Thoracic Outlet Syndrome (TOS) is a group of disorders impacting the brachial plexus, subclavian artery, or subclavian vein, as they pass through the scalene triangle, costoclavicular space, and coracopectoral tunnel. The variable anatomy, biomechanics, signs and symptoms of TOS pose a particularly unique challenge to diagnosis. I can't think of another MSK condition that encompasses such widely variable and complex anatomy that defines it. The expansive region from which it arises, multiple anatomical structures that may be afflicted, and varied biomechanics of the three regions that can cause it, relates to a myriad of signs and symptoms, which can include any combination of the following:

 

1. neck pain
2. shoulder pain
3. chest pain
4. upper extremity pain
5. upper extremity numbness
6. upper extremity parasthesia
7. Raynaud's phenomenon
8. heaviness of upper extremity
9. swelling
10. claudication
11. pallor
12. coldness
13. digital ischemia and ulceration
14. cyanosis

 

In this two-part article I will unravel the complexity of TOS to add specificity to its diagnosis and improve treatment outcomes. Part 1 of this article will focus on "the challenge" of diagnosing TOS. Part 2 - coming soon - will focus on the solutions to that challenge.

 

Defining a Diagnosis

 

Before discussing TOS we should pause to define what a meaningful and specific diagnosis includes. One of my early mentors, Dale Buchberger, was fond of saying:

 

Non-Specific Diagnosis = Non-Specific Treatment = Non-Specific Outcome

 

The goal of evaluating a patient is to render a diagnosis that allows for a clear definition of the treatment according to the diagnosis. A thorough history and exam aims to define what is known as the 4-part diagnosis:

1. Stage and state of the condition - chronic, acute, irritable, non-irritable
2. Structural Diagnosis - the most probable anatomical and tissue specific diagnosis
3. Functional Diagnosis - functional factors and causal mechanisms. Although not typically written in the clinicians chart notes, a functional diagnosis should include a comprehensive understanding of the mechanisms that aggravate and ease the patients symptoms.
4. Complicating Factors - such as diabetes, heart disease, obesity, social/emotional/cognitive factors, etc.

 

We will leave the 4-part diagnosis behind for now - keep it in consideration as we move through these next sections - and we will come back to defining it in TOS terms in part 2.

 

8 Reasons TOS Poses a Unique Challenge to a Good Diagnosis

 

 

1. Distinct anatomy, and three locations: 

 

Figure 1 from: Thoracic Outlet Syndrome: A Comprehensive Review of Pathophysiology, Diagnosis, and Treatment (3)

 

a) The Scalene Triangle is a vertically narrow and triangular space formed by three borders: anteriorly by the anterior scalene - which attaches from the anterior transverse process of C3-C6 and the first rib, anterior to the subclavian groove (formed by the subclavian artery); posteriorly by the middle scalene - attaching to the posterior transverse process of C2-6 and the 1st rib, lateral / posterior to the subclavian groove; and inferiorly by the 1st rib, between its anterior and middle scalene attachments.

 

The subclavian artery enters the scalene triangle after passing through the superior thoracic aperture below.

 

The 3rd through 7th cervical and 1st thoracic spinal nerves enter the scalene triangle, but only the C5-T1 spinal nerves form the brachial plexus. The spinal nerves then form the superior trunk (C5-6), middle trunk (C7), and inferior trunk (C8-T1) within and lateral to the scalene triangle, at which point the trunks divide to become cords and enter the costo-clavicular space along with the subclavian artery.

 

b) The Costoclavicular Space is formed by borders created inferiorly by the 1st rib; antero-superiorly by the the undersurface of the clavicle the subclavius muscle. The brachial plexus and subclavian artery are joined by the subclavian vein, which conveniently omits the scalene triangle, as they pass through the costoclavicular space toward...

 

c) The Coracopectoral Tunnel - also commonly referred to as the subcoracoid space, which I find lacks description for this space - extends from inferior clavicle to the axilla. 

 

The anterior border of the coracopectoral tunnel is formed for a short section by the pectoralis major, after which the anterior border is formed the pectoris minor and fascia that envelopes it. The posterior border is  traditionally defined as the 2nd-4th ribs, however, with careful inspection I believe this is true only at the proximal, most superior-medial aspect of the tunnel. As the plexus, axillary artery and vein traverse laterally and inferiorly the rib cage soon becomes medial and inferior, and the subscapularis and glenohumeral joint (GHJ) become posterior-superior. We may recall a patient presenting after glenohumeral joint dislocation with signs and symptoms of numbness and tingling into the upper extremity. Glenohumeral joint instability and/or dislocation can injury or irritate the brachial plexus subject to its position - posterior and superior to the brachial plexus. I would also like to remind the reader that there is something attaching onto the ribs that precedes them - the serratus anterior; perhaps it is more accurate to state the serratus anterior and overlying fascia as the posterior then medial-inferior border of coracopectoral tunnel.

 

Distal to the clavicle the subclavian artery and vein are now the axillary artery and vein that traverse the coracopectoral tunnel, during which several branches arise to and from, providing blood supply to the pectoral muscles, shoulder girdle and upper extremity.

 

I encourage the clinician to look closely at the images below, as the standard definitions of the subcoracoid space and coracopectoral tunnel fail to match its complexity, which may be especially pertinent for the manual therapy practitioner. 

 

By Henry Vandyke Carter - Henry Gray (1918) Anatomy of the Human Body (See "Book" section below)Bartleby.com: Gray's Anatomy, Plate 523, Public Domain, https://commons.wikimedia.org/w/index.php?curid=559532

 

 

From Edgelow PI: Neurovascular consequences of cumulative trauma disorders affecting the thoracic outlet: a patient-centered treatment approach. In Donatelli RA, editor: Physical therapy of the shoulder, ed 4, St Louis, 2004, Churchill Livingstone. Courtesy of Peter Edgelow.)

 

 

The brachial plexus starts its journey into the thoracic outlet as spinal nerves entering the scalene triangle, where they become the superior (C5-6), middle (C7) and inferior trunks (C8-T1). The dorsal scapular and long thoracic nerves arise from spinal nerves C5 and C5-6-7, respectively. The suprascapular nerve branches from the superior trunk (C5-6). As the plexus passes under the clavicle and through the costoclavicular space the trunks divide into the posterior, lateral and medial cords, and gives rise to several peripheral nerves (see figure above). I encourage you to consider mechanisms that may impede good function of one, or many of the nerves passing through this region, and the symptoms that may present if that were the case.

 

 

By Chris Talbot - Own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=22561787

 

 

 

2. Variant Anatomy:

 

Congenital variations in anatomy such as a scalene minimus or fascial bands can cross the scalene triangle. They can pass between nerve roots or over the inferior trunk of the brachial plexus and can cause entrapment to the brachial plexus or compression on the subclavian artery. There can be anomaly's to the 1st ribs, or the addition of cervical ribs (complete or partial; osseous or fibrous) may also contribute to TOS. These anatomical variants can occur in people who do not have TOS, but they are considered to increase the risk of developing it and it may complicate recovery in some. Stewman et al (6) report that 20% of patients with nTOS have a cervical rib, the presence of which can also increase the risk for subclavian artery stenosis or aneurysm. 

 

 

3. Three distinct mechanisms: 

 

 

a) Scalene Triangle: There are both non-traumatic and traumatic mechanisms at the scalene triangle. Trauma, including whiplash injury, can strain the scalenes and brachial plexus, resulting in fibrosis and scarring and potential entrapment neuropathy. Non-traumatic mechanisms relate to overactivity and/or hypertrophy of the scalenes, often associated with excessive elevation of the upper ribs secondary to breathing pattern and/or stabilization system dysfunction. Of course a traumatic mechanism may invoke a protective pattern that may also create overactivity of the scalene muscles, breathing and stabilization dysfunction. 

 

b) Costoclavicular Space: Both traumatic and non-traumatic mechanisms may occur. Traumatic mechanisms relates to clavicular fracture with resulting inferior displacement of the clavicle or hypertrophic callus formation. This can occur acutely, at the time of injury to the clavicle, but can also present weeks to months after injury. I have had several patients who reported no symptoms until 3-6 months after injury, presenting when resuming daily or sporting activities. Non-traumatic mechanisms involve repeated or prolonged exposure to inferior and posterior translation of the clavicle in scapular depression and retraction. Posterior rotation of the clavicle may also be a contributing factor, and coincides with scapular retraction. Hyperabduction is also a described mechanism due to extension, posterior rotation, and potentially depression of the clavicle, with the potential for neurovascular compression. This is commonly associated with athletes in overhead sports (i.e. baseball). It is worth noting the subclavian artery, vein and plexus move a lot when this happens. Case reports with imaging have demonstrated the tortuous route of the subclavian artery/vein during elevation of the shoulder girdle, causing compression and stenosis. 

 

c) Coracopectoral Tunnel / Subcoracoid Space: Neurovascular compression also
occurs at the subcoracoid space by stretching the overlying pectoralis minor
muscle when the arm is brought overhead - especially when combines with horizontal extension and external rotation. Excessive tension of the pectoral muscles, restriction or fibrosis within the pectoral or axillary fascia can accentuate this compression. Glenohumeral anterior translation, instability, or a history of dislocation can exacerbate this mechanism, along with the potential for fibrosis and entrapment that may impact the brachial plexus passing by. Similar to the costoclavicular space, the axillary artery and vein are also subject to stretch and strain in positions of abduction, extension and external rotation. 

 

 

4. Three anatomic structures:

 

Classification systems have divided TOS based on the impaired anatomical structures, including: Neurogenic (nTOS), Arterial (aTOS), and Venous (vTOS). The estimated distribution of each variant are: 

• nTOS encompasses 85-95%
• vTOS 3-10%
• aTOS 1-5%
  

 

The diagnosis of vascular TOS (vTOS) indicates a referral for diagnostic imaging and potential medical intervention. Due to this fact, not to mention the much higher prevalence of nTOS and its plentiful complexity; we will focus the majority of our attention in this article to nTOS. Before doing so, I will outline the key signs and symptoms that lead to the diagnosis vascular TOS.

 

vTOS is more likely to include symptoms #8-14 from the above list, but may also include symptoms #1-7. Exam findings in a vTOS patient can include:

 

• supraclavicular pulsatile mass +/1 subclavian bruit on auscultation
• a blood pressure differential of ≥20 mmHg
• upper extremity swelling and cyanosis (2).

 

In the presence of these signs and symptoms I will advise the advise the patient according to positions of relief and provocation and engage limited treatment to promote the same, while pursuing further evaluation via referral to their medical provider. A sudden onset of swelling and cyanosis of the upper extremity may indicate effort vein thrombosis - “Paget-Schroetter” syndrome - and warrants urgent referral. 

 

Note that Raynaud's phenomenon can be caused by neurodynamic dysfunction and may be present in patients with nTOS; therefore Raynaud's on its own is insufficient to conclude a vascular TOS diagnosis. 

 

 

5. Variable Symptom Presentation - Variable Sections of Brachial Plexus Involvement:

 

 

Sanders et al (1) reported the symptoms of 50 consecutive TOS cases. Although it is unclear from the article, based on the authors discussion and symptoms, I believe all 50 of these patients fall within the category of neurogenic TOS.

 

We should remind ourselves that patients; depending on the stage, state, location and mechanism of their TOS diagnosis; may present with any one, two, several, or all of the symptoms Sanders et al reported. As an example to the former, I have had patients with TOS present with:

 

• lateral elbow, dorsal radial wrist and 1st-2nd finger pain, parasthesia and partial numbness (mimicking radial tunnel syndrome or C6 radiculopathy)
• anterior lateral arm, elbow to forearm pain and parasthesia (mimicking C6 radiculopathy or musculocutaneous nerve entrapment)
• medial distal forearm to 4-5th finger pain, parasthesia and numbness (mimicking C8 radiculopathy or cubital tunnel syndrome)
• neck, upper back and shoulder pain (mimicking a number of conditions including C5 radiculopathy, postural dysfunction with rotator cuff syndrome, etc)

 

 

6. Overlapping Mechanisms and Less Than Perfect Diagnostic Tests:

 

Injury and provocation of the neural structures from nerve roots to plexus and upper extremity peripheral nerves can respond negatively to: 

 

• movement of the neck, shoulder girdle, and arm
• tension or pressure on the brachial plexus 

 

This brings me to the commonly accepted diagnostic tests, which are all provocative maneuvers for nTOS: the Upper Limb Tension Test (ULTT), Elevation Abduction Stress Test (EAST), and Adson's test.

 

The ULTT is a sensitive test for mechanical sensitivity of the brachial plexus, proximal nerve roots, and median nerve. It is a highly sensitive test for nTOS. However, it is not specific.

 

The ULTT is synonymous with the Median Neurodynamic Test 1 (MNT1). This test uses the median nerve - the largest peripheral nerve in the upper extremity with the greatest effect on the brachial plexus and nerve roots. The MNT1 can provoke symptoms related to mechanical sensitivity of the C5-T1 nerve roots, brachial plexus and its distal extensions. In a patient with whole hand symptoms, an abnormal MNT1 can help confirm the diagnosis nTOS.  However, in more isolated symptom presentations an abnormal ULTT may only provide confirmation of a mechanically sensitive nerve that is intolerant to a tension mechanism. For example, envision a patient that presents with lateral elbow to radial wrist and 1st-2nd finger pain and parasthesia. You perform an ULTT and provoke the patients chief complaint symptoms. You now have evidence the nerves that innervate structures of the lateral elbow to 1st-2nd fingers are mechanically sensitive. But you have no evidence to delineate between C6 radiculopathy and nTOS.

 

The Elevation Arm Stress Test (EAST), also referred to as Roo's test, combines sliding and lengthening of the brachial plexus and proximal nerve roots, tensioning of the pectoral muscles and clavi-pectoral fascia, and closing the costoclavicular and subcoracoid space. The patient is instructed to hold this position while opening and closing their hands, for up to 3 minutes. The clinician may palpate the radial artery at the wrist during this maneuver, but evidence suggests a high false positive rate in normal subjects. Pesser et al studied the EAST test via retrospective analysis of 428 patients and demonstrated a positive predictive value ranging between 65% and 66%, and the negative predictive value between 53% and 58% (5). 

 

 

 

Adson's test involves ipsilateral neck rotation + extension with extension of an outstretched arm, while the practitioner evaluates the radial pulse. The proposed mechanism is elevation of the 1st rib, increased tension in the scalene muscles, and depression/extension of the clavicle. Unfortunately, Adson's test has not faired well in research. It has both poor sensitivity and specificity, and a high number of asymptomatic subjects will have a diminished radial pulse during the test. Sanders comments that "The Adson Test of noting a radial pulse deficit in provocative positions has been shown to be of no clinical value and should not be relied upon to make the diagnosis of any of the three types. The test is normal in most patients with nTOS and at the same time can be positive in many control volunteers."

 

Despite its poor performance, I would like to point out potentially relevant mechanics of the Adson's test as it relates to cervicothoracic and scapulothoracic function. In good function the 1st-2nd rib arthro/osteokinematics should roll into extension + posterior rotation, which should reduce the amount of closing at the costoclavicular space. In poor stabilization the cervicothoracic spine and associated upper ribs can paradoxically move to flexion and lack normal posterior rotation/extension, which further closes the costoclavicular space. There is also some neural tension during this test, as the extended arm draws tension distally on the plexus.

 

 

7. Double crush mechanisms:

Another quote from my former mentor, Dale Buchberger: 

 

"A patient maintains the right to have more than one thing"

 

Double crush, whereby neural compression or irritation is occurring at two or more locations, within and outside the aforementioned locations included in the diagnosis of TOS. Also consider the possibility of referred pain from somatic structures - glenohumeral joint, facet joints, intervertebral discs, ligaments, muscles and fascia. 

 

Postural stabilization, breathing and movement pattern dysfunction influences the region of the cervicothoracic spine, rib cage and shoulder girdle. Therefore one patient with one underlying cause - significant postural stabilization dysfunction - may present with multiple areas compromised by this function, including the cervical intervertebral discs and facet joints, associated intervertebral foramen, and, the thoracic outlet.

 

 

8. Considerations with Diagnostic Testing:

 

• all patients with suspected vascular TOS are referred for imaging. 
• mild to moderate nTOS with no signs of vascular compromise and no motor deficit is treated with a trial of conservative care for 3 months before imaging is indicated. The patient is carefully monitored for response to treatment; an unresponsive case of nTOS is referred for imaging. The decision to wait for imaging varies depending on chronicity and persistence of symptoms, age of the patient, prior care, and of course exam findings that lead me to suspect 'something else is going on' or a soft diagnosis that needs confirming. When in doubt, get imaging. When symptoms are persistent and/or unresponsive to care, get imaging. Radiological imaging is low risk and inexpensive; it is often well worth the added specificity in diagnosis and treatment. However, it is essential we understand anatomical variants such as cervical ribs or degenerative changes are frequently present AND asymptomatic. Response to treatment trumps imaging findings any day of the week.
• For both neurogenic and vascular TOS - X-ray is first in line - to inspect morphology of the cervical and upper thoracic spine, ribs and clavicle.
• For vascular TOS - Diagnostic ultrasound is first in line, and maintains a reasonable degree of sensitivity and specificity, is inexpensive and readily available. MR angiography can add more detail and differentiate cases the diagnostic ultrasound cannot confirm.
• For neurogenic TOS - nerve conduction studies and EMG, and MR neurogram may be ordered.
• Nerve conduction studies and EMG testing are NORMAL in the majority of nTOS patients. When positive, it indicates greater severity and compromise. Recent evidence suggest nerve conduction studies focusing on the medial antebrachial nerve are more sensitive and specific to nTOS (3).

 

In Conclusion: 

There's a lot of possibilities in patient presentation, relevant anatomy and biomechanics of the thoracic outlet, making a specific and meaningful diagnosis challenging. It is enough to drive you a little batty 😳.

 

Of course GOOD patients present with the typical signs and symptoms, and respond to diagnostic tests in a typical manner. "Order me up a side of true positives and negatives, please!". But this is not the reality of clinical practice. Many patients defy us with variations that defy the 'norm'. We define them as atypical because we don't see them as often. Which is why relying on statistics and probabilities can really mess with good diagnostic skills. On the one hand, knowing our statistical probabilities based on patient presentation and diagnostic testing can be super helpful and necessary. On the other hand, if we rely to heavily on them, we close our eyes to the less likely or not as common presentations, leading to missed diagnoses.

 

In part 2 of this article I will aim to reduce the complexity of this TOS hot-mess. See you then!

 

1. Sanders RJ, Hammond SL, Rao NM. Diagnosis of thoracic outlet syndrome. J Vasc Surg. 2007 Sep;46(3):601-4. doi: 10.1016/j.jvs.2007.04.050. PMID: 17826254.
   
   
2. Povlsen S, Povlsen B. Diagnosing Thoracic Outlet Syndrome: Current Approaches and Future Directions. Diagnostics (Basel). 2018 Mar 20;8(1):21. doi: 10.3390/diagnostics8010021. PMID: 29558408; PMCID: PMC5872004.
   
   
3. Jones, M.R., Prabhakar, A., Viswanath, O. et al. Thoracic Outlet Syndrome: A Comprehensive Review of Pathophysiology, Diagnosis, and Treatment. Pain Ther 8, 5–18 (2019). https://doi.org/10.1007/s40122-019-0124-2
   
   
4. Delaune LA, Wehrli L, Maeder Y, Vauclair F, Moerenhout K. Acute brachial plexus deficit due to clavicle fractures. JSES Int. 2020 Oct 31;5(1):46-50. doi: 10.1016/j.jseint.2020.09.004. PMID: 33554163; PMCID: PMC7846699.
   

5. Pesser, Neils, et al. Reliability and validity of the elevated arm stress test in the diagnosis of neurogenic thoracic outlet syndrome. Clinical research study | Thoracic outlet syndrome. Volume 76, Issue 3p814-820September 2022.

6. Stewman C, Vitanzo PC, Harwood MI. Neurologic thoracic outlet syndrome: summarizing a complex history and evolution. Curr Sports Med Rep. 2014;13(2):100–6.