To evaluate the relationship between the type of vertebral artery steal and posterior circulation ischemia by carotid Doppler ultrasonography (CDU) in the case of moderate to severe subclavian artery stenosis or occlusion.

A total of 163 patients with moderate to severe subclavian artery stenosis or occlusion confirmed by CDU and digital subtraction angiography (DSA) at the Stroke Center of the First Affiliated Hospital of Soochow University from January 2017 to July 2020 were retrospectively enrolled. All patients underwent cranial magnetic resonance imaging (MRI) and National Institute of Health Stroke Scale (NIHSS) score evaluation. According to the type of vertebral artery steal on the affected side of patients examined by CDU, 163 patients were divided into type Ⅰ (latent type; 51 cases), type Ⅱ (partial type; 70 cases), and type Ⅲ (complete type; 42 cases) cases. CDU was used to assess the type of blood steal and compared with DSA findings to analyze the correlation between different degrees of subclavian artery stenosis and the type of vertebral artery steal, and the incidence of cerebral infarction in patients with different types of vertebral artery steal.

The results of CDU and DSA in evaluating the types of vertebral artery steal were: type Ⅰ (51 vs 0); type Ⅱ (70 vs 53); type Ⅲ (42 vs 42). Among 46 patients with moderate subclavian artery stenosis, 36 (78.3%, 36/46), 10 (21.7%, 10/46), and 0 were evaluated by CDU as having type Ⅰ, type Ⅱ, and type Ⅲ blood steal, respectively. Among 83 patients with severe stenosis of SA, 15 (18.1%, 15/83), 60 (72.3%, 60/83), and 8 (9.6%, 8/83) were evaluated by CDU as having type Ⅰ, type Ⅱ, and type Ⅲ blood steal, respectively. All 34 patients with SA occlusion were evaluated by CDU as having type Ⅲ blood steal. The degree of subclavian artery stenosis was not completely consistent, but positively correlated, with the type of vertebral artery steal (r_s=0.802, P＜0.001). All patients with type Ⅰ showed transient ischemic attack (TIA), the incidence of cerebral infarction was 0 (0/51), and the NIHSS score was <10. The incidence of cerebral infarction in type Ⅱ patients was 28.6% (20/70), and those with an NIHSS score ≥10 accounted for 4.3% (3/70). The incidence of cerebral infarction in type Ⅲ patients was 52.4% (22/42), and those with an NIHSS score ≥10 accounted for 16.7% (7/42). There was a significant difference in the incidence of cerebral infarction in type Ⅰ~Ⅲ patients (H=33.337, P＜0.001).

With the development of the type of blood steal from the affected vertebral artery, the incidence of cerebral infarction gradually increases. CDU allows for early, noninvasive, real-time, dynamical, and accurate evaluation of subclavian artery blood steal syndrome, so as to provide a basis for clinical individualized treatment.