SURGERY OF THE THORACIC OUTLET

For reasons that will become much clearer later in this
chapter, I believe that the diagnosis of TOS is primarily clinical.
Although some laboratory examinations are helpful, diagnosis is usually
made on the basis of a careful history and physical examination.

The symptoms of TOS are quite variable, depending on
which structures are being compressed. The majority of patients
complain of pain and paresthesias that radiate from the neck, upper
chest, or shoulder region, down the medial aspect of the arm and into
the little and ring fingers, although some experience numbness of the
entire upper limb. Much less commonly, it is the radial aspect of the
arm and hand that is numb. Symptoms are often nocturnal; elevation of
the arm in sleep or during activities aggravates them. Complaints such
as the inability to hold a hair dryer or to work with the arms above
the head are common. Some patients have numbness when driving or when
carrying heavy objects. These symptoms are caused by compression of the
lower trunk of the brachial plexus.

The signs are usually subtle and may be confined to
weakness of the ulnar-innervated intrinsic muscles unless compression
is severe. There may be weakness of all the intrinsic musculature and
loss of power of the long flexors of the little and ring fingers. When
the upper trunk of the brachial plexus is involved, differential
diagnosis becomes more difficult, and there may be confusion with
cervical radiculopathy or carpal tunnel syndrome.

Signs and symptoms of TOS may include acute arterial
insufficiency or gangrene in the hand, but such severe symptoms are
rarely seen in an orthopaedic practice. Such patients usually have
acute occlusion of the subclavian or axillary arteries and often have
aneurysms caused by compression by significant cervical ribs (13,27). This complication is more often found in older patients with long histories of often undiagnosed symptoms.

Venous compression can cause swelling of the hand and
upper limb, which, if intermittent, is difficult to document until its
severity motivates the patient to have it examined. Intermittent venous
compression in the absence of thrombosis produces cyanosis as well as
swelling in the limb.

Occasionally, a patient exhibits acute venous thrombosis, the so-called Paget-Schroetter syndrome (11,13,34).
This “effort thrombosis” of the subclavian and axillary veins results
from underlying subclinical thoracic outlet compression and repetitive
or strenuous use of the arms, particularly in the overhead position.
The symptoms are acute pain and swelling of the limb and ipsilateral
chest wall. Prompt recognition of the condition is imperative—delay may
result in considerable disability due to chronic pain and swelling.
Acute therapy consists of administration of intravenous thrombolytics
by catheter, with mechanical dilatation, if needed. Maintain the
patient on anticoagulants for 3 or 4 months, and follow up with
elective first rib resection to diminish the likelihood of recurrent
thrombosis (35).

Unilateral Raynaud’s phenomenon has been occasionally
diagnosed as being consistent with TOS. In this case, it is attributed
to selective compression of the sympathetic innervation of the limb
within the lower trunk of the brachial plexus. The etiology of
unilateral Raynaud’s phenomenon is not collagen disease. More likely,
this phenomenon represents episodes of distal embolization from an
aneurysm located within the subclavian artery (13).

Patients with TOS may have significant neck or chest
pain. When TOS is in the left hemithorax, such pain may mimic angina or
myocardial infarction and bring the patient repeatedly to the emergency
room or cardiac care unit (42).

The physical examination is most important, not only to
achieve a positive diagnosis but also to rule out other conditions that
may be confused with TOS or that may coexist with it. Cervical
radiculopathy is unusual in the C8—T1 distribution in the arm and hand (5).
Nevertheless, when the symptoms are difficult to interpret, the
cervical spine must be ruled out as a site of disease. There may be
tenderness over the brachial plexus in the supraclavicular fossa, but
this nonspecific finding may accompany either thoracic outlet
compression or cervical radiculopathy.

Of the various provocative maneuvers used to elicit
thoracic outlet compression, the overhead exercise test and Wright’s
maneuver (49) are the most consistently
positive. The first test is performed by having the patient, in the
“hands-up” position, rapidly flex and extend the fingers. In
susceptible patients, this exercise produces cramping on the affected
side within 30 seconds. In addition, placing the arm in the abducted
and laterally rotated position not only obliterates the pulse at the
wrist but also reproduces the symptoms. Lowering the arm to the side
restores the pulse and alleviates the symptoms.

The mere obliteration of a distal pulse with any of the
provocative positions is not diagnostic of TOS, because in many young
women, some position of the arm can obliterate a pulse. The
reproduction of the symptoms is crucial! Sometimes they can be elicited
by rotating the neck to the opposite side or by having the patient take
a deep breath, particularly while the affected extremity is abducted
and externally rotated at the shoulder.

The costoclavicular maneuver or military brace position
reproduces symptoms in patients whose costoclavicular interval has been
narrowed, particularly by a clavicular fracture. The classic Adson’s
maneuver is performed by having the patient, with the arm dependent,
rotate and hyperextend the neck to the affected side and take a deep
breath. In my experience, this has been the least rewarding of the
provocative tests.

It is important to examine the shoulder girdle carefully
to verify the stability of the glenohumeral joint, because patients
with instability may complain that the arm “feels dead” (33).
In addition, finding an increased venous pattern on the side of the
lesion is presumptive evidence of venous hypertension and possible
compression or thrombosis.

Manual muscle testing must include the intrinsic muscles
of the hand. The only motor deficit may be just-perceptible atrophy of
the hypothenar muscles and weakness of adduction of the little finger
to the ring finger. Rarely, all of the intrinsics are weak or atrophic.
The cutaneous distribution of the first thoracic nerve is the medial
aspect of the forearm, thus it is differentiated from the distribution
of the ulnar nerve.

Adequate radiographs are essential. It may be difficult
to assess the lower cervical spine for the presence of adventitious
ribs or long transverse processes at C7 because the patient’s
identification plate may have been placed there. Hypoplastic, true
first ribs may be present, and sometimes, a cervical rib can be
confused with a hypoplastic rib unless you are sure of the vertebral
level. Count from the atlas caudalward to establish that there is, in
fact, a cervical rib. Look carefully at the disc spaces and the
intervertebral foramina, and note abnormalities. In the lateral view of
patients with droopy shoulders, the physician may be able to clearly
define the second dorsal vertebra. Such patients may resemble those
with TOS but may have no peripheral neural deficits in the limbs (37).

Apical lordotic radiographs of the chest can rule out
lung tumors, which can mimic TOS in their presentation. Adequate
radiographs of the shoulder are also necessary. When there is
significant concern about possible cervical radiculopathy, magnetic
resonance imaging (MRI) or myelography may be indicated. They are not
needed in most cases.

Noninvasive vascular studies have been advocated as a
reliable diagnostic test for TOS. In my experience, however, because
pulses can be positionally obliterated in many normal people, the
incidence of false-positive results is substantial. Interpret such
studies with caution. For patients with intrinsic vascular disease,
noninvasive studies can be extremely useful.

Arteriography and venography have limited application in
the general diagnosis of TOS, and I do not routinely use them. Because
the occlusion of arterial outflow with the position of the arm may be
observed in asymptomatic patients, an arteriogram would provide little
information unless there is serious consideration of the presence of an
aneurysm in the subclavian artery or intrinsic vascular disease.
Patients with complete cervical ribs may have an increased incidence of
such aneurysms (13,27).
Therefore, in such patients, it is prudent to obtain arteriographic
studies if the surgeon intends to explore the thoracic outlet through
the axilla. Venography is useful in demonstrating thrombosis of the
axillary or subclavian vein and in following the process of
recanalization, should it occur. Noninvasive testing may be of value in
this situation as well.

Although some have claimed success using measurement of
the velocity of conduction of the ulnar nerve through the thoracic
outlet to diagnose TOS, the experience of many workers has failed to
substantiate this claim (40,42,44).
Reports from electrodiagnostic laboratories may say, “There is no
evidence of TOS”; interpret such statements based on nerve conduction
velocity (NCV) with caution, however, because the test is not of value
in this situation.

Where NCV is very useful is in differential diagnosis.
Conditions such as ulnar neuropathy at the elbow and carpal tunnel
syndrome are readily identified by means of

NCV
determinations. Because NCV determination is not a reliable means of
assessing the plexus within the thoracic outlet, the use of
somatosensory evoked potentials and F responses has received attention
as a means of obtaining further objective evidence of neural
dysfunction to diagnose TOS (10,23).
Some workers have found the measurement of the amplitude of the evoked
response of the medial cutaneous nerve of the forearm and C8 root
stimulation to be of value in making the diagnosis (24).
In my opinion, the diagnostic value of such measurements has not yet
been firmly established, and the diagnosis remains clinical (4,44).

Electromyographic examination of patients with TOS is
usually not markedly abnormal unless there has been sufficient
compression to cause denervation. In such cases, it is often possible
clinically to detect atrophy and weakness of the ulnar-innervated
interosseous and hypothenar muscles. In other patients, however, these
muscles may show fibrillation potentials at rest that appear to be
clinically normal (36).

The question of double-crush syndrome and TOS is controversial. Although Carroll and Hurst (7)
have written that the coexistence of TOS and carpal tunnel syndrome is
rare if it exists at all, patients are seen nonetheless with
well-demonstrated signs of both entities (47). Ulnar neuropathy may coexist with TOS (16).
Patients may have hard-to-define combinations of signs and symptoms
that appear to be the result of more than TOS. I these cases, look for
additional lesions to explain the entire picture; otherwise treatment
will fail.

The general condition of patients with TOS varies
enormously, and in many cases, the symptoms have been present for
months or years. Not uncommonly, such patients have been told that
their symptoms are of emotional origin, or they have been subjected to
unsuccessful conservative or operative treatment. Such patients require
objective, sympathetic evaluation and sometimes psychological
consultation. Some patients with TOS are clinically depressed. The very
deleterious effect depression has on their posture aggravates the
anatomic problem. If such is the case, the surgeon may, according to
training and inclination, treat the patient with antidepressants or
refer him or her for psychotherapy. It is not rare to see patients
whose symptoms of TOS can be lessened by such measures, even though
some may ultimately require surgery.

The general level of fitness of patients with TOS is
often poor. It is worthwhile to inquire about their level of physical
activity; often, it is quite limited. Sometimes, patients are afraid to
exercise for fear of worsening the condition, or they have been so
advised by their physicians. If the patients have no other serious
medical problems, make an effort to mobilize them with aerobic
exercises. Probably the easiest generalized exercise for such patients
is walking, because it does not usually cause much discomfort. Some
patients may experience increased symptoms with walking, and in these
cases, the use of a shoulder support such as the Biomet Hook
Hemi-Harness (Biomet, Warsaw, IN) (Fig. 61.2) or the Roylan sling (Smith & Nephew, Germantown, WI) (Fig. 61.3) may help. Also, such patients may be able to benefit from use of a stationary bicycle.

The problem of obesity can be extremely difficult to
manage; patients do not like to be told that they are overweight. Yet,
excess soft tissue can place additional strain on the shoulders,
particularly in women who have large breasts, which can aggravate TOS
by increasing traction on the structures crossing the first rib. In
cases of gigantomastia, reduction mammoplasty as a first step in the
treatment of very debilitating TOS may be successful. Kay has reported
on neurologic deficits in women with large breasts (15).
The mechanism of thoracic outlet compression makes it likely that this
is the locus of the problem. Even if weight reduction does not produce
the desired alleviation of compression, the patient who is thinner will
be easier and safer to operate on, especially through the axilla.

The most important part of the conservative management
of TOS involves correcting postural abnormalities that can be
identified as contributing to the compression, and exercises to
strengthen the shoulder girdle when it is determined that weakened
muscles are a significant factor (Fig. 61.4).
Unfortunately, when the physician refers patients with TOS to a
physical therapy department, it is not always guaranteed that the
therapist caring for the

patient will understand the genesis of the problem and apply appropriate therapy. Very often, stereotyped routines (25)
such as stretching, soft-tissue massage, “nerve mobilization,” and
cervical traction are used, and these methods will actually worsen the
symptoms in many patients. The TOS patient must have a thorough
analysis by a knowledgeable therapist, and an individual therapy
program should be designed, implemented, and monitored to avoid
provocative maneuvers (25).

Patients with significant shoulder girdle disorders may
not tolerate even the most gentle conservative program. Particularly in
the case of anterior glenohumeral instability, the exercises often
prove provocative because they tend to reproduce the subluxations. Such
patients may be differentiated from those with dead arm syndrome. In
all likelihood, many patients with the diagnosis of dead arm syndrome
owing to glenohumeral instability really have TOS. If the condition
goes unrecognized, the patients will

continue
to be symptomatic even after shoulder repair. Leffert and Gumley
studied an interesting series of patients in whom these two entities
coexisted (17).

If an activity of daily living or employment seems to be
related to symptoms of TOS, modify such activities, if possible.
Overhead activities, the carrying of heavy loads, or the use of
backpacks can be quite provocative to patients with compression within
the thoracic outlet.

Use conservative measures and review progress
periodically so that the condition does not drag on interminably. The
conservative approach with periodic review is particularly important in
patients whose cases are complicated by litigation, insurance, or open
workers’ compensation cases. Caregivers must try to be objective in
interpreting symptoms so that patients are not unfairly deprived of the
benefits of treatment. In addition to periodic reviews, set a time
limit beyond which conservative therapy should be viewed as having
failed. If there is no positive response, the clinician must decide
whether to advise surgery or to accept the status quo with hopes that
the condition may improve with time.

Failure of a carefully supervised program that includes
the previously described treatment measures is the usual indication for
surgery.

Patients with significant neurologic deficits—usually of
the intrinsic muscles of the hand but sometimes in the long flexors as
well—almost never respond positively to conservative therapy. If they
are subjected to surgery, however, they should know preoperatively that
postoperative improvement in the power of the long flexors may actually
increase the muscle imbalance in the fingers and cause clawing or make
it worse. Advise such patients that they may require secondary hand
reconstruction for the muscle imbalance if it occurs. For those who
have significant sensory loss, the outlook is somewhat unpredictable
but generally favorable.

Patients with impending or established gangrene in the
hand are candidates for immediate surgery. Fortunately, they are rare
and usually are not seen primarily by the orthopaedic surgeon but more
likely by vascular surgeons.

The most problematic indication for surgery in the
thoracic outlet is intractable pain. Only the patient can feel and
describe the pain; the surgeon must be able to interpret these reports
appropriately. Constraints of daily routine, vocational or avocational
adjustments, sleep disturbance, and history of analgesic use are all
important avenues of inquiry necessary to formulate a decision for
surgery. A complete and in-depth review of what was done in prior
physical therapy sessions is most important, because the patient may
not have had optimal care, or may actually have gotten worse because of
provocative positioning the exercises require (25).

Patients with TOS often have very complicated medical
records. Review the differential diagnosis very carefully before
proceeding. Discuss with the patient and family the mechanics and
objectives of surgery. Explain all possible complications in detail so
that the patient can give an informed consent. If you believe that a
second opinion would be useful, made certain that the surgeon or
physician rendering this opinion has sufficient experience in the area
to provide a valid opinion. Too often that opinion may be rendered on
the basis of little personal experience or knowledge.

An obese patient who has made little genuine attempt to
lose weight despite specific evaluation for the cause of obesity and
instructions to correct it evinces poor cooperation. I have found it
prudent to inform such patients that their surgery is elective and will
not be done until they demonstrate genuine evidence of being willing or
able to participate in their own rehabilitation.

Because of the possibility of significant intraoperative
hemorrhage, the patient’s blood should be typed and cross-matched
before surgery. Whether to operate on a patient who will not accept
blood transfusion or blood products, should they be needed
intraoperatively, is a difficult and individual decision for which I do
not have a confident answer.

Depending on the surgeon’s training and inclinations and
local custom, as well as the nature of the surgery to be performed,
either a vascular or general thoracic surgeon may be part of the
operating team. Although I have performed the vast majority of my
surgeries without need for such assistance, it is valuable to ensure
that such help is available on an if-needed basis. For patients who
have particular problems such as successive surgeries in which
complications can be anticipated, consultants may be invited to
participate at the beginning of the case. In addition, the operating
team should be experienced and adequate in number. I require three
scrubbed assistants for surgery by the axillary route. Do not use
overhead arm traction with weights, as is used in shoulder arthroscopy;
it carries a risk of inducing an intraoperative traction injury of the
brachial plexus.

Finally, make sure proper instruments are available and
in working order before you make the skin incision. Have vascular and
thoracic surgical instrument packs in the operating room should they be
needed.

General anesthesia for surgery of the thoracic outlet by
the transaxillary route requires that the patient be profoundly
relaxed. The newer, short-acting muscle relaxants are particularly well
suited to this situation. Discuss the time frame for the surgery with
the anesthesiologist before commencing surgery so that muscular
relaxation may be reversed by the time the incision is closed, thus
allowing extubation without delay.

Surgeons with inadequate training or experience should
not undertake thoracic outlet surgery because it has the potential for
very serious complications. However, with proper identification of
patients suffering from significant compression and technically
adequate surgery, the results are very gratifying in most cases.

There is no universal agreement as to which surgical
procedure is best for the treatment of TOS. Scalenotomy was the first
procedure to gain favor. If the pathology resided entirely between the
scalene muscles (1,2 and 3),
as was theorized, release of the anterior scalene should have
permanently cured the condition. Unfortunately, the incidence of
recurrence following scalenotomy was sufficiently high that surgeons
had to develop other procedures.

Many workers have considered the first rib to be the
common denominator of compression and have concentrated their efforts
on eliminating it as well as adventitious ribs or congenital bands that
might be encountered in the course of exploration (9).
Since Roos reported his experience with the transaxillary first rib
resection in 1966, this approach has been the procedure most often
performed, and as the mainstay of the surgical approach to the problem,
it is described in detail (28,29,30,31 and 32). A variety of anatomic approaches, however, may be used for exploration of the thoracic outlet and removal of the first rib.

Significant reduction in the costoclavicular interval
due to fracture angulation or hypertrophic subclavicular callus may
occasionally necessitate claviculectomy. It is worth considerable
effort to retain the clavicle if to do so does not materially increase
the surgical risks (14). Removal of the strut
function of the intact clavicle from a shoulder girdle with poor
muscular support can markedly increase symptoms because of the
superimposed traction effect of the ptotic shoulder. Patients with a
good trapezius may function well following claviculectomy, but they
experience weakness in overhead use of the arm. The enthusiasm for
claviculectomy for uncomplicated thoracic outlet compression has been
very limited. The procedure is not recommended (19).

The use of scalenectomy to treat TOS depends on the
local pathology. Some authors consider it the preferred method of
surgical treatment in certain cases (31).
Although it is desirable to avoid reattachment of the scalenes to the
bed of the first rib after it has been resected, I do not routinely
resect a portion of the anterior and middle scalenes because of the
possibility of injury to branches of the brachial plexus, which may
actually pass through the middle scalene. The phrenic nerve must be out
of harm’s way if you elect to resect part of the anterior scalene.

Scalenectomy may be indicated in patients in whom a
recurrence has been deemed due to scarring within the scalene muscles
themselves. An anterior approach is best. In addition to obvious
branches of the brachial plexus and phrenic nerve, the long thoracic
nerve is particularly vulnerable, because it may not always be located
lateral to the middle scalene. It may pierce the muscle or even present
as two branches, each of which must be gently retracted and preserved.

In the next section, I describe in detail three
operative approaches to the thoracic outlet specifically for resection
of the first rib and then will comment briefly on the others.

In uncomplicated exploration and decompression of the
thoracic outlet by the axillary route, there is minimal blood loss and
physiologic disturbance. Although there may be moderate postoperative
pain, within a day most patients are sufficiently comfortable to move
about without the support of a sling. If a pleural tear has occurred
intraoperatively, then a chest tube is used for 24 hours. Most patients
are discharged from the hospital on the second or third day after
surgery.

Within 48 hours of surgery, the dressing may be reduced
to a single layer of gauze or removed altogether. Patients may shower
as long as the wound is kept dry; underarm deodorants and powders are
prohibited until the sutures are removed. Patients may be ambulatory at
home but are cautioned against heavy lifting or any strenuous

activities
for 4 weeks after surgery. Then they may resume their preoperative
exercise program, if there is no significant postural abnormality or
weakness of the scapular muscles. It is most important that these
exercises do not involve either overhead use of the arms or shoulder
bracing, because both of these types of exercises essentially duplicate
the provocative maneuvers that are used in diagnosis. Unfortunately,
they are still prescribed and used in many physical therapy facilities (25).