Tiakis/i Neiirosiirgery 9: 137 - 143, 1999 Erbayrnklar: Siiprnc/aviciilar Approac11 lo NelIrogeHic Tas

Neurogenic ThoracicCompression at

Outlet Syndrome due tothe Interscalene Triangle

the

Interskalen Üçgendeki Sikismalara Bagli NörojenikTorasik Çikis Sendromu

SERHAT ERBA YRAKTAR, ENGIN UÇAR, BURAK SADE, UMIT ACAR, ME TIN GÜNER

Dokuz Eylül University, School of Medicine, Department of Neurosurgery (SE, EU, BS, ÜA, MG), Izmir, Turkey

Gelis Tarihi: 13.3.1998 ~ Kabul Tarihi: 6.8.1999

Abstract: The characteristics of neurogenic thorasic outletsyndrome due to compressions at the interscalene trianglewere evaluated in 17 patients. Sixteen of the patients werefemale, and the mean age of the group was 34,6 years. Eightcases had only osseous anomalies (Group 1). Another sixhad no bony variations, and the scalene musdes had causedthe compression (Group II). The other 3 cases had both bonyanomalies and muscular compression (Group III). The mostcommon symptoms were painful myofascitis (88%) andparesthesia (77%), and the most frequent physical findingswere motor deficits (82%) and positive Adson's tests (82%).In cases with osseous anomalies cervicocephalic symptomswere often seen but paresthesia was relatively rare (p<O.05).Also in these patients, a positive hyperadduction test wasoften accompanied by a fibrous band. Double crush orreverse double crush syndromes were detected in 7 of the17 cases. Of the electrophysiological tests performed,electromyography gaye the most information. Twenty-oneextremities in 17 patients were operated via a supradavicularappproach sparing the first rib, and no complications wereencountered. Fibrous bands were found only in patients whohad osseous anomalies (p<O.05). Satisfactory and tolerableresults were achieved in nine (53%) and seven (41%) patients,respectively.

Key Word s: Double crush syndrome, supradavicularapproach, thoracic outlet syndrome.

INTRODUCTION

Symptoms of upper extremity neurovascularcompression can arise from changes or abnormalities

Özet: Interskalen üçgendeki kompresyonlara baglinörojenik torasik ÇikiSsendromunun özellikleri 17 hastadadegerlendirildi. Bunlardan 16'sl kadin olup ortalama yas34.6 idi. Sekiz olguda yalniz kemik anomalileri vardi(l.Grup). Diger 6 olguda ise herhangi bir kemik varyasyonolmayip sikisikliklarin nedenleri skalen adelelerdi (2. Grup).Kalan 3 olguda ise hem kemik anomalileri hem de kassikisikliklan mevcuttu (3. Grup). En sik izlenen semptomlaragrili myofasitis (%88) ve paresteziydi (%77). Motordefisitler (%82) ve pozitif Adson testleri (%82) de en sikgözlenen fizik bulgulard!. Kemik anomalili olgulardaservikosefalik semptomlar sik oldugu halde parestezilergöreceli olarak düsük oranda izlenmekteydi (p<O.05). Pozitifhiperadduksiyon testi de siklikla fibröz bandlara esliketmekteydi. Ikili yada ters ikili sikisma sendromlari da 7olgumuzda gözlendi. Elektrofizyolojik testler arasinda iseelektromyografi, digerlerine göre daha bilgilendiricigöründü. Onyedi olguda 21 ekstremite supraklaviküleryolla birinci kaburga çikarilmaksizin tedavi edildi veherhangi bir komplikasyon ile karsilasilmad!. Fibrözbandlar sadece kemik anomalili olgularda izlendi (p<O.05).Tatminkar ve tolere edilebilir sonuçlar ise sirasiyla dokuz(%53) ve yedi (%41) hastada elde edildi.

Anahtar Sözcükler: Ikili sikisma sendromu,supraklaviküler yaklasim, torasik çikis sendromu.

in three spaces, namely, the interscalene triangle, thecostoclavicular space and the subpectoral minorspace (L, 13). Congenital fibromuscular (l7, 18) andbony anatomic (lO) variations in these spaces may

137

Turkish Neurosiirgery 9: 137 - 143, 1999 Erbayraklar: Siipraclaviciilar Approacli lo Neurogeiiic Tos

PATIENTS and METHOD S

Surgical technique:

Table I: Demographic and clinical characteristics of17 patients with TOS.

procedure and yields satisfactory results in certainpatients (2,20).

34.7±9.6

16/1 •32.9±23.9

4 (24%)4 (24%)9 (52%)

76.0±49.5

Mean age (years)Gender (Female/Male)Mean duration of symptams (manths)Patients with signs andi or symptoms

BilateralUnilateral rightUnilateralleft

Mean duration of fallow-up (manths)+: P<O.05

The supraclavicular approach is performedund er general anesthesia. We placed the patient inthe supine pasition and turned the neck toward the

During the 14-year period from 1984 through1997,16 female and 1 male ranging in age from 22 to63 years (rnean 34.6) underwent primary surgicaldecompression of the interscalene triangle (Table 1).The patients were subdivided into three groupsaccording to the nature of the cornpressing structure,namely osseous (group 1), muscular (grup 11),andcombined (group III). In all cases, thesyrnptomatology, physical examination findings, andresults of provocative tests were consistent with TOS.Osseous anomalies were investigated with x-raystudies. Electrophysiological tests and computerizedtomography were used for either confirmation ofTOS or differentiation of it from the clinics of other

compressive neuropathies. Initally, all TOS caseswere treated conservatively, and surgery wasproposed only in cases where there was noimprovement. In all of our patients, surgery wasindicated due to brachial plexus cornpressian at theinterscalene triangle. The operations were done viathe supraclavicular approach. As part of the protocol,patients were prescribed physical exercisespostoperatively to minimize scar formatian aroundthe brachial plexus.

In this report, we present our experience with17 patients who had neurogenic TOS due tocompression at the interscalene triangle, and whounderwent surgery via the supraclavicular route,with the first rib left intact.

Although a causative association betweenthoracic outlet compressions and arm symptoms isoften suspected clinically, this is difficult to prove.Various electrophysiological tests has beenevaluated, but all have a low level of sensitivity forTOS diagnosis and a negative finding does notexclude the condition (3, 9, 22). In patients who haveno visible osseous anomaly on x-ray, magneticresonance imaging (MRI) has recently been suggestedto be of potential value in diagnosing TOS. UsingMRI, it is possible to demonstrate deviation ordistortion of nerves or blood vessels, findings thatsuggest the presence of radiologically invisible bands,and to revealother causes of TOS (11). Currently,the cornbination of interpretation of the symptoms,physical examination findings, and provocative testsseems to be the most reliable way to identify thecompression site and the affected structures. Thetraditional approach to decompression of theinterscalene triangle has been transaxillary first ribresection (16), either with or without transcervicalscalenectomy (1). Recently, a more selective methodconsisting of anterior scalenectomy and brachialplexus neurolysis via the supraclavicular approachhas also be en widely proposed, since it is a safe

The interscalene triangle, which is bordered bythe anterior scalene muscle anteriorly, the middlescalene muscle posteriorly, and the first rib inferiorly,is the most common site of compression bypredisposing fibrosseous or fibromuscular anornahes(1). The brachial plexus and the subclavian arterypass through this triangle. Compression of thesubclavian artery is ra re, and its symptoms areusually caused by arterial insufficiency. They includeextremity weakness, cold sensation, and pain causedby ischemic neuritis of the plexus. The finding ofbruitin the supraclavicular area, in addition to anarteriogram that reveals a cornpressed area withpoststenotic dilatation of the subclavian artery, maygive the clinical impression of vascular TOS (5,6).However, 90% of all TOS cases are not vascular, butneurogenic in type. In these patients, the mostcommon seen symptoms are pain and paresthesia(90-95%) distributed according to the level of theplexus involvement, weakness with an easilyfatigued extremity, cold sensation, and coldintolerance and swelling (23).

predispose individuals to thoracic outlet syndrome(TOS) after trauma, inflammation, or as a result ofother factors, such as the dynamic and functionaldemands of the upper extremity, shoulder, and neckthat cause postural changes (1,21).

138

Tllrkis/i Neiirosiirgery 9: 137 - 143, 1999

nonoperative side. The surgery started with a 6-8 cmneck crease incision parallel to and 2 cm above theclavicle. We divided the platysma, and thenidentified and mobilized the supraclavicularnerves.The omohyoid and the lateral portion of theclavicular head of the sternocleidomastoid muscle

were then divided. The supraclavicular fat pad waselevated proximally, the scalenus anticus muscle wasapproached, and the phrenic nerve was thenidentified on its anterior surface.Next, the upper,middle, and lower truncus of the brachial plexuswere mobilized. Any muscular, fibrous, or osseousstructure observed to be causing compression wasresected at this stage, and varying degrees ofneurolysis of the brachial plexus were performed ineach case. In all patients, the first rib was left intact.The procedure was finished with repair of thesternocleidomastoid muscle and closure of the skinwound.

Clinical data was assessed using chi-square test,and Fisher's exact test was used when needed. P­values less than 0.05were accepted significant duringthe statistical analysis.

Erbnyrnklnr: SlIprnclnviciilnr Approncli lo Neiirogeiiic Tos

RESULTS

Group I included eight patients, six who hadcervical ribs and two had enlarged transverseprocesses. Fibrous bands were present in six of thesepatients. Grup ii consisted of six patients, and nofibrous bands were found during these operations.In Group III included three patients, two with cervicalribs and one with enlarged transverse process. Anassociated fibrous band was also excised in one case.

The duration of symptoms varied from 1monthto 6 years (mean 33 months). The most commonsymptoms in all groups were pain and paresthesia.Painful myofascitis and radicular pain were presentin 88% and 71% of patients, respectively. Paresthesias(77%) and weakness of the upper extremity (53%)were also frequently reported (Table 2). Complaintsof easily fatigued extremity, especially while carryingobjects, nocturnal aggrevation of paresthesias, andcervicocephalic symptoms were relatively ra re(below 30%). Prior to surgery, one of the patients hadpreviously undergone abilateral sympathectomyand first rib resection, and another had undergone

Table ii: Summary of the syrnptomatology and physical findings in patients with TOS.

Groupl (N=8)Groupll (N=6)GroupIII (N=3)Total (N=17)SYMPTOMS PainMyofascitis

76215 (88%)Radicular pain

63312 (71%)Paresthesia

4+6313 (77%)Weakness

4329 (53%)Easily fatigued limb

32O5 (29%)Aggravation with carrying

32O5 (29%)Noeturnal aggravation

21O3 (18%)Cervicocephalic symptoms

4+OO4 (24%)

PHYSICAL EXAMINA TION Motor deficits74314 (82%)

Sensorial deficits63211 (65%)

Bruit1113 (18%)

Absence of DTRs1O12 (12%)

Atrophy11O2 (12%)

PROVOCA TIVE TESTS Adson's65314 (82%)

Neck tilting5117 (41%)

Hyperabduction2114 (24%)

Hyperadd uetion3eOie4 (24%)

PLEXUS INVOL VEMENT Upper11O2 (12%)

Lower3115 (29%)

Combined44210 (59%)

Abbr: DTR:deep tendon reflexe~: P<O.05;e: hyperadduction test was positive only in cases with fibrous bands (p<O.05).

139

Tiirkisli Neiirosiirgery 9: 137 - 143, 1999 Erbayrnklar: Siiprnclaviciilar Approacli lo Neiirogeiiic Tos

The mean follow-up period was 76 m.onths.Improvement of neurogenic TOS symptoms wassatisfactory in nine (53%) cases. Seven of the othercases were left with persistent minor symptoms thatcould be tolerated well during daily life, but onepatient experienced no improvement postoperatively(Table 4).

5 (29%)7 (41%)

1 (6%)2 (12%)1 (6%)

5/7 (71%)

4/6 (67%)ELECTRODIACNOSTICTESTS(S/NS)

ElectromyographyF-waveSEPENC

_ 1/1(100%)3/6 (50%)

DOUBLECRUSH SYNDROME 7 (41%)CT (CDH) 5 (29%)Electrodiagnostic tests (CTS) 2 (12%)

Abbr: S/N:specific/nonspecific findings confirmingTGS;SEP:somatosensoryevoked potential study;ENG:electroneurography; CDH:cervicaldisc hemiation;CTS:carpaltunnel syndrome.

patient, who had undergone previous surgeryon thecontrlateral side, we found no compression at theinterscalene triangle. Likely, this indi vid ual' ssymptoms were caused by overuse of the extremitydue to disease in the contrlateral arm. Fibrous bands,which were encountered in 7 of the 11 patients withosseous anomalies, were also excised. We found nofibrous bands in the patients with muscularcompression. In group II and III patients, weeliminated muscular compression via anteriorscalenectomy, and performed middle scalenectomywhen indicated (Table 4). Five cases were operatedbilaterally, and the scalene muscles were identifiedas causing the compression in three of these patients.

Table ID: Results of the diagnostic laboratory studies.

X-RAYCervical rib

UnilateralBilateral

Enlarged transverse processUnilateralBilateral

Combined

When we considered both the symptomatologyand physical examination, lower (C8-Th1) andcombined (C5- Thl) types of brachial plexusinvolvement were identified in 5 and 10 patients,respectively. Bilaterality of the osseous anomaly didnot seem to affect the symptomatology. Although x­rays revealed bilateral osseous anomaly in fourpatients from group I, three from group II, and twofrom group III, these were symptomatic in only halfof these patients, and were excised unilaterally onthe affected side. Computed tomography showedvarious intensities of disc herniation in five patients,but onlyone of the five underwent similtaneouscervical discectomy and thorasic outletdecompression procedures. Electrophysiologicaltests, including electromyography (EMG),somatosensory evoked potential study (SEP) andelectroneurography (ENG), were also used in sevenof the 17 patients. Five of the seven exhibited specificfindings of thoracic outlet compression. EMG wasthe most informatiye, and yielded these specificfindings for four of six cases (Tabi e 3). Findingsassociated with another type of compression,suggesting double crush or reverse double crushsyndrome, were also recorded in five of the six whounderwent EMG, and in one case severe carpaltunnel compression was released during the samesurgery, immediately after the brachial plexus wasdecompressed.

Physical examination showed motor andsensory deficits in 82% and 65% of the patients,respectively, and neck tilting was painful in 41%. Ofthe provocative tests, Adson's test was positive in82% of the c ises (Table 2). Atrophy, aggravation ofsymptoms on percussion of the supraclavicular areaand on hyperabduction or hyperadduction, and bruitin the supraclavicular region were not commonlyobserved (below 25%).

ipsila teral carpal tunnel release. Associa tedinflammatory conditions, such as radialepicondylitis, and bicipital and rotator cuff tendonitissecondary to disuse of the affected extremity, wereobserved in four patients.

Seventeen patients underwent 21 surgicalprocedures involving the supraclavicularapproaches. Four of the patients had bilateralprocedures, with the intervals between theoperations for the two sides varying from 1 monthto 4 years. In nine of the group I and III patients, wewere able to totally resect the osseous anomaly. Inone case the resection was subtotal, and in another

Statistica/ niin/ysis:

All but one of our patients was female, whichindicates agender bi as toward woman in TOS(p<0.05). In patients with TOS, only the complaintsof paresthesia and cervicocephalic symptoms werestatistically significant (p<0.05). In patients with

140

Turkish Neurosurgery 9: 137 - 143, 1999 Erbayrakiar: Supraclavicular Approacli lo Neurogenic Tos

Table IV: Summary of the surgical findings, procedures and their effects on the complaints of the patientswith TOS.

9 (53%)7 (41 %)

1 (6%)

Total (N=17)

8 (47%)

3 (18%)

7 (41 %)

9 (53%)

5 (29%)

3 (18%)

1 (6%)

1 (34%)2 (66%)

GroupIII (N=3)

2 (66%)

1 (34%)

1(34%) •

3 (100%)

1 (34%)

3 (50%)3 (50%)

6 (100%)

2 (34%)

1 (17%)

GroupII (N=6)

2 (25%)

2 (25%)

1 (12%)

GroupI (N=8)

6 (75%)

2 (25%)

6(75%) •

Enlarged transverse process excision

Fibrous band excision

Cervical rib resection

Bilateral approaches

Surgery for a second compression

Negative explorationPostoperative results

Satisfactory 5 (63%)Tolerable 2 (25%)

No improvement 1 (12%)+: Fibrous bands were observed only in cases with osseos anomalies (p<O.05).

Scalenectomy

osseous anomalies, cervicocephalic smyptoms werecommon, but paresthesia was relatively rare (p<0.05).AIso, compression due to fibrous bands wasstatistically significant in this patient group (p<0.05).In analyzing physical findings, with regarding todiagnosing the cause of compression, we found apositive hyperadduction test to be more meaningfulthan a positive hyperabduction test. All four of thepatients whose symptoms were aggravated byhyperadduction had fibrous bands (p<0.05). Therewas no statistical association between postoperativeresults and cause of compression at the interscalentriangle or applied surgical technique.

DISCUSSION

The reported incidence of TAS in the generalpopulation is approximately 0.3% to 0.7% (1,22). Theaffected typically range from 25 to 40 years in age(mean 34,6) and a higher incidence (4:1) has beenreported in females (1, 19). In our patients, the agerange was similar to that report ed by otherinvestigators, but the bias toward females was moreextreme (16:1).

Apart from the detection of radiologicallyvisible osseous anomalies, the combination of clinicalexamination and provocative testing seems the bestway to determine which of the three outlets of thoraxis the site of interest. Symptoms and findings due tocompression of all three structures in theneurovascular bundle can be aggravated either byhyperabduction, when the pectoralis minor space is

narrowed (24), or by costoclavicular com pres sion,when the costoclavicular space is diminished (7).

Although combined upper and lower plexusinvolvement occurs in 85-90% of all TAS cases (19),the symptoms and neurological findings of theinvolved plexus level and the response tohyperabduction and hyperadduction tests may varyaccording to the specific compression problem at theinterscalene triangle. The indi cation for scalenectomyrequires careful interpretation of the symptoms andphysical findings, provocative tests,electrophysiologic tests, and probably MR! (11) forpatients with x-rays that show no visible osseousanomaly. In addition, the anterior scalene muscleinjection test may be used as a diagnostic andconfirmatory test for thoracic outlet compressioncaused by the scalene muscles (21). This type ofcompression usually affects the upper plexus (C5,6,7)and relief of symptoms occurs during adduction (19).In contrast, since cervical ribs and associatedligaments act as a fu1crum beneath the plexus,exerting pressure upward against the nerves,complaints in these cases relate to the lower plexusinvolvement (C8, Th1), and thus increase duringadduction, and resolve during hyperabduction (1).We also noted positive hyperadduction tests in ourpatients with fibrous bands.

Electrophysiologic studies may help eitherconfirm the existence of compression at interscalenetriangle or predict a possible cause. Of our sevenpatients who were examined with EMG, SEP and

141

Tiirkish Neurosiirgenj 9: 137 - 143, 1999

ENG, we observed specific findings of TOS in five.Of the three tests done, EMG was most informatiye.

This may be explained by the very low incidence ofsevere atrophy and longstanding disease in ourpatients. Since EMG is able to detect the early changes(12), it was most helpful in our patients. Others formsof electrodiagnostic testing are usually accepted ascomplementary (12, 22). In the two patients whoregistered normal electrodiagnostic test results,severe compression was revealed intraoperatively.Thus, a negative finding should never exclude thepresence of TOS when the clinical picture is highlysuggestive of thoracic outlet compression (12, 22).We have no experience with the utility of MRI fordiagnosing TOS, but, this technique mayaIso behelpful for confirming the diagnosis. Panegyres. etal, demonstrated the deviation of the brachial plexusin 19 out of the 24 symptomatic sides in affectedpatients, and detected a band-like structure in 25 outof 33 sides (11).

Sometimes, it is difficult to differentiate TOS

from carpal tunnel syndrome (CTS). it is noteworthythat the symptoms, especially the parestesia, are notusually nocturnal in TOS, but are very frequently soin CTS (22). We also found that nocturnal aggravationof symptoms was ra re in our patients (18%), and thethree patients with this complaint had no carpaltunnel compression. Double-crush or reverse double­crush syndrome is more common in TOS patients(22). The treatment of another compression distal orproximal to the thorasic outlet may delay the intensityof the symptoms of thorasic outlet compression, aswell as the surgery (4). The best results are achievedwhen coexisting compressions are investigated andtreated simultaneously O). Seven of our casesexhibited varying degrees of a separate typeof neuralcompression. In three of these patients, bothcompressed areas were treated, and we achievedexcellent results in these cases. Of the other four, onlyone patient had no complaints postsurgery.

Most series report some relief of symptoms inthe 70% to 80% range, regardless of the operatiyeapproach used (2, 14, 16,20). Currently, resection ofthe first rib is the most po pul ar surgical approach(16). This is suitable for cases with signs of lowerplexus involvement, and can be performed via eithera transaxillar or the supraclavicular route. However,this is limited in that it can only treat problemsass6ciated with the lower plexus, and it is alsoassociated with considerable risks. In contrast, the

supraclavicular approach offers better visualizationand resection of the compressing structures in the

142

Erbnyraklar: Siipraclnviciilnr Approncli lo Neurogeiiic Tas

interscalene triangle, thus, a scalenectomy can beperformed when compression on the upper plexusis identified. Recent reports have demonstrated thatsignificant relief of neurogenic symptoms can beachieved with decompressive operations that leavethe first rib intact, and that this also mini miz es

complications (2, 8, 14, 15). Resection of the first ribhas been recommended only for vascular TOS (8).For our patients, we used a more selective approach,consisting of excision of the osseous and fibrousanomalies, anterior scalenectomy and brachial plexusneurolysis via the supraclavicular route, sparing thefirst rib in all cases. Direct visualization of

neurovascular and bony structures was excellent, andallowed to a precise anatomic decompression.

. Fortunately, we did not encounter any significantpostoperative complications, and our results werecomparableto those reported with other traditionaltechniques.

Correspondence: Uzm. Dr.5erhat ErbayraktarDokuz Eylül Üniversitesi Tip FakültesiNörosirürji ABD 35340 Balçova /IzmirTel: 0/232/2595959

Fax: 0/232/2787595

REFERENCES

1. Atasoy E: Thoracic outlet compression syndrome.Orthop CIin North Am 27:265-303, 1996.

2. Cheng SWK, ReiIly LM, Nelken NA, et al: Neurogenicthoracic outlet decompression: rationale for sparing thefirst rib. Cardiovasc Surg 3:617-623, 1995.

3. Dale WA: Thoracic outlet syndrome: Critique in 1982.Arch Surg 117:1437-1445,1982.

4. Daube JR: Nerve conduction studies in thoracic outletsyndrome. Neurology, 25:345, 1975.

5. Darazio RA, Ezzel F: Arterial compIications of the TOe.Am J Surg 138:246-250, 1979.

6. Etheredge S, WiIber B, Stoney GJ: Thoracic outletsyndrome. Am J Surg 138:175-182, 1979.

7. Falconer MA, Weddel LG: CostocIavicularcompression of the subcIavian ar tery and vein. Lancet2:539-543, 1943.

8. Fantini GA: Reserving SupracIavicular First RibResection for Vascular Complications of ThoracicOutlet Syndrome. Am J Surg 172:200-204, 1996.

9. Glover JL, Worth RM, Bendick DJ, et al: Evokedresponses in the diagnosis of TOe. Surgery 89:86-93,1981.

10. Juvonen T, Satta J, Laitala P, Luukkonen K, Nissinen J:AnomaIies at the Thoracic Outlet Are Frequent in theGeneral Population. Am J Surg 170:33-37, 1995.

11. Panegyres PK, Moore NiaII, Gibson R, Rushworth G,Donaghy M: Thoracic outlet syndromes and magneticresonance imaging. Brain 116:823-841, 1993.

Tiirkis/i NeltrOSiirgery 9: 137 - 143, 1999

12. Passero S, Paradiso C, Giannini F, Cioni R, BurgalassiL, Battistini N: Diagnosis of thoracic outlet syndrome:Relative value of electrophysiological studies. ActaNeurol Scand 90:179-185, 1994.

13. Pollak EW: Surgical anatomy of the thoracic outletsyndrome. Surg Gynecol Obstet 150:97-102,1980.

14. Qvarfordt PG, Ehrenfeld WK, Stoney RJ:Supraclavicular radical scalenectomy and transaxillaryfirst rib resection for the thoracic outlet syndrome. AmJ Surg 148:111-116, 1984.

15. Reilly LM, Stoney RJ: Supraclavicular approach forthoracic outlet decompression. J Vasc Surg 8:329-334,1988.

16. Roos DB: Transaxillary approach for first rib resectionto relieve thoracic outlet syndrome. Ann Surg 163:354­358, 1966.

17. Roos DB: Congenital anomalies associated withthoracic outlet syndrome; anatomy, symptoms,diagnosis and treatment. Am J Surg 132:771-778, 1976.

18. Roos DB: New concepts of thoracic outlet syndromethat explain etiology, symptoms, diagnosis andtreatment. Vasc Surg 13:313-321, 1979.

Erbnyrnklnr: Siiprnclnviciilnr Approncli lo Ne1lrogwic Tas

19. Roos DB: The place for scalenectomy and first ribresection in thoracic outlet syndrome. Surgery 92: 1077­1085, 1982.

20. Sanders RJ,Pearce WH: The treatment of thoracic outletsyndrome: a comparison of different operations. J VascSurg 10:626-634, 1989.

21. Sanders RJ: Thoracic outlet syndrome-A commonsequela of Neck Injuries. Philadelphia, JB Lippincott,1991.

22. Tindall SC: Chronic injuries of peripheral nerves byentrapment, Youmans JR (ED), NeurologicalSurgery,Vol 4, third edition, Philadelphia, W.B.Saunders, 2511-2542, 1990

23. Urschel HC, Hylan JW, Solis RM, et al: Thoracic outletsyndrome masquerading as coronary artery disease(pseudoangina). Am Thoracic Surg 16:239-248, 1973.

24. Wright is: The neurovascular syndrome produced byhyperabduction of the arms: the immediate changesproduced in 15normal controls, and the effects on somepersons of prolonged hyperabduction of the arms, asin sleeping, and certain occupations. Am Heart J 29:1,1945.

143