Hemorrhagic Complications Following Endovascular Treatment for Atherothrombotic Large Vessel Occlusion

Article information

J Stroke. 2025;27(1):149-153

Publication date (electronic) : 2025 January 31

doi : https://doi.org/10.5853/jos.2024.01935

Satoru Fujiwara¹, Kazutaka Uchida^,2, Tsuyoshi Ohta³, Nobuyuki Ohara¹, Michi Kawamoto¹, Hiroshi Yamagami⁴, Kazunori Toyoda⁵, Yuji Matsumaru⁶, Yasushi Matsumoto⁷, Kenichi Todo⁸, Mikito Hayakawa⁹, Seigo Shindo^10,11, Shinzo Ota¹², Masafumi Morimoto¹³, Masataka Takeuchi¹⁴, Hirotoshi Imamura¹⁵, Hiroyuki Ikeda¹⁶, Kanta Tanaka^5,17, Hideyuki Ishihara¹⁸, Hiroto Kakita¹⁹, Takanori Sano^20,21, Hayato Araki²², Tatsufumi Nomura²³, Mikiya Beppu², Fumihiro Sakakibara², Manabu Shirakawa², Shinichi Yoshimura², Nobuyuki Sakai^24,25

¹Department of Neurology, Kobe City Medical Center General Hospital, Kobe, Japan

²Department of Neurosurgery, Hyogo Medical University, Nishinomiya, Japan

³Department of Neurosurgery, Kobe City Medical Center General Hospital, Kobe, Japan

⁴Division of Stroke Prevention and Treatment, University of Tsukuba, Tsukuba, Japan

⁵Department of Cerebrovascular Medicine, National Cerebral and Cardiovascular Center, Suita, Japan

⁶Department of Neurosurgery, University of Tsukuba, Tsukuba, Japan

⁷Division of Development and Discovery of Interventional Therapy, Tohoku University Hospital, Sendai, Japan

⁸Department of Neurology, Osaka University Graduate School of Medicine, Suita, Japan

⁹Department of Neurology, Institute of Medicine, University of Tsukuba, Tsukuba, Japan

¹⁰Department of Neurology, Japanese Red Cross Kumamoto Hospital, Kumamoto, Japan

¹¹Department of Neurology, Kumamoto University, Kumamoto, Japan

¹²Department of Neurosurgery, Brain Attack Center Ota Memorial Hospital, Fukuyama, Japan

¹³Department of Neurosurgery, Yokohama Shintoshi Neurosurgical Hospital, Yokohama, Japan

¹⁴Department of Neurosurgery, Seisho Hospital, Odawara, Japan

¹⁵Department of Neurosurgery, National Cerebral and Cardiovascular Center, Suita, Japan

¹⁶Department of Neurosurgery, Kurashiki Central Hospital, Kurashiki, Japan

¹⁷Stroke Center, Kindai University, Osaka-Sayama, Japan

¹⁸Department of Neurosurgery, Yamaguchi University School of Medicine, Ube, Japan

¹⁹Department of Neurosurgery, Seijinkai Shimizu Hospital, Kyoto, Japan

²⁰Department of Neurosurgery, Japanese Red Cross Ise Hospital, Ise, Japan

²¹Department of Neurosurgery, Mie Prefectural General Medical Center, Yokkaichi, Japan

²²Department of Neurosurgery, Araki Neurosurgical Hospital, Hiroshima, Japan

²³Department of Neurosurgery, Ohkawara Neurosurgical Hospital, Muroran, Japan

²⁴President, Seijinkai Shimizu Hospital, Kyoto, Japan

²⁵Department of Neurovascular Research, Kobe City Medical Center General Hospital, Kobe, Japan

Correspondence: Kazutaka Uchida Department of Neurosurgery, Hyogo Medical University, 1-1 Mukogawa, Nishinomiya, Hyogo 663-8501, Japan Tel: +81-798-45-6458 E-mail: kuchidans@gmail.com

Received 2024 May 22; Revised 2024 September 26; Accepted 2024 October 25.

Dear Sir:

Approximately 15% of large vessel occlusions (LVOs) treated with endovascular treatment (EVT) are classified as atherothrombotic stroke-related LVOs (AT-LVOs), which encompass two underlying etiologies: tandem and intracranial atherosclerotic occlusions [1]. EVT for AT-LVOs commonly necessitates the intraoperative or early postoperative addition of antithrombotic drugs, even after thrombolytic therapy [2], thereby increasing the risk of hemorrhagic complications. Herein, we investigated the differences in the incidence and characteristics of hemorrhagic complications following EVT for AT-LVOs according to etiology.

We conducted a post hoc analysis of the Recovery by Endovascular Salvage for Cerebral Ultra-acute Embolic and Atherothrombotic Stroke with Large Vessel Occlusion (RESCUE AT-LVO), that enrolled 783 patients who underwent EVT for AT-LVOs at 51 centers across Japan between January 2017 and December 2019 [3]. AT-LVOs were defined as follows: (1) residual stenosis of ≥50% on final angiography during EVT, and (2) residual stenosis <50% with the appearance of a tapered sign and/or flow impairment during EVT [4]. We further included patients with anterior circulation AT-LVOs, excluding those with (1) a non-atherosclerotic etiology, (2) multiple acute infarctions in multiple vascular territories, and (3) an unknown onset time. According to the etiology of intracranial artery occlusion, patients with artery-to-artery embolism from tandem extracranial atherosclerosis were classified into the tandem group, while those with in situ occlusion of intracranial atherosclerosis were classified into the intracranial group. The primary outcome was any intracranial hemorrhage (ICH), while secondary outcomes included ICH subtypes, symptomatic ICH (a worsening of ≥4 on the National Institutes of Health Stroke Scale [NIHSS]) [5], decompressive craniectomy, and all-cause mortality within 90 days. ICH was assessed by brain computed tomography or magnetic resonance imaging 24±8 hours following EVT, and categorized into hemorrhagic infarction, parenchymal hematoma, and subarachnoid hemorrhage [6]. We further constructed Firth’s bias-reduced logistic regression model to estimate the adjusted odds ratios (aORs) and 95% confidence intervals (CIs). Considering the substantial differences in patient backgrounds, we further calculated the propensity score and performed a comparison between the two groups using inverse probability of treatment weighting. Details of the statistical analyses are presented in the Supplementary Table 1.

We analyzed 635 patients (mean age, 73.1 yr; 72% male); of these, 254 (40%) were classified into the tandem group, of whom 166 (65%) underwent carotid stenting. Regarding treatment of the intracranial artery, thrombectomy was performed in 474 (75%) patients, while stenting was performed in 65 (10%) patients (Table 1). A comparison of the characteristics of the two groups is presented in Table 1. ICH after EVT was observed in 53 (8.3%) patients, with a significantly higher incidence in the tandem group compared to the intracranial group (11.8% vs. 6.0%, aOR, 1.94; 95% CI, 1.08–3.48) (Table 2). The distribution of hemorrhagic complications revealed significant differences between the two groups (Figure 1), with a higher proportion of parenchymal hematomas (6.3% vs. 0.8%, aOR, 6.71; 95% CI, 2.03–22.2) (Table 2). The tandem group had a higher incidence of symptomatic ICH (6.3% vs. 1.3%, aOR, 4.93; 95% CI, 1.80–13.6). There were no significant differences in the incidences of hemorrhagic infarction, subarachnoid hemorrhage, decompressive craniotomy, or all-cause mortality within 90 days (Table 2). Similar trends were observed in additional analyses (Supplementary Table 2).

Table 1.

Comparison of patient characteristics between the tandem and intracranial groups

Table 2.

Differences in type of ICH following endovascular treatment for AT-LVO

Figure 1.

Intracranial hemorrhages following endovascular treatment for AT-LVO. AT-LVO, atherothrombotic stroke-related large vessel occlusion; ICH, intracranial hemorrhage; PH, parenchymal hematoma; HI, hemorrhagic infarction; SAH, subarachnoid hemorrhage.

The incidence of ICH following EVT for AT-LVOs has been reported in previous studies conducted in various settings. However, tandem and intracranial atherosclerotic occlusions have typically been discussed separately, while the differences in ICH between these two etiologies have not been studied. Our study is unique in that it examined the differences in the incidence and subtypes of hemorrhagic complications after EVT for AT-LVOs according to the two etiologies using a large-scale registry. There are several possible reasons for different types of ICH following EVT in patients with AT-LVOs. For example, patients in the intracranial group may have better collateral circulation at the distal site of the acute occlusion, facilitated by chronic atherosclerotic stenosis, compared to acute occlusion caused by an embolus, as demonstrated in a prior study [6]. This may mitigate ischemic damage, and consequently lower the risk of parenchymal hematoma formation that could become symptomatic. Indeed, the ASPECTS scores on admission tended to be higher in the intracranial group. Further, our results underscore the importance of managing ICH risk during EVT and recognizing differences based on the diagnosed etiologies, even within the same category of AT-LVOs. While there is currently no robust evidence suggesting that additional antiplatelet or thrombolytic therapies adversely affect outcomes after EVT for AT-LVOs, antithrombotic medications may be a concern for ICH. Our results indicate that strategies to reduce the risk of ICH, particularly in cases of tandem occlusion, may be important for improving the outcomes of patients with AT-LVOs. Indeed, a weighted analysis of a registry study of patients with tandem occlusion showed that patients who underwent carotid stenting had a higher incidence of any and symptomatic ICH than those who did not undergo stenting [7]. While considering the risk of re-occlusion, skipping stenting during emergent EVT or staged angioplasty to avoid hyperperfusion, which may lead to hemorrhagic complications [8], could be a useful option.

This study had several limitations. Firstly, this multicenter registry study did not include eligibility criteria or a predefined protocol for EVT in AT-LVOs; as such, selection bias was unavoidable. Second, our analysis did not consider some potential confounding factors between post-EVT hemorrhage and AT-LVOs etiology, due to the small number of events. Several factors, such as the skill level of the operator, EVT procedures, collateral state or ischemic volume assessed on perfusion images, and changes in blood pressure during EVT, may all have influenced the occurrence of ICH. Third, the number of patients who developed ICH was relatively small; therefore, although we used what we considered to be the most appropriate analytical method, caution is required when interpreting the results. Finally, the findings from our registry may not be generalizable to other settings.

In conclusion, the frequency and subtypes of ICH following EVT for AT-LVOs vary according to the underlying etiology. Perioperative management based on awareness of differences in ICH risk may be beneficial for further improving the outcomes of AT-LVOs.