Background The major concern for bridging intravenous thrombolysis (IVT) before endovascular thrombectomy (EVT) is the potentially increased risk of symptomatic intracerebral hemorrhage (sICH). Thus we conducted this study to clarify whether evaluation of individual bleeding risk could assist in the decision to perform IVT before EVT.
Methods The study was a subgroup analysis of a randomized trial evaluating the safety and efficacy of IVT before EVT. The SEDAN (blood Sugar, Early infarct signs and (hyper) Dense cerebral artery sign, Age, and National Institutes of Health Stroke Score) score, GRASPS (Glucose, Race, Age, Sex, systolic blood Pressure, and Severity of stroke) score, and SITS-SICH (Safe Implementation of Thrombolysis in Stroke-Symptomatic Intracerebral Hemorrhage) score were used to evaluate individual bleeding risk. The primary outcome was functional independence, defined as a modified Rankin Scale (mRS) score of 0–2 at 90 days. Binary logistic regression with an interaction term was used to estimate treatment effect modification to clarify whether direct EVT was more beneficial in patients with a higher sICH risk, while adjunctive IVT before EVT was more beneficial in patients with a lower sICH risk.
Results Among 658 randomized patients, 639 (361 men, 56.5%; median age 69 (IQR 61–76) years) were included in the study. With the SITS-SICH score as an example, adjusted OR for functional independence with EVT alone was 1.12 (95% CI 0.68 to 1.82) in patients with a lower sICH risk (SITS-SICH score 0–4) and 0.92 (0.53 to 1.60) in those with a higher sICH risk (SITS-SICH score 5–15). There were no treatment-by-bleeding-risk interactions for all dichotomized mRS outcomes based on the three scores (all p>0.05).
Conclusions We found no evidence that clinicians can decide whether to omit IVT before EVT based on an individualized assessment of bleeding risk.
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.
What is already known on this topic
Several randomized trials have recently investigated the adjunctive benefit of intravenous thrombolysis (IVT) before endovascular thrombectomy (EVT) by using non-inferiority designs, with mixed results in supporting or refuting the benefits.
What this study adds
Patients with a higher bleeding risk did not benefit more from EVT alone than bridging therapy, while those with a lower bleeding risk did not benefit more from bridging therapy than EVT alone.
How this study might affect research, practice, or policy
We found no evidence that clinicians can decide whether to omit IVT before EVT based on an individualized assessment of bleeding risk.
Endovascular thrombectomy (EVT) has become the standard of care for patients with acute ischemic stroke due to large vessel occlusion in the anterior circulation.1 Current guidelines also recommend intravenous thrombolysis (IVT) before EVT in eligible patients.2 3 The potential advantage of IVT prior to EVT is that alteplase can partially dissolve the thrombi, achieving earlier and more complete recanalization, especially in cases of delay in EVT.4 5 In contrast, administration of alteplase might delay initiation of the EVT procedure, and increase the risk of intracranial hemorrhage (ICH) and medical expenses.6
Several randomized trials,7–12 including DIRECT-MT (Direct Intra-arterial Thrombectomy in Order to Revascularize AIS Patients With Large Vessel Occlusion Efficiently in Chinese Tertiary Hospitals),7 have recently investigated the adjunctive benefit of IVT before EVT by using non-inferiority designs, with mixed results in supporting or refuting the benefits. Since outcomes were similar for both treatment arms across the published trials,13 the heterogeneity of the treatment effect of IVT on subgroups based on patient characteristics is worth exploring further. The major concern for bridging IVT is the increased risk in symptomatic intracerebral hemorrhage (sICH). Hence we hypothesized that evaluation of individual bleeding risk could assist in the decision to perform IVT before EVT.
Because both treatment arms received EVT, we used a prediction scale of bleeding risk for IVT alone, to separately evaluate the bleeding risk from IVT. There are several scoring systems available to predict the risk of sICH after IVT.14–19 Three scores have better reliability and validity due to the large sample size of both the derivation and validation cohorts.16–18 Based on these scores, we aimed to investigate whether direct EVT was more beneficial in patients with a higher risk of sICH, and whether adjunctive IVT before EVT was more beneficial in patients with a lower risk of sICH, in this subgroup analysis of DIRECT-MT.
Study design and patients
DIRECT-MT was a randomized, controlled, open label trial, assessing the non-inferiority of EVT alone versus bridging therapy (IVT before EVT) in patients from 41 centers in China. Patients were aged ≥18 years with a proximal arterial occlusion in the anterior circulation, and treated within 4.5 hours of symptom onset. Detailed study methods and patient eligibility criteria have been reported previously.7 20 In this subgroup analysis, those who did not undergo catheter angiography were exclude. Written informed consent was obtained from all patients or their legal representatives before randomization. The study protocol was approved by a central medical ethics committee and the research board of each participating center.
Demographic, laboratory, radiological, and clinical characteristics were recorded at the time of enrollment. Imaging was evaluated by an independent imaging core laboratory that was blinded to the treatment assignments, with a consensus reached in the event of discrepancies. Collateral flow was graded with baseline CT angiography, using a 4 point scale, with 0 representing absent collateral flow (absent filling of the occluded territory), 1 representing poor collateral flow (<50% filling of the occluded territory), 2 representing intermediate collateral flow (between 50% and 100% filling of the occluded territory), and 3 representing excellent collateral flow (100% filling of the occluded territory). Outcome data at 3 months were obtained from structured interviews using standardized forms, and were then verified with the score on the modified Rankin Scale (mRS) by an outcome committee.
Three scores were used to evaluate individual bleeding risk. The SEDAN (blood Sugar, Early infarct signs and (hyper)Dense cerebral artery sign, Age, and National Institutes of Health Stroke Score (NIHSS)) score (range 0–6) included baseline glucose (1 point for 145–216 mg/dL and 2 points for >216 mg/dL), early infarct signs (1 point), hyperdense cerebral artery sign (1 point), age (1 point for >75 years), and baseline NIHSS score (1 point for ≥10).16 The GRASPS (Glucose, Race, Age, Sex, systolic blood Pressure, and Severity of stroke) score (range 45–105) included baseline glucose (2 points for <100 mg/dL, 6 points for 100–149 mg/dL, and 8 points for ≥150 mg/dL), ethnicity (9 points for Asian), age (8 points for ≤60 years, 11 points for 61–70 years, 15 points for 71–80 years, and 17 points for >80 years), gender (4 points for men), baseline systolic blood pressure (10 points for <120 mm Hg, 14 points for 120–149 mm Hg, 18 points for 150–179 mm Hg, and 21 points for ≥180 mmHg), and baseline NIHSS score (25 points for 0–5, 27 points for 6–10, 34 points for 11–15, 40 points for 16–20, and 42 points for >20).17 The SITS-SICH (Safe Implementation of Thrombolysis in Stroke-Symptomatic Intracerebral Hemorrhage) score (range 0–15) included baseline NIHSS score (1 point for 7–12 and 2 points for>12), baseline glucose (2 points for >180 mg/dL), baseline systolic blood pressure (1 point for ≥146 mm Hg), age (1 point for 72 years), body weight (1 point for ≥95 kg), onset to treatment time (1 point for ≥180 min), prior antiplatelet therapy (2 points for aspirin monotherapy and 3 points for combined aspirin and clopidogrel), and history of hypertension (1 point).18 The cut-off value for each score was determined by the relatively balanced sample size in both groups, while considering approximately 10% of predicted sICH (European Cooperative Acute Stroke Study (ECASS II) criteria) risk in the high bleeding risk group based on data from previous studies.16–18 21
The primary outcome of this subgroup analysis was functional independence defined as an mRS score of 0–2 at 90 days (within a window of ±14 days) after randomization. Secondary outcomes were the following: favorable functional outcome (defined as an mRS score of 0–1 at 90 days after randomization), successful reperfusion (defined as an extended Thrombolysis in Cerebral Infarction score of ≥2 b) before thrombectomy (assessed on initial DSA),22 and successful reperfusion on final DSA. Safety outcomes were sICH and asymptomatic ICH according to the Heidelberg criteria,23 assessed on follow-up non-contrast CT at 24–72 hours, and mortality within 90 days.
The analyses were primarily based on the as-treated population. Study participants were dichotomized according to the prediction scores of sICH risk. Data are presented as mean (SD), median (IQR), or number (%). The χ2 test or Fisher’s exact test was used to compare the dichotomous variables between groups, while the independent sample t test or Mann–Whitney U test was used for continuous variables, as appropriate. We analyzed binary outcomes with logistic regression and reported them as both unadjusted and adjusted ORs with 95% CIs, for age, baseline NIHSS score, baseline mRS score, time from stroke onset to randomization, and cerebral collateral status per the DIRECT-MT statistical analysis plan. Statistical significance was set at a p value of <0.05. All statistical analyses were performed with SAS software, version 9.2 (SAS Institute).
The trial profile of the subgroup analyses is shown in figure 1. A total of 656 patients were randomized in the DIRECT-MT trial between February 23, 2018 and July 2, 2019. Seventeen patients did not undergo catheter angiography and thus were excluded from the subgroup analyses. Finally, 639 patients (361 men (56.5%); median age 69 (IQR 61–76) years), 316 (49.5%) in the EVT alone group and 323 (50.5%) in the bridging therapy group, were included. Median baseline NIHSS score was 17 (IQR 13–22) and mean time interval from stroke onset to puncture was 204.9±58.8 min.
Comparison of actual sICH rate and predicted sICH risk
Table 1 shows the comparison of sICH rate in patients with a high or low predicted sICH risk (for IVT) based on the SEDAN, GRASPS, and SITS-SICH scores. In the whole cohort, more sICH occurred in patients who had a higher score, with significant or marginally significant difference. Regardless of treatment approach and rating scale, the sICH rate was numerically higher in patients who had a higher score, although this was not statistically significant.
Comparison of characteristics stratified by sICH risk prediction score
Baseline characteristics of the enrolled cohort, stratified by SITS-SICH score, are shown in table 2. A total of 626 patients had available data for the SITS-SICH score, among whom 323 (51.6%) were rated as having a SITS-SICH score of 0–4 (165 (51.1%) received EVT alone; 158 (48.9%) received bridging therapy) and 303 (48.4%) a score of 5–15 (145 (47.9%) received EVT alone; 158 (52.1%) received bridging therapy). Baseline characteristics were similar in the treatment groups stratified by SITS-SICH risk score, except that the time interval from onset to puncture was shorter and the presence of early infarct sign was more often in patients who received EVT alone and had a higher SITS-SICH score, with marginally significant differences (table 2). Similar findings for SEDAN and GRASPS scores are shown in online supplemental tables 1 and 2.
The mRS score at 90 days was missing for two patients (one received EVT alone and one received bridging therapy), and data were not imputed. Patients who were rated as having an SITS-SICH score of 5–15 achieved a lower rate of functional independence than those rated as having a score of 0–4 (80 (26.5%) vs 151 (46.7%), p<0.001). The adjusted OR for functional independence with EVT alone was 1.12 (95% CI 0.68 to 1.82) in patients with a lower SITS-SICH score (0–4), and 0.92 (95% CI 0.53 to 1.60) in those with a higher SITS-SICH score (5–15) (table 3). The adjusted OR for functional independence with EVT alone was 1.00 (95% CI 0.58 to 1.71) in patients with a lower SEDAN score (0–2), and 1.14 (95% CI 0.69 to 1.88) in those with a higher SEDAN score (3–6) (online supplemental table 3). The adjusted OR for functional independence with EVT alone was 1.20 (95% CI 0.69 to 2.06) in patients with a lower GRASPS score (45–80), and 0.88 (95% CI 0.54 to 1.43) in those with a higher GRASPS score (81–105) (online supplemental table 4). There were no treatment-by-bleeding-risk interactions for all dichotomized mRS outcomes based on the three scores, in both adjusted and unadjusted analysis (table 4).
Secondary and safety outcomes
The adjusted ORs and 95% CIs for the secondary and safety outcomes stratified by SITS-SICH score are shown in table 3. For the majority of secondary and safety outcomes, such as favorable functional outcome, successful reperfusion after EVT, death, and sICH, no significant differences were found between the EVT alone group and the bridging therapy group, stratified by SITS-SICH risk score. Successful reperfusion before EVT (3 (2.1%) vs 13 (8.3%); OR 0.27, 95% CI 0.07 to 0.99) and asymptomatic ICH (44 (30.3%) vs 73 (46.2%); OR 0.53, 95% CI 0.33 to 0.87) occurred less frequently in patients who received EVT alone and had a higher SITS-SICH score (table 3). The adjusted ORs and 95% CIs for the secondary and safety outcomes stratified by SEDAN and GRASPS scores are shown in online supplemental tables 3 and 4.
In this subgroup analysis, we found no evidence of treatment effect modification by predicted sICH risk for functional outcome, based on the SEDAN, GRASPS, and SITS-SICH scores. Specifically, patients with a higher bleeding risk did not benefit more from EVT alone than bridging therapy, while those with a lower bleeding risk did not benefit more from bridging therapy than EVT alone.
The results of six recent randomized trials that compared the effect of EVT alone versus bridging therapy were inconsistent.7–12 The DIRECT-MT and DEVT (Direct Endovascular Treatment vs Standard Bridging Therapy for Patients With Acute Stroke With Large Vessel Occlusion in the Anterior Circulation) trials,7 8 both from China, showed that EVT alone was non-inferior to alteplase followed by EVT, with liberal non-inferiority margins. The SKIP (Direct Mechanical Thrombectomy in Acute LVO Stroke) trial,9 conducted in Japan, was underpowered and did not show the non-inferiority of EVT alone. The MR CLEAN-NO IV (Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic Stroke in the Netherlands) trial failed to show superiority or non-inferiority.10 SWIFT-DIRECT (Bridging Thrombolysis vs Direct Mechanical Thrombectomy in Acute Ischemic Stroke) and DIRECT-SAFE (DIRECT Endovascular Clot Retrieval vs Standard Bridging Thrombolysis With Endovascular Clot Retrieval) did not show the non-inferiority of omitting IVT, using generous non-inferiority margins (12% and 10%).11 12 In an expedited guideline from the European Stroke Organisation, a study level meta-analysis of all six trials showed that a prespecified 1.3% benefit on functional independence of IVT could not be ruled out.24
To separately evaluate the potentially increased ICH risk of alteplase administration before EVT, we selected three scales for assessing the risk of sICH after IVT: SEDAN, GRASPS, and SITS-SICH scores.16–18 Although patients with a higher score were more likely to have sICH after EVT, the prediction was not very accurate in our study, considering these scales were developed for patients with IVT alone. In this subgroup analysis, we could not confirm the hypothesis that an individual's bleeding risk modifies the effect of alteplase when applied as bridging therapy before EVT. Patients treated with EVT alone or bridging therapy had similar functional outcomes in all subgroups based on the sICH risk prediction scores. There are several possible explanations: (1) the non-significantly higher hemorrhage (asymptomatic ICH or sICH) rate might offset the benefits of partial thrombi dissolution; (2) the development of sICH might mainly relate to reperfusion of brain tissue (reperfusion injury), but not the specific treatment approach (limited influence); and (3) alteplase was usually still running during the thrombectomy procedure due to the short time interval between the beginning of IVT to groin puncture in the DIRECT-MT trial, which might reduce the benefits of alteplase.
In most clinical trials, the risk of sICH did not increase significantly when alteplase was applied as adjunctive therapy before EVT. For example, there was no significant difference in the occurrence of sICH (Heidelberg definition) between the EVT alone group and the bridging therapy group in the DIRECT-MT trial (4.3% vs 6.1%),7 DEVT trial (6.1% vs 6.8%),8 and MR CLEAN-NO IV trial (5.9% vs 5.3%).10 The current study also revealed that omitting IVT before EVT would not lead to better functional outcomes even in patients with a high bleeding risk, indicating that clinicians should not be overly concerned about the hemorrhagic events of bridging therapy, especially in cases of delay in EVT. On the other hand, successful reperfusion before EVT occurred more frequently in patients who received bridging therapy. Therefore, the cost efficiency of bridging therapy is evaluated, both the increased cost of alteplase use and the reduced cost due to recanalization prior to the EVT attempt should be considered.
This subgroup analysis had several limitations. First, the statistical tests for interactions between individual bleeding risk and treatment allocation may be underpowered. If the sample size allows, we could further focus on patients with an extremely high bleeding risk. Second, as mentioned above, time from IVT infusion to EVT procedure may not have been adequate to state conclusively the lack of benefit of alteplase. Third, the three scales for predicting sICH after IVT are not specific to ischemic stroke patients with large vessel occlusion.
In conclusion, there is currently no evidence that clinicians can decide whether to omit IVT before EVT based on an individualized assessment of bleeding risk. Pooled analysis of randomized trials may provide greater clarity.
Data availability statement
Data are available upon reasonable request.
Patient consent for publication
This study involves human participants and was approved by Shanghai Changhai Hospital Ethics Committee (CHEC2018-003). Participants gave informed consent to participate in the study before taking part.
SY and MZ contributed equally.
Contributors SY and MZ: conceptualization, data curation, methodology, software, and writing-original draft. HZ: Data curation, formal analysis, investigation, and writing-review and editing. YZ: data curation, formal analysis, supervision, and writing-review and editing. YC, XZ, ZC, PY, YZ, LZ, ZL, and PX: data curation, validation, and writing-review and editing. JS: conceptualization, project administration, resources, supervision, and writing-review and editing. ML: conceptualization, funding acquisition, project administration, resources, supervision, writing-review and editing. JL: conceptualization, project administration, and writing-review and editing. ML (guarantor) accepted full responsibility for the conduct of the study, had access to the data, and controlled the decision to publish.
Funding This work was supported by the National Natural Science Foundation of China (82171274, 81971101, 82171276) and the Science Technology Department of Zhejiang Province (2018C04011).
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.