Background and purpose It remains controversial if endovascular treatment (EVT) can improve the outcome of patients with acute basilar artery occlusion (BAO). This study aims to compare the functional outcomes between EVT with and without intravenous thrombolysis (IVT) first in patients who had acute ischaemic stroke (AIS) due to BAO.
Methods Patients who had AIS with BAO who underwent EVT within 24 hours of onset were enrolled in this multicentre cohort study, and the efficacy and safety were compared between IVT+EVT and direct EVT. The primary outcome was 90-day functional independence. All outcomes were assessed with adjusted OR (aOR) from the multivariable logistic regression. In addition, a meta-analysis was performed on all recently published pivotal studies on functional independence after EVT in patients with BAO.
Results Of 310 enrolled patients with BAO, 241 (78%) were treated with direct EVT and 69 (22%) with IVT+EVT. Direct EVT was associated with a worse functional outcome (aOR, 0.46 (95% CI 0.24 to 0.85), p=0.01). IVT+EVT was associated with a lower percentage of patients who needed ≥3 passes of stent retriever (10.14% vs 20.75%). The meta-analysis regression revealed a potential positive correlation between bridging with IVT first and functional independence (r=0.14 (95% CI 0.05 to 0.24), p<0.01).
Conclusions This study showed that compared with direct EVT, EVT with IVT first was associated with better functional outcomes in patients with BAO treated within 24 hours of onset. The meta-analysis demonstrated similar favourable efficacy of IVT first followed by EVT in patients with BAO.
Data availability statement
Data are available upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
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.
Basilar artery occlusion (BAO) accounts for about 2% of all stroke cases and has a high mortality rate and risk of disability, even after successful recanalisation.1 2 Reperfusion therapy might improve the prognosis of patients with BAO but with very limited evidence.3
Pivotal randomised controlled trials (RCTs) confirmed that patients who had acute ischaemic stroke (AIS) with anterior-circulation large vessel occlusion (LVO) could benefit from intravenous thrombolysis (IVT) and endovascular treatment (EVT) up to 24 hours from stroke onset.4 Safety of EVT in patients with BAO has been proven in recent two RCTs. However, the efficacy of EVT was not superior to that of medical treatment in patients with BAO by the current evidence. Only one cohort study found that EVT had a marginal effect on improving functional outcomes.5–7 Most centres apply recanalisation therapies for BAO up to 12–24 hours after symptom onset, which is a substantially longer time window than the 6 hours recommended by current guidelines.8
Recent RCTs indicated that direct endovascular treatment (DEVT) was non-inferior to bridging treatment within 4.5 hours of onset in patients who had AIS with an anterior-circulation LVO.9 10 However, the specific effect of IVT first followed by EVT in patients with posterior-circulation LVO remains unclear. The proportion of prior IVT ranged from 18.4% to 78.4% in the reported BAO trials,5–7 which is much lower than in the anterior-circulation LVO trials (approximately 80%).4 A recently published subgroup analysis of the Basilar Artery International Cooperation Study (BASICS) illustrated that the efficacy of EVT was marginally better with bridging therapy in patients with BAO (OR, 1.16–2.70 (95% CI 0.88 to 8.11) vs 1.08 (95% CI 0.53 to 2.23)), but without statistical significance.5
The initial onset of BAO is usually insidious and often progresses rapidly. Owing to the different degrees of involvement of the brainstem, patients with BAO can present with symptoms that vary from isolated cranial nerve palsies or hemiplegia to a locked-in state or coma. Even small infarcts would often result in poor outcomes. Moreover, the collateral compensatory ability of the perforating branches of the basilar artery is worse than that of the anterior-circulation large vessels. Therefore, patients with BAO have more irreversible and severe infarction, and even with successful recanalisation, reperfusion therapy in BAO often fails to improve disability. Furthermore, in some patients with BAO with intracranial stenoses and tortuous vessels in the vertebrobasilar artery system, EVT becomes more difficult. Hence, clinical assessment and therapeutic options for BAO are different from that of anterior-circulation occlusions.
This study investigates the functional outcomes of IVT+EVT and DEVT in patients who had AIS due to BAO in the real world.
This prospective cohort study was conducted in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.11
Study design and setting
Details of the Registration Study for Critical Care of Acute Ischemic Stroke after Recanalization (RESCUE-RE) have been reported previously.12 Briefly, RESCUE-RE is an ongoing, investigator-initiated, nationwide, prospective, observational cohort study recruiting patients who had acute stroke with an LVO who received EVT at 18 comprehensive stroke centres across China. Details on demographic features, clinical variables and laboratory data were collected by trained research coordinators and investigators. All imaging files were stored in a centralised manner and reviewed independently by two radiologists (WGu and XHo) blinded to the clinical data. All records at each participating centre were examined and audited by full-time quality control coordinators.
Participants and treatments
From July 2018 to October 2020, patients who had AIS treated with EVT within 24 hours of stroke onset in RESCUE-RE and who met the following criteria were recruited to the current study: (1) age >18 years old; (2) AIS diagnosed based on brain imaging with documented LVO in the basilar artery, which was confirmed by either a CT angiography or magnetic resonance angiography or digital subtraction angiography (DSA), and patients with occluded distal intracranial vertebral artery (V4 segment) resulting in no flow to the basilar artery (eg, functional BAO); (3) IVT was initiated within 4.5 hours according to the standard criteria in qualified patients; (4) a prestroke modified Rankin Scale (mRS) score ≤2; and (5) followed up for 3 months.
All patients were admitted to the intensive care unit at the participating centre, and periprocedural management was conducted according to the local clinical care protocols and latest national guideline.8 IVT was given with a standard dosage of alteplase (0.9 mg/kg over 1 hour with a 10% initial bolus, maximum 90 mg). Alteplase infusion should be completed during EVT, even if the procedure was successful in establishing revascularisation. Mechanical thrombectomy, aspiration, balloon angioplasty, stent implantation and combination therapies (two or more of the above procedures) were all considered as a form of EVT.
Variables and outcome measurement
The study variables were collected prospectively, including baseline characteristics, details of workflow, EVT procedure, perioperative treatment, and primary and secondary outcomes. National Institutes of Health Stroke Scale (NIHSS) score on admission was used to assess stroke severity. BAO was divided into distal, middle and proximal BAO based on the occlusion site. The posterior-circulation collateral status was assessed using both the Basilar Artery on Computed Tomography Angiography (BATMAN) score and the Posterior Circulation Collateral Score (PC-CS).13 Successful pre-EVT or post-EVT reperfusion was defined as a score of ≥2b on the modified Thrombolysis in Cerebral Ischemia (mTICI) scale on DSA. Patients were followed up for 3 months through face-to-face or over-the-phone communications by staff who were blinded to patient information and treatment.
The primary outcome was functional independence, defined as a dichotomised mRS score of 0–2 at 90 days. The secondary outcomes included mortality, successful reperfusion, culprit vessel occlusion at 24 hours after EVT, any intracranial haemorrhage (ICH) and symptomatic intracranial haemorrhage (sICH). sICH was defined as any ICH on post-EVT cerebral imaging with an increase of ≥4 points on the NIHSS within 7 days, according to the Second European-Australasian Acute Stroke Study criteria.14
Categorical variables were presented as numbers (proportions) and continuous variables as median (IQR) or mean±SD. χ2 test or Fisher’s exact test was used for univariate analysis of categorical variables. Continuous variables were analysed with Student’s t-test or Mann-Whitney U test. Multivariate logistic regression was performed to obtain the OR and 95% CI of the associations between the two groups and their outcomes. The mRS scores of patients treated with IVT+EVT were compared with those of patients treated with DEVT by conducting ordinal logistic regression to obtain the common OR. Potential confounders were adjusted based on the results of univariable analyses to p<0.05 while taking into consideration the important risk factors in combination with clinical judgement and expert knowledge: age, baseline NIHSS score, onset to puncture time, prestroke mRS score, history of stroke and transient ischaemic attack. The functional outcomes in subgroups were divided based on age, baseline NIHSS score, key intervals and other key variables and were compared between two groups. Univariate logistic regression was applied to calculate the OR in all subgroups.
Only randomised trials or high-quality cohort studies in recent years that reported original data on functional independence (mRS score of 0–2 at 90 days) after EVT in patients with BAO plus the current cohort study were included in this meta-analysis. The occurrence of the following events in the EVT group was extracted and analysed: total patients, treated with IVT, an mRS score of 0–2 and deceased patients at 90 days (the IVT+EVT group and the DEVT group of RESCUE-RE were analysed separately). To combine the data from all studies, the restricted maximum likelihood random-effects method was used to estimate a summary across all included studies. The percentage of mRS score of 0–2 estimates and the associated 95% CI for each of the endpoints were calculated. Between-study heterogeneities were evaluated by the I2 statistics and the Cochran Q (χ2) statistics, with a p value of 0.10 set to be significant for heterogeneity. The results of the meta-analysis regression would demonstrate the influence of IVT on functional independence in each study as well as the overall meta-analysis summary estimate.
All analyses were performed with SAS V.9.4 and Stata V.16.0 software. A two-sided p<0.05 was considered statistically significant.
From July 2018 to October 2020, 1701 patients were consecutively enrolled in the RESCUE-RE cohort. A total of 321 patients met the inclusion criteria and were included in the current study (figure 1). After excluding 11 patients who were lost to follow-up, 310 patients were eligible for the final analysis, with 241 (78%) treated with DEVT and 69 (22%) with IVT+EVT. Patients were allocated to the DEVT group (241 patients) if they did not receive IVT first. The reasons for withholding IVT included contraindications to IVT, beyond the treatment time window, over 4.5 hours of in-hospital delay, refusal by patients or their families, and other reasons.
Table 1 shows the baseline characteristics of the participants in the two groups. The mean age was 61.39±10.92 years and 77% of the participants were male. The median NIHSS score was 21 (IQR 11–27). The median BATMAN score was 6 (IQR 4–7) and the PC-CS was 5 (IQR 2–6). V4 occlusion was found in 36.13% of the patients and BAO in 63.87%, including 31.94% in the middle segment. As for the aetiology, 80.97% were from atherosclerosis, 14.19% were cardioembolic and 15% were of other types. The baseline characteristics of the participants in the two treatment groups were similar.
Workflow intervals and EVT procedures
The median onset to needle time in the IVT+EVT group was 190 min (160–233). The median onset to arrival time was 293 min (135–520). The onset to puncture time was 485 min (333–835); the puncture to recanalisation time was 122 min (78–175); and the onset to recanalisation time was 630 min (457–1012). There was no difference in the time interval parameters between the two groups. Thrombectomy was a little more common in the DEVT group, with statistical significance (86.72% vs 76.81%, p=0.04). In addition, in the DEVT group, more patients received combination therapy (53.53% vs 39.13%, p<0.01), and the proportion of patients with ≥3 retriever passes (20.75% vs 10.14%, p=0.05) was marginally higher. The median residual stenosis after EVT was 20%. No difference was found in workflow and procedures between the two groups (table 2).
Primary and secondary outcomes
Analysis of the primary outcome revealed fewer patients with an mRS score of 0–2 at 90 days in the DEVT group after adjusting for potential confounders (aOR, 0.46 (95% CI 0.24 to 0.85), p=0.01). The distributions (shift) of mRS score at 90 days (figure 2) differed between the two groups after the adjustment (adjusted common odds ratio (acOR), 0.56 (95% CI 0.34 to 0.90), p=0.02).
The primary and secondary outcomes are presented in table 3. Successful reperfusion after EVT, the culprit vessel occlusion at 24 hours, and the frequency of any ICH and sICH were not statistically different between the DEVT group and the IVT+EVT group. Mortality at 90 days was marginally higher in patients treated with DEVT (26.14% vs 18.84%), without statistical significance (aOR, 1.79 (95% CI 0.87 to 3.71), p=0.11).
The subgroup analyses (figure 3) did not show any characteristics significantly associated with treatment effect, except for patients with baseline NIHSS score of 10–21. In this subgroup with an NIHSS score between 10 and 21 (96 patients, 32.11% of the total subjects), the 90-day functional outcome was better in the IVT+EVT group compared with the DEVT group (OR, 0.15 (95% CI 0.05 to 0.48), p interaction=0.04). Patients of younger age (≤75 years), with atherosclerotic stroke, with shorter time interval from onset to door (≤3 hours), to groin puncture (≤6 hours) and to recanalisation (≤12 hours), and with baseline NIHSS score ≤21, BATMAN score >7, PC-CS >5 or direct to EVT centre had better outcomes in the IVT+EVT group, although without statistical significance.
A total of two randomised trials and two cohort trials,5–7 including RESCUE-RE, with 1176 patients were eligible for the meta-analysis (online supplemental figure S1). In the other three trials, EVT was not considered an inclusion criterion, but as an intervention, and for these three trials the original data needed for the current analysis were extracted from the EVT group.5–7 The average duration of follow-up was 90 days. The mean duration from onset to treatment was 246–485 min and the NIHSS score at baseline was between 21 and 32. The proportion of IVT first ranged from 18.4% to 78.4%. More details on the design, characteristics and quality assessment of the included trials are provided in online supplemental tables S1 and S2. After EVT, 31.2% (95% CI 26.0 to 36.4) of the patients had functional independence (mRS score 0–2 at 90 days), while the mortality rate was between 18.9% and 46.2%. Successful reperfusion (mTICI ≥2 b) after EVT was achieved in 68.1%–80.7% of the patients. About 4.5%–8.7% of the patients had sICH (table 4, figure 4A). A stratified analysis by onset to EVT time (<24 hours and <8 hours) is provided in online supplemental figure S2. Of the patients, 30.40% (95% CI 22.13 to 38.68) had functional independence in the subgroup of onset to EVT time <24 hours, while the functional independence rate was 34.46% (95% CI 28.17 to 40.76) in the subgroup of onset to EVT time <8 hours. There was substantial heterogeneity in the overall analysis (I2=82.10%), but there was no statistical difference between the two subgroups (p=0.44). The meta-analysis regression model (figure 4B) revealed significant positive correlation between the proportion of patients who had IVT first and functional independence after EVT (r=0.14 (95% CI 0.05 to 0.24), p<0.01). There was also no significant publication bias detected with regard to functional independence on examination of funnel plots (online supplemental figure S3).
Previous studies had inconclusive and conflicting conclusions regarding the comparison between EVT and standard medical treatment in patients who had AIS due to BAO.5–7 A recent RCT revealed that EVT and medical therapy were not significantly different in achieving favourable functional outcome in patients with BAO.5 However, a cohort registry trial, the EVT for Acute Basilar Artery Occlusion Study, indicated that EVT provided considerable benefit compared with standard medical care alone in patients with BAO (aOR, 3.08 (95% CI 2.09 to 4.55), p<0.01).7 DEVT was known to be non-inferior to bridging treatment within 4.5 hours of onset in patients who had AIS with anterior-circulation LVO.9 10 However, data comparing IVT+EVT and DEVT in patients with BAO are scarce. As the first multicentre cohort study comparing the effectiveness and safety of IVT+EVT versus DEVT in patients with BAO treated within 24 hours of onset, our study found that the combination of EVT with IVT first (bridging) was associated with a better clinical outcome. Moreover, the regression meta-analysis identified a potential positive correlation between the proportion of patients with bridging treatment and improved functional independence in patients with BAO. Our study therefore provided more data and evidence to support the use of bridging treatment.
There is no consensus about the time window to begin EVT in BAO. Although current guidelines suggest that EVT within 6 hours could be reasonable,8 the treatment window of 6–24 hours after the onset for patients with BAO might be beneficial in recent trials.5–7 Based on these real-world data, the time from IVT to groin puncture in our bridging group was significantly longer than those with an anterior-circulation LVO. However, we found that patients in the bridging group had significantly better outcomes than DEVT if IVT was given <4.5 hours, even though EVT was initiated up to 24 hours after onset. The benefit of IVT first appeared to persist, without increasing the risk of haemorrhage, even if EVT was delivered at a later time, up to 24 hours after onset. This may have provided some evidence that in patients with acute BAO and those who need to be transferred to a comprehensive stroke centre, bridging with IVT first should be considered.15 Nevertheless, the earlier the EVT is initiated, the more effective the therapy will be and with better safety profile.16 17 In our subgroup analysis, we also found that patients with shorter onset to door time (≤3 hours), fast groin puncture time (≤6 hours) and faster recanalisation time (≤12 hours) had a better outcome after bridging treatment.
Potential advantages of bridging therapy include early thrombus fragmentation, microvascular reperfusion and enhanced recanalisation.18–20 Subgroup analysis of the ESCAPE trial (Small Core and Anterior Circulation Proximal Occlusion With Emphasis on Minimizing CT to Recanalization Times Trial) indicated that pretreatment with IVT reduced approximately by two-thirds the likelihood of infarct in a new previously unaffected territory complicating DEVT, whereas infarct in a new previously unaffected territory was in turn associated with substantial reduction in the odds of functional improvement quantified using the shift in mRS score.21 Posterior-circulation occlusions are significantly associated with distal emboli in the same and previously unaffected territories during thrombectomy. This may be partially ameliorated with bridging tissue plasminogen activator and result in better outcomes.22 23 The bridging approach has possible benefits in certain individual cases (eg, patients within the very early time window and with short dense clot, distal thrombus location, high residual flow permeability, good collaterals, or drip and ship thrombolysis), but they seem to be low as revealed by a previous study.24 However, based on the results of the meta-analysis, the overall reperfusion status after EVT in BAO was lower than in anterior-circulation LVO. Early reocclusion and EVT failure were more common in BAO, causing a lower reperfusion rate.1 Besides, 80.97% of the stroke cases in BAO were from atherosclerosis, a higher proportion compared with that in the anterior-circulation LVO. Therefore, theoretically, BAO might benefit more from IVT first.
Several studies reported that IVT was associated with increased risk of haemorrhagic transformation, delayed initiation of the EVT procedure, adverse thrombus migration and fragmentation, resulting in thrombi that were out of reach for mechanical thrombectomy.25–27 IVT prior to EVT seemed useless for patients with long clots, low clot burden scores, proximal occlusions, long onset to treatment time or with calcific emboli.28 However, in the current study, the 90-day mortality and sICH were not statistically different between the two groups, similar as in treating anterior-circulation LVO. Furthermore, no difference was found in problems with workflow between the two groups. Patients with BAO are often found with a higher percentage of intracranial stenoses and tortuous vessels in the vertebrobasilar artery system,29 which would increase the difficulty of EVT. This potential difficulty in EVT operation was confirmed by the current study, as the DEVT group had more patients who needed ≥3 passes with stent retriever.30
While the results of the current study may not be sufficient to affirm or negate the value of IVT prior to EVT for anterior-circulation LVO, it is important to acknowledge that there are many individual variables in the decision-making process that may not be captured in clinical trials.28 Recently, several studies have reported prognostic factors for clinical outcome in patients with acute BAO after mechanical thrombectomy.13 31 32 Although there are some differences in the cut-off value, BASICS reported an initial modest NIHSS score before EVT as a possible predictor of good outcome.5 Our study indicated that IVT+EVT might provide a more favourable outcome for patients with BAO with an NIHSS score between 10 and 21. In addition, in younger patients (≤75 years), good collateral status (BATMAN >7 or PC-CS >5) and direct to EVT centre, IVT+EVT could provide a better outcome compared with DEVT. However, further validation through large and randomised trials is needed since the data were from a subgroup analysis.
On the site of basilar occlusion, previous studies13 33 identified that distal BAO was a possible favourable predictor of outcome due to probably better collaterals. Distal BAO may present with fewer initial neurological symptoms, a shorter clot burden and a higher probability of successful recanalisation. The subgroup analysis in the current study showed that bridging treatment might not have any advantage in patients with distal BAO compared with DEVT.
The current study has several limitations. First, this study was an observational study. How treating clinicians select a specific treatment is a complex issue in the real world. Multivariable analyses cannot adjust completely for systematic differences between treatment groups without a randomised trial. Second, there was potential selection bias as the treatment group was stratified based on the selective clinical criteria (IVT contraindications, for instance). About 30% of patients in the DEVT group had contraindications to IVT, which may carry a poor prognosis.34 Third, more patients received DEVT, a potential of lack of equipoise among participating centres.7 Fourth, the time delay between hospital arrival and arterial puncture for EVT was longer than that in RCTs and guidelines, which might influence the results. In addition, the percentage of atherosclerosis was higher than reported in other BAO studies, which may reflect the Asian Intracranial atherosclerotic disease (ICAD) preponderance, but reduces the generalisability of this study. Lastly, the study enrolled patients with BAO who were chosen to receive EVT based on limited level of evidence and thus the findings should be treated with caution and be further confirmed by well-designed RCTs. There were also several limitations to the meta-analysis, such as substantial heterogeneity in the overall analysis, selection bias and other possible bias and confounding bias, which might impact the results. An ongoing RCT (NCT03494920) enrolling both patients with anterior-circulation and posterior-circulation LVO should be able to provide conclusive results.
In conclusion, a significant difference in terms of functional outcome was observed in patients with BAO-AIS treated within 24 hours of onset with either bridging therapy or DEVT. The data showed bridging therapy might bring better functional outcomes at 90 days than those treated with EVT alone after adjusting for potential confounders, notably in patients with an initial modest NIHSS score (10–21) at onset. IVT first may provide additional benefits. Furthermore, the regression meta-analysis revealed a potential positive correlation between IVT first and functional independence. Overall, the results of the current study indicated that bridging therapy may be potentially favourable for BAO and in improving functional outcomes. More RCTs are needed to confirm these findings.
Data availability statement
Data are available upon reasonable request.
Patient consent for publication
The study protocol was approved by the ethics committee of the Beijing Tiantan Hospital and all participating centres.
The authors thank Hang Yu, Yanyan Ma, Xiaoyu Cui, Lei Yu and other study quality control coordinators for their meticulous work on data quality control and are grateful to all subjects for their participation in the RESCUE-RE trial.
Contributors All authors contributed to this paper. Study concept and design: LLi and ZMi. Drafting of the manuscript: XNi and XLe. Statistical analysis: YPa and HYa. Study supervision and organisation of the project: YWe, QLu, XLi, LZe, JLi, XYa, YDi, DLi, WDu, ZZa, WGu and XHo. Revision of the manuscript: DWa, YPa, ZYa and MWe. Responsible for the overall content as the guarantor: LLi.
Funding This work was supported by the National Key R&D Program of China (2016YFC1307301) and the National Natural Science Foundation of China (81820108012). This study was partially funded by Boehringer Ingelheim China.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.