THORACIC OUTLET
SYNDROME AS A DISORDER
OF NEUROGENIC INFLAMMATION Wladislaw
Ellis M.D. 510-849-4703 ABSTRACT Neurogenic thoracic
outlet syndrome continues to be controversial largely, because the
pathophysiology remains undefined. The analysis of the progression
of symptoms in 1500 patients coupled with the discovery of innervated
fibrotic tissue adherent to the trunks of the brachial plexus and secreting
proinflammatory peptides, prompted the following hypothesis. Stretch activates
nervi nervorum to discharge inflammatory peptides and to degranulate
resident mast cells producing vasospasm, ischemia, swelling and leukocyte
recruitment with a resultant, localized, neural compartment syndrome.
Repeated injury initiates a vicious cycle of increasing neural sensitization
to neuropeptide expression, symptom intensification and spread, and
evolving innervated fibrosis. This proposed cascade is sufficient to
account for the pleiotropic symptoms and contended findings in this
disorder
KEY WORDS Thoracic outlet
syndrome, nerve entrapments, innervated fibrosis, mast cells, pain.
BACKGROUND Neurogenic thoracic
outlet syndrome (TOS) continues to be controversial. Its variable symptomatology, a paucity of “objective”
clinical findings, together with its varied and often intermittent
course, may be confusing and viewed with skepticism by many examining
physicians [1]. The lack of a clearly defined and effective treatment
strategy is a major factor in the progression of symptoms and ensuing
disability in affected patients.
This may occur
because the pathophysiology of this disorder remains undefined. Symptom
onset commonly follows single force cervical flexion/extension (whiplash)
injury or repetitive trauma associated with poor upper extremity positioning
in the work place. Related predisposing factors in this predominantly
female population in their third and fourth decades include musculoskeletal
abnormalities, fibrous bands and post traumatic mal-positioning of
the plexus roots in the scalene triangle. However, the presence of
these same factors observed in normal individuals during routine autopsy,
and the absence of clinical symptoms in numerous individuals who have
experienced similar traumatic events are not easily reconciled and
underscore their insufficiency as explanations [2]. The accurate diagnosis
and a better understanding of the actual disease mechanisms becomes
increasingly important in the face of a growing
incidence of individuals experiencing hand, shoulder, and arm pain
as well as paraesthesias referable to the brachial plexus.
HYPOTHESIS Based on the
evaluation and treatment of over 1500 patients exhibiting brachial
plexus irritability, the following pathophysiological mechanism is
proposed. The plexus roots, following contusive or repetitive trauma,
develop a neurogenically induced inflammation and
innervated fibrosis which provide a cytokine based mechanism for continued, repetitive
neural inflammation without demyelinization or frank neuronal death.
This would explain the lack of findings on electrodiagnosis or imaging
modalities given the small size of the fibrotic lesions (see below)
[3].
JUSTIFICATION The typical
time course of symptom formation is instructive. Distal paresthesias usually begin to occur
following prolonged repetitive work with the arms outstretched. Paresthesias are paradigmatic for neural
ischemia.
Aching pain
follows with increasing use of the arm(s), or as an initial result
of contusive injury, improving with rest. As
the upper extremities, shoulders, and neck continue being used in ways
that increase traction on the trunks of the brachial plexus, symptoms
begin to increase over time and take longer to improve with rest. At some point, often three to six months after
their initial appearance, symptoms become constant. This course argues for the increasing sensitization
of the nerves involved.
This sensitization
of the nerves is further evidenced by the spread of symptoms. Swelling
in the dorsi of the hands, as well as supraclavicularly becomes a common
occurrence. Pain analogous to that of “sympathetically mediated disorders”,
namely burning, lancing pains with areas of hyper- and hypoesthesia
makes its appearance. Intermittent
motor weakness becomes a frequent accompaniment of these increasing
symptoms, as does symmetrically contralateral involvement.
This symptom
spread is not mononeural, nor restricted to myotomes or dermatomes,
although frequently there is preponderance of ulnar involvement. Within a year or so of symptom onset, the clinical
presentation commonly progresses to involve all three trunks. Double
and triple crush injuries are common. Mirror symptoms follow. This
spread also activates the cervical plexus (paravertebral neck pain),
the greater occipital nerves (radiating headaches), and often vagal
(GI symptoms) and recurrent laryngeal (hoarseness) nerves. Raynaud’s
phenomenon frequently occurs with exposure to temperature extremes,
implying the involvement of sympathetic and nocifensive C-fibers.
Another important
observation, besides symptom spread, that needs explanation is that
symptoms increase in a highly time variable fashion. Pain can increase
in seconds with abrupt arm abduction or in many hours following five
minutes of writing. This argues for a time differential release or
induction of algesic compounds.
This symptom
sequence, whether initiated by either of the principal traumatic mechanisms,
implies a process which, over time, involves nerves adjacent to the
brachial plexus and, ultimately, the central nervous system (CNS).
This pattern of accelerated neural re-injury may explain the increase
in symptom formation with use.
The direct observation
of the scalene triangle and its contents, exposed during supraclavicular
decompression procedures, helps clarify this seemingly confusing panorama. Fibrous
bands, and persistent adhesions are visualized which limit the normal
movement of the plexus and impose deformation
and swelling, adhering it to adjacent structures
[4]. Most important, in terms
of the pathophysiology and continued symptom formation, is the occurence
of richly innervated small (<1cm. sq.) perineural adhesions. Careful resection of these perineural adhesions
results in the normalization of blood flow in the appropriate dermatomes
with resolution of the distal vasospasm as confirmed thermographically.
The relief of these symptoms and signs of vasospasm following neurolysis
implies that they originated in the relevant plexus roots [5]. In over one
hundred cases utilizing intraoperative thermography, decompressive
surgery documented the presence of both directly mechanical (fibrous
bands, fibrotically transformed muscle, adhesions) and neurogenic (small
perineural adhesions) mechanisms for neural constriction, inflammation,
and distal vasospasm. Substance P (SubP) and calcitonin gene related
peptide (CGRP), present in the resected perineural adhesions as substantiated
by histological immunoassay provide the basis for neurogenic inflammation
[5]. Re-operation underscores the relevance of these perineural adhesions
by the observed presence of exuberant recurrence.
Evidence from
similar disorders in ulnar and median nerve entrapments provides corroborating
evidence for the primacy of fibrosis in initiating and continuing symptomatology
in these disorders. Surgical
decompressions of the ulnar and median nerves in the elbow, forearm,
wrist, and hand all note the presence of significant fibrosis, occurring
both peri- and endoneurally. Endoneural
fibrosis was felt to be most damaging because of its primary appearance
and more direct effect on nerves and blood vessels [6].
Based on our
own intraoperative findings of highly innervated perineural fibrosis
secreting inflammatory factors, it is highly likely that the endoneural
fibrosis is similarly innervated. Neuropeptides
expressed by this innervated fibrosis are well recognized as important
mast cell activators [7]. Mast cells reside in significant numbers
endoneurally, clustering at C-fiber convolutions, synaptic junctions
of the nervi nervorum, and peri-vascularly about the vasa nervorum
[8]. Their activation is biphasic with either explosive extrusion of
granules which can volume increase by a factor of 1000 in milliseconds
or by a slow tonic release lasting hours to days [9]. Messengers include
histamine, proteoglycans, proteases, leukotrienes, prostaglandins,
interleukins, interferon gamma, tumor necrosis factor (TNF) alpha and
tryptases which initiate fibroblast proliferation and collagen deposition
[10, 11].
These mast cell
mediators, coupled with the direct effect of neuropeptides in the vasa
nervorum produce loosening of endothelial junctions, localized edema,
and recruitment of lymphocytes and monocytes [12].
The result is a localized compartment syndrome inside the relatively
rigid perineurium. The tortuous arrangement of fascicles in peripheral
nerves embedded in a fibrous matrix of resident mast cells, macrophages, Schwann cells, and perineural cells is made
to order for the mechanical propagation of this, neurogenically induced
local compartment syndrome [13]. Lengthy symptom occurrence, as well
as fibroblast proliferation and fibrosis follow.
This fibrosis,
spreading perineurally with reinjury, fixes the nerve, which is normally
free to travel significant distances, focally increasing the mechanical
strain on the nerve, and thus causing vasospasm of the vasa nervorum
[14]. Delayed symptom occurrence could be explained
by the slow tonic release of vasospastic mast cell mediators noted
above. In and of itself, this facilitates the creation of intermittent
symptoms, much like a pressure cuff. What
is more worrisome, and certainly more debilitating over time, is the
consequent proliferative response, which worsens symptoms, the ease
with which they arise, and lengthens any recuperative processes. What
is originally a neurogenically induced localized inflammation with
it’s consequent pain begins to spread both endoneurally and perineurally,
most likely via cytokine mediated mechanisms promoting adhesions, their
innervation and, ultimately, the activation of the
CNS.
Neural ischemia
gives rise to yet another dimension of injury:
ischemia-reperfusion, which has been shown to produce entrapment symptomatology
even with minimal ischemia as long as it is repetitive [15]. This could
augment and consolidate the above tissue changes
Additionally, pharmacological manipulations
corroborate the primacy of neurogenic inflammation in TOS.
Mu receptor
agonists are minimally effective in controlling the chronic pain of
TOS, producing a generalized numbness, but little amelioration of the
more intense symptoms. This
is characteristic of neuropathic/neurogenic disorders [16]. Interestingly, nalbuphine, a kappa and sigma
agonist, is consistently more effective and has minimal side effects.
Octreotide,
a somatostatin analog which quickly down regulates neuropeptide production
(SubP, CGRP, VIP, etc.}, is effective during early flare-ups, corroborating
the importance of neurogenic peptides in initiating symptoms [17]. Octreotide labeled positron emission tomography
is a potential diagnostic method because of its ability to image localized
neurogenic inflammation.
Topical nitroglycerin
can lessen pain significantly, more so than opiates. This points to
the importance of endothelial factors that are nitric oxide mediated as well as the importance
of vascular integrity [18].
Anti-inflammatory
agents modulating cyclo-oxygenases show
little symptomatic benefit, possibly quieting the effects of direct
mechanical irritation.
Substantiating
the importance of TNF-alpha, a major cytokine of mast cells, Etanercept,
a TNF-alpha blocker, eliminates the diffuse and spreading pain accompanying
flare ups (personal communication) [19].
More dramatically,
low-dose heparin produces relief, lasting weeks to months. Heparin is a complicated mix of glycose aminoglycans
with pleiotropic functions. Only
3% of the heparin functional sites are dedicated to anticoagulation. The
rest regulate the extracellular matrix, cytokines, chemokines, growth
factors, normalize the endothelium, leukocyte migration, cellular junctions,
and stabilize mast cells. Its
tissue half life in the extracellular matrix is on the order of several
weeks [20]. Heparin has been
shown to reduce: post-radiation neuropathy, inflammation and pain in
severe burns, asthmatic bronchoconstriction, the symptoms
of irritable bowel syndrome, and the
pain of nerve entrapments by its direct action on peripheral nerves
[21, 22]. CONCLUSION Stretch and
micromovement activate nervi nervorum to discharge inflammatory neuropeptides
with consequent vasospasm and ischemia. These
neuropeptides (SubP, CGRF and others) can also produce sudden mast
cell degranulation with acute symptoms, or, alternatively, a more gradual
release, increasing symptomatology over time and explaining the clinically
observed time variable symptom formation. These cytokines and vaso-active
compounds, released by mast cells and nervi nervorum (primarily nocifensive
C-fibers), result in increased neural irritability and sensitivity,
progressively involving the local microvasculature with increased vasospasm,
initiating endothelial dysjunction, edema, leukocyte and macrophage
migration with a resulting endoneural compartment syndrome [23]. Fibroblast
and mast cell migration follow as does growth factor (NGF, BDNF, FGF,
etc.) induced innervated fibrosis [24, 25]. Reinjury occurs easily
in these sensitized nerves , the fibrosis spreads, resulting
in a vicious cycles of progressive
neuronal activation, edema, stasis, and more extensive, innervated
fibrosis. Given the relatively rigid perineurium, and the sinuous and
interdigitating course of the fascicles, a compartment syndrome results
Continued cytokine elaboration is sufficient to evoke
symptom spread, the establishment of double or triple crush injuries,
CNS sensitization as well as the
appearance of a mirror syndrome [26, 27]. Kindling, hypo/hyperesthesia,
allodynia, and Raynaud’s phenomenon follow.
Based on the
well-documented etiology of TOS in the workplace, constant stretch
and small movements are enough to set off this vicious cycle. Repetitive
non deforming stretch (e.g. keyboarding) has been shown to induce neural
ischemia and bleb formation, sensitizing the nerves and providing the
initial insult for the neurogenic cascade described above [28].
Direct traumatic
onset (excessive flexion/extension and seatbelt contusions during motor-vehicle
accidents, clavicular fractures, thrombi or aneurysms) can initiate
enough of an inflammatory response through tissue necrosis, to also
allow the above perturbations to take hold [29].
Neurogenically
induced perineural inflammation could induce scalene muscle fibrosis
and transformation as well as initiate or enhance the pathological
effect of existing interdigitations or fibrous bands [30].
The current etiologic explanation for TOS of volume increase in the
scalene triangle and consequent compression of the trunks of the plexus
is insufficient to account for the late onset of symptoms and the consequent
progression and elaboration of symptoms [31]. Direct irritation of
the plexus is certainly important but it is not sufficient. Neurogenic
inflammation with its cascade of swelling, cellular recruitment and
fibrosis provides a more comprehensive explanation.
This description
of pathological events and their consequences explains much of the
often cryptic initiation and
clinical progression of TOS.
The spread of innervated fibrosis with continued reinjury presents
a mechanism for increasing symptoms, the involvement of sympathetic
and motor nerves, as well as distal, proximal and adjacent neural sensitization.
Reflection extends this process to other entrapments as indicated by
the prevalence of intraoperative reports of localized fibrosis, perineural
thickening, and edema. Idiopathic low back pain, postoperative pain,
and sacrococcygeal syndrome, among others, could have similar pathophysiologies.
This hypothesis
allows for greater diagnostic and clinical clarity and redefines the
direction of treatment. Long term mast cell stabilizing agents, cytokine
and neuropeptides inhibitors, and endothelium stabilizing agents can offer the hope of better treatment in the
future.
REFERENCES 1) Amadio P: Editorial comment. In Yearbook of Hand
Surgery. Edited by Berger R, Ludd A. Chicago: Moseby; 2002:160.
2) Juvonen T,
Satta J, Laitala P, Luukkonen K, Nissinen J: Anomalies of the thoracic outlet are frequent in the general population. Am
J Surg 1985, 170:33-37.
3) Gillard J,
Perez-Cousin M, Hachulla E, Remy J, Hurtevant J, Vinckier L, Thevenon
A, Duquesnoy B: Diagnosing thoracic
outlet syndrome: contribution of provocative tests, ultrasonography,
electrophysiology, and helical computed tomography in 48 patients. J
Bone Spine 2001, 68(5):416-424.
4) Cheng S,
Ellis W, Stoney R: Thoracic
outlet syndrome: supraclavicular approach. In Vascular Surgery-
5th edition. Edited by Jamieson C, Yao J. London: Chapman
and Hall; 1994:475-482.
5) Ellis W,
Cheng S: Intraoperative thermographic
monitoring during thoracic outlet decompressive surgery. Vasc Endovasc
Surg 2004, 37(4):253-257,
6) Rosenbaum
R, Ochoa J: Acute and chronic
mechanical nerve injury: pathologi, physiologic, and clinical correlations. In
Carpal Tunnel Syndrome and Other Disorders of the Median Nerve. Boston:
Butterworth- Heinemann; 1993:197-230.
7) Jasin L,
Janni G: Experimental neurogenic
cystitis. Adv Exp Med Biol 2003, 539(P&A):319-335.
8) Dines K,
Powell H: Mast cell interactions
with nervous system; relationship to mechanisms of disease. J Neuropathol
Exp Neurol 1997, 56(6):627-640.
9) Theoharides
T: Mast cells and stress- a
psychoneuroimmunological perspective. J Clin Psychopharm 2002, 22(2):103-114.
10) Siddiqui
A, Miner P: The role of mast
cells in common gastrointestinal diseases. Current Science 2004, 4: 47-54.
11) Claman H: Mast cells and fibrosis: the relevance to
scleroderma. Rheum Clin North Am, 16:141-151.
12) Richardson
J, Vasko M: Cellular mechanisms
of neurogenic inflammation. Pharm Experi Therapeutics 2002, 302(3):839-825.
13) Sunderland
S: Fascicular anatomy. In
Nerve Injuries and their Repair. Edinburgh: Cherchill-Livingstone;
1991:31-40.
14) Greening
J, Smart S, LearyR, Hall-Craggs M, O’Higgins P, Lynn B: Reduced movement of median nerve in carpal tunnel during wrist flexion
in patients with non-specific arm pain. Lancet 1999, 354:217-218.
15) Suo V, Freeland
A: Biochemistry of carpal tunnel
syndrome. Microsurgery 2004 Oct 12 [e-pub ahead of print].
16)Kupers R,
Gybels J: Responsiveness of
chronic pain to morphine. Lancet 1992, 340:310-311.
17)Ellis W: Octreotide, a small peptide, alleviates
burning pain and hyperesthesia: a preliminary study. The Pain
Clinic 1990, 3(4):239-242.
18) Ferreira
S, Lorenzetti B, Faccioli L: Blockade
of hyperalgesia and neurogenic edema by topical application of nitroglycerin. European
J Pharm 1992, 217:207-209.
19)Wagner R,
Myers R: Endoneurial injection
of TNF-alpha produces neuropathic pain behaviors. Neuro Report
1996, 7:2897-2901.
20) Ellis W: Heparin alleviates pain in nerve entrapments. AJPM
2003, 13(2):54-59.
21) Glantz M,
Burger P, Friedman A, Radtke R, Massey E, Schold S: Treatment of radiation-induced nervous system injury with heparin and
warfarin. Neurol 1994, 40:2020:2027.
22) Saliba M: The effects and uses of heparin in the care
of burns that improves treatment and enhances the quality of life. Acta
Chirurgiae Plasticae 1997, 39(1):13-16.
23) Kruger L: Morphological features of thin sensory efferent
fibers: a new interpretation. Prog Brain Res 1988, 74:253-257.
24) Welker P,
Grabb J, Grutzkau A, Heinz B. Effects
of nerve growth factor (NGF) and other fibroblast-derived growth factors
on immature human mast cells (HMC-1). Immunology 1998, 94(3):310-317.
25) Anand P: Nerve growth factor regulated nociception
in human health and disease. Brit J Anesth 1995, 75:201-208.
26) Golovchinsky
V: Relationship between damage
of cervical nerve roots on brachial plexus and development of peripheral
entrapment syndrome in upper extremity (double crush syndrome). J
neurol Orthop Med Surg 1995, 16:61-69.
27) Twining
C, Sloane E, Milligan E, Chacur M, Martin D, Poole S, Marsh H, Maier
S, Watkins L. Peri-sciatic pro-inflammatory cytokines,
reactive oxygen species, and complement induced mirror-image neuropathic
pain in rats. Pain 2004, 110:299-309.
28) Markin V,
Tanelian D, Jersild R, Ochs S: Biomechanics
of stretch-induced beading. Biophysical J 1999, 76(5):2852-2860.
29) Sanders
R, Haug C: Etiology. In
Thoracic Outlet Syndrome- a common sequela of neck injuries. Philadelphia:
Lippencott; 1991:21-23.
30) Gruber B,
Kew S, Jelaska A, Marchese M, Garlick J, Ren S, Schwartz L, Korn J: Human mast cells activate fibroblasts.
J Immunology 1997, 158:2310-2317.
31) Brantigan
C, Roos D: Etiology of neurogenic
thoracic outlet syndrome. Hand Clin 2004, 20:17-22. |