Clinical Progression in Patients With Ischemic Stroke Associated With Intracranial Vertebrobasilar Artery Dissection

Article information

J Stroke. 2025;27(3):409-412

Publication date (electronic) : 2025 July 15

doi : https://doi.org/10.5853/jos.2024.05638

Hung-Yu Liu^,1,2, Feng-Chi Chang^2,3, Shih-An Tang¹, Ronda Lun⁴, Chih-Ping Chung^1,2

¹Department of Neurology, Neurological Institute, Taipei Veterans General Hospital, Taipei, Taiwan

²School of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan

³Department of Radiology, Taipei Veterans General Hospital, Taipei, Taiwan

⁴Department of Neurology, Stanford Hospital, Palo Alto, CA, USA

Correspondence: Hung-Yu Liu Department of Neurology, Neurological Institute, Taipei Veterans General Hospital, No. 201, Sec. 2 Shih-Pai Rd, Taipei, Taiwan Tel: +886-2-2871-2121#7578 E-mail: christie393@yahoo.com.tw

Received 2024 December 25; Revised 2025 February 26; Accepted 2025 March 17.

Dear Sir:

Intracranial vertebrobasilar artery dissection (iVBD) is a rare cause of posterior circulation stroke [1] and is associated with a poorer prognosis compared to extracranial vertebral artery (VA) dissection [2,3]. Due to the limited number of studies on iVBD associated with ischemic stroke (iVBDIS), both the frequency and long-term effects of stroke progression remain unclear. Moreover, an optimal strategy for treating iVBDIS has yet to be established [3,4]. Given the potential rapid vascular changes and diverse stroke mechanisms in dissection [5], patients with iVBDIS may be prone to clinical progression. The identification of high-risk subgroups and providing aggressive treatment may prevent adverse consequences.

This study investigated the frequency, impact, and clinical and imaging predictors of progression in patients with iVBDIS. Diagnosis of iVBD requires at least one of the following characteristics of the intracranial VA or basilar artery (BA) on magnetic resonance imaging (MRI): (1) intramural hematoma demonstrated on T1-weighted MRI or high-resolution vessel wall imaging; (2) intimal flap; (3) double-lumen sign; and (4) pearl and string sign or string sign on magnetic resonance angiography (MRA) with no other identified etiologies, such as atherosclerotic plaque, vasculitis, or vasospasm, or showing rapid temporal changes in vascular shape. At Taipei Veterans General Hospital (VGHTPE), we use high-resolution vessel wall MRI to determine the etiology of stroke when iVBD is suspected and other causes cannot be ruled out. Extracranial VA dissection extending to the intracranial portion was classified as iVBD in this study, and the patient selection process is illustrated in Figure 1. Two stroke specialists, a neurologist and neuroradiologist, reviewed the neuroimages and reached a consensus on the final study of patients with iVBDIS. We excluded patients with ruptured iVBD and subarachnoid hemorrhage, asymptomatic dissection unrelated to the current stroke, extracranial VA dissection without intracranial involvement, and those with insufficient clinical information at follow-up. This study was approved by the Institutional Review Board of VGHTPE (2021-04-007AC).

Figure 1.

Patient selection process for intracranial vertebrobasilar artery dissection with ischemic stroke (iVBDIS). MRI, magnetic resonance imaging; VA, vertebral artery; BA, basilar artery.

A total of 128 patients with iVBDIS were included, of whom 81 (63.3%) underwent high-resolution vessel wall MRI to confirm the etiology. Infarct locations were categorized as spinal cord, cerebellum (unilateral/bilateral), medulla, pons, midbrain, thalamus, and temporo-occipital lobes supplied by the posterior cerebral artery territory, with multiple regional infarcts involving combinations of these areas. Clinical progression was defined as rapid neurological deterioration, indicated by an increase of ≥4 points on the National Institutes of Health Stroke Scale (NIHSS) or the emergence of new neurological symptoms or signs attributed to iVBD, within 1 month of stroke onset. Functional outcomes were followed up through structured telephone interviews with senior stroke case managers at 1, 3, 6, and 12 months.

At baseline, 71.1% of patients had mild stroke (NIHSS ≤4), 21.9% had moderate stroke (NIHSS 5–15), and 7% had severe to very severe stroke (NIHSS ≥16). Unilateral VA dissection without BA involvement (50.8%) was the most common MRA feature at diagnosis, followed by VA dissection with BA involvement (35.2%). The infarcts mainly affected the cerebellum (50.8%), medulla (35.9%), and pons (30.5%), with 44.5% involving multiple regions and 3.1% affecting the spinal cord (Table 1). Most patients (95.3%) were administered antiplatelet agents. Thirty-four patients received dual antiplatelets and 25 received anticoagulants, either used before or added after progression.

Table 1.

Clinical and imaging features in patients with iVBDIS with and without clinical progression

Clinical progression occurred in 32 (25%) patients, with 19 (14.8%) showing early progression within 1 week. Among the 32 patients with clinical progression, 21 had new infarcts on follow-up imaging consistent with their symptoms, while two had no new infarcts. In the remaining nine patients, four experienced early stroke progression and underwent MRI after deterioration, while five lacked follow-up imaging. Twenty-two patients with clinical progression underwent follow-up vascular imaging, and 13 showed dissection progression. Supplementary Figure 1 shows a representative case of clinical progression.

A total of 87 patients underwent at least one follow-up vascular imaging within 1 year. The duration ranged from 1 to 364 days, with a median of 30 days, and an interquartile range of 7–175 days. Twenty patients showed dissection progression, including extension, progressive narrowing, vessel occlusion, or development of a new dissecting aneurysm; 12 showed regression; and 55 showed no significant changes. Dissection progression was linked to clinical progression compared to regression or no change (65% vs. 11.9%, P<0.001). Nine of the 128 patients underwent stenting: three for clinical progression, one for recurrent transient ischemic attacks, one for recurrent stroke after 2 months, one to prevent dissecting aneurysm rupture, and three for prophylactic stenting to preserve BA flow. During the follow-up period, none of the patients experienced dissection-associated hemorrhage.

Progression was a predictor of poor 1-year functional outcome (modified Rankin Scale [mRS] score ≥3) after adjusting for age, sex, and initial NIHSS (adjusted odds ratio [aOR] 10.1, 95% confidence interval [CI]: 3.1–34.9, P<0.001). Patients who were older, had VA dissection with BA involvement, and had infarcts involving the bilateral cerebellum, thalamus, spinal cord, or multiple regions were more likely to experience clinical progression (P<0.05). In multivariable analysis, patients aged over 65 years (aOR 3.29 [1.24–8.75], P=0.017), spinal cord infarcts located (aOR 14.63 [1.13–190.25], P=0.040), and multiple infarcts involving the bilateral cerebellum in combination with the thalamus, pons, or temporo-occipital regions (aOR 3.48 [1.29–9.44], P=0.014) independently predicted clinical progression (Supplementary Table 1). Moreover, VA dissection involving BA independently predicted early progression within 1 week (aOR 4.17 [1.30–13.35], P=0.016).

Our study revealed that most patients with iVBDIS initially presented with mild strokes; however, one-third became functionally dependent (mRS score ≥3), with progression predicting poor functional outcomes. During the acute and subacute phases of stroke, a 25% risk of clinical progression was identified in patients with iVBDIS. These findings suggest the need for proactive and aggressive treatments to prevent deterioration in high-risk individuals.

Clinical progression in our patients was primarily due to recurrent stroke caused by initial dissection or progression. The potential mechanisms of new infarcts include dissection-associated thromboembolism, perforator artery occlusion, insufficient blood flow due to severe narrowing or occlusion of the dissecting parent artery, or a combination of these mechanisms. Previous research has identified BA extension and posterior inferior cerebellar artery involvement as predictors of progression in iVBD [6]. Our study further showed that BA extension predicted stroke progression within 1 week, highlighting the need for close monitoring and proactive interventions. Bilateral cerebellar infarcts suggest simultaneous bilateral VA dissection or dissection extending to the contralateral side of the VA or BA. Additional infarcts in the pons, thalamus, or temporo-occipital regions suggest that artery-to-artery emboli drift to the distal vascular territory or that dissection extends to the BA, causing perforator artery occlusion. These MRI infarct patterns implied that a longer, more extensive, or unrecovered dissection is associated with clinical progression. Additionally, age was a predictor of progression, potentially due to increased platelet activation associated with aging and atherosclerosis, thereby raising the risk of thrombus formation.

The limitations of this study include potential underestimation of clinical progression frequency, as posterior circulation stroke may manifest with atypical and nonspecific symptoms; underdiagnosis of iVBDIS with atypical vascular features [7], and difficulty in distinguishing intramural hematoma with coexisting atherosclerosis from intraplaque hemorrhage, although controversial cases were excluded.

In conclusion, dissection extending to the BA and MRI infarct patterns can help identify patients with iVBDIS at risk of clinical deterioration. Future trials of early aggressive medications or interventions in this target population could help determine the most effective treatment strategy for patients with iVBDIS.