Ischemic stroke due to basilar artery occlusion (BAO) is a catastrophic condition that is associated with high rates of disability and mortality (1). Whether endovascular treatment (EVT) improves the prognosis of these patients remains unclear. Recent clinical trials on BAO treatment–BASICS (Basilar Artery International Cooperation Study) and BEST trial (Basilar Artery Occlusion Endovascular Intervention vs. Standard Medical Treatment)–failed to show superiority of EVT over standard medical treatment (2, 3). However, the high crossover rate in the BEST trial and a large number of screened patients treated outside of the trials might have affected the neutral results. Given the better outcome after EVT in per-protocol-treated patients in the BEST trial and in patients with a higher disease severity in the BASICS trial, EVT may be somewhat beneficial for patients with BAO who are likely to have poor prognosis with standard medical treatment. A more detailed analysis of prognostic factors is needed to achieve good outcomes after EVT in patients with BAO stroke.

Infarct location may play a critical role in functional impairment (4–6), especially in posterior circulation stroke where eloquent neural pathways are densely packed together (7). The corticospinal tract (CST) is a major neural pathway that is associated with post-stroke motor outcomes (8). It travels through the cerebral crus and the anterior part of the pons that are supplied by perforators originating from the basilar trunk, which accordingly is vulnerable to BAO. Sparing the CST through reperfusion therapy was associated with improved outcomes in anterior circulation stroke, independent of overall salvage of ischemic tissues (9, 10). Despite the encouraging result, this association has not been well-investigated in posterior circulation stroke. This study aimed to evaluate the impact of salvaging the CST on functional outcomes and its related factors in patients with BAO treated with EVT.

Eighty-eight patients with BAO who were treated with EVT were enrolled in this study [mean age: 71.7 ± 9.7 years; 49 (55.7%) were men]. In 39 (44.3%) patients, post-interventional imaging revealed CST salvage. The clinical characteristics were not significantly different between the groups with CST- and CST+, except for the baseline NIHSS total and motor scores, stroke mechanism, and AF. The baseline NIHSS total and motor scores were lower in patients with CST- than with CST+ (total: 16.3 ± 8.1 vs. 20.5 ± 6.9, P = 0.010; motor: 7.4 ± 4.8 vs.9.5 ± 4.6, P = 0.034, respectively). Those with CST- were more associated with cardioembolic stroke, whereas those with CST+ were more associated with atherothrombotic stroke. Moreover, AF was more frequent in those with CST- than in those with CST+ (69.2 vs. 36.7%, P = 0.002; Table 1).

A comparison of imaging parameters, including baseline imaging scores, the occlusion type and site, is shown in Table 1. Compared with the groups with CST+, the groups with CST- had higher pc-ASPECTS [median (IQR), 9.0 (8.0–10.0) vs. 8.0 (6.0–9.0), P < 0.001], lower pc-CTA scores [median (IQR), 2.0 (1.0–3.0) vs. 3.0 (2.0–4.0), P = 0.011] and higher BATMAN scores [median (IQR), 6.0 (5.0–7.0) vs. 5.0 (3.0–7.0), P = 0.036]. Those with CST- showed higher rate of branching-site occlusion than with CST+ (87.2 vs. 65.3%, P = 0.019). Patients with CST- had higher rate of distal BAO (89.7 vs. 71.4%, P = 0.034), lower rates of mid-BA occlusion (15.4 vs. 59.2%, P < 0.001), proximal BA occlusion (7.7 vs. 30.6%, P = 0.008), and occlusion in the multiple segments of BA (12.8 vs. 57.1%, P < 0.001). Successful recanalization was achieved more frequently in patients with CST- than in those with CST+ (97.4 vs. 77.6%, P = 0.007). With respect to time variables, the rates of onset-to-puncture time and onset-to-recanalization time within 180 min were higher in the CST- group than in the CST+ group (48.7 vs. 12.2%, P < 0.001; 33.3 vs. 6.1%, P = 0.001, respectively; Table 1).

Thirty-nine (44.3%) patients achieved good functional outcomes at 3 months. Patients with CST- had a significantly higher rate of 3-month good functional outcomes than those with CST+ (79.5 vs. 16.3%, P < 0.001, Table 2). Moreover, good to moderate functional dependence and functional independence at 3 months were higher in the groups with CST salvage than in the CST involvement group (94.9 vs. 34.7%, P < 0.001 and 69.2 vs. 14.3%, P < 0.001, respectively). Those with CST- had lower NIHSS scores on the day of discharge or day 7 [median (IQR), 3.0 (1.0–5.0) vs. 11.0 (7.0–26.5), P < 0.001] and tended to have higher rates of early neurological improvement (79.5 vs. 42.9%, P = 0.001). Regarding safety outcomes including 3-month mortality, any hemorrhagic transformation, and sICH, patients with CST salvage had significantly lower rates of adverse outcomes than patients with CST involvement, except for sICH. Furthermore, those with CST- had a smaller final infarct volume than those with CST+ [median (IQR), 6.5 mL (0.6–14.3) vs. 17.3 (5.3–46.6), P = 0.001, Table 2]. The associations of CST salvage and 3-month good functional outcomes remained statistically significant after adjusting for pre-specified potential confounders {adjusted odds ratio (aOR), 18.52 [95% confidence interval (CI), 4.31–79.67], P < 0.001; Table 3}.

Various clinical and imaging parameters were analyzed to identify factors associated with CST salvage. Before performing multivariable logistic regression analysis, potential factors were selected through the binary logistic regression analyses (Table 4). Baseline NIHSS total score over motor score and pc-ASPECTS as a continuous rather than dichotomized variable were selected. For thrombus burden measurement, pc-CTA score as a continuous variable over BATMAN score was selected. Moreover, for time variable, onset-to-puncture time within 180 min was selected. The location and length of BAO were entered in multivariable models 1 and 2, respectively. In the subsequent multivariable analyses, AF [aOR 3.92 (95% CI, 1.18–13.00), P = 0.026 in model 1], location of occlusion [mid-BAO; aOR 0.21 (95% CI, 0.06–0.72), P = 0.013], length of occlusion [involved segment of BAO <2; aOR 4.77 (95% CI, 1.30–17.59) P = 0.019], and onset-to-puncture time ≤180 min [aOR 4.84 (95% CI, 1.13–20.75), P = 0.034 in model 1] showed independent associations with the CST salvage after EVT (Table 4).

Similar results were observed in the sensitivity analysis in patients whose CST lesions were assessed with MRI. The baseline characteristics of this subgroup were presented in Supplementary Table 1. Patients with CST- had better clinical outcomes in terms of 3-month good functional outcomes [aOR, 17.54 (95% CI, 4.48–68.71), P < 0.001] and other secondary outcomes than patients with CST+ (Supplementary Table 2). Location (mid-BAO) and length of occlusion (involved segment of BAO <2) were still independently associated with CST in this subgroup, whereas pc-ASPECTS and successful reperfusion (modified treatment in cerebral ischemia grade of 2b−3) were newly detected as independent predictors of CST salvage (Supplementary Table 3).

In the present study, the CST could be salvaged by EVT in 44.3% of the patients with acute BAO. Moreover, sparing the CST was associated with a clinical benefit, including good functional outcome and early neurological improvement. Although several scoring systems with pretreatment imaging that weighted higher scores in brainstem lesions compared to other lesion locations have been developed for outcome prediction in patients with acute BAO treated with EVT (7, 25), to the best of our knowledge, this is the first study to evaluate the direct relationships between CST salvage and predictors and clinical outcomes in these patients. Our findings suggested that a clinical parameter (AF) and clot characteristics (location and length of BAO at baseline CTA) may enable early prognostication of CST salvage after EVT without advanced imaging, and that rapid initiation of EVT may protect this eloquent tract.

The strong association between the CST salvage and good clinical outcomes in our study indicated that the eloquence of certain brain areas could have profound effects on outcome after EVT (10, 26) and it could be explained by the fact that the intact CST was a predictor of independent walking after stroke (27). We found that the extent of BAO was associated with CST salvage. Longer clots may obstruct more perforating branches of the BA and consequently have a greater chance of inducing extensive ischemic damage in the brainstem. Moreover, the most vulnerable segment of BA to CST involvement was the mid-BA, where penetrating pontine arteries that supply the CST are located. These characteristics of BAO were robust predictors for CST salvage confirmed by sensitivity analysis. AF is also associated with CST salvage. In contrast to those with intracranial atherosclerosis, patients with AF were less likely to have underlying severe stenosis in the target artery, which may require a longer thrombectomy procedure time and additional rescue treatments such as balloon angioplasty or stenting (28). Furthermore, early initiation of EVT was linked to a higher chance of CST salvage. Of 25 patients with an onset-to-puncture time of ≤180 min, 19 (76.0%) achieved CST salvage after EVT, which is in line with the results regarding anterior circulation large-vessel occlusion stroke (10). However, these data should be interpreted with caution, as large heterogeneity exists in the infarct growth rate among patients with BAO and slow progressors may benefit from EVT beyond this time window (29).

The strength of this study is the identification of the mechanism of good prognosis in patients with acute BAO treated with EVT. Predictors of CST salvage may maximize the therapeutic effect of EVT in this patient population, and in-hospital workflow management to facilitate the rapid initiation of EVT is warranted. However, our study has several limitations. First, we included patients with premorbid mRS scores ≤3; thus, the patients in our study cohort were older and had more AF than in other study cohorts (2, 3). The generalizability of our study results to different study cohorts should be validated. Second, measurement of CST salvage/involvement was performed by direct visual interpretation. Although it could be easily applied in clinical practice and showed excellent interobserver agreement, quantitative assessment of CST salvage could not be performed using this method. Third, CT scan was used for evaluating CST salvage/involvement in 18.2% of the study participants, which is less sensitive in detecting infarcted lesions compared to MRI. However, the proportions of CST involvement in groups measured by CT was no less than those in groups measured by MRI [75.0% (12/16) vs. 51.4% (37/72), P = 0.085, respectively]. Moreover, due to the small number of the patients whose CST lesions were assessed with CT, whether the associations between CST salvage and clinical outcomes and the predictors of CST salvage were consistent in this subgroup could not be reliably tested (Supplementary Tables 1, 4). Fifth, the effect of different lesion locations within the CST on clinical outcomes could not be analyzed. As the CST runs down to the lower brainstem, the compactness of the tract increases. Thus, lesions of similar size in the CST may lead to different outcomes depending on the lesion location (30). Sixth, bias may have been introduced from differences in initial stroke severity (NIHSS score, pc-ASPECTS, and thrombus burden) between the CST- and CST+ groups when investigating the relationship of CST salvage and outcome, although potential confounders were adjusted. Further robust studies are warranted to confirm our knowledge of the effects of CST salvage. Seventh, the study enrollment period of 9 years might affect the study results given the changes in practice with the advance of EVT technology. However, the major study outcomes among groups sorted by 3-year-interval admission years were not significantly different (Supplementary Table 5). Eighth, a small proportion (16.3%) of patients with CST+ showed a good functional outcome in this study, which could be achieved partly due to peri-infarct areas compensating for the CST damage and facilitating motor recovery (31). Therefore, prognostic evaluation based solely on CST involvement should be avoided. Finally, whether the therapeutic time window may vary according to the clot length and location could not be evaluated. Further studies are required to develop more elaborate prognostic models that integrate imaging and time variables to identify EVT candidates in patients with BAO.

In conclusion, the occlusion site and length of BA at baseline CTA may predict CST salvage after EVT, and the CST could be protected by the rapid initiation of EVT in patients with acute BAO. Given that CST salvage is strongly associated with good functional outcomes, it may be considered as one of the potential therapeutic goals in patients with acute BAO who have undergone EVT; further well-designed studies are warranted.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Kyungpook National University Hospital. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

D-SG established the study protocol, analyzed and interpreted the data, and drafted the manuscript. WCC, Y-WK, D-HK, and WS analyzed and interpreted the data and revised the manuscript for intellectual content. Y-HH established the study idea, analyzed and interpreted the data, and made revisions to the manuscript with intellectual input. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

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.