Bridging versus Direct Mechanical Thrombectomy in Acute Ischemic Stroke: A Subgroup Pooled Meta-Analysis for Time of Intervention, Eligibility, and Study Design

Previous randomized controlled trials (RCTs) showed that the association between intravenous thrombolysis (IVT) and endovascular thrombectomy (EVT) is superior to thrombolysis alone in patients with large vessel occlusion (LVO) [1‒4]. Current international guidelines have been updated accordingly, strongly advising combined treatment under appropriate circumstances [5]. However, it is still unclear if IVT adds some benefit in patients with stroke due to a LVO treated with EVT. Indeed, available RCTs included only few patients with direct EVT (dEVT), limiting the interpretation of the impact of previous IVT on outcomes. If reperfusion is the target, previous IVT would be helpful at the cost of increasing the risk of bleeding [6]. Therefore, the definition of risk/benefit profile of IVT before EVT is crucial to identify the optimal treatment. Previous meta-analyses found no significant differences on functional outcome after mechanical thrombectomy depending on previous IVT [7, 8]. However, there is still paucity of data on outcomes depending on time to intervention, which might indeed be crucial given the nonlinear relationship between delay in intervention, bleeding, and effectiveness [9, 10].

Here we provide a systematic review and pooled data meta-analysis of studies comparing dEVT versus bridging therapy, including subgroup analysis by treatment timing, eligibility to IVT, and study type.

The methods and guidelines of this study-level meta-analysis followed PRISMA guidelines and study protocol registered with PROSPERO (CRD42019135915). Two reviewers systematically searched PubMed, EMBASE, and Cochrane Central register of Controlled Trials for studies comparing dEVT to bridging therapy published between January 1990 and March 1, 2019. Search strategy included the combination of terms, such as “cerebrovascular disorder,” “stroke,” “thrombolysis,” “thrombectomy,” as either keywords or MeSH terms. Specific search algorithm for PubMed is reported in online supplementary material (for all online suppl. material, see www.karger.com/doi/10.1159/000507844). Reference lists and citing articles were also reviewed to increase the identification of relevant studies.

We included RCTs, prospective, and retrospective studies reporting the clinical efficacy and safety of dEVT or combined treatment among adult (≥18) patients with acute ischemic stroke, independently from the device used. We limited the studies to English language and excluded case reports, small case series (<20), conference proceedings, and reviews. The interventional group comprised patients treated with bridging therapy, while the control group was represented by dEVT.

The primary endpoint was functional independence at 90 days from stroke onset, defined as modified Rankin Scale <3. Secondary endpoints were (i) rate of good recanalization according to thrombolysis in cerebral infarction grade (2b or 3), (ii) mortality at 90 days from stroke onset, (iii) and the occurrence of symptomatic intracranial hemorrhage (sICH) as defined by ECASS 2 criteria [10].

Two reviewers (S.V. and M.R.) independently extracted data concerning baseline and outcome characteristics of each included study, as well as its methodological design. We reported the lack of data on outcome, when appropriate. Risk of bias was assessed and reported according to the recommendations of the Cochrane Handbook for Systematic Reviews of Intervention.

We performed a statistical analysis pooling data in the intervention group and the control group. Outcome heterogeneity was evaluated with Cochrane’s Q test and I². An overall p value of <0.05 was considered statistically significant. Fixed-effects models (Mantel-Haenszel method) and random-effects models (DerSimonian and Laird method) were applied depending on quantification of heterogeneity. OR and 95% CI values were calculated for primary and secondary endpoints. Subgroup analysis was predefined for (i) difference of the onset-to-groin time (OGT) between interventional and control groups, (ii) study design, and (iii) eligibility to IVT. For time-dependent analysis, difference in onset-to-treatment <30 min was considered as nonsignificant [9, 11]. We report the analysis results graphically using forest plots for outcomes of single included trials and the total treatment effects. We also calculated the number needed to treat for the primary endpoint. Finally, we introduced covariates to reduce heterogeneity of meta-analysis, and we performed a meta-regression analysis to detect the influence of these variables on the primary endpoint. In particular, we considered the differences of means or medians between interventional and control groups of each included study concerning age, clinical severity summarized by the score of National Institute of Health Stroke Scale, pretreatment size of infarction using the ASPECT score, OGT, and onset-to-recanalization time. We reported graphical representation of final results by bubble plots. Data analysis was performed using Review Manager version 5.3 (The Cochrane Collaboration 2012, Copenhagen, Denmark) and R software.

We included in this analysis 35 studies with a total of 9,117 patients (PRISMA flowchart, Fig. 1) [12‒45]. Most of the included studies were observational and prospective. Only 4 studies were RCTs. No significant differences emerged comparing patients undergoing dEVT and bridging treatment for gender, hypertension, diabetes, National Institute of Health Stroke Scale score at admission. On the contrary, dEVT patients were slightly younger (p = 0.013), had longer OGT, and higher prevalence of atrial fibrillation (p < 0.001; Table 1). The mean of the onset-to-needle time was 129.4 min (SD 74 min).

Considering the primary endpoint, bridging therapy was superior to dEVT (OR 1.44, 95% CI 1.22–1.69, p < 0.001, p_{heterogeneity}<0.001; Fig. 2).Good outcome was reported in 41 and 37% of patients receiving bridging therapy and dEVT, respectively, with number needed to treat being 18 in favor of IVT + EVT (Fig. 2). In the subgroup analyses according to the study type, observational studies favored bridging treatment with 6,391 patients included (OR 1.32, 95% CI 1.19–1.46, p < 0.001), while pooling data from RCTs with a total of 1,003 considered subjects, the difference between treatments was not significant (OR 0.91, 95% CI 0.6–1.39, p = 0.2; Fig. 2).

In subgroup analysis depending on differences in onset-to-groin time, the pooled analysis showed better results for bridging therapy versus dEVT (OR 1.55, 95% CI 1.29–1.87, p < 0.001, p_{heterogeneity}<0.001), which was confirmed among studies with similar OGT (OR 1.66; 95% CI 1.21–2.28, p = 0.008) but with high heterogeneity (I² = 79%; Fig. 3). Among studies with shorter OGT for IVT + EVT (n = 2,948), bridging was superior to dEVT (OR 1.53, 95% CI 1.27–1.85, p < 0.001, p_{heterogeneity} = 0.69). Vice versa, reperfusion strategies had similar outcome (OR 1.29, 95% CI 0.92–1.79) considering studies with shorter OGT for dEVT than bridging therapy (n = 943; Fig. 3).

Meta-regression analysis of the difference between OGT for single study did not show significant association with functional outcome (SE 0.001; p = 0.361), as represented in online supplementary Figure S1. Similarly, differences of age and clinical severity between study groups did not have significant correlation with functional outcome (online suppl. Fig. S2, 3), as well as the difference for ASPECT score and onset-to-recanalization time (online suppl. Fig. S4, 5).

Better primary endpoint for bridging treatment was also observed in the pooled analysis of patients with or without eligibility to IVT (OR 1.43, 95% CI 1.21–1.69, p < 0.001, p_{heterogeneity}<0.001). However, considering only patients eligible to IVT (n = 1,848), no difference in outcome distribution depending on reperfusion strategy was detected (OR 1.06, 95% CI 0.86–1.32, p_{heterogeneity} = 0.06; Fig. 4). Finally, considering only studies with patients eligible to IVT and similar OGT between groups (number of studies: 2) [23, 30], we observed a neutral result (OR 1.00; 95% CI 0.71–1.42).

Successful recanalization rates did not differ between groups when we considered RCTs and observational retrospective studies. We observed better results for bridging when we considered observational prospective studies only (OR 1.47; 95% CI 1.16–1.87, p < 0.01) (online suppl. Fig. S6). Overall, IVT + EVT associated with higher rate of good recanalization compared to direct mechanical thrombectomy (OR 1.33, 95% CI 1.07–1.66; p < 0.01).

Regarding further secondary endpoints, mortality at 90 days (n = 5,939) was higher in dEVT group (OR 1.38; 95% CI 1.09–1.75, p < 0.01, p_{heterogeneity} = 0.002). Such result was driven by observational prospective studies (n = 2,935, OR 1.5, 95% CI 1.25–1.8), while substantial equivalence was noted taking into account RCTs only (n = 630) or retrospective observational studies only (n = 2,374; online suppl. Fig. S7). A similar results regarding mortality was found considering OGT differences across studies (OR 1.49, 95% CI 1.23–1.8, p < 0.001; online suppl. Fig. S8).

Overall, no significant difference was noted in terms of sICH depending on reperfusion strategy (OR 0.85, 95% CI 0.68–1.06). Such results were similar also when pooling data depending on study type (online suppl. Fig. S9) and OGT differences and timing. In particular, considering OGT (n = 7,702), IVT + EVT was associated with marginally lower risk of sICH (OR 0.83, 95% CI 0.68–1.01, p_{heterogeneity} = 0.31), although considering studies with shorter OGT for dEVT, reperfusion strategies had similar sICH risk (online suppl. Fig. S10).

In this study-level meta-analysis comparing dEVT vs bridging therapy, we observed that the latter associates with a 1.5-fold increase in 3-month good functional outcome compared to the former. Study design impacted on results, with observational studies reporting net benefit from bridging therapy over dEVT, while RCTs supported substantial equivalence between the 2 strategies. When we considered studies enrolling only patients eligible to IVT, the results did not significantly differ between groups. On the contrary, the pooled analysis of studies including patients ineligible to IVT showed a better outcome for the bridging therapy compared to dEVT.

Our results refine and substantially update previous reports summing up evidences on the risk/benefit profile of IVT before EVT in acute ischemic stroke. Comparable effectiveness and safety of reperfusion strategies were reported in previous meta-analysis, which however were limited by small sample size and inclusion of studies with first-generation thrombectomy devices only [8], or did not provide time-to-intervention subgroup analysis, limiting the interpretation of results [46]. Benefit from bridging over direct mechanical thrombectomy emerged from a 2017 meta-analysis [6], which however was limited to 13 studies and an overall sample size <3,000 patients, and did not provide time-to-intervention subgroup analysis, therefore preventing from drawing firm conclusions [8, 46]. Improved reperfusion rates with bridging were reported in recent meta-analysis as well, though data on functional outcome and OGT were not provided [47]. In our meta-analysis, the most comprehensive provided to date, including 35 studies and >9,000 patients, we were able to confirm an overall benefit of bridging over dEVT, which was confirmed in studies with shorter OGT for IVT + EVT than dEVT.

In our pooled analysis, the differences of results between observational studies and randomized trials could be related to lower selection criteria in nonrandomized studies. Evidences deriving from real-life setting observational studies further support the convenience of bridging over dEVT, supporting its current use in daily practice, independently from stroke network organization into drip-and-ship or mothership.

Regarding onset-to-groin time subgroup analysis, when we compared studies and trials with similar median OGT, patients undergoing bridging had better functional outcome than those receiving dEVT. Although differences in workflow times exist depending on study type, these findings seem consistently replicated in overall analysis. Therefore, a benefit from previous IVT emerges, with IVT then plausibly contributing to successful reperfusion, early recanalization and better outcome, which would marginally be supported also by a 10% successful recanalization already after IVT [45]. The fact that dEVT seems marginally superior to bridging in studies with shorter OGT for direct mechanical thrombectomy needs to be confirmed by larger RCTs.

In this pooled analysis, we considered also some important covariates for functional outcome. Indeed, the differences for age and clinical severity in interventional and control groups for each study did not significantly correlate with the probability of good outcome. Therefore, this finding suggests that the single-study level comparisons of these variables were not different between the 2 groups, and our results concerning the 2 types of treatments are substantial.

Our study has some limitation. First, we conducted only a study-level analysis rather than a patient-level pooled. Second, data from RCTs derive mainly from post hoc analyses, and, as well as observational studies, might be susceptible to selection bias. Third, imaging endpoints were not consistently adjudicated, even if the minimum core was determined by the detection of a LVO. Fourth, since delayed presentation and coagulopathy are usually identified as causes of IVT ineligibility, patients receiving dEVT would be possibly less prone to recanalization, and therefore the benefit of direct mechanical thrombectomy could have been slightly underestimated. However, since our results are supported through multiple subgroup analysis, and meta-regression analysis was used to adjust for confounders, it seems reasonable to offer bridging treatment as advised by current guidelines. The fact that multiple meta-analysis came out with different results depending mostly on sample size and inclusion criteria for systematic review points to the need of specific RCTs. This issue is also emerged in this study with an observed neutral result in the subgroup analysis of patients eligible to IVT and similar onset-to-groin time. Therefore, the results from 2 ongoing RCTs comparing the added benefit of IVT prior to endovascular treatment (SWIFT DIRECT and the MR CLEAN-NOIV) are critically needed to refine treatment of acute LVO strokes.

Ethical approval was not required because it is a systematic review of previous published studies. The research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki.

The authors have no conflicts of interest to declare.

There are no funding sources to declare.

S.V.: study concept. S.V. and M.R.: acquisition, analysis, interpreting of data, and drafting of the manuscript. D.C. and E.C.A.: critical revision and final approval of the manuscript.