diagnosis and thoracic outlet...

PERIPHERAL NERVE ISSUES: CONTROI-ERSIES AND EVOLVING TREATMEh'TS 1042-3680/01 $15.00 + .00

DIAGNOSIS AND TREATMENT OF THORACIC OUTLET SYNDROME

Rishi N. Sheth, MD, and Allan J. Belzberg, MD, FRCSC

Thoracic outlet syndrome (TOS) is one of the most controversial subjects in entrapment neuropathies. Views differ regarding the pathogenesis, diagnosis, and management of TOS. Some authorities even question its existence. In spite of lack of understanding of the disease, clinicians often encounter patients whose pain or neurologic deficits in the upper limb cannot be explained after ruling out cervical spine diseases and other common causes of peripheral neuropathies. The goals of this article arc to provide a comprehensive understanding of the syndrome by reviewing the anatomy of the region and the pathomecl~anism of the disease. Diagnosis can be established confidently in cases where the neurologic deficits are obvious and tests sho~t- positive findings. Issues surrounding cases TI-here the objective findings are minimal are discussed. The duration of conservative treatment, indications for surgery, surgical approach, and patient selection for surgery are not standardized, and experiences vary across institutions. A number of studies in the past have shown conflicting results on the outcome of various surgical approaches, which has only added to the confusion. A systematic scientific approach aimed at understanding the pathogenesis of TOS, elucidating the criteria

for diagnosis, and defining the treatment proto- col has never been undertaken. Future endeav- ors should include multicentric, randomized, prospective clinical trials so as to ansrver these questions.

Until the first half of the nineteenth cen- tury, several terms were assigned to describe the symptom complex caused by neurovascular compression in the upper thoracic region based on the physiologic mechanism and structure that caused the compression (Table 1). Such rigid classification was the basis for explaining the pathogenesis, diagnosis, and treatment.22 4"5 80 In 1956, Peet et al-' coined a common term, thomcic outlet s y f d r . o : w , to en- compass these various entities. It is now realized that strict classification is not practical. Patients tend to have multiple abnormalities, which may be responsible for the compression of the neurovascular bundle, and the clinical presentation is not unique to one ~yndrome.~"

ANATOMY

It has to be noted that TOS is a misnomer. Anatomically speaking, the true thoracic outlet is at the level of the diaphragn~.~' Early

From the Department of Seurosurger!; Iohns Hopkins University School of Medicine, Baltimore, Maryland

NEUROSURGERY CLINICS OF NORTH A-MERICA

VOLUME 17. NUMBER 7 APRIL 2001 295

296 SHETH & BELZBERG

Table 1. SYNDROMES INCLUDED UNDER THE DESIGNATION "THORACIC OUTLET SYNDROME"

Cervical rib syndrome Costoclavicular syndroine First thoracic rib s!-ndrome Scalenus anticus s!-n~lromr Hyperabduction s!.~-drorne Cervicobracliial neuro\-ascular compression s p d r o m e Shoulder-arm syndrome Effort vein t11rombor;is syndrome Pi~eumatic hammer 51.ndrome

Bracllial pIe\us syndroine Adson's syndrome Paget-Scliroetter syndrome Shoulder-girdle syn~ircmw Fractured cla\.icle syndrome Cervical brachial compression syndrome Pectoralis ininor syndrome IHumeral head syndrome Rucksack paralysis

authors described the syndrome of neurovascular compression at the superior aperture of the thorax as TOS, and this incorrect term has since remained in the medical v o c a b ~ l a r y . ~ ~ ~ The anatomy of this region is . complex, and understanding the relation between the neurovascular structures and the sur ro~~nding muscles and bony structures is critical in understanding the cause of compression.

The neurovascular complex traverses through three narroI\- straits within the thoracic outlet before entering the arm (Fig. The first passage, the so-called "scalene triangle," is mainly formed by muscles. The thoracic rib forms the base, and the anterior and middle scalene muscles form the anterior and posterior borders, respecti\.el!: The scalene muscles arise from the tubercles of the transverse processes of the cervical \-ertebrae and insert on the dorsal surface of the first rib. Although the hiatus transmits all the trunks of the brachial plexus, only the subcla1-ian artery and lower trunk are closely related to the first rib. The subclavian vein passes in front of the anterior scalene muscle." After passage through the in- terscalene triangle, the neurovascular bundle passes through the costoclavicular triangle. Bordered by the first rib below, the clavicle and the subclavius muscle superiorly, and the upper border of the scapula and subscapularis muscle posteriorly, this hiatus transmits all the nerve trunks, the subclavian artery, and the vein. It has to be noted that the first rib is a structure common to both passages, and one of the surgical options is to resect the normal rib so as to pave the Ivay to a wider passage. Finally, the neurovascular bundle traverses the subcoracoid space, a short tunnel just beneath

the coracoid process of the scapula. The bundle is closely related to the fibrous clavipectoral fascia and the dense costocoracoid ligament anteriorly. As the flat tendon of pectoralis minor muscle ascends laterally to insert on the upper border of the coracoid process, it crosses the subcoracoid space."' The ligament and tendon becoine taut during hyperabduction to significantly compress the space.

PATHOMECHANISM

Compression of the neurovascular structures as they course through the narrow passages of the upper thoracic aperture before entering the arm causes TOS. Any factor that compromises this narrow passage, which eventually leads to compression of neurovascular structures, can be included as a possible cause of TOS. These factors can be classified as:

Anatomic or structural abnormalities: congenital (cerl-ical ribs and fibromuscular bands) and acquired causes Pl~ysiologic or dynamic factors: developmental and postural factors Traumatic factors Idiopathic factors

A number of hypotheses have been proposed to explain the mode of compression in cases where structural abnormalities are not obvious.

Structural Anomalies

Congenital Anomalies

Although congenital anomalies of all structures that form the borders and contents of the

DIAGNOSIS AND TREAT\IENT OF THORACIC OUTLET 5\?23OME 297

- C , Vertebra

Figure 1. The anatomic relation of the neurovascular bundle with adjacent structures. As the brachial plexus and the axillary artery emerge between the scalenes, they loop above the first rib then pass behind the clavicle. Note that the subclavian vein lies outside the scalene triangle. The spinal nerves are numbered on the left and the vertebrae on the right. (Adapted from Travel1 JG, Simons DG (eds): Myofascial pain and dysfunction, vol 1. Baltimore, William and Wilkins, 1983: with permission.)

thoracic outlet have been known to compromise the space through which the neurovascular structures course, the common anomalies are described here. Cervical ribs are supernu- merary ribs and can be fully or partially de- veloped. If a cervical rib is formed partially, it is associated with a ligamentous extension that inserts on the manubrium or the rib below. Incidence varies from 0.0004% to 1.5% according to various studies.' '' 31 4y Fifty percent of cervical ribs occur bilaterally, and, cu- riously, there is a female predominance. The mere presence of this anomaly does not con- fer the diagnosis of TOS because only 5% to 10% of individuals with cervical ribs actually develop the symptoms of TOS.4X The presence of this anomaly creates a barrier over which the neurovascular bundle has to acutely ascend

to reach into the arm, which can result in mi- crotrauma to the nerves from stre~ching and compression."

Fibrous bands are the most common congenital anomaly encountered thar cause the sy~nptoms of TOS. Various kinds of fibrous bands have been encountered, SF?, attempts lm\-e been made to classify tbsm according to their origin, insertion, a n i associated s t ruc t~ res . ' ~ '' As these anomalous structures tr'I\-erse through the thoracic outl?: they tend to elevate, kink, and compress the Rtlurovascu- lar bundle against the surrounding anatomic borders of the thoracic outlet. Anonlalous mus- cular insertion and muscle hyperrrophy have been described to cause TOS, as :bey tend to trap the loxver trunk of the plexus ayd the subcla\.ian artery.'j


Acquired Anomalies and age all tend to sag the scapula, which puts

These anomalies are space-0ccup.i-ing lesions and can arise from the struct&-es that form the borders or the thoracic outlet or from structures outside the outlet. Such lesions limit the neurovascular bundle into a narrow con- finement. Fractures of the clavicle leading to malunion or formation of a large callus have been reported to cause TOS by reducing the costoclavicular space. Although uncommon, benign or malignant tumors of the first rib or the clavicle ha\-e been described as causing TOS.-" Among the Iesions arising outside the outlet, a Pancoast tumor can compress the neurovascular structure. Rarely, malignant or inflammatory regional lyinpl~adenopathy and tumors of tlie spinal cord or bracl~ial plexus may present as TOS.

Dynamic Factors

Developmental Factors

It has been postulated that certain changes taking place during development can predis- pose to compression of the nerves and vessels. Tlie position of the scapula dictates tlie inclination of the clavicle at the acromioclavic- ular joint, which, in turn, determines the width of the costoc1a.i-iiular space. In infants, the scapula is liigl~ in the thorax, and the costoclavicular space is wider. As growth proceeds, the scapula descends as a result of the weight of the bearing limb. This, in turn, brings the clavicle closer to tlie first rib, therebv narrow- ing the costoc1a.i-icular space. ~ e s c i n t of the scapula is more pronounced in women, and this may explain the predominance of TOS in wome11.'~

Body Position and Habitat

Any factor thar lengthens the descent of the neurovascular bundle so that the nerves and vessels become tensed as they tral-erse more obliquely against the first rib can cause TOS. Patients with ten-icodorsal scoliosis often develop TOS symptoms on the side t\,here the plexus has to travel a greater length. Poor physique of the shoulder girdle, debilitation,

extra tension on the plexus." -

Certain postures of the body tend to exacerbate or provoke the symptoms of TOS. Any position of tlie arm that stretches or compresses the plexus against the bones may consistently reproduce symptoms. Hyperabduction with external rotation of the arm is the most common position with ~\~hicll patients associate their symptoms. At arm abduction above 90°, the costoclavicular triangle closes, thereby com- pressing the cords between the bones.71 Pa- tients usually complain that symptoms are aggravated at lrork when they use their arm overhead such as when a teacher writes on the blackboard or a painter paints the ceiling. Cer- tain daily activities or l~ousel~old chores that involve using a repetitive motion of the upper extremity can x\,orsen the symptoms.

~ i m i l a & , arm positions tliat depress the shoulder girdle (e.g., carrying heavy luggage, wearing a heavy bag over the shoulder) place pressure over tlie neurovascular structure by drawing the clavicle against the first rib.Is

Hyperabduction and depression of the shoulder are maneuvers used by the clinician to provoke symptoms in patients wit11 the diagnosis of TOS.

Traumatic Factors

The role of trauma is least understood of all, yet it has to be realized that it plays an important role. Most patients complain of the onset of TOS symptoms after an automobile accident or work-related injury that involves a strain of the ipsilateral neck or shoulder r e g i ~ n . ' ~ , ~ ~ , ~ ' Not all patients who have trauma to the neck or shoulder region develop symptoms of TOS. Conversely, trauma in itself may be insuffi- cient to caise TOS. It is most likel; that trauma may be a precipitating factor in developing the symptoms in patients who have a predis- position because of congenital anomalies of the region. It has to be noted that there are few patients in whom the onset of symptoms occurs without any recognizable precipitating event." j6 Whatever the cause, the final path- way in the pathogenesis of symptoms is the compression of the neurovascular bundle.

DIAGSOSIS AKD TREAThlENT OF THORACIC OLTLET SYNDRCXIE 299

CLINICAL PRESENTATION

Clinical features of the syndrome depend on the anatomic structures compressed. Accord- ingly, three distinct syndromes are recognized. These syndromes also help to delineate the pre- senting clinical signs and symptoms. They include (1) the arterial form caused by compression of the subclavian artery, (2) the venous form caused by compression of the subclavian vein, and (3) the neurogenic form caused by brachial plexus compre~sion.'~ The last form is by far the most common one encountered,'* and these syndromes rarely overlap.

Arterial Syndrome

The severity of symptoms depends on the extent of compression and injury to the subclavian artery. The symptoms and signs are secondary to the loss of blood supply to the arm.

Mild ischemic syn~ptoms may be the result of intermittent arterial compression at the thoracic outlet. Symptoms caused by compression may appear during certain postures of the arm (hyperabduction) or during exertion of the arm when the demand for blood supply in- creases. Patients are usually symptom-free between the attacks or during the summer season. Cold temperature always worsens the symptoms. During an acute episode, patients complain of pain followed by parestl~esias; the pain is poorly localized and may involve the whole hand. Signs include a cool and pale extremity and the feeling of a stiff hand. Patients recover from the symptoms between the attacks. Such seasonal variation in the severity of symptoms is not seen with neurogenic or subclavian vein syndron~es .~~ 4h "

Sustained compression is a form of trauma to the arterial wall that can result in the formation of a mural thrombus leading to partial occlusion. Small emboli can dislodge from the thrombus and cause tissue necrosis, ulcers, and gangrene of the digits. These ischemic insults are more localized in the radial aspect of the hand, affecting mainly the index finger and the thumb.;'"

Complete obstruction caused by a thrombus at the site of the injury secondary to compres-

sion or poststenotic dilatation resulting in tur- bulent blood flow is a rare complication. The severity of symptoms depends on :he extent of collateral circulation in the arn1 In acute cases, the presentation is typical of complete blood loss to the arm. Patients complain of se- vere pain and numbness. The limb is cool, pale, and pulseless and may become para!yzed and gangrenous. Formation of an aneun-sm of the subclavian artery secondary to com~ression is rare.37, 58

Diagnosis is mainly based on ?resenting signs and symptoms. Se1-era1 manecl-ers have been described to aid in the diagnosis of arterial TOS. The following are the tests xost commonly used:

Adson test: with the patient in a sitting position, arms by the side, neck in extension, and facing either aT\-ay or toward the affected side. The examiner feels the obliteration of the radial pulse of the aiiected arm when the patient is asked to inspire deeply. 90' abduction external rotation test: the examiner feels the radial pulse \\-hen the affected arm is abducted to 90' and externally rotated with the forearm in the 9i' position. When the patient is asked to look in the op- posite direction, the pulse may obliterate. Hyperabduction test: this s i rn~ le test is positive when the radial pulse is dimin- ished after elevating the arm above the head. Shoulder brace: many variants of this test have been described, but the test consists of bracing the shoulder backward 2nd feeling for obliteration of the pulse.

Although these tests are emp1o~-?a to diag- nose TOS, the specificity of these t;.sts needs to be questioned. Studies have shox..-n that positional obliteration of the radial pc:se is seen in normal asymptomatic subjects. rositive results can be interpreted confidently only if they simulate the symptoms of the patient." ''

Arteriography of the subclaviar system is the gold standard. This can aid i:- delineat- ing the site of obstruction, demonsrrate intra- mural lesions, and shoil- dilatatior . Accord- ing to Sadler et al,h"ositional arreriograms are of limited value, because positional arterial

300 iHETH & BELZBERG

coinpressloll occurs in normal asympton~atic subjects Radiographs may reveal an associated cen ical rib."" There seeins to be a general agreement about the indications for surgical in- terventlon in cases of thromboeinbolic complication, poststenotic dilatation, and aneurysm format~nn. Transaxillary resection of the first rib is ad~.ocated along with resection of bony anomahes like a cervical rib or bony exostosis, if present, to decompress the passageway. Vas- cular repair may be necessary in certain cases.

Venous Syndrome

Venous compression at the thoracic outlet may be intermittent (positional), or venous obstruction may be associated with thrombosis. V e n o ~ ~ s thrombosis can develop at the site of con~press ion .~~ Acute symptoms develop after stre~wous physical activity of the arm and are referred to as effort tl~rombosis or Paget- Schroetter syndrome. The affected limb feels heavy and becomes weak, tender, swollen, and cyanot i~ . '~ Excruciating deep pain is felt in the chest, shoulder, and arm. Tingling paresthesias in the arm are common. Certain positions of tlie arm or strenuous physical activities can evoke intermittent compression. Synlptoins are similar to the ones described previously.

In addition to compression, other causes of thron~bosis have to be considered in the differential diagnosis, which includes iatrogenic (as a result of venous catheterization) or hyperco- agulable states (malignancy, glomerulonephri- tis, cr!-~globulinemia), vessel wall webs, and idiopathic factors.2947

Diagnosis depends on clinical suspicion. Venography is confirmatory and can show the site of obstruction or the presence of thrombus. Intermittent compression of the subclavian vein may be demonstrated during provocative maneuvers. The use and interpretation of Doppler flow measurement are difficult and of limited ~ a l u e . ~ ~ , " Venous pressure measurement has been used to support the diagnosis. Venous pressure higher than 10 mm Hg may be demonstrated in the affected limb compared with the contralateral limb.

The treatment in the intermittent type is to avoid provocative arm positions and to rest the

arm. Anticoagulation is advocated to recai~al- ize the thrombus. Arm elevation aids in reducing the edema. Venous throinbectolny has been suggested when a gangrenous or post- phlebitic complication of thrombosis is an im- minent danger. Along with thrombectomy, the first rib should be removed to relie\-e the obstruction of the venous system.i6

Neurogenic Syndrome

Compression of the brachial ple\us is the most common form of TOS." Neurologic symptoms relate to the distribution of the lo~ver trunk and the C8 through TI roots, the eleinents of the brachial plexus that are closely related to the first rib and are prone to compression against it. Because the lower elements of the brachial plexus carry motor, sensor!-, and sympathetic fibers to the arm, synlptoms relate to the fibers that are compressed.

Sensory Symptoms

Pain is the most common and earliest coin- plaint. Pain is usually located in the supraclavicular region, axilla, back of tlie neck, and shoulder region going down into the arm. Some patients may have atypical pain in the retrosternal or parascapular region. If the retrosternal pain is felt on the left side, it can be confused with anginal p a i i ~ . ~ V a i n is usually described as a dull constant ache that is difficult to localize. On most occasions, it does not re- spect the C8 through T1 dermatomes.

Paresthesias are frequently more localized to the C8 through T1 dermatomes. Patients may complain of a "pins and needles" sensation and numbness on the medial side of the arm, forearm, hand, and fourth and fifth digit^.'^ 69 ''

The sensory symptoms can be 01: a sudden or insidious onset. In most cases, patients are able to relate a major accident, fall, or any injury involving the shoulder region. In cases, where the onset is subtle, symptoms are recognized either during work (a barber may complain of pain and numbness in the hand after keeping the arm in a prolonged abducted position) or during routine household chores.

Although pain and paresthesias may be felt constantly, symptoms are aggravated by any

DIAGNOSIS AND TREATMENT OF THORACIC OUTLET SYNDROME 301

activities (e.g., carrying heavy loads on the shoulder or in the dependent arm) or position (e.g., combing hair, reaching toward a shelf) of the arm that put traction or pressure on theneu- ral structures. These ma!- be related to occupa- tion, sports, or daily activities. Symptoms are usually worse at the end of a tiring day. Pain may be seI7ere enough to disrupt a patient's sleep during the night. In many instances, patients are forced to quit or change jobs to a less demanding one. In a fen- patients, cold weather seems to \\-orsen symptoms.

Relief is bought about by discontinuing ag- gravating activities. Patients learn to relieve pressure over the brachial plexus by elevating the shoulder or resting the elbow on the table or on the arm of a chair.

Objective signs of sensory loss are usually a late finding. The percentage of patients with sensory s)-mytoms tvho demonstrate sensory loss is ~unknol\.n. Hypesthesia to pinprick and touch sensation is distributed in theC8 through TI dermatomes. This 11)-pesthesia eventually spreads to cover the u-hole hand. Vibratory sense, proprioception, and two-point discrim- ination may be affected variably.

Motor Symptoms

Complaints of subjective motor weakness may be elicited on questioning. Activities in- \-olved include hand grip, and fine hand move- ments are clumsy. Most notably, the intrinsic l ~ a n d muscles innervated by the ulnar and median nerl-es are affected, which, in turn, have contributions from the C8 through TI spinal nerves. Objective signs of muscle wasting are uncommon and seen in long-standing cases only. Muscle bulk and signs of paresis of the small intrinsic hand muscles should be as- sessed. There is a predilection for the first dorsal interosseous muscle (innervated by the ulnar nerve) and thenar n~uscles (abductor pollicis brevis and opponens pollicis innervated b ~ , the median nerve). The pattern of thenar muscle atrophy gives a characteristic guttering ~m the lateral aspect of the hand. There are several unsettled explanations as to the highly se- lective nature of muscle atrophy. The muscles ot the medial forearm may also be affected, but this is usually not conspicuous.2sh7

Sympathetic lnvolvemenl

The C8 through T1 spinal roots carry sympathetic fibers, which supply most of the upper limb.66 Vasomotor disturbances accom- pany sensory and motor symptoms. Isolated cases of sympathetic involvement are rare but have been reported in the past." Symptoms of vasomotor instability include bluish red discol- oration of the dependent arm and blanching of the hand affecting the .n,hole hand. Intermit- tent attacks of cyanosis or pallor of the hand are usually precipitated by emotional disturbances or by exposure to cold, and the severity of symptoms is independent of arm position. Trophic changes of the hand have been described in long-standlng cases, including thinning of skin, loss of hair, and thickening of nails.-14 h7

Vasomotor involvement caused by TOS has to be differentiated from other causes of vasomotor instability. Disturbances iike reflex sympathetic dystrophy, causalgia, Raynaud's disease, secondary causes of Raynaud's phe- nomenon (e.g., connective tissue disorder, obstructive arterial diseases, neurogenic lesions elsewhere, drug intoxication, dyspro- teinemias), acrocyanosis, and erythromelal- gia should be kept in mind. Patients with Raynaud's disease usually present with bilateral involvement for longer than 2 years without much progression of the disease. At times, a careful history and physical examination can elicit the secondary causes as a result of un- derlying disease. Acrocyanosis usually occurs in women and presents tvith bilateral cyanosis of the hands and, occasionally, the feet. Ery- thromelalgia is rare and in1,olves sudden burn- ing pain along with redness and increased temperature of the feet and 11ands.~ '

DIFFERENTIAL DIAGNOSIS

The symptoms of neurogenic TOS overlap with the clinical presentation of commonly encountered entities. Clinicians are faced xvith the challenge of first excluding these causes before giving a diagnosis of TOS.

Cervical degenerative diseases are common and important in the differential diagnosis. The


following signs are helpful:

Patients 1vit11 cervical disk disease have tenderness over the spinous processes and paraspinal muscles, whereas TOS patients tend to have tenderness over the supraclavicular fossa. Valsalva maneuvers and neck hyperexten- sion usually provoke radiating pain going down the sympton~atic arm in patients with degenerative disk disease. Downward pressure on the top of the head with the neck hyperextended and flexed toward the symptomatic side worsens the symptoms in patients with cervical herni- ated disks. Signs of upper motor neuron involvement might be elicited on examination in cases where the osteophytes compress the spinal cord.

Entrapment of peripheral nerves distal to thoracic outlet present with a confusing picture. The clinical picture caused by ulnar nerve neuropathy closely mimics the presentation caused by TOS. Examinations oriented specif- ically toward distinguisl~ing ulnar entrapment from TOS are as follows:

Tinel's sign at the medial epicondyle elic- its radiating paresthesias along the ulnar nerve distribution. Nerve conduction velocity studies show slo~ving of conduction of the ulnar nerve at the level of the elbow. Normal pinprick and light touch sensation are demonstrated on the radial aspect of the fourth finger in the case of ulnar neuropathy.

Certain lion-neurologic diseases that can mimic the symptoms of TOS include muscu- loskeletal (arthritis, bursitis, tendinitis of the shoulder or elbow region) and visceral (angina pectoris, esophageal diseases, gastric ulcer, pul~nonary diseases) causes.

It is in~portant to recognize secondary causes of brachial plexus compression. Pancoast tumors can invade the thoracic outlet and compress the plexus. The diagnosis is made on eliciting a history of smoking, and a plain chest radiograph reveals an apical mass. The presence of Horner's syndrome is an indication for brachial plexus involvement."

The supraclavicular and akillary regions are rich in lymph nodes. Enlargement of these nodes as a result of metastasis or inflamma- tion can compress the braclual plexus. Palpat- ing the supraclavicular and a d l a r y regions can reveal adenopathy. Metastasis to the brachial plexus from breast carcinoma can result in TOS. Trauma to the upper thoracic reg1011 can lead to compression. Malunited clavicular fractures can result in callus overgro~t-t11, \vhicll can im- pinge on the plexus.'" "' Atasov7 reported a case of scl~wannoma of the C7 root causing coin- pression of the tlloracic outlet.

DIAGNOSIS

Suspicion for TOS should be raised in women with a long neck and sloping shoulders who present with a history of some trauma to the shoulder region and symptoms referred to the C8 through T1 dermatomes. TOS remains a diagnosis of exclusion; thus, a detailed history and physical exainination should aim at ruIing out other common entities that present with an overlapping or similar clinical picture.12 5"y

The authors like to distinguish two forins of neurogenic TOS. Most patients rvith brachial plexus con~pression present with the main complaint of pain. There are no objective signs of neurologic deficits present. Physical examination usually rex~eals no a~preclable sensory loss or motor atrophy in the hand. Radio- logic studies, nerve conduction tests, and elec- tromyograpl~ic recordings show no abnormalities. Wilbourne and Porter-" 11al.e called the entity a disputed form of TOS. M L K ~ contro- versy exists as to its existence, because the diagnosis of thrs entity is based purely 011 clinical judgment. Physical signs specific to TOS that should help a clinician to support the diagnosis of TOS include (1) palpating the supra- clavicuIar fossa to elicit tenderness and (2) tap- ping the supraclavicular fossa to elicit tingling paresthesias radiating dow-n the arm. A positive Tinel's sign at the supraclavicular regioi~ can be considered specific for TOS.

Stress tests are maneuvers that provoke the symptoms of pain and paresthesias on the medial side of the hand and forearm. Such provoative tests further compress the neurovascular structures as they pass through

DIAGNOSIS AND TREATLIEKT OF THORACIC OUTLET SYNDROME 303

the narrowed Commo:; tests are as follows:

Hyperabduction test: the a r n of the patient is raised above the level of t?.e Costoclavicular test: press-lre over the shoulder region may provoh? symptoms. Military maneuver: eliciting rlxe symptoms by elevation of the chin ar.d pulling the shoulder joint behind in an eltreme "attention" position.34

Because of the lack of syste:natic clinical study, the reliability and specificity of these tests in diagnosing TOS are unknown.

In the true neurogenic type of TOS, there may be signs of intrinsic hand muscl? atrophy and reduced pinprick and touch sens3tion in the C8 through T1 distribution. Radiologic and electrodiagnostic studies may indicate some abnormality. Unfortunately, there is no test that is pathognomonic of TOS. Positive findings in the ancillary tests are associated witlx the true form of TOS. The main purposs for ordering these tests is to exclude other dizgnoses.

Radiologic studies are helpfu: in excluding the secondary causes of comprssion caused by structural anomalies. Plain r3diographs of the cervical spine are helpful in showing a cervical rib, an elongated transx.-erse process, or scoliosis of tlxe region. M a l ~ r i t e d fractures and malignant lesions arising trorn tlxe bony wall of the thoracic outlet can be demonstrated. The cervicotl~oracic ju:lction should be carefully inspected for mal:-;nant lesions such as a Pancoast tumor. Cervical degenerative diseases can be ruled out sing plain radiographs. Discograms, my elogranxs, or MR imaging of the spine may be needed to further delineate the level of disk disease. CT is helpful in excluding intraspilx?.: (e.g., tumor, syringomyelia) causes of senscry and motor symptoms.WR imaging of the csrvicothoracic region has been shown to be useful by demonstrating distortion of the plexus. In addition to demonstrating bony anomalies sf the region, MR imaging has the advantage of disclosing lxypertrophy of muscles and radiologically in- visible bands that can compress the neurovascular bundle." "

Electrodiagnostic studies can help to support the diagnosis of TOS by exc1udir.g the common

causes of neuropathies at sites other than the plexus such as ulnar neuropathy at the elbow and median neuropathy in the carpal tunnel.

Abnormal patterns of neuroelectric disturbances have been described in TOS patients. Electromyography may reveal denervation of Intrinsic hand muscles innervated by the median and ulnar nerves.27 Prolongation of F- \ \ w e latency has been described in patients with atrophy of the hand muscles.20 81 Abnor- malities in the somatosensory evoked poten- tlals obtained after stimulating the ulnar nerve have been noted in a few patients." Urschel et al-' and Caldwell et a19 described the slow- Ing of ulnar nerve motor conduction velocity across the thoracic outlet with enthusiastic results. They stimulated the plexus at the supraclavicular fossa and found abnormal conduction in tlxe ulnar nerve in almost all their cases. This test is controversial, because many investi- gators have not been able to replicate the same r e s ~ ~ l t s . ' ~ I 7 " The significance of electrodiagno- sis in TOS is limited because most patients do not sho\v any kind of abnormality. Numerous studies show a high variability in the outcome of these tests. Until a systematic clinical trial re- solves the question, caution must be exercised when interpreting the data.

TREATMENT

Patients witlx TOS have often been misdi- agnosed or passed off as histrionic. They are comforted to know that their symptoms are consistent with a diagnosis and that treatment can be instituted. As in most medical con- ditions, treatment can be divided into nonoperative and operative categories. The patient should avoid postures that exacerbate symptoms. A change in work habits, including attention to the work place can help those patients ~\.ith repetitive stress disorder. Physical therapy likely plays a role, but the literature is lacking in outcome data.77 The general approach involves measures to strengthen the shoulder girdle and improve posture. Cervi- cal traction can be added. Nonoperative care can help many patients to avoid surgery."he surgical approach to TOS varies depending on the philosophy and training of the surgeon. If the pathology is thought to be caused by


compression of the 10~1-er trunk from fibrous bands or a cervical rib, a supraclavicular approach is taken.@ If, however, the pathologic findings are thought to be more distal, resec; tion of the first rib is performed. Within the vascular literature, treatment of vascular coin- pression is approached most often with first rib resection.

Access to the first nb can be gained from a supra- or infraclavicular approach. Murphyi' described the supraclavicular approach to the first rib in 1910, but tlm approach was aban- doned as being too dangerous. It was redis- covered in 1962, when Clagett16 introduced the posterior thoracotomy rib resection. Roos" reported the transaxillar? approach, which has become the most popular surgical approach for first rib removal. For some, a combined approach affords the best surgical outcon~e.~"

Supraclavicular Approach

For most neurosurgeons, this approach is the most familiar, prmiding direct visualiza-

tion of the brachial plexus. The patient is positioned supine wit11 the arms adducted at the side, and a general anesthetic is delivered. Muscle paralysis is avoided to allow electric stimulation. The incision is made just above and parallel to the clavicle. The platysina is sl~arply divided, and the lateral one third of the sternocleidomastoid muscle is divided. The omohyoid muscle can either be retracted or divided, and the fat pad of the posterior triangle is elevated. The phrenic nerve is located, often aided by electric stimulation, on the anterior scalene muscle and is protected for the duration of the case (Fig. 2). The anterior scalene ~nuscle is then divided, and the bulk of the muscle is removed. The upper trunk of the brachial plexus is located along the lateral border of the anterior scalene muscle. The middle and lower trunk lies inferior and slightly posterior to the upper trunk. A branch of the thyro- cervical trunk may have to be divided to allow exposure of the lower trunk.

A neurolysis is performed encompassing the upper, middle, and lower trunk of the brachial plexus. Moving proximal along the

Spina~ ccessory n

Levator scapulae m

Trans cerv~cal 3 & V

Sternocleidomastoid m /

Scalenus ant m

Brachtal plexus

Figure 2. Structures encountered during supraclavicular approach to neurolysis of the brachial plexus. Often :he anterior scalene muscle is removed before the neurolysis of the plexus is performed. Note the close relation of the phrenic nerve with the anterior scalene muscle. Ext = external; Int = internal; Trans = transverse; v = vein; m = muscle; ant =anterior; n = nerve; a = artery. (Adaptedfrom Weinshel SS: Surgical exposure of the brachial plexus. In Benzel EC (ed): Practical Approaches to Peripheral Nerve Surgery. Illinois, AANS publications; with permission.)

DIAGNOSIS AND TREATMENT OF THORACIC OUTLET SYSDROME 305

lo~ver trunk, Sibson's fascia over the dome of the lung is released, and, separated by the first rib, the C8 and T1 roots are identified. In this area, several anomalous structures can be encountered, including (1) fibrous bands arising from the C7 transverse process, first rib, or apical lung pleura; (2) arterial vessels or a high arching subclavian artery; and (3) a cervical rib with fibrous bands arising from the end of the rib and inserting on the first rib. A cervical rib can be readily resected using this approach.

The first rib can be exposed by retracting the elements of the plexus. The rib can be removed through this exposure; however, this is a technically challenging procedure that places

the nerves and 1-essels between the surgeon and the rib. As such, most surgeons prefer the transaxillarv approach if the first rib is to be resected.

Transaxillary Approach

After delivery of general anesthesia, the patient is positioned prone, with the arm abducted, external11- rotated, and suspended (3-lb weight). A11 inciiion is placed at the caudal extent of the axilla, [vhere the skin breaks away from the chest \$-all (Fig. 3). Dissection is taken to the chest ~vall and up toward the axilla (see Fig. 3'4). The interiostobrachial nerve, a branch

Figure 3. A. Structures seen from the transaxillary surgical approach. B F~brous band presses the neurovascular bundle as it extends from the transverse process to attach onto the first rib. C. View of the cervical rib from the transaxillary approach. The bundle can be seen to lcop above the cervical rib in an abnormal fashion. m = muscle; Lat. = laternal; n = nerve. (Adapted from PAacHleder HI (ed): Vascular disorders of the upper extremity. New York, Futura Publishing. 1998. pp 112-1 13; with permission.)

of the T2 intercostai nerve, traverses the area on its way to the arm and needs to be identified and protected. The first rib is palpated as a broad flat structure. A localizing radiograpli can be obtained if there is any question as to its identification. Deaver retractors are used to elevate the soft tissr.es of the axilla, including the vessels and nei-1-es. The intercostal muscles are stripped from the rib using a combination of blunt dissection and cautery. The pleura is gently stripped awTalr. Blunt dissection is used to identify the inserxion of the anterior scalene lnuscle on the rib, \\-hich is then di~.ided. Prox- imal on the rib, the middle scalene muscle is located and di17ided. Palpation is used to identify any anomalous bands inserting on the rib (see Fig. 3B) .

Reinm.al of the first rib can be accomplished using a varietv of te;l-uniques. The authors prefer a rib cutter in cambination lvith rongeurs used to remove the rib piecemeal. Resection proceeds to r<ithin 2 cm of the transverse process. An intraoperarive radiograph can be obtained to confirm the extent of the proximal resection. The lower trunk courses over the rib and needs to be protected as does the T1 root as it emerges from the foramen just below the insertion of the first ri':~ on the transverse process. The anterior extent of rib resection terminates distal to the subcla1-ian vein. A cervical rib can be resected from the transaxillary approach after completion of removal of the first rib; however, the authors prefer to remove cervical ribs from a supraclavicular approach (see Fig. 3C).

If the pleura has been violated, it does not have to be closed; howerrer, the patient is main- tained on positive-pressure ventilation until the wound is closed. If the lung has been violated, a chest tube is placed. If there is a question of lung injury, a postoperative chest radiograph is obtained to look for a pneumotl~orax while the patient is in the recovery room.

Outcome

As is the case xsiih most surgical treatments, patient selection remains the most important criterion for a good outcome. \\'it11 some ex- ceptions, studies have suggested a high suc- cess rate for surgical intervention using either

a supraclavicular or transaxillary approac1.1.'~~ 72

Patients ~vith profound intrinsic muscle atrophy do not normally regain function; rather, the surgery is performed to arrest the yrogres- sive deterioration.

In one study, an orthopedic surgeon and a vascular surgeon jointly conducted 30 operations for TOS in 27 patients. Anterior scalenectomy was performed by the supraclavicular route, except in one case, where the infraclavicular route was used. The further surgical procedure was tailored to the abnormalities identified (i.e., resection of a cervical rib or band, medial scdenectomy). The first rib lvas spared. At a median follow-up of 37 montl~s, results rvere judged excellent or good on 26 of 30 sides (87%); on the three occasions when scalenectoiny alone was performed, the results 11-ere only fair or poor. There were no major complicatioi~s. The long-term outcome in this series suggests that good results can be obtained by the supraclavi&lar route without resection of the first rib."3

Wenz and Husfeldt7"reated 67 patients with a total of 80 TOSS between 1980 and 1994 (average follow-up of 6.6 years). Each of these patients ivas treated with transaxillary resection of the first rib, excision of fibrous bands, and, if necessary, excision of cervical ribs.

A total of 84.570 of the patients had complete resolution or improvement of s y m p t o n ~ s . ~ ~ Edwards et a12' studied 52 rib resections performed in 46 patients with a median follow-up interval of 33 months. Postoperative symptomatic improvement was accepted a s a confir- mation of correct preoperative diagnosis. The indications for surgery were neurologic symptoms (n = 42 excisions), arterial compromise (n = J), and venous compromise (n = 5). A total of 1 2 patients (48 resections) showed imine- diate improvement in symptoms after surgery, although symptoms recurred in three patients (four resections) between 6 and 8 months after surgery. In the final 2 years of this study, 20 resections resulting in symptomatic improvement T\.ere performed, suggesting a prevalence for TOS of at least 10 per 100,000 population per year. Thoracic outlet decompression was performed more frequently in this series than in many previous studies, suggesting that TOS may be ~nderdiagnosed.~'

DIAGNOSIS A S D TREATMENT OF THORACIC OL TLET SYNDROME 307

Donaghy et all\erformed 49 operations in 40 patients (33 women, seven men; age range, 22-62 years) with follow-up at 3 months to 20 years after surgery. Cervical ribs were removed in 23 patients together with fibrous band excision in nine. In the 17 patients xvith- out a cervical rib, the thoracic outlet xvas de- compressed by resection of the first thoracic rib in nine and by supraclavicular operations in eight. After surgery, patients reported reduced pain (33 of 36 patients), and improved sensory disturbance (30 of 35 patients), hand muscle strength (14 of 27 patients), and hand function (23 of 24 patients). After surgery, TOS recurred in two patients, and symptolns con- tinued to progress in three patients in 1vhom other diagnoses eventually emerged. Surgical con~plications were recorded in 10 patients but were transient and did not result in per- manent symptomatic sequelae. These inves- tigators concluded that surgical treatment of suspected neurogenic TOS relieves pain and sensory disturbance (90%) but is less effective for muscle weakness (50%). Surgery relieved sensory and motor abnormalities to a similar degree in patients with and without a cervical rib.'"

Franklin et a F studied all injured workers in the Washington State Workers' Con~pensation system who received TOS surgery from 1986 to 1991 (n = 158). The main outcome measure was work disability status 1 year after surgery. Additional functional status and quality of life outcomes were determined by telephone sur- vey an average of 4.8 years after surgery. A sample of workers wit11 a TOS diagnosis who did not receive surgery during the period from 1987 to 1989 were identified as a comparison group (n = 95). It cannot be stressed enough that this was not a randomized study, and as such, comparison of the surgical group lvith the nonsurgical group is of limited value. Sixty percent of workers were still work-disabled 1 year after surgery. The strongest predictors of remaining disability were the amount of work disability before surgery, longer time bet~veen injury and TOS diagnosis, and older age at injury. There was no relation between type of surgery, presence of any provocative tests, experience of the surgeon, and work disability outcome. In follow-up surveys, 72.5% of lvork-

ers still reported that thel- I\-ere "limited a lot" in vigorous activities. Conlparec'i xt.ith a nonsurgical sample of TOS patients, those receiv- ing surgery had 50% greater medical costs and were three to four times more likely to be I\-ork- disabled. The nonspecific ~~eurogenic TOS diagnosis, complexity of ~vorkers' compensation cases, and adverse event profiles are 1ikel~- sub- stantial contributors to the I\-orse outcomes reported l~ere.~%off et al' compared surgery for neurogenic TOS between laborers and nonlaborers. Laborers with TO5 .itrere less likely to have a good result from surgical intervention and were unlikely to return to thelr original occ~~pation.~'

SUMMARY

TOS represents a spectrum of disorders encompassing four related syndromes: arterial compression, venous comyression, neurogenic compression, and a pnor1~- defined pain syndrome. Patients can present ~vith signs of arterial insufficiency, venous obstruction, painless wasting of intrinsic llar-td muscles, and pain. History and physical examination are the most important diagnostic studies, and radiographs of the chest and cervical spine and electromyo- grapl~y/nerve conduction studies are useful to identify other causes oi: pain and disability Surgical intervention is indicated for patients failing nonoperative maneuvers and can usually yield satisfactory results. TOS may also be the most underrated, 01-erlooked, and misdi- agnosed, and the most important and difficult to manage peripheral nen-e iompression in the upper e~tremity.~

References

I . Adson AW:Surgical treatnwr.: : . , r syn~p to~ns prod~iced b!. cervical ribs and the scale:-.:i anticus rnusclt. Surg Gyiiecol Obstet 85685-7011. :-1-

2. Atasoy E: Thoracic outlet ;-,:npression s v ~ ~ i i r o m r . Orthop Clin North Am 27.2-7--313, 1996

3. Atamy E: Thoracic outiet ;.mpressioi~ synLirome caused by a schwaniwma 13: ::-e C7 nerx e r w t . 6 r I Hand Surg 22662-663,19CJ;

4. Beers MH, Berkow R: Cari,:. ascular iilsor~lers. 111 Berkow R, Beers MH (edsi: Ti.- Merck hlmual of Di- agnosis and Treatinmt, ed 1- S e l v Jerse!-. hlerik Re- search I.aboratories, 1999, p~ : T-'0-1794


5. Bilbey JH, Muller NL, Connell DG, ct al: Thoracic outlet syndrome: Evaluation with CT. Radiology 171:381- 384,1989

6. Blair Dx, Rapport S, Sostman HD, et al: Normal brachial plexus: MR imaging. Radiology 165:763-767, 1987

7. Braunn.ald E: Disorders of cardiovascular system. 111 Fauci AS, Braunwald E, Isselbacher KJ, et a1 (eds): Harrison's Principles of Internal Medicine, ed 14. New York, RlcGra~\.-Hill, 1998, pp 1401-1403

8. Buonocore XI, Manstretta C, Mazzucchi G, et al: The clinical evaluation of conservative treatment in patients with the thoracic outlet syndrome. G Ital Med Lav Er- gon 20:249-231,1998

9. Cald~vell JR, Crane CR, Krusen EM: Nerve conduction studies; An aid in the diagnosis of thoracic outlet syndrome. South Med J 64:210-212,1971 Campbell E, Hornyard WP, Burklund CW: Delayed brachial plexus palsy due to ununited fracture of the clavicle: A case report. JAMA 139:Yl-92, 1949 Campbell J S : Thoracic outlet syndrome. 111 Frylnoyer JW, Ducker TB, Hadler NM, et a1 ( 4 s ) : The Adult Spine: Principles and Practice, ed 2. Philadelphia, Lippincott- Raven, 1997, pp 1319-1355 Camphcll IS, Naff NJ, Dellon AL: Thoracic outlet synclrome: Scurosurgical perspective. Neurosurg Clin North Am 2227-233, 1991 Capistrant TD: Thoracic outlet syndrome in whiplash injury Ann Surg 183: 175-1 78, 1977 Cherington M: Ulnar conduction velocity in thoracic outlet syndrome [letter]'. N Engl J Med 294:1185-1186, 1976 Chodoroff D, Lee DW, Honet JC: Dynamic approach in the diagnosis of thoracic outlet syndrome using somatosensory evoked responses. Arch Physiol Med Re- habil66:3-7, 1985 Clagett OT: Research and proresearch. J Thorac Cardio- vasc Surg 14:153, 1962 Dale \2'A, Lewis MR: Management of thoracic outlet syndrome. Ann Surg 18575-585,1975 Dauhe JR: Rucksack paralysis. JAMA 208:2447-2452, 1969 Dona$~v hi, Matkovic Z, Morris P: Surgery for suspected neurogenic thoracic outlet syndrome: A follow- up study. J Neurol Neurosurg Psychiatry 67:602-606, 1999 Dorfman L\': F-m.ave latency in the cervical-rib-and- band s!mirome. Muscle hTerve 2:158-159, 1979 Eduwcis DP, Mulkcrn E, Raja AN, et al: Trans-axillary first rib resection for thoracic outlet syndrome. J R Coll Surg Edinb 44:362-365,1999 Falconar hlA, Weddell G: Costoclavicular comyression of the subclavian artery and vein: Relation to the scalenus anticus syndrome. Lancet 2:539-544, 1941 -. --

23. Felson B: A reviexv of 30,000 chest x - r aw 111 Chest -,

pp 19.5-495 24. Franklin GM. Fulton-Kehoe D, Bradlev C, et al: Out-

come of surgery for thoracic outlet syndrome in \\lash~ngton State Workers' Compensation. Neurology j1:132-1237,2000

25. Geroudis R, Barnes RW: Thoracic outlet arterial compression: Prevalence in normal persons. Angiology 31538-541,1980

26. Gilliat RW: The classical neurological syndrome associated rtrith a cervical rib and band. 111 Greep JM,

Lemmens HA, Roos DB, t.t a1 (ecis): Pain in Slmulder and Arm. The Hague, XIartinus Nijhoii Publisher, 1979, pp 173-18.:

27. ~ i l l i a t t Ril'. Willison RG, Dietz V, et al: Peripheral nerve conduction in patients I\-ith a cervical rib and bnncl. Ann Neurol4:124-129, 19;s

28. Gilliat RM'. LeQuesne P\.\.. Logue \J, et al: M'asting of the hand associated with a cervical rib or band. J Neurol Neurosur~ Psychiatry 33:615-626, lQ70

29. Glass BA: The relationship of axillary vein thrombosis to the thoracic outlet syniirome. . inn Thorac Surg 19:613-621. 1975

30. Goit CD, Parent FN, Sato DT, et al: A comparison of surgery t ~ ~ r neurogenic tl~oracic outlet s ~ n d r o m e he- twe~n.laborers a14 nonlaborers. Am ] &rs 176:215- 218,1998

31. Healey JE. Seybold WD: Thorax. 111 A Synopsis of Clin- cal Anatomy. Philadelphia, WB Saunders, 1969, p p 81- 126

32. Heyden 6. Vollmar J: Thoracic outlet syndrome 1vlt11 vascular complications. I Cardiovasc Surg 20:531-536. 1979

33. Leuk MR. Dale WA: Thoriicic outlet syndrome. Surgi- cal Digest. 1973, pp 16-17

34. Lord JM'. Iiosati LM: Thoracic outlet syndromes. Clin Symp 233-32,1971

35. Luoma .A. Nelerns B: Tlioracic outlet syndrome: Thn- racic surgery perspective. Neurosurg Clin North Am 2:137-22~.. 1991

36. Makhoul RG, Machleder HI: Developmental anoma- l i e at the thoracic outlet: An analysis of 700 consecu- tive cases. J Vasc Surg 16:531-542,1992

37. Martin \-. Gaspard DJ, Iohnston PW, ct al: Vascular manifestations of the thoracic outlet syndrome. Arch Surg 111 :779-782,1976

38. Martinez NS: Traumatic thoracic outlet syndrome, 111 Pollak E\V (ed): Thoracic Outlet Syndrome: Diagno- sis and Treatment. Kelt- York, Futura Publishing Com- pany, lYP6, pp 125-134

39. hlcCarthy MJ, Varty K, London XJ, et al: Experience of supraclavicular exploration and decompression for treatment of thoracic outlet syndrome. Ann Vasc Surg 13:268-274,1999

40. hiolina IE: Combined posterir.:. and transaxillary approach for neurogenic thoracic outlet syndrome. J .Am Coll Surs 187:3943,1998

41. 5iurph)- T: Brachial neuritis c: 11seiI by pressure of first rib. A U S ~ Med J 15:582. 1910

42. Ochsner A, Gage ?/I, DeBakey 11: Scalenus anticus (Saffziger) syndrome. .Am J Surg X669-695,1935

43. Panegyres PK, Moore S, Gibson R, et al: Thoracic outlet synciroines and magnetic resonance imaging. Brain 116:823-841,1993

44. Pang D, Wessel HB: Thoracic outlet syndrome. Neuro- surgery 22:105-121,19RR

45. Peet D, Henriksen JD, Anderson TP, et al: Thoraiic- outlet iyndrome: Evaluation of a therapeutic exercise program. Mayo Clin Proc 31:281-287,1956

46. Pisko-Dubienski ZA, Hollingsl\.orth J: Clinical application of Doppler ultras,:nography in the thoracic outlet synclrome Cali I Surg 21:145-140. 1978

47. Pollak EW: Clinical Presentation. 111 Thoracic Outlet Syndrome: Diagnosis and Treatment. S e w York, Fu- tura Publishing Company, 1986, p p 37-70

48. Pollak EW: Surgical anatomy of the thoracic outlet s!-r~- drome. Surg Gynecol Obstet 150:97-103,1980

DIAGNOSIS hlD TREATMENT OF THORACIC OUTLET SYNDROSlE 309

49. Pollak El\.: Surgical anatcny. 111 Thoracic Outlet Syn- drome: Diagnosis and Treatment. Nerv York, Futura Publishing Company, 198-, pp 3-37

50. Raaf J: Surgery for cervii:l rib and scalenus anticus syndrome. JAMA 13219-123,1955

51. Razi D\I, Wassel HD: Tr:ffic accident incluced thoracic outl~.t syndrome: Dei,.mpression without first rib resection, correction of associated recurrent thoracic aneurysm. Int Surg 78:25-:7, 1993

52. Rob CG, Standeven A: Arterial occlusion complicat- ing thoracic outletconipreision syndromes. BMJ2:709- 712,1958

53. Rogers L: Upper limb ra in due to lesions of the thoracic outlet: The scalenus syndrome, cervical rib, and costoclavicular c; mpression. B\!J 2:956-958, 1949

54. Roos DB: Congenital anc?:l~alies associated rvith thoracic outlet syndron~e: An:tomy symptoms, diagnosis and treatment. Am J Surp 132:771-778, 1976

55. Roos DB: S e w concepts cf thoracic outlet syndronie that explain etiology, syn-;?toms, diagnosis and treatment. \.asc Surg 13313-3-1, 1979

56. Roos DB: Thoracic outlet s:.-ndromes: Update 1987. Am J Surg 1 %568-573,1987

57. Roos DB: Transaxillary ay;lroach for first rib resection to relieve thoracic outlet syndrome. Ann Surg 163:356- 358,1966

58. Ross JP: The vascular con~ylications of cervical rib. Ann Surg li0:340-345,1959

59. Rowland LP: Thoracic o.:tlet syndrome. 111 hlerritt's Textbook of Neurology, t.d 8. Philadelphia, Lea & Febiger, 1989, p p 413-444

60. Sadler TP, Rainer WG, T:..-on-tbley G: Thoracic outlet compression: Application of positional arteriography and ner\.e conduction studies. Am J Surg 130:704-706, 1975

61. Salnions 5: Muscle. lii Williams I'L (ed): Gray's Anaton~y. New York, Churchill Livingstone, 1995, pp 838-S40

62. Shaliani BT, I'otts F, J ~ ~ g i o : ~ m A, et al: Electrophysiolog- ical studies. Muscle Nerve 3:192-193, 1980

63. Sh,lr,ln D, \loulton A, Gri3tre\ GH, et al: T1f.o-surgeon approach to thoracic outlcr syndronie: Long-term outcome. J R Soc Med '1?.:23"-243, 1999

64. Sumner DS: Noninvasil-c val;cular laboratory assess- ment. 111 \lechlecler i 11 (&): \-ascular Disorders of the Upper Eltremity. Nccv \I?rk, Futura Publishing Corn- pany, 199S, pp 15-6:

65. Sunderland S: Brachial exus us lesions due to abnor- ribs: The 'cer~.ical rib' syndrome. 111 Nerves and

Nerve Injuries. Ne\\. Yor!, Churchill Livingstone, 1978, pp 920-943

66. Sunderland S: The bracI-:~l plexus: Normal anatomy. 111 Nerves and Nerx.e Jnjur:es, Setv York, Churchill Liv-

67. Sunderland S: Disturbances of brachial plexus origin associated with unusual anatomical arrangements in the cervico-brachial region: The thoracic outlet syndrome. In Nerves and Nerve Injuries. S e w York, Churchill Livingstone, 1978, pp 901-919

68. Swanson WM, Vigesaa RE, Hieb RE, et al: Thoracic outlet syndrome: The supraclavicular approach to first- rib resection. Surgical Rounds 73:79, 1986

69. Tindall SC: Chronic injuries of peripheral nerves by entrapment. In Youmans JR (ed): Neurological Surgery, ed 4. Philadelphia, WB Saunders, 1996, p p 2182-2187

70. Todd TW: The descent of the shoulder after birth: Its significance in the production of pressure-symptoms on the lowest brachial trunk. Anat Anz 41:385-397, 1912

71. Todd TW: Posture and cervical rib syndrome. Ann Surg 75:105-109,1922

72. Toso C, Robert J, Berney T, et al: Influence of personal history and surgical technique on long-term results. Eur J Cardiothorac Surg 16:44-47,1999

73. Urschel HC, Razzuk MA: Improved ~nanagement of the Paget-Schroetter syndrome secondary to thoracic outlet compression. Ann Thorac Surg 521217-1221, 1991

74. Urschel HC, Razzuk MA: Management of the thoracic outlet syrtdrome. N Engl J Med 28b:1140-1143, 1972

75. Urschcl HC, Razzuk MA, Wood RE, et al: Objective diagnosis (ulnar conduction velocity) and current therapy of the thoracic outlet syndrome. Ann Thorac Surg 12:608-620,1971

76. Urschel HC, Hyland JW, Solis RM, et al: Thoracic outlet syndrome masquerading as coronar>- artery disease (pseudoangina). Ann Thorac Surg 16:239-248, 1973

77. Walsh MT: Therapist management of thoracic outlet syndrome. J Hand Ther 7:131-144, 1994

78. Wenz W, Husfeldt KJ: Thoracic outlet syndrome- an interdisciplinary topic. Experience with diagnosis and therapy in a 15-year patient cohort (80 transaxillary resections of first rib in 67 patients) and a literature review. Z Orthop Ihre Grcnzgeh 135:84-90, 1997

79. Wilbourne AJ, Porter JM: Thoracic outlet syndromes. 111 Weiner MA (ed): Spine: State of the Art Re- views. Philadelphia, Hanky and Belfus, 1988, pp 598- 626

80. Wright IS: The neurovascular syndrome produced by hyperabduction of the arms. An1 Heart J 29:l-19,1945

81. Wuff CH, Gilliatt R: F-waves in patients with hand wasting caused by a cervical rib and band. Muscle Nerve 2:452-457,1979

82. Yiannikas C, Walsh JC: Son~atosensory evoked responses in the diagnosis of thoracic outlet syndrome. J Neurol Neurosurg Psychiatry 46:231-2111,1983

A d d w s r c p : i : t r.~.q~iests t o

Allan J. Belzberg, \ID, FRCSC Johns Hopkins Hospital

600 North Wolfe Street, \!eyer 5-109 Baltinlore, MD 21287-7509

diagnosis and thoracic outlet...

Documents