Thoracic outlet: assessment with MR imaging in asymptomatic and symptomatic populations

To compare the dynamic modifications of the thoracic outlet in asymptomatic volunteers and symptomatic patients and assess the presence and location of vasculonervous compressions in these two populations.

Thirty-five healthy volunteers and 54 patients with clinical symptoms of thoracic outlet syndrome (TOS) underwent magnetic resonance (MR) imaging of the thoracic outlets with their arms alongside their bodies and after a postural maneuver. Measurements were obtained at the interscalene triangle (thickness of anterior scalene muscle, interscalene angle), at the costoclavicular space (minimum costoclavicular distance, distance between inferior border of subclavius muscle and the anterior chest wall, maximum thickness of subclavius muscle, angle between first rib shaft and horizontal), and at the retropectoralis minor space (distance between posterior border of pectoralis minor muscle and posterior lining of axilla at the passage of the axillary vessels, thickness of pectoralis minor muscle). The presence and location of vasculonervous compressions were also assessed. Group data were analyzed with the Student t test.

Patients with TOS had a smaller costoclavicular distance after the postural maneuver (P <.001), a thicker subclavius muscle in both arm positions (P <.001), and a wider retropectoralis minor space after the postural maneuver (P <.001) than did volunteers. Venous compressions after the postural maneuver were observed in 47% of volunteers and 63% of patients at the prescalene space, in 54% of volunteers and 61% of patients at the costoclavicular space, and in 27% of volunteers and 30% of patients at the retropectoralis minor space. Arterial and nervous compressions, respectively, were seen in 72% and 7% of patients. No arterial or nervous compression was seen in volunteers. Except for venous thrombosis, vasculonervous compressions were demonstrated only with arm elevation. Only three thoracic outlet measurements differed significantly in both populations.

MR imaging appeared helpful in demonstrating the location and cause of arterial or nervous compressions.