Abstract
Objective:
By developing and comparing various preoperative multimodal CT-based models to predict prognosis after successful recanalization in acute anterior circulation occlusive stroke, this study aimed to ascertain the prognostic assessment of patients following successful recanalization.
Methods:
Patients with acute anterior circulation large vessel occlusion who underwent mechanical thrombectomy and achieved successful recanalization were consecutively enrolled from Sanming First Hospital Affiliated to Fujian Medical University between January 2022 and October 2024. Based on the 90-day modified Rankin Scale (mRS) scores, the patients were categorized into a favorable clinical outcome group (mRS 0–3) and an unfavorable clinical outcome group (mRS 4–6). Following the identification of statistically significant variables from the analyzed clinical and imaging data, optimal cut-off values were computed. Five multivariate logistic regression models were subsequently constructed: Clinical-Imaging (C-I), Clinical-non-Perfusion (C-NP), Clinical-non-Angiography (C-NC), Clinical-only (C), and Imaging-only (I). Models performance were compared using receiver operating characteristic (ROC) curve analysis, with comparisons based on the area under the curve (AUC), sensitivity, and specificity. Using the Delong test for model comparison and selection, a nomogram of the optimal model was developed. Model performance was subsequently assessed by means of a confusion matrix and 5-fold cross-validation.
Results:
Of the 131 enrolled patients, 77 (58.78%) were classified into the favorable clinical outcome group and 54 (41.22%) into the unfavorable clinical outcome group. Statistically significant differences were identified in age (59.5 years), preoperative blood glucose (PBG) (7.21 mmol/L), National institutes of health stroke scale (NIHSS) (18.5), Alberta Stroke Program Early CT Score (ASPECTS) (7.5), collateral score (3.5), infarct core volume (19.25 mL), and hypoperfusion volume (180.65 mL). The AUCs for the five models were C-I Model (0.851, 95% CI: 0.785–0.917), C-NC Model (0.865, 95% CI: 0.804–0.926), C-NP Model (0.861, 95% CI: 0.798–0.923), C Model (0.713, 95% CI: 0.626–0.801) and I Model (0.772, 95% CI: 0.688–0.855). The accuracy of the optimal C-NP Model was 0.771, and the mean AUC from 5-fold cross-validation was 0.828.
Conclusion:
The combined model revealed strong predictive power regarding outcomes following successful mechanical thrombectomy. Furthermore, the C-NP Model holds greater practical utility, as it enables clinicians to quickly assess the prognosis after a successful intervention.
Introduction
Stroke is characterized by high incidence, mortality, and disability rates, imposing a substantial economic burden, and represents the leading cause of death and disability among Chinese adults (1, 2). Acute ischemic stroke (AIS), the most common type, is attributable to large vessel occlusion in 35–40% of cases, which is responsible for 95.6% of all stroke-related deaths (3, 4). Mechanical thrombectomy (MT) has been established as the most effective treatment for large vessel occlusive stroke (LVOS) (5–11). However, despite achieving high recanalization rates of 71–95%, approximately 45–54% of patients experience unfavorable clinical outcome, resulting in poor clinical outcomes (12–16). Unfavorable clinical outcome can be attributed to a variety of factors, such as patient age, preoperative blood glucose (BPG), National institutes of health stroke scale (NIHSS) score, preoperative imaging characteristics, the number of thrombectomy passes, and postoperative management strategies (17, 18). This time sensitivity is underscored by evidence showing that a mere one-hour delay in achieving recanalization reduces the chance of a good clinical prognosis by approximately 38% (19). Hence, it is crucial to rapidly provide a preliminary assessment of the likelihood of a favorable postoperative outcome based on limited preoperative data. This assessment is instrumental in facilitating rapid consensus between clinicians and patients’ families regarding the surgical decision.
Although both multimodal CT and MRI are used for pre-MT assessment (20, 21), CT is often more practical and widely accessible than MRI in the acute stroke setting, given the latter’s relative contraindications and limited availability in many institutions. Multimodal CT primarily includes non-contrast CT (NCCT), CT angiography (CTA), and CT perfusion (CTP). NCCT is primarily used to assess for the presence of intracranial hemorrhage, as well as the location and extent of infarction. CTA is employed to determine the site of vascular occlusion and evaluate collateral compensation. CTP provides information on the volume of the infarct core and the ischemic penumbra. The acquisition of complete multimodal CT, especially CTP, is often challenging in severe stroke due to poor patient cooperation. Moreover, performing CTP on AIS patients without large vessel occlusion, who will not undergo MT, results in unnecessary radiation exposure. To address this, our study aims to develop preoperative models combining different clinical and imaging variables. We seek to explore their predictive ability for the prognosis of successful mechanical thrombectomy, ultimately providing a basis for clinicians and patients to decide whether to undergo intervention.
Materials and methods
Study strategy and data collection
A consecutive series of patients with acute anterior circulation occlusion who underwent MT and achieved successful recanalization at Sanming First Hospital Affiliated to Fujian Medical University between January 2022 and October 2024 were retrospectively enrolled. All patients underwent preoperative multimodal CT imaging. Prior to the examination, the purpose and procedures of the study were explained to the patients and/or their families, and written informed consent was obtained from all participants. This study was reviewed and approved by the Ethics Committee of Sanming First Hospital Affiliated to Fujian Medical University.
The inclusion criteria were as follows: (1) age ≥18 years; (2) time from onset to treatment ≤ 24 h; (3) a pre-stroke modified Rankin Scale (mRS) score of 0 (22); (4) CTA-confirmed unilateral acute anterior circulation large vessel occlusion, including the internal carotid artery (ICA), the A1 segment of the anterior cerebral artery (ACA), and the M1-M3 segments of the middle cerebral artery (MCA); (5) successful post-procedural recanalization (mTICI ≥ 2b) (23, 24). Patients were excluded for any of the following reasons: (1) a history of multiple strokes; (2) presence of posterior circulation occlusion; (3) unsuccessful recanalization (mTICI ≤ 2a); (4) patient refusal or lack of recommendation for thrombectomy; (5) poor image quality (Figure 1).
Figure 1
Research method
CT protocol
Imaging was acquired on the Revolution CT system (GE Healthcare, United States) utilizing a low-dose protocol with adaptive statistical iterative reconstruction (level = 50%). The imaging protocol was as follows and was completed in one continuous session lasting approximately 10 min: head NCCT, followed by head-and-neck CTA, multiphase CTA (mCTA) of the intracranial vasculature, and finally, whole-brain CTP. The detailed scan parameters are summarized in Table 1.
Table 1
| Parameter | NCCT | Head and neck CTA/mCTA | CTP |
|---|---|---|---|
| Scan mode | Axial | Helical | Axial (Cine) |
| Detector coverage (mm) | 160 | – | 160 |
| Rotation time (s) | 1 | 0.5 | 0.5 |
| Tube voltage (kV) | 120 | 100 | 80 |
| Tube current (mA) | Automated (200–300) | Automated (80–350) | Fixed (130) |
| Pitch | – | 0.992:1 | – |
| Slice thickness (mm) | 1.25 | 1.25 | 5.0 |
| Contrast volume (mL) | – | 50 | 40 |
| Saline flush (mL) | – | 30 | 30 |
| Injection rate (mL/s) | – | 5 | 5 |
| Acquisition timing | – | Arterial phase from aortic arch to vertex; mCTA: 2 additional scans at 8 s intervals at skull base-vertex. | 12 scans @ 2 s interval, then 8 scans @ 3 s interval, post 5–6 min delay. |
Multimodal CT scan protocols.
mCTA, multiphase CTA.
Imaging post-processing
The NCCT, CTA, and mCTA images were post-processed on an Advantage Workstation 4.7 (GE Healthcare) utilizing the FastStroke application. This software was used to create axial maximum intensity projection (MIP) images with a 20-mm slab thickness, yielding separate arterial, venous, and late venous phase images of the cerebral vasculature and their corresponding color-fused MIP maps, where the three phases were assigned red, green, and blue, respectively.
The raw CTP images were analyzed by the Computer-Aided Analysis System for Cerebral CTP Imaging (Shanghai United Imaging Intelligence Co., Ltd.). The system automatically generated perfusion parameter maps. If the time-density curve was aberrant, the arterial input (M1 segment of the unaffected middle cerebral artery) and venous outflow (Torcular Herophili) were manually redefined before the data were reprocessed.
Data collection
Patients’ clinical data were recorded, including sex, age, and history of hypertension, diabetes mellitus, coronary heart disease, and atrial fibrillation. PBG levels and the NIHSS scores were also documented. The Alberta stroke program early CT score (ASPECTS) was determined from the head NCCT images. The length of the occluded vessel and the collateral score were obtained from the head-and-neck CTA and mCTA datasets. Based on CTP, the following volumetric parameters were derived: hypoperfusion volume (Tmax > 6 s), infarct core volume (CBF < 30%), and the ischemic penumbra volume (defined as the mismatch between the two).
Two radiologists with more than a decade of experience independently performed all quantitative and qualitative image analyses under blinded conditions, with interobserver consistency assessed by the intraclass correlation coefficient (ICC). The mean values of continuous measurements from both readers were adopted for further statistical analysis. In case of disagreements on ordinal ratings, a final consensus was reached through deliberation.
Follow-up and outcomes
The 90-day modified Rankin Scale (90D-mRS) score was assessed by a neurologist through either outpatient visits or telephone interviews. Based on the 90D-mRS scores, patients were stratified into two groups: a favorable clinical outcome group (scores 0–3) and an unfavorable clinical outcome group (scores 4–6).
Statistical analysis
We performed statistical analyses with R software (v4.5.1) and IBM SPSS Statistics 25, setting the significance level at p < 0.05. For univariate comparisons, categorical variables were analyzed using Pearson’s chi-square or Fisher’s exact test. Continuous variables were first tested for normality with the Shapiro–Wilk test; those normally distributed were compared with an independent t-test, and non-parametric data were compared with the Mann–Whitney U test. Optimal cut-off values were determined for the statistically significant variables. These variables were then included in a multivariable logistic regression analysis to construct predictive models. The predictive performance of the models was compared using receiver operating characteristic (ROC) curve analysis by examining the area under the curve (AUC), sensitivity, and specificity. Using the Delong test for model comparison and selection, a nomogram of the optimal model was developed. Model performance was subsequently assessed by means of a confusion matrix and 5-fold cross-validation.
Results
Baseline data analysis
Among the 131 enrolled patients, 77 (58.78%) were classified into the favorable clinical outcome group and 54 (41.22%) into the unfavorable clinical outcome group. Among them, severe postoperative intracranial hemorrhage occurred in 5 patients, 3 of whom died. Overall, 14 patients died (90D-mRS 6), accounting for 10.7% of the total and 25.9% of the unfavorable clinical outcome group. The ICC values for length of thrombus, ASPECTS, and collateral score were 0.998 (95%CI: 0.997–0.998), 0.907 (95%CI: 0.871–0.933), 0.877 (95%CI: 0.830–0.911), respectively. These ICC values demonstrate substantial interobserver agreement for all three imaging variables, supporting the robustness of our image-based predictors. Compared with patients in the favorable clinical outcome group, those in the unfavorable clinical outcome group were older (p = 0.001), had higher PBG and NIHSS scores (p = 0.028 and p = 0.007, respectively), lower ASPECTS and collateral scores (both p < 0.001), and larger infarct core volumes and hypoperfusion volumes (p < 0.001 and p = 0.029, respectively). The optimal cutoff values for these parameters were 59.5 years (age), 7.21 mmol/L (PBG), 18.5 (NIHSS), 7.5 (ASPECTS), 3.5 (collateral score), 19.25 mL (infarct core volume), and 180.65 mL (hypoperfusion volume), as presented in Tables 2, 3.
Table 2
| Variate | Favorable (n = 77) (%) | Unfavorable (n = 54) (%) | X2/t/Z | P-value |
|---|---|---|---|---|
| Male | 46(59.7) | 36(66.7) | 0.650 | 0.420 |
| Age (y) | 64.3 ± 10.8 | 70.7 ± 9.7 | 0.416 | 0.001 |
| Hypertension | 43(55.8) | 32(59.3) | 0.151 | 0.697 |
| Diabetes | 14(18.2) | 16(29.6) | 2.356 | 0.125 |
| AF | 26(33.8) | 22(40.7) | 0.665 | 0.415 |
| PBG (mmol/L) | 6.35(5.72, 7.27) | 7.24(5.94, 9.87) | −2.190 | 0.028 |
| NIHSS | 15 ± 6 | 18 ± 6 | 1.628 | 0.007 |
| Length of thrombus (mm) | 17(12, 28) | 19.5(13, 29.5) | −1.383 | 0.167 |
| ASPECTS | 8(7, 9) | 6(5, 8) | −4.725 | <0.001 |
| Collateral score | 4(3, 4) | 3(2, 3) | −4.916 | <0.001 |
| Infarct core volume (ml) | 3.9(0.7, 15.0) | 26.3(8.9, 75.8) | −5.286 | <0.001 |
| Hypoperfusion volume (ml) | 138.4(92.0, 174.1) | 156.6(93.4, 245.2) | −2.179 | 0.029 |
| Ischemic penumbra volume (ml) | 119.8(83.0, 159.4) | 108.8(73.7, 141.8) | −0.650 | 0.516 |
Univariate analysis of all data.
AF, atrial fibrillation; PBG, preoperative blood glucose; NIHSS, national institutes of health stroke scale; ASPECTS, Alberta stroke program early CT score; X2/t/Z, statistical value.
Table 3
| Variate | Sensitivity | Specificity | Maximum Youden index | OCV |
|---|---|---|---|---|
| Age | 0.870 | 0.390 | 0.260 | 59.5 |
| PBG | 0.519 | 0.753 | 0.272 | 7.21 |
| NIHSS | 0.593 | 0.688 | 0.281 | 18.5 |
| ASPECTS | 0.688 | 0.722 | 0.411 | 7.5 |
| Collateral score | 0.597 | 0.815 | 0.412 | 3.5 |
| Infarct core volume | 0.648 | 0.818 | 0.466 | 19.25 |
| Hypoperfusion volume | 0.463 | 0.818 | 0.281 | 180.65 |
The optimal cut-off value of the effective variable.
PBG, preoperative blood glucose; NIHSS, national institutes of health stroke scale; ASPECTS, Alberta stroke program early CT score; OCV, optimal cut-off values.
Model building
Variables with p < 0.05 in the baseline data analysis were included in a multivariable logistic regression analysis to construct five distinct models: a clinical-imaging model (C-I Model), a clinical-non-collateral score model (C-NC Model), a clinical-non-perfusion model (C-NP Model), a comprehensive clinical model (C Model), and a comprehensive imaging model (I Model), as detailed in Table 4. Receiver operating characteristic (ROC) curves were plotted for each model. The area under the curve (AUC), sensitivity, and specificity were calculated based on these ROC curves, with the results presented in Table 5 and Figure 2.
Table 4
| Variate | C-I model | C-NC model | C-NP model | ||||||
|---|---|---|---|---|---|---|---|---|---|
| β | OR (95%CI) | P | β | OR (95%CI) | P | β | OR (95%CI) | p value | |
| Age | 0.108 | 1.114 (1.053, 1.177) | <0.001 | 0.110 | 1.116 (1.055, 1.181) | <0.001 | 0.109 | 1.116 (1.057, 1.178) | <0.001 |
| PBG (mmol/L) | 0.119 | 1.127 (1.001, 1.268) | 0.048 | 0.124 | 1.131 (1.006, 1.273) | 0.039 | 0.120 | 1.128 (1.003, 1.268) | 0.044 |
| NIHSS | 0.024 | 1.026 (0.952, 1.105) | 0.505 | 0.025 | 1.026 (0.952, 1.104) | 0.506 | 0.030 | 1.030 (0.956, 1.110) | 0.432 |
| ASPECTS | −0.436 | 0.646 (0.465, 0.898) | 0.009 | −0.502 | 0.606 (0.443, 0.828) | 0.002 | −0.504 | 0.604 (0.441, 0.826) | 0.002 |
| Collateral score | −0.403 | 0.668 (0.313, 1.426) | 0.298 | – | – | – | −0.696 | 0.499 (0.251,0.989) | 0.047 |
| Infarct core volume | 0.016 | 1.016 (0.996, 1.036) | 0.119 | 0.019 | 1.019 (1.001, 1.038) | 0.043 | – | – | – |
| Hypoperfusion volume | −0.001 | 0.999 (0.993, 1.005) | 0.755 | −0.001 | 0.999 (0.993, 1.005) | 0.726 | – | – | – |
| Variate | C model | I model | ||||
|---|---|---|---|---|---|---|
| β | OR (95%CI) | P | β | OR (95%CI) | P | |
| Age | 0.065 | 1.068 (1.026, 1.111) | 0.001 | – | – | – |
| PBG (mmol/L) | 0.100 | 1.105 (0.991, 1.231) | 0.071 | – | – | – |
| NIHSS | 0.073 | 1.076 (1.009, 1.148) | 0.025 | – | – | – |
| ASPECTS | – | – | – | −0.234 | 0.792 (0.608, 1.031) | 0.083 |
| Collateral score | – | – | – | −0.542 | 0.581 (0.298,1.134) | 0.111 |
| Infarct core volume | – | – | – | 0.021 | 1.021 (1.002, 1.041) | 0.030 |
| Hypoperfusion volume | – | – | – | −0.001 | 0.999 (0.993, 1.005) | 0.635 |
Multivariate logistic regression analysis of five models.
PBG, preoperative blood glucose; NIHSS, national institutes of health stroke scale; ASPECTS, Alberta stroke program early CT score.
Table 5
| Model | AUC | 95%CI | Sensitivity | Specificity |
|---|---|---|---|---|
| C-I | 0.851 | 0.785–0.917 | 0.944 | 0.675 |
| C-NC | 0.865 | 0.804–0.926 | 0.796 | 0.818 |
| C-NP | 0.861 | 0.798–0.923 | 0.796 | 0.792 |
| C | 0.713 | 0.626–0.801 | 0.722 | 0.649 |
| I | 0.772 | 0.688–0.855 | 0.648 | 0.818 |
Prediction efficiency of the five models.
AUC, area under curve.
Figure 2
Model evaluation
A comparison of the AUC values among the five models reveals that the combined models outperform the individual models. The C-NC Model and the C-NP Model achieved the highest AUCs and were therefore selected for further comparison. Delong’s test showed no statistically significant difference in the predictive performance between the C-NC Models and C-NP Models (Z = 0.334, p = 0.739). In clinical practice, however, the C-NP model streamlines the workflow and provides essential cerebrovascular information, making it more suitable for interventional intervention assessment. Based on the above findings, we constructed a nomogram for the C-NP model and presented its corresponding confusion matrix (Figures 3, 4). The accuracy of the optimal C-NP Model was 0.771, and the sensitivity and specificity were 0.722 and 0.805, respectively. We employed 5-fold cross-validation to reassess the performance of our model (Figure 5). This process yielded a set of five performance metrics, from which the mean and standard deviation were calculated. The mean AUC from 5-fold cross-validation was 0.828. Given the limited number of outcome events (n = 54), our models face a potential risk of overfitting. The moderate variability in cross-validated AUCs (SD 0.065) suggests some model instability. Therefore, our findings should be considered hypothesis-generating, and external validation in larger independent cohorts is necessary before clinical application. We demonstrate the application of our nomogram for personalized risk prediction with case examples (Figures 6, 7).
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Discussion
Endovascular therapy has made remarkable progress, with continual innovation in devices leading to a significant increase in recanalization rates for LVOS patients. However, favorable outcomes remain elusive for about half of all treated patients, and a substantial number suffer from neurological deterioration or mortality (12, 25). By retrospectively collecting preoperative patient data, we developed and compared five models to evaluate their predictive value for prognosis after successful recanalization following MT, aiming to provide evidence for initial decision-making during preoperative discussions. Our results demonstrated that the combined model offered the best performance in predicting post-MT prognosis, thereby helping to avoid unnecessary surgical interventions.
In our study, the incidence of poor outcomes after MT in patients with anterior circulation LVOS was 41.22%, which is consistent with previous reports. Among them, the death toll accounted for 10.7% of the total and 25.9% of the unfavorable clinical outcome group. The predictors of death may differ from those of severe disability (90D-mRS 4–5). For example, while large hypoperfusion volume and older age might predict both outcomes, other factors such as post-recanalization hemorrhage, malignant cerebral edema, or systemic complications could be more specific for mortality (17, 18). Future research should more carefully distinguish between mortality and disability to better elucidate prognostic factors and guide clinical decision-making.
While most studies consider a 90D-mRS score of 0–2 as indicative of a favorable outcome in patients with acute ischemic stroke (AIS), some research has shown that AIS patients with 90D-mRS scores of 2 and 3 exhibit similar 7-year survival rates (26). Additionally, data suggest that AIS patients with 90D-mRS scores of 2 and 3 have comparable health-related quality of life scores (27). Therefore, in our study, categorizing a 90D-mRS score of 0–3 as representing a favorable outcome appears more appropriate for patients with large vessel occlusion. This is beneficial for preoperative risk–benefit assessments with patients and families who may consider moderate disability an acceptable outcome.
In the combined model of this study, age and PBG were identified as independent risk factors for futile recanalization following MT. Within the HIAT-2 score, an age greater than 59 years is associated with an increased risk of unfavorable outcomes (28). A study focusing on acute LVOS in the Chinese population indicated that an age younger than 66 years is associated with favorable prognosis (16). In our study, the optimal cutoff value for age was determined to be 59.5 years, which aligns with the aforementioned findings. Hyperglycemia increases lactate production by cerebral cells in the ischemic penumbra, leading to acidosis that exacerbates brain injury and expands the infarct volume. Furthermore, it impairs cerebral autoregulation, predisposing patients to reperfusion injury, and increases insulin resistance, thereby intensifying inflammatory and oxidative stress responses. Consequently, hyperglycemia serves as an independent risk factor for poor prognosis after MT (29, 30). Multiple studies on outcomes following successful thrombectomy for anterior circulation strokes have consistently shown that elevated PBG is associated with unfavorable outcomes (31, 32). Goyal’s study demonstrated that admission hyperglycemia was associated with a higher likelihood of 3-month mortality and lower odds of 3-month functional improvement in terms of shift in mRS score in ELVO patients treated with MT (33). The study by Filipov demonstrated that hyperglycemia on admission, rather than a history of diabetes, is an independent predictor of poor prognosis after MT in patients with anterior circulation large vessel occlusion (34). Our findings are consistent with this result. Therefore, patients with advanced age and elevated PBG require closer attention and stricter intraoperative as well as postoperative management.
The NIHSS score reflects the severity of stroke symptoms and serves as an important indicator in various prognostic prediction models (25, 28, 35). Patients who develop malignant middle cerebral artery infarction after intervention tend to have higher preoperative NIHSS scores (36). In the univariate analysis of this study, the NIHSS score was a significant assessment indicator and remained an important predictor in C Model. However, it showed no statistically significant difference in the combined models. We speculate that it may correlate with imaging indicators, which warrants further investigation with larger sample sizes.
The ASPECTS is a semi-quantitative method for assessing the volume of cerebral infarction. Brain tissue appearing as hypodense areas on NCCT is considered to represent irreversible ischemic injury (37, 38). Research by Haussen et al. (39) demonstrated that the NCCT-based ASPECTS is inversely correlated with the infarct core volume on CTP. Furthermore, studies have shown that an ASPECTS ≥7 corresponds to an infarct core volume of <70 mL, while an ASPECTS <4 corresponds to a volume >100 mL, indicating that ASPECTS can quantitatively reflect the size of infarct core (40). Raza et al. (41) replaced the NCCT-based ASPECTS in the PRE model—which originally comprised age, NIHSS, and NCCT-based ASPECTS—with infarct core volume and reevaluated the prognosis of AIS patients. They found that this substitution did not improve prognostic performance, indicating that ASPECTS and infarct core volume have comparable prognostic value for patient outcomes (41). In our study, NCCT-based ASPECTS showed statistical significance in combined models, with an optimal cutoff value of 7.5, suggesting that an ASPECTS ≥8 points is associated with favorable prognosis. These findings are generally consistent with the aforementioned research.
Multiple studies have indicated that infarct core volume is more dependent on collateral status, with accelerated progression of cerebral infarction occurring under poor collateral conditions (42–44). Therefore, collateral status serves as a critical factor for revascularization in patients beyond the standard time window. In recent years, collateral status assessment methods based on CTA or DSA have been widely used in patients with AIS. Among these, the Menon scoring system based on mCTA has been extensively studied and validated (45, 46). Studies have shown that the Menon score outperforms single-phase CTA and is comparable to simplified CTP in predicting clinical outcomes, while avoiding complex post-processing (47). Therefore, this study adopted the Menon scoring method to evaluate the collateral status of patients. Research has indicated that poor collateral status is one of the reasons why patients with LVOS do not undergo MT and is the baseline imaging feature most strongly associated with poor outcomes in wake-up stroke patients (48). In our study, the C-NP Model demonstrated that collateral score is an independent risk factor for predicting ineffective recanalization after MT, which is consistent with previous findings. It exhibits predictive capability comparable to the C-NC Model and is more suitable for patients unable to cooperate with imaging examinations.
Raza and Rangaraju (49) further demonstrated that infarct core volume, rather than ischemic penumbra volume, serves as a significant predictor of outcome in patients with LVOS. Infarct core volume is a key criterion for selecting patients for MT. For instance, the SWIFT PRIME trial excluded patients with an infarct core volume exceeding 50 mL; the DAWN trial required an infarct core volume no greater than 51 mL; and the DEFUSE 3 trial stipulated an infarct core volume below 70 mL (8, 10, 11). These criteria underscore that infarct core volume is the most critical variable among perfusion parameters. In the present study, infarct core volume emerged as an important parameter in the C-NC Model and was the sole variable in the I Model, indicating its essential role in predicting post-MT prognosis. Furthermore, the C-NC model achieved the highest AUC, highlighting the pivotal significance of infarct core volume in imaging assessment. These findings are consistent with previous research.
In a recent multinational survey, factors most strongly associated with the decision to perform MT included stroke severity (NIHSS score), patient age, guideline-based evidence level, pre-procedural brain imaging, and physician experience (50). Multiple studies have demonstrated that combined models integrating clinical and imaging variables can effectively predict post-MT outcomes (51–54). Consistent with previous findings, the combined models in the present study demonstrated the highest predictive performance for post-procedural prognosis. The HIAT-2 score, a classic predictor of poor outcome following anterior circulation intra-arterial therapy, incorporates age, pre-treatment glucose level, NIHSS score, and ASPECTS, achieving an AUC of 0.748 for predicting poor outcome (28). Another model based solely on imaging variables—comprising the clot burden score, multisegment clot burden, CBV-ASPECTS, and collateral score—yielded an AUC of 0.714 for predicting unfavorable outcomes post-MT (55). This indicates that combined models generally outperform models based solely on imaging data. Similarly, in the current study, the AUC of the combined models were also higher than those of the purely clinical or imaging-based models, which aligns with the findings from the aforementioned research.
Recent large-scale trials focusing on large infarct cores have increasingly favored simplified imaging protocols to enable rapid assessment (56, 57). Although MT provides clinical benefits for critically ill patients with large cerebral infarctions, it does not significantly reduce mortality (58). Our C-I model, which incorporates only NCCT examination, may therefore be better suited for the rapid evaluation of patients with large infarct cores. While the C-NC model demonstrated the highest predictive value, the success rate of CTP scan in critically ill patients is often limited. Moreover, for AIS patients without large vessel occlusion, CTP may unnecessarily increase radiation exposure. In contrast, the C-NP model allows rapid assessment of the presence of large vessel occlusion and collateral status without requiring CTP. This model simplifies the preoperative workflow, achieves a higher examination success rate, offers broader applicability, and reduces radiation exposure for patients. Therefore, we propose that the C-NP model is the most suitable for preoperative assessment, providing clinicians with a practical tool to identify eligible candidates for MT.
Limitations of this study include
(1) Stroke etiology data were not available in our study. Most patients came from rural or county-level areas with poor pre-admission healthcare access, and they were often unaware of their baseline vascular status. As a result, etiological classification could not be reliably determined within the short preoperative window, and the high rate of missing data precluded its inclusion. The absence of etiology may introduce unmeasured confounding and limit the generalizability of our models to populations with known or different etiological profiles. Future studies incorporating etiological stratification are needed to validate and refine our findings, which would enhance the model’s accuracy.
(2) Procedural variables (e.g., number of passes, device type, time to reperfusion, anesthesia type) and post-procedural management were not recorded or adjusted for. Because our model is intended for preoperative prediction, these factors are inherently unavailable at the time of prediction. However, their absence may affect model transportability, and future studies should integrate preoperative-available proxies or develop dynamic models that incorporate intra-procedural updates.
(3) The study population was selected based on corresponding diagnostic and therapeutic guidelines (59) and patient preference, making it difficult to avoid selection bias.
(4) Furthermore, the relatively small sample size necessitates further validation with larger cohorts. The present model was developed from a single-center cohort with a limited sample size. Although internal validation via cross-validation was performed to assess its performance, the risk of overfitting remains a concern. Its prospective and multi-center external validation is imperative before any widespread implementation can be contemplated. Our next step will involve conducting external validation across multiple stroke centers of different levels.
(5) It should be acknowledged that some studies have indicated that mTICI 2C/3 may confer better clinical outcomes than mTICI 2B (60). And advances in endovascular materials may facilitate more complete reperfusion in the future. Further research is needed to elucidate the effect of varying reperfusion grades on clinical prognosis.
(6) While our study used the Menon multiphase collateral score to assess arterial collateral status, we did not incorporate dedicated venous outflow models for comparison with other metrics. Recent literature has demonstrated that combined assessment of cortical and medullary venous outflow can improve the predictive performance of multiparametric models in stroke patients (61–63). Given that multiphase CTA data were available, future studies should explore whether integrating venous outflow assessment could provide additional prognostic value beyond arterial collateral grading alone. This remains an important direction for future research.
(7) Different CTP processing platforms employ proprietary algorithms for deconvolution, arterial input function selection, and baseline calibration. Consequently, even when applying the same relative cerebral blood flow (rCBF) < 30% threshold to define the ischemic core, absolute infarct volume measurements can differ substantially across platforms. This limits the direct comparability of our volumetric data with studies that used other software. Future studies should consider cross-platform validation or the use of standardized phantoms to harmonize measurements.
Conclusion
The combined models based on multimodal CT demonstrated a robust predictive ability for the prognosis of patients with acute anterior circulation occlusive stroke following MT. The C-NP model streamlines the examination process, enhances assessment speed, and reduces patient radiation exposure, thereby enabling clinicians to better assess prognosis after successful intervention and holding considerable promise for broader adoption in primary care settings.
Statements
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Ethics statement
The studies involving humans were approved by the Ethics Committee of Sanming First Hospital Affiliated to Fujian Medical University. 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.
Author contributions
GL: Conceptualization, Data curation, Formal analysis, Investigation, Writing – original draft. MD: Funding acquisition, Project administration, Resources, Software, Supervision, Writing – review & editing. LS: Data curation, Formal analysis, Methodology, Validation, Visualization, Writing – original draft. GC: Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Resources, Writing – review & editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. The study was sponsored by startup fund for scientific research, Fujian Medical University (Grant number: 2022QH1254).
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that Generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fneur.2026.1818377/full#supplementary-material
References
1.
WuSWuBLiuMChenZWangWet al. Stroke in China: advances and challenges in epidemiology, prevention, and management. Lancet Neurol. (2019) 18:394–405. doi: 10.1016/S1474-4422(18)30500-3,
2.
PeultierACPandyaASharmaRSeverensJLRedekopWK. Cost-effectiveness of mechanical Thrombectomy more than 6 hours after symptom onset among patients with acute ischemic stroke. JAMA Netw Open. (2020) 3:e2012476. doi: 10.1001/jamanetworkopen.2020.12476,
3.
MiaoZHuoXGaoFLiaoXWangCPengYet al. Endovascular therapy for acute ischemic stroke trial (EAST): study protocol for a prospective, multicentre control trial in China. Stroke Vasc Neurol. (2016) 1:44–51. doi: 10.1136/svn-2016-000022,
4.
MalhotraKGornbeinJSaverJL. Ischemic strokes due to large-vessel occlusions contribute disproportionately to stroke-related dependence and death: a review. Front Neurol. (2017) 8:51. doi: 10.3389/fneur.2017.00651
5.
JansenIGHMulderMJHLGoldhoornRB. Endovascular treatment for acute ischaemic stroke in routine clinical practice: prospective, observational cohort study (MR CLEAN registry). BMJ. (2018) 360:k949–9. doi: 10.1136/bmj.k949
6.
GoyalMDemchukAMMenonBKEesaMRempelJLThorntonJet al. Randomized assessment of rapid endovascular treatment of ischemic stroke. New Engl J Med. (2015) 372:1019–30. doi: 10.1056/NEJMoa1414905
7.
CampbellBCMitchellPJKleinigTJDeweyHMChurilovLYassiNet al. Endovascular therapy for ischemic stroke with perfusion-imaging selection. New Engl J Med. (2015) 372:1009–18. doi: 10.1056/NEJMoa1414792
8.
SaverJLGoyalMBonafeADienerHCLevyEIPereiraVMet al. Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke. New Engl J Med. (2015) 372:2285–95. doi: 10.1056/NEJMoa1415061,
9.
JovinTGChamorroACoboEde MiquelMAMolinaCARoviraAet al. Thrombectomy within 8 hours after symptom onset in ischemic stroke. New Engl J Med. (2015) 372:2296–306. doi: 10.1056/NEJMoa1503780,
10.
NogueiraRGJadhavAPHaussenDCBonafeABudzikRFBhuvaPet al. Thrombectomy 6 to 24 hours after stroke with a mismatch between deficit and infarct. New Engl J Med. (2018) 378:11–21. doi: 10.1056/NEJMoa1706442
11.
AlbersGWLansbergMGKempSTsaiJPLavoriPChristensenSet al. A multicenter randomized controlled trial of endovascular therapy following imaging evaluation for ischemic stroke (DEFUSE 3). Int J Stroke. (2017) 12:896–905. doi: 10.1177/1747493017701147
12.
GoyalMMenonBKZwamWHDippelDWMitchellPJDemchukAMet al. Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. (2016) 387:1723–31. doi: 10.1016/S0140-6736(16)00163-X,
13.
LapergueBBlancRGoryBLabreucheJDuhamelAMarnatGet al. Effect of endovascular contact aspiration vs. stent retriever on revascularization in patients with acute ischemic stroke and large vessel occlusion. The ASTER randomized clinical trial. J Vasc Surg. (2017) 66:1910. doi: 10.1001/jama.2017.9644
14.
FiehlerJThomallaGBernhardtMKniepHBerlisADornFet al. ERASER. Stroke. (2019) 50:1275–8. doi: 10.1161/STROKEAHA.119.024858
15.
KaesmacherJDobrockyTHeldnerMRBellwaldSMosimannPJMordasiniPet al. Systematic review and meta-analysis on outcome differences among patients with TICI2b versus TICI3 reperfusions: success revisited. J Neurol Neurosur Ps. (2018) 89:910–7. doi: 10.1136/jnnp-2017-317602,
16.
JiaBRenZMokinMBurginWSBauerCTFiehlerJet al. Current status of endovascular treatment for acute large vessel occlusion in China: a real-world Nationwide registry. Stroke. (2021) 52:1203–12. doi: 10.1161/STROKEAHA.120.031869,
17.
NieXLengXMiaoZFisherMLiuL. Clinically ineffective reperfusion after endovascular therapy in acute ischemic stroke. Stroke. (2022) 54:873–81. doi: 10.1161/STROKEAHA.122.038466,
18.
WuSYuanRWangYWeiCZhangSYangXet al. Early prediction of malignant brain edema after ischemic stroke. Stroke. (2018) 49:2918–27. doi: 10.1161/STROKEAHA.118.022001
19.
MenonBKAlmekhlafiMAPereiraVMGrallaJBonafeADavalosAet al. Optimal workflow and process-based performance measures for endovascular therapy in acute ischemic stroke: analysis of the solitaire FR thrombectomy for acute revascularization study. Stroke. (2014) 45:2024–9. doi: 10.1161/STROKEAHA.114.005050
20.
HuoXMaGZhangXPanYTongXSunDet al. Endovascular therapy in acute anterior circulation large vessel occlusive patients with a large infarct core (ANGEL-ASPECT): protocol of a multicentre randomised trial. Stroke Vasc Neurol. (2022) 8:169–74. doi: 10.1136/svn-2022-001865
21.
SarrajAHassanAEAbrahamMGOrtega-GutierrezSKasnerSEHussainMSet al. Trial of endovascular Thrombectomy for large ischemic strokes. N Engl J Med. (2023) 388:1259–71. doi: 10.1056/NEJMoa2214403
22.
BamfordJMSandercockPAWarlowCPSlatteryJ. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. (1989) 20:828. doi: 10.1161/01.STR.20.6.828,
23.
MokinMKhalessiAAMoccoJLanzinoGDumontTMHanelRAet al. Endovascular treatment of acute ischemic stroke: the end or just the beginning?Neurosurg Focus. (2014) 36:E5. doi: 10.3171/2013.10.FOCUS13374
24.
XuZGrewalSMofattehMDmytriwAAZhaoDChenBet al. Can ischemic stroke patients with mTICI of 2b achieve similar outcomes compared to those with complete recanalization following endovascular therapy?Front Neurol. (2024) 15:1486586. doi: 10.3389/fneur.2024.1486586
25.
LeeSHKimBJHanMKParkTHLeeKBLeeBCet al. Futile reperfusion and predicted therapeutic benefits after successful endovascular treatment according to initial stroke severity. BMC Neurol. (2019) 19:11. doi: 10.1186/s12883-019-1237-2,
26.
MagalhãesRAbreuPCorreiaMWhiteleyWSilvaMCSandercockP. Functional status three months after the first ischemic stroke is associated with long-term outcome: data from a community-based cohort. Cerebrovasc Dis. (2014) 38:46–54. doi: 10.1159/000364938
27.
RangarajuSHaussenDNogueiraRGNahabFFrankelM. Comparison of 3-month stroke disability and quality of life across modified Rankin scale categories. Interv Neurol. (2017) 6:36–41. doi: 10.1159/000452634,
28.
SarrajAAlbrightKBarretoADBoehmeAKSittonCWChoiJet al. Optimizing prediction scores for poor outcome after intra-arterial therapy in anterior circulation acute ischemic stroke. Stroke. (2013) 44:3324–30. doi: 10.1161/STROKEAHA.113.001050,
29.
LuGDRenZQZhangJXZuQQShiHB. Effects of diabetes mellitus and admission glucose in patients receiving mechanical Thrombectomy: a systematic review and Meta-analysis. Neurocrit Care. (2018) 29:426–34. doi: 10.1007/s12028-018-0562-4,
30.
Perez-VegaCDomingoRATripathiSRamos-FresnedoAKashyapSQuinones-HinojosaAet al. Influence of glucose levels on clinical outcome after mechanical thrombectomy for large-vessel occlusion: a systematic review and meta-analysis. J Neurointerv Surg. (2021) 14:neurintsurg-2021-017771. doi: 10.1136/neurintsurg-2021-017771
31.
WeylandCSMokliYVeyJAKieserMHerwehCSchönenbergerSet al. Predictors for failure of early neurological improvement after successful Thrombectomy in the anterior circulation. Stroke. (2021) 52:1291–8. doi: 10.1161/STROKEAHA.120.030519
32.
YedavalliVSSalimHAMusmarBAdeebNEssibayiMAElNaamaniKet al. Symptomatic intracerebral hemorrhage in proximal and distal medium middle cerebral artery occlusion patients treated with mechanical thrombectomy. J Neurointerv Surg. (2025) 17:803–10. doi: 10.1136/jnis-2024-021879,
33.
GoyalNTsivgoulisGPandhiADillardKKatsanosAHMagoufisGet al. Admission hyperglycemia and outcomes in large vessel occlusion strokes treated with mechanical thrombectomy. J Neurointerv Surg. (2017) 10:112–7. doi: 10.1136/neurintsurg-2017-012993
34.
FilipovAAndermannMLepskiGArévaloAHilgenfeldTSchönenbergerSet al. Admission glucose is a significant outcome predictor in anterior circulation stroke: approaching the sweet spot. Neurol Res Pract. (2025) 7:39. doi: 10.1186/s42466-025-00393-0
35.
RangarajuSAghaebrahimAStreibCSunCHRiboMMuchadaMet al. Pittsburgh response to endovascular therapy (PRE) score: optimizing patient selection for endovascular therapy for large vessel occlusion strokes. J Neurointerv Surg. (2015) 7:783–8. doi: 10.1136/neurintsurg-2014-011351,
36.
XuXZhangGJiaZZhaoLCaoYShiHet al. Predictors of malignant middle cerebral artery infarction after endovascular thrombectomy: results of DIRECT-MT trial. Eur Radiol. (2023) 33:135–43. doi: 10.1007/s00330-022-09013-w
37.
AlMajaliMDibasMGhannamMGalecio-CastilloMQudahAAKhasiyevFet al. Does the ischemic Core really matter? An updated systematic review and Meta-analysis of large Core trials after TESLA, TENSION, and LASTE stroke. Vasc Interv Neurol. (2024) 4. doi: 10.1161/SVIN.123.001243
38.
WangJQiuJWangY. Neurological functional Independence after endovascular Thrombectomy and different imaging modalities for large infarct Core assessment: a systematic review and meta-analysis. Clin Neuroradiol. (2023) 33:21–9. doi: 10.1007/s00062-022-01202-w
39.
HaussenDCDehkharghaniSRangarajuSRebelloLCBouslamaMGrossbergJAet al. Automated CT perfusion ischemic Core volume and noncontrast CT ASPECTS (Alberta stroke program early CT score): correlation and clinical outcome prediction in large vessel stroke. Stroke. (2016) 47:2318–22. doi: 10.1161/STROKEAHA.116.014117
40.
de Margerie-MellonCTurcGTisserandMNaggaraOCalvetDLegrandLet al. Can DWI-ASPECTS substitute for lesion volume in acute stroke?Stroke. (2013) 44:3565–7. doi: 10.1161/STROKEAHA.113.003047,
41.
RazaSABarreiraCMRodriguesGMFrankelMRHaussenDCNogueiraRGet al. Prognostic importance of CT ASPECTS and CT perfusion measures of infarction in anterior emergent large vessel occlusions. J Neurointerv Surg. (2019) 11:670–4. doi: 10.1136/neurintsurg-2018-014461,
42.
WinkelmeierLKniepHThomallaGBendszusMSubtilFBonekampSet al. Arterial collaterals and endovascular treatment effect in acute ischemic stroke with large infarct: a secondary analysis of the TENSION trial. Radiology. (2025) 314:e242401. doi: 10.1148/radiol.242401,
43.
KimHJRohHG. Imaging in acute anterior circulation ischemic stroke: current and future. Neurointervention. (2022) 17:2–17. doi: 10.5469/neuroint.2021.00465,
44.
Puhr-WesterheideDTiedtSRotkopfLTHerzbergMReidlerPFabritiusMPet al. Clinical and imaging parameters associated with Hyperacute infarction growth in large vessel occlusion stroke. Stroke. (2019) 50:2799–804. doi: 10.1161/STROKEAHA.119.025809
45.
MenonBKd'EsterreCDQaziEMAlmekhlafiMHahnLDemchukAMet al. Multiphase CT angiography: a new tool for the imaging triage of patients with acute ischemic stroke. Radiology. (2015) 275:510–20. doi: 10.1148/radiol.15142256
46.
SekerFPotreckAMöhlenbruchMBendszusMPhamM. Comparison of four different collateral scores in acute ischemic stroke by CT angiography. J Neurointerv Surg. (2016) 8:1116–8. doi: 10.1136/neurintsurg-2015-012101,
47.
LinYKangNKangJLvSWangJ. Predictive value of time-variant color-coded multiphase CT angiography (mCTA) regarding clinical outcome of acute ischemic stroke: in comparison with conventional mCTA and CT perfusion. Acta Radiol. (2022) 63:84–92. doi: 10.1177/0284185120981770,
48.
YuFBaiXShaAZhangMShanYGuoDet al. Multimodal CT imaging characteristics in predicting prognosis of wake-up stroke. Front Neurol. (2021) 12:702088. doi: 10.3389/fneur.2021.702088,
49.
RazaSARangarajuS. Prognostic scores for large vessel occlusion strokes. Neurology. (2021) 97:S79–90. doi: 10.1212/WNL.0000000000012797,
50.
SaposnikGMenonBKKashaniNWilsonATYoshimuraSCampbellBCVet al. Factors associated with the decision-making on endovascular thrombectomy for the management of acute ischemic stroke. Stroke. (2019) 50:2441–7. doi: 10.1161/STROKEAHA.119.025631
51.
EspinosaRMParrillaGManzano-FernándezSGarcía-VillalbaBZamarroJHernández-FernándezFet al. Combined multimodal computed tomography score correlates with futile recanalization after Thrombectomy in patients with acute stroke. Stroke. (2015) 46:2517–22. doi: 10.1161/STROKEAHA.114.008598
52.
LansbergMGMlynashMHamiltonSYeattsSDChristensenSKempSet al. Association of Thrombectomy with Stroke Outcomes among Patient Subgroups: secondary analyses of the DEFUSE 3 randomized clinical trial. JAMA Neurol. (2019) 76:447–53. doi: 10.1001/jamaneurol.2018.4587,
53.
SallustioFToschiNMascoloAPMarramaFMorosettiDda RosVet al. Selection of anterior circulation acute stroke patients for mechanical thrombectomy. J Neurol. (2019) 266:2620–8. doi: 10.1007/s00415-019-09454-2
54.
LyuJXiaoSMengZWuXChenWWangGet al. Collateral-Core ratio as a novel predictor of clinical outcomes in acute ischemic stroke. Transl Stroke Res. (2022) 14:73–82. doi: 10.1007/s12975-022-01066-9,
55.
ZhongWChenZYanSZhouYZhangRLuoZet al. Multi-mode imaging scale for endovascular therapy in patients with acute ischemic stroke (META) [J/OL]. Brain Sci. (2022) 12:871. doi: 10.3390/brainsci12070821
56.
CostalatVJovinTGAlbucherJFCognardCHenonHNouriNet al. Trial of Thrombectomy for stroke with a large infarct of unrestricted size. N Engl J Med. (2024) 390:1677–89. doi: 10.1056/NEJMoa2314063,
57.
YooAJZaidatOOShethSARaiATOrtega-GutierrezSGivenCAet al. Thrombectomy for stroke with large infarct on noncontrast CT: the TESLA randomized clinical trial. J Am Med Assoc. (2024). 332, 1355–1366. doi: 10.1001/jama.2024.13933
58.
HuoXSunDNguyenTNMaGPanYTongXet al. Endovascular therapy versus medical Management for Large Ischemic Infarct: 1-year outcomes of the ANGEL-ASPECT trial. Stroke. (2025) 56:2398–407. doi: 10.1161/STROKEAHA.124.050086
59.
Chinese Society of Neurology, Chinese Stroke Society, Neurovascular Intervention Group of Chinese Society of Neurology. Chinese guidelines for the early endovascular intervention in acute ischemic stroke 2022. Chin J Neurol. (2022) 55:565–80. doi: 10.3760/cma.j.cn113694-20220225-00137
60.
LeCouffeNEKappelhofMTreurnietKMLingsmaHFZhangGvan den WijngaardIRet al. 2B, 2C, or 3: what should be the angiographic target for endovascular treatment in ischemic stroke?Stroke. (2020) 51:1790–6. doi: 10.1161/STROKEAHA.119.028891
61.
KoneruMMeiJYLakhaniDASalimHAShahriariMDmytriwAAet al. Prolonged venous transit on perfusion imaging is associated with longer lengths of stay in acute large vessel occlusions. AJNR Am J Neuroradiol. (2025) 46:887–91. doi: 10.3174/ajnr.A8611,
62.
ParthasarathyRKateMRempelJLLiebeskindDSJeerakathilTButcherKSet al. Prognostic evaluation based on cortical vein score difference in stroke. Stroke. (2013) 44:2748–54. doi: 10.1161/STROKEAHA.113.001231,
63.
DrozdovAAAroraMLeon GuerreroCRSparksADReza TaheriM. Appearance of medullary and cortical veins on multiphase CT-angiography in patients with acute ischemic stroke. Clin Neurol Neurosur. (2022) 224:107523. doi: 10.1016/j.clineuro.2022.107523
Summary
Keywords
acute ischemic stroke, collateral score, large vessel occlusion, mechanical thrombectomy, multimodal CT
Citation
Luo G, Deng M, She L and Chen G (2026) Preoperative multimodal CT for selection of acute anterior circulation occlusion stroke patients for mechanical thrombectomy. Front. Neurol. 17:1818377. doi: 10.3389/fneur.2026.1818377
Received
26 February 2026
Revised
22 April 2026
Accepted
24 April 2026
Published
08 May 2026
Volume
17 - 2026
Edited by
Bryan Good, The University of Tennessee, Knoxville, United States
Reviewed by
Romil Singh, Mayo Clinic, United States
Aleksandr Drozdov, University of Vermont, United States
Updates
Copyright
© 2026 Luo, Deng, She and Chen.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Guiyun Luo, fjmulgy@163.com; Guiling Chen, fjmucgl@163.com
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.