As the second most lethal and disabling disease worldwide, acute ischemic stroke (AIS) has a serious impact on patient survival and quality of life (1). Approximately 20% of patients with AIS arrive at the hospital with a large infarct core, characterized by high mortality and poor prognosis, posing a global challenge in clinical management for this population (2). Endovascular therapy (EVT) has emerged as the preferred treatment option for AIS due to large vessel occlusion.

A previous study reported that, compared with patients treated with medication alone, those with large infarct-core AIS treated with EVT plus medication had a higher likelihood of reduced disability, independent ambulation, and favorable functional outcomes at 90 days (3). The Chinese Guidelines for Endovascular Treatment of Acute Ischemic Stroke 2023 recommend EVT for patients with large-vessel occlusion-induced AIS in the pre-circulation stage among those who meet the enrolment criteria of the ANGEL-ASPECT (4), RESCUE-Japan LIMIT (5), or SELECT 2 (6) studies (7).

Despite the success associated with EVT, intracranial hemorrhage (ICH) is a common complication, with a prevalence of 46.0–49.5% (8). In particular, symptomatic ICH (sICH) can reduce the benefit–risk ratio of treatment and increase the risk of mortality (9). However, limited data are available regarding the predictors and clinical relevance of sICH after EVT. Therefore, risk factors for sICH after EVT should be identified for the prevention of sICH and improvement in the efficacy of this new treatment strategy. Utilizing the database from the Prospective Multicenter Cohort Study of Early Treatment in Acute Stroke (MAGIC) study, the present study aimed to analyze potential predictors of sICH after EVT in patients with pre-circulation large infarct-core AIS.

A total of 490 patients with pre-circulation large infarct-core AIS who were treated with EVT were enrolled in this study. They were divided into three groups: without ICH, with aICH, and with sICH. Overall, 115 patients (23.5%) had aICH after EVT within 24 h, whereas 65 patients (13.3%) had sICH. The proportions of reaching favorable functional outcomes in the three groups were 43.5, 32.2, and 13.8%, respectively (p < 0.001) (Figure 1 and Table 1). Significant differences were noted in the probability of developing brain herniation (21.6% vs. 30.4% vs. 58.5%, p < 0.001) and mortality at 90 days (35.2% vs. 44.3% vs. 69.2%, p < 0.001) (Table 1).

When comparing sICH based on the level of blood glucose at admission, a significant intergroup difference was noted (7.6 ± 2.8 vs. 8.0 ± 2.9 vs. 9.25 ± 3.5 mmol/L, respectively, p < 0.001) (Table 1). The one-way ANOVA revealed that the levels of blood glucose at admission in the sICH group were significantly different from those in the without ICH and with aICH groups (p < 0.001; p = 0.008, respectively). The count showed that the levels of blood glucose at admission in the sICH group were significantly different from those in the without ICH and with aICH groups (p < 0.001; p = 0.024, respectively, by the Bonferroni correction) (Figure 2). In multivariate analysis, higher blood glucose levels at admission (OR 1.169, p < 0.001, CI 1.076–1.269) were associated with sICH after EVT in patients with pre-circulation large infarct-core AIS (Table 2).

Figure 3 presents the results of the ROC analysis determining the prognostic value of blood glucose levels at admission to predict sICH. The area under the curve for the model was 0.685, with a 95% CI of 0.616–0.754, indicating that blood glucose levels at admission have a good discriminative ability. The optimal threshold was 6.95 mmol/L, corresponding to a sensitivity and specificity of 82.0 and 49.8%, respectively, with Youden’s index of 0.312. The positive predictive value was 19.2%, and the negative predictive value was 84.1%.

In our study, we explored potential predictors of sICH after EVT in patients with pre-circulation large infarct-core AIS and observed that the probability of sICH was 13.3% in patients who underwent EVT. Furthermore, sICH greatly reduced the probability of patients having a 90-day favorable prognosis and increased the risk of brain herniation and mortality. In addition, we demonstrated that the blood glucose level at admission was an independent predictor of sICH after EVT in AIS patients with large infarct cores, and the risk of sICH increased with higher blood glucose levels.

In our study, the occurrence of sICH was similar to that in the SELECT trial, which reported an increased probability of sICH in patients with large-vessel occlusion of large-core infarcts who were treated with EVT with an onset greater than 24 h (11). However, our results showed a higher value (13.3%) than that reported in a previous randomized controlled trial (4.4%) (12) and in the North American Solitaire Acute Stroke Study (9.9%) (13). This discrepancy may stem from the inclusion of patients in our study with large infarct cores in the anterior circulation and an ASPECTS score of ≤5, whereas previous EVT studies have largely excluded patients who had large infarct cores already present at the time of preoperative imaging. Several studies have demonstrated an association between larger cores on imaging and an increased risk of sICH after EVT (14, 15). In most studies, an ASPECTS score of ≤5 was reported to be associated with an approximately 2-fold risk of ICH (16, 17). Patients with large infarct cores have a high degree of blood–brain barrier disruption, and the development of ICH is closely related to blood–brain barrier damage. In previous models of cerebral ischemia in rats, loss of integrity of the matrix and basement membrane connecting the endothelial cells was observed by electron microscopy, and in some cases, glial cell protrusions were observed to be swollen or degenerative. These alterations affect the structure and function of the blood–brain barrier, which induces ICH by increasing its permeability (18).

Experimental results have shown that the ischemia–reperfusion-induced release and activation of metalloproteinases can cause rupture of the basement membrane, leading to ICH (19, 20). The results of a previous meta-analysis demonstrated that, compared to pharmacological treatment, EVT did not increase the risk of sICH but improved the functional outcome in patients with large infarct cores (21, 22). Based on the above observations, it is reasonable to attribute the higher incidence of sICH in our study to the larger infarct core rather than the EVT procedure itself.

Similar to the findings of Shen et al. (23), our results showed that patients with sICH had a higher risk of brain herniation and mortality and a greater likelihood of unfavorable outcomes than patients without ICH and with aICH. This emphasizes the critical nature of sICH as a potentially fatal complication that significantly influences patient functional outcomes, warranting emphasis on preventive measures.

In our study, blood glucose was identified as a predictor of sICH after EVT in AIS patients with large infarct cores in the anterior circulation. The blood glucose levels >6.95 mmol/L at admission were associated with an increased risk of sICH after EVT, comparable to a level of 6.6 [5.7–7.7] mmol/L reported in a previous study (24). Several previous studies have suggested that patients with hyperglycemia are at a high risk of ICH from different perspectives, such as a meta-analysis that retrospectively analyzed the clinical data of patients undergoing EVT and found that higher blood glucose levels at admission were associated with a higher incidence of sICH (25, 26). Previous studies have shown that acute hyperglycemia increases blood–brain barrier disruption and ICH incidence in rat models (27). The results of Shen et al.’s (23) study suggest that high blood glucose levels are the strongest predictors of sICH after EVT. Another study showed that an elevated glycosylated hemoglobin level was significantly associated with an unfavorable prognosis and mortality in 90 days with AIS treated with EVT (28). The results of these studies are consistent with our findings; however, the mechanisms underlying how hyperglycemia enhances ICH remain unclear. The possible causes of ICH are as follows: (1) hyperglycemia exacerbates vascular wall dystrophy and hypoxia, resulting in more susceptible degeneration and necrosis of the vascular wall (29, 30); (2) hyperglycemia leads to disruption of cellular metabolism, resulting in increased plasma osmolality and intracellular lactic acid buildup, ultimately leading to endothelial cell injury and acidosis (31); and (3) hyperglycemia can reportedly increase the activity of the matrix metalloproteinases (MMP-9 and MMP-3) in ischemic areas and aggravate blood–brain barrier dysfunction and post-reperfusion ICH (32–34). While it is hypothesized that maintaining stable blood glucose levels post-successful EVT may be crucial in preventing sICH (35), further studies are needed to confirm this.

This study has some limitations that should be considered when interpreting the results. First, as the study adopted a prospective observational design, we did not evaluate other potentially relevant variables, such as changes in perioperative blood pressure, use of heparin during EVT, and the effect of antiplatelet regimens on hemorrhagic conversion, which may influence the risk of ICH. Second, an incomplete evaluation index may not fully determine a causal connection, and the practical implications of these unresolved findings require future validation. Despite these limitations, the findings of this study provide insights into the predictive value of blood glucose levels at admission for sICH after EVT in pre-circulation large infarct core-AIS to facilitate early management.

In this multicenter study, the incidence of sICH after EVT was higher than that previously reported. Patients with pre-circulation large infarct core AIS who developed sICH after EVT had a higher incidence of unfavorable prognosis and mortality than those who did not develop sICH. Higher blood glucose levels at admission increased the risk of sICH after EVT in AIS patients with large infarct cores in the anterior circulation. It remains unclear whether controlling blood glucose levels prior to EVT can reduce the risk of sICH, and further investigation is required.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

The studies involving humans were approved by the ethics committee of Suzhou Dushu Lake Hospital. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

YY: Writing – original draft, Data curation, Formal analysis, Investigation. LY: Writing – original draft, Data curation, Methodology. XS: Writing – original draft, Formal analysis, Investigation, Software. XN: Formal analysis, Writing – original draft, Data curation, Funding acquisition, Methodology. SF: Data curation, Writing – original draft, Conceptualization, Investigation, Supervision. XX: Investigation, Supervision, Data curation, Writing – original draft, Formal analysis, Methodology. JM: Methodology, Data curation, Writing – original draft, Software, Validation. SY: Methodology, Writing – original draft, Conceptualization, Project administration, Resources, Supervision. ZW: Methodology, Writing – original draft, Formal analysis, Software, Validation. WZ: Validation, Writing – review & editing, Data curation, Funding acquisition, Project administration, Resources. DY: Funding acquisition, Project administration, Resources, Writing – review & editing, Data curation, Investigation, Formal analysis. YH: Investigation, Project administration, Writing – review & editing, Supervision, Visualization.