-4.35 versus ASITN 3-4, all p=0.003).

Conclusions: Neuroprotection combined with endovascular therapy may optimally target stroke patients with worse collaterals, diabetes or increased age. Trial design for neuroprotection with revascularization in AIS should leverage these data.

Abstract WMP4: Fast versus Slow Progressors in Real-World Data From the Trevo Retriever Registry: Collaterals Dominate Time to Reperfusion in Clinical Outcome After Thrombectomy

Background: Time to reperfusion (TTR) is commonly cited in clinical outcome after endovascular treatment of acute ischemic stroke, yet collaterals may set the pace of ischemia. Real-world data on fast and slow progressors also remain scarce. We analyzed the impact of TTR on clinical outcome in real-world data using core lab adjudicated angiography, interventional steps and corresponding reperfusion.

Methods: 16 key time intervals were calculated from workflow (time of symptom onset, door, picture, puncture) and core lab metrics (clot visualization, first deployment, first reperfusion, final angiography) in real-world data from the Trevo Retriever Registry. These 16 variations of TTR were analyzed overall and by collateral status (ASITN 0-1 versus 2 versus 3-4) to determine the relationship with 90-day clinical outcomes.

Results: Real-world data on endovascular therapy from 1,441 subjects in the Trevo Retriever Registry were analyzed to relate TTR with clinical outcomes. Overall metrics for TTR are shown in Table 1. TTR was not linked with collateral status. Using a multivariate model incorporating known predictors, there was no influence of TTR using any of the 16 definitions on clinical outcome. Better collateral status on DSA prior to revascularization showed a potent relationship with 90-day mRS (p<0.001) and better probability of functional independence (aOR 1.4, 95% CI 1.2, 1.7) per grade of collateral flow.

Conclusions: Collaterals transform time to reperfusion, linking fast and slow progressors with subsequent clinical outcomes. TTR may be standardized based on these 16 key epochs in endovascular stroke therapy to document workflow metrics. Time is relative, even when measured with detailed, standardized metrics.

Abstract 63: Collaterals in Thrombectomy for MCA Occlusion: Mapping the Collaterome in the Trevo Retriever Registry

Background: The delay and dispersion of collateral circulation to the MCA territory is readily available prior to endovascular thrombectomy for acute ischemic stroke (AIS). Factors associated with collateral grade have never been established in such an extensive population, reflecting diverse subjects from around the world. Largescale data from the Trevo Retriever Registry enables mapping of the MCA collaterome for the first time.

Methods: The Imaging and Angiography Core Lab of the Trevo Retriever Registry independently conducted prospective evaluation of angiography in more than 1,500 subjects. Collaterals were systematically scored using ASITN grade prior to thrombectomy. Descriptive statistics detailed the distribution of collateral grades and analyses with demographic, imaging and clinical variables to explore key associations with routinely acquired registry data.

Results: 890 subjects (68.5 ± 15.1 years; 54.6% women; baseline NIHSS median 15 (10-19)) with AIS due to MCA occlusion had angiography of collateral circulation centrally adjudicated. Proximal M1 MCA occlusion was noted in 671/890 (75.4%). Collateral grade prior to thrombectomy included grade 4 or most robust collaterals in 38/890 (4.3%), 3 in 294/890 (33.0%), 2 in 467/890 (52.5%), 1 in 80/890 (9%) and 0 or none in 11/890 (1.2%). Baseline collaterals at angiography and pre-procedure ASPECTS were closely correlated (r=0.439, p<0.001) with more modest correlation between collateral grade and CTP/DWI (rrCBF<30% or ADC<620) infarct core (r=-0.31, n=391; p<0.001) or CTP/PWI (Tmax>6s) hypoperfusion at-risk (r=-0.10, n=391; p=0.043) volumes. First pass mTICI≥2b occurred in 553/858 (64.5%) with final mTICI≥2b in 824/890 (92.6%). Collateral grade prior to thrombectomy (each 1-point increment, after adjustment for other predictors) was strongly associated (OR 1.38 95%CI (1.12-1.7), p=0.002) with good clinical outcomes (mRS 0-2) at 90 days.

Conclusions: Largescale mapping of collaterals prior to MCA thrombectomy reveals marked variation in the extent and functional impact of the collaterome. The preponderance of partial perfusion in the downstream ischemic territory prompts the need to investigate and leverage the protective nature of the collaterome in AIS.

Abstract 114: Aspects versus Perfusion in the Trevo Retriever Registry: Defining the Core on the Largest Scale to Date

Background: Routine use of advanced imaging versus ASPECTS for imaging triage of endovascular thrombectomy candidates has not been evaluated on a large scale to date. Ischemic core may be defined by ASPECTS or perfusion imaging, yet these variable definitions likely reflect different pathophysiology as perfusion may fluctuate and ASPECTS lesions may be time-dependent.

Methods: The Trevo Retriever Registry was a prospective, open-label, consecutive enrollment, multicenter, international registry with more than 65 enrolling sites worldwide. The Imaging and Angiography Core Lab systematically adjudicated more than 1,500 subjects, scoring ASPECTS and separately processing perfusion imaging. Ischemic core volume on perfusion imaging was defined as rrCBF<30% (CTP) and analyzed with respect to ASPECTS.

Results: 488 subjects (68.3±14.4 years; 53.3% women; baseline NIHSS median 15 (10-19) with anterior circulation occlusions were evaluated with both ASPECTS and perfusion imaging prior to thrombectomy. Arterial occlusions included 87/487 (17.9%) ICA and 296/487 (60.8%) proximal M1 MCA, treated with thrombectomy at median 4.7 (3.3-7.9) hours from time last known well (TLKW). ASPECTS was median 8 (7-9) with ischemic core lesions of median 18 (4.9-39.2) cc. At-risk hypoperfusion (Tmax>6s) lesions were median 109.8 (62-156.9) cc. TLKW was associated with ASPECTS (r=-0.18, p<0.001) yet no time relationship was noted with either ischemic core or at-risk hypoperfusion on perfusion imaging. ASPECTS correlated modestly with perfusion imaging-derived ischemic core (r=-0.35, p<0.001) and at-risk hypoperfusion (r=-0.24, p<0.001). Post-procedure mTICI≥2b occurred in 448/488 (91.8%). Each increment in baseline ASPECTS was associated with an adjusted OR of 1.21 (95%CI (1.05-1.39), p<0.008 for good clinical outcomes (day 90 mRS 0-2), whereas the perfusion lesion volume for ischemic core and at-risk hypoperfusion did not predict outcomes.

Conclusions: Largescale, systematic evaluation of ASPECTS and perfusion imaging prior to thrombectomy reveals discrepancy in the definition of ischemic core and the prediction of clinical outcomes after revascularization. ASPECTS is time-dependent, yet reliably predicts outcomes in routine clinical practice.

Abstract WP5: The Transfer Score May Aid Decisions Whether to Transfer Patients with Large Vessel Occlusions for Endovascular Therapy

Background: While faster reperfusion with EVT leads to better outcomes in acute ischemic stroke due to large vessel occlusion (LVO), most LVO patients present to outside hospitals without EVT capability. Treating physicians are often unsure if EVT would confer benefit upon arrival to tertiary hospitals given inter-facility transfer delays.

Objective: We evaluated independent predictors of good outcome in transferred patients treated with EVT to devise a score that may assist treating physicians to make transfer and treatment decisions.

Methods: Transfer patients were analyzed in a multicenter international prospective cohort study of LVO patients treated with stent retriever thrombectomy (TREVO Registry) from 11/2013 to 4/2016. Independent factors correlating with good outcome after EVT were identified using univariate and multivariate analyses. We devised a score to identify patients with LVO at the referral facility who may benefit from EVT.

Results: Of 1000 patients enrolled, 226 were anterior circulation occlusions, transferred and treated within 0-8 hrs (Table 1). Age, stroke severity, glucose level, M2 occlusion and achieving onset to groin puncture ≤ 5 hr were independent factors associated with good outcome (Table 2). Other clinical variables were analyzed, as in ASPECTS, but were not significant. A 10 point score was devised (Table 3). Patients with a score of 0-4 had 4 times the odds of good outcome compared to a score of 5-9 (aOR 4.3, 95% CI 1.9-9.9; p <0.001). These results were maintained after adjustment for mTICI and IV-tPA (aOR 4.0, 95% CI 1.7-9.4; p <0.001). Fig 1 shows good outcome rates stratified by score points. ROC curves showed better score performance (AUC= 0.8) compared to THRIVE (AUC=0.74) and HIAT (AUC=0.69) certifying good predictability.

Conclusion: A simple transfer score may be an effective triage method to identify patients at remote facilities who may benefit from EVT upon transfer. Further validation is necessary to confirm these findings.

Abstract WMP2: Trevo 2000: Real-World Experience in the First 1247 Patients

Objective: The Trevo Registry is designed to assess real world outcomes of the Trevo Retriever in patients experiencing ischemic stroke. This is the largest prospective study for acute stroke intervention, with 1247 patients currently enrolled and 90 day outcomes in 1021 patients. The primary endpoint is revascularization status based on post-procedure TICI score and secondary endpoints include 90-day mRS, 90-day mortality, neurological deterioration at 24 hours and device/procedure related adverse events.

Methods: The study is a prospective, open-label, consecutive enrollment, multi-center, international registry of patients undergoing mechanical thrombectomy for acute stroke using the Trevo stent retriever as the initial device. Enrollment is expected to reach 2000 subjects at up to 100 sites.

Results: As of August 13, 2016 a total of 1247 patients were enrolled. The median NIHSS at admission was 16 (IQR 11-20). Most patients (66.2%) were treated at >/= 6 hours from last known normal with a median procedure time of 50 minutes (32-77). The occlusion site was M1 or M2 in 74.5%. General anesthesia was employed in 46.6% of procedures. TICI 2b or 3 revascularization was 92.8% with an average of 1.6 passes with the device. Median NIHSS at 24 hours and discharge was 6 and 4 respectively. Fifty-five percent of patients had mRS ≤2 at 3 months and the overall mortality rate was 15.4%. Patients treated after 8 hours of symptom onset had a 94.9% revascularization rate and 52.8% mRS ≤2 at 3 months. The symptomatic ICH rate was 1.2%. Patients who met the revised AHA criteria for thrombectomy were found to have 58.4% mRS 0-2 at 90 days.

Conclusions: The Trevo Retriever Registry represents the first real world data with stent retriever use in the era of clinical trials showing the overwhelming benefit of stent retrievers to treat acute ischemic stroke. Due to the fact that this data represents real world use of the Trevo Retriever, (e.g. subjects pre-stroke mRS >1 (16.5%) and those treated 6-24 hours after stroke symptoms (33.8%), this data cannot be compared to the results from recent trials with restricted eligibility criteria. Future subgroup analysis of this large cohort will help to identify areas of future research to enhance outcomes further with this treatment modality.

Abstract WMP9: Endovascular Thrombectomy Impact in the First Three “Golden” Hours

Background: Endovascular thrombectomy (EVT) substantially increases the likelihood of good outcome in acute ischemic strokes due to large vessel occlusion (LVO). Expediting EVT to achieve faster reperfusion is an important factor that correlates with good outcome. Ultra-early intervention in the first 3 “golden” hours from onset was not well characterized in recent trials.

Objective: We sought to assess the impact of early treatment within the first 3 hours on clinical outcomes in large, real life, world-wide practice.

Methods: We analyzed a multicenter international prospective cohort study of LVO patients treated with stent retriever thrombectomy (TREVO Registry) between11/2013 and 4/2016. We stratified patients based on treatment time, onset to groin puncture (GP), into 3 groups: 0-3, 3-6, >6 hrs. 90 day mRS was the primary outcome (0-2 good outcome). Logistic regression modeling was performed to evaluate the impact of treatment within the golden 3 hours on outcomes and to determine the independent factors associated with EVT initiation within 3 hours.

Results: In the 905 patients, GP occurred in: 23.1% 0-3 hrs, 44.3% 3-6 hrs and 32.6% >6 hrs. Table 1 shows similar baseline characteristics among the groups. Patient-level predictors of treatment within 3 hrs were age (aOR 1.1 per decade of age ≥18) and good ASPECTS (aOR 1.2 per point). No hospital-level predictors of early treatment were found. Patients treated within 3 hrs have a higher likelihood of good outcome as compared to those treated >3 hrs (aOR 2.0, 95% CI 1.4-2.9; p <0.001) after adjustment for age, NIHSS, IV tPA and mTICI ≥2b (Table 2). No differences were found in mortality and sICH. Treatment in the golden hours had the highest impact on excellent outcome rates (mRS 0-1) (Fig 1).

Conclusion: Early thrombectomy of LVO strokes, within the first three hours provides the highest impact compared with later time windows. Streamlining processes to deliver rapid intervention within 3 hours would improve clinical outcomes.

Abstract TP20: Uncertainties of Endovascular Therapy Outside the AHA Guidelines

Background: The efficacy of endovascular therapy (EVT) in randomized clinical trials (RCTs) for acute strokes due to large vessel occlusion (LVO) led to AHA guidelines recommending EVT as standard of care for selected patients. However, many conditions were under-represented in the RCTs: ASPECTS <6, age ≥80 yo, NIHSS <6, onset to treatment >6 hrs and M2/ distal/ posterior circulation occlusions.

Objective: We evaluated EVT outcomes in these populations compared to counterparts represented in the RCTs.

Methods: A large multicenter international prospective cohort study of LVO patients treated with stent retriever thrombectomy (TREVO Registry) between 11/2013 and 4/2016 was analyzed. 90 day mRS was the primary outcome (0-2 good outcome). Multivariate logistic regression modeling was employed to evaluate EVT impact in the different groups.

Results: Of 1000 patients, 81 had NIHSS <6 and 81.5% of those achieved a good outcome (aOR 3.6, 95% CI 1.9-6.8; p< 0.001 compared with NIHSS ≥6) (Table 1). Over 80 yo, however, had low odds of independence (aOR 0.3, 95% CI 0.2-0.5; p <0.001 compared with <80 yo). Among 212 patients treated >6 hrs, 51% had a good outcome (aOR 0.78, 95% CI 0.55-1.1; p =0.17) compared to ≤6 hrs. Nearly half of patients with ASPECTS <6 (3-5) had a good outcome. Fig 1 illustrates mRS distributions stratified by the different subgroups. There were low rates of sICH for treated patients with NIHSS<6, age≥80, ASPECTS <6 or treatment >6 hrs. Fig 2 demonstrates the likelihood of good outcome by clot location. M2 and distal occlusions had the highest good outcome probabilities while proximal ICAs had the lowest (48.1%). More than half of vertebrobasilar patients achieved independence (54.8%).

Conclusion: While effectiveness cannot be determined in the absence of medically treated controls, our analyses of real world data show several groups outside AHA guidelines may benefit from EVT. In particular, further study is needed to examine EVT benefits for mild stroke and M2 occlusions.

Abstract WMP11: Joint Commission Certified Stroke Centers Treat More Severe Strokes with Faster Procedure Times Compared to Non-joint Commission Certified Stroke Centers in the Trevo Registry

Introduction: Endovascular stroke therapy has become the gold standard treatment for large vessel occlusion. The Joint Commission has certified hospitals as Comprehensive stroke centers (JCCSC) based on rigorous standards in the hopes of identifying centers of excellence. We sought to determine if JCCSC have faster door to reperfusion times compared to non-JCCSC.

Methods: The TREVO registry is a multicenter international real world registry assessing angiographic and clinical outcomes with the Trevo device being used in the first pass. We defined a CSC as certified by the Joint Commission as of July 1, 2016. Demographic information, times within the hospital, angiographic results and clinical outcomes were analyzed between the JCCSC and non-JCCSC institutions.

Results: A total of 507 patients (329 JCCSC, 178 non-JCCSC) have completed data in the Trevo registry to date. There are a higher proportion of patients with ASPECTS < 7 being treated at JCCSC vs. non-JCCSC (8.8% vs. 0.0%, p<0.02). There were no differences in outcomes, reperfusion rates or symptomatic hemorrhage rates between the two groups. Demographics were similar except patients treated at a JCCSC had a higher median NIHSS [17 vs. 15, p<0.003] compared to the non-JCCSC group. Median (IQR) door to puncture times did not differ between the two groups [85(57-132) vs. 91(59-137), p<0.96], but patients treated at a JCCSC had lower mean angiographic procedure times [59 ± 34 minutes vs. 66±44 minutes, p<0.05]. The analysis did not change when we looked at the subset of patients who were not transferred with anterior circulation strokes less than 8 hours from onset.

Conclusions: Patients treated at a JCCSC had faster procedural times, without faster door to procedure times when compared to non-JCCSC centers. Outcomes were no different, due to imbalances in stroke severity at baseline and a higher proportion of patients with ASPECTS < 7 being treated.

Abstract 37: ASPECTS and Stratified Outcomes After Endovascular Therapy in the Trevo Retriever Registry: Benefit in Low ASPECTS

Background: Most endovascular stroke therapy studies and subsequent guidelines restrict intervention based on ASPECTS. A wide range of ASPECTS scores may be encountered in practice and individual patient benefit may be realized even at low ASPECTS. We examined large-scale data on outcomes after endovascular therapy, stratified by baseline ASPECTS in the Trevo Retriever Registry.

Methods: The independent Imaging Core Lab of the Trevo Retriever Registry prospectively determines ASPECTS on baseline imaging acquired immediately prior to endovascular thrombectomy. ASPECTS scores and regional involvement were analyzed with respect to site of arterial occlusion, effect of time from symptom onset, co-morbidities and clinical outcomes, based on ASPECTS strata.

Results: Baseline ASPECTS data was reviewed by the Imaging Core Lab in 426 subjects with anterior circulation stroke enrolled in the Trevo Retriever Registry, as of July 2016. Mean age was 68.8 ± 13.7 yrs, with 20.9% > 80 years old. Baseline NIHSS was median 15.0 (10.0, 19.0). Onset to CT was median 3.8 (1.5, 9.0) hrs, with median ASPECTS of 8.0 (7.0, 9.0), ranging from 3-10. Baseline ASPECTS 0-7 occurred in 118/426 (27.7%) subjects, including 39.0% of ICA, 27.1% M1 and 16.9% M2/3 arterial occlusions at angiography. Baseline clinical variables predicting ASPECTS included age and NIHSS, whereas the ASPECTS score was mildly associated with final TICI2C reperfusion (r=0.24, p<0.001). Subsequent symptomatic ICH was 1.7% with baseline ASPECTS 0-7 versus 2.0% with ASPECTS 8-10. The distribution of mRS at 90 days based on individual ASPECTS strata from 10 to 3 revealed a trend to worse outcomes with lower ASPECTS, yet good outcomes (mRS 0-2) were 60.7% (ASPECTS 10), 55.3% (9), 60.2% (8), 54.9% (7), 55.1% (3-6).

Conclusions: Discrete ASPECTS strata may influence outcomes of endovascular therapy conducted in routine practice around the world, yet individuals with low ASPECTS may still achieve reasonable outcomes.

Abstract WP6: Real-World Data on Reperfusion: Evidence of Good Outcomes in the International Trevo Retriever Registry

Background: Endovascular thrombectomy trials established efficacy in acute ischemic stroke, yet real-world data on device effectiveness is warranted. Core lab adjudication of angiography is required to validate reperfusion, providing evidence and detailed data beyond randomized, controlled trials. We report the largest endovascular therapy registry data linking independent core lab data on reperfusion with clinical outcomes.

Methods: The Trevo Retriever Registry is a prospective, open-label, consecutive enrollment, multicenter, international registry with more than 65 enrolling sites worldwide. An independent Imaging Core Lab, blind to all other data, evaluates the angiography with a battery of various TICI scores (mTICI, oTICI, oTICI2C) to define reperfusion. Statistical analyses examined TICI reperfusion and association with clinical outcomes considering expansive data collected in the registry.

Results: 506 enrolled subjects (mean age 68.2 ± 14.2 yrs; 53% female) had core lab adjudicated angiography as of July 2016, including 21.5% > 80 years old. Baseline NIHSS was median 15.0 (9.0, 20.0) with time from onset to CT of median 4.0 (1.7, 9.7) hrs. Core lab adjudicated arterial occlusion sites were: 53% M1, 24% ICA, 16% M2, 4% Basilar and 2% other. Time to reperfusion (oTICI ≥ 2A) was median 30.0 (19.0, 42.0) min. Core lab adjudicated revascularization was mTICI ≥ 2B in 90.4% (95%CI 87.4, 92.9), oTICI ≥ 2B in 82.3% (95%CI 78.6, 85.6) and oTICI2C ≥ 2C in 45.0% (95%CI 40.5, 49.6). mRS of 0-2 at 90 days was achieved in 57.3% (95%CI 52.5, 62.1). Extensive clinical, laboratory and stroke workflow variables were considered, yet only male sex (OR 0.62 (95% CI 0.38, 0.99) was an independent predictor of successful reperfusion (oTICI ≥ 2B) while age (OR 0.96 (95% CI 0.94, 0.97), NIHSS (OR 0.91 (95% CI 0.88, 0.94) and diabetes (OR 0.54 (95% CI 0.33, 0.88) predicted mRS 0-2 at 90 days.

Conclusions: Proven reperfusion rates after endovascular stroke therapy excel in the real-world translation of thrombectomy devices around the globe, leading to good outcomes after stroke.