Prediction of Occlusion Duration in Nonacute Intracranial Large Artery Through High‐Resolution Vessel Wall Imaging

Abstract

Background

Accurately assessing the duration of nonacute intracranial large vessel occlusion (ILVO) is critical, because occlusion exceeding 3 months significantly reduces endovascular treatment efficacy. This study explores high‐resolution vessel wall imaging to differentiate subacute (3 days–1 month), short‐duration chronic (1–3 months), and long‐duration chronic (>3 months) ILVOs.

Methods

A retrospective analysis was conducted on consecutive patients with nonacute ILVOs with well‐documented clinical courses (June 2018–March 2022). High‐resolution vessel wall imaging pre‐ and postcontrast features, including traditional and histogram characteristics, were evaluated. Logistic regression identified key determinants for distinguishing occlusion durations. Endovascular treatment outcomes were assessed in a subset of patients.

Results

Among 218 ILVOs (77 subacute, 69 short‐duration chronic, 72 long‐duration chronic), multivariate analysis identified signal intensity (odds ratio [OR], 3.71; P<0.001), enhancement index (OR, 1.79; P<0.001), precontrast skewness (OR, 4.41; P=0.04), and postcontrast coefficient of variation (OR, 9.34; P<0.001) as significant for differentiating subacute and long‐duration chronic ILVOs. Short‐ versus long‐duration chronic ILVOs were distinguished by signal intensity (OR, 4.02; P<0.001), enhancement index (OR, 2.02; P<0.001), lumen diameter (OR, 1.70; P=0.002), and precontrast skewness (OR, 4.91; P=0.047). Short‐duration ILVOs showed higher recanalization success (91.7% versus 61.9%, P=0.02) and required fewer stents (45.8% versus 90.5%, P=0.02) compared with long‐duration ILVOs.

Conclusions

High‐resolution vessel wall imaging features effectively differentiate ILVO durations, providing structural insights into occluded vessels. These findings can guide endovascular recanalization strategies, emphasizing the importance of occlusion duration in treatment planning.

Nonstandard Abbreviations and Acronyms

CV

coefficient of variation

EI

enhancement index

HR‐VWI

high‐resolution vessel wall imaging

ILVO

intracranial large vessel occlusion

SI

signal intensity

Clinical Perspective.

What Is New?

• The duration of intracranial large vessel occlusions can be effectively differentiated using high‐resolution vessel wall imaging and further categorized into the subacute phase, short‐duration chronic phase, and long‐duration chronic phase.

• The long‐duration chronic intracranial large vessel occlusions demonstrate lower signal intensity and enhancement index compared with subacute and short‐duration chronic intracranial large vessel occlusions and can be accurately distinguished by incorporating traditional imaging features with histogram analysis.

What Are the Clinical Implications?

• High‐resolution vessel wall imaging can noninvasively differentiate subacute and short‐duration chronic occlusions from long‐duration chronic occlusions, and its clinical significance lies in identifying patients with shorter occlusion durations who are more likely to benefit from endovascular recanalization.

Symptomatic nonacute intracranial large vessel occlusion (ILVO) within the anterior circulation poses a significant risk of morbidity and mortality due to ischemic stroke. Patients with objectively compromised cerebral perfusion face an elevated risk of persistent neurological deficits, recurrent stroke, and potential cognitive impairment resulting from prolonged hypoperfusion caused by nonacute ILVO. ¹ Endovascular recanalization has demonstrated feasibility and efficacy in highly selective cases for treating symptomatic nonacute ILVO, offering potential improvements in patient symptoms. ² The selection criteria for this therapy require a comprehensive assessment of technical feasibility and the likelihood of successful recanalization, based on thorough preoperative clinical and neuroimaging examinations. The duration of occlusion significantly influences the maneuverability of the microguidewire within the occluded segment, making recanalization more challenging due to increased calcification and fibrosis. Successful rates are notably higher when the occlusion duration is <3 months. ³ A recent study ⁴ revealed that each additional 10 days of occlusion duration before the 51‐day inflection point reduces the probability of successful recanalization by 57%. This finding underscores the critical importance of elucidating the temporal dynamics of the underlying biology of these occlusions.

Despite the potential benefits, accurately determining the duration of occlusion in patients with ILVO remains challenging due to the diverse neurologic manifestations associated with this condition. Endovascular recanalization for nonacute ILVO presents significant technical challenges, particularly because conventional imaging modalities, such as computed tomography angiography, magnetic resonance angiography, ultrasonography, and digital subtraction angiography, are inadequate for assessing vessel segments with minimal or no blood flow. Unlike stenting for stenosis, conventional imaging techniques cannot reliably determine the nature of an occlusion.

High‐resolution vessel wall imaging (HR‐VWI) has emerged as a practical and accurate solution for diagnosing ILVO. Its attributes, including black‐blood effects, excellent scan efficiency, high isotropic resolution, signal‐to‐noise ratio, and extensive spatial coverage, address the limitations of conventional modalities. ⁵ Multiple studies ⁶ , ⁷ , ⁸ have demonstrated that HR‐VWI is capable of not only characterizing various vascular lesion components, such as plaque enhancement, fibrous matrix, hemorrhage, and thrombus, but also assessing their progression across different stages. Histogram analysis has been used to investigate quantitative biomarkers of lesions observed through HR‐VWI. ⁹ This analysis focuses on compositional features to enhance our understanding of the internal nature of ILVO. Furthermore, Zhang et al ¹⁰ found the compositional characteristics of occlusions, such as extensive hyperintense signals and the absence of extensive calcification, were associated with successful recanalization.

In line with these advancements, we hypothesize that the signal intensity (SI) of the occluded segment demonstrates temporal fluctuations. Through a retrospective study, we aimed to perform a comprehensive histogram analysis to identify potential quantitative biomarkers for the duration of ILVO as visualized by HR‐VWI. This approach seeks to uncover the underlying internal pathophysiology of ILVOs, ultimately contributing to improved success rates in recanalization procedures. ¹¹

METHODS

Study Patients

This retrospective study was approved by the ethics committee of our institute (2023–052), and informed consent was waived due to its retrospective nature. Data from 218 patients with nonacute ILVOs who underwent HR‐VWI between June 2018 and March 2022 at our hospital were analyzed. The study included data from subacute, short‐duration chronic, and long‐duration chronic groups to observe the complete evolution of arterial occlusion over time. Inclusion criteria consisted of a clear onset time and course, occlusive lesions in the middle cerebral and distal internal carotid arteries, and performance of magnetic resonance imaging including HR‐VWI. Exclusion criteria encompassed nondiagnostic studies due to patient motion artifacts, other cerebral vasculopathies, and a history of transluminal intervention. The data supporting the findings of this study are available from the corresponding author upon reasonable request.

Arterial occlusions were categorized as subacute, short‐duration chronic, and long‐duration chronic groups based on clinical and imaging criteria. Subacute occlusions were defined as those occurring between 3 days and 1 month following the onset of acute occlusion, as determined by diffusion weighted imaging and HR‐VWI examinations. Chronic occlusions were identified by the absence of infarcts or presence of chronic infarcts in the downstream vascular territory, or through reports from previous angiographic studies conducted >1 month before HR‐VWI indicating an existing occlusion. Based on the patients' clinical histories, chronic occlusion was subsequently classified into 2 categories: short‐duration (1–3 months) and long‐duration (>3 months). Data S1 and Figures S1 detail methodologies for differentiating occlusion durations.

Magnetic Resonance Imaging Data Acquisition

Magnetic resonance imaging was conducted using on a 3T magnetic resonance imaging scanner (MAGNETOM Prisma; Siemens Healthcare, Erlangen, Germany) equipped with a 64‐channel head–neck coil. HR‐VWI was conducted using a precontrast 3‐dimensional turbo spin‐echo technique known as T1w–sampling perfection with application‐optimized contrast using different flip angle evolutions. The parameters for HR‐VWI were as follows: repetition time of 900 ms, echo time of 15 ms, field of view of 200×200 mm², matrix size of 320×320, slice number of 224, slice thickness of 0.53 mm, and an acquisition time of approximately 7 minutes. Chemical shift fat suppression was applied during image acquisition. A partial Fourier technique (6/8 in the slice direction) and parallel imaging acceleration with a factor of 2 generalized autocalibrating partially parallel acquisition, (GRAPPA) were also used. Postcontrast HR‐VWI was performed 5 minutes after the administration of a single‐dose gadolinium‐based contrast agent (0.1 mmol per kg of body weight; Magnevist; Bayer Healthcare Pharmaceuticals).

HR‐VWI Analysis

Two independent neuroradiologists, comprising a radiology resident with 8 years of experience and a senior radiologist with >15 years of expertise in neuroradiology diagnostics, assessed imaging features using ImageJ software (National Institutes of Health, Bethesda, MD). The occluded segment was defined as a focal region exhibiting abnormal intraluminal SI that entirely replaced the normal intraluminal black‐blood signal on both nonenhanced and contrast‐enhanced HR‐VWI images. ¹² Disagreements between the 2 evaluators were resolved through consensus. The intraclass correlation coefficient was used to evaluate both intrarater and interrater agreements. Additionally, the intraclass correlation coefficient was used in the sample size calculation. For detailed information, refer to Table S1. Various features, including the enhancement index (EI), intraplaque hemorrhage, and lumen diameter, were quantified. Statistical information derived from histograms was characterized using skewness, kurtosis, and the coefficient of variation (CV). Skewness describes the degree of asymmetry in the distribution shape; positive skewness indicates a longer tail on the right side, whereas negative skewness indicates a longer tail on the left side. Kurtosis measures the peakedness or flatness of a distribution relative to a normal distribution; higher kurtosis signifies a sharper peak with heavier tails, whereas lower kurtosis indicates a flatter peak with lighter tails. The CV reflects the dispersion of SI within an image region, providing a normalized measure of variability relative to the mean. These parameters are further defined below.

$$\begin{array}{c}1.\text{Enhancement index}=\left(\frac{\mathrm{SI}\_{\text{occlusion}}_{\text{Postcontrast}}}{\mathrm{SI}\_{\text{white matter}}_{\text{Postcontrast}}}\right.\hfill \\ -\left.\frac{\mathrm{SI}\_{\text{occlusion}}_{\text{Precontrast}}}{\mathrm{SI}\_{\text{white matter}}_{\text{Precontrast}}}\right)/\frac{\mathrm{SI}\_{\text{occlusion}}_{\text{Precontrast}}}{\mathrm{SI}\_{\text{white matter}}_{\text{Precontrast}}},\hfill \end{array}$$

where EI is the percentage of contrast enhancement. SI, normalized to adjacent white matter, was measured. ¹³

2. Intraplaque hemorrhage was defined as SI >150% of adjacent white matter on the precontrast HR‐VWI images.

3. The lumen diameter of the proximal one‐fourth, distal one‐fourth, and middle occluded segments were measured, and their mean values were used.

4. Skewness = $\frac{{\sum }_1^n{\left(x_i-\overline{x}\right)}^3}{n{\sigma }^3}$

5. Kurtosis = $\frac{{\sum }_1^n{\left(x_i-\overline{x}\right)}^4}{n{\sigma }^4}-3$

6. Coefficient of variation = $\sigma /\overline{x}$

where $x$ is the SI value of each pixel within the region of interest; $\overline{x}$ and $\sigma$ are the mean and standard deviation of $x$, respectively; and n is the pixel number.

Endovascular Reperfusion Therapy

In our hospital, endovascular treatment was administered to a subgroup of patients presenting with symptomatic nonacute ILVOs. These patients exhibited hemodynamic compromise and experienced progressive or recurrent ischemic symptoms despite optimal medical therapy. Ischemic core and penumbral regions were assessed using RAPID software (iSchemaView). Balloon angioplasty and stenting were performed as clinically indicated. Revascularization status was evaluated using the Modified Thrombolysis in Cerebral Infarction scale, with successful revascularization defined as achieving a Modified Thrombolysis in Cerebral Infarction grade of 2b or 3.

Statistical Analysis

All statistical analyses were conducted using MedCalc software (version 11.1.1.0; Mariakerke, Belgium). A P value of <0.05 was considered statistically significant. Normally distributed variables were expressed as mean±SD, and the ANOVA test was used. Abnormally distributed variables were shown as the median (interquartile range), and nonparametric (Kruskal‐Wallis) tests were used. Categorical data were expressed as numbers and percentages. Multiple testing correction was performed using Bonferroni correction, with a significance threshold of P<0.0167 (0.05/3) due to 3 pairwise comparisons among the 3 groups. Univariate analyses were performed to evaluate the association between parameters and occlusion duration. Variables with a P value <0.1 in univariate analyses were included in multivariate binary logistic regression models. Odds ratios (ORs) with corresponding 95% CIs were calculated to quantify the strength of associations. Receiver operating characteristic curves were constructed to assess model performance, and the area under the curve (AUC) was calculated to provide a quantitative measure of diagnostic accuracy. Optimism‐corrected AUCs were calculated using the bootstrapping method, with 1000 replications to compensate overfitting of the models using Python version 3.10. The optimism percentage was defined as the relative difference between the apparent AUC and the optimism‐corrected AUC, calculated as $\frac{{\mathrm{AUC}}_{\text{apparent}}-{\mathrm{AUC}}_{\text{optimism}-\text{corrected}}}{{\mathrm{AUC}}_{\text{apparent}}}\times 100\%$. A lower optimism percentage indicates less overestimation of model performance.

RESULTS

Participants

A total of 251 patients were initially enrolled in the study. Of these, 33 patients were excluded due to dissection (n=2), Moyamoya disease (n=7), and inadequate image quality (n=24). Consequently, the final analysis included 218 patients with ILVOs. The cohort comprised 76.6% men, with a mean age of 57 years. Patients were categorized into 3 groups based on the duration of occlusion: 77 patients in the subacute group, 69 in the short‐duration chronic group, and 72 in the long‐duration chronic group. Detailed demographic information for the patients is provided in Table 1.

Table 1.

Patient Demographics

Demographic	Subacute (n=77)	Short‐duration chronic (n=69)	Long‐duration chronic (n=72)	P value
Age, y	58 (51–67)	60 (53–65)	58 (47–64)	0.37
Men, n (%)	62 (89.5%)	56 (81.2%)	49 (68.1%)	0.11
Hypertension	53 (68.8%)	39 (56.5%)	44 (61.1%)	0.30
Diabetes	19 (24.7%)	23 (33.3%)	14 (19.4%)	0.16
Hyperlipidemia	6 (8.8%)	14 (20.3%)	10 (13.8%)	0.09
Atrial fibrillation	6 (8.8%)	8 (11.6%)	3 (4.2%)	0.26
Occlusion site				0.94
Proximal MCA	53 (68.8%)	48 (69.6%)	49 (68.1%)
Distal MCA	10 (13.0%)	4 (5.8%)	6 (8.3%)
Intracranial ICA	14 (18.2%)	17 (24.6%)	17 (23.6)

Data are presented as n (%) or as median (interquartile range). ICA indicates internal cerebral artery; and MCA, middle cerebral artery.

HR‐VWI Characteristics of ILVO

The characteristics of HR‐VWI are summarized in Table 2, with representative cases illustrated in Figure 1. HR‐VWI identified several parameters significantly associated with occlusion duration, including SI_preVWI (P<0.001), intraplaque hemorrhage (P=0.02), EI (P<0.001), lumen diameter (P=0.002), and CV_postVWI (P<0.001). Violin plot analysis was used to compare imaging characteristics across different occlusion durations, as detailed in Figure 2.

Table 2.

HR‐VWI Characteristics

Characteristics	Subacute (n=77)	Short‐duration chronic (n=69)	Long‐duration chronic (n=72)	P value
SIpreVWI	0.81 (0.73 to 0.89)	0.85 (0.73 to 0.96)	0.62 (0.52 to 0.75)	<0.001*
IPH	6 (7.8%)	12 (17.4%)	3 (3.8%)	0.02*
EI	0.38 (0.20 to 0.73)	0.81 (0.59 to 1.17)	0.07 (0.02 to 0.26)	<0.001*
Diameter, mm	1.84 (1.59 to 2.07)	1.99 (1.70 to 2.15)	1.82 (1.40 to 2.02)	0.002*
CVpreVWI	0.11 (0.04 to 0.14)	0.10 (0.04 to 0.13)	0.11 (0.07 to 0.14)	0.07
SkewnesspreVWI	0.06 (−0.16 to 0.27)	0.03 (−0.34 to 0.52)	−0.06 (−0.48 to 0.18)	0.06
KurtosispreVWI	−0.30 (−0.80 to −0.01)	−0.33 (−0.68 to 0.28)	−0.40 (−0.82 to 0.13)	0.76
CVpostVWI	0.20 (0.12 to 0.26)	0.09 (0.06 to 0.13)	0.10 (0.07 to 0.12)	<0.001*
SkewnesspostVWI	0.24 (−0.06 to 0.53)	−0.01 (−0.34 to 0.39)	0.17 (−0.23 to 0.51)	0.07
KurtosispostVWI	−0.47 (−0.89 to 0.19)	−0.47 (−0.81 to 0.05)	−0.34 (−0.80 to 0.32)	0.36

Data are presented as n (%) or as median (interquartile range). CV indicates coefficient of variation; EI, enhancement index; HR‐VWI, high‐resolution vessel wall imaging; IPH, intraplaque hemorrhage; PostVWI, postcontrast vessel wall images; PreVWI, precontrast vessel wall images; and SI, signal intensity.

^* P<0.05.

Figure 1. HR‐VWI and histogram analysis of ILVOs.

TOF‐MRA demonstrated occlusions in the left middle cerebral artery in all patients. Hyperintense signals on DWI indicate the presence of cerebral infarction lesions. HR‐VWI images, including pre‐ and postcontrast imaging at occluded segments, along with quantitative histogram parameters derived from postcontrast imaging (the contours of the occluded segments and the regions of interest of white matter for normalization), are shown on the right. The histogram shows a higher coefficient of variation in subacute ILVO than in chronic ILVO, with the lowest enhancement index in long‐duration chronic ILVO. A, Subacute ILVO. B, Short‐duration chronic ILVO. C, Long‐duration chronic ILVO. Histogram plot: x axis represents signal intensity of the occluded segment; y axis represents pixel count. DWI indicates diffusion‐weighted imaging, HR‐VWI, high‐resolution vessel wall imaging; ILVO, intracranial large vessel occlusion; Kurt, kurtosis; Skew, skewness; StdDev, standard deviation; TOF‐MRA, time‐of‐flight magnetic resonance angiography; and VWI, vessel wall imaging.

Figure 2. Violin plots comparing imaging characteristics of subacute, short‐duration chronic, and long‐duration chronic ILVOs.

A, Normalized SI on precontrast HR‐VWI images. B, Enhancement index. C, CV on postcontrast HR‐VWI images. D, Mean lumen diameter. Bonferroni's correction was applied (*P<0.0167 [0.05/3]). CV indicates coefficient of variation; HR‐VWI, high‐resolution vessel wall imaging; ILVO, intracranial large vessel occlusion; postVWI, postcontrast vessel wall images; preVWI, precontrast vessel wall images; and SI, signal intensity.

Factors Associated With Occlusion Duration

Potential predictive parameters listed in Table 2 with P<0.1 were included in the regression analyses (Table 3). The univariate logistic regression analysis showed that SI_preVWI, EI, lumen diameter, skewness_preVWI, kurtosis_postVWI, and CV_postVWI were significantly different between subacute and long‐duration chronic group, and SI_preVWI, intraplaque hemorrhage, EI, lumen diameter, CV_preVWI, and skewness_preVWI were significantly different between the short‐duration and long‐duration chronic group. Multiple logistic regression analyses identified EI (OR, 1.79 [95% CI, 1.35–2.36]; P<0.001), SI_preVWI (OR, 3.71 [95% CI, 2.31–5.98]; P<0.001), skewness_preVWI (OR, 4.41 [95% CI, 1.09–17.76]; P=0.04), and CV_postVWI (OR, 9.34 [95% CI, 3.31–26.41]; P<0.001) as significant features different between subacute and long‐duration chronic group. The AUC values of EI, SI_preVWI, skewness_preVWI, and CV_postVWI were 0.820, 0.811, 0.612, and 0.825, respectively. The combination of these features improved the AUC to 0.962 (Figure 3A). Enhancement index (OR, 2.02 [95% CI, 1.44–2.83]; P<0.001), SI_preVWI (OR, 4.02 [95% CI, 2.10–7.69]; P<0.001), lumen diameter (OR, 1.70 [95% CI, 1.21–2.38]; P=0.002), and skewness_preVWI (OR, 4.91 [95% CI, 1.02–23.58]; P=0.047) were significantly associated with occlusion duration in the short‐duration and long‐duration chronic group. Their AUC of receiver operating characteristics were 0.870, 0.822, 0.662, and 0.576, respectively. The combination of these 4 features improved the AUC to 0.973 (Figure 3B). To prevent overfitting and optimism, we conducted bootstrap validation for each AUC. The optimism‐corrected AUCs for the combined features were 0.961 (95% CI, 0.928–0.990) and 0.972 (95% CI, 0.946–0.992), respectively. The optimism percentage was <0.3%, and the apparent AUCs matched the optimism‐corrected AUCs within the reported precision. Detailed values, including 95% CIs, are provided in Table S2.

Table 3.

Univariate and Multivariate Binary Logistic Regression for Factors Associated With Occlusion Duration

Characteristics	Univariate logistic regression			Multivariate logistic regression
	OR	95% CI	P value	OR	95% CI	P value
Differentiating subacute from long‐duration chronic group
SIpreVWI	2.36	1.74–3.18	<0.001	3.71	2.31–5.98	<0.001*
IPH	1.94	0.47–8.08	0.36	NA
EI	1.55	1.31–1.83	<0.001	1.79	1.35–2.36	<0.001*
Diameter, mm	1.12	1.02–1.23	0.02	1.03	0.87–1.21	0.74
CVpreVWI	0.92	0.48–1.75	0.80	NA
SkewnesspreVWI	2.60	1.22–5.55	0.01	4.41	1.09–17.76	0.04*
KurtosispreVWI	0.97	0.60–1.56	0.89	NA
CVpostVWI	6.70	3.45–13.02	<0.001	9.34	3.31–26.41	<0.001*
SkewnesspostVWI	1.13	0.65–1.95	0.67	NA
KurtosispostVWI	0.71	0.50–1.02	0.06	0.76	0.34–1.68	0.50
Differentiating short‐duration from long‐duration chronic group
SIpreVWI	2.16	1.65–2.83	<0.001	4.02	2.10–7.69	<0.001*
IPH	4.84	1.30–18.00	0.02*	0.15	0.01–3.09	0.22
EI	1.55	1.35–1.78	<0.001	2.02	1.44–2.83	<0.001*
Diameter_mm	1.21	1.09–1.34	<0.001	1.70	1.21–2.38	0.002*
CVpreVWI	0.55	0.28–1.07	0.08	2.12	0.50–8.98	0.31
SkewnesspreVWI	1.75	0.96–3.20	0.07	4.91	1.02–23.58	0.047*
KurtosispreVWI	1.08	0.66–1.77	0.75	NA
CVpostVWI	0.66	0.33–1.31	0.23	NA
SkewnesspostVWI	0.67	0.38–1.18	0.17	NA
KurtosispostVWI	0.76	0.54–1.07	0.12	NA

CV indicates coefficient of variation; EI, enhancement index; IPH, intraplaque hemorrhage; OR, odds ratio; PostVWI, postcontrast vessel wall images; PreVWI, precontrast vessel wall images; and SI, signal intensity.

^* P<0.05.

Figure 3. Receiver operating characteristic curves.

Receiver operating characteristic curves for distinguishing subacute (A) and short‐duration chronic (B) from long‐duration chronic intracranial large vessel occlusions. CV indicates coefficient of variation; EI, enhancement index; postVWI, postcontrast vessel wall images; preVWI, precontrast vessel wall images; and SI, signal intensity.

In the multivariate binary logistic regression analysis conducted to differentiate subacute from short‐duration chronic occlusions, both EI (OR, 1.22 [95% CI, 1.09–1.37]; P=0.001) and CV_postVWI (OR, 0.09 [95% CI, 0.03–0.22]; P<0.001) were identified as independent predictors. Detailed results are presented in Table S3.

Technical Success

A total of 45 patients underwent endovascular recanalization. Given that only 2 patients with subacute occlusion received endovascular therapy, patients were categorized into 2 groups based on the duration of occlusion: 24 patients in the short‐duration group (3 days to 3 month) and 21 patients in the long‐duration group (>3 months). The overall technical success rate was 77.8%, with initial recanalization achieved in 35 patients. Stent implantation was more frequently performed in the long‐duration group (90.5%) compared with the short‐duration group (45.8%; P=0.002) to achieve complete recanalization. Additionally, a higher proportion of patients in the long‐duration group (38.1%) failed to revascularize ILVOs compared with the short‐duration group (8.3%; P=0.02). No significant differences were observed between the 2 groups on clinical data, volume of ischemic core, or penumbral regions (Table S4, Figure S5).

DISCUSSION

In our extensive examination of patients with nonacute ILVOs, facilitated by 3.0T HR‐VWI, our study unveiled a nuanced relationship between magnetic resonance signals and occlusion duration. The dynamic nature of this association suggests an evolution in the biology of occluded vessels, as reflected in magnetic resonance signals over time. By integrating key parameters, including traditional and histogram‐based vascular wall features, subacute and short‐duration chronic ILVOs can be effectively distinguished from long‐duration chronic ILVOs. Understanding the duration of nonacute ILVOs emerges as pivotal in gauging the likelihood of treatment success.

The protean clinical presentation of nonacute ILVO presents challenges in accurately determining the precise timing of occlusion based solely on clinical symptoms before recanalization. Notably, our study reveals no significant differences in clinical parameters, ischemic core volume, or penumbral regions between short‐duration and long‐duration ILVOs. This finding highlights the complexity involved in estimating occlusion timing using clinical symptoms alone. Additionally, we propose that the SI of the occluded segment may vary over time, which adds another dimension to understanding the diverse pathological characteristics associated with the temporal evolution of ILVOs.

Intracranial atherosclerosis, more prevalent in Asian populations, ¹⁴ constitutes a major cause (≈70%) of nonacute ILVOs. ¹⁵ The chronological progression of intraplaque hemorrhage within atherosclerotic lesions, as demonstrated in previous studies, ¹⁶ suggests that variations in SI within the occluded segment may reflect distinct pathological characteristics. The interplay between arterial thrombosis and atherosclerotic plaques leads to unstable plaque formation, disrupting the endothelial layer and triggering local platelet accumulation and thrombosis. The EI, independently associated with unstable plaques, is lower in newly formed thrombi due to the absence of nutrient arteries. ¹⁷ , ¹⁸ Coexisting atherosclerotic plaques and new clots in subacute occlusion contribute to a higher coefficient of variation post‐VWI, gradually evolving over time as new thrombi develop capillary ingrowth. Consequently, the EI in short‐duration chronic occlusion increased significantly. In the subsequent long‐duration chronic phase, organized thrombi are characterized by macrophage dominance, increased calcification, and fibrosis, presenting distinct imaging features on HR‐VWI. Long‐duration chronic ILVOs demonstrate reduced SI_preVWI, a lower EI, and a decreased lumen diameter, which may be associated with increased calcification and fibrosis within the occluded segment. A diagram illustrating the biological changes in occluded vessels over time is provided in Figure S6.

The evaluation of imaging characteristics is crucial, particularly in the context of endovascular recanalization for nonacute ILVOs. ¹⁹ , ²⁰ Factors influencing the success of this procedure include retrieval methodology, operator experience, anatomical features, vessel type, and notably, the duration of occlusion. Prolonged occlusion is associated with reduced rates of successful recanalization and may necessitate the use of additional stents. ²¹ Thrombus fibrosis, which develops over time, reduces the likelihood of microguidewire passage, especially when encountering calcified plaques, thereby further impacting recanalization success. ⁶ , ²² Kitano et al ²³ reported that older thrombi were associated with longer puncture‐to‐reperfusion times and poorer functional outcomes compared with fresh thrombi. Imaging biomarkers indicative of occlusion duration emerge as essential tools for predicting recanalization success, offering the potential for personalized and effective treatment strategies.

In contrast to stenotic lesions, the distal blood vessels of the occluded artery are not visible, and the characteristics of the occluded segment are challenging to determine. Our study demonstrates that HR‐VWI‐defined features play a significant role in differentiating subacute and short‐duration chronic from long‐duration chronic occlusions. Ex vivo studies with atherosclerotic plaques from the middle cerebral artery confirm HR‐VWI's contrasting capabilities in delineating various atherosclerotic components. ²⁴ , ²⁵ This imaging modality provides detailed visualization of the occluded segment, including occlusion length, angle, SI, and enhancement characteristics. ¹² Furthermore, HR‐VWI has demonstrated effectiveness in guiding endovascular recanalization for nonacute large vessel occlusion, offering predictive value for successful outcomes. Hou et al ²⁶ reported that residual luminal patency and shorter occlusion length on HR‐VWI were predictors of successful endovascular recanalization. Additionally, Hou et al ²⁷ noted that HR‐VWI can identify true tandem occlusions in patients with absent magnetic resonance angiography signals, compared with digital subtraction angiography, and help select better candidates for recanalization therapy.

Despite the valuable insights provided, this study has certain limitations. It is retrospective and single‐centered, with a relatively small sample size, especially among patients undergoing endovascular treatment. A prospective, longitudinal magnetic resonance imaging study is necessary to comprehensively capture changes in imaging characteristics over time and to clarify their association with successful recanalization.

CONCLUSIONS

In a comprehensive exploration of nonacute ILVOs, this study systematically delved into both traditional and histogram characteristics across different durations of occlusion. The results significantly enhance our understanding of the complexities inherent in nonacute ILVOs, potentially refining our interpretation of these occlusions. The HR‐VWI technique, leveraging its intrinsic advantages, emerges as a promising and essential tool for optimizing endovascular treatment of nonacute ILVOs. Insights from this study may lead to improved diagnostic accuracy and personalized therapeutic strategies, highlighting the potential transformative impact of HR‐VWI on the management of nonacute ILVOs.

Sources of Funding

This study was supported, in part, by the following sources: the Natural Scientific Foundation of China (82171916), the Natural Scientific Foundation of Tianjin (22JCYBJC00520), and the Tianjin Research Project (TJWJ2023QN058). Dr Cao is a postdoctoral researcher jointly trained in Tianjin First Central Hospital and Nankai University.

Disclosures

None.

Supporting information

Data S1 Supplemental Methods

Tables S1–S4

Figures S1–S6