Clavicle-induced narrowing of the thoracic outlet during shoulder abduction as imaged by computed tomographic angiography and enhanced by three-dimensional reformation.

OBJECTIVE: This study aimed to confirm the location and degree of compromise of the subclavian vessels within the thoracic outlet during ipsilateral arm abduction in patients with clinical evidence of thoracic outlet syndrome and to identify both the physical and physiologic source of neurovascular compromise that induces the symptoms of thoracic outlet syndrome.

DESIGN: After a neuromuscular and vascular examination, all of the subjects underwent a two-part high-resolution computed tomographic angiography with three-dimensional reformation. The initial study was performed with the arm held at the side in an anatomical neutral position. Subsequently, the arm was abducted to 90 degrees with external rotation (ABER). In each position, 60 ml of iodinated nonionic contrast medium was injected in the opposite arm at 4 ml/sec. Three-dimensional volume-rendered images were obtained. Each image was subsequently reviewed by a musculoskeletal radiologist (S. Yadavalli). Patients were initially evaluated in the physiatrist's private office (M.M. LaBan). The computed tomographic scans were obtained from the participants as outpatients in an academic community-based medical center (William Beaumont Hospital). Seventeen outpatients with clinical signs and symptoms of thoracic outlet syndrome were evaluated, including seven men and ten women. This group has an average age of 48 yrs (range, 17-73 yrs).

RESULTS: The level of vessel occlusion varied in the costoclavicular space as well as in demonstrating the alterations in the diameter of both the subclavian artery and vein both in the neutral and ABER positions. The possible levels of occlusions included the costoclavicular space, the interscalene triangle, and the retropectoralis minor space. The narrowing of the subclavian vessel was considered significant if the percentage change of the vessel's diameter between the neutral and the ABER positions was 30% or greater for the subclavian artery and 50% or greater for the subclavian vein.

CONCLUSIONS: The average change in the costoclavicular space between the neutral and ABER positions was 18.2 mm or 55.6%. The degree of subclavian artery occlusion was significant in 8 (47%) of the 17 patients. The average change in artery diameter was 28% (5.5 to 7.5 mm). Significant subclavian vein occlusion was present in 12 (75%) of 16 patients. The average change in venous diameter was 54.1% (5.7 to 12.6 mm). In two cases, venous occlusion occurred in the retropectoralis minor space, one of which was significant at 79%. The vast majority of patients, that is, 13 (76.5%) of 17, demonstrated a compression of either the subclavian vein or artery, whereas 6 (35.3%) of 17 demonstrated a compression in both. In each of these cases, the asymptomatic side failed to demonstrate a significant change in either the venous and/or arterial caliber.