Clinical data were collected at each hospital through chart review or contact with patients or relatives. This study was conducted in accordance with the Declaration of Helsinki and conformed to Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (17). The complete STROBE checklist is included in Supplementary material. Furthermore, the requirement for written informed consent was waived because the study was retrospective and used anonymized data.

All patients with AIS due to large vessel occlusion (LVO) caused by intracranial atherosclerosis or extracranial carotid atherosclerosis admitted within 7 days of the last known well (LKW) were retrospectively registered in the Recovery by Endovascular Salvage for Cerebral Ultra-acute Embolic and Atherothrombotic Stroke with Large Vessel Occlusion (RESCUE AT-LVO) (18, 19), a historical multicenter registry that included data from 51 hospitals in Japan from January 2017 to December 2019.

For the present sub-study, we reviewed the findings of consecutive patients enrolled in this registry who met the following criteria: (1) AIS due to anterior circulation LVO of the extracranial or intracranial ICA and the M1 or M2 segment of the MCA; (2) underwent EVT; (3) showed TOs (occlusion or stenosis at the c-ICA with ipsilateral intracranial artery occlusion); and (4) available for modified Rankin Scale (mRS) score at 90 days. Patients were excluded if they met any of the following criteria: (1) stenosis caused by a non-atherosclerotic etiology, such as moyamoya disease, arterial dissection, or vasculitis; (2) multiple acute infarctions in multiple vascular territories, excluding artery-to-artery embolism due to c-ICA occlusion or stenosis; (3) underwent EVT alone for cervical lesions; (4) had an unknown onset time; or (5) had posterior circulation LVO.

The following clinical data were collected: age, sex, pre-stroke mRS score, baseline systolic blood pressure, baseline National Institutes of Health Stroke Scale (NIHSS) score, medical history (atrial fibrillation, hypertension, diabetes mellitus, dyslipidemia, IS/transit ischemic attack prior to index stroke, and ischemic heart disease), antithrombotic drugs prior to index stroke (any antiplatelet drugs, single antiplatelet drug, dual antiplatelet drugs, any anticoagulant drugs, and statins), statin use prior to index stroke, and imaging (the Alberta Stroke Program Early Computed Tomographic Score [ASPECTS] on diffusion-weighted magnetic resonance imaging [MRI] or non-contrast computed tomography [CT]). Procedural variables included details of thrombectomy (stent retriever/combined contact aspiration and stent retrievers, contact aspiration, angioplasty, and CAS), antegrade/retrograde thrombectomy, and additional antithrombotic medication during the perioperative period (aspirin, clopidogrel, cilostazol, ticagrelor, prasugrel, intravenous ozagrel sodium, warfarin, and direct oral anticoagulants). Imaging findings included the presence of c-ICA lesions (c-ICA occlusion and stenosis) and the degree of stenosis of the cervical lesion at baseline according to the North American Symptomatic Carotid Endarterectomy Trial (20). Intravenous thrombolysis was performed using alteplase (0.6 mg/kg: the dose approved in Japan) (21). Time delays included time from LKW to hospital arrival, time from LKW to groin puncture, and time from groin puncture to modified Thrombolysis In Cerebral Infarction scale (mTICI) ≥2a reperfusion. Procedural variables included details regarding thrombectomy, antegrade/retrograde thrombectomy, and additional antithrombotic medications during the perioperative period. The perioperative period included the period in which antithrombotic therapy was administered immediately before thrombectomy and during the operative period.

All EVT procedures were performed by physicians certified by the Japanese Society for Neuroendovascular Therapy (22), as recommended by the American Heart Association/American Stroke Association guidelines (7) and the guidelines from the Japan Stroke Society, the Japan Neurological Society, and the Japanese Society for Neuroendovascular Therapy (22). EVT procedures included stent retriever/combined contact aspiration, stent retriever application (23), contact aspiration, angioplasty, and CAS. Procedural device selection was at the discretion of the treating physician, although limited to those approved in Japan. Additionally, the decision to perform antegrade or retrograde thrombectomy or angioplasty/CAS for TO was made at the physician’s discretion. The reperfusion status after EVT was assessed using the mTICI (24).

The type and dosage of APT regimens in the perioperative period (preoperative and intraoperative) were determined by the treating physician according to the institution’s protocol and included aspirin, P2Y₁₂ inhibitors (clopidogrel or prasugrel), and/or cilostazol. Since glycoprotein (GP) IIB/III A inhibitors (tirofiban, epifibatide, or abciximab) and other P2Y₁₂ inhibitors (ticagrelor and cangrelor) are not approved in Japan for ischemic stroke, they were excluded from this sub-study. In this study, DAPT was defined as APT with any two of aspirin, clopidogrel, cilostazol, or prasugrel, and triple APT (TAPT) was defined as APT with any three of aspirin, clopidogrel, cilostazol, or prasugrel. Based on the results of the PRASTRO integrated study (25), prasugrel was approved in Japan in December 2021 for the treatment of non-cardioembolic ischemic stroke within 7 days of onset.

The primary outcome was an mRS score of 0–2 at 90 days, indicating a good functional outcome. Secondary outcomes were defined as death within 90 days, any hemorrhagic event, any intracranial hemorrhage (ICH), SICH, any ischemic event, recurrent IS, post-procedure re-occlusion, and mRS shifts (an increase of 1 point in the mRS score). ICH was assessed using non-contrast CT or gradient-echo MRI 24 ± 8 h after the procedure. Any ICH was defined as any new ICH on imaging, irrespective of the symptoms. SICH was defined as any ICH with a ≥ 4-point increase in the NIHSS score from baseline according to the Heidelberg classification (26). Procedural outcomes were final mTICI ≥2b reperfusion, final mTICI ≥2c reperfusion, and re-occlusion during the procedure.

Data were summarized as median (interquartile range [IQR]) for continuous variables and frequencies and percentages for categorical variables. The number of missing observed variables is also presented. Patients were categorized into those who received a P2Y₁₂ inhibitor (clopidogrel or prasugrel) in the perioperative period (P2Y₁₂ inhibitor [+] group) and those who did not (P2Y₁₂ inhibitor (−) group). Statistically significant differences between groups were assessed using the Mann–Whitney U test, Student’s t-test, Wilcoxon rank-sum test, χ2 test, or Fisher’s exact test, as appropriate. Multivariate logistic regression models were used to evaluate the association between primary and secondary outcomes and P2Y₁₂ inhibitor administration in the perioperative period. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated using the P2Y₁₂ inhibitor (−) group as a reference. The following prespecified variables were included: sex (27), age (28–30), pre-stroke mRS (30), baseline NIHSS score (27), atrial fibrillation (31), hypertension (32), diabetes mellitus (33), dyslipidemia (34), ASPECTS (35), intravenous thrombolysis (28), statin use (36), aspirin administration during the perioperative period (37), and angioplasty and CAS (38). An ordinal logistic regression model was used to analyze shifts in mRS scores at 90 days. Regarding sensitivity analysis, we used inverse probability of treatment weighting (IPTW) to adjust for differences in baseline characteristics. The propensity scores of IPTW analyses were calculated using a mixed-effect logistic regression model. Patients with anterior circulation TO were classified as having occlusion or stenosis of the c-ICA with ipsilateral intracranial artery occlusion, and the same analysis was performed for each group with or without perioperative P2Y₁₂ inhibitors. Moreover, the same analysis was performed for patients who underwent CAS or those who received perioperative aspirin. Patients were categorized into four groups (no APT, SAPT, DAPT, and TAPT) according to the number of antiplatelet drugs administered in the perioperative period, and patient backgrounds and outcomes were compared. All analyses were performed using the Stata/IC statistical package, version 17.1 (Stata Corp LLC, College Station, TX, United States).

Among the 770 patients undergoing EVT for AIS due to extracranial carotid atherosclerosis with anterior circulation TOs enrolled in the RESCUE AT-LVO registry, after excluding 514 with intracranial atherosclerotic stenosis-related LVO stroke, 2 with acute posterior-circulation TOs, and 12 with missing mRS scores at 90 days, the remaining 242 patients with acute anterior circulation TOs (42 females [17.4%]; median age, 76 years [IQR, 69–81 years]; median NIHSS score, 15 [IQR, 10–21]) who underwent EVT were analyzed in this study (Figure 1).

The baseline characteristics of the patients with TOs treated with and without P2Y₁₂ inhibitors during the perioperative period are shown in Table 1. Patients who received P2Y₁₂ inhibitors in the perioperative period showed a lower frequency of atrial fibrillation (1.5% vs. 8.1%, p = 0.03), any anticoagulant drugs prior to the index stroke (1.5% vs. 9.0%, p = 0.01), and perioperative administration with cilostazol (3.8% vs. 10.8%, p < 0.01), and a higher frequency of CAS (86.3% vs. 42.3%, p < 0.01) and treatment with aspirin during the perioperative period than those who did not receive P2Y₁₂ inhibitors (93.1% vs. 20.7%, p < 0.01). The two groups showed no significant differences in procedural complications during EVT (Table 2). Patients with TO treated with P2Y₁₂ inhibitors in the perioperative period showed a higher frequency of good functional outcomes than those who were not treated with P2Y₁₂ inhibitors (mRS scores of 0–2 at 90 days; 42.0% vs. 32.4%; adjusted OR: 4.08, 95% CI: 1.31–12.63). The two groups showed no significant difference in the incidence of death within 90 days (3.8% vs. 9.9%; adjusted OR: 0.82, 95% CI: 0.14–4.90), any ICH (9.2% vs. 15.3%; adjusted OR: 0.33, 95% CI: 0.09–1.06), SICH (5.3% vs. 8.1%; adjusted OR: 0.49, 95% CI: 0.11–2.19), recurrent IS (8.4% vs. 8.1%; adjusted OR: 0.96, 95% CI: 0.21–4.26), and re-occlusion during the procedure (8.4% vs. 6.3%; adjusted OR: 3.57, 95% CI: 0.90–14.29; Table 3). Figure 2 shows the distribution of the mRS scores at 90 days between the two groups. The clinical outcomes were also compared according to the timing of P2Y₁₂ inhibitor administration, and no significant differences were observed (Supplementary Table 1).

Of the total, 143 (59.1%) patients had occlusion at the c-ICA with ipsilateral intracranial artery occlusion. For both patients with occlusion and stenosis at the c-ICA with ipsilateral intracranial artery occlusion, the rate of CAS was significantly higher in the patients who received P2Y₁₂ inhibitors than in those who did not receive P2Y₁₂ inhibitors (Supplementary Table 2). In patients with occlusion at the c-ICA with ipsilateral intracranial artery occlusion, a significantly higher rate of good functional outcomes was found in the patients who received P2Y₁₂ inhibitors than in those who did not (39.0% vs. 28.8%; adjusted 5.33, 95% CI, 1.27–22.29). However, in patients with stenosis at the c-ICA with ipsilateral intracranial artery, no significant difference was found between patients who received P2Y₁₂ inhibitors in the perioperative period and those who did not (Table 4).

Of all 242 patients undergoing EVT for AIS due to extracranial carotid atherosclerosis with anterior circulation TOs, CAS was performed in 160 patients (66.1%), of whom 57/160 (35.6%) underwent angioplasty, and 103 (64.4%) underwent CAS alone. Patient background with and without carotid artery stenting is shown in Supplementary Table 3. Patients receiving P2Y₁₂ inhibitors (n = 113) had a higher frequency of good functional outcome than those not receiving P2Y₁₂ inhibitors (n = 47; 39.8% vs. 34.0%; adjusted OR: 4.79, 95% CI: 1.19–19.19), whereas the two groups showed no significant difference in death within 90 days (2.7% vs. 6.4%; adjusted OR: 0.16, 95% CI: 0.01–111.1) and SICH (4.4% vs. 8.5%; adjusted OR: 0.57, 95% CI: 0.06–5.34; Supplementary Table 4; Supplementary Figure 1A). Among patients who did not undergo CAS, no significant differences between patients receiving P2Y₁₂ inhibitors (n = 18) and those not receiving P2Y₁₂ inhibitors (n = 64) in good functional outcome (55.6% vs. 31.3%; adjusted OR: 3. 56, 95% CI: 0.74–17.13), death within 90 days (11.1% vs. 12.5%; adjusted OR: 1.13, 95% CI: 0.18–7.18) and SICH (11.1% vs. 7.8%; adjusted OR: 1.62, 95% CI: 0.26–10.04) are shown in the Supplementary Table 4 and Supplementary Figure 1B. In the overall cohort, there were no significant statistical differences between patients who underwent CAS and those who did not for good functional outcome (38.1% vs. 36.6%; adjusted OR: 0.67, 95% CI: 0.31–1.46), death within 90 days (3.8% vs. 12.2%; adjusted OR: 0.37, 95% CI: 0.10–1.43), and SICH (5.6% vs. 8.5%; adjusted OR: 0.80, 95% CI: 0.23–2.73).

Among patients receiving aspirin in the perioperative period, patients receiving aspirin and P2Y₁₂ inhibitors (n = 123) showed a higher frequency of CAS (86.9% vs. 60.9%, p < 0.01) and a lower frequency of cilostazol administration during the perioperative period than those receiving aspirin alone or aspirin and APT other than P2Y₁₂ inhibitor (n = 23; 3.3% vs. 47.8%, p < 0.01; Supplementary Table 5). Patients receiving aspirin and P2Y₁₂ inhibitors in the perioperative period had a higher frequency of good functional outcomes than those receiving aspirin alone or aspirin and APT other than P2Y₁₂ inhibitor (41.0% vs. 8.7%; adjusted OR: 7.29, 95% CI: 1.64–35.48). The two groups showed no significant difference in death within 90 days (3.3% vs. 8.7%; adjusted OR: 089, 95% CI: 0.04–20.00) and SICH (4.9% vs. 13.0%; adjusted OR: 0.31, 95% CI: 0.06–1.68; Supplementary Table 6). Furthermore, the distribution of mRS scores at 90 days in patients receiving aspirin in the perioperative period in the two groups is shown in Supplementary Figure 2.

In the perioperative period, of all 242 patients, 87 (36.0%) did not receive any APT, 21 (8.7%) received SAPT, 130 (53.7%) received DAPT, which was the highest of the four groups, and 4 (1.6%) received TAPT. The most common regimen was 200 mg of aspirin (47.1%) among patients receiving SAPT. Among patients receiving DAPT, the most common regimen was 200 mg of aspirin and 300 mg of clopidogrel (40.8%), followed by 300 mg of aspirin and 300 mg of clopidogrel (16.2%; Figure 3; Supplementary Table 7). Of the four patients who received perioperative TAPT, one received 300 mg of aspirin, 30 mg of clopidogrel, and 300 mg of cilostazol; and three received 200 mg of aspirin, 30 mg of clopidogrel, and 300 mg of cilostazol. No significant difference was found in the rate of achieving a good functional outcome or death within 90 days between the perioperative antiplatelet drug regimens (Table 5). Details of the additional antithrombotic drugs administered before onset and during the perioperative period are shown in Supplementary Table 8. Furthermore, details of the pre-antithrombotic and perioperative additional antithrombotic doses are shown in Supplementary Table 9.