How to truly identify and treat thoracic outlet syndrome (TOS)

The infamous thoracic outlet syndrome. TOS is considered to be one of modern medicine’s most difficult issues, because of the complex and variable nature of its symptoms. It has potential to cause numerous types and areas of pain, such as neuralgia in the arms, chest, between the shoulder blades and in the back (figure 1), dizziness, brain fog, migraine, headaches, a feeling of being “heavy-headed”, etc.

The reason why the potential symptoms are all over the spectrum, is because it in addition to compression of the entire brachial plexus nerve network which innervates the arms as well as parts of the chest, neck and back, also may compress the subclavian artery & vein. A branch of the subclavian artery include a key vessel, the vertebral artery. The VA supplies the brain with blood, and is therefore especially important to assess for symptoms of vertebrobasilar insufficiency. It has also been shown that TOS may cause secondary dysautonomic symptoms.

Despite more than 2600 references to TOS on pubmed, there is still wide controversy regarding TOS; no concrete diagnostic criteria have been established, and many practitioners claim that the whole problem is a fad which does not really exist. Below are some interesting quotes related to thoracic outlet syndrome.

Fig. 1

This article is concerned with thoracic outlet compression syndrome (TOCS), one of the most controversial subjects in medicine. It may also be the most underrated, overlooked, misdiagnosed, and probably the most important and difficult to manage peripheral nerve compression in the upper extremity. – Atasoy, 1996

This review was complicated by a lack of generally accepted diagnostic criteria for the diagnosis of TOS. Thoracic outlet syndrome is one of the most controversial diagnoses in clinical medicine. – Povlsen et al., 2014

Thoracic outlet syndrome (TOS) is controversial in terms of definition, anatomy, aetiology and treatment. – Redman & Robbs, 2015

Actually it [TOS] is not widely known and it is also a controversial issue for some physicians. Moreover, it is sometimes strongly denied by those who have not had the opportunity of identifying it as a disease or even when they have not dealt with TOS patients….  We are confronted with a disease that is commonly undiagnosed by the majority of physicians. – Silva & Selmonosky, 2011

The diagnosis of neurogenic TOS is more challenging because its symptoms of nerve compression are not unique – Sanders et al., 2008

Conversely, no valid standard diagnostic test is available for disputed neurogenic TOS, resulting in controversies in the frequency of TOS diagnosis – Hooper et al., 2010

Diagnosis and treatment of thoracic outlet syndrome (TOS) involves neurologists, physiatrists, family physicians, orthopedic surgeons, vascular surgeons, thoracic surgeons, neurosurgeons and sometimes psychiatrists.  – Köknel, 2005

The most commonly recommended interventions are strengthening and stretching of the shoulder girdle musculature.2,7,19,21 However, little agreement exists on which muscles need strengthening and which ones need lengthening.5 These types of exercises do not detail how they address functional TOS as a result of respiratory alterations and they do not aim to inhibit muscle.1,5,19 – Robey & Boyle, 2009

Neurogenic thoracic outlet syndrome (NTOS) is an oft-overlooked and obscure cause of shoulder pain that regularly presents to the office of shoulder surgeons and pain specialists. – Boezaart et al., 2010

Taking the research above into account, the reader can probably start to understand that it’s often very difficult to be properly diagnosed and treated if one has thoracic outlet syndrome. This article will shed light on what I consider a very effective approach to both diagnosis and treatment, that have cured thoracic outlet syndrome for most of our patients. Be aware though, that the actual treatment is a demanding procedure that will have to be managed through cooperation with a qualified therapist.

DISCLAIMER: This article is written for educational purposes only

Causes and consequences

The name thoracic outlet syndrome suggests chronic irritation (compression) of the brachial plexus and the subclavian vessels, as mentioned initially. “Thoracic” means region of the thorax (chest), and “outlet” is self explanatory. Why do they become irritated or compromised? Due to continuous compression within spaces that the nerves and vessels pass through. The cause of the compression is mainly tightness of the surrounding muscles and clavicular depression, strangulating the thoracic outlet vascular and nervous structures.

It is generally accepted that TOS is caused by compression of brachial plexus elements or subclavian vessels in their passage from the cervical area toward the axilla and proximal arm either at the interscalene triangle, the costoclavicular triangle, or the subcoracoid space – Köknel, 2005

Fig. 2

In turn, the main cause of the the muscle tightness and clavicular depression, is faulty movement and postural strategies. This can be rooted in habits alone, or triggered by injuries such as a clavicular fracture (Moon Jib Yoo et al., 2009; Ishimaru et a., 2012; Connolly & Dehne, 1989), whiplash injury (Schenardi, 2005) or similar. Slouching of the neck (forward head posture) and shoulders (Vanti et al., 2007), belly-(only)-breathing (Simon & Travell, 1999), and lack of diverse movement will cause the scalenes that form the interscalene triangle of which the brachial plexus pass through, to inhibit/deactivate. This in turn may cause severe tightening of the scalenes, compressing all of the thoracic outlet’s structures and may thus (with potential) cause all of the initially mentioned symptoms.

The (anterior and medial) scalenes are involved in many actions. They elevate the ribs during inspiration (inhalation), ipsilaterally rotate, cause lateral translation, laterally flex and forward flex (bend) the neck. In normal breathing patterns, the ribs and clavicle should elevate slightly during inspiration, and this is done in syncronization by the scalenes, trapezius and several other muscles. Severe slouching habits will inhibit this pattern as well as proper cervical (axial) rotation, causing degeneration of the involved muscles. In turn, severe inhibition of the scalenes will often develop over time.

Because the trapezius muscle holds the scapula and clavicle, the loss of optimal function of this muscle will cause chain reactions of muscular inhibition down the line (arm), creating the potential for several nervous and vascular entrapment points, such as the triangular interval in the posterior shoulder. This is especially important when there is pre-compression within the scalenes and costoclavicular passage, as this sensitize the whole nervous chain and make the distal branches more vulnerable to additional irritation.

Additionally, because the scalenes attach to the ribs, they may elevate the first rib, greatly increasing the potential of secondary compression between the 1st rib and the clavicle. Compression within the scalenes often attribute to between 60-80% of the patients’ direct symptoms in my experience; a considerable amount.

Fig. 3 – Venstre utløp

The myth of  mandated muscular “release”

The weaker a muscle gets, the tighter it will feel. The patient may feel like stretching a steel wire that won’t budge when stretching a weak and inhibited muscle. It may get better for an hour or so, but then comes back with a vengeance. … And sadly, most repeat this process over and over until the only choice left is surgery. I have also addressed this topic in my lumbar plexus compression syndrome article.

Fig. 4 – Stretching is NOT the solution to your problems!

Extreme muscular inhibition will cause severe abrasiveness and tightening, greatly increasing its potential of irritating / compressing nearby structures such as nerves and blood vessels. In my experience, it’s a great and even potentially dangerous myth to assume that these tight muscles are “over active” and mandate release. This will make them even weaker and even tighter, as they are exposed to a stress that they can not handle.

Garrick and Webb1 in their excellent book, Sports Injuries: Diagnosis and Management, state that a weak muscle is a tight muscle. Whenever a weak muscle is forced to work beyond its capacity, it will tighten and, therefore, be more subject to stress and strain. – Warren Hammer, 1990

Bluntly, the myth of stretching (releasing) is one of the main reasons why most therapists are not able to cure thoracic outlet syndrome (or other nervous compression issues of muscular origin, for that matter) with conservative measures. The muscles that entrap the nerves and vascular structures must be strengthened significantly, so that they no longer reflexively tighten due to the unduly stress they’re exposed to.

That said, I can understand why people still do it. The stretching makes the client feel better! While strengthening on the other hand, makes it feel worse. And we want it to feel better, right? Wrong! The reason the strengthening makes it feel worse, is because the muscles are so utterly weak that any stimulus will cause exacerbation of the symptoms. They have minimal work capacity, which is why they severely tighten and irritate the surrounding nervous structures. The only way (that I know of) to deal with this, is slowly rehabbing the muscles by strengthening them steadily and easily over time.

Here’s an ultrasound image of a patient’s scalenes, clearly showing atrophy (degeneration w. fatty infiltration) of the muscle, especially the anterior scalene. Massaging such extremely weakened muscles will only exacerbate the situation. Note the difference in echogenicity between the sternocleidomastoid (scm) and scalenes (white structures = fat; the muscle should be relatively dark).

Fig. x

To systematically evaluate the muscles’ functions, it’s necessary to a testing tool. Such a tool is manual muscle testing (MMT), palpation, and strengthening exercises which are specific to the point of entrapment. I recommend David Weinstock’s book “Neurokinetic Therapy”, as it demonstrates the MMT tests well.

I have three rules that need to be fulfilled before I decide to release a muscle. If the muscle in question fits all of these rules, it’s probably safe to release. These “safe” (read: relatively healthy) muscles are usually not relevant to the patient’s complaint, in my personal experience, which is why I don’t perform releases all that often (many may, of course, disagree with this).

  1. The respective muscle is thick (hypertrophy)
  2. It has a high muscle tone (contractile status when resting)
  3. Tests very strong on a muscle test

Identification and correction

So far, the key points that we have talked about are:

  • The importance of proper cervical and clavicular posture, and breathing patterns
  • That the main compression occurs between the scalenes, and beneath the clavicle
  • That the muscles causing the entrapment are usually extremely weak, and not over-active

It is absolutely critical to establish proper breathing habits, clavicular resting position and cervical posture, in order to resolve thoracic outlet syndrome. Supplementary, strengthening of all the involved inhibited structures should take place. Let us now go into detail about the underlying causes of all of these elements, and how they can be corrected.

Swayback posture (SBP)

Swayback posture is the most common stabilisation strategy I see utilised by clients with thoracic outlet syndrome. When the pelvis is tucked down and in (posterior pelvic tilt, lumbosacral flexion), it causes a shift in the body’s gravitational points so that the mid back hyperextends and the shoulders and head comes forward. It is almost impossible for a client to change their head and shoulder postural habits without addressing the root cause of it all, namely the pelvic tucking and thoracolumbar hinging. Additionally the pelvic tucking and forward head posture may cause breathing dysfunction, as it causes gripping of the abdominal muscles, making it hard to breathe diaphragmatically, and because it depresses the clavicle (as mentioned earlier).

I have written extensively about the topic of correcting swayback posture numerous times in my other norwegian articles, but also in this lower back article in english. Therefore it will not be elaborated further in this article, but it is paramount that the reader understands the chain reactions of pelvic misalignment on the head, neck and shoulders. And, of course its relation to breathing dysfunction.

Fig. 5

Spotting forward head posture is not difficult, but spotting clavicular and scapular misalignment on the other hand is often missed even by experienced therapists. As I have said already, the key to solving forward head posture is correcting pelvic and thoracic alignment. We will have a closer look on clavicular and scapular misalignment patterns, and how it can be identified and corrected shortly. But first, some elaboration with regards to swayback posture and breathing dysfunction is necessary.

How to correct your posture:

SBP and breathing

As we have already seen, SBP will affect our breathing strategy. Hanging forward with the head and slouching with the shoulders will inhibit the scalenes’ ability to elevate the ribs during inspiration, exacerbating the dysfunction.

To assess breathing, lie down comfortably on the back and evaluate whether or not there is adequate thoracic vertical expansion during moderate breathing intensity. Often, a very reduced vertical expansion will be noted. This can also be compared to standing up. A 50/50% expansion of the abdomen and thorax is a well-balanced way to go, in my experience. To help this, it will be beneficial to strengthen the muscles that assist in thoracic inspiration: The sternocleidomastoid, scalenes, (and sometimes the pectoralis minor, but this will absolutely require proper scapular stability first)

Fig. 6

Many breathing “experts” claim that diaphragmatic (belly)-breathing is the ultimate cure to virtually anything. Ok, I am exaggerating a little, and I agree that diaphragmatic breathing ability is important, but teaching the client to reduce thoracic expansion may often lead to detrimental consequences (I learned this the hard way!). Thoracic expansion is normal, and abdominal expansion is normal. We need both.

So, in addition to the strengthening work that was mentioned above, we will of course need to work directly on our breathing habits. First, make sure that the clavicle is properly positioned (read more on that below). Then, try to make the thorax and abdomen expand in all 360 degrees as you inhale, getting into a calm rhythm of balanced respiration. If this is too difficult for you, either find a coach or work solely on thoracic vertical expansion, as this is most important element for resolving TOS. Remember that the clavicle should elevate gently as you breathe in, and gently depress as you breathe out. This cycle will need to be practiced over and over until it feels more normal or occurs automatically.

Proper scapular / clavicular resting position

The scapula should be located between the T2 and T7 vertebrae, with its superior angle levelled with T2 on the longitudinal line. This may however be cheated, by anteriorly rotating the scapula, which is a main trait when in slouching shoulders. Thus it is very important to be aware that the scapula should also be in mild upward and posterior rotation while positioned in height with T2 & T7. Most of the time, however, the scapula is so depressed that even with anterior rotation it will not be in line with T2, such as with the person in the picture below.

A typical TOS patient will often present with similar scapular resting position, as many studies (cited below) also show. The superior scapular angle is significantly inferior (lower than) the T2 vertebrae, and they rest in considerable anterior and downward rotation. Scapular depression and anterior tilt will cause the clavicle to jam into the brachial plexus and subclavian vessels, compressing them. Additionally, (as mentioned) inhibition of normal breathing patterns, cervical posture and rotation. A terrible combination that’s almost always found present in clients with thoracic outlet syndrome.

These patients are often cued by their therapist to pull the shoulders “back and down”, but this is very harmful and must never be done, as it causes compression of the costoclavicular space, and may result in nerve damage. Please consider that “back and down” is a provocative (orthopaedic) test for costoclavicular space syndrome (Magee, DJ. Orthopedic physical assessment, 2014).

Fig. 7

Regardless of what you have heard, no amount of strengthening will solve this problem. It’s rooted in habits, and must be corrected primarily by habitual changes. In practice that means relearning proper scapular resting position, by raising them into the proper height and rotational alignment and staying there. It’s hard work, but well worth it. Don’t get me wrong though; strengthening work is important. It’s just much less important than optimization of habits. I’ve written more about the scapular positioning topic in this shoulder pain article.

Optimal resting position should look something like the picture below. Ignore the muscle size, it is not important nor a criteria for proper positioning. However; the trapezius is clearly active, holding the scapula in proper height while also upwardly & posteriorly rotating it. This will ensure that the clavicle rests above the thoracic outlet, instead of crushing into it.

Fig. 8

Why you should NEVER pull the shoulders “back and down”

How to correct improper scapular and cervical positions:

Below are some relevant citations.

In our experience, droopy shoulder syndrome has accounted for most cases of thoracic outlet syndrome but is largely unrecognized by physicians. Recognition of this syndrome should lead to a better understanding of the underlying pathophysiology and prevent unnecessary surgery. – Swift & Nichols, 1984

Here’s a large quote collection from Watson et al., 2010 regarding the scapula’s relation to thoracic outlet syndrome. A great article that’s worth reading.

Thoracic outlet syndrome (TOS) is a symptom complex attributed to compression of the nerves and vessels as they exit the thoracic outlet. Classified into several sub-types, conservative management is generally recommended as the first stage treatment in favor of surgical intervention. In cases where postural deviations contribute substantially to compression of the thoracic outlet, the rehabilitation approach outlined in this masterclass will provide the clinician with appropriate management strategies to help decompress the outlet. The main component of the rehabilitation program is the graded restoration of scapula control, movement, and positioning at rest and through movement.

The conservative physiotherapy regimen outlined in this article will be suitable for patients presenting with TOS where there is a strong postural contribution to their symptoms. In particular, in cases of TOS where the scapula mechanics are poor and the patient presents with the dropped shoulder condition (scapula depressed and/or downwardly rotated, and/or anteriorly tilted) (Ranney,1996).

Many forms of scapula asymmetry may well exist in TOS populations, but in the limited research that has been done, scapula or shoulder girdle depression or “drooping” has been consistently observed (Kenny et al., 1993; Walsh, 1994; Pascarelli and Hsu, 2001; Skandalakis and Mirilas, 2001). Scapula depression will lead to

Scapula depression will lead to an alteration of the anatomical alignment of the structures in both the cervical and thoracic outlet (Telford and Mottershead, 1948; Kai et al., 2001; Skandalakis and Mirilas, 2001) (Fig. 2). It may potentially lead to tractional stress being placed on the nerve, vascular and muscular elements as well as compression as the clavicle descends closer towards either the first rib or any other bony element present. Elevation of the shoulder girdle can alleviate these stressors and potentially lead to “decompressing” the thoracic outlet (Kitamura et al., 1995).

One of the consistent objective findings that we have observed and measured in cases of sTOS is that the scapula can be depressed at rest (Fig. 3) on the symptomatic side compared to the other side (in unilateral TOS) and to the normative data in cases of bilateral TOS (Kai et al., 2001). Increased anterior tilt of the scapula is also commonly identified in sTOS (Sucher, 1990; Aligne and Barral, 1992; Press and Young, 1994; Walsh, 1994) and it is frequently coupled clinically with increased downward rotation of the scapula. – Watson et al., 2010

Neurogenic TOS

The symptoms of TOS may greatly vary. They are not unique, and this is one of the main reasons why making a diagnosis is difficult. Many of the same clues are however often present, and this is what we need to use as a measure of probability.

Neurogenic TOS is very easy to trigger, and this is tremendously helpful while diagnosing and identifying nervous entrapment points down the branches of the brachial plexus. To do this, I use a pressure-testing “technique” as means of provocation. The concept is simple: Push into the entrapment point and see if it reproduces the pain. This is called a positive Tinel’s sign. If it does, this is a region that’ll need corrections. There are potential entrapment points all the way down the arms, in the route of the nervous branches. We will now look more closely on these, and how each branch can be addressed.

The tinel’s sign has been shown to have poor specificity in the literature, but because plexopathic problems are so controversial, there is not reason to rely on this. The Tinel’s sign is a very good indicator of entrapment. If it hurts, there is a problem. It should not hurt! A reason why surgeons require high specificity testing for TOS (although such does not exist) is simple: They do not want to operate unless clearly warranted. This understandable! However, with proper conservative treatment, such risks are not present, and we need to be so wary of false positives. If it hurts, we strengthen the muscle which is most likely to irritate the nerve. Risk free!

With regards to diagnosis of N-TOS, it has been shown that EMG, NCV and MR neurographies are not reliable diagnostic criteria (Tolson 2004, Passero 1994, Veilleux 1988, Aminoff 1988, Rousseff 2005, Kwee 2014) There have also been reports of EMGs only being positive when the patient is in certain positions (Fishman 2002), and reports that motor nerve NCVs have been negative while sensory segments positive (Machanic 2008). I have seen several patients with severe pain upon pressure to the interscalene triangle, positive myotome tests etc., who still did not have any findings upon EMG. The sensitivity of these tests are simply inadequate and should not be used to exclude pathology. They may be used to quantify the problem, once already implicated, however.

NCV can be prolonged by injury or simple extrinsic pressure against a nerve.41 NCV prolongation is especially seen in patients with long-standing NTOS that results in muscle atrophy.42 However, other articles have reported that NCV is often normal in patients with symptoms of NTOS.42,43

Somatosensory evoked potentials studies have been found useful in some reports.46,47 However, somatosensory evoked potential has also been criticized as nonspecific, nonlocalizing, and rarely abnormal.43,44,48

Findings showed denervation activity, increased mean action potential amplitude, and/or duration and reduced recruitment at maximum effort. However, there is still some question as to whether EMG is adequately sensitive to detect changes in NTOS patients with milder symptoms.42,45 – Sanders et al., 2008

Somatosensory evoked potentials (SEPs) are used in the diagnosis of thoracic outlet syndrome (TOS), even as an indication for surgery. The purpose of this study was to evaluate the use of SEPs in the diagnosis of TOS. Twenty-one patients (mean age, 37 years) with TOS and 23 control subjects (mean age, 34 years) were included. Somatosensory evoked potentials of median and ulnar nerves were measured bilaterally in patients in both a relaxed and arms-elevated provocative position. A three-way analysis of variance showed no significant difference between the interpeak latencies of the TOS and control groups (p = .352). Significant differences were found in testing positions (p = .0014) and nerve tested (p = .001) in both groups. Therefore, this study suggests that SEPs are not helpful in the diagnosis of TOS. – Komanetsky et al., 1996

Fig. 9

Brachial plexus compression sites

Compression directly to the brachial plexus is the most common driver of thoracic outlet syndrome. It is comprised of two main entrapment zones, which are the interscalene triangle and the costoclavicular passage.

The interscalene triangle is usually the main entrapment point (culprit), and will often stand for 60-80% of the patients’ symptoms. When scalenes are very very tight, they also elevate the first rib, furtherly reducing the space between the rib and the clavicle, increasing the potential for compression  within the costoclavicular passage. Compressive forces within the interscalene triangle will affect all of the thoracic outlet’s structures and may thus cause all of the symptoms that were mentioned in the beginning of this article.

Fig. 10

To evaluate the scalenes’ involvement, the therapist pushes the thumb into the brachial plexus, in the middle of the distal anterior and middle scalene fibers. Start light and gradually go hard(er), to see if the symptoms reproduce. This is called the Morley’s test (Sanders 2007, Laulan 2011). You need to push directly into the brachial plexus. If you miss the right spot on a patient with TOS, you’ll get a false negative. It’s a generally a good idea to move the thumb around a little to make sure that your test results are accurate.

Some may argue that pressure directly into a muscle that lies on top of a nerve, always will cause nervous symptoms, but this is NOT true. I use these tests almost every day, and they will show up negative if there is not nervous irritation in the region you’re testing.

If the pressure reproduce the symptoms, you’ll want to muscle test (MMT) the surrounding muscles. For the anterior scalene, resist above the eyebrow while client the head toward the shoulder. Medial scalene, resist at temple while client moves head toward the shoulder. Watch out for clenching of the jaw, breath-holding, etc, as the body would try to cheat and use any synergist rather than the scalenes to protect the already irritated brachial plexus from the activation of the scalenes.

If the pressure test reproduced the pain but the scalenes test strong, most of the time that means the test is skewed. MMT is a skill that takes time to develop, but is extremely useful when you get good at it. Tell the patient to relax and to resist your pressure naturally, without engaging all the muscles of the neck. The point here is to assess the specific muscles’ functions, not to “win”.

Fig. 11-12 – Scalenus anterior (left) & medius (right) MMT


Here are the exercises for scalene strengthening. Make sure that the person doing it starts very, very easy. Going on hard on these exercises may trigger tremendous pain and significant worsening of the symptoms. It should get a little worse as the scalenes are worked, but not cause excruciating pain. 10 reps for 1-2 sets once per day is usually a safe start.

To check for entrapment within the costoclavicular passage, I use a clavicular depression test. Either with the patient sitting, or supine, the therapist strongly depresses the shoulder manually to see if this will reproduce the pain. Again, a strong pressure will usually be required. If the test reproduce the pain, which it often will if the scalenes are affected, this means that the clavicle is too posturally depressed and is irritating the thoracic outlet within the costoclavicular passage. This test can also be falsely negative if there is numbness of the nerves (a consequence of long term compression), so don’t rely fully on it.

As I’ve said many times now, this is a postural and breathing related issue. The trapezius may be strengthened by performing shrugs or similar exercises, but the habitual changes are what will yield long lasting results in this case. Raising the shoulders slightly in posture (and staying there) will decompress the thoracic outlet. Optimization of thoracic vs. diaphragmatic breathing balance will also stimulate the scalenes, as mentioned earlier.

Fig. 13

If the costoclavicular space (CCS) is compromised, which is more serious than muscular entrapment (as bones will be compressing the nerves, as opposed to myofascial irritation), there will usually be subsequent myotome weakness. Myotome testing is therefore important to do on these patients, to evaluate the degree of compression. Most commonly, the inferior trunk of the brachial plexus will be affected. This is because it lies most anteriorly of the trunks, making it more susceptible to compression. Inferior trunk  compression will usually cause weakness of the 5th finger (ulnar nerve), and sometimes triceps and axillary nerves (radial and axillary nerves). Sometimes the middle trunk may be affected as well, which causes weakness of the biceps (musculocutaneous nerve). The median nerve is rarely affected by costoclavicular space compression (superior trunk).

Myotome tests:

  • 1st finger opposition – Median nerve – Superior trunk
  • Biceps – Musculocutaneous nerve – Middle trunk
  • Lateral deltoid – Axillary nerve – Inferior trunk, middle trunk
  • Triceps – Radial nerve – Inferior trunk
  • 5th finger opposition – Inferior trunk
  • Finger abduction – Inferior trunk

If significant weakness is discovered, it is an utmost high priority to decompress the CCS.

Back to Tinel’s sign. You can also push into the pectoralis minor to see whether it reproduce any symptoms or not. This test, however, is not all that useful. The subcoracoid space-compression (beneath pectoralis minor) is rarely a big player in the dysfunction, and will almost always resolve on its own when the posture, scalenes and clavicle have been corrected.

The retropectoralis minor space is a very rare potential site of compression. – Demondion et al., 2006

As I mentioned earlier, postural dysfunction will cause scapular instability. This, in turn, will often cause a chain reaction of inhibition down the lines of the arm, as these structures mostly depend on the stability of the scapula to be able to generate force safely. It’s very important to also address these secondary sites of compression. What you’ll likely come to notice is that “carpal tunnel syndrome” and similar issues are often just a secondary TOS-symptom. Let’s have a closer look at these secondary sites of compression, and how they can be assessed and corrected. These principles also apply if TOS is negative, it is just not as common.

The radial & axillary nerve compression sites

The main compression site for the radial nerve, is within the triangular interval and between the fibers of the supinator muscle. To test for affection, squeeze your thumb into the interval in the posterior armpit, and/or into the supinator muscle. If pain is reproduced, you can evaluate the muscles that surround the nerve’s function by using palpation and MMT. Weakness and hypotonus of the teres minor, lateral & long heads of the tricep will usually be present for the posterior shoulder. It’s virtually always appropriate to initiate a strengthening protocol on these structures.

Fig. 14

The axillary nerve passes through the quadrangular interval, and will usually be compressed between the teres minor and teres major. As usual, squeeze into the interval with your thumb to see whether the symptoms reproduce. If they do, you can MMT the teres major and minor, or just initiate a strengthening protocol right away as they’ll test weak anyway. Strong, healthy muscles are rarely responsible for neuralgia.

Fig. 15

As explained, the supinator and triangular interval are by far the most common regions of radial nerve compression. The two most useful MMTs are provided here, for the teres minor and supinator muscles. To test the supinator, client resist the therapist’s attempt to pronate his wrist. For the teres minor, the same principle, but by resisting internal humeral rotation.

Fig. 16-17 – Supinator MMT (left), Teres minor MMT (right)


Exercise for the lateral rotators

Exercise for the supinator muscle

The median nerve compression sites

Most of the same principles of both identification and correction apply to the median nerve. The entrapment points of the median nerve are underneath the pronator teres muscle, and within the carpal tunnel.

The carpal tunnel is a little different than the rest of the compression points in this article. Eleven tendons pass through the CT, and even slight hypertrophy of these will greatly reduce the space within the tunnel. Flexor dominance will lead to hypertrophy, and may thus lead to strangulation of the median nerve within the carpal tunnel. Stretching the finger flexors followed by strengthening of the finger and wrist extensors may be a very beneficial and rewarding protocol.

Fig. 18

Pronator teres “syndrome”. Squeeze into the pronator teres and see whether it reproduces median neuralgia. If it does, MMT it by having the client resist your attempt to supinate their wrist. If it’s weak, strengthen it with the exercise provided in the video about wrist supination and pronation, further up.

Fig. 19 – MMT for the pronator teres

The musculocutaneous nerve compression sites

Musculucutaneous nerve compression often cause misleading symptoms in the lateral arm, mimicking radial nerve pain. It’s actually quite common, but it took me some time to figure this out. The nerve passes through the coracobrachialis, and then between the biceps and brachialis muscles. The patient will often lack significant medial humeral rotation when the MCN is affected, often appearing to be a mobility problem at first.

To evaluate compression between the biceps, squeeze into the distal biceps. If symptoms reproduce, test the biceps and brachialis muscles. Commonly I find that the biceps are weak and brachialis is strong, in which you may release the brachialis and strengthen the biceps (remember to force supination during elbow flexion). No absolutes, though.

Fig. 20

The same assessment protocol applies to the coracobrachialis. Elevate the arm and squeeze into the musculocutaneous nerve. If this reproduces the pain, test the muscle. If it’s weak, and it usually is, strengthen it.

Fig. 21 – Coracobrachialis MMT

Myotome tests

Nerve compression neuropathy may lead to muscle weakness. Such weakness in the sequela of neuropathy is called a positive myotome test. The best way to evaluate myotomes are to look for relative weaknesses, as utter paralyzation is usually not present. Weakness is usually not a cause of muscular entrapment, but rather of costoclavicular space compression (i.e. osseous compression of the brachial plexus).

The inferior trunk of the brachial plexus lies most susceptible placed within the costoclavicular space, i.e. more forward. With depression of the scapulae, this may cause weakness of the fifth finger and finger abduction (C8 and T1 nerve roots). Thus relative weakness of the fifth finger with regards to opposition and abduction (Selmonosky 1981, 2002, 2008) is a good criteria for detection of TOS. Such weakness indicates inferior trunk compression unless there is C8 or T1 radiculopathy (disc herniation).

As the problem progresses, weakness of the triceps and wrist flexors (radial nerve, C7 nerve root) and medial deltoid (C5 nerve root) may occur. Usually the median nerve is not affected (weakness of the 1st finger). There may sometimes be weakness of the biceps (musculocutaneous nerve, C5-6 nerve roots).

Compare the affected and unaffected sides to evaluate relative weakness and thus estimate degree of weakness sequelar to nerve compression. As mentioned, if there is weakness, the most common cause is costoclavicular space compression (depressed scapulae and/or scapular dyskinesis).

Fig. 22 – Source:

The ulnar nerve compression sites

The ulnar nerve is often just a side effect from the compression in the thoracic outlet. However it may be slightly compressed beneath the flexor carpi ulnaris muscle, and within the arcade of struthers which is a passage between the medial triceps and medial intermuscular septum. When the medial triceps is weak, the struther’s passage tightens, often causing the typical neuralgic symptoms of the meidal elbow and into the little- and ring fingers.

The same protocol applies: Test the medial tricep and FCU. If they’re weak, strengthen them by performing elbow extensions in slight lateral humeral rotation and wrist flexion with ulnar deviation. The medial tricep can be tested by having the patient resist elbow flexion while in slight lateral humeral rotation. The FCU, by having the patient resist wrist extension by flexing it with ulnar deviation.

Fig. 22

Once in a while, the pressure test will be positive but the MMT truly negative. This generally means that the compression is stemming from another structure, and that the area that you’re working on is not that important. No absolutes, though. Use MMT, palpation and provocative pressure tests to find the answers.

Arterial / Vascular TOS

Arterial thoracic outlet syndrome is thought to be very rare. It has infact been estimated that approximately 95% of the thoracic outlet syndrome cases are related to neurogenic symptoms (Wilbourn et al., 1990). But, how reliable is this estimate? First of all, neurogenic TOS is in general misdiagnosed, overlooked, etc even though it is the most easily triggered type of pain. Arterial TOS is much more subtle, and may mimic many other issues. It is therefore extremely difficult to quantify its involvement and thus, in my view, highly unlikely that this estimate is reliable.

The most common symptoms of arterial and/or venous TOS is:

  • Dizziness / vertigo
  • Migraine and/or headaches
  • Cold hands and arms
  • Feeling “heavy-headed” or as if wearing a tight helmet
  • Difficult to concentrate
  • Fatigue
  • More rare; ischemia of the arm or hand
  • Also rare; swelling

Most of these symptoms may have several other potential causes, which is why you need to do a probability estimate of whether thoracic outlet compression may be involved or not. One factor that often holds true, is visible increase of pressure in the external jugular vein. This is almost always caused by tightness of the SCM and scalenes, and/or depression of the clavicle (we now know that these two often go hand in hand), as it compresses the subclavian artery and thus compromises these structures. You can also have the patient elevate the arm, then evaluate whether or not the radial pulse diminishes, which would indicate compromisation of blood flow and thus also arterial TOS.

Fig. 23 – Source:

Selmonosky (1981, 2002, 2008) describes a simple test for brachial ischemia or cyanosis which involves maximal elevation of the arms. The patient leaves the arms up for 1-2 minutes, and the therapist looks for a White hand sign (WHS), which implies cadaveric paleness of the affected hand, usually along with tiredness and/or pain. There may also be venous insufficiency, causing venous distention and purpuric skin color indicative of cyanosis.

Fig. 24

If the posture, breathing, and neurogenic pressure-testing all have indications of dysfunction, and of course that the patient presents with additional vascular symptoms, they may very well be caused by vascular thoracic outlet compression. I noticed this connection especially as some clients were complaining of dizziness and migraine-like symptoms during strengthening regimes for the scalenes.

I did some research and realised that compression of the subclavian artery had been shown to greatly compromise the vertebral artery, that supplies the brain with blood. Here are some interesting quotes.

Thoracic Outlet Syndrome (TOS) causes dizziness because of positional compression of the vertebral artery with resultant symptoms of vertebrobasilary insufficiency. Compression of  7,C8,and T1 nerves fibers is responsible for the neck pain. – Selmonosky, 2007

The cases of 17 patients with vertigo, tinnitus, deafness, supraclavicular bruit, and a diminished radial pulse are reported. All the patients had an anomalous vertebral artery. All had subclavian-vertebral arteriograms preoperatively. Each patient showed an anomaly of the vertebral artery system which allowed intermittent compression of either the origin or cervical course of the artery. The compression was usually aggravated by rotation or hyperextension of the neck. In most cases, the vertebral artery arose at the level of the thyrocervical trunk and the compression was relieved by section of the scalenus anticus muscle and by division of the inferior thyroid artery. – Powers et al., 1961

We report a patient who developed occasional vertigo when turning his head to the right side. Occlusion of the right vertebral artery occurred at the narrowed “scalenovertebral angle” with this rotational head movement. This triangular tunnel consisted of the hypertrophied ligament of the longus colli muscle and the anterior scalene muscle. – Kojima et al., 1985

Rotation-induced vertebrobasilar artery hypoperfusion causes transient ischemic attacks (TIAs), affecting the cerebellum, brainstem and spinal cord. When these symptoms occur transiently due to head movement, compression of the vertebral artery by an extraluminal lesion should be suspected. Cervical spondylotic spurs and anterior scalene muscle or deep cervical fascia are among the factors which can compress the vertebral artery. – Dadsetan & Skerhut, 1989

Rotational positioning of the head showed vertebral obstruction in one direction, and unobstructed filling of the vessel when the head was turned to the opposite side. Fifteen patients showed rotational vertebral artery occlusion. The site of obstruction occurred at the origin of the vertebralartery or cephalad to the level of C5. Two patients had bilateral fascial band obstruction, one patient had left only, and the remaining 10 were obstructed on the right side. An anterior scalenotomy was done with preservation of the phrenic nerve. We were more impressed with the deep cervical fascia as the cause of intermittent rotational obstruction rather than the anterior scalene muscle. Only two patients showed unequivocal poststenotic dilatation as evidence of severe anterior scalene muscle compression. The obstructing extra-luminal fascia was quite dense, fibrotic and often completely encircling the artery. – Hardin & Poser, 1963

Subclavian steal symptoms presents secondary to arterial insufficiency, created by a retrograde flow that “steals” blood from the brain circulation, more specifically from the basilar artery via the vertebral artery. Classically it presents with neurological symptoms from the posterior brain and cerebellum [4,6]. Decreased flow over the basilar artery gives rise to symptoms like lightheadedness, ataxia, vertigo, dizziness, confusion, headache, nystagmus, hearing loss, presyncope and syncope, visual disturbances, focal seizures, and in extremely rare cases, death [610]. However the vast majority of patients are asymptomatic and rarely require any intervention [3,5,11]. – Alcocer et al., 2013

This article describes migraine without aura since childhood in a patient with bilateral cervical ribs. In addition to usual migraine triggers, symptoms were triggered by neck extension and by arm abduction and external rotation; paresthesias and pain preceded migraine triggered by arm and neck movement. Suspected thoracic outlet syndrome was confirmed by high-resolution bilateral magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA) of the brachial plexus. An unsuspected aberrant right subclavian artery was compressed within the scalene triangle. Left scalenectomy and rib resection confirmed the MRI and MRA findings; the scalene triangle contents were decompressed, and migraine symptoms subsequently resolved. – Saxton et al., 1999

Thoracic outlet syndrome (TOS) refers to the compression of the neurovascular bundle within the thoracic outlet. Cases are classified by primary etiology-arterial, neurogenic, or venous. In addition to the typical symptoms of arm swelling and paresthesias, headaches have been reported as a potential symptom of TOS. In this report, we describe a patient with debilitating migraines, which were consistently preceded by unilateral arm swelling. Resolution of symptoms occurred only after thoracic outlet decompression. Patients with migraines and concomitant swelling and/or paresthesias, especially related to provocative arm maneuvers, should be considered a possible atypical presentation of TOS and evaluated in more detail. – Chahwala et al., 2017

It is also noteworthy that the hypertrophied and contracted anterior scalenus muscle exerts a strong although intermittent compression of the vertebral artery, causing in severe TOS diverse symptoms that are very characteristic of vertebrobasilary insufficiency. – Silva & Selmonosky, 2011

Reports of transient blindness resulting from this condition are even more rare. The authors describe the case of a middle-aged woman who presented with transient blindness when she turned her head excessively to the left. She also exhibited other less severe brainstem symptoms. Arteriography demonstrated occlusion of the left vertebral artery only when her head was rotated to the left. Surgical exploration revealed entrapment of the left vertebral artery by a tight anterior scalene muscle, release of which resulted in complete resolution of her symptoms. – Sell et al., 1994

Triggering the symptoms may be a little challenging. If it’s headaches, try to rotate and flex the head contralaterally while in cervical extension and lying supine, to tighten the scalenes around the thoracic outlet. The therapist may also force the clavicle caudally. Hold this for a few minutes and have the patient stand up. Dizzy? Drowsy eyed? Heavy-headed? If any relevant symptoms appear after the provocation, that’s a confirmation that there may be vascular compression of the thoracic outlet.

If symptoms appear within 15-30 seconds while still lying on the table, this may indicate vertebral artery dissection (VAD). This is a potential emergency, and must be screened and/or treated as soon as possible at a hospital. Read more about VAD HERE.

Fig. 24

The approach of corrections remain the same, however. We want a posture that remains the head, cervical spine and clavicle in optimal position. Breathing habits will need to be worked on, especially with regards to thoracic vertical expansion during inhalation. Additionally, the scalenes and sternocleidomastoid will need strengthening, along with any relevant compression you may find in the extremities.

Here’s a patient with ipsilateral migraine and facial numbness. In normal position, there is nice normal flow within the vertebral artery, with a strong signal. Because ultrasound is not quantitative, meaning that it can not reliably quantify blood volume, it is generally used for qualitative assessments, meaning that evaluation of flow speeds and waveforms are used to estimate whether or not the flow is “normal”. The problem is that the reference ranges for these scans are very wide, and it is very easy to get a false negative. Thus one needs to evaluate changes between the foraminal levels, as well as with rotation in both directions while in cervical extension.

Signal strength indicates the amount of blood that travels at the given speeds, and is thus quantitative. Did I not just say that ultrasound is not quantitative? Here’s the problem. Signal strength is very, very easily altered. Too much or too little gel, poor probe position or insonation angle, changed by gain levels, etc. Thus, one needs to keep the same insonation angle, depth, as well as gel amount, and MOST IMPORTANTLY keep the same gain setting when evaluating changes. Usually slight speed changes, but large signal changes are seen in patients with non-acute pathology, such as TOS-related migraines or similar.

Fig. x

Turned head to the right, i.e. in a position similar to that of DeKleyn’s (VAD) test shows significant loss of flow volume, indicated by obliteration of signal.

Fig. x

TOS may also lead to migraines in the absence of vertebral artery compression. Tightness (due to weakness) of the scalenus muscles will compress the subclavian artery, especially during ipsilateral rotation and extension of the neck. As the subclavian artery compresses, the blood that is supposed to enter the arm is forced to redirect into the head. In result, intermittent or sometimes even chronic hyperperfusion of the carotid and vertebral arteries may occur. Carotid hyperperfusion syndrome is a phenomenon usually associated with carotid stent placement, i.e. in relation to surgical intervention of atherosclerosis. However, musculoskeletally induced hyperperfusion may also occur, as stated, if the inlet to the arm is obstructed.

Similar to that of hypopefusion (flow deficit), hyperperfusion is also associated with migraines, headaches, dizziness, transient bell’s palsy, nausea, hemiplegia palsy and more (Adhiyaman 2007, Tehindrazanarivelo 1992, Coutts 2003, Sundt 1981).

TOS and autonomic dysfunction

Another very interesting aspect of thoracic outlet syndrome, though somewhat more rare, is its potential for autonomic nervous system irritation. The somatic nervous system and autonomic nervous system is interconnected through something called gray rami communicans. When the somatic nerves such as the brachial plexus are entrapped, the dysfunction may “bleed over” into the autonomic chains, just as a lumbar disc herniation may cause pain from the back down into the foot.

A relatively common symptom is chronic cough, but I’ve also seen chronic hiccups, increased heart rate upon cervical rotation, dry throat syndrome, clogged ears, tinnitus, burning tongue and even pseudoangina symptoms occur in some of these patients. I have also seen associations between autonomic irritation and atrial fibrillation. Because these nerves innervate virtually all organs in the body, it is difficult to list all the possible symptoms that could occur when they are irritated. See some interesting evidence below.

Fig. x

In neurogenic TOS, neurogenic symptoms occur in the upper extremity and may radiate to the shoulder, neck, and occipital regions if the upper trunk is involved; Raynaud phenomenon is frequently seen due to an overactive sympathetic nervous system, whose fibers run along the C8 and T1 nerves. – Aralasmak et al., 2010

Chest pain or pseudoangina can be caused by TOS. Dorsal sympathectomy is helpful for patients with sympathetic maintained pain syndrome or causalgia and patients with recurrent TOS symptoms who need a second procedure.

Compression of the sympathetic nerves in the thoracic outlet may occur alone or in combination with peripheral nerve and blood vessels. The sympathetics are intimately attached to the artery as well as adjacent to the bone. They may be compressed or irritated in primary or recurrent TOS. Atypical chest pain (pseudoangina) simulates cardiac pain (48). Major indications for dorsal sympathectomy include hyperhidrosis, Raynaud’s phenomenon or disease, causalgia, SMPS, reflex sympathetic dystrophy, and vascular insufficiency of the upper extremity.

To explain chest pain from TOS compression, it is important to remember there are at least two types of pain pathways in the arm: the commonly acknowledged (C5 to T1) somatic fibers, which transmit more superficial pain, and the afferent sympathetic nerve fibers, which transmit deeper painful stimuli. The cell bodies of the two types of neurons are situated in the dorsal root ganglia of the corresponding spinal segments. They synapse in the dorsal gray matter of the spinal cord, and the axons of the second-order neurons ascend in the spinal cord up to the brain. Compression of the superficial C8 to T1 cutaneous afferent fibers elicits stimuli that are transmitted to the brain and are recognized as integumentary pain or paresthesias in the ulnar nerve distribution. In contrast, compression of the predominantly deeper sensory fibers elicits impulses that are appreciated by the brain as deep pain originating in the arm or the chest wall, even if the source of the impulses is cardiac (referred pain). – Urschel & Kourlis, 2007

Cough attacks elicited by movement of the neck and right arm are reported in a patient who had sustained several shoulder injuries and who had an anterior scalenectomy. The coughing was accompanied by weakness in the right upper limb. At exploration, the phrenic nerve was found adhered to the brachial plexus. The cough attacks disappeared, and the weakness of the right upper limb improved somewhat after lysis of the adhesions between the phrenic nerve and the plexus and after external neurolysis of the upper, middle, and lower trunks. Postoperatively, the patient could elevate his right arm without coughing. – Yamagami et al., 1994

In this case report, we rendered a 22 year old woman with the diagnosis of neurogenic thoracic outlet syndrome. We have evaluated her symptoms of palpitation with Holter monitorization during Roos test before and after surgery where transaxillary first rib resection and scalenectomy were performed. Postoperatively she improved and the tachycardia resolved. We propose that stellate ganglion or postganglionic efferent sympathetic fibers forming the cardiac plexus are exposed to compression while Roos test is being performed. Due to this irritation, there can be an increase in the cardiac sympathetic activity. – Kaymak et al. 2004

Four patients with elevated creatine phosphokinase (CPK) values and recurrent chest pain were found to have thoracic outlet syndrome. This association of abnormal CPK levels and chest pain due to thoracic outlet syndrome has not been previously reported. Symptoms and CPK values improved with anti-inflammatory medications and/or proper posture instruction. It is proposed that CPK values become elevated by ischemic or neurologic compromise of muscles supplied by the subclavian artery or brachial plexus respectively. Accordingly, chest pain in the same dermatomal distribution as that of angina pectoris may be simulated by ischemic skeletal muscle. – Godfrey et al., 1983

Forty-four patients presenting with chest pain suggesting coronary artery disease had normal exercise stress tests and selective coronary angiography and subsequently were found to have an unsuspected thoracic outlet syndrome. Pathways of pain in angina pectoris and afferent stimuli originating from brachial plexus compression at the thoracic outlet stimulate the same autonomic and somatic spinal centers that induce referred pain to the chest wall and arm. – Urschel et al., 2010

There has been increasing evidence that dysfunction of the autonomic nervous system that encompasses the sympathetic, parasympathetic and intrinsic neural network is involved in the pathogenesis of AF (atrial fibrillation).

Sympathetic system may promote arrhythmia by increasing Ca2+ transient. Activated β-adrenergic signal pathways increase Ca2+entry and the spontaneous release of Ca2+ from sarcoplasmic reticulum (36). However, vagal stimulation or perfusion of ACh in experiments contributes to development of AF by heterogeneous shortening of action potential duration and refractory period. With vagal hyperactivity, the atrial repolarization is abbreviated by ACh-activated potassium current (IKACh) (37), and/or non-cholinergic and non-adrenergic neurotransmitters, such vasoactive intestinal polypeptide VIP (38). Furthermore, studies have demonstrated that the interaction between sympathetic and parasympathetic nervous systems in developing AF by recording nerve activities directly from stellate ganglia, and vagal nerve (39).

Parasympathetic stimulation has long been associated with increased propensity to AF (40,41). The onset of paroxysmal AF often may be preceded by evidence of increased vagal tone, especially in patients with lone AF who otherwise have structurally normal heart (29). – Xi & Cheng, 2015

Symathetically mediated atrial fibrillation is observed in the presence of any heart disease, the first effect of which is to provoke a vagal withdrawal. The role of the autonomic influences should be taken into consideration every time conventional antiarrhythmic treatment is insufficient. – Coumel, 1994

Pathways of pain in angina pectoris and afferent stimuli originating from brachial plexus compression at the thoracic outlet stimulate the same autonomic and somatic spinal centers that induce referred pain to the chest wall and arm. – Urschel et al., 2010

A 60-year-old man experienced arrhythmia when he turned his head to the left and had these symptoms for 7 years. The patient attributed his symptoms to TOS. The arrhythmia was triggered while performing an Adson test during the clinical evaluation. … The cardiac plexus receives parasympathetic fibers from the superior and inferior cardiac branches and the recurrent laryngeal nerves that are branches of the vagus nerve. Occasionally, the postganglionic sympathetic fibers may pierce the anterior scalene muscle. Therefore, the authors believe that abnormalities in this muscle may cause sympathetic cardiac hyperactivity. Increased cardiac sympathetic activity appears to be linked with arrhythmias. – Shreeve & La Rose, 2011

Confusion regarding the differentiation between arterial and neurogenic TOS is common because many patients with neurogenic TOS have symptoms of coldness and color changes in their hands along with their other symptoms. These symptoms do not establish a diagnosis of arterial or vascular TOS. They are the result arteriolar vasoconstriction brought on by sympathetic nerve stimulation from compression of the sympathetic nerve fibers that accompany the C7 and C8 nerve roots [2]. – Sanders, 2007

It is clear that the irritation of the cervical sympathetic plexus comes from entrapment of the thoracic outlet. However, the vagus and phrenic nerves have a different course than the above-mentioned, yet are also related to the scalenes. More specifically, the anterior scalene and the clavicular portion of the sternocleidomastoid muscle.

When treating patients with stiff necks, I noticed how some of these had an aggressive cough mechanism occur every time the patient’s head was rotated maximally to one side, usually the side of more significant TOS-related symptoms. It took me a while, but in turn I realized that the vagus nerve as well as the phrenic nerves may get caught between the SCM and anterior scalene, especially when extending or rotating the head. This is often occurring if the patient has a prominent external jugular vein when lying supine, which is indicative of dysfunction.

The treatments are of course the same; the scalenes and SCM requires significant strengthening over a period of time.

Fig. x – 1: m. SCM, 2: m. scalenus anterior, 9: n. vagus, 10: n. phrenicus

Use great caution when treating TOS

Although I am more than confident that my protocol that’s written in this article works, it is important to emphasize that treating TOS is not simple, nor easy. There is a great level of detail that goes into the exercises, as the patient’s body will have learned many compensatory strategies, often for years on end, in order to cope with daily life. The body has especially learned to NOT use the scalenes, as it knows that will lead to a bad time.

Often times the patient will have a difficult time performing the exercises properly. I have spent up to 10 sessions with certain clients until they’ve got it right. That’s not because they are not intelligent, but perhaps had a slight lack of attention to detail, and of course because the body was working against them rather than with them. We have to force the body to re-engage those scalenes.

More so, once the patient does engage the scalenes properly during their homework, their symptoms will exacerbate. This period of exacerbation of symptoms can last all from 2 weeks to 6 months depending on the severity of the situation, and presuming everything is performed correctly (exercises, posture, breathing, etc), and this may of course become a difficult period for the client. It is, however, better than having no treatment at all. And once this period is finished, the muscles can be strengthened without symptoms, and the symptoms themselves will also be gone.

Do not go too hard, too fast

A middle aged woman, dentist and tennis player, came to see me for many issues. Pretty much wide spread pain, much of which was nerve pain stemming from the thoracic outlet. She was having difficulty breathing, clogged ears, neck and shoulder pain, and dizziness. She was also very tired. All symptoms of significant TOS. She was also very, very stressed, worked 10 hour days (with a horrible posture as a dentist), almost without breaks, for 30 years. Of course, time was starting to take its toll.

When she laid supine on the bench, I could see the external jugular vein greatly distending. And even though I hadn’t touched her yet, I knew based on this and the history that this was TOS. I squeezed into the interscalene triangle (into the plexus brachialis) and it caused great pain even with moderate pushing.

I knew that starting to strengthen those scalenes was going to be really rough for her, but because there was so many things going on, we just had to get started. I told her very clearly that her symptoms will surely exacerbate as we start training these muscles; she concurred. We did 5 repetitions the first day, and I texted her the day after and asked how bad her symptoms were. She said that she was fine, and as you know, this implies going a little harder. The next day she did 7 reps, still no symptoms. The day after, she did 10 reps. The next morning, 8 am she calls me; extreme dizziness, can barely stand, a throat so dry that not even water could moist it, difficulty breathing and almost fainting. And of course, big time neck pain. She was stressed out of her mind because patients were waiting for her.

Well, there wasn’t much I could do, as the damage was already done. I told her to take some NSAIDS, which helped some. This is a very unique case and I’ve never experienced something so dramatic before, and I’ve treated many severe TOS sufferers, but that’s also why I bring it up so that you’re aware that this may occur. I don’t know if she trained them (the scalenes) more properly the last day, or if it was the cumulative loading that made the muscles inflammate, but these symptoms are of course vagus nerve irritation as well as vertebrobasilar insufficiency.

She was fine a few days after, but was of course mortified of starting those exercises again. The exercises really aren’t dangerous or scary if adequate intensity is used, but it may take some trial and error to find that “adeuqate intensity”. I’ve already done the trial and error, though, so that you don’t have to. The moral of the story is that if it looks really bad, it probably is, and it may be well worth going easy the first weeks. For this patient 2-3 repetitions PER DAY would be sufficient the first 2 weeks.


Thoracic outlet syndrome is caused by continuous compression of the nerves and vascular structures. The underlying reasons are often postural and breathing abnormalities that need to be corrected. Secondary to the postural and breathing correctives, it will be important to address all the symptoms; the muscle inhibition. It’s important to work on both the cause and the symptoms in order to resolve thoracic outlet syndrome as swiftly as possible.

Strengthening the muscles that surround the irritated nervous fibers will trigger and worsen the symptoms. This may seem contra intuitive, which is probably why so few are able to manage these types of issues in the first place. However it’s necessary the increase the work capacity of the given muscles to such extent that they no longer irritate the nervous structures that either pass through, or next to them. Some pain in the process is inevitable, so don’t let it scare you. Regulate exercise volume and intensity based on how much it hurts (it should just hurt a little), and start very easy.

Provocative pressure testing is a very reliable way of diagnosing  thoracic outlet syndrome, because it shows the therapist exactly where the nerves are irritated. This takes the guess-work away, and the therapist will know where the further assessment and correctives should be initiated in order to resolve the issue.Manual muscle testing of muscles that are responsible for nervous compression, will often reveal a false negative (appear strong) at first. Don’t trust this, as it’s just the body’s protective response. The body knows that firing off that muscle will cause pain and irritation, and often does everything it can to avoid using it.

A quick summary, by points:

  • Thoracic outlet syndrome is usually caused by extremely weak scalenes and posturally depressed clavicle
  • Underlying causes for the above are often swayback posture, belly-breathing, poor scapular control
  • Pressure tests can be performed to identify the exact areas of compression
  • The muscles that surround the irritated nerves are almost always weak, and need strengthening
  • Strengthening the muscles will make the symptoms worse before it gets better – this is normal and necessary, but shouldn’t be done so intensely that the symptoms become excruciating

32 thoughts on “How to truly identify and treat thoracic outlet syndrome (TOS)

  1. Mona Vandescheur says:

    Hi , we spoke about a month ago on my TOS from Canada’s . Just wondering what are you studying on TOS ? Have you heard of this TOSMRI? I did give Dr. Werden your FB link and told him you have amazing case studies. He was intrieged! Keep up the good work . PS I never did get your physio links…….Mona

    • Kjetil Larsen says:

      Dear Mona

      I sent you everything on Skype, it is still there in the chatbox.
      We are currently studying TOS and its mechanism of cerebrological comorbidities.


  2. Mona Vandescheur says:

    Hi , we spoke about a month ago on my TOS from Canada’s . Just wondering what are you studying on TOS ? Have you heard of this TOSMRI? I did give Dr. Werden your FB link and told him you have amazing case studies. He was intrieged! Keep up the good work . PS I never did get your physio links…….Mona

  3. Tom says:

    Dear Kjetil,

    I just want to know what are your thoughts about trigger points deep massages in case of TOS ?

    In other words, besides all your recommendations, could trigger points massaging bring something positive to TOS recovery ?

    I thank you in advance for your thoughts

    Best regards


  4. Tom says:

    Dear Kjetil,

    I just want to know what are your thoughts about trigger points deep massages in case of TOS ?

    In other words, besides all your recommendations, could trigger points massaging bring something positive to TOS recovery ?

    I thank you in advance for your thoughts

    Best regards


  5. Ehsan says:

    Dear Kjetil
    Thanks for your helpful artikle about TOS. I want to know more about exercises for strengthening Scalen and SCM muscles. Can you help me? Thank you very much.

  6. Ehsan says:

    Dear Kjetil
    Thanks for your helpful artikle about TOS. I want to know more about exercises for strengthening Scalen and SCM muscles. Can you help me? Thank you very much.

  7. Rachel says:

    I Have a 10 year old with EDS, POTS and more. there is a difference of opinion if it’s VTOS or NTOS. Would you be able to give me an opinion based on her ultrasound resukts?

    Much Gratitude

  8. Rachel says:

    I Have a 10 year old with EDS, POTS and more. there is a difference of opinion if it’s VTOS or NTOS. Would you be able to give me an opinion based on her ultrasound resukts?

    Much Gratitude

  9. Mario says:

    Thank you for the helpful information! I know you mention that when you start strengthening the scalenes and other supporting muscles, symptoms could get worse at first. About how long does that worsening last and at what point do you decide that the worsening symptoms indicate that the TOS is getting worse, not better? Thanks!

  10. Mario says:

    Thank you for the helpful information! I know you mention that when you start strengthening the scalenes and other supporting muscles, symptoms could get worse at first. About how long does that worsening last and at what point do you decide that the worsening symptoms indicate that the TOS is getting worse, not better? Thanks!

  11. John says:

    This article connected a lot of dots for me and I really appreciate the information. I have TOS and in therapy we have found that my arm becomes very full, fatigued and discolored when I do external rotation. Nothing else really makes it do this. Any thoughts on what may be being compressed here? Is it possible that the external rotators are pressing on a vein or artery? Thanks

  12. John says:

    This article connected a lot of dots for me and I really appreciate the information. I have TOS and in therapy we have found that my arm becomes very full, fatigued and discolored when I do external rotation. Nothing else really makes it do this. Any thoughts on what may be being compressed here? Is it possible that the external rotators are pressing on a vein or artery? Thanks

  13. Paul Fontana says:

    I wish you were a doctor around here. For me its neck, shoulders, upper arm and fingers mainly index and thumb. I haven’t noticed any bulging vessels or swollen anything but i do have tinnitus and stuffed ears. My doctor has me doing standard PT and it has relived the pain somewhat. but after reading this I’m not sure if its the right thing.

  14. Renee Johnson says:

    Would a knotted muscle in the neck or suprascrapular area cause symptoms similar to TOS? I went to therapy for TOS, but didn’t seem to help – but worsen my neck it seemed. I found your site and did the head exercise, not letting it reach the floor – seemed to have helped a lot. But problem hasn’t gone away. I was told it may be a knotted muscle in neck, so I am wondering if this could be just a knotted muscle in shoulder neck area.

    • Kjetil Larsen says:

      The scalenus muscle is in the neck. Please read the article before asking questions. “My problem hasn’t gone away”, well, you don’t know what you’re suffering from nor what muscle to treat.

  15. Rajiv says:

    Kjetil, thank you very much for the detailed article. I’ve been working on the scalene exercises with a fairly low number of reps (5) and I’ve been noticing some numbness/tingling on my face (near the chin and side of my cheek), even when resting for three days between sets. Is this something I should be concerned about, or have you seen this before? Should I reduce the exercise intensity?

    • Kjetil Larsen says:

      That’s fine, you’re just doing too many reps or the frequency is too high. Find a rep range / frequency ratio where you get worse only 1 day after training. I usually have my patient train twice per week. See my “reps and sets” video on youtube.

Leave a Reply

Your email address will not be published. Required fields are marked *