Thoracic Outlet Syndrome and Vascular Disease of the Upper Extremity

History

Although symptoms associated with neurovascular compression of the upper extremity have been present throughout the ages, it has only been in relatively recent times that the concept of TOS has taken a unified form. Galen is attributed with the first mention of cervical ribs in the second century AD, but descriptions of cervical rib syndrome in the English literature date to the early 19th century. Sir Astley Cooper is credited with description of cervical rib syndrome (London 1821). Coote reported resection of a cervical rib in 1861, heralding the recognition of skeletal abnormalities in upper extremity pain syndromes. With time, the role of soft tissue compression by the scalene muscles came to the fore after reports by Adson and Coffey at the Mayo Clinic. This was named Scalenus Anticus (Nafziger) Syndrome by Ochsner, Gage, and Debakey in 1935, giving credit to Naffziger for the concept.

The recognition of venous thrombosis of the subclavian vein was initially described independently by Sir James Pager (London 1875) and Leopold Von Shcroetter (Vienna 1884). The designation Paget-Schroetter syndrome (PSS) was devised by Hughes (London 1948) when reviewing 320 cases of spontaneous subclavian vein thrombosis. Arterial presentations of TOS (aTOS) were likewise evolving during this period.

The unifying term TOS was used by Peet (1956) and Rob (1958) to describe neurovascular compression at the thoracic outlet. Their idea was to consider compression of neurovascular structures at the thoracic outlet as a common etiology that presented different manifestations according to those elements most severely impacted.

This concept of TOS preceded a further shift in therapeutic approach toward these maladies. Instead of identifying bony anomalies (cervical rib syndrome) or muscular compression (scalenus anticus syndrome), the concept of thoracic outlet compression emphasized the interaction of these elements.

The surgical approach to TOS evolved apace with these conceptual developments. The cervical rib syndrome was managed by cervical rib resection. This was first reported by Coote in 1861. Scalenus anticus syndrome was managed with scalenectomy. TOS led to the adoption of first rib resection.

Murphy performed a first rib resection in 1908, but concerns with nerve complications led to its abandonment until it was reintroduced by Clagett in 1962. He applied his experience with thoracoplasty done for treatment of tuberculosis (TB) to a posterior approach to first rib resection. This was a technically difficult operation with considerable morbidity. The description of the transaxillary approach by Roos in 1966 presented a dramatically superior approach. Transaxillary surgery had far less morbidity than the posterior rib resection and was subsequently widely adopted. The principal problem was that it was a technically challenging operation.

The supraclavicular approach to first rib resection experienced a surge of interest because it combined supraclavicular surgical exposure and scalenectomy, which was more familiar to vascular surgeons, with partial removal of the posterior aspect of the first rib. Subsequent evolution of the supraclavicular approach has led to inclusion of a secondary infraclavicular incision to allow further resection of the anterior portion of the first rib.

Anatomy

The skeletal thoracic outlet is that space defined by the first ribs, sternum, and spine circumscribing the top of the thoracic cage. It forms the floor to which are attached the muscular elements of the thoracic outlet. The clavicle forms a superior boundary, riding over the neurovascular structures. The muscular elements include the anterior scalene, middle scalene, and subclavius muscles ( Fig. 33.1 ).

Taken together these structures give rise to two areas of potential compression: the scalene triangle and the costoclavicular space . The scalene triangle is formed by the anterior scalene muscle, the middle scalene muscle, and the first rib. Through this triangle pass the brachial plexus and the subclavian artery. The anatomic configuration where the brachial plexus and subclavian artery travel through the scalene triangle accounts for some of the confusion in defining symptoms of TOS. Any compression of the brachial plexus will also result in some compression of the subclavian artery. Consequently, arterial and neurogenic TOS (nTOS) are at times confused. The costoclavicular space is formed by the first rib and the clavicle. These form an open-ended triangle that joins at the hinged junction of these two bones. The subclavian vein crosses through the apex of this triangle and may be subject to compression at this location.

The anterior scalene muscle is unique in separating the subclavian artery and vein: the subclavian vein travels outside the scalene triangle in a venous channel bordered anteriorly by the subclavius muscle and posteriorly by the anterior scalene muscle. The anterior scalene muscle forms the anterior border of the scalene triangle. The phrenic nerve courses on the anterior aspect of the anterior scalene muscle. It descends from the superior lateral border of the muscle and crosses toward the inferior medial portion of the scalene and enters the chest behind the subclavian vein.

The middle scalene muscle forms the posterior border of the scalene triangle. The anterior scalene muscle is innervated by the C5 and C6 cervical spinal nerves. It originates from the anterior transverse processes of C3 to C6 and inserts onto the first rib on the scalene tubercle. The middle scalene muscle is the largest of the scalene muscles. It is innervated by nerves arising from nerve roots C4 to C6. The middle scalene originates from the transverse processes of the C2 to C7 cervical vertebrae. The muscle occupies a broad insertion on the dorsal aspect of the first rib from the posterior aspect of the rib extending toward the mid portion of the rib. The dorsal scapular, suprascapular, and long thoracic nerves traverse the body of the middle scalene muscle. In this location, they are at risk of injury in the course of middle scalene resection and supraclavicular rib resection.

Both scalene muscles are important elements in the genesis of thoracic outlet compression because spasm and contraction of these muscles will result in elevation of the first rib and muscular compression of the structures traversing the scalene triangle.

The scalenus minimus is a small muscle that most often arises from the middle scalene muscle, traverses between the nerve roots, and inserts on the first rib and thickened dome of the pleura (Sibson fascia). As such, it is a muscular sling that envelops the lower trunks of the brachial plexus and may contribute to genesis of symptoms.

The subclavian vein travels in a venous channel that is bordered anteriorly by the subclavius muscle tendon and posteriorly by the middle scalene muscle . The subclavius muscle arises from the undersurface of the clavicle and inserts on the first rib in front of the costoclavicular ligament. It is innervated by a nerve to the subclavius arising from nerve roots C5 and C6.

The brachial plexus arises from nerve roots C5 to C8, and T1. These roots then form three trunks (upper [C5 and C6], middle [C7] and lower [C8 and T1]). The trunks then intermix to form divisions (three anterior and three posterior), which are intermediaries to the formation of cords. The cords (posterior, medial, and lateral) then lead to the terminal nerves, which supply the arm (median, radial, ulnar, and musculocutaneous).

Several important nerve branches arising from the brachial plexus traverse this space: among these are the phrenic nerve, the long thoracic nerve, the dorsal scapular nerve, the suprascapular nerve, and the thoracodorsal nerve.

The suprascapular nerve arises from nerve roots C5 and C6 and innervates the supraspinatus and infraspinatus muscles. It courses through the body of the middle scalene muscle. The supraspinatus muscle forms part of the rotator cuff and abducts the arm at the shoulder. The infraspinatus muscle forms part of the rotator cuff of the shoulder and serves to externally rotate the arm and stabilize the shoulder.

The dorsal scapular nerve arises from the C5 nerve root. It courses through the body of the middle scalene muscle and innervates the rhomboid and levator scapulae. The rhomboid muscles pull the scapular toward the midline. The levator scapula elevates the scapula and lifts the scapula toward the head.

The phrenic nerve arises from nerve roots C3 to C5 and innervates the diaphragm. The nerve arises on the superior lateral border of the anterior scalene muscle, courses across the front of the anterior scalene muscle, and reaches the medial inferior border of the anterior scalene. Loss of function will result in paralysis of the hemidiaphragm. Most patients are quite symptomatic with loss of the hemidiaphragmatic function, making this an easily identified nerve injury. In about 13% of the population the phrenic nerve forms two separate bundles, which course in parallel across the anterior scalene muscle.

The long thoracic nerve arises from nerve roots C5 to C7, travels through the body of the middle scalene muscle, and innervates the serratus anterior muscle. Damage to this nerve results in instability of the scapula when the arm is raised or used to push. This defect is noted with the medial border of the scapula pushing backward and is called a winged scapula. Because of the prominence of this defect, it is easy to identify.

The thoracodorsal nerve arises from nerve roots C6 to C8, and innervates the latissimus dorsi muscle. The latissimus dorsi muscle is a broad muscle that attaches from spine to humerus and serves to extend, abduct, and stabilize the shoulder. The latissimus dorsi may occasionally give rise to the axillary arch—a slip of the muscle that may cross in front of the axillary vessels resulting in neurovascular compression.

The following are additional structures of importance in upper extremity neurovascular syndromes:

The pectoralis minor muscle arises from the chest wall over the anterior aspect of ribs 3 to 5. It inserts on the coracoid process in the undersurface of the scapula. Along its course it crosses over the neurovascular bundle as this exits the thoracic outlet and enters the arm. In this manner it may generate symptoms of nerve or vascular compression in the arm, giving rise to pectoralis minor syndrome.

The quadrilateral space is a muscular space bordered by the heads of the teres major and teres minor muscles, as well as the triceps muscle and the neck of the humerus. Coursing through this space is the axillary nerve and the posterior humeral circumflex artery. The posterior humeral circumflex artery is subject to injury at this location resulting in a well-documented thromboembolic syndrome.

Anatomical Variants

Many TOS presentations are associated with acquired or developmental anomalies of the muscular elements within the thoracic outlet. Hypertrophy of the scalene and subclavius muscles is associated with venous TOS (vTOS). Overdevelopment or chronic spams of the anterior and middle scalene muscles may be associated with arterial and nTOS.

The most clinically significant variations in thoracic outlet structure consist of alterations in the size and insertion of the muscles, variations in the number and shape of the bony elements, and presence of tendinous bands. David Roos described nine variants, which he termed bands. These include varying attachments of cervical ribs, scalene muscles, and ligaments onto the pleura and first rib. Machleder recorded anatomical variants in 200 consecutive cases. Variations in scalene muscle development or insertion accounted for 43% of cases. Anomalies of the subclavian muscle and tendon accounted for 19%, and an accessory scalene minimus muscle was noted in 10%. The scalenus minimus muscle is an accessory muscle that separates the brachial plexus and subclavian artery. It most often arises as a fascicle of the middle scalene muscle and traverses the lower roots of the brachial plexus. The net effect is that it may form a muscular tendinous band that essentially encircles the lower brachial plexus roots ( Fig. 33.2 ).

Cervical ribs are a well-recognized anatomical variant associated with TOS. The cervical rib is a supernumerary rib that arises from the transverse process of the C7 vertebral body. The presence of cervical ribs is noted in about 0.5% to 1% of the population. These are noted to occur in women with twice the frequency as with men and are bilateral in almost half of the cases. Cervical ribs arise as a developmental anomaly in early embryogenic development. In the course of body segmentation, early embryogenic cervical ribs C5 to C7 will normally regress as the nerve roots of the brachial plexus form. If this process does not occur, or if it occurs with incomplete regression, then a cervical rib remains. There is a considerable variety in the formation of cervical ribs, from fully ossified ribs joining with the sternal border to shortened partial ribs with fibrous bands completing their course to the first rib, to an elongated transverse process of the C7 vertebral body associated with a fibrous band. The Gruber classification is a simple system that organizes cervical ribs according to size: less than 2.5 cm (class I), greater than 2.5 cm (class II), and junction of a full-sized rib with the first rib: fibrous union (class III; Fig. 33.3 ) or full articular junction (class IV; Fig. 33.4 ). Although as many as 45% of people with cervical ribs are asymptomatic, cervical ribs are associated with arterial and nTOS symptoms. The presence of a cervical rib reduces the space within the scalene triangle. Partially formed cervical ribs are often associated with fibrous bands that constrict the lower trunks of the brachial plexus.

A developmental consequence of cervical rib persistence is noted in abnormal first rib growth. In cases with fully formed cervical ribs, the first rib will simulate a normal second rib in its growth: it will reside more laterally and may have a form similar to the normal second rib. Variations of this effect are seen with the various permutations of cervical ribs: smaller cervical ribs or an elongated transverse process may not alter the first rib development. Intermediate-sized cervical ribs will have an intermediate effect.

Neurogenic Thoracic Outlet Syndrome

nTOS is the symptomatic presentation of compression of elements of the brachial plexus as they cross the thoracic outlet. Cardinal symptoms include pain, paresthesia (numbness, tingling, and alteration of sensation). Weakness and loss of dexterity are also common. When compression at the thoracic outlet affects the sympathetic fibers, color changes, temperature instability (mostly coldness), and mottling are noted. It should be noted that some of these symptoms are strongly associated with arterial insufficiency, but in this instance arise from nerve compression. Atrophy of the muscles of the hand and (less commonly) the forearm are noted in cases where motor denervation results from nerve compression.

Venous Thoracic Outlet Syndrome

vTOS commonly arises from compression of the subclavian vein as it crosses between the clavicle and first rib, bound by the anterior and middle scalene muscles. The most common of these presentations is that of acute subclavian vein thrombosis. In addition to acute thrombosis, some presentations include acute chronic thrombosis, chronic obstructive venous congestion, and nonthrombotic positional intermittent venous occlusion (McLeery syndrome).

Arterial Thoracic Outlet Syndrome

aTOS are a result of chronic compression with damage to the subclavian artery. These presentations include aneurysmal degeneration, acute and chronic embolization to the upper extremity, and arterial thrombosis. Intermittent positional compression, without evidence of arterial thrombosis or embolization (which may occur when a patient raises their arm overhead), is not considered an arterial presentation of TOS.

Neurogenic Thoracic Outlet Syndrome

The typical presentations of nTOS include numbness, tingling, weakness, and pain in the upper extremity. These are usually along the distribution of the nerve trunks. The symptoms tend to course along the ulnar or radial aspect of the limb with involvement of the hand and fingers.

In addition to paresthesia, sensation of coldness and color changes are often noted in the affected hand. Headaches are a frequent accompanying symptom. There seems to be an association of migraine where the development of nTOS will exacerbate migraine.

Atrophy of the musculature of the hand is a relatively unusual but well-recognized presentation of nTOS. When atrophy is associated with presence of a cervical rib and shows evidence of denervation on electromyography (EMG) the presentation is sometimes termed the Gilliatt-Sumner (1970) hand.

Core symptoms of nTOS include pain or paresthesia from the base of the neck to the fingers. In addition to these, many patients will present with associated symptoms including migraine, ocular, ear, facial, parascapular, pectoral, and axillary symptoms. A further subdivision of symptoms into upper plexus and lower plexus presentations is based on the predominance of radial or ulnar symptom distribution. This distinction is sometimes used to direct the choice of operations.

Atypical presentations may include incomplete presentations as well as presentation with unusual symptoms. Incomplete presentations might be pain and paresthesia limited to the neck and shoulders but not including the arm and hand. Unusual presentations may include chest pain with repeated emergency room evaluations for angina along with pain or paresthesia in the arm.

Diagnosis of Neurogenic Thoracic Outlet Syndrome

TOS is a clinical syndrome. The diagnosis of TOS is based on the summation of patient symptoms and the findings of physical examination. The diagnosis does not rest on any one pathognomonic symptom or finding. The diagnosis may be supported by appropriate testing. Testing is directed toward evaluating both the diagnosis of TOS and the competing differential diagnosis. There is no test that is considered a gold standard. There is no single finding on exam that is clearly diagnostic. Clinical acumen and appreciation of diagnostic patterns is probably most important in identifying patients who have TOS. Testing for TOS needs to be individualized—not all patients require all tests. In addition to supporting the diagnosis, testing may also provide insight into the likely result of treatment.

Patients presenting with nTOS represent the most complex group to diagnose. This is partially due to the overlap of symptoms and exam findings between nTOS and other neurocompressive syndromes affecting the upper extremity. The first step in diagnosis is to recognize elements of a typical presentation. The typical symptoms of nTOS include pain, numbness, tingling, and weakness in the upper extremity. The pain and paresthesia tend to be centered over the brachial plexus in the base of the neck and will then radiate down the arm. In many instances there may be symptoms in the upper trapezius, parascapular area, chest wall, and the side of the head. In a minority of cases, atrophy of the musculature of the hand is a prominent feature.

Differential Diagnosis

The differential diagnosis of nTOS includes any cervical spine disease that may result in radicular symptoms: spondylolisthesis, disk herniation, foraminal stenosis, and ostephyte formation. In addition, peripheral nerve compression syndromes such as carpal tunnel, cubital tunnel, Guyon tunnel, and radial tunnel compression may result in neuralgia and should be considered in the differential diagnosis. Other disease processes that are characterized by pain including fibromyalgia and myofascial pain syndromes should be considered. Shoulder pathology may cause positional triggering of symptoms but would not be expected to result in pain or paresthesia radiating down the extremity along a neural pathway.

Physical Examination

Physical evaluation requires a complete neurovascular exam of the upper extremity. This should include evaluation of sensory function, motor function, and reflexes. The next step is evaluation of sensitivities over the compressive sites by palpation and percussion. The hands should be assessed for strength and muscular atrophy. Function of the opponens muscles in the thumb and fifth fingers, as well as grip strength and interosseous function are routinely done.

Percussion over the nerves eliciting radiating pain or paresthesia is referred to as Tinel sign and is an indication of irritability of a given nerve. At minimum, Tinel testing is performed over the Erb point (at the anterior base of the neck over the scalene muscle), over the ulnar nerve at the elbow as it courses through the olecranon groove, and over the median nerve at the wrist where it travels through the carpal tunnel. Additional points of evaluation may include the radial tunnel in the forearm or Guyon canal at the wrist, if clinically indicated.

In addition, palpation and percussion should evaluate tenderness over the pectoralis muscle. If pressure over the pectoral muscle results in pain in the muscle along with pain or paresthesia down the arm, then pectoralis minor syndrome should be considered.

Provocative examination includes maneuvers that use different positions of the arm to elicit positional compression of arteries or to place tension on the nerves and elicit positional exacerbation of symptoms. These tests include the Adson maneuver, the abduction external rotation (AER) test, the costoclavicular compression test, the elevated arm stress test (EAST), the upper extremity limb tension test (UELTT), and the Wright hyperextension test.

The Adson test consists of abduction of the limb with head turning, deep breath holding, and assessment of radial pulse. Turning the head and taking a deep breath engages the scalene muscles resulting in loss of pulse. In a positive test the pulse is lost and symptoms may be reproduced.

The AER test raises the arm laterally to the level of the shoulder or above while assessing the radial pulse and auscultating for subclavian bruit. A positive test includes loss or severe reduction of pulse, development of a bruit, and pallor of the hand.

The costoclavicular compression test, or military brace test, places arms at the side while retracting the shoulders and extending the neck. The radial pulse is assessed. Loss of pulse and reproduction of symptoms indicates a positive test.

Positional loss of pulse must be evaluated in the context of presenting symptoms. Almost 50% of the normal population will demonstrate positional loss of radial pulse. The positional loss of pulse is significant only in those patients who have nTOS symptoms. The rationale for these tests is recognition that the brachial plexus and subclavian arteries are in immediate proximity to one another as they course through the scalene triangle. Thus the positional loss of pulse is seen as a proxy for nerve compression.

The EAST requires the patient to raise the arms above the head with the arms at shoulder level, then open and close the hands repeatedly. This is continued for 1 to 3 minutes while recording symptoms. A positive test includes reproduction of radiating symptoms in the arms and hands. Pallor of the hands is often noted with hyperemia when the arms are lowered.

The UELTT is designed to place gradually increased stretch on the nerves in the upper extremity to try to reproduce symptoms of pain or paresthesia radiating down the arm. This is performed by having the patient extend the arms out to the sides at shoulder level. The next step is to have the patient dorsiflex the hands. The third step is to have the patient lean their head from side to side. A typically positive test will rapidly result in pain or paresthesia down the arm.

The Wright hyperextension test is performed by having the patient reach both arms downward behind the torso. This places the pectoralis minor muscle on tension and results in accentuating compression of the neurovascular bundle as it courses down the arm. A positive test will result in pain in the pectoral region and pain or paresthesia down the arm.

In addition to the above noted tests, evaluation of cervical spine disease is done with rotation, extension, and flexion of the neck with evaluation of any resulting symptoms. The Spurling maneuver is an additional test done to assess possible nerve root compression. This is done by exerting downward pressure on the patient's head while leaning the head slightly toward each side. A positive result is that of pain or paresthesia down the arm.

Evaluation of these tests may show sensitivity over the thoracic outlet and not in the arm and thus may be helpful in supporting the diagnosis of nTOS. As noted before, there may be considerable overlap of nerve compression syndromes of different source. Once a nerve is irritated at one point, it may be irritated along the entire course, thus making differentiation of a peripheral nerve compression and TOS clinically difficult. Finally, not all presentations are typical. Some patients may have pain in the neck, shoulder, and upper arm without continuation to the hand and fingers.

Testing

Testing as part of evaluation of nTOS serves two purposes: to identify concurrent disease processes that may mimic nTOS presentations and to support the diagnosis of nTOS.

Tests that are done to assess possible concurrent problems include cervical spine magnetic resonance imaging (MRI) and peripheral nerve conduction and EMG testing.

Testing that supports the diagnosis of nTOS includes imaging studies: x-ray evaluation for presence of cervical ribs and abnormal first ribs, and MRI to assess anatomical evidence of nerve and vascular compression. Additional testing that may confirm a diagnosis of nTOS includes electrophysiologic testing and scalene muscle blocks.

Imaging

Cervical x-ray imaging is fundamental to evaluation of TOS patients. Chest x-rays should be reviewed to assess the presence of cervical ribs and abnormal first ribs. If there is uncertainty, a cervical spine series should be done in addition to the chest x-rays. Chest or cervical films taken with a 15-degree apical-lordotic view will offset the cervical rib and make it more readily apparent.

Imaging used for evaluation of nTOS has included MRI, duplex ultrasound, and catheter-based arteriography. MRI imaging is popularly considered to be a definitive modality to assess most neurological maladies. In the instance of nTOS, however, its promise remains unfulfilled. In general, TOS MRI studies have not been standardized across the nation. Accordingly, there is a wide variation in how these tests are performed and evaluated. The most complete evaluations will include a magnetic resonance arteriogram (MRA) and a magnetic resonance venogram (MRV) along with MRI of the brachial plexus. Careful examiners will note the course of the brachial plexus structures, the presence of edema within the nerves, and anatomical abnormalities. Compression of the subclavian artery or vein may be documented. Further refinement includes measurement of muscle sizes (anterior scalene, middle scalene, subclavius), and measurements of the distances between structures within the thoracic outlet. Because of the variation in how these exams are conducted, these tests may not prove helpful unless performed at a center with a dedicated interest in diagnosis of TOS.

Catheter-based imaging has been used as a proxy to identify neurological compression. Past experience shows this to be an unreliable means of identifying nTOS. Catheter-based imaging does not establish a diagnosis of nTOS. In the absence of evidence suggesting arterial thromboembolic disease or venous thrombosis, it serves no diagnostic role ( Fig. 33.5 ).

Similarly, arterial duplex ultrasonography is not a reliable means of establishing the diagnosis of nTOS. Although it may identify positional loss or reduction of arterial signals, this may also be seen in normal asymptomatic subjects and so is not considered sufficiently specific to be diagnostic.

Electrophysiologic testing may provide evidence of nTOS. Indirect evidence of compression at the thoracic outlet may be provided by findings of delayed antidromic f-waves. Although the presence of positive findings among these tests does not exclude TOS, the presence of coexisting problems needs to be taken into account in formulating a treatment plan.

Somatosensory evoked potential (SSEP) testing has been shown to have considerable specificity in the diagnosis of nTOS. Unfortunately, the sensitivity of this test may limit the number of patients diagnosed. The test involves using a peripheral stimulating electrode on the hand and recording electrodes placed over the Erb point, the cervical spine, and the contralateral cerebral cortex. The test is done to assess both median and ulnar conduction both with the arm at the side and with the arm in a stress position. Evaluation includes measurement of conduction velocity and amplitude across the thoracic outlet. In a review of experience at the University of California–Los Angeles (UCLA), Machleder and Newer reported return of normal amplitude and latency in up to 92% of patients following surgical relief of symptoms.

More recently, assessment of median antebrachial cutaneous sensory nerve action potentials (MAC SNP) has generated interest. This test, evaluating a branch of the median nerve, has been used to help establish the diagnosis of TOS. As with all nerve conduction testing, there are significant limitations in sensitivity and specificity. Also, familiarity with the testing methodology is not widespread and so the exam is not generally available.

The anterior scalene muscle block is the only test that relates the patient's symptoms to the dynamic anatomy at the thoracic outlet. It is the only test that associates a test intervention with the effect on the patient's symptoms. With this test, a local anesthetic such as lidocaine or Marcaine is injected into the anterior scalene muscle and the patient is asked to report the impact of the injection on upper extremity symptoms. It is important that the test be performed with the assistance of imaging such as ultrasound, fluoroscopy, or computed tomography (CT) to accurately guide needle placement in the body of the scalene muscle. Variations in the test include addition of associated muscles to the target set (middle scalene, subclavius, upper trapezius) and the use of blinded controls (long-duration versus short-duration local anesthetics).

The methods of performing an anterior scalene muscle block have been carefully reported by Jordan and Machleder. Anesthetic blocks of the anterior scalene muscle have been used as a method of confirming TOS and predicting the likely benefit from surgical decompression. When a local anesthetic agent is injected into the anterior scalene muscle and paralyzes it temporarily, the patient's symptoms caused by compression are relieved for a few hours to a few days. However, the technique of using surface landmarks often results in inadvertent somatic and sympathetic block because there is no reliable way to verify needle tip localization. For this reason, a guided technique using electrophysiology, ultrasound, CT, or fluoroscopy is recommended.

Jordan and Machleder reported that electrophysiologic guidance facilitates accurate needle tip placement in the performance of anterior scalene muscle blocks. A Teflon-coated, 25-gauge hypodermic needle, bared at the tip, is advanced through the sternocleidomastoid muscle. Electromyographic activity is monitored as the needle is advanced through the tissue layers; the anterior scalene muscle can be activated with lateral neck bending against resistance and with deep inspiration. In most patients, a twitch of the scalene muscles is visible with electrical stimulation. The patient is asked whether pain is produced and whether insertion at this depth produces pain that is similar in quality and location to the usual pain experienced. After an injection of 2 mL of 2% lidocaine into the anterior scalene muscle, the arm is placed into a stress position and exercised for 1 minute; the patient is then asked to rate the pain. Attempts to activate the anterior scalene muscle are performed again with lateral neck movement and with deep inspiration, but only distant motor action potentials can be identified. Electrical stimulation can no longer produce a visible twitch of the scalene muscles. A positive test occurs if the patient has greater than 50% improvement in the elevated arm stress pain score after anesthetic injection of the anterior scalene muscle compared with a baseline examination. The results of these blocks correlate with surgical outcomes; a positive result predicts a good outcome to surgical decompression. Note that the absence of pain relief does not exclude the presence of TOS as long as the patient has symptoms and clinical findings, but the probability of surgical success is significantly reduced.

Treatment

nTOS is usually treated initially with physical modalities and medication. A large number of these patients will require no surgical intervention.

Initial management of nTOS should include evaluation of causative factors. Many repetitive motion injuries at work are related to poor ergonomic conditions. In the office this may be the result of ergonomically improper workstations, computers, and desks. Ergonomic workplace assessment and correction is important for such instances. Workplace restrictions may help alleviate symptoms in some. Avoidance of overhead reaching and lifting may help allow an injured limb to rest and recover.

Physical modalities are an important part of treating TOS patients. Included in this are physical therapy, acupuncture, massage, and chiropractic care. Physical therapy is believed to be helpful in about 85% of TOS patients. The fundamental goals of PT include balancing of muscle strength, stretching of the scalene muscles, and correction of posture.

Postoperative PT is helpful in the majority of patients. The postoperative PT goals include assistance with range of motion, strengthening, and posture correction.

Medication used in managing TOS patients include antiinflammatories, muscle relaxers, pain medication, and medication to reduce nerve sensitivity. In more severe and chronic situations, consultation with pain management is advised. Therapeutic scalene muscle blocs may be indicated in patients who are not interested in surgery, or for some reason cannot undergo surgery. These are done with combinations of local anesthetics (Marcaine), steroids, and antiinflammatory agents (platelet-rich plasma [PRP]). The goals are to reduce inflammation, edema, and muscle spasm.

Botox has been used with the intention of limiting the scalene muscle's ability to contract. Success rates of Botox are believed to be about 85% with a duration of about 3 to 4 months. Botox often requires ongoing administration. The number of patients whose nTOS is permanently relieved by Botox administration is not clearly established.

Surgical Goals for Neurogenic Thoracic Outlet Syndrome

Surgery for nTOS is reserved for cases that are severe and do not respond to other less invasive modalities. Physical therapy, workplace accommodation, avoidance of triggering activities, use of medication, and injections should generally precede a decision for surgery. Ultimately the decision for surgery is based on an assessment of the severity of symptoms, the degree of disability, and the likely benefit of surgery.

The surgical goals vary for each of the TOS presentations. For those patients presenting with nTOS (pain and paresthesia), the goal is to alleviate symptoms. For those whose presentation includes atrophy and weakness, the goal is to forestall progressive loss of function. The fundamental surgical goal is decompression of the neural elements as they cross the scalene triangle. To accomplish nTOS decompression, the scalene triangle is deconstructed and anomalous muscular and fibrous bands are resected. Removal of either muscle or rib accomplishes this goal. If a cervical rib is present, then the goal includes addressing the cervical rib as well. In instances where a long or complete cervical rib is present (Gruber II, III, or IV), the cervical rib should be removed. For small partial cervical ribs (Gruber I) or elongated transverse processes, the residual fibrous band arising from the rib is the source of compression and is the target of decompression and must be removed. Cases involving anomalous first ribs should be managed with removal of the anomalous first rib.

Choice of Operations

There are a surprising number of surgical alternatives for TOS decompression. These are essentially combinations of rib resection and scalene muscle resection and vary as to the extent of resection and the approach to the resection. The surgical approaches for TOS decompression include scalenectomy (supraclavicular anterior scalene muscle resection or anterior and middle scalene muscle resection), posterior rib resection (partial first rib resection from a posterior approach), claviculectomy, transaxillary first rib resection (partial scalene muscle resection and complete first rib resection via axillary approach), supraclavicular rib resection (complete anterior and middle scalene muscle and partial first rib resection), paraclavicular first rib resection (complete anterior and middle scalene muscle and complete first rib resection by means of additional infraclavicular incision), and anterior infraclavicular partial first rib resection. Kashyap and colleagues reviewed some of the advantages and disadvantages of different surgical approaches.

A surgeon's choice of operation is largely based on familiarity with the procedure. Additional elements may favor one approach over the other: presence of a cervical rib, presence of an anomalous first rib, vTOS presentation, aTOS presentation may require specific interventions.

Some authors have suggested that presentations be grouped as upper plexus (symptoms along radial distribution) and lower plexus (symptoms along ulnar distribution), and that surgery be guided according to these groupings: supraclavicular scalenectomy approach for upper plexus symptoms, and first rib resection for lower plexus symptoms. Finally, an estimation of risk for each operation may favor one approach over another; the supraclavicular approaches have greater risk of injury to the brachial plexus and its branches, and if such injuries would be particularly difficult in a given patient, then perhaps it is better proceed with the transaxillary approach.

Scalene muscle resections include resection of the anterior scalene muscle alone or both anterior and middle scalene muscles. The decision between these two operations is based largely on familiarity of the surgeons and considerations of risk. Scalenectomy is accomplished via a supraclavicular approach. Resection of the anterior scalene muscle dates to the era of Adson, Nafziger, and Oschner. This operation is fairly successful in relieving TOS symptoms, but a number of patients may have residual symptoms related to the middle scalene muscle and the first rib. The principal risk of anterior scalene muscle resection is injury to the phrenic nerve. Additional concerns include lymphatic leak from thoracic duct injury and vascular injury to the subclavian artery and vein. The principal limitation is the rate of significant persistent and recurrent symptoms.

Resection of the anterior and middle scalene muscles is a more extensive operation. It requires mobilization of the brachial plexus to reach the middle scalene muscle. As the muscle is resected, there is risk of injury to the long thoracic nerve, suprascapular nerve, and the dorsal scapular nerve. These risks are in addition to those noted with anterior scalene muscle resection alone. The result of resection of both anterior and middle scalene muscles are slightly better that those of anterior scalene resection alone.

Scalenectomy without resection of the first rib is generally attended by a reduced success rate when compared to operations that include resection of the first rib. First rib resections are most commonly accomplished via the supraclavicular or transaxial approach. A posterior approach to first rib resection has been largely abandoned due to its potential morbidity and the superiority of other approaches. An anterior infraclavicular resection is used by some as treatment of vTOS.

Supraclavicular first rib resection requires resection of the anterior and middle scalene muscles to reach the first rib. Supraclavicular rib resection does provide effective decompression of the neural structures across the thoracic outlet. Another advantage is the ability to address cervical ribs from this approach. Because the cervical ribs attach to the transverse process of the C7 vertebral body, the attachment of the cervical rib to the spine can be exposed in the course of a supraclavicular operation. This is particularly helpful in small (Gruber I) cervical ribs, which are not as easily approached by transaxillary means.

The limitations of the supraclavicular approach lies principally in the inability to resect the anterior portion of the first rib. Supraclavicular rib resection normally allows the anterior portion of the first rib to remain in place. Thus, in cases of vTOS, it does not provide adequate decompression. In cases of vTOS, surgeons who prefer the supraclavicular approach will add an anterior infraclavicular incision to remove the anterior portion of the first rib and resect part of the subclavius muscle. This approach has been referred to as a “paraclavicular approach.” Another concern with supraclavicular rib resection is the potential risk to the brachial plexus and its cervical branches, especially the phrenic nerve, the long thoracic nerve, and the dorsal scapular nerve.

Transaxillary first rib resection approaches resection of the first rib from beneath the arm, traveling along the chest wall beneath the brachial plexus and its nerve branches. The entire first rib may be resected from the transaxillary approach. Larger cervical ribs (Gruber II, III, and IV may also be resected via the transaxillary approach. The transaxillary approach allows for effective decompression of the subclavian vein in cases of vTOS. Because the approach courses below the neural structures, the incidence of brachial plexus, long thoracic, and phrenic nerve injuries is accordingly lower than that seen with supraclavicular operations.

The principal limitations of transaxillary rib resection arise from the inability to reach the upper portions of the scalene muscles. In the course of transaxillary rib resection, the lower 25% of the scalene muscle is resected. This allows removal of the first rib and allows for resection of that portion of the muscles that overlaps and attaches to the neural and vascular structures. The residual portion of the scalene muscle may eventually require removal in a minority of cases. Similarly, small cervical ribs are not readily removed via the trans-axillary approach. If these are believed to result in persistent symptoms, then a supraclavicular approach is required.

Description of Operations