Clinical Research

PresentedMeeting in San

No compet

Division oHopkins Medic

CorrespondSurgery, RossE-mail: klikes1

Diagnostic Accuracy of Physician andSelf-referred Patients for Thoracic OutletSyndrome Is Excellent

Kendall Likes, Danielle H. Rochlin, Quinn Salditch, Thadeus Dapash, Yen Baker,

Roxanne DeGuzman, Shalini Selvarajah, and Julie Ann Freischlag, Baltimore, Maryland

Background: The purpose of this study was to categorize patients referred to a specializedthoracic outlet syndrome (TOS) practice to determine the diagnostic accuracy of those whoare physician and self-referred.Methods: Demographic and clinical data on all patients who were referred for TOS between2006 and 2010 were retrospectively reviewed from a prospectively maintained institutional re-view boardeapproved database and patient records.Results: Between 2006 and 2010, 621 patients were referred for TOS (433 women and 188men; mean age 39 years [range 10e87]). Five hundred seventy-one patients (92%) were diag-nosed with TOS, with 421 (74%) neurogenic, 126 (22%) venous, and 24 (4%) arterial TOScases. Of the 525 physician referrals, 478 (91%) had TOS, and of the 93 self-referrals, 90(97%) had TOS. The 421 patients with neurogenic TOS (NTOS, 304 women and 117 men)had symptoms on average for 56 months (range 1e516). Two hundred seventy-one patients(64%) were initially treated with TOS-specific physical therapy (PT), and 100 (37%) improved.One hundred seventy-eight patients (42%) underwent a lidocaine block, and 145 patients(81%) had a positive block. Seventy-four patients (18%) underwent Botox injections 44 (60%)of which were positive and the average number of Botox injections was 1.3. One hundred fortypatients (33%) underwent transaxillary first rib resection and scalenectomy (FRRS), and 128patients (91%) improved. Of patients undergoing FRRS, 92 (66%) had a lidocaine block, 82(89%) of which were positive. Of patients with a positive lidocaine block, 74 (90%) improved afterFRRS. Of patients undergoing FRRS, 31 (22%) underwent Botox injections, 15 (48%) of whichwere positive. Of patients with a positive Botox block, 14 (93%) improved after FRRS. Averagelength of time between initial visit and operation was 6.4 months (range 2 weeks to 34 months),and average follow-up time was 13 months (range 1 week to 49 months).Conclusions: 1) Both referring physicians and patients are very accurate in their preliminarydiagnosis of TOS (neurogenic, venous, or arterial TOS). 2) In a specialized TOS practice,two-thirds of patients are sent to TOS-specific PT and one-third improve from this treatmentalone. 3) One-third of patients referred for NTOS eventually undergo FRRS with a 91% successrate.

at the Peripheral Vascular Surgery Society AnnualFrancisco, California on May 29, 2013.

ing interest declared.

f Vascular Surgery and Endovascular Therapy, Johnsal Institutions, Baltimore, MD.

ence to: Kendall Likes, BS, Johns Hopkins Department of759, 720 Rutland Avenue, Baltimore, MD 21205, USA;@jhmi.edu

Ann Vasc Surg 2014; -: 1–6http://dx.doi.org/10.1016/j.avsg.2013.12.011� 2014 Elsevier Inc. All rights reserved.

Manuscript received: November 13, 2013; manuscript accepted:

December 20, 2013; published online: ---.

1

2 Likes et al. Annals of Vascular Surgery

INTRODUCTION

Thoracic outlet syndrome (TOS) is a clinical disorder

that can manifest with neurogenic, venous, and/or

arterial symptoms. The most common subtype is

neurogenic TOS (NTOS, 95%)which often develops

in adults at their 20s to 40s, with a greater preva-

lence among women. These patients often present

with pain and numbness resulting from compres-

sion of the brachial plexus by the anterior scalene

muscle, the first rib, or fibrous bands.1 Often ana-

tomic compression is due to history of traumatic in-

juries or chronic and repetitive motion activities.2

Patients presenting with venous TOS (VTOS, 4%)

can present with subclavian vein thrombosis, most

frequently as an acute episode of unilateral arm

swelling and discoloration. These symptoms are

frequently associated with a history of vigorous

physical activity but can also be associated with an

underlying hypercoagulable disorder.3,4 Arterial

TOS (ATOS) is rare (1%), and patients often present

with acute or chronic ischemia of the arm. Patients

with ATOS have a high prevalence of bony abnor-

malities, such as cervical or rudimentary ribs.5

Treatment strategies for the various TOS subtypes

remain controversial and are often complex, espe-

cially in the neurogenic subtype. The diagnosis

and treatment can be particularly difficult, and as

more patients are being diagnosed as having TOS,

more are being referred for treatment options. Little

is known about the means by which high-volume

institutions, and more specifically specialized TOS

practices, manage a large number of referrals.

Thus, there is a clear need to better understand

the presentations, treatment algorithms, and out-

comes in the population of patients presenting

with symptoms of TOS. Our institution treats a large

volume of TOS patients, and thus, we were able to

examine a sizable cohort of patients referred for

TOS. The detailed analysis and characterization of

patients referred for TOS will help guide the course

of treatment for patients presenting with TOS. The

data and outcomes reports will help to identify can-

didates for further treatment, such as physical ther-

apy (PT), lidocaine and/or Botox blocks, and first rib

resection and scalenectomy (FRRS), all of which

have been shown to be effective treatment strategies

for TOS.

METHODS

Data Collection

A retrospective review of a prospectivelymaintained

databasewas performed. Patient datawere compiled

from January 1, 2006 to December 31, 2010. This

database has approval from the Johns Hopkins Insti-

tutional Review Board to follow the presentation,

treatment course, and outcome of patients present-

ing with symptoms of TOS at the Johns Hopkins

Medical Institutions. The database includes patient

demographic and clinical characteristics, such as

initial presentation, previous therapies, and treat-

ment course, in addition to patient outcomes, post-

operative complications, and patient follow-up.

This database is prospectively maintained by the

Johns Hopkins Division of Vascular Surgery and

Endovascular Therapy. Any additional information

was obtained from the electronic patient records.

Patient Selection Criteria

Patients were included in this study if they were

either self-referred or referred by a physician for

symptoms of TOS. Patients were diagnosed as hav-

ing VTOS and ATOS using duplex scanning and

the presence of symptoms consistent with compres-

sion, thrombosis, or ischemia. Patients were diag-

nosed as having NTOS if they had symptoms of

prolonged pain, weakness, numbness, or tingling

in the neck, shoulder, arm, and/or ulnar distribu-

tion; had a positive elevated arm stress test; had a

history of trauma or repetitive motion injury; and/

or if they experienced symptomatic relief and

decreased discomfort with a scalene block 3e5 hrs

after the procedure. Lidocaine injections were used

as a diagnostic measure to predict whether patients

would improve after an FRRS procedure. Botox in-

jections were used for painmanagement in older pa-

tients and in patients who needed to delay surgery.

Surgical Management

Patients referred to our institution with VTOS and

ATOS underwent preoperative duplex scanning. Pa-

tients with NTOS underwent 8 weeks or more of PT

before being considered for FRRS. TOS-specific PT

included exercises and maneuvers to stretch the

scalenemuscle, increase range ofmotion, strengthen

the back muscles and core, and massage the soft tis-

sue.2 Patients who did not see improvement after

this protocol underwent FRRS through a transaxil-

lary approach.

Patients with VTOS underwent venography at

2 weeks postoperatively to assess subclavian vein

patency. Those with >50% stenosis of the vein un-

derwent balloon dilatation and anticoagulation.

Those with chronic occlusion received anticoagula-

tion therapy. Those with a patent subclavian vein

(<50% stenosis) received no further interventions.

Vol. -, No. -, - 2014 Determination of diagnostic accuracy for TOS 3

Follow-up included clinic visits at 1, 3, and

6months and 1 and 2 years postoperatively. In addi-

tion to an assessment of symptoms in all patients,

follow-up testing for patients with vascular symp-

toms included duplex scans at postoperative visits

to assess vessel patency and flow velocities.

Fig. 1. Life table analysis displaying rate of improve-

ment, as defined by freedom from symptoms, after

FRRS for patients with at least 12-month follow-up

with neurogenic thoracic outlet syndrome.

RESULTS

Six hundred twenty-one patients were referred for

TOS between January 2006 and December 2010.

There were 433 women and 188 men with an

average age of 39 years (range 10e87 years). Of

these 621 patients, 571 patients (92%) were diag-

nosed with TOS. Of these, 421 (74%) were diag-

nosed with NTOS, 126 (22%) with VTOS, and 24

(4%) with ATOS. Of the 525 physician referrals,

478 (91%) were diagnosed with TOS, and of the

93 self-referrals, 90 (97%) were diagnosed with

TOS during their assessment.

Neurogenic TOS

The 421 patients who had NTOS included 304

women (72%) and 117 men (28%) with an average

age of 40 years (range 10e82 years). Of the neuro-

genic patients, 389 (92%) were referred by their

physician and the other 32 (8%) were self-referred.

These patients experienced neurogenic symptoms

for an average of 56 months (range 1e516 months)

before presentation. Of the 421 patients with NTOS,

140 (33%) underwent transaxillary FRRS, and 128

(91%) improved with surgery. The average length

of time between presentation to clinic and FRRS

operation was 6.4 months (range 0.5e24 months).

Of patients undergoing FRRS, 10 (7%) were found

to have a prominent cervical rib that was also

removed during the operation. Neurogenic patients

were followed for an average of 16.3 months after

FRRS (median 13 months, range 0.25e87 months;

life table analysis, Fig. 1).

Of the neurogenic patients, 271 (64%) were first

sent to TOS-specific PT, and of these, 100 (37%)

improved fromPT alone. Thosewho did not improve

after PT were considered for surgery. Of this patient

subset, 67 (39%) underwent transaxillary FRRS and

65 (97%) improved after the operation.

Of patients presenting with NTOS, 178 (42%) un-

derwent a lidocaine block. Of patients who under-

went a lidocaine block, 145 (81%) had a positive

block, 27 (15%) had a negative block, and the

remaining patients had an unknown outcome. Of

patients undergoing surgical intervention for NTOS,

92 (66%) had undergone a lidocaine block before

surgery, 82 (89%) of which exhibited a positive

block. Of patients undergoing transaxillary FRRS

who had a positive lidocaine block, 74 (90%)

improved after the operation. Of the 27 patients

with a negative block result, 7 (26%) underwent

transaxillary FRRS and 6 (86%) of them improved

after surgery.

Of the neurogenic patients, 74 (18%) underwent

Botox injections. Of these, 44 patients (60%) had a

positive block, 19 patients (26%) had a negative

block, and the outcomes of the 11 remaining pa-

tients’ blocks were unknown. The average number

of Botox injections administered was 1.3, and suc-

cessive injections were given 3 months apart. Of pa-

tients undergoing transaxillary FRRS, 31 (22%) had

undergone Botox injectons before surgery. Of these,

15 patients (48%) exhibited a positive Botox block

before surgery. Of patients with a positive Botox

block, 14 (93%) improved after surgical interven-

tion. Of the 19 patients with a negative Botox block,

11 (58%) underwent transaxillary FRRS, and 9

(82%) of them improved after the operation.

Venous TOS

The 126 patients who were diagnosed with VTOS

included 77 women (61%) and 49 men (39%) with

an average age of 33 years (range 15e65 years). Of

the venous patients, 113 (90%) were referred by a

physician, whereas the other 13 (10%) were self-

referred. These patients experienced vascular symp-

tomsanaverageof22months (range1e324months).

Of patients diagnosed with VTOS, 13 (10%) were

found to carry an underlying hypercoagulable disor-

der. The average length of time between the initial

Fig. 2. Life table analysis displaying rate of improve-

ment, as defined by freedom from symptoms, after

FRRS for patients with at least 12-month follow-up

with venous thoracic outlet syndrome.

Fig. 3. Life table analysis displaying rate of improve-

ment, as defined by freedom from symptoms, after

FRRS for patients with at least 12-month follow-up

with arterial thoracic outlet syndrome.

4 Likes et al. Annals of Vascular Surgery

presentation to clinic and surgery was 1.8 months

(range 1 day to 12 months). Ultimately, 89 patients

(71%) found to have VTOS were operated on, and

77 (87%) of these improved in the postoperative

period. Of patients undergoing FRRS to relieve

compression of the subclavian vein, only 1 had a

prominent cervical rib that was also removed during

the operation. Venous patients were followed for an

average of 19.1 months (median 14 months, range

0.5e70 months; life table analysis, Fig. 2).

Arterial TOS

The 24 patients who were diagnosed with ATOS

included 15 women (63%) and 9 men (37%)

with an average age of 29 years (range 12e54 years). Of the arterial patients, 22 (92%) were

referred by a physician, whereas the other 2 (8%)

were self-referred. These patients experienced

symptoms for an average of 31.7 months (range

1e144 months). Of the patients with ATOS, 18

(75%) underwent transaxillary FRRS to relieve

arterial compression, and 16 (89%) improved in

the postoperative period. Of patients undergoing

FRRS to relieve compression of the subclavian ar-

tery, 7 (39%) had a prominent cervical rib that

was also removed during the operation. The

average length of time between their initial clinic

visit and date of surgery was 2.3 months (4 days

to 10 months). Arterial patients were followed for

an average of 17.3 months (median 12 months,

range 1e71 months; life table analysis, Fig. 3).

One patient diagnosed with ATOS underwent

FRRS but was lost to follow-up.

Other Diagnosis than TOS

There were 50 patients presented to clinic with TOS-

like symptoms who were not diagnosed with TOS.

There were 33 women and 17men, with an average

age of 47 years (range 14e87). Of these, 47 (94%)

were referred by a physician and 3 (6%) were self-

referred. These patients were found to carry a num-

ber of different diagnoses, including 15 patients

(30%) with other neck or shoulder problems, 13 pa-

tients (26%) with other vascular issues such as arte-

rial occlusive disease or arteritis, 6 patients (12%)

with other pain syndromes, 5 patients (10%) with

musculoskeletal injury or pain, 3 patients (6%)

with other nerve issues, and 3 patients (6%) with

neurologic issues. These patients were referred to

other specialists (orthopedic, neurosurgery, neurol-

ogist, or pain management) to create an appropriate

treatment plan for their discomfort and pain.

DISCUSSION

Little is known about the spectrum of TOS referrals

encountered and managed by a high-volume insti-

tution and more specifically by a specialized TOS

practice. We examined a large cohort of patients

referred to our practice for TOS to determine trends

in clinical diagnoses, treatment patterns, and thera-

peutic outcomes. The detailed analysis and charac-

terization of patients referred for TOS and the

outcomes reported in our study serve to help iden-

tify candidates for further management and guide

the course of treatment for patients presenting

with symptoms of TOS.

Vol. -, No. -, - 2014 Determination of diagnostic accuracy for TOS 5

Our analysis reveals a number of patterns con-

cerning patient presentation and referral. On

average, patients with NTOS were older than pa-

tients with VTOS (P� 0.05) or ATOS (P� 0.05). Pa-

tients with NTOS were on average 40 years old as

compared with patients with VTOS, who were on

average 33 years old, and patients with ATOS,

who were 29 years old on average. In addition, pa-

tients with NTOS have a longer duration of symp-

toms than those with either ATOS (P � 0.05) or

VTOS (P � 0.05) and fewer have surgery.

In our patient population, just 33% of patients

diagnosed with NTOS underwent transaxillary

FRRS, whereas a greater percentage of patients

with VTOS and ATOS underwent surgical interven-

tion to relieve vascular compression 71% (P� 0.05)

and 75% (P � 0.05), respectively. For many pa-

tients, neurogenic symptoms can be relieved with

more conservative treatments such as TOS-specific

PT2 and/or scalene muscle blocks.6 In our patient

population, neurogenic patients had symptoms for

an average of 56 months before presentation,

whereas venous and arterial patients had symptoms

for an average of 22 and 31.7 months, respectively.

Prior studies have suggested that duration of symp-

toms is associatedwith a poorer outcome.7,8 Howev-

er, our study indicates that surgical management

yields excellent results in all patients with all sub-

types of TOS.

We find that patients with vascular symptoms are

treated more aggressively than patients with neuro-

genic symptoms. On average, patients presenting

with VTOS and ATOS waited a shorter amount of

time before surgical intervention, 1.8 months

(P � 0.05) and 2.3 months, (P � 0.05) respectively,

as compared with patients with NTOS, who waited

for an average of 6.4 month filled with further PT

and scalene blocks before undergoing FRRS. All pa-

tients presentingwith venous symptoms underwent

our standard protocol of transaxillary FRRS and

postoperative venography at 2 weeks with possible

balloon dilatation and/or anticoagulation therapy,

which has been shown to rapidly resolve symptoms

so that patients can quickly return to baseline activ-

ity levels. Patients with arterial symptoms should

also be treated aggressively to prevent complications

such as axillosubclavian aneurysms and emboliza-

tion from subsequently developing. Patients with

NTOS should initially be treated conservatively

with PT before undergoing FRRS for symptom relief

because, in a specialized TOS practice, two-thirds of

patients are sent to TOS-specific PT, and of those,

one-third improve from PT alone.9

We emphasize the importance of easy access to

office staff by phone or e-mail for the purpose of

referral. In our practice, the patient coordinator is

a specialized informed professional working with

and under the direction of the physician. Although

the physician is primarily responsible, the coordi-

nator supports, facilitates, and manages daily pa-

tient interaction and plays a critical role in the

conduct of the practice. The coordinator is knowl-

edgeable and seeks to aid patients who contact us

by phone or e-mail to discuss symptomatology and

discern treatable candidates from individuals un-

likely to harbor a diagnosis of TOS. The individual

helps to prevent poor outcomes by helping to iden-

tify factors predictive of success and failure with our

TOS treatments and protocols. In our practice, we

also stress the importance of obtaining and using pa-

tient records ahead of time to direct treatable indi-

viduals to our practice. By performing these duties,

the coordinator works with the physician, depart-

ment, and institution to support the many facets of

the practice.

We find that surgical management in patients

with all 3 forms of TOS yields excellent outcomes

and a quick return to baseline activity levels. This

compares favorably to previously recorded out-

comes of surgical intervention for patients with

various types of TOS.8,10e13 The surgery success

rates are comparable among groups. In our patient

population, 91% of neurogenic patients, 87% of

venous patients, and 89% of arterial patients

improved after transaxillary FRRS. We follow all 3

groups of patients for about the same amount of

time; in this study, neurogenic patients were fol-

lowed postoperatively for an average of 13 months,

venous patients were followed postoperatively for

an average of 16 months, and arterial patients

were followed postoperatively for an average of

13.2 months. We have found that it takes patients

w1 year to return to baseline activity after FRRS.9,10

Both referring physicians and patients are very

accurate in their preliminary diagnosis of TOS.

We found that 92% of physician referrals were

confirmed to have TOS, suggesting that our medi-

cal and surgical colleagues are successfully making

the appropriate diagnoses and referrals. Similarly,

the even higher rate of accurate patient self-

referrals (P < 0.05) suggests that patients are

educated about TOS. It is our belief that this is

occurring through social media and/or other

Internet-based sources. In our practice, we support

the use of Web-based social media as a diagnostic

aid. New media tools are reengineering the way

patients obtain information, and we have found

that social media, such as YouTube,14 can be

used as an informative and educational platform

to empower patients.

6 Likes et al. Annals of Vascular Surgery

CONCLUSIONS

Both referring physicians and patients are accurate

in making the diagnosis of NTOS, VTOS, and

ATOS. Utilization ofmedical literature, socialmedia,

and electronic communication may play a role in

informing patients and physicians about TOS. A

knowledgeable intake staff member is critical to

screening patients and organizing their records

before their visit. Excellent outcomes are seen in pa-

tients with NTOS, VTOS, and ATOS using both con-

servative treatments and surgical intervention.

REFERENCES

1. Brooke B, Freischlag J. Contemporary management of

thoracic outlet syndrome. CurrOpinCardiol 2010;25:535e40.

2. Fugate MW, Rotellini-coltvet L, Freischlag JA. Current man-

agement of thoracic outlet syndrome. Curr Treat Options

Cardiovasc Med 2009;11:176e83.3. Martinelli I, Cattaneo M, Panzeri D, Taioli E, Mannucci PM.

Risk factors for deep venous thrombosis of the upper ex-

tremities. Ann Intern Med 1997;126:707e11.

4. Likes K, Rochlin D, Nazarian SM, Streiff MB, Freischlag JA.

Females with subclavian vein thrombosis may have an

increased risk of hypercoagulability. JAMA Surg 2013;148:

44e9.5. Chang KZ, Likes K, Davis K, Demos J, Freischlag JA. The sig-

nificance of cervical ribs in thoracic outlet syndrome. J Vasc

Surg 2013;57:771e5.

6. Lum Y, Brooke B, Likes K, et al. Impact of anterior scalene

lidocaine blocks on predicting surgical success in older

patients with neurogenic thoracic outlet syndrome. J Vasc

Surg 2012;55:1370e5.

7. Yavuzer S, Atinkaya S, Tokat O. Clinical predictors of surgi-

cal outcome in patients with thoracic outlet syndrome oper-

ated on via transaxillary approach. Eur J Cardiothorac Surg

2004;25:173e8.

8. Altobelli GG, Kudo T, Haas BT, Chandra FA, Moy JL,

Ahn SS. Thoracic outlet syndrome: pattern of clinical success

after operative decompression. J Vasc Surg 2005;42:122e8.

9. Chang DC, Rotellini-Coltvet LA, Mukherjee D, De Leon R,

Freischlag J a. Surgical intervention for thoracic outlet syn-

drome improves patient’s quality of life. J Vasc Surg 2009;

49:630e5. discussion 635e7.

10. Rochlin DH, Gilson MM, Likes KC, et al. Quality-of-life

scores in neurogenic thoracic outlet syndrome patients un-

dergoing first rib resection and scalenectomy. J Vasc Surg

2010;57:436e43.

11. Rochlin DH, Likes KC, Gilson MM, Christo PJ, Freischlag JA.

Management of unresolved, recurrent, and/or contralateral

neurogenic symptoms in patients following first rib resection

and scalenectomy. J Vasc Surg 2012;56:1061e8.

12. Roos D, Owens C. Thoracic outlet syndrome. Archives of

Surgery 1966;93:71e4.

13. Degeorges R, Reynaud C, Becquemin JP. Thoracic outlet

syndrome surgery: long-term functional results. Ann Vasc

Surg 2004;18:558e65.

14. Johns Hopkins Medicine (2011, October 4). What is thoracic

outlet syndrome, its diagnosis and treatment? j Q&A. Re-

trieved August 21, 2013, from http://www.youtube.com/

watch?v¼35UI4Pmzpoc.