Original Paper Cerebrovasc Dis 2013;36:218–225 DOI: 10.1159/000353990
Received: February 25, 2013 Accepted: June 20, 2013 Published online: October 12, 2013
Neurothrombectomy for the Treatment of Acute Ischemic Stroke: Results from the TREVO Study Olav Jansen a Juan Macho b Monika Killer-Oberpfalzer c David Liebeskind d Nils Wahlgren e on behalf of the TREVO Study Group a
Department of Neuroradiology, University Hospital Schleswig-Holstein Campus Kiel, Kiel, Germany; Angioradiology and Interventional Neuroradiology Unit, Radiology Department, Imaging Diagnostic Center, Clinic University Hospital, Barcelona, Spain; c Department of Neurology/Research Institute of Neurointervention, Paracelsus Medical University, Salzburg, Austria; d Department of Neurology, Stroke Center, University of California, Los Angeles, Calif., USA; e Department of Neurology, Karolinska University Hospital, Stockholm, Sweden b
Key Words Ischemic stroke · Endovascular treatment · Retrieval device
Abstract Background: Fast recanalization has been shown to be one of the most important factors for good clinical outcome in stroke patients with acute large vessel occlusion. While intravenous thrombolysis has been shown to be of limited effect in patients with large clot burden, intra-arterial neurothrombectomy offers a new and promising possibility to achieve high recanalization rates within a short time. The Trevo device is a stent-like retriever and was primarily designed to remove thrombus in patients experiencing an acute ischemic stroke. We report on the results of the TREVO Study, which was a prospective, multicenter study in acute stroke patients treated with the Trevo device. Methods: Patients were recruited in seven European centers under the control of an external monitor. Centers were selected because of their long experience with interventional stroke therapy especially with neurothrombectomy. We included adults aged 18–85 years with angiographically confirmed large vessel occlusion strokes and NIHSS scores of 8–30 and treatable within 8 h of symptom onset. The primary endpoint was revascularization, defined as at least TICI (throm-
© 2013 S. Karger AG, Basel 1015–9770/13/0363–0218$38.00/0 E-Mail [email protected] www.karger.com/ced
bolysis in cerebral infarction) 2a. The revascularization scores were assessed by an independent core lab. Secondary endpoints were clinical outcome at 90 days (mRS 90), any devicerelated serious adverse events and the rate of symptomatic intracerebral hemorrhages. Results: 60 patients were enrolled. The overall recanalization rate (≥TICI 2a) was 91.7% and TICI 2b and 3 was achieved in 78.3%. At 90 days, 55% of the patients had a favorable neurological outcome (mRS 0–2) and 20% had died. Patients with successful recanalization (TICI 2a,b/3) had a good 90-day neurological outcome (mRS 0–2) in 60%, whereas no patient without recanalization had a mRS 90 3.0 Treated with Heparin within 48 h with a PTT greater than 2 times the lab normal Baseline platelet count 185 mm Hg or DBP >110 mm Hg) Note: If the blood pressure can be successfully reduced and maintained at the acceptable level using medication (i.e. nipride), the patient can be enrolled Woman of child bearing potential who is known to be pregnant Patient participating in another clinical study or protocol For anterior circulation strokes: strokes involving greater than 1/3 of the MCA territory, as determined by hypodensity on the baseline noncontrast CT, or low CBV on CT perfusion imaging, or restricted diffusion on DWI images For posterior circulation strokes within the midbrain and/or pons, extensive hypodensity on the baseline CT, or low CBV on CT perfusion imaging, or extensive restricted diffusion on DWI images Baseline CT/MR evidence of significant mass effect with midline shift Baseline CT/MR evidence of hemorrhage Baseline CT/MR evidence of intracranial tumor (except small meningioma) Angiographic evidence of vasculitis or arterial dissection High-grade stenosis that cannot be treated safely or which prevents access to the thrombus with the Trevo device Angiographic evidence of excessive arterial tortuosity that precludes the Trevo device from reaching the thrombus
Cerebrovasc Dis 2013;36:218–225 DOI: 10.1159/000353990
or to the study all endovascular operators completed an in vitro training with the Trevo device as well as a review of the protocol inclusion/ exclusion criteria, requirement to collect neuroimaging to send to the core lab, and familiarity with a standard angiography acquisition protocol. Only CE-marked devices were used in the study. Inclusion and Exclusion Criteria The inclusion criteria for the study were diagnosis of an acute ischemic stroke involving occlusion of a proximal artery (ICA, M1, M2 or vertebrobasilar) with an NIHSS score of 8–30 in a subject between 18 and 85 years of age who could be treated within 8 h from symptom onset (see table 1 for a complete listing of inclusion/exclusion criteria). Screening, Consent and Enrollment Patients were enrolled consecutively in the participating centers. All patients presenting with stroke symptoms were screened at each site to determine if they met entry criteria for the study. After obtaining signed informed consent, all eligible patients were evaluated for neurological and functional status, and an angiographic assessment of the suspected vascular occlusion. For patients who met all of the inclusion criteria and none of the exclusion criteria and who were eligible for the study, the following baseline data were collected: medical history and admission details, patient demographics/characteristics, CT or MR neuroimaging scan, baseline blood values, baseline NIHSS assessment and prestroke mRS assessment. Patients who did not qualify for participation in the study because they did not meet the inclusion/exclusion criteria were documented as screen failures, and were excused from further study participation. Enrollment in this study was defined as the moment when the Trevo device was deployed through the microcatheter and enters the patient’s neurovasculature. During the study, no patients were treated by the investigators with the TREVO device outside the trial. Trevo Procedure Using appropriate (general or local) anesthesia, transfemoral arterial access was obtained per standard practices at the treating institution. A diagnostic angiogram was performed in order to determine the appropriateness of the occlusion for treatment with the Trevo device. Visual estimation of the vessel diameter and occlusion location was recorded. Prior to the start of the procedure, the TICI score in the vascular territory being treated was assessed. If thrombus was identified in a treatable vessel, defined as the internal carotid, middle cerebral M1 and/or M2 branches, basilar or vertebral arteries, the Trevo thrombectomy procedure was initiated. For this, a microcatheter was navigated through the occlusion and its tip was placed distally to the thrombus. The Trevo device was delivered through the Trevo microcatheter and the shaped section was unsheathed via pulling back on the microcatheter, leaving the device directly in the thrombus and allowing it to immediately begin expanding and incorporating into the thrombus, thereby anchoring the device to the thrombus and facilitating retrieval of the thrombus into the guide catheter. The intra-arterial dwell time to let the device struts integrate into the thrombus was 3–5 min. During the pull-back maneuver, aspiration was performed either via a distal access catheter (DAC 0.0057) placed in the distal
Jansen et al.
Table 2. Time and events schedule
Assessment
Preprocedure Trevo procedure
Consent
✓
Patient demographics/characteristics
✓
Medical history and admission details
✓
Baseline labs
✓
CT or MRI
✓1
NIHSS
✓
Angiogram
✓
TICI and TIMI scores
✓
Unscheduled follow-up
Day 7–10 or day of discharge (whichever occurs first)
✓1
✓2
✓
✓
✓
✓
✓
✓ ✓
✓ ✓
✓ ✓
✓2
1
✓
Procedure details mRS AEs
90 days (±14 days)
24 h (–6/+12 h)
✓3
✓
1 Images are de-identified and submitted for archiving and core lab review. 2 If clinically indicated, e.g. a significant neurological worsening, or suspicion of an intracranial hemorrhage. 3 Preprocedure mRS is based upon patient’s prestroke abilities and life style. AEs = Adverse events.
ICA (C1–3) or proximal M1 or via a balloon occlusion guide catheter (typically 8F) placed in the proximal ICA. The same Trevo device could be used several times to a maximum of six retrievals; however, after each deployment of the device, it was thoroughly inspected and cleaned before reloading. Use of the Trevo device was terminated if there was any angiographic evidence leading to the suspicion of an intracranial hemorrhage, e.g. extravasation of contrast during the procedure. At the end of the Trevo procedure, biplane angiography was performed immediately afterwards in order to document the TICI score in the vascular territory being treated. Investigators were instructed to use Trevo as the first and primary device, but if the device was unsuccessful, the operator had the option of using other adjunctive treatment, including intra-arterial thrombolysis, if they had performed at least one Trevo pass. Postprocedure Follow-Up Patients were managed postprocedure according to each institution’s usual practice. It was recommended that administration of anticoagulants and antiplatelets be suspended for 24 h postthrombectomy in patients who were not directly in need of these agents. At 24 (–6/+12) h postprocedure, the following study assessments were performed: NIHSS score, CT or MR imaging to assess the presence/absence of intracranial hemorrhage, adverse events and adverse device effects were documented. At day 7–10, or at discharge if earlier, the NIHSS score and mRS score was evaluated and again at day 90 (±14 days) postprocedure. Time and events schedule is shown in table 2 in more detail.
propriately de-identified, and transmitted for data archiving and central core lab evaluation. For each enrolled subject in the trial, the angiography dataset was reviewed to ensure complete content of the diagnostic and therapeutic phases of the procedure. No information other than personal identifiers was removed from the dataset prior to independent angiography core lab review. The core lab utilized the modified TICI grading scale for assessment of the primary outcome measure. Revascularization outcomes immediately after use of the Trevo device and postprocedure (if additional adjunctive/rescue treatments were done) were adjudicated. Additional information acquired by the core lab included potential angiographic complications including perforation, dissection, vasospasm, and contrast extravasation. Study Monitoring Under the supervision and at the direction of Concentric Medical, a monitor conducted on-site monitoring to ensure that all sites were in compliance with the Declaration of Helsinki 23, the Clinical Study Plan, the Clinical Study Agreement, Ethics Committee requirements, and applicable local laws and regulations. Standardized case report forms were completed for all patients enrolled in the study. Source data verification was performed by the monitor for key data points throughout the study.
Results
Core Lab All enrolled patient’s imaging data, including baseline and postprocedure CT/MRI and the full angiography datasets were ap-
Enrollment began in February 2010 and was completed in August 2011, with the last patient follow-up completed in November 2011.
Neurothrombectomy for the Treatment of Acute Ischemic Stroke
Cerebrovasc Dis 2013;36:218–225 DOI: 10.1159/000353990
221
60 patients were enrolled among the seven participating sites. No patients were lost to follow-up and a 90-day mRS was obtained for all surviving subjects. In 58 cases angio core lab reads were completed, in 2 cases site-reported angio data were used due to technical reasons that precluded the transfer of the images to the Core Lab. Table 3 summarizes the results on patient demographics, baseline stroke score and site of target vessel. The suspected stroke etiology included atrial fibrillation in 41.7%, large artery atherosclerosis in 20%, other cardioembolic disease in 9% and other/unknown reason in 28.3%. Time from symptom onset to first pass with the Trevo retriever was 3.9 ± 1.4 h (mean ± SD). In 45%, the femoral arterial puncture to start the procedure was performed within the first 3 h, in 31.7% this puncture was between hours 3 and 4.5 and in 23.3% this was done between hours 4.5 and 8. 60% of the patients (36/60) had received intravenous r-tPA prior to the endovascular procedure with a mean dosage of 63.9 mg r-tPA. The mean time from symptom onset to start with intravenous thrombolysis in this subgroup was 1.9 h. Periprocedural adjuvant treatment was done with intra-arterial t-PA in 6/60 patients (10%), before thrombectomy extracranial carotid stenting was performed in 6/60 patients (10%) and extracranial angioplasty due to a proximal stenosis was performed in 3/60 patients (5%). Another 3 patients (5%) received a IIb/IIIa inhibitor and in 13/60 patients (21.7%) an additional thrombectomy device was used as second-line endovascular treatment. The interventionalist performed a diagnostic angiogram after deployment of the Trevo device, showing a perfusion channel in 73.5% (111/151 passes). The mean dwell time to integrate the struts of the device into the thrombus was 2.62 ± 1.50 min. A mean of 2.62 ± 1.37 passes with the Trevo device were performed (range 1–7). Mean time from femoral arterial puncture to the end of the procedure was 86.6 ± 44 min with a mean treatment time (introduction of Trevo device to end of the procedure) of 59.8 ± 40 min. The sites reported all cases starting with either TICI 0 (98.3% 59/60) or TICI 1 (1.7% 1/60). Evaluation of the initial angiograms by the Core Lab showed a complete vessel occlusion (TICI 0) in 95% (57/60), a TICI 1 score in 1.7% (1/60) and a TICI 2a score in 3.3% (2/60). At the end of the Trevo passes, the Core Lab reading of the an222
Cerebrovasc Dis 2013;36:218–225 DOI: 10.1159/000353990
Table 3. Baseline subject demographics and stroke characteristics
Characteristics
TREVO (n = 60 subjects)
Age, years Mean ± SD Median (range) % male gender NIHSS Mean ± SD Median (range) 8–14 15–22 23–30 Baseline mRS 0 1 Clot location (per core lab) ICA M1 M2 VB Suspected stroke etiology Large artery atherosis Cardioembolic Afib Other Unknown Other
64.7±13.4 65.2 (21.5–84.5) 45.0 17.7±4.8 18.0 (8.0–28.0) 25.0% 56.7% 18.3% 78.3% 21.7% 21.7% 60.0% 10.0% 8.3% 20.0% 51.7% 41.7% 8.3% 25.0% 3.3%
Afib = Atrial fibrillation.
giogram showed a recanalization (TICI 2a, 2b and 3) in 90.0% (54/60). At the conclusion of the entire procedure, including any adjunctive treatment, the recanalization rate was 91.7% and TICI 2b/3 was achieved in 78.3%. Recanalization rate by target vessel is shown in table 4. Clinical Outcome At 90 days, 55% of the patients had a favorable neurological outcome (mRS 0–2) and 20% had died. Favorable outcome was very often apparent at 24 h with a decrease of the median NIHSSS from initial 18 to 9. At day 7 the NIHSSS was decreased again to a median of 4.0. Patients with successful recanalization (TICI 2a,b/3) had a good 90-day neurological outcome (mRS 0–2) in 60%, whereas no patient without recanalization had a mRS less than 3 at day 90. Mortality at day 90 in successfully recanalized patients was 18 and 40% in patients without recanalization. Figure 2 shows the outcome results in more detail and by occlusion location. The mean length of hospital stay was 8.2 ± 2.9 days with a range from 2 to 22 days. Jansen et al.
Outcome measure TICI, preprocedure, by core lab 0 1 2a TICI, immediately after Trevo, by core lab 0 1 2a 2b 3 TICI, end of procedure, by core lab 0 1 2a 2b 3 Postdevice revascularization success (TICI 2a, 2b, or 3) End of procedure revascularization success (TICI 2a, 2b, or 3)
TREVO (n = 60 subjects) 95.0 1.7 3.3 6.7 3.3 18.3 65.0 6.7 3.3 5.0 13.3 70.0 8.3
ICA (n = 13)
92% 93% 80%
80%
MCA (n = 40) VB (n = 5)
54% 52%
23%
Revasc
mRS 0–2
19% 20%
mRS 6
Fig. 2. Revascularization and 90-day mRS by occlusion location.
90.0 91.7
mRS
7 days (n = 59)
90 days (n = 60)
0 1 2 3 4 5 6 mRS 0–2 10+ improvement in NIHSS* At 24 h At 7 day At 90 day
13.6 18.6 13.6 6.8 20.3 16.9 10.2 45.8
15.0 33.3 6.7 15.0 10.0 0.0 20.0 55.0
42.6 55.4 68.3
* If baseline NIHSS was 10 or less, then a score of 0 or 1 was counted as success. Expired subjects are scored as a 42 for this analysis. Values are percentages.
Complications Serious procedure-related adverse events were reported in 5/60 patients (8.3%). In 3 of these patients (5%) the SAEs were also adjudicated as device-related: two early symptomatic intracranial hemorrhages SICH (perforation, parenchymal hemorrhage grade 2 and SAH; hemorrhagic infarction grade 1 and SAH) and one late SICH. The overall SICH rate, using the pre-specified SITSMOST definition was 5% (3/60); only one of them was Neurothrombectomy for the Treatment of Acute Ischemic Stroke
% 100 90 80 70 60 50 40 30 20 10 0
Color version available online
Table 4. Primary results
device related. The SICH rate, as defined by the ECASS III criteria, was 8.3% (5/60). There were four other nonserious device-related events (1 minor vessel perforation, and three asymptomatic hemorrhages). Adjunctive Treatment In 13 (21.7%) of the cases an adjunctive device as a second-line treatment was used. Among the 13 cases, 9 cases had a single adjunctive device used and 4 had more than one device used. The devices used were Solitaire (11 cases), Merci (2 cases), Phenox (2 cases) and Penumbra (2 cases). Comparing the revascularization rates after Trevo alone to the final (postadjunctive), 5 had the TICI scores improve by at least 1 grade (3 ‘unsuccessful’ cases became ‘successes’ while the other 2 remained a success (n = 1) or a failure (n = 1)). Six cases remained unchanged following adjunctive use, and 2 cases got worse (both were TICI 2a following Trevo, but resulted in TICI 0 or 1 after adjunctive treatment). The net change from postTrevo recanalization (90%) to final recanalization (91.7%) was only 1.7%, but the 90-day outcomes for these subjects getting adjunctive treatment was much worse than the overall cohort (mRS 0–2 = 23% (3/13); mortality = 38% (5/13)).
Discussion
This study showed a recanalization rate (TICI 2b and 3) of 78% with good clinical outcome (mRS 90: 0–2) in 55% in patients with acute occlusion of greater intracranial vessels who were treated with the TREVO system. Cerebrovasc Dis 2013;36:218–225 DOI: 10.1159/000353990
223
Following international guidelines intravenous thrombolysis is still the only accepted treatment within the first 4.5 h after stroke onset. However, intravenous thrombolysis is known to be of limited value in proximal arterial occlusion, particularly in cases with extensive clot burden. Recanalization success on the other hand has been shown to be the most important variable for good clinical outcome. Rha and Saver [2] were able to demonstrate in a meta-analysis of 55 studies with more than 2,000 patients that successful recanalization has an OR for good clinical outcome (mRS 90
Received: February 25, 2013 Accepted: June 20, 2013 Published online: October 12, 2013
Neurothrombectomy for the Treatment of Acute Ischemic Stroke: Results from the TREVO Study Olav Jansen a Juan Macho b Monika Killer-Oberpfalzer c David Liebeskind d Nils Wahlgren e on behalf of the TREVO Study Group a
Department of Neuroradiology, University Hospital Schleswig-Holstein Campus Kiel, Kiel, Germany; Angioradiology and Interventional Neuroradiology Unit, Radiology Department, Imaging Diagnostic Center, Clinic University Hospital, Barcelona, Spain; c Department of Neurology/Research Institute of Neurointervention, Paracelsus Medical University, Salzburg, Austria; d Department of Neurology, Stroke Center, University of California, Los Angeles, Calif., USA; e Department of Neurology, Karolinska University Hospital, Stockholm, Sweden b
Key Words Ischemic stroke · Endovascular treatment · Retrieval device
Abstract Background: Fast recanalization has been shown to be one of the most important factors for good clinical outcome in stroke patients with acute large vessel occlusion. While intravenous thrombolysis has been shown to be of limited effect in patients with large clot burden, intra-arterial neurothrombectomy offers a new and promising possibility to achieve high recanalization rates within a short time. The Trevo device is a stent-like retriever and was primarily designed to remove thrombus in patients experiencing an acute ischemic stroke. We report on the results of the TREVO Study, which was a prospective, multicenter study in acute stroke patients treated with the Trevo device. Methods: Patients were recruited in seven European centers under the control of an external monitor. Centers were selected because of their long experience with interventional stroke therapy especially with neurothrombectomy. We included adults aged 18–85 years with angiographically confirmed large vessel occlusion strokes and NIHSS scores of 8–30 and treatable within 8 h of symptom onset. The primary endpoint was revascularization, defined as at least TICI (throm-
© 2013 S. Karger AG, Basel 1015–9770/13/0363–0218$38.00/0 E-Mail [email protected] www.karger.com/ced
bolysis in cerebral infarction) 2a. The revascularization scores were assessed by an independent core lab. Secondary endpoints were clinical outcome at 90 days (mRS 90), any devicerelated serious adverse events and the rate of symptomatic intracerebral hemorrhages. Results: 60 patients were enrolled. The overall recanalization rate (≥TICI 2a) was 91.7% and TICI 2b and 3 was achieved in 78.3%. At 90 days, 55% of the patients had a favorable neurological outcome (mRS 0–2) and 20% had died. Patients with successful recanalization (TICI 2a,b/3) had a good 90-day neurological outcome (mRS 0–2) in 60%, whereas no patient without recanalization had a mRS 90 3.0 Treated with Heparin within 48 h with a PTT greater than 2 times the lab normal Baseline platelet count 185 mm Hg or DBP >110 mm Hg) Note: If the blood pressure can be successfully reduced and maintained at the acceptable level using medication (i.e. nipride), the patient can be enrolled Woman of child bearing potential who is known to be pregnant Patient participating in another clinical study or protocol For anterior circulation strokes: strokes involving greater than 1/3 of the MCA territory, as determined by hypodensity on the baseline noncontrast CT, or low CBV on CT perfusion imaging, or restricted diffusion on DWI images For posterior circulation strokes within the midbrain and/or pons, extensive hypodensity on the baseline CT, or low CBV on CT perfusion imaging, or extensive restricted diffusion on DWI images Baseline CT/MR evidence of significant mass effect with midline shift Baseline CT/MR evidence of hemorrhage Baseline CT/MR evidence of intracranial tumor (except small meningioma) Angiographic evidence of vasculitis or arterial dissection High-grade stenosis that cannot be treated safely or which prevents access to the thrombus with the Trevo device Angiographic evidence of excessive arterial tortuosity that precludes the Trevo device from reaching the thrombus
Cerebrovasc Dis 2013;36:218–225 DOI: 10.1159/000353990
or to the study all endovascular operators completed an in vitro training with the Trevo device as well as a review of the protocol inclusion/ exclusion criteria, requirement to collect neuroimaging to send to the core lab, and familiarity with a standard angiography acquisition protocol. Only CE-marked devices were used in the study. Inclusion and Exclusion Criteria The inclusion criteria for the study were diagnosis of an acute ischemic stroke involving occlusion of a proximal artery (ICA, M1, M2 or vertebrobasilar) with an NIHSS score of 8–30 in a subject between 18 and 85 years of age who could be treated within 8 h from symptom onset (see table 1 for a complete listing of inclusion/exclusion criteria). Screening, Consent and Enrollment Patients were enrolled consecutively in the participating centers. All patients presenting with stroke symptoms were screened at each site to determine if they met entry criteria for the study. After obtaining signed informed consent, all eligible patients were evaluated for neurological and functional status, and an angiographic assessment of the suspected vascular occlusion. For patients who met all of the inclusion criteria and none of the exclusion criteria and who were eligible for the study, the following baseline data were collected: medical history and admission details, patient demographics/characteristics, CT or MR neuroimaging scan, baseline blood values, baseline NIHSS assessment and prestroke mRS assessment. Patients who did not qualify for participation in the study because they did not meet the inclusion/exclusion criteria were documented as screen failures, and were excused from further study participation. Enrollment in this study was defined as the moment when the Trevo device was deployed through the microcatheter and enters the patient’s neurovasculature. During the study, no patients were treated by the investigators with the TREVO device outside the trial. Trevo Procedure Using appropriate (general or local) anesthesia, transfemoral arterial access was obtained per standard practices at the treating institution. A diagnostic angiogram was performed in order to determine the appropriateness of the occlusion for treatment with the Trevo device. Visual estimation of the vessel diameter and occlusion location was recorded. Prior to the start of the procedure, the TICI score in the vascular territory being treated was assessed. If thrombus was identified in a treatable vessel, defined as the internal carotid, middle cerebral M1 and/or M2 branches, basilar or vertebral arteries, the Trevo thrombectomy procedure was initiated. For this, a microcatheter was navigated through the occlusion and its tip was placed distally to the thrombus. The Trevo device was delivered through the Trevo microcatheter and the shaped section was unsheathed via pulling back on the microcatheter, leaving the device directly in the thrombus and allowing it to immediately begin expanding and incorporating into the thrombus, thereby anchoring the device to the thrombus and facilitating retrieval of the thrombus into the guide catheter. The intra-arterial dwell time to let the device struts integrate into the thrombus was 3–5 min. During the pull-back maneuver, aspiration was performed either via a distal access catheter (DAC 0.0057) placed in the distal
Jansen et al.
Table 2. Time and events schedule
Assessment
Preprocedure Trevo procedure
Consent
✓
Patient demographics/characteristics
✓
Medical history and admission details
✓
Baseline labs
✓
CT or MRI
✓1
NIHSS
✓
Angiogram
✓
TICI and TIMI scores
✓
Unscheduled follow-up
Day 7–10 or day of discharge (whichever occurs first)
✓1
✓2
✓
✓
✓
✓
✓
✓ ✓
✓ ✓
✓ ✓
✓2
1
✓
Procedure details mRS AEs
90 days (±14 days)
24 h (–6/+12 h)
✓3
✓
1 Images are de-identified and submitted for archiving and core lab review. 2 If clinically indicated, e.g. a significant neurological worsening, or suspicion of an intracranial hemorrhage. 3 Preprocedure mRS is based upon patient’s prestroke abilities and life style. AEs = Adverse events.
ICA (C1–3) or proximal M1 or via a balloon occlusion guide catheter (typically 8F) placed in the proximal ICA. The same Trevo device could be used several times to a maximum of six retrievals; however, after each deployment of the device, it was thoroughly inspected and cleaned before reloading. Use of the Trevo device was terminated if there was any angiographic evidence leading to the suspicion of an intracranial hemorrhage, e.g. extravasation of contrast during the procedure. At the end of the Trevo procedure, biplane angiography was performed immediately afterwards in order to document the TICI score in the vascular territory being treated. Investigators were instructed to use Trevo as the first and primary device, but if the device was unsuccessful, the operator had the option of using other adjunctive treatment, including intra-arterial thrombolysis, if they had performed at least one Trevo pass. Postprocedure Follow-Up Patients were managed postprocedure according to each institution’s usual practice. It was recommended that administration of anticoagulants and antiplatelets be suspended for 24 h postthrombectomy in patients who were not directly in need of these agents. At 24 (–6/+12) h postprocedure, the following study assessments were performed: NIHSS score, CT or MR imaging to assess the presence/absence of intracranial hemorrhage, adverse events and adverse device effects were documented. At day 7–10, or at discharge if earlier, the NIHSS score and mRS score was evaluated and again at day 90 (±14 days) postprocedure. Time and events schedule is shown in table 2 in more detail.
propriately de-identified, and transmitted for data archiving and central core lab evaluation. For each enrolled subject in the trial, the angiography dataset was reviewed to ensure complete content of the diagnostic and therapeutic phases of the procedure. No information other than personal identifiers was removed from the dataset prior to independent angiography core lab review. The core lab utilized the modified TICI grading scale for assessment of the primary outcome measure. Revascularization outcomes immediately after use of the Trevo device and postprocedure (if additional adjunctive/rescue treatments were done) were adjudicated. Additional information acquired by the core lab included potential angiographic complications including perforation, dissection, vasospasm, and contrast extravasation. Study Monitoring Under the supervision and at the direction of Concentric Medical, a monitor conducted on-site monitoring to ensure that all sites were in compliance with the Declaration of Helsinki 23, the Clinical Study Plan, the Clinical Study Agreement, Ethics Committee requirements, and applicable local laws and regulations. Standardized case report forms were completed for all patients enrolled in the study. Source data verification was performed by the monitor for key data points throughout the study.
Results
Core Lab All enrolled patient’s imaging data, including baseline and postprocedure CT/MRI and the full angiography datasets were ap-
Enrollment began in February 2010 and was completed in August 2011, with the last patient follow-up completed in November 2011.
Neurothrombectomy for the Treatment of Acute Ischemic Stroke
Cerebrovasc Dis 2013;36:218–225 DOI: 10.1159/000353990
221
60 patients were enrolled among the seven participating sites. No patients were lost to follow-up and a 90-day mRS was obtained for all surviving subjects. In 58 cases angio core lab reads were completed, in 2 cases site-reported angio data were used due to technical reasons that precluded the transfer of the images to the Core Lab. Table 3 summarizes the results on patient demographics, baseline stroke score and site of target vessel. The suspected stroke etiology included atrial fibrillation in 41.7%, large artery atherosclerosis in 20%, other cardioembolic disease in 9% and other/unknown reason in 28.3%. Time from symptom onset to first pass with the Trevo retriever was 3.9 ± 1.4 h (mean ± SD). In 45%, the femoral arterial puncture to start the procedure was performed within the first 3 h, in 31.7% this puncture was between hours 3 and 4.5 and in 23.3% this was done between hours 4.5 and 8. 60% of the patients (36/60) had received intravenous r-tPA prior to the endovascular procedure with a mean dosage of 63.9 mg r-tPA. The mean time from symptom onset to start with intravenous thrombolysis in this subgroup was 1.9 h. Periprocedural adjuvant treatment was done with intra-arterial t-PA in 6/60 patients (10%), before thrombectomy extracranial carotid stenting was performed in 6/60 patients (10%) and extracranial angioplasty due to a proximal stenosis was performed in 3/60 patients (5%). Another 3 patients (5%) received a IIb/IIIa inhibitor and in 13/60 patients (21.7%) an additional thrombectomy device was used as second-line endovascular treatment. The interventionalist performed a diagnostic angiogram after deployment of the Trevo device, showing a perfusion channel in 73.5% (111/151 passes). The mean dwell time to integrate the struts of the device into the thrombus was 2.62 ± 1.50 min. A mean of 2.62 ± 1.37 passes with the Trevo device were performed (range 1–7). Mean time from femoral arterial puncture to the end of the procedure was 86.6 ± 44 min with a mean treatment time (introduction of Trevo device to end of the procedure) of 59.8 ± 40 min. The sites reported all cases starting with either TICI 0 (98.3% 59/60) or TICI 1 (1.7% 1/60). Evaluation of the initial angiograms by the Core Lab showed a complete vessel occlusion (TICI 0) in 95% (57/60), a TICI 1 score in 1.7% (1/60) and a TICI 2a score in 3.3% (2/60). At the end of the Trevo passes, the Core Lab reading of the an222
Cerebrovasc Dis 2013;36:218–225 DOI: 10.1159/000353990
Table 3. Baseline subject demographics and stroke characteristics
Characteristics
TREVO (n = 60 subjects)
Age, years Mean ± SD Median (range) % male gender NIHSS Mean ± SD Median (range) 8–14 15–22 23–30 Baseline mRS 0 1 Clot location (per core lab) ICA M1 M2 VB Suspected stroke etiology Large artery atherosis Cardioembolic Afib Other Unknown Other
64.7±13.4 65.2 (21.5–84.5) 45.0 17.7±4.8 18.0 (8.0–28.0) 25.0% 56.7% 18.3% 78.3% 21.7% 21.7% 60.0% 10.0% 8.3% 20.0% 51.7% 41.7% 8.3% 25.0% 3.3%
Afib = Atrial fibrillation.
giogram showed a recanalization (TICI 2a, 2b and 3) in 90.0% (54/60). At the conclusion of the entire procedure, including any adjunctive treatment, the recanalization rate was 91.7% and TICI 2b/3 was achieved in 78.3%. Recanalization rate by target vessel is shown in table 4. Clinical Outcome At 90 days, 55% of the patients had a favorable neurological outcome (mRS 0–2) and 20% had died. Favorable outcome was very often apparent at 24 h with a decrease of the median NIHSSS from initial 18 to 9. At day 7 the NIHSSS was decreased again to a median of 4.0. Patients with successful recanalization (TICI 2a,b/3) had a good 90-day neurological outcome (mRS 0–2) in 60%, whereas no patient without recanalization had a mRS less than 3 at day 90. Mortality at day 90 in successfully recanalized patients was 18 and 40% in patients without recanalization. Figure 2 shows the outcome results in more detail and by occlusion location. The mean length of hospital stay was 8.2 ± 2.9 days with a range from 2 to 22 days. Jansen et al.
Outcome measure TICI, preprocedure, by core lab 0 1 2a TICI, immediately after Trevo, by core lab 0 1 2a 2b 3 TICI, end of procedure, by core lab 0 1 2a 2b 3 Postdevice revascularization success (TICI 2a, 2b, or 3) End of procedure revascularization success (TICI 2a, 2b, or 3)
TREVO (n = 60 subjects) 95.0 1.7 3.3 6.7 3.3 18.3 65.0 6.7 3.3 5.0 13.3 70.0 8.3
ICA (n = 13)
92% 93% 80%
80%
MCA (n = 40) VB (n = 5)
54% 52%
23%
Revasc
mRS 0–2
19% 20%
mRS 6
Fig. 2. Revascularization and 90-day mRS by occlusion location.
90.0 91.7
mRS
7 days (n = 59)
90 days (n = 60)
0 1 2 3 4 5 6 mRS 0–2 10+ improvement in NIHSS* At 24 h At 7 day At 90 day
13.6 18.6 13.6 6.8 20.3 16.9 10.2 45.8
15.0 33.3 6.7 15.0 10.0 0.0 20.0 55.0
42.6 55.4 68.3
* If baseline NIHSS was 10 or less, then a score of 0 or 1 was counted as success. Expired subjects are scored as a 42 for this analysis. Values are percentages.
Complications Serious procedure-related adverse events were reported in 5/60 patients (8.3%). In 3 of these patients (5%) the SAEs were also adjudicated as device-related: two early symptomatic intracranial hemorrhages SICH (perforation, parenchymal hemorrhage grade 2 and SAH; hemorrhagic infarction grade 1 and SAH) and one late SICH. The overall SICH rate, using the pre-specified SITSMOST definition was 5% (3/60); only one of them was Neurothrombectomy for the Treatment of Acute Ischemic Stroke
% 100 90 80 70 60 50 40 30 20 10 0
Color version available online
Table 4. Primary results
device related. The SICH rate, as defined by the ECASS III criteria, was 8.3% (5/60). There were four other nonserious device-related events (1 minor vessel perforation, and three asymptomatic hemorrhages). Adjunctive Treatment In 13 (21.7%) of the cases an adjunctive device as a second-line treatment was used. Among the 13 cases, 9 cases had a single adjunctive device used and 4 had more than one device used. The devices used were Solitaire (11 cases), Merci (2 cases), Phenox (2 cases) and Penumbra (2 cases). Comparing the revascularization rates after Trevo alone to the final (postadjunctive), 5 had the TICI scores improve by at least 1 grade (3 ‘unsuccessful’ cases became ‘successes’ while the other 2 remained a success (n = 1) or a failure (n = 1)). Six cases remained unchanged following adjunctive use, and 2 cases got worse (both were TICI 2a following Trevo, but resulted in TICI 0 or 1 after adjunctive treatment). The net change from postTrevo recanalization (90%) to final recanalization (91.7%) was only 1.7%, but the 90-day outcomes for these subjects getting adjunctive treatment was much worse than the overall cohort (mRS 0–2 = 23% (3/13); mortality = 38% (5/13)).
Discussion
This study showed a recanalization rate (TICI 2b and 3) of 78% with good clinical outcome (mRS 90: 0–2) in 55% in patients with acute occlusion of greater intracranial vessels who were treated with the TREVO system. Cerebrovasc Dis 2013;36:218–225 DOI: 10.1159/000353990
223
Following international guidelines intravenous thrombolysis is still the only accepted treatment within the first 4.5 h after stroke onset. However, intravenous thrombolysis is known to be of limited value in proximal arterial occlusion, particularly in cases with extensive clot burden. Recanalization success on the other hand has been shown to be the most important variable for good clinical outcome. Rha and Saver [2] were able to demonstrate in a meta-analysis of 55 studies with more than 2,000 patients that successful recanalization has an OR for good clinical outcome (mRS 90
