Acute vascular insufficiency in the upper limb suggests the possibility of
an embolic
phenomenon1-3.
The most common source of upper-extremity emboli is from the subclavian
artery, with less likely potential sources being the superficial palmar arch
and cardiac
vessels1. Thoracic
outlet compression represents the most frequently reported cause of subclavian
arterial
disease1,2,4,
and arterial-related problems associated with thoracic outlet compression,
although extremely uncommon, have also been
reported5. In
addition to the detection of historical features and the performance of
provocative maneuvers during the clinical examination, the use of noninvasive
magnetic resonance angiography has been reported to be particularly helpful in
the diagnosis of thoracic outlet
compression6,7.
We report an unusual clinical case in which a peripheral embolic disorder
was identified in a young adult with an associated history of recurrent
shoulder instability as well as a deformity of the first rib. Although
magnetic resonance angiography failed to identify the pathology in the
subclavian artery, it was subsequently confirmed by arteriography.
The patient was informed that data concerning this case would be submitted
for publication.
Ahealthy nineteen-year-old college freshman had pain in the right wrist
after being struck by a football during a friendly competition. The symptoms
persisted for several weeks. When the patient returned home from college, he
was evaluated by a hand specialist who suspected the diagnosis of extensor
carpi ulnaris tendinitis and treated the condition with a corticosteroid
injection. The wrist pain continued and became associated with diffuse
numbness, alteration in temperature, and pain extending up the forearm.
Of note is the fact that, following a traumatic ipsilateral shoulder
dislocation at twelve years of age, the patient had experienced three to four
episodes of shoulder subluxation during strenuous activities, although the
last episode occurred two years prior to the current problems and never
required surgical intervention.
Because of the change in the symptom complex, the patient underwent an
evaluation by a vascular surgeon. On physical examination, the results of the
Allen test were notable for delayed filling in both the ulnar and radial
arteries of the involved limb compared with the filling time in the uninvolved
limb, and the distal radial pulse was absent during an audible Doppler
examination. No sensory or motor defects were noted.
Because there was a concern for the presence of vascular compromise with
possible distal embolization, an evaluation was performed with use of magnetic
resonance angiography with gadolinium enhancement and three-dimensional
reformations along with magnetic resonance imaging of the thoracic outlet. The
subclavian and axillary arteries and the proximal portion of the brachial
artery appeared to be normal (Fig.
1). An evaluation of the distal portion of the arm and the forearm
with use of magnetic resonance angiography revealed a patent brachial artery
and ulnar artery with complete occlusion of the radial and interosseous
arteries.
Because of concern for the presence of a distal arterial occlusion, the
patient was referred to our care and was seen approximately one month
following the magnetic resonance angiography study. Our examination revealed
an absence of palpable and audible pulses in both the radial and the ulnar
artery with the arm by the side as well as pallor of the hand with overhead
positioning of the arm. An audible Doppler signal was found only in the
mid-forearm.
As a result of the previous findings and because the findings from our
examination were suggestive of progressive vascular changes consistent with
the possibility of embolization, a formal transfemoral arteriogram was
performed. It revealed a small amount of thrombus in the right axillary artery
at the level of the first rib, with a mild stenosis in the subclavian artery
with minimal aneurysmal dilation (Fig.
2), as well as embolic occlusion of the distal brachial artery
with complete occlusion of the radial and interosseous arteries. During the
study, overhead positioning of the arm clearly revealed the stenotic region of
the subclavian artery to be at the junction of the clavicle and first rib
(Fig. 3) with an approximate 1
× 2-cm area of cortical prominence noted in the anterior third of the
first rib (Fig. 4).
Shortly after the arteriogram and approximately two months after noting the
absence of a radial pulse, a vascular surgeon performed embolectomy of the
brachial, radial, and interosseous arteries and intraoperative administration
of tissue plasminogen activator. At the time of exploration, thrombus that
appeared to be chronic and embolic in nature was retrieved from all three
vessels. An intraoperative arteriogram, with the arm abducted at 90°, was
then performed, revealing no appreciable aneurysmal dilation, residual
thrombus, or hemodynamically important stenosis in the axillary artery.
Intravascular ultrasonography showed no evidence of [appreciable] stenosis,
aneurysmal dilation, residual thrombus, or compression of the subclavian
artery at the level of the first rib with the arm in the neutral position.
However, a calcified plaque was noted in the posterior wall of the axillary
artery at the level of the first rib, consistent with some type of repetitive
arterial trauma to the area. The radial artery was noted to appear chronically
occluded, with a large amount of thrombus or embolus in the lumen.
Because the arteriogram demonstrated thoracic outlet compression when the
shoulder was positioned in abduction and because embolic sequelae were
confirmed during the embolectomy, a transaxillary resection of the right first
rib was performed, but at a later date in order to avoid the risk of increased
bleeding that accompanies the administration of intravenous heparin during
embolectomy. There was a moderate amount of inflammation around the first rib
at the time of the surgery, and the pleural surface was stuck to the
undersurface, suggesting some type of previous inflammatory process. The rib
otherwise did not appear remarkable and, since it was removed in pieces along
with the periosteum, it was not sent for pathologic examination. Since the
artery appeared normal at the time of surgery and had no appreciable dilation
or thrombus formation, resection of the artery was not performed. Following
the procedure, the patient was placed on Coumadin (warfarin) for six months,
during which time he remained asymptomatic. Follow-up noninvasive arterial
studies revealed normal blood flow in the right axillary artery with the arm
at the side, with the arm abducted at 90°, and with the arm abducted and
externally rotated. However, noninvasive arterial studies suggested that there
was appreciable compression of the left arterial system with the arm at the
side as well as with the arm at 90° of abduction, with worsening of blood
flow with external rotation of the shoulder. After a discussion with the
patient and his family, the patient chose to undergo elective left
transaxillary resection of the first rib to minimize the chance of
experiencing a similar problem on the left side. Postoperative noninvasive
arterial duplex studies, performed after the transaxillary resection of the
left first rib, revealed normal arterial flow in both axillary arteries with
the arms abducted and externally rotated.
Arterial thoracic outlet compression, although rare, may have severe
complications. The fundamental arterial lesion involves repetitive compression
of the second portion of the subclavian artery with intimal injury or stenosis
and resultant poststenotic dilation and associated turbulence with mural
thrombus formation and
degeneration5. With
a mural thrombus, distal embolization is always a threat; thus, a timely
diagnosis along with initiation of prophylactic or therapeutic intervention is
necessary.
The presentation may range from acute ischemia to insidious unilateral
upper-extremity pain, claudication, and coolness of the digits. Careful
attention must be paid to the arterial pulses with the shoulder hyperabducted
and externally rotated. In addition, delayed filling on the Allen test, skin
coolness, fingertip ulcerations, and subungual hemorrhages are important
indicators of vascular
compromise5.
Imaging modalities may play a key role in diagnosis and management. Plain
radiographs of the shoulder may demonstrate osseous abnormalities leading to
thoracic outlet compression. The first rib creates the posteromedial border of
the costoclavicular space and, with deformity, may directly compress the
neurovascular structures. In our patient, the deformity of the first rib was
consistent with an exostosis, either posttraumatic or congenital in origin. It
is plausible that the same initial contact injury that caused the shoulder
dislocation could also have caused a fracture of the first rib. There have
been reports of acute fractures of the first rib, congenital anomalies of the
first rib, clavicular malunions, and retrosternal clavicular dislocation
causing thoracic outlet arterial
compression3,8,9.
A review of the literature revealed no reports of malunion of the first rib as
a cause of this problem. Unfortunately, with regard to the case of our
patient, there was no comment in the operative report regarding the appearance
of the first rib during excision.
The previous traumatic shoulder dislocation and recurrent shoulder
subluxation, although an unlikely explanation due to the eventual bilateral
involvement, may have contributed to the thoracic outlet compression in our
patient. The bilateral nature of the problem, as revealed on the arterial
studies, suggested that a congenital problem with the first rib and scalene
musculature may also have contributed to this process; however, the symptoms
occurred only in the subluxating shoulder and not in the contralateral
shoulder. A study performed by Leffert and Gumley showed a correlation between
anterior subluxation of the shoulder and neurogenic thoracic outlet syndrome
in eight of twenty-seven patients (30%) in a consecutive series of Bankart
operations for treatment of
subluxation10. The
patients were diagnosed with thoracic outlet compression clinically by a
history of numbness or paresthesias in the hand and by physical examination
with reproduction of symptoms and weakening of the radial pulse when the
shoulder was abducted and the humerus externally rotated (described as a
positive Wright
maneuver10). No
arteriograms were performed. The proposed mechanism was a disturbance in the
kinesiology of the shoulder-joint complex that altered the position of the
scapula relative to the rib cage and the neurovascular supply to the upper
limb. There are few reported cases of axillary or subclavian artery aneurysm
secondary to blunt shoulder trauma or repetitive throwing motions in
pitchers11,12.
Magnetic resonance angiography has been shown to be a reliable noninvasive
alternative to conventional arteriography for evaluation of upper-extremity
vasculature13. In
the case of our patient, magnetic resonance angiography was used as an initial
imaging step to evaluate progressive peripheral vascular changes that were
consistent with the possibility of embolization. However, magnetic resonance
angiography proved to be inadequate in detecting the subclavian aneurysm
containing thrombus, which was later noted on the arteriogram. The magnetic
resonance angiogram did reveal occlusions in the distal vasculature, which
were confirmed on the arteriogram. In their prospective study, Kransdorf et
al. showed that magnetic resonance angiography is a reliable noninvasive
method to evaluate patients with severe upper-extremity ischemic disease,
accurately depicting the peripheral vessels through the level of the palmar
arch13. In another
prospective study, Cosottini et al. reported a sensitivity and specificity of
90% and 95%, respectively, in detecting stenoocclusive disease of the
subclavian artery in fifty
patients14. Other
reports have shown that magnetic resonance imaging along with magnetic
resonance angiography does an excellent job of demonstrating the anatomy of
the brachial plexus as well as any vascular compression or occlusion in the
abducted upper
extremity6,7.
On the contrary, Estilaei and Byl, in their recent evidence-based review of
seven studies, revealed that the current evidence in support of magnetic
resonance angiography as a valid test for diagnosing arterial thoracic outlet
syndrome is weak and not based on strong
design15. Among the
studies reviewed was the paper by Cosottini et al., which was the only
prospective study and had the best level of evidence in the
review14,15.
In the case of our patient, the arm was in the neutral position during the
magnetic resonance angiography. Compression might better have been
demonstrated with the arm placed in an abducted position, as was done for the
arteriogram16.
Since the patient was not being evaluated specifically for thoracic outlet
syndrome, an abduction protocol was not ordered. Two separate boluses of
contrast are necessary to perform magnetic resonance angiography with the
shoulder in the abducted and neutral position due to the clearing of contrast
during patient repositioning. When two consecutive acquisitions are obtained
with the arm in two different positions, contrast from the first injection
increases the background signal for the second acquisition and the image
quality of the second acquisition may be
impaired16.
Therefore, if a vascular thoracic outlet syndrome is suspected, the first
imaging sequence should occur with the shoulder abducted to avoid degradation
of this most important portion of the study. Interpretive difficulties may
also be seen with magnetic resonance angiography secondary to patient movement
and overlapping
structures17. Ways
of minimizing interpretive difficulties may depend on the specific magnetic
resonance angiography sequence used. An evaluation of the magnetic resonance
angiography technique (three-dimensional contrast-enhanced imaging compared
with two-dimensional imaging) was reported by Charon et
al.18. They found
that three-dimensional contrast-enhanced magnetic resonance angiography was
less prone to artifact and frequently demonstrated the underlying cause of
thoracic outlet syndrome when
reformatted18.
Conventional angiography has been the mainstay in the diagnosis of vascular
thoracic outlet
syndrome16,18,
but it is invasive and requires iodinated contrast and ionizing
radiation19. In
comparison with iodinated contrast, magnetic resonance paramagnetic contrast
agents rarely provoke anaphylaxis and are not
nephrotoxic20. In
our patient, an arterial duplex examination was also helpful in documenting
arterial compression in the axillary artery with abduction and external
rotation of the upper extremity. In particular, improvement in the arterial
flow and relief of the extrinsic compression were able to be identified after
resection of the first rib on both sides.
In conclusion, while magnetic resonance angiography continues to be used as
a noninvasive imaging technique for the upper-extremity vasculature,
arteriography is still viewed as the gold standard for delineating proximal
artery abnormalities when evaluating the thoracic outlet syndrome. Even if a
patient has negative results on magnetic resonance angiography, an arteriogram
should be made if there is a high clinical suspicion for arterial thoracic
outlet compression. Further prospective studies examining the reliability of
magnetic resonance angiography for detecting proximal upper-extremity
pathology are necessary. Orthopaedic surgeons should be aware of this new
technology in case they wish to use a less-invasive imaging modality, although
they should be wary of the possible shortcomings and knowledgeable of the ways
in which such shortcomings can be minimized. ?