Futile recanalization after endovascular treatment in acute ischemic stroke with large ischemic core

Abstract

Background

Endovascular therapy (EVT) is the treatment of choice for acute ischemic stroke (AIS) with large vessel occlusion. However, in many patients, successful EVT recanalization does not correspond to a clinical improvement, called futile recanalization (FR). We aimed to identify stroke risk factors and patient characteristics associated with FR in AIS with large core infarct (LCI).

Methods

A total of 137 patients with AIS with LCI treated by EVT at a single stroke center were retrospectively included from January 2016 to June 2023. LCI was defined by Diffusion-Weighted Imaging-Alberta Stroke Program Early Computed Tomography Score (DWI-ASPECT) < 6. Patient age, sex, modified Rankin Scale (mRS), National Institutes of Health Stroke Scale (NIHSS), time to treatment, risk factors, and radiologic findings were collected, and potential associations with FR were analyzed. FR was defined as successful reperfusion with modified Thrombolysis in Cerebral Infarction (mTICI) ≥ 2b but without functional independence at 90 days (mRS ≥ 3). A multivariate logistic regression analysis was conducted on the clinical characteristics of patients, based on the presence or absence of FR, and the factors influencing FR.

Results

Of 137 patients, 120 showed successful recanalization (mTICI ≥ 2b). All patients were divided into FR (n = 80) and no FR (n = 40) groups. Older age (odds ratio [OR] 1.052, 95% confidence interval [CI] 1.002–1.105; p = 0.041), the higher the initial NIHSS score (OR 1.181, 95% CI 1.037–1.344; p = 0.012), and prior intravenous plasminogen activator (OR 0.310, 95% CI 0.118–0.813, p = 0.017) were independent influencing factors of FR.

Conclusions

The older age, the higher the initial NIHSS, and not receiving intravenous plasminogen activator were independently associated with FR in AIS with LCI. These factors could identify poor responders to EVT recanalization.

Introduction

Recent advancements in endovascular interventions have changed the management of patients with large core infarcts (LCI), specifically those with an Alberta Stroke Program Early Computed Tomography Score (ASPECT) of 5 or lower, for whom the benefits of endovascular therapy (EVT) remain uncertain. This cohort was traditionally omitted from early trials due to safety apprehensions. Notably, trials such as SELECT2 (Randomized Controlled Trial to Optimize Patient’s Selection for Endovascular Treatment in Acute Ischemic Stroke) and ANGEL-ASPECT (Study of Endovascular Therapy in Acute Anterior Circulation Large Vessel Occlusive Patients With a Large Infarct Core) corroborated findings from RESCUE-Japan LIMIT (Recovery by Endovascular Salvage for Cerebral Ultra-Acute Embolism-Japan Large Ischemic Core Trial), affirming that patients with large ischemic stroke attain superior functional outcomes after EVT compared to best medical treatment [1,2,3]. These randomized controlled trials provide compelling evidence supporting the extension of EVT to specific subgroups of patients with substantial baseline stroke up to 24 h after the last known well.

There is much that early recanalization can improve the prognosis of AIS with LCI [1, 3]. However, recent randomized clinical trials have shown that many patients undergoing successful recanalization have poor clinical outcomes, termed futile recanalization (FR) [4]. The difference between recanalization success and patient outcomes requires an investigation into the factors that determine the overall efficacy of EVT.

While numerous prior studies have identified predictors of FR following endovascular treatment [5,6,7,8], few have specifically examined this issue in patients with LCI. Consequently, this study aimed to identify potential predictors of FR in AIS patients with LCI who underwent endovascular thrombectomy. By identifying precise predictors of FR in this high-risk cohort, our study aims to inform clinical decision-making and optimize therapeutic strategies, ultimately enhancing the efficacy of EVT in AIS patients with LCI.

Methods

Patients

Medical records of AIS patients treated with EVT between January 2016 to June 2022 were identified from the Chonnam National University Hospital (CNUH) stroke registry. We analyzed the data of patients with ischemic anterior circulation stroke with large ischemic core who underwent artery puncture within 24 h from symptom onset and achieved successful angiographic recanalization (defined as modified Thrombolysis in Cerebral Infarction [mTICI] grade 2b or 3). Enrolled patients were > 18 years old, examined within 24 h of the last known time the patient was observed to be well, had occlusions of the internal carotid artery (either extracranial or intracranial), M1, or proximal M2 segment of the middle cerebral artery on magnetic resonance angiography (MRA), and had large ischemic cores with Diffusion-Weighted Imaging-Alberta Stroke Program Early Computed Tomography Scores (DWI ASPECTS) < 6. At our center, the stroke evaluation protocol includes diffusion MRI and MR angiography as part of the initial examination for ischemic stroke, unless contraindicated. All patients included in this study underwent these imaging procedures. n intravenous thrombolytic (Actilyse) was administered to patients who were diagnosed within 4.5 h of stroke onset. Trevo stent retrievers were used first, and aspiration devices and ballooning devices were utilized as determined by the interventionist’s judgement. We examined the modified Rankin Scale (mRS) score at 90 days post-treatment. FR was defined as the poor outcome (mRS score, 3–6) at 90 days despite successful revascularization, and favorable recanalization was defined as functional independence (mRS score, 0–2). Patients were divided into the FR group and the non-FR group accordingly. The Institutional Review Board (IRB) of CNUH approved this study, which was conducted according to the ethical standards laid down by the Declaration of Helsinki (IRB no. CNUH-2020-024).

Data collection

Clinical variables were age, gender, initial National Institutes of Health Stroke Scale (NIHSS), and the presence of cerebrovascular risk factors (hypertension, diabetes mellitus, dyslipidemia, smoking history, and atrial fibrillation). Blood pressure, blood sugar, lipid profile, and complete blood count with differential were measured at arrival. Underlying medical conditions, including hypertension and dyslipidemia, were determined based on the patient’s previous medical history and medication use. Additionally, diabetes and dyslipidemia were diagnosed if blood test results, such as lipid profiles and hemoglobin A1c, taken at the time of admission, met the relevant criteria. Hypertension was diagnosed if elevated blood pressure was observed during the hospital stay. A smoker was defined as anyone with a history of smoking within the past five years. Patients were classified into subgroups according to Trials of Org 10,172 in Acute Stroke Treatment (TOAST) classification.

Patients who underwent emergency magnetic resonance imaging (MRI) were targeted, and a large ischemic core was defined by DWI ASPECTS < 6. The collateral grade before EVT from cerebral arteriography was defined as described in Higashida et al. [9] In our study, collateral flow grading was assessed using MRA and categorized into four grades. Grade 0 indicates no visible collateral flow to the ischemic area. Grade 1 represents slow collateral flow reaching only the periphery, with some perfusion deficit. Grade 2 shows faster collateral flow but covering only part of the ischemic territory. Grade 3 reflects slow but complete perfusion of the ischemic bed by the late venous phase. Grade 4 indicates rapid and complete collateral flow to the entire ischemic territory. Successful vascular recanalization was defined as a modified thrombolysis in cerebral ischemia (mTICI) scale of 2b or 3 after EVT. Computed tomography or MRI scans were usually performed 24 h after EVT, or when an intracranial hemorrhage was suspected based on clinical symptoms. The radiographic definition of hemorrhagic transformation was classified by the European Cooperative Acute Stroke Study (ECASS2) based on MRI after EVT. Symptomatic intracerebral hemorrhage (ICH) was defined as MRI evidence of ICH and rapid neurological deterioration (an increase in NIHSS by ≥ 4 points). Our stroke center conducts regular monthly meetings for data quality control, and the imaging analyses of the patients included in this data were performed by two or more neurologists in an anonymized and blinded manner. Any ambiguous findings were discussed during the meetings to minimize errors.

Statistical analysis

All statistical analyses were conducted using SPSS 22.0. A two-sided p-value < 0.05 indicated statistical significance. Patients were divided into two groups based on mRS at 3 months after stroke. Univariate analysis was performed to identify clinical and imaging variables associated with functional outcomes. The student’s t-test was used for continuous variables, and Chi-squared or Fisher’s exact tests were used for categorical variables. Multivariate logistic regression analysis was used for risk factors and adjusted for confounders. We conducted a multivariate analysis and selected variables with significant differences between the two groups in the General Characteristics as covariates. The adjusted variables included age, atrial fibrillation, glomerular filtration rate, initial SBP, initial NIHSS, postprocedural hemorrhagic transformation, and collateral grade before EVT.

Results

A total of 662 patients underwent screening between January 2016 and June 2023. Of 137 patients with AIS with large ischemic core treated by EVT, 120 showed successful recanalization (mTICI ≥ 2b).

These patients were divided into those who had FR (n = 80) versus those with no FR (n = 40) (Fig. 1). The included patient cohort consisted of 75 men (62.5%), with a mean age of 71.1 ± 13.8 years and a mean initial NIHSS of 13 (11–17). The mean duration between the last known time the patient was observed to be well and arterial puncture was 342.1 ± 261.2 min, while the mean duration between the last known time the patient was observed to be well and recanalization was 372.4 ± 260.2 min. Forty-eight patients (40.0%) received intravenous thrombolysis, and 91 patients (75.8%) had a poor collateral grade before EVT (0–2). The mean DWI ASPECTS was 4 (3–5). Hemorrhagic transformation post-EVT occurred in 73 patients (60.8%), and symptomatic ICH was observed in 8 patients (6.7%). The vascular occlusion sites were as follows: 76 patients (63.3%) had isolated MCA occlusion, while 44 patients (36.7%) had occlusion extending from the ICA. (Table 1).

Fig. 1

Flowchart of patient selection

Table 1 Patient baseline Characteristics*

The FR group comprised 80 patients (75.1%) (Table 2). In univariate analysis, the FR group was significantly older (p < 0.001), had poorer collateral grades before EVT (p = 0.016), more frequent atrial fibrillation (p = 0.036), more instances of post-procedure hemorrhagic transformation (p = 0.001), higher initial NIHSS scores (P < 0.001), and higher initial systolic blood pressure (p = 0.035) compared to the non-FR group. No significant differences between the two groups were observed for inflammatory markers, including the neutrophil-to-lymphocyte ratio, use of intravenous recombinant tissue plasminogen activator (rt-PA), and TOAST classification.

Table 2 Comparison of factors between favorable and unfavorable group*

After adjusting for potential confounders in the multivariate logistic regression analysis, factors associated with FR were older age (odds ratio [OR] 1.052, 95% confidence interval [CI] 1.002–1.105; p = 0.041) and higher initial NIHSS scores (OR, 1.181; 95% CI, 1.037–1.344; p = 0.012). Additionally, treatment with intravenous rt-PA usage was negatively associated with FR (OR, 0.310, 95% CI, 0.118–0.813, p = 0.017) (Table 3).

Table 3 Multivariate analysis of the association between factors and functional outcomes after EVT

Discussion

This study identified predictors of FR in patients with LCI who do not have clinical benefit even after EVT with adequate recanalization. FR after EVT was relatively common in AIS with LCI. Older age, high initial NIHSS score, and no rt-PA before EVT were associated with FR.

The definition of FR is still controversial, but it is generally defined by functional mRS ≥ 3 despite successful vascular recanalization (mTICI 2b–3) [10]. The prevalence of FR may vary depending on the definition, and FR occurs in 32.4–69.6% of patients with AIS due to large vessel obstruction after EVT [5]. Nevertheless, the mechanism, clinical predictors, and management for FR are still lacking. Many trials recently have created some backgrounds for performing EVT even in the presence of a large ischemic core [1,2,3]. With advancements in endovascular therapy (EVT), successful vascular recanalization can now be achieved in up to 90% of cases. However, despite the administration of appropriate treatment, futile recanalization (FR) remains relatively common. Identifying the factors associated with FR is crucial, as it will provide clinicians with valuable insights for optimizing patient management and facilitating the effective distribution of medical resources [11].

While many studies have already identified predictors of FR in AIS, our study is unique in that it focuses exclusively on patients with LCI. As the criteria for EVT become increasingly liberal and the eligible patient expands, identifying specific characteristics in LCI, where treatment outcomes remain uncertain, could aid in resource allocation, improve decision-making, and better manage patient expectations. In addition, we assessed the ASPECT score using DWI instead of CT. While this may overestimate the true ischemic core, our study’s focus on LCI allowed us to detect the ischemic core more sensitively and evaluate its prognostic factors.

Risk factors for FR were assessed in various EVT studies. In the subgroup analysis of DIRECT-MT investigating risk factors for FR, the incidence of FR was higher with increasing age and higher initial systolic blood pressure [6, 12]. In our study, FR was more likely to occur with older age, and the mean age of the FR group was significantly higher than that of the non-FR group (75.1 vs. 63.1, p < 0.001). Another meta-analysis similarly reported that older age, severity of initial NIHSS score, degree of collateral flow, and initial ASPECT score were associated with FR. In our study, good collateral flow showed significance in univariate analysis; however, it was not associated with FR in multivariate analysis [13].

Unlike previous studies, our research specifically identified the factors most likely to contribute to FR in patients with AIS and LCI. Since earlier studies on FR did not focus exclusively on patients with LCI, particularly those with low ASPECT scores, discrepancies in the findings between our study and prior research are expected. Our study’s emphasis on this distinct patient subgroup provides a more targeted understanding of FR in cases with significant infarct burden.

In our registry, FR occurred in approximately 66% of cases, showing the frequency slightly higher than in other studies. This is likely due to the relatively longer door-to-recanalization time compared to other studies. In a well-designed study, DIRECT-MT, the door-to-recanalization time was around 280 min, whereas it was approximately 380 min in our registry [14]. This high duration is explained by the geographical challenges of our stroke center, which encompasses numerous islands, mountainous terrain, and rural areas. It will serve as a foundation for highlighting the importance of pre-hospital management support in the future.

Previous research has suggested a relationship between FR and high NIHSS scores [7, 8, 15,16,17]. In patients with severe stroke, while the benefit of EVT was greater compared to medical treatment alone, FR tended to increase as stroke severity escalated. However, despite this increase in FR, the overall advantage of EVT remained more pronounced [8, 18]. This trend was similarly noted in our study. Therefore, it would be unnecessary to exclude patients with large ischemic cores from EVT based on the high NIHSS score. There are controversies about the effect of prior collateral flow status on the FR [5, 17, 19]. In our study, we observed that patients who experienced FR tended to have poor collateral flow. Collateral flow is known to develop poorly in elderly patients or cases of cardioembolic stroke [20,21,22,23]. The relatively high prevalence of atrial fibrillation in the FR group, along with older age, might contribute to poor collateral flow within this group. In patients already presenting with a large ischemic core with a small penumbra, the efficacy of collateral flow is diminished, suggesting that the advantage provided by collateral flow might have decreased somewhat. However, although not statistically significant, good collaterals showed a trend of protective effect with an adjusted odds ratio of 0.386. Therefore, a well-designed study with a larger sample size is necessary in the future.

Thrombolysis administered before EVT also exhibited a protective effect against FR (OR 0.310, 95% CI 0.118–0.813, p = 0.017). Recent studies have excluded thrombolysis before EVT [24,25,26], but our data suggest that omitting thrombolysis in patients with large ischemic core does not yield significant benefits and remains advantageous.

Our study had several limitations. First, the study had generalization issues as it was conducted at a single comprehensive stroke center. Second, the retrospective design limited the control of all variables that may influence patient treatment or outcomes. Third, the sample size was relatively small, indicating that a multicenter study targeting a larger population is necessary in the future. Fourth, our study assessed the ASPECT score using DWI, but since most previous studies have typically used CT-based ASPECT scoring, there may be limitations in generalizing our results. Fifth, although we defined functional independence as mRS 0–2 based on the definition of FR, this is stricter than the outcome of mRS 0–3 used in previous large-core trials. Hence, caution is required in interpreting the results, and further studies focusing only on LCI patients may be necessary. Last, since our study lacks information about the final infarct volume, further analysis will be necessary.

In conclusion, FR was prevalent among patients undergoing EVT with a large ischemic core. Older age, high initial NIHSS score, and prior thrombolytic treatment were identified as factors influencing FR. In the future, a well-designed prospective multicenter study is warranted for analyzing the impact of FR.