Article Text


Research paper
Outcomes of stent retriever thrombectomy in basilar artery occlusion: an observational study and systematic review
  1. Benjamin Gory1,
  2. Islam Eldesouky1,
  3. Rotem Sivan-Hoffmann1,
  4. Murielle Rabilloud2,
  5. Elodie Ong3,
  6. Roberto Riva1,
  7. Dorin Nicolae Gherasim1,
  8. Alexis Turjman1,
  9. Norbert Nighoghossian3,
  10. Francis Turjman1
  1. 1Department of Interventional Neuroradiology, Hôpital Neurologique Pierre Wertheimer, Bron, France
  2. 2Department of Biostatitics, UMR 5558, Hospices Civils de Lyon, Lyon, France
  3. 3Department of Neurology, Stroke Unit, Hôpital Neurologique Pierre Wertheimer, Bron, France
  1. Correspondence to Dr Benjamin Gory, Department of Interventional Neuroradiology, Hôpital Neurologique Pierre Wertheimer, 59 Boulevard Pinel, Bron 69677, France; benjamin.gory{at}


Background Basilar artery occlusion (BAO) remains one of the most devastating subtypes of stroke with high mortality and poor outcome. Early recanalisation is the most powerful predictor of favourable outcome in patients with stroke, and may be improved with mechanical thrombectomy using stent retriever devices. However, the benefit in functional outcome and safety of stent retrievers are not yet well known. The aim of this study was to assess efficacy and safety profiles of stent retriever thrombectomy in BAO patients with stroke.

Methods We analysed data retrospectively from our consecutive clinical series and conducted a systematic review and meta-analysis of all previous studies of stent retriever thrombectomy in BAO patients with stroke between November 2010 and April 2014.

Results From March 2010 to March 2013, 22 patients with acute BAO were treated with a Solitaire stent retriever in our series. Favourable outcome was significantly associated with younger age and distal BAO. The literature search identified 15 previous studies involving a total of 312 subjects. In the meta-analysis, including our series data, the recanalisation rate (Thrombolysis In Cerebral Infarction (TICI) score ≥2b) reached 81% (95% CI 73% to 87%). The rate of symptomatic intracranial haemorrhage was 4% (95% CI 2% to 8%), favourable outcome (modified Rankin Scale (mRS) ≤2 at 3 months) was found in 42% (95% CI 36% to 48%) and mortality rate was 30% (95% CI 25% to 36%).

Conclusions Stent retriever thrombectomy is a safe treatment modality for patients with stroke presenting with BAO. Although the stent retrievers showed a good recanalisation rate, there are currently no randomised clinical trials to assess its clinical efficacy in comparison with the reference treatment.

View Full Text

Statistics from


Strokes secondary to acute basilar artery occlusion (BAO) are associated with high mortality and poor outcome.1 ,2 Early recanalisation has been demonstrated to be the major goal of acute ischaemic stroke therapy.3–5 Although intravenous recombinant tissue plasminogen activator (rt-PA) remains the first-line treatment, its efficacy is limited by the low recanalisation rate in the setting of BAO.6 In patients with acute ischaemic stroke caused by a proximal intracranial occlusion of the anterior circulation, early mechanical thrombectomy (MT) using stent retriever devices, as compared with intravenous rt-PA alone, improved functional outcome.7–9 However, in acute BAO patients with stroke, MT benefit has not yet been proven, although it seems to be a promising adjuvant or stand-alone therapy.10–24 Pending the results of future randomised controlled trials,25 the benefit and safety of stent retrievers MT remain to be determined in patients with stroke with acute BAO.

We therefore analysed data from our clinical consecutive series and conducted a systematic review and meta-analysis to assess efficacy and safety profiles of stent retriever thrombectomy in BAO patients with stroke.


Neurological hospital series

All consecutive patients treated with a Solitaire device (Covidien/ev3, Dublin, Ireland) at the Lyon Neurological Hospital between 1 March 2010 and 1 March 2013 were retrospectively analysed. Patients received intravenous rt-PA according to current guidelines (0.9 mg/kg), and a systematic approach including MT was implemented for patients. In patients with a contraindication to rt-PA, a direct MT approach was considered.

Patients were eligible for MT if they (1) had a clinical diagnosis of acute stroke in the posterior circulation; (2) were admitted within 24 h after onset of symptoms; (3) had a significant clinical deficit following physician evaluation (no National Institutes of Health Stroke Scale (NIHSS) limit); (4) underwent an MRI or CT before treatment (no pc-ASPECTS limit); (5) presented with acute basilar ischaemia and BAO in MR or CT angiography. Patients were excluded if they (1) were pregnant; (2) were aged <18 years; (3) had cerebral imaging that revealed an intracranial haemorrhage (ICH), tumour or a subacute infarct.

Systematic review

We searched MEDLINE between 1 November 2010 and 1 April 2014 using the following search terms: “acute basilar artery occlusion”, or “stroke”, or “stent retriever”, or “solitaire”, or “revive,” or “trevo”, or “mechanical thrombectomy”. Inclusion criteria were the following: ≥10 patients, English language, assessed data on recanalisation rates and clinical outcomes for patients with BAO only. All observational or interventional studies that met these criteria were included regardless of the study design.

Two authors (BG and DNG) identified potentially relevant articles based on the title and abstract and obtained the full text for a detailed review. The electronic search was supplemented by obtaining the full text for a detailed review and by emailing the authors for collection of additional data.

Data analysis

Successful recanalisation was defined as ‘Thrombolysis In Cerebral Infarction’ (TICI) ≥2b revascularisation of the basilar artery. Favourable outcome was defined as modified Rankin Scale (mRS) ≤2 at 90 days. Mortality was evaluated at discharge or 90 days as determined by each author. According to European Cooperative Acute Stroke Study II (ECASS-II),26 symptomatic intracranial haemorrhage (sICH) was defined as any parenchymal haematoma, subarachnoid haemorrhage or intraventricular haemorrhage associated with a worsening of the NIHSS score by ≥4 within 24 h. These criteria were commonly used for anterior circulation strokes; however, since there are no standardised criteria for posterior circulation sICH, we used the same criteria as mentioned.


The percentage of patients presenting the outcome was estimated for each study included in the systematic review and for each outcome. The 95% CIs of the estimates were built using the Wilson method. The estimate and the 95% CI of the mean percentage over all the studies were obtained for each outcome using a logistic mixed model with a random effect on the intercept in order to take into account the heterogeneity between the studies. All the analyses were carried out using the statistical software R, V.3.1.0. We prepared this report with reference to MOOSE (meta-analyses of observational studies) guidelines.


Neurological hospital series

From March 2010 to March 2103, 22 consecutive patients suffering from acute BAO stroke were treated with a Solitaire stent retriever. The baseline characteristics, recanalisation rates, time to recanalisation and clinical outcomes are presented in online supplementary table S1. Ten patients were admitted in a state of coma, and the other 12 patients’ mean NIHSS was 13.6 (range, 4–22). The procedure was performed under general anaesthesia in 16 patients (73%) and conscious sedation in 6 patients (27%). The average number of attempts with the device was 1.7 (range, 1–5). The median time to recanalisation was 8 h after symptoms onset (range 2:50–12:45). Nine patients (41%) had an underlying atherosclerotic stenosis (≥70% of normal diameter), eight of them treated by angioplasty and one by stenting. Recanalisation (TICI 2b or 3) was achieved in 73%. Overall, 27% had a 90-day favourable outcome (mRS score 0–2), 45% died and 14% had haemorrhagic complications within 24 h. In the group of patients admitted in a coma, six patients died, three had an unfavourable outcome and one a favourable one. In the other 12 patients, there were 4 deaths, 3 unfavourable outcomes and 5 favourable ones.

Favourable outcome was significantly associated with younger age (47 vs 66 years; t test p=0.0003) and with third distal BAO localisation (41% vs 0%, Fisher's exact test p=0.05). Outcome was also better in patients with cardioembolic aetiology, but this did not reach significance (50% vs 20%; Fisher's exact test p=0.29). In 5 out of the 6 patients with a favourable outcome, recanalisation was achieved more than 6 h after symptoms onset.

Systematic review

The literature search identified 191 citations. After reviewing the titles and abstracts, 113 articles were read in full and 15 were judged eligible for inclusion (figure 1). In one of the included studies, only recanalisation and mortality rates could be extracted, since the main purpose was to describe all patients treated by the Solitaire stent.11 The included studies, which enrolled 312 subjects (mean sample size, 21; range, 10–36), were conducted in Europe, South Korea, the UK and the USA. All of the studies were case series; most of them (n=9) were classified as retrospective case series analysis.

Figure 1

Flow diagram showing screening and selection of studies for systematic analysis.

As shown in table 1, most studies used bridging therapy with intravenous thrombolytic before MT (mean, 38.7%; range, 0–87; 10 studies) and the vast majority used the solitaire stent retriever device. In two studies, both solitaire and revive were used, while one used only the revive stent. Combined methods (Penumbra, intra-arterial rt-PA, thromboaspiration) were used with stent retrievers in three studies. The admission NIHSS score ranged between 11 and 26.3 (10 studies). The mean time to recanalisation was 8 h (range, 5.5–13.6; 10 studies).

Table 1

Characteristics of included studies

Table 2 shows the pooled estimates of the rate of recanalisation and clinical outcomes. In the meta-analysis, including the data from our centre, the overall recanalisation (TICI score ≥2b) rate was 81% (95% CI 73% to 87%; 15 studies) with a significant heterogeneity (p=0.03), and clinical outcomes were favourable (with a modified Rankin Scale score ≤2) at 3 months in 42% (95% CI 36% to 48%; 14 studies; figure 2). The recorded mortality was 30% (95% CI 25% to 36%; 15 studies) either on discharge or after 3 month follow-up, and the symptomatic ICH was 4% (95% CI 2% to 8%; 15 studies; figure 3).

Table 2

Pooled rates of recanalisation and clinical outcomes

Figure 2

Crude ORs for (A) recanalisation, and (B) favourable outcome.

Figure 3

Crude OR for (A) mortality, and (B) symptomatic intracranial haemorrhage.

We found no evidence of major publication and selection bias by examining the funnel plots of Freeman-Tukey transformed proportions against their SE (see online supplementary figure S1).


Our systematic review and meta-analysis including our observational series data suggest high successful recanalisation rates and acceptable favourable outcome rates for acute BAO patients with stroke treated with stent retriever devices. The overall recanalisation rate was 81% (95% CI 73% to 87%), and the pooled estimate of favourable clinical outcome was 42% (95% CI 36% to 48%; 14 studies).

As reported previously, complete and early recanalisation is the primary objective to be achieved in the setting of acute intracranial occlusion, since a strong association between recanalisation rates and improved functional outcomes has been demonstrated in patients with stroke treated with either intravenous rt-PA or an endovascular approach.3–5 ,26 ,27 The reported frequency of recanalisation after stent retrievers (81%) exceeds that of previously published recanalisation rates of 65% or 68% after intravenous rt-PA and concomitant heparin, respectively.28 ,29 These findings are in accordance with recent data from the ENDOSTROKE registry, which showed arterial recanalisation in 79% of patients with a 90-day favourable outcome of 34%.30 Compared to the last generation of mechanical devices, the stent retrievers improve the rate of recanalisation and favourable clinical outcomes in large vessel occlusion strokes of predominantly anterior circulation.31 ,32 In the SWIFT trial, recanalisation was achieved more often in the Solitaire group than it was in the Merci group (61% vs 24%, p=0.0001) and more patients had a good 90-day neurological outcome with Solitaire than with Merci (58% vs 33%, p=0.02).31 Similar advantages over the Merci device were recently reported in a randomised comparison trial with another stent retriever device, the Trevo.32 However, none have been evaluated in a head-to-head basis versus intravenous rt-PA in the setting of BAO.

Patients successfully recanalised with stent retrievers do not systematically present a favourable functional outcome. This finding may probably be explained by differences in admission characteristics of the patients.33 ,34 In fact, severe neurological condition on admission, with reduced consciousness and already permanent brain tissue damage, may lessen the benefits of recanalisation and hamper prognosis. It was demonstrated that poor 1-month outcome after BAO can be reliably predicted by older age, absence of hyperlipidaemia, presence of prodromal minor stroke, higher NIHSS score and longer time to treatment.34 Two main criteria that seemed to impact directly on clinical outcomes were (1) baseline ischaemic imaging lesions; and (2) time to recanalisation. However, such data were rarely described in the series we analysed or some studies did not provide full details of some baseline characteristics likely to be important for outcome, for example, hyperlipidaemia, pc-ASPECTS (posterior circulation Acute Stroke Prognosis Early CT score), NIHSS score or time to recanalisation. Although we recently reported substantial reversal of early ischaemic damage in posterior circulation after Solitaire thrombectomy,35 the pretreatment infarct size seems to be a potent prognostic factor after either intravenous rt-PA or an endovascular therapy.36 ,37 In the prospective series of 184 consecutive patients with BAO treated with intravenous rt-PA, successful recanalisation led to a functionally favourable 90-day outcome in 50% of patients with no extensive brain infarction (pc-ASPECTS ≥8), compared with 5.9% of patients with extensive brain infarction (pc-ASPECTS <8; p<0.001), irrespective of recanalisation.28 However, MRI was performed at baseline in only 44% of patients and recanalisation status was evaluated approximately 24 h after thrombolysis. Baseline imaging for patient selection varied between studies, and may explain the relatively wide variability of the clinical prognosis observed in the present meta-analysis. Furthermore, there is a relationship between the time from symptom onset to arterial recanalisation and clinical effectiveness of recanalisation procedure. In fact, the BASICS registry patients had an increased risk of poor functional outcome as the time to recanalisation therapy became longer.38 A significantly increased risk of poor outcome seems to occur when recanalisation therapy is started >6 h after the estimated time of BAO. However, although the time from symptom onset to recanalisation is associated with good clinical outcome, recanalisation of BAO up to 48 h after onset was found to be beneficial as well in the absence of extensive baseline ischaemia.28

Although the mortality rate is not negligible, MT in patients with BAO did not impact negatively on clinical outcomes. We showed a pooled estimate of 30% (95% CI 25% to 36%; 15 studies) mortality without significant heterogeneity (p=0.34). The frequency of recanalisation could partly explain the relatively low mortality of patients with BAO treated with stent retrievers. In fact, the majority of patients with no or minimal recanalisation died after intravenous rt-PA or endovascular approach.1 ,6 ,28–30 ,34 In addition, stent retrievers achieved faster recanalisation, whereas a time-dependent relationship was reported between time to recanalisation therapy and risk of death.38 Similar safety profiles between endovascular therapy after intravenous rt-PA and intravenous rt-PA alone in the management of patients with large intracranial artery occlusions was also recently reported in the IMS-3 trial.39

The rate of sICH in the BASICS registry was 6% after intravenous rt-PA and 14% after intra-arterial therapy,1 whereas it was observed in only 4% after stent retrievers. This higher frequency of sICH could also explain the higher fatal evolution as it was previously reported to be independently associated with death, especially after intravenous rt-PA (OR, 14.6; 95% CI 1.4 to 157).29 However, this frequency of sICH may predominantly be caused by full-dose anticoagulation started concomitantly.

The limitations of this study are due to the fact that the articles included in the systematic review contained both prospective and retrospective studies with a limited number of patients; some included combined anterior and posterior circulation occlusion patients with stroke. A range of different protocols were used: for example, Baek et al12 included patients with a baseline NIHSS score that was ≥4, no bilateral diffuse pontine ischaemia on the diffusion sequence, and start of the procedure within 8 h after symptom onset. The study of Mourand et al21 included patients with no NIHSS and ischaemic damage on the diffusion sequence limit, and the recanalisation could be obtained within 24 h after onset of symptoms. These variables should be considered when interpreting the results of our systematic analysis. It is possible that some relevant studies were not taken into account in our systematic review. However, it is unlikely that this potential publication bias strongly distorted our findings because we found no evidence of such bias by examining the funnel plots.


In patients presenting with an acute symptomatic BAO, stent retriever thrombectomy is a safe treatment modality for patients with stroke. Although the stent retrievers showed a good recanalisation rate, there are currently no randomised clinical trials to assess its clinical efficacy in comparison with the reference treatment.


View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Contributors BG took part in the study concept and design. BG, IE, RS-H, EO, DNG and AT took part in the acquisition of data. BG, DNG, MR, NN and FT took part in the analysis and interpretation of data. BG, NN and took part in a critical revision of the manuscript for important intellectual content. BG and FT took part in the study supervision.

  • Competing interests None declared.

  • Ethics approval Ethics committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

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.

Linked Articles