Association of Estimated Arterial Stiffness with Collateral Status and Functional Outcome after Endovascular Thrombectomy: A Secondary Analysis of the OPTIMAL-BP Trial

Article information

Neurointervention. 2026;21(3):173-184
Publication date (electronic) : 2026 October 8
doi : https://doi.org/10.5469/neuroint.2026.00822
1Department of Neurology, Yonsei University College of Medicine, Seoul, Korea
2Integrative Research Center for Cerebrovascular and Cardiovascular Diseases, Yonsei University College of Medicine, Seoul, Korea
3Department of Neurology, Kyungpook National University Chilgok Hospital, School of Medicine, Kyungpook National University, Daegu, Korea
4Department of Neurology, Keimyung University Dongsan Hospital, Keimyung University School of Medicine, Daegu, Korea
5Department of Neurology, Seoul Medical Center, Seoul, Korea
6Department of Neurology, Kyung Hee University Hospital at Gangdong, Seoul, Korea
7Department of Radiology, Yonsei University College of Medicine, Seoul, Korea
8Department of Neurology, Korea University Anam Hospital, Korea University College of Medicine, Seoul, Korea
9Department of Neurology, Chosun University School of Medicine, Gwangju, Korea
10Department of Neurology, Ewha Womans University Seoul Hospital, Ewha Womans University College of Medicine, Seoul, Korea
11Graduate Program in System Health Science and Engineering, Ewha Womans University, Seoul, Korea
12Department of Neurology, Ewha Womans University Mokdong Hospital, Ewha Womans University College of Medicine, Seoul, Korea
13Department of Neurology, National Health Insurance Service Ilsan Hospital, Goyang, Korea
14Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea
15Department of Neurology, Dong-A University Hospital, Busan, Korea
16Department of Neurology, Busan Paik Hospital, Inje University College of Medicine, Busan, Korea
17Department of Neurology, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Korea
18Department of Neurology, Pusan National University School of Medicine, Busan, Korea
19Department of Neurology, Gachon University Gil Medical Center, Incheon, Korea
20Department of Neurology, Yongin Severance Hospital, Yonsei University College of Medicine, Yongin, Korea
21Department of Neurology, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Korea
22Department of Neurology, Kyungpook National University Hospital, School of Medicine, Kyungpook National University, Daegu, Korea
23Department of Neurology, Korea University Guro Hospital, Korea University College of Medicine, Seoul, Korea
24Department of Neurology, Daejeon Eulji Medical Center, Eulji University School of Medicine, Daejeon, Korea
25Department of Health Promotion, Severance Hospital, Yonsei University College of Medicine, Seoul, Korea
26Cardiovascular Research Institute, Yonsei University College of Medicine, Seoul, Korea
27Department of Research Affairs, Biostatistics Collaboration Unit, Yonsei University College of Medicine, Seoul, Korea
28Department of Neurology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
29Department of Neurology, CHA Bundang Medical Center, CHA University School of Medicine, Seongnam, Korea
Correspondence to: Hyo Suk Nam, MD, PhD Department of Neurology, Yonsei University College of Medicine, 50-1 Yonsei-ro, Seodaemun-gu, Seoul 03722, Korea E-mail: hsnam@yuhs.ac
Received 2026 July 15; Revised 2026 September 21; Accepted 2026 September 23.

Abstract

Purpose

Estimated pulse wave velocity (ePWV) is a surrogate of arterial stiffness that may influence stroke outcomes and potentially affect collateral circulation. We examined the association of ePWV with collateral status and 3-month functional outcome after successful endovascular thrombectomy and explored the contribution of collateral status to this relationship across blood pressure (BP) management strategies.

Materials and Methods

In this secondary analysis of the Outcome in Patients Treated With Intra-Arterial Thrombectomy–Optimal Blood Pressure Control trial, 285 patients were analyzed. Collateral status was assessed using the Tan scale (poor, 0–1; good, 2–3). Poor functional outcome was defined as 3-month modified Rankin Scale score of 3–6. Multivariable logistic and mediation analyses were performed to evaluate the association between ePWV and functional outcomes and to explore direct and indirect pathways.

Results

A total of 285 patients were included (mean age, 73.3±11.3 years; 40.4% women). Each 1 m/s increase in ePWV was associated with poor collaterals (adjusted odds ratio [OR] 1.276, 95% confidence interval [CI] 1.083–1.503) and poor functional outcome (adjusted OR 1.338, 95% CI 1.121–1.597). In mediation analyses, collateral status accounted for only a small proportion of the association between ePWV and poor functional outcome (indirect effect: adjusted β 0.022, 95% CI 0.001–0.053), whereas the direct effect remained significant (adjusted β 0.237, 95% CI 0.115–0.359). Neither the direct effect of ePWV on functional outcome nor the indirect effect through collateral status differed according to BP management strategy.

Conclusion

Higher ePWV was associated with poor collateral status and unfavorable functional outcome after successful reperfusion. Collateral status accounted for only a small proportion of the observed association between ePWV and functional outcome.

INTRODUCTION

Acute ischemic stroke due to large vessel occlusion (LVO) remains a major cause of disability and death worldwide, even in the era of endovascular thrombectomy (EVT) [1]. Among the determinants of outcome after EVT, collateral circulation is one of the most critical factors, as it preserves penumbral tissue, limits infarct growth, and extends the therapeutic window [2,3]. Patients with robust collaterals generally experience slower infarct progression and better functional recovery [4], whereas poor collateral status is associated with rapid tissue injury and unfavorable outcomes [3]. Despite its importance, collateral function is not solely determined by vascular anatomy but is also influenced by systemic physiological conditions, suggesting that underlying vascular properties may modulate collateral efficiency and clinical outcomes [5-7].

The estimated pulse wave velocity (ePWV), a non-invasive surrogate marker of arterial stiffness, has gained attention as a potential biomarker in cerebrovascular disease due to its ease of calculation using only age and blood pressure (BP) [8]. Increased arterial stiffness could impair cerebral autoregulation [9] and influence the development and maintenance of collateral flow [10]. In a previous secondary analysis of the Outcome in Patients Treated With Intra-Arterial Thrombectomy–Optimal Blood Pressure Control (OPTIMAL-BP) trial, higher ePWV was independently associated with poorer functional outcomes after EVT [11]. That study also demonstrated a crude association between higher ePWV and poorer collateral status, suggesting that collateral circulation may represent one potential pathway linking arterial stiffness to clinical outcomes. However, the association between ePWV and collateral status has not been evaluated using multivariable and mediation analyses, and the extent to which collateral status contributes to the relationship between ePWV and functional outcome remains unknown.

Building on our previous analysis of the OPTIMAL-BP cohort, we aimed to determine whether the observed association between ePWV and collateral status persists after multivariable adjustment and to quantify the contribution of collateral status to the relationship between ePWV and functional outcome. In addition, we assessed whether these associations differed according to post-procedural BP management strategies.

MATERIALS AND METHODS

Study Design

This post hoc secondary analysis used the data from the OPTIMAL-BP trial, a multicenter, randomized, open-label study with blinded outcome assessment, conducted in 19 stroke centers across South Korea from June 2020 to November 2022. The OPTIMAL-BP trial enrolled patients with acute ischemic stroke caused by LVO who achieved successful reperfusion (modified Thrombolysis in Cerebral Infarction grade 2b–3) after EVT. These patients presented with a systolic BP greater than 140 mmHg within 2 hours after EVT. Participants were randomized to either intensive systolic BP management (target <140 mmHg) or conventional systolic BP management (target 140–180 mmHg) during the initial 24-hour post-procedure period (ClinicalTrials.gov NCT04205305) [12]. For this secondary analysis, we specifically included patients with available collateral images and a 3-month modified Rankin Scale (mRS) score.

Estimated Pulse Wave Velocity Calculation

BP measurements obtained at trial enrollment (within 2 hours after EVT) were used to calculate ePWV. We employed a validated equation developed by the Reference Values for Arterial Stiffness Collaboration [13], as described by Greve et al. [14]. This model uses age and mean BP to approximate arterial stiffness. The formula applied was: ePWV = 9.587 − 0.402 × age + 4.560 × 10−3 × age2 − 2.621 × 10−5 × age2 × mean BP + 3.176 × 10−3 × age × mean BP − 1.832 × 10−2 × mean BP. Mean BP was calculated as: mean BP = diastolic BP + 0.4 × (systolic BP − diastolic BP) [14]. To promote broader adoption and practical application, we developed a user-friendly web-based ePWV calculator, publicly accessible at https://epwv.netlify.app.

Tan Scale Assessment and Classification of Collateral Status

Collateral circulation was assessed using the Tan collateral grading scale, a validated angiographic method for assessing leptomeningeal collateral flow in patients with acute ischemic stroke [15]. This scale categorizes collateral status based on the extent of contrast opacification in the territory distal to the occluded vessel, as visualized on baseline computed tomography angiography. Grades range from 0 (no collateral supply) to 3 (complete collateral filling of the affected territory). Collateral status was dichotomized into 2 groups: poor collaterals (Tan grade 0–1) and good collaterals (Tan grade 2–3). All imaging assessments were independently reviewed by 2 blinded neurointerventionists, with excellent inter-rater agreement (weighted κ=0.830) and all discrepancies resolved through discussion.

Assessment of Functional Outcome

The primary clinical outcome was the functional status at 3 months following stroke onset, assessed with the mRS. Structured telephone or in-person interviews using the Korean version of the mRS (https://k-stroke-academy.co.kr) were performed by locally certified medical staff blinded to treatment allocation. A good functional outcome was defined as an mRS score of 0–2, indicating independence or slight disability, whereas a poor functional outcome was defined as an mRS score of 3–6, indicating moderate to severe disability or death.

Statistical Analysis

To compare clinical characteristics between groups, continuous variables were analyzed using either the independent two-sample t-test or the Mann–Whitney U-test, depending on the distribution of the data. These variables were reported as mean±standard deviation or median with interquartile range, as appropriate. Categorical variables were compared using the chi-square test or Fisher’s exact test and presented as frequencies and percentages. Participants were stratified into high and low ePWV groups using a cutoff derived from the receiver operating characteristic curve for poor functional outcome. The cutoff was defined as the value that maximized the Youden index (sensitivity + specificity − 1).

The association between ePWV and collateral status was assessed using binary logistic regression models, both univariable and multivariable. Similarly, the relationship between ePWV and functional outcome was evaluated using logistic regression analyses. These models estimated the odds of poor collaterals or poor functional outcomes per 1 m/s increase in ePWV, as well as for the high ePWV group compared to the low ePWV group. Five models were used to control for potential confounding: Model 1 was unadjusted; Model 2 adjusted for age (<60 vs. ≥60 years) and sex; Model 3 further included baseline National Institutes of Health Stroke Scale (NIHSS 0–5, 6–15, ≥16), time from symptom onset to randomization, and BP management strategy (intensive vs. conventional); Model 4 additionally adjusted for Alberta Stroke Program Early CT Score (ASPECTS) and internal carotid artery occlusion; and Model 5 was further adjusted for smoking, atrial fibrillation, previous stroke, and stroke subtype. To avoid overfitting, effect modification (via interaction terms with ePWV) and subgroup analyses were restricted to Model 4. Subgroups included age, sex, stroke subtype, baseline NIHSS score, ASPECTS, onset‑to‑randomization time, occlusion site, systolic BP, diastolic BP, and BP management strategy.

To explore whether collateral status contributed to the association between ePWV and functional outcome, mediation analyses were performed using the PROCESS macro for SPSS. ePWV was modeled as a continuous variable, collateral status was represented by the Tan collateral grade (0–3), and functional outcome was represented by the original 3-month mRS score (0–6), modeled numerically without dichotomization. Indirect effects were estimated using 5,000 bootstrap resampling. The proportion of the total effect mediated was calculated as (indirect effect / total effect) × 100%. Potential effect modification by BP management strategy was evaluated by testing the interaction between ePWV and BP management for the direct pathway and by calculating the index of moderated mediation for the indirect pathway. All statistical analyses were conducted using IBM SPSS software, version 30 (IBM Co.). A two‑sided P-value <0.05 was considered statistically significant.

RESULTS

Baseline Characteristics

Of the 306 enrolled patients in the OPTIMAL-BP trial, 21 patients were excluded (1 withdrew consent, 3 were lost to follow-up, and 17 had missing or poor-quality collateral images). As a result, 285 patients were included in the final analysis and divided into high ePWV group (>12.6 m/s, n=184) and low ePWV group (≤12.6 m/s, n=101) (Fig. 1).

Fig. 1.

Flowchart of patient inclusion and exclusion. OPTIMAL-BP, Outcome in Patients Treated With Intra-Arterial Thrombectomy–Optimal Blood Pressure Control.

Patients in the high ePWV group were older (79.6±6.7 vs. 61.8±8.5 years) and more likely to be women (48.9% vs. 24.8%, P<0.001 for both) than those in the low ePWV group. They also had lower body mass index (23.5 [21.4, 25.4] vs. 24.5 [22.6, 26.1] kg/m2, P=0.010), a higher prevalence of hypertension (81.0% vs. 67.3%, P=0.010), atrial fibrillation (60.9% vs. 28.7%, P<0.001), and previous stroke (26.1% vs. 12.9%, P=0.009). In contrast, smoking was more common in the low ePWV group (41.6% vs. 12.0%, P<0.001). Regarding stroke subtype, cardioembolic stroke was more frequent in the high ePWV group (59.2% vs. 35.6%, P<0.001), while large artery atherosclerosis was more common in the low ePWV group. Baseline NIHSS score was higher in the high ePWV group (13.0 [8.0, 19.0] vs. 11.0 [7.3, 16.0], P=0.026) (Supplementary Table 1).

Baseline Characteristics according to Collateral Status and Functional Outcome

Based on Table 1, patients with poor collaterals and those with poor functional outcomes showed broadly consistent clinical and imaging profiles. Both groups were older and had more severe strokes, as reflected by higher baseline NIHSS scores, larger infarction volumes, and lower ASPECTS (all P<0.05). ePWV was significantly higher in the poor collateral group (13.7±2.1 vs. 12.9±2.1 m/s, P=0.004) and in the poor functional outcome group (13.6±2.1 vs. 12.6±2.1 m/s, P<0.001).

Baseline characteristics according to collateral status and functional outcome

Compared with patients with good outcomes, those with poor functional outcomes additionally had a higher prevalence of diabetes and previous stroke, longer procedure times, and were more frequently managed with intensive BP control. They also more often exhibited poor collaterals and internal carotid artery occlusions, along with fewer middle cerebral artery occlusions (all P<0.05).

Association between Estimated Pulse Wave Velocity and Poor Collaterals

Poor collaterals were observed in 93 patients (32.6%) and were more frequent in the high ePWV group than in the low ePWV group (37.5% vs. 23.8%, P=0.018). In continuous analyses, each 1 m/s increase in ePWV was associated with poor collaterals in both unadjusted (odds ratio [OR] 1.191, 95% confidence interval [CI] 1.054–1.347, P=0.005) and fully adjusted models (adjusted OR 1.276, 95% CI 1.083–1.503, P=0.004) (Table 2, Fig. 2A). Similar findings were observed in dichotomized analyses (Table 2, Fig. 3A). No significant subgroup interactions were identified (all P for interaction >0.05; Supplementary Table 2).

Association of ePWV with poor collaterals or poor functional outcome

Fig. 2.

Restricted cubic spline analyses of the association of ePWV with collateral status and functional outcome. Panel (A) shows the adjusted association between ePWV and poor collateral status. Panel (B) shows the adjusted association between ePWV and poor functional outcome at 3 months. Solid lines represent adjusted ORs, and shaded areas indicate 95% CIs. ePWV, estimated pulse wave velocity; OR, odds ratio; CI, confidence interval.

Fig. 3.

Distribution of collateral status and functional outcome by ePWV. Panel (A) illustrates the distribution of collateral grades according to the Tan scale (0–3) in patients stratified by ePWV. Poor collateral status was defined as Tan scale ≤1. The corresponding unadjusted and adjusted ORs for poor collateral status according to ePWV group are also presented. Panel (B) shows the distribution of 3-month functional outcomes based on the mRS (0–6), with poor functional outcome defined as mRS ≥3. The corresponding unadjusted and adjusted ORs for poor functional outcome according to ePWV group are also presented. ePWV, estimated pulse wave velocity; OR, odds ratio; CI, confidence interval; mRS, modified Rankin Scale.

Association between Estimated Pulse Wave Velocity and Poor Functional Outcome

Poor functional outcome at 3 months occurred in 151 patients (53.0%), including 62.0% of patients in the high ePWV group and 36.6% in the low ePWV group (P<0.001). Higher ePWV was associated with poor functional outcome in both continuous (adjusted OR 1.338, 95% CI 1.121–1.597, P=0.001) (Table 2, Fig. 2B) and dichotomized analyses (Table 2, Fig. 3B). No significant subgroup interactions were identified (all P for interaction >0.05) (Supplementary Table 3).

Mediation Effect of Estimated Pulse Wave Velocity on Functional Outcome via Collateral Status

Mediation analysis revealed that the effect of ePWV on functional outcome was partially mediated by collateral status in the overall cohort. The indirect effect of ePWV on mRS through the Tan scale was statistically significant (adjusted β 0.022, 95% CI 0.001–0.053, P=0.040), while the direct effect also remained significant (adjusted β 0.237, 95% CI 0.115–0.359, P<0.001). The total effect of ePWV on functional outcome was also significant (adjusted β 0.259, 95% CI 0.137–0.381, P<0.001). Collateral status accounted for 8.6% of the total effect of ePWV on functional outcome (Fig. 4A).

Fig. 4.

Mediation effects of ePWV on functional outcome through collateral status. Panels (A–C) show mediation results in the overall cohort and stratified by BP management strategy. Panel (D) summarizes effect modification by BP management strategy, assessed using the interaction term between ePWV and BP management for the direct pathway and the index of moderated mediation for the indirect pathway. ePWV, estimated pulse wave velocity; mRS, modified Rankin Scale; Boot SE, bootstrapped standard error; CI, confidence interval; BP, blood pressure. *Adjusted for age (<60 vs. ≥60 years), sex, National Institutes of Health Stroke Scale score, onset-to-randomization, BP management strategy, Alberta Stroke Program Early CT Score, internal carotid artery occlusion, smoking, atrial fibrillation, previous stroke, and stroke subtype. †Adjusted for age (<60 vs. ≥60 years), sex, National Institutes of Health Stroke Scale score, onset-to-randomization, BP management strategy, Alberta Stroke Program Early CT Score, and internal carotid artery occlusion.

In the conventional BP management group, while the total effect of ePWV on functional outcome was statistically significant (adjusted β 0.174, 95% CI 0.003–0.338, P=0.046), neither the direct effect (adjusted β 0.142, 95% CI –0.024 to 0.308, P=0.092) nor the indirect effect through the collateral status (adjusted β 0.032, 95% CI –0.004 to 0.081, P=0.168) reached statistical significance (Fig. 4B).

In the intensive BP management group, both the total effect (adjusted β 0.297, 95% CI 0.120–0.474, P=0.001) and the direct effect (adjusted β 0.286, 95% CI 0.108–0.465, P=0.002) were statistically significant, whereas the indirect effect via collateral status was not (adjusted β 0.011, 95% CI –0.013 to 0.047, P=0.466) (Fig. 4C). However, no statistically significant differences between BP management strategies were observed for either the direct effect (P for interaction=0.257) or the indirect effect (95% CI –0.055 to 0.009, P=0.260) (Fig. 4D).

DISCUSSION

In this secondary analysis of the OPTIMAL‑BP trial, higher ePWV was associated with poorer functional outcomes after successful EVT. Extending our previous observations from the same cohort, we demonstrated that the association between ePWV and collateral status remained significant after multivariable adjustment. We further quantified the contribution of collateral status to the relationship between ePWV and functional outcome using mediation analysis, which showed that collateral status accounted for only a small proportion of the observed association.

These findings indicate that mechanisms beyond collateral circulation may contribute to the association between ePWV and clinical outcomes. Increased arterial stiffness reduces vascular compliance and impairs buffering of pulsatile flow [16], leading to enhanced transmission of hemodynamic stress to the cerebral microcirculation [10]. These hemodynamic changes may contribute to endothelial dysfunction, oxidative stress [17], and impaired autoregulation [9], thereby exacerbating neurovascular injury [18]. Although arterial stiffness could theoretically influence collateral function through reduced vasodilatory capacity and microvascular remodeling [17,19,20], our results indicate that this pathway contributes only to a limited extent to this association. To further explore other potential mechanisms, we performed additional exploratory mediation analyses using infarct volume and baseline NIHSS as candidate mediators. Neither variable demonstrated a significant indirect effect linking ePWV to functional outcome (Supplementary Table 4). Taken together, these findings suggest that the association between ePWV and clinical outcomes is not fully explained by collateral status, infarct volume, or baseline NIHSS.

We found no statistically significant effect modification by BP management strategy. Although stratified analyses suggested a numerically stronger association between ePWV and functional outcome in the intensive BP group, neither the interaction term for the direct effect nor the index of moderated mediation was statistically significant. Therefore, these findings should be considered exploratory. We initially hypothesized that intensive BP reduction after EVT might differentially influence the association between ePWV and clinical outcome because arterial stiffness is associated with impaired vascular compliance and cerebral autoregulatory capacity [19,20]. Consequently, patients with higher ePWV may be less able to tolerate post-EVT BP reduction and may be more susceptible to reductions in cerebral perfusion [12,21-23]. However, the present analyses did not demonstrate significant differences according to BP management strategy.

Because age is a major determinant of ePWV, its association with clinical outcomes may partially reflect the effects of chronological aging [1]. While the associations between ePWV and both collateral status and functional outcome remained significant in the multivariable analyses using dichotomized age adjustment, they were attenuated and no longer reached statistical significance when age was modeled as a continuous variable (Supplementary Tables 5, 6). Furthermore, the addition of ePWV did not meaningfully improve model discrimination beyond continuous age (Supplementary Table 7). Taken together, these findings suggest that ePWV may be better interpreted as a surrogate marker of vascular aging derived from age and BP rather than as an age-independent prognostic marker.

Given that the Tan scale was originally developed for anterior circulation occlusions, we performed a sensitivity analysis restricted to patients with anterior circulation occlusions. The associations of ePWV with both collateral status and functional outcome remained significant and directionally consistent with the primary analyses (Supplementary Table 8). These findings support the robustness of our results within the population for which Tan collateral grading is most applicable.

The present findings nevertheless have potential clinical relevance. Although collateral status accounted for only a modest proportion of the association between ePWV and functional outcome, the observed associations remained directionally similar across the primary and sensitivity analyses. In addition, ePWV can be readily calculated from routinely available clinical variables [24], and the study was conducted using a well-characterized multicenter randomized clinical trial cohort with standardized imaging and outcome assessment. However, several limitations should be acknowledged. First, collateral status was assessed on baseline computed tomography angiography, whereas ePWV was calculated using BP measured at trial enrollment after EVT, introducing a temporal mismatch that limits causal interpretation. Therefore, the mediation findings should be interpreted as an exploratory decomposition of the observed associations rather than evidence of a causal pathway, and further studies using temporally aligned measurements are warranted. Second, ePWV is an estimated surrogate rather than a direct measure of arterial stiffness and may be influenced by acute hemodynamic conditions. Third, residual confounding cannot be excluded despite adjustment for relevant clinical variables. Finally, the study population consisted exclusively of Korean patients, which may limit generalizability to other populations.

CONCLUSION

Higher ePWV, a surrogate marker of vascular aging, was associated with poorer collateral status and unfavorable functional outcome after EVT. Collateral status accounted for only a small proportion of the observed association between ePWV and functional outcome, highlighting the need for further studies to identify additional mechanisms linking arterial stiffness to clinical outcomes.

SUPPLEMENTARY MATERIALS

Supplementary materials related to this article can be found online at https://doi.org/10.5469/neuroint.2026.00822.

Supplementary Table 1.

Baseline characteristics according to ePWV

neuroint-2026-00822-Supplementary-Table-1.pdf
Supplementary Table 2.

Subgroup analysis of the association between high ePWV and poor collaterals

neuroint-2026-00822-Supplementary-Table-2.pdf
Supplementary Table 3.

Subgroup analysis of the association between high ePWV and poor functional outcome

neuroint-2026-00822-Supplementary-Table-3.pdf
Supplementary Table 4.

Exploratory multivariable-adjusted mediation analyses using collateral status, infarct volume, and baseline NIHSS as candidate mediators of the association between ePWV and 3-month functional outcome

neuroint-2026-00822-Supplementary-Table-4.pdf
Supplementary Table 5.

Sensitivity analysis of the association of ePWV with poor collaterals or poor functional outcome using continuous age adjustment

neuroint-2026-00822-Supplementary-Table-5.pdf
Supplementary Table 6.

Mediation effects of ePWV on functional outcome through collateral status with continuous age adjustment

neuroint-2026-00822-Supplementary-Table-6.pdf
Supplementary Table 7.

Incremental predictive value of ePWV beyond continuous age

neuroint-2026-00822-Supplementary-Table-7.pdf
Supplementary Table 8.

Sensitivity analysis of the association of ePWV with poor collaterals and poor functional outcome among patients with internal carotid, middle cerebral, or anterior cerebral artery occlusions

neuroint-2026-00822-Supplementary-Table-8.pdf

Notes

Fund

This research was supported by a grant from Patient-Centered Clinical Research Coordinating Center (PACEN) funded by the Ministry of Health & Welfare, Republic of Korea (grant number: HC19C0028).

Ethics Statement

The OPTIMAL-BP trial received ethical approval from the Institutional Review Board (IRB) of Severance Hospital (approval number 4-2019-1208), as well as from the IRBs of all participating institutions. Written informed consent was obtained from all participants or their legal representatives. This post hoc secondary analysis adhered to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines to ensure rigorous and transparent reporting. We have anonymized all patient information that could identify an individual.

Conflicts of Interest

BMK, KDS, JHB have been the editor and assistant editors of the Neurointervention since 2022. No potential conflict of interest relevant to this article was reported.

Author Contributions

Concept and design: MH and HSN. Analysis and interpretation: MH, SJ, HSL, and HSN. Data collection: MH, EP, JWJ, HL, JH, YDK, HJ, KHK, Jaeseob Yun, JKC, IHL, BMK, DJK, NYS, BHC, SHA, HP, SIS, Jeong-Ho Hong, TJS, YC, GSK, KDS, KL, JYC, JHS, SL, JHB, HJC, DHS, JK, Joonsang Yoo, MB, YHJ, YHH, CKK, JGK, CJL, SP, SUK, OYB, Ji Hoe Heo, and HSN. Writing the article: MH. Critical revision of the article: MH, JWJ, YDK, KHK, Jaeseob Yun, and HSN. Final approval of the article: HSN. Statistical analysis: MH. Obtained funding: HSN. Overall responsibility: HSN.

References

1. Prabhakaran S, Gonzalez NR, Zachrison KS, Adeoye O, Alexandrov AW, Ansari SA, et al. 2026 guideline for the early management of patients with acute ischemic stroke: a guideline from the American Heart Association/American Stroke Association. Stroke 2026;57:e316–e436.
2. Maguida G, Shuaib A. Collateral circulation in ischemic stroke: an updated review. J Stroke 2023;25:179–198.
3. Gensicke H, Al-Ajlan F, Fladt J, Campbell BCV, Majoie CBLM, Bracard S, et al. Comparison of three scores of collateral status for their association with clinical outcome: the HERMES collaboration. Stroke 2022;53:3548–3556.
4. Bosshart SL, Kappelhof M, Stebner A, Fujiwara S, Cimflova P, Schüngel MS, et al. Decision-making for endovascular thrombectomy in patients with large vessel occlusions and mild neurological deficit: a consensus statement. J Stroke 2025;27:338–349.
5. Zhang X, Ren X, Zhang Y, Zhang Y, Zhang L, Shen H, et al. Interaction of brain imaging features and effects of intensive blood pressure lowering after endovascular treatment for acute ischaemic stroke: the pre-specified secondary analyses of ENCHANTED2/MT trial. EClinicalMedicine 2025;83:103197.
6. Olthuis SGH, Pinckaers FME, Robbe MMQ, de Ridder IR, Hoving JW, Venema E, et al. CT perfusion imaging after selection for late-window endovascular stroke treatment: secondary analysis of the MR CLEAN-LATE randomized trial. JAMA Neurol 2025;82:589–596.
7. Winkelmeier L, Kniep H, Thomalla G, Bendszus M, Subtil F, Bonekamp S. Arterial collaterals and endovascular treatment effect in acute ischemic stroke with large infarct: a secondary analysis of the TENSION Trial. Radiology 2025;314e242401.
8. Vlachopoulos C, Terentes-Printzios D, Laurent S, Nilsson PM, Protogerou AD, Aznaouridis K, et al. Association of estimated pulse wave velocity with survival: a secondary analysis of SPRINT. JAMA Netw Open 2019;2e1912831.
9. Reeve EH, Barnes JN, Moir ME, Walker AE. Impact of arterial stiffness on cerebrovascular function: a review of evidence from humans and preclincal models. Am J Physiol Heart Circ Physiol 2024;326:H689–H704.
10. Scheuermann BC, Parr SK, Schulze KM, Kunkel ON, Turpin VG, Liang J, et al. Associations of cerebrovascular regulation and arterial stiffness with cerebral small vessel disease: a systematic review and meta-analysis. J Am Heart Assoc 2023;12e032616.
11. Han M, Joo H, Lee H, Heo J, Jung JW, Kim YD, et al. Association between estimated pulse wave velocity and endovascular thrombectomy outcome: a secondary analysis of the OPTIMAL-BP Trial. Stroke Vasc Interv Neurol 2025;5e001678.
12. Nam HS, Kim YD, Heo J, Lee H, Jung JW, Choi JK, et al. Intensive vs conventional blood pressure lowering after endovascular thrombectomy in acute ischemic stroke: the OPTIMAL-BP randomized clinical trial. JAMA 2023;330:832–842.
13. Reference Values for Arterial Stiffness’ Collaboration. Determinants of pulse wave velocity in healthy people and in the presence of cardiovascular risk factors: ‘establishing normal and reference values’. Eur Heart J 2010;31:2338–2350.
14. Greve SV, Blicher MK, Kruger R, Sehestedt T, Gram-Kampmann E, Rasmussen S, et al. Estimated carotid-femoral pulse wave velocity has similar predictive value as measured carotid-femoral pulse wave velocity. J Hypertens 2016;34:1279–1289.
15. Tan BY, Wan-Yee K, Paliwal P, Gopinathan A, Nadarajah M, Ting E, et al. Good intracranial collaterals trump poor ASPECTS (Alberta Stroke Program Early CT Score) for intravenous thrombolysis in anterior circulation acute ischemic stroke. Stroke 2016;47:2292–2298.
16. Han M, Joo H, Lee H, Heo J, Jung JW, Kim YD, et al. Arterial stiffness predicts the outcome of endovascular treatment in patients with acute ischemic stroke. J Clin Med 2024;13:4198.
17. Herzog MJ, Müller P, Lechner K, Stiebler M, Arndt P, Kunz M, et al. Arterial stiffness and vascular aging: mechanisms, prevention, and therapy. Signal Transduct Target Ther 2025;10:282.
18. Ling YH, Pan LH, Lin CH, Wang YF, Wu CH, Lirng JF, et al. Association of increased central arterial stiffness with BBB disruption in patients with reversible cerebral vasoconstriction syndrome. Neurology 2025;105e214318.
19. Acampa M, Romano DG, Lazzerini PE, Leonini S, Guideri F, Tassi R, et al. Increased arterial stiffness is associated with poor collaterals in acute ischemic stroke from large vessel occlusion. Curr Neurovasc Res 2018;15:34–38.
20. Kasal DA, Sena V, Huguenin GVB, De Lorenzo A, Tibirica E. Microvascular endothelial dysfunction in vascular senescence and disease. Front Cardiovasc Med 2025;12:1505516.
21. Jung JW, Ko EL, Heo J, Lee H, Kim B, Kim YD, et al. Association between hyperacute blood pressure lowering and outcomes in patients with endovascular thrombectomy. J Stroke 2026;28:136–149.
22. Yang P, Song L, Zhang Y, Zhang X, Chen X, Li Y, et al. Intensive blood pressure control after endovascular thrombectomy for acute ischaemic stroke (ENCHANTED2/MT): a multicentre, open-label, blinded-endpoint, randomised controlled trial. Lancet 2022;400:1585–1596.
23. Writing Committee Members, Jones DW, Ferdinand KC, Taler SJ, Johnson HM, Shimbo D, et al. 2025 AHA/ACC/AANP/AAPA/ABC/ACCP/ACPM/AGS/AMA/ASPC/NMA/PCNA/SGIM guideline for the prevention, detection, evaluation and management of high blood pressure in adults: a report of the American College of Cardiology/American Heart Association Joint Committee on clinical practice guidelines. Hypertension 2025;82e212. –e316.
24. Kakaletsis N, Hosomi N, Nezu T, Michel P, Guillaume T, Strambo D, et al. Estimated pulse wave velocity predicts outcomes after acute ischemic stroke: a multinational cohort analysis. Am J Hypertens 2026;39:1142–1150.

Article information Continued

Fig. 1.

Flowchart of patient inclusion and exclusion. OPTIMAL-BP, Outcome in Patients Treated With Intra-Arterial Thrombectomy–Optimal Blood Pressure Control.

Fig. 2.

Restricted cubic spline analyses of the association of ePWV with collateral status and functional outcome. Panel (A) shows the adjusted association between ePWV and poor collateral status. Panel (B) shows the adjusted association between ePWV and poor functional outcome at 3 months. Solid lines represent adjusted ORs, and shaded areas indicate 95% CIs. ePWV, estimated pulse wave velocity; OR, odds ratio; CI, confidence interval.

Fig. 3.

Distribution of collateral status and functional outcome by ePWV. Panel (A) illustrates the distribution of collateral grades according to the Tan scale (0–3) in patients stratified by ePWV. Poor collateral status was defined as Tan scale ≤1. The corresponding unadjusted and adjusted ORs for poor collateral status according to ePWV group are also presented. Panel (B) shows the distribution of 3-month functional outcomes based on the mRS (0–6), with poor functional outcome defined as mRS ≥3. The corresponding unadjusted and adjusted ORs for poor functional outcome according to ePWV group are also presented. ePWV, estimated pulse wave velocity; OR, odds ratio; CI, confidence interval; mRS, modified Rankin Scale.

Fig. 4.

Mediation effects of ePWV on functional outcome through collateral status. Panels (A–C) show mediation results in the overall cohort and stratified by BP management strategy. Panel (D) summarizes effect modification by BP management strategy, assessed using the interaction term between ePWV and BP management for the direct pathway and the index of moderated mediation for the indirect pathway. ePWV, estimated pulse wave velocity; mRS, modified Rankin Scale; Boot SE, bootstrapped standard error; CI, confidence interval; BP, blood pressure. *Adjusted for age (<60 vs. ≥60 years), sex, National Institutes of Health Stroke Scale score, onset-to-randomization, BP management strategy, Alberta Stroke Program Early CT Score, internal carotid artery occlusion, smoking, atrial fibrillation, previous stroke, and stroke subtype. †Adjusted for age (<60 vs. ≥60 years), sex, National Institutes of Health Stroke Scale score, onset-to-randomization, BP management strategy, Alberta Stroke Program Early CT Score, and internal carotid artery occlusion.

Table 1.

Baseline characteristics according to collateral status and functional outcome

Collateral status
Functional outcome
Poor (n=93) Good (n=192) P-value Poor (n=151) Good (n=134) P-value
Age (y) 76.1±10.6 72.0±11.4 0.007 76.0±11.0 70.3±10.9 <0.001
Women 36 (38.7) 79 (41.1) 0.694 66 (43.7) 49 (36.6) 0.220
Body mass index (kg/m2) 24.4 [22.4, 26.6] 23.4 [21.4, 25.5] 0.010 23.8 [21.5, 26.0] 23.9 [21.7, 25.5] 0.980
Hypertension 70 (75.3) 147 (76.6) 0.810 115 (76.2) 102 (76.1) 0.994
Diabetes 42 (45.2) 77 (40.1) 0.417 75 (49.7) 44 (32.8) 0.004
Dyslipidemia 33 (35.5) 73 (38.0) 0.678 50 (33.1) 56 (41.8) 0.130
Smoking 18 (19.4) 46 (24.0) 0.383 31 (20.5) 33 (24.6) 0.408
Atrial fibrillation 51 (54.8) 90 (46.9) 0.207 79 (52.3) 62 (46.3) 0.308
Coronary artery disease 9 (9.7) 24 (12.5) 0.485 18 (11.9) 15 (11.2) 0.848
Peripheral artery disease 3 (3.2) 5 (2.6) 0.719 2 (1.3) 6 (4.5) 0.153
Previous stroke 19 (20.4) 42 (21.9) 0.780 40 (26.5) 21 (15.7) 0.026
Active cancer 2 (2.2) 10 (5.2) 0.348 6 (4.0) 6 (4.5) 0.832
Stroke subtype
 Cardioembolism 54 (58.1) 91 (47.4) 0.466 79 (52.3) 66 (49.3) 0.514
 Large artery atherosclerosis 24 (25.8) 54 (28.1) 41 (27.2) 37 (27.6)
 Stroke of other determined etiology 1 (1.1) 3 (1.6) 2 (1.3) 2 (1.5)
 Negative evaluation 10 (10.8) 30 (15.6) 17 (11.3) 23 (17.2)
 Two or more causes 4 (4.3) 14 (7.3) 12 (7.9) 6 (4.5)
NIHSS score just before EVT 14.0 [9.0, 19.0] 12.0 [7.0, 17.0] 0.037 16.0 [11.3, 20.0] 9.0 [6.0, 12.3] <0.001
ASPECTS 9.0 [7.0, 9.0] 9.0 [8.0, 10.0] 0.004 8.0 [7.0, 9.0] 9.0 [8.0, 10.0] <0.001
Intravenous tPA 28 (30.1) 65 (33.9) 0.527 57 (37.7) 36 (26.9) 0.050
Onset-to-puncture (min) 388.0 [215.0, 678.0] 373.0 [208.0, 731.0] 0.613 335.0 [215.5, 737.3] 435.5 [198.5, 722.5] 0.677
Onset-to-randomization (min) 502.0 [290.0, 780.0] 494.0 [325.0, 840.0] 0.516 450.0 [315.0, 840.0] 515.0 [313.8, 812.5] 0.703
Procedure time (min) 42.0 [25.0, 77.0] 48.0 [33.0, 80.0] 0.107 50.0 [34.3, 85.0] 39.0 [25.0, 68.5] 0.008
Infarction volume (mL) 25.9 [8.5, 130.5] 15.1 [5.3, 40.7] 0.003 32.8 [10.7, 105.9] 8.5 [4.0, 21.0] <0.001
First-pass recanalization 65 (69.9) 123 (64.1) 0.330 96 (63.6) 92 (68.7) 0.366
Occlusion site
 ICA 23 (24.7) 34 (17.7) 0.165 38 (25.2) 19 (14.2) 0.021
 MCA 59 (63.4) 140 (72.9) 0.102 97 (64.2) 102 (76.1) 0.029
 ACA 2 (2.2) 2 (1.0) 0.599 0 (0.0) 4 (3.0) 0.048
 BA 9 (9.7) 14 (7.3) 0.488 14 (9.3) 9 (6.7) 0.429
 VA 2 (2.2) 4 (2.1) >0.999 3 (2.0) 3 (2.2) >0.999
 PCA 0 (0.0) 1 (0.5) >0.999 0 (0.0) 1 (0.7) 0.470
SBP (mmHg) 153.0 [145.0, 160.0] 150.0 [145.0, 164.0] 0.480 150.0 [145.0, 163.0] 151.5 [145.0, 162.0] 0.888
DBP (mmHg) 86.0 [76.0, 95.0] 83.0 [76.0, 92.0] 0.575 81.0 [74.0, 91.0] 87.5 [77.0, 95.0] 0.128
ePWV (m/s) 13.7±2.1 12.9±2.1 0.004 13.6±2.1 12.6±2.1 <0.001
Intensive BP management 42 (45.2) 102 (53.1) 0.207 87 (57.6) 57 (42.5) 0.011
Poor collaterals (Tan scale ≤1) 60 (39.7) 33 (24.6) 0.007

Continuous and categorical variables are shown as mean±standard deviation or median [interquartile range] and number (%), respectively.

NIHSS, National Institutes of Health Stroke Scale; EVT, endovascular thrombectomy; ASPECTS, Alberta Stroke Program Early CT Score; tPA, tissue-type plasminogen activator; ICA, internal carotid artery; MCA, middle cerebral artery; ACA, anterior cerebral artery; BA, basilar artery; VA, vertebral artery; PCA, posterior cerebral artery; SBP, systolic blood pressure; DBP, diastolic blood pressure; ePWV, estimated pulse wave velocity; BP, blood pressure.

Table 2.

Association of ePWV with poor collaterals or poor functional outcome

Outcome Per 1 m/s increase P-value High vs. low (>12.6 m/s) P-value
Poor collaterals
 Model 1 1.191 (1.054–1.347) 0.005 1.925 (1.114–3.326) 0.019
 Model 2 1.243 (1.068–1.446) 0.005 2.024 (1.071–3.826) 0.030
 Model 3 1.247 (1.069–1.454) 0.005 2.207 (1.149–4.239) 0.017
 Model 4 1.289 (1.098–1.513) 0.002 2.421 (1.228–4.775) 0.011
 Model 5 1.276 (1.083–1.503) 0.004 2.383 (1.178–4.822) 0.016
Poor functional outcome
 Model 1 1.263 (1.122–1.422) <0.001 2.817 (1.704–4.656) <0.001
 Model 2 1.204 (1.058–1.370) 0.005 2.346 (1.359–4.050) 0.002
 Model 3 1.237 (1.056–1.449) 0.008 2.510 (1.316–4.788) 0.005
 Model 4 1.273 (1.081–1.500) 0.004 2.689 (1.372–5.270) 0.004
 Model 5 1.338 (1.121–1.597) 0.001 2.813 (1.368–5.783) 0.005

Values are presented as OR (CI). Model 1: unadjusted. Model 2: adjusted for age (<60 vs. ≥60 years) and sex. Model 3: adjusted for age (<60 vs. ≥60 years), sex, National Institutes of Health Stroke Scale score, onset-to-randomization, and blood pressure management strategy. Model 4: adjusted for age (<60 vs. ≥60 years), sex, National Institutes of Health Stroke Scale score, onset-to-randomization, blood pressure management strategy, Alberta Stroke Program Early CT Score, and internal carotid artery occlusion. Model 5: Model 4 plus smoking, atrial fibrillation, previous stroke, and stroke subtype.

ePWV, estimated pulse wave velocity; OR, odds ratio; CI, confidence interval.