Late neuro events in asymptomatic vs symptomatic isolated subclavian steal

Objective: Natural history of subclavian steal (SS) is not well defined. This is the largest study to date, to analyze late neuro events of patients with asymptomatic (asx) vs symptomatic (sx) isolated SS in an over 5-year period. Methods: SS detected during 11,762 routine carotid duplex ultrasound exams (CDUEs) done over a 2-year period were analyzed. Late neuro events, including transient ischemic attack/stroke (hemispheric anterior cerebral vs posterior stroke), were recorded for isolated SS (no associated carotid disease). Kaplan-Meier analysis for freedom from stroke for isolated sx (posterior circulation symptom/arm claudication) and isolated asx patients with SS and for patients with combined SS and carotid disease was done. Results: A total of 285 SS were detected during 11,762 CDUEs; 191 had retrograde vertebral flow (complete steal), and 94 had bidirectional flow (partial steal), 64% on left and 33% on right. Indications of CDUE testing included post carotid endarterectomy/carotid artery stenting surveillance (27%), surveillance of carotid stenosis (45%), asx carotid screening/bruit (8%), hemispheric neuro symptoms (9%), and arm claudication/posterior circulation symptoms (12%). One hundred fifteen (40%) had isolated SS, and 60% were associated with >70% carotid stenosis. The mean systolic arm pressure differential was 32.4 mmHg in the whole series, and it was 31.8 vs 34.4 for asx vs sx patients (P = .537). Mean systolic pressure for patients with retrograde flow was 105 mmHg vs 129 mmHg for bidirectional flow and 126.8 for normal antegrade flow (P < .0001). Fifty-one interventions were done for sx patients: 27 subclavian percutaneous transluminal angioplasties/stents, 5 carotid subclavian bypasses (for claudication/posterior circulation symptoms), and 19 carotid endarterectomies for associated carotid disease. At a mean follow-up of 34 months (range, 1-79 months), 10 late strokes were noted in the whole group; however, in 115 patients with isolated SS (85 were initially asx and 30 sx), there was one transient ischemic attack and two late strokes in the asx SS group: both were not posterior strokes (one lacunar hypertensive stroke and one cardiac emboli), and there were no strokes in the sx SS group. Kaplan-Meier analysis (including patients with carotid artery disease) showed the rates of freedom from stroke at 1, 3, and 5 years were 99%, 97%, and 97% for the whole series. The rates of freedom from stroke in the isolated SS group at 1, 3, and 5 years were 100%, 97%, and 97% in the whole series; for patients with no symptoms (posterior circulation or arm claudication), they were 100%, 96%, and 96%, and for patients with posterior circulation symptoms and/or claudication, they were 100%, 100%, and 100%, respectively. Conclusions: SS in patients undergoing CDUE is rare (2%), and the majority were asymptomatic. Isolated SS had a relatively benign course with no posterior circulation strokes up to 5 years.