Background No randomized trial has investigated the effect of mechanical thrombectomy (MT) alone in patients with acute stroke. There are conflicting results as to whether prior intravenous thrombolysis (IVT) facilitates subsequent MT, and data in patients treated with MT alone owing to contraindications to IVT are limited.
Objective To compare consecutive patients treated with MT alone or with preceding IVT in a large tertiary neurointerventional center, with special emphasis on contraindications to IVT.
Methods Retrospective analysis of 283 consecutive patients with acute ischemic stroke treated with MT in a tertiary neurovascular center over 14 months. Data on characteristics of periprocedural times, recanalization rate, complications, and long-term functional outcome were collected prospectively.
Results Information on prior IVT and functional outcome was available in 250 patients. Mean (SD) follow-up period was 5.7 (5.1) months and 105 (42%) patients received both IVT and MT. No significant differences were found in successful recanalization rates (Thrombolysis in Cerebral Infarction (TICI) 2b/3, 73.8% vs 73.1, p=0.952), complication rates, and long-term favorable outcome (modified Rankin Scale 0–2, 35.2% vs 40%, p=0.444) between patients receiving MT plus IVT and those receiving MT alone. A favorable outcome in patients directly treated with MT alone who were eligible for IVT was achieved in 48.2%. Thrombectomy was safe and resulted in a favorable outcome in 32% of patients with absolute contraindications to IVT.
Conclusions Preceding use of IVT was not an independent predictor of favorable outcome in patients with acute stroke treated with MT and complication rates did not differ whether or not IVT was used. MT is safe and achieved a favorable outcome in one-third of patients with stroke ineligible for IVT.
Statistics from Altmetric.com
Recent publications provide evidence for a second treatment option in patients with acute ischemic stroke in addition to the well-established intravenous thrombolysis (IVT) with recombinant tissue plasminogen activator.1 Mechanical thrombectomy (MT) with stent retrievers was shown to be superior to IVT alone in five randomized clinical trials in patients with occlusion of a large artery in the anterior brain circulation.2–6 Preceding or ‘bridging’ IVT was a prerequisite in two of these five randomised MT trials,4 ,5 and most patients included in the other three randomized trials were also treated with both IVT and MT.2 ,3 ,6 Subgroup analyses in these trials did not show significant different therapeutic effects between patients with or without bridging IVT, but the number of patients treated with MT alone was small in each trial. In contrast, data from three retrospective studies suggested a better functional outcome in patients treated with MT and bridging IVT.7–9 Moreover, a substantial number of patients with stroke are not eligible for IVT owing to an extended period of time after known symptom onset, unknown onset of stroke symptoms (eg, wake-up stroke), or contraindications to IVT (eg, recent major surgery or stroke, anticoagulation, history of major hemorrhage). But MT alone might still be a possible treatment in such patients with proven large artery occlusion. No randomized trial has investigated the effect of MT alone or in comparison with bridging IVT in patients with or without contraindications to IVT, and the randomized thrombectomy in patients ineligible for IV tissue plasminogen activator (tPA) (THRILL) trial was recently stopped with only a few patients included.10
The aim of this single-center retrospective analysis was to describe differences in long-term functional outcome and intraprocedural complications between patients treated with MT alone or with bridging IVT, taking contraindications to IVT into account.
Between June 2012 and August 2013, all consecutive patients with acute ischemic stroke treated with MT in our tertiary neurovascular center were prospectively documented using a standardized case report form, which included demographics, vascular risk factors, periprocedural time management and complications, and data on hospital outcome. MT was performed by five experienced neurointerventionalists according to criteria established by neurologists and neurointerventionalists in the Neurovascular Net Ruhr.11 IVT use in eligible patients with stroke before MT was at the discretion of the treating neurointerventionalist and neurologist since it was unclear at this time point whether prior IVT might lead to higher bleeding rates or vessel complications in patients receiving MT. IVT was initiated within 4.5 h after stroke symptom onset in eligible patients secondarily referred for MT in the initial neurological hospital (‘drip-and-ship’) or at our institution in directly admitted patients. The presence of intracranial hemorrhage was assessed on brain imaging performed between 20 and 30 h after MT. Symptomatic intracranial hemorrhage (sICH) was defined as any intracranial hemorrhage seen on brain imaging associated with an increase of four points or more on the National Institutes of Health Stroke Scale (NIHSS) score, according to the European Cooperative Acute Stroke Study (ECASS)-3 criteria.12 Data quality was monitored centrally by the Institute of Epidemiology and Social Medicine of the University of Münster for completeness and consistency. Information on functional outcome using the modified Rankin Scale (mRS) was assessed at 3 months with mailed questionnaires by a blinded investigator of the Institute of Epidemiology and Social Medicine at the University of Münster in all patients who had given informed written consent. For patients unable to respond to the questionnaire, mRS status was assessed by phone by one local investigator (JH). If no follow-up information could be obtained, a query was sent out to the local death registry to assess vital status. The study was approved by the ethics committees of the Chambers of Physicians of Westphalia and North Rhine in Germany.
Baseline characteristics, periprocedural times, intraprocedural complications, and outcome parameters were compared between patients receiving thrombectomy with and without IVT, and between subgroups of patients eligible for IVT but primarily treated with thrombectomy alone and those patients receiving thrombectomy alone owing to contraindications to IVT, using the Mann–Whitney-U test or χ2 test, as appropriate. Given percentages are based on patients with available information. To determine independent predictors of favorable clinical outcome (defined as mRS 0–2), a multivariate binary logistic regression including all variables with a p value of <0.1 in univariate analysis was performed. These variables were age, NIHSS score at admission, arterial hypertension, diabetes mellitus, current smoking, atrial fibrillation, occlusion of the internal carotid artery, time from symptom onset to first cerebral imaging, time from symptom onset to end of the thrombectomy procedure, successful reperfusion (defined as Thrombolysis in Cerebral Infarction (TICI) score 2b or 313), number of passes of the thrombectomy device, and sICH. A p value of <0.05 was considered significant. All statistical analyses were performed with SPSS V.21.
A total of 283 patients were treated with MT during the 14-month study period. Information on clinical outcome and use of IVT was available in 250 (88.3%) patients (see table 1 for baseline characteristics). Mean (SD) follow-up period was 5.7 (5.1) months. All baseline demographics including age and stroke severity, time from stroke onset (or patient last seen well) to first cerebral imaging and to groin puncture, duration of angiography, device use, and treatment complications did not differ significantly between patients with and without follow-up information (data not shown). An exception was a significant lower proportion of successful recanalization (74.3% vs 50.0%, p=0.007) and a higher percentage of procedural subarachnoid hemorrhage (2.8% vs 13.4%, p=0.005) in the group without follow-up information.
Of the 250 patients with follow-up information, 210 (84.0%) patients had anterior circulation occlusions (see table 1 for exact site of symptomatic occlusion), and 105 (42%) patients received IVT followed by thrombectomy, while 145 (58%) patients were treated with thrombectomy alone. In the MT alone group, 75 of the patients had contraindications for IVT, while thrombectomy was directly performed in 70 patients who were also eligible for IVT within 4.5 h after stroke symptom onset. The most common contraindications for IVT were wake-up stroke with unknown stroke onset or >4.5 h between known stroke onset and effective anticoagulation (table 2). Patients receiving IVT were significant more often secondarily referred from another hospital for thrombectomy (with IVT started in the referring hospital), had significantly less frequent atrial fibrillation and fewer had a history of previous cerebrovascular events, while other baseline characteristics were not significant different (table 1). Stent retrievers were used in 233 (93.2%) of all patients for thrombectomy with no differences in their frequency between patients receiving IVT or not. Of all interventions, 98% were performed under general anesthesia. CT angiography was the primary brain vessel imaging modality in 88 (84.6%) patients with IVT and 132 (91.7%) patients not receiving IVT (p=0.083). The median time delay between stroke symptom onset and first cerebral imaging was also not statistically different between the two patient groups (77 min vs 87 min, p=0.516).
Procedural characteristics and intraprocedural complications
There was a slight, but not significant, difference between patients receiving both IVT and MT or MT alone in the median time delay between stroke symptom onset and groin puncture (233 min vs 210 min; p=0.103) or end of thrombectomy procedure (280 min vs 272 min; p=0.221) (table 3). The time interval between first brain imaging and groin puncture was significantly longer in patients with prior IVT (148 min vs 112 min, p<0.001). During the time period of 14 months, only one patient receiving prior IVT showed ultra-early recanalization on angiography and no MT was performed.
Successful reperfusion was achieved in 73.8% of patients with IVT, which was similar to the 73.1% in the group with no IVT. Furthermore, there were no significant differences in the median number of passes of the thrombectomy device (1 (range 1–11) vs 2 (range 0–9), respectively, p=0.904). The proportions of important intraprocedural complications were lower in the MT alone group, but none of these differences was statistically significant between the groups (emboli in further vessel territory, 5.7% vs 3.4%, p=0.374; iatrogenic dissection, 6.7% vs 2.8%, p=0.238; subarachnoidal hemorrhage, 2.9% vs 2.8%, p=0.947; table 3). Symptomatic ICH on control brain imaging (5.9% vs 3.5%, p=0.387) and postinterventional occurrence of groin hematoma (1.9% vs 1.4%, p=0.732) also did not significantly differ between the two treatment groups.
Clinical long-term outcome
Favorable functional outcome at follow-up was not significantly different in patients receiving bridging IVT or thrombectomy alone (mRS 0–2, 35.2% vs 40%, p=0.444). Prior use of IVT was not associated with favorable outcome in univariate analysis (p=0.444). Multivariate regression analysis showed that younger age (OR=0.95, 95% CI 0.92 to 0.99; p=0.015), lower NIHSS score on admission (OR=0.90, 95% CI 0.84 to 0.95; p≤0.001), and successful reperfusion (OR=3.67, 95% CI 1.22 to 11.03, p=0.021) were independent predictors of favorable outcome at follow-up.
Comparison of patients who had contraindications to bridging IVT with those who primarily received thrombectomy
Baseline characteristics and vascular risk factors were not statistically different between patients who were eligible for IVT but were primarily treated with thrombectomy and patients with contraindications for IVT (table 1). The median time delays between stroke symptom onset to first imaging (73 min vs 119 min; p=0.001), symptom onset to groin puncture (183 min vs 271 min; p≤0.001) and symptom onset to end of thrombectomy (231 min vs 325 min; p≤0.001) were significantly shorter in patients primarily treated with thrombectomy and no contraindications for IVT. The rates of successful reperfusion (74.3% and 74.0% p=0.966), sICH (2.9% vs 4.1%, p=0.708), and intraprocedural complications did not significantly differ between the groups (table 3). Significantly more patients who were primarily treated with thrombectomy and no contraindications to IVT had a favorable functional outcome at follow-up (48.6% vs 32%, p=0.042).
Our study showed no differences in recanalization rate, intraprocedural complications, and long-term functional outcome in patients receiving MT plus prior bridging IVT versus MT alone. Preceding use of IVT was not an independent predictor of favorable functional outcome in our cohort. Three smaller retrospective studies had reported that bridging IVT facilitated MT in patients with acute stroke.7–9 Pfefferkorn et al7 first described that 26 patients with preceding IVT and successful recanalization were more likely to be functionally independent than 23 patients not receiving IVT (42% vs 13%). Recanalisation rates were similar in both treatment groups (74% and 77%) and similar to those from our study. Owing to the small number of patients included, regression analysis was not performed, and the authors did not report the overall results for all 65 analyzed patients, which might have resulted in selection bias. Furthermore, only 2 of the 49 patients were treated with stent retrievers.
A retrospective multicenter study including our center, comprised 141 consecutive patients with stroke treated with the Solitaire stent through June 2010, also reported a significantly higher favorable functional outcome after 3 months in patients receiving prior IVT (66% vs 42%).8 However, follow-up information at 3 months was available in only 58.2% of the included patients, which was considerable lower than our follow-up rate of 88.3%. No regression analysis was performed to check whether bridging IVT was an independent predictor of favorable functional outcome. Recanalization rates were higher (88% with IVT and 81% without IVT) compared with our study. The median number of stent retriever passes and intraprocedural complications including sICH did not differ between patients with and without preceding IVT and were similar in comparison to our study.
Guedin et al9 analyzed 68 patients with occlusion of the proximal middle cerebral artery (28 of whom received IVT) treated within 6 h after stroke onset with the Solitaire stent and reported a significantly higher recanalization rate (85.7% vs 55%) and a significantly shorter endovascular procedure time in patients with IVT, while the rates of sICH (7.1% vs 5%) and favorable outcome (67.9% vs 52.5%) were not statistically different. In view of the small study size, the authors did not perform multivariable analysis.
In contrast to the aforementioned studies, and in line with our findings, subgroup analyses of three of the five recently published prospective randomized trials with stent receivers in which the use of IVT was not a study prerequisite,2 ,3 ,6 as well as the prospective population-based SONIIA registry13 showed no significantly different therapeutic effects between patients receiving previous IVT or not. However, the number of patients treated with MT alone was small in the randomized trials (n=55 in MR CLEAN, n=45 in ESCAPE, n=33 in REVASCAT) and only patients with anterior stroke were included. In SONIIA, 285 (53.2%) of the analyzed 536 patients received IVT and this registry also included patients with both anterior and posterior stroke.14 The rate of sICH and favorable outcome in SONIIA did not differ between patients with and without previous IVT (4.3% vs 6.8% and 45.8% vs 40.4%, respectively). Similar to our study, previous IVT was not an independent predictor of favorable outcome at 3 months in multivariable logistic regression models (OR=1.14, 95% CI 0.73 to 1.77).14 Leker et al15 also reported no differences in favorable outcome or survival in 57 patients with occlusion of the proximal middle cerebral artery receiving stent-retriever thrombectomy with our without IVT. Similar to our study, only age and NIHSS score at admission were variables associated with outcome in multivariate analysis.
When patients were first included in our registry, it was unclear whether prior IVT might lead to higher bleeding rates or vessel complications in patients receiving MT. Similar to all the aforementioned studies, complication rates were not higher in patients receiving both bridging IVT and MT than in those treated with MT alone in our study. In particular, the rates of sICH and subarachnoidal hemorrhage, embolization in further vessel territories during the thrombectomy procedure, and number of passes of the thrombectomy device required did not differ statistically between patients with and without bridging IVT. This finding is of great importance since it corroborates the safety of combined IVT and MT in patients with acute stroke.
The highest rate of favorable functional outcome in our study was seen in patients with no contraindications for IVT who were directly treated with MT. Median symptom onset to groin puncture time in these 70 patients (of whom 29 were directly admitted at our center and 41 were secondarily referred for MT) was 183 min which was similar to the 185 min in the ESCAPE trial3 and shorter than the median time delays of >200 min in the other four randomized thrombectomy trials,2 ,4–6 and 314 min in the SONIIA registry.14 The median time from initial brain imaging to groin puncture was significantly shorter in patients receiving MT alone without prior IVT in our study. This finding may have been biased by the higher number of patients directly admitted at our center in this patient group. Nevertheless, previous use of IVT might have contributed to this time delay in our patient cohort. The STAR study also reported a significantly prolonged time from baseline imaging to groin puncture in patients receiving prior IVT (32 min mean delay).16 These findings highlight again the importance of starting the thrombectomy procedure as early as possible, not waiting for a potential improvement in patients receiving IVT, and warrant a prospective randomized trial comparing thrombectomy alone with IVT plus MT which is being planned.
The overall lower rate of favorable functional outcome in our study compared with some of the published randomized trials (ESCAPE, EXTEND-IA, SWIFT PRIME) is due to inclusion of consecutive patients with stroke with no selection/exclusion of patients based on age limits, unknown stroke onset, comorbidities premorbid functional status, or enhanced brain imaging parameters.3 ,4 ,5 Long-term functional outcome in the overall study cohort was similar to that found in SONIIA14 and better than the outcome in the MR CLEAN trial, which had a pragmatic study design resulting in inclusion of a wide range of patients with stroke.2
Patients with acute stroke with contraindications to IVT are of considerable interest since MT is the only acute treatment option for these patients. No data from randomized trials are available in this patient group and enrollment in the THRILL trial was recently stopped with only a few patients included.10 Compared with a multicenter retrospective observational study which included 130 patients with anterior stroke who were ineligible for IVT,17 recanalisation rate (74% in our study vs 77%) and favorable functional outcome (32% in our study vs 37.9%) were slightly lower in patients with contraindications to IVT in our study, while sICH occurred less frequently (4.1% in our study vs 13.1%). Concomitant diseases including recent ischemic or hemorrhagic stroke are likely to have contributed to the worse outcome in patients with contraindications to IVT. Furthermore, patients with contraindications to IVT had the longest median symptom onset to groin puncture time (271 min), which was also substantially longer than the 175.7 min (mean) in the study from Dorn et al.17
Our prospective observational single-center study has several limitations. Imaging data were not evaluated by a blinded investigator, whereas long-term functional outcome was evaluated in the majority of patients by a blinded investigator. Owing to inclusion of consecutive patients and no randomization, there is the inherent risk of selection bias and only indirect comparisons between patients with and without bridging IVT in our study and with the results of other studies are possible. Furthermore, the group of patients who did receive MT alone without bridging IVT was heterogeneous for several reasons (patients primarily admitted at our center or secondarily referred; patients who were eligible for IVT but were treated with MT alone; patients with contraindications for IVT; inclusion of patients with pre-stroke mRS ≥3 and severe comorbidities).
Our prospective observational study showed no differences in intraprocedural complications or long-term functional outcome in patients with stroke receiving MT with or without bridging IVT. Preceding use of IVT was not an independent predictor of favorable outcome in patients treated with MT. Favorable functional outcome in patients directly treated with MT without prior IVT who had no contraindications for IVT was achieved in almost 50% of our patients. Thrombectomy alone was safe and resulted in favourable outcome in one-third of patients with stroke with absolute contraindications to IVT and large brain artery occlusion and appears to be a valuable treatment option for these patients.
Contributors RW was involved in the design of the study, monitored data collection for the whole study, acquired patient data, wrote the statistical analysis plan, analyzed and interpreted the data, drafted and revised the paper. He is guarantor. HN was involved in the design of in the design of the study and patient data acquisition, analyzed and interpreted data, and revised the draft paper. JH wrote the statistical analysis plan, collected, analysed, and interpreted data, and revised the draft paper. MH, MS, and PS collected, analysed, and interpreted data and revised the draft paper. KB was involved in the design of the study, monitored data collection, analyzed and interpreted data, and revised the draft paper. RC was involved in the design of the study, analyzed and interpreted data and revised the draft paper.
Competing interests RW received speaker honoria from Covidien and Boeheringer Ingelheim and from serving on a scientific advisory board of Covidien. HN received speaker honoria from Covidien and Microvention. PS received consulting fees from Acandis and Covidien, speaker honoraria from Microvention, and proctering agreement from Balt and Microvention. RC received consulting fee or speaker honoria from Balt, Medtronic/Covidien, Microvention, Neuravi, Siemens and Stryker.
Ethics approval Ethics committees of the Chambers of Physicians of Westphalia and Northrhine, Germany.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement The data discussed in this article are taken from our institution. Unpublished anonymized data may be available on request to the corresponding author.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.