Carpal Tunnel Syndrome
Carpal Tunnel Syndrome
Carpal Tunnel Syndrome
Chapter 4
1
Westmead Clinical School, University of Sydney, Sydney, NSW, Australia
2
Northern Beaches Clinical School, Macquarie University, Sydney, NSW, Australia
Abstract
Median neuropathy at the wrist, commonly referred to as carpal tunnel syndrome (CTS), is the most com-
mon entrapment neuropathy. It is caused by chronic compression of the median nerve at the wrist within the
space-limited carpal tunnel. Risk factors that contribute to the etiology of compression include female gen-
der, obesity, work-related factors, and underlying medical conditions, such as hypothyroidism, pregnancy,
and amyloidosis. The diagnosis is made on clinical grounds, although these can be confounded by ana-
tomical variations. Electrodiagnostic studies, which are specific and sensitive in diagnosing CTS, support
the diagnosis; however, a subgroup may present with normal results. The advent of imaging techniques,
including ultrasound and MRI, further assists the diagnostic process. The management of CTS is divided
into the nonsurgical approaches that include hand therapy, splinting and corticosteroid injection, and
surgical decompression of the carpal tunnel. Although several surgical techniques have been developed,
no one method is more effective than the other. Each of these management approaches are effective at
providing symptom relief and are utilized at different severities of the condition. There is, however, a lack
of consensus on standardized diagnostic criteria, as well as when and to whom to refer patients for surgery.
⁎
Correspondence to: Prof. Neil Simon, MBBS, PhD, FRACP, Suite 6a, Northern Beaches Hospital, 105 Frenchs Forest Rd W,
Frenchs Forest, Sydney, NSW, 2086, Australia. Tel: +61-2-9982-2270; Fax: +61-2-9981-7880, E-mail: neil@nbneuro.com.au
62 N. HARINESAN ET AL.
ANATOMY border of the TCL (Madhav et al., 2009). Inferiorly, the
tunnel is bound by the anterior surface of the carpal
Structure of the carpal tunnel bones, which is convex in structure and termed the sulcus
The carpal tunnel is an osteofibrous channel situated in carpus. The sulcus carpus is bound laterally by the radial
the ventral wrist, through which passes the median nerve, carpal eminence and medially by the ulnar carpal emi-
along with nine tendons; four from the flexor digitorum nence, to which the transverse carpal ligament attaches.
superficialis (FDS), four from the flexor digitorum pro- The tendons of the flexor muscles of the hand consti-
fundus (FDP), and the tendon from the flexor pollicis tute the bulk of the contents of the carpal tunnel. These
longus (FPL)) (Fig. 4.1). Superiorly, the transverse carpal muscles originate from the medial epicondyle of the
ligament (TCL), a strong fibrous ligament, forms the roof humerus and the anterior aspects of the radius, ulnar,
of the tunnel and is attached to the pisiform bone and and interosseous membrane. By the time these muscles
hook of hamate bone on the ulnar side. On the radial side, enter the proximal portion of the carpal tunnel, they
it attaches to the scaphoid bone and the ridge of the are mainly tendinous in nature. The arrangement of the
trapezium bone. The TCL splits into a superficial lamina contents of the carpal tunnel can be variable, though
(with attachments to the tubercle of the scaphoid and are usually easily identifiable on imaging studies through
trapezium) and deep lamina (with attachments to the the surrounding ulna (which separates the superficial and
medial lip of the ridge of the trapezium). These laminae deep tendons) and radial bursae (which separates the
along with the ridge of the trapezium, form a small sep- FPL) (Middleton et al., 1987). The most superficial mus-
arate fibro-osseous tunnel lined with a synovial sheath cle in this region is the FCR, though it is technically not
containing the tendon to flexor carpi radialis (FCR) within the carpal tunnel proper as it has its own synovial
(Rotman and Donovan, 2002). Proximal to the TCL is sheath as outlined above. Deep to the FCR lie the four
a thickening of the deep investing fascia of the forearm, tendons to the flexor digitorum superficialis (FDS),
and distally is the aponeurosis between the thenar and with the tendons to the middle and ring fingers lying
the hypothenar muscles (Cobb et al., 1993). This triseg- slightly superficial to those of the index and the little
mented structure is collectively termed as the flexor fingers (Mitchell et al., 2009; Ghasemi-rad et al.,
retinaculum (FR), though many authors use the term 2014) (Fig. 4.1).
TCL synonymously (Stecco et al., 2010). The function The size of the carpal tunnel varies in healthy subjects.
of the FR is to serve as a flexor pulley at the wrist, The mean width of the tunnel has been reported as
preventing tightening of the tendons. 25 1.2 mm proximally, 20 1.2 mm at the hook of
Immediately proximal to the distal end of the TCL, in the hamate, and 25 1.5 mm at its distal extent, though
line with the axis of the ring finger, is a palmar fat pad. these measurements vary depending on the population
This is a reliable anatomic landmark that, in the surgical studied (Cobb et al., 1993; Pacek et al., 2010). The length
setting, requires retraction prior to visualizing the distal of the carpal tunnel has been reported to be 12.7 2.5 mm
Fig. 4.1. Cross-sectional anatomy of the carpal tunnel. The tendons of their respective muscles are labeled. PL, palmaris longus;
FCR, flexor carpi radialis; FPL, flexor pollicis longus; FDS, flexor digitorum superficialis; FDP, flexor digitorum profundus; TCL,
transverse carpal ligament.
CARPAL TUNNEL SYNDROME 63
when measuring the strict boundaries of the TCL (Cobb 2021). The reduction in space and subsequent increase
et al., 1993; Pacek et al., 2010). The cross-sectional area in pressure that follows can impair conductivity of the
of the carpal tunnel ranges between 151–175 mm2 proxi- enclosed median nerve through mechanical pressure
mally and 173–187 mm2 distally (Dekel et al., 1980; and impaired blood flow through the epineurium (Sevy
Yoshioka et al., 1993; Monagle et al., 1999; Bower and Varacallo, 2022).
et al., 2006). Similarly, the carpal tunnel volume is from
3430 to 4760 mm3 using the proximodistal boundaries
of the TCL (Richman et al., 1987; Cobb et al., 1992;
Median nerve anatomy
Bower et al., 2006). The median nerve is one of the main and most superficial
The size and shape of the carpal tunnel also varies structures in the carpal tunnel. The median nerve is
depending on wrist position. With wrist extension, the formed by the union of lateral and medial roots, which
carpal tunnel area increases distally and decreases prox- originate from the lateral and medial cords of the brachial
imally, while wrist flexion decreases the cross-sectional plexus, respectively (Fig. 4.2). The nerve then runs down
area throughout the tunnel (Bower et al., 2006). In the anteromedial aspect of the arm (initially lateral to the
addition, gendered differences have been noted, with brachial artery, then medial once it crosses over in the
females having consistently lower carpal tunnel areas middle of the upper arm), before descending into the
and volumes (Sassi and Giddins, 2016). Anatomical cubital fossa deep to the aponeurosis of biceps brachii,
anomalies can also narrow the tunnel, as structures that remaining medial to the brachial artery (Fig. 4.2). It then
typically run outside of it intrude into it. In the case of gives off a purely motor branch, the anterior interosseous
tendons, this can include palmaris profundus, FDS, or nerve (AIN), before entering the forearm between the
lumbricals, and in the case of vessels, this can include heads of pronator teres. The AIN goes on to innervate
the superficial branch of the radial artery or the presence FPL, FDP (index and middle fingers), and pronator quad-
of a persistent median artery (Cobb et al., 1994; Olave ratus (PQ), while the median nerve continues between
et al., 1997; Takata and Roll, 2019; Pezas and Jose, FDS and FDP, before entering the carpal tunnel just
Fig. 4.2. Median nerve anatomy in the arm, depicting its branches and the muscles they supply.
64 N. HARINESAN ET AL.
lateral to the tendons of FDS and between the tendons of TCL, as seen in approximately 9% of patients undergo-
FCR and palmaris longus (PL). At this stage, it is rela- ing carpal tunnel surgery (Hurwitz, 1996).
tively superficial with only the tendon of PL and fascia Another classification system is described as Lanz
covering it (Iskra et al., 2013). Another branch, the pal- Types 1–4 (Lanz, 1977). The Type 1 variations represent
mar cutaneous, arises from the radial edge of the median the differences in the course of the thenar branch, while
nerve in the distal forearm, approximately 5–6 cm prox- Types 2–4 include an accessory branch distal to the
imal to the distal transverse flexion crease of the wrist. tunnel, which is a high division of the median nerve
It travels alongside the median nerve and then the tendon and an accessory branch proximal to the tunnel, respec-
of FCR, before progressing more superficially through tively (Lanz, 1977). A high division of the median nerve
the forearm fascia, prior to the carpal tunnel, to supply (or bifid median nerve) is present in 1%–3% of individ-
sensory branches to the deep dermis palm in 2 or 3 uals undergoing carpal tunnel release and up to 17% of
terminal divisions (Smith and Ebraheim, 2019). healthy subjects (Lanz, 1977; Lindley and Kleinert,
As the median nerve enters the carpal tunnel, it is cov- 2003; Mitchell et al., 2009; Shinagawa et al., 2019). This
ered only by the TCL and lies superficial to the nine can be isolated, but is often also associated with a persis-
flexor tendons, traveling between the tendons of FCR tent median artery, which can travel between the two
and PL. After it traverses the TCL, it flattens and divides nerve branches (Spagnoli et al., 2017).
into a medial and lateral branch. The cross-sectional area The median nerve is also involved in clinically rele-
of the median nerve at this point in the palm is slightly vant anastomoses. The Riche-Cannieu anastomosis is
larger than at the distal wrist crease (Jain et al., 2020). an interconnection between the recurrent branch of the
The medial branch terminates as two common palmar median nerve and deep branch of the ulnar nerve, with
digital nerves supplying motor innervation to the 2nd involvement of both motor and sensory fibers, with an
lumbrical and sensory innervation to the palm and estimated pooled prevalence of 55% (Henry et al.,
fingers. The lateral branch gives rise to the thenar motor 2015). There are three possible presentations; a hand
branch (TMB) before terminating in the two proper that is solely supplied by the ulnar nerve, a hand with
digital nerves, supplying motor innervation to the 2nd motor supply only through the ulnar nerve, and a hand
lumbrical and sensory innervation to the lateral hand. with traditionally median supplied muscles being par-
The TMB supplies motor innervation to opponens polli- tially supplied by the median nerve (Felippe et al.,
cis, abductor pollicis brevis, and the superficial part of 2021). In these cases, weakness of thenar musculature
flexor pollicis brevis, before terminating in sensory cannot be used to accurately predict median nerve com-
branches. promise. The Berrettini anastomosis is a neural connec-
tion between the common digital nerve of the ulnar and
median nerves, often occurring between the middle and
Anatomical variants
ring fingers and often bilateral, and is purely sensory. The
Several anatomic variations of the median nerve in the pooled prevalence is estimated at 61% (Henry et al.,
carpal tunnel have been reported, in particular regarding 2015). This is often seen just distal to the TCL and is
the variance of the path of the TMB. These variations at risk of iatrogenic injury during carpal tunnel surgery
have been described using Poisel’s classification system, in addition to other hand surgeries (Felippe et al.,
distinguishing three types of TMB branching: extra- 2021). Lesioning of the anastomosis results in reduced
ligamentous (type 1), subligamentous (type 2), and sensation in the region between the middle and ring
transligamentous (type 3) (Poisel, 1974) (Fig. 4.3A–C, fingers.
respectively). The standard anatomy of the TMB is extra-
ligamentous, as described above, with a pooled preva- CARPAL TUNNEL SYNDROME
lence of 75.2% (55.4%–84.7%) of patients (Henry
Epidemiology
et al., 2015). In the subligamentous type, the branch
arises from the median nerve within the tunnel and Carpal tunnel syndrome is the most frequent entrapment
remains deep to the TCL until it reaches its distal end, neuropathy (Padua et al., 2016). The female to male ratio
before bending back around to supply the motor compo- is 3:1 (Stevens et al., 1988; Atroshi et al., 1999, 2011a;
nents. In the transligamentous type, the branch arises Pourmemari et al., 2018) with an estimated annual inci-
from the median nerve within the tunnel and then pierces dence of 2.2–5.4 per 1000 in females and 1.1–3.0 per
the TCL to supply the motor components (Mitchell et al., 1000 in males (Gelfman et al., 2009; Atroshi et al.,
2009). While the TMB is generally located on the radial 2011a; Pourmemari et al., 2018). Prevalence in the com-
side of the median nerve, a rarer origin from the ulnar side munity ranges from 1% to 5%, though this is higher in
of the median nerve has been described (Graham, 1973) older patients and those with an elevated body mass
(Fig. 4.3D). It may also take a course superficial to the index (BMI) (Werner et al., 1994; Atroshi et al., 1999;
CARPAL TUNNEL SYNDROME 65
Fig. 4.3. Common anatomical variants of the origin of the thenar motor branch. (A) Extraligamentous thenar motor branch (TMB)
(Poisel type 1) is the most common configuration. (B) Subligamentous TMB (Poisel type 2). (C) transligamentous TMB (Poisel
type 3). (D) Rare configuration of the TMB arising from the ulnar side of the median nerve.
Shiri et al., 2015). Higher rates have also been reported symptoms and pain becoming more prominent. In fact,
in working populations (Dale et al., 2013). Life-time pain is reported in 52% of patients (Padua et al.,
prevalence of clinical CTS is estimated to be 1 in 10 per- 1999). With further progression, there can be a loss of
sons (Foley et al., 2007; AAN (American Academy of sensation in the hand, weakness of thenar muscles, and
Neurology), 1993). Each of these estimates is influenced atrophy of the thenar eminence, corresponding to the
by the differences in diagnostic criteria used in different degree of axonal degeneration (Padua et al., 2016).
reports and is highest in the setting of diagnoses made The distribution of sensory symptoms in CTS have
with clinical criteria alone. Regardless, it is clear that this been extensively studied and demonstrated to occur in
is a highly prevalent disorder. a much broader region than just the aspect of the hand
innervated by the median nerve (Stevens et al., 1999;
Tecchio et al., 2002; Zanette et al., 2006; Clark et al.,
Clinical characteristics
2011). A distribution of symptoms primarily in a median
At its onset, CTS is characterized by the presence pattern (that is, involving the first three digits) has been
of intermittent, primarily nocturnal, paresthesia, and associated with increased neurophysiological severity
dysesthesia. These symptoms increase in frequency of median neuropathy at the wrist (Caliandro et al.,
and severity as the syndrome progresses, with daytime 2006). The most common subjective sensory symptom
66 N. HARINESAN ET AL.
is nocturnal acroparesthesia, which is a painful tingling In severe cases, this can lead to constriction and tethering
in the hand at night often relieved by a change in limb of the median nerve, which can further lead to injury with
position, shaking, massage, or immersion in cold water. repetitive movements (MacDermid and Doherty, 2004).
Symptoms may also occur during the day, triggered by Increased pressure may also be a result of other anoma-
activities that increase compression on the median nerve lous contents within the canal that can decrease the avail-
by narrowing the tunnel, such as sewing or driving. In able canal space (see Other Causes section).
addition, symptoms proximal to the wrist are common; Animal models of chronic nerve compression have
up to 45% of patients experience pain that radiates from demonstrated a dose–response relationship between
the wrist proximally as high as the shoulder (Cherington, the duration of compression and neural injury, with the
1974). This was also shown to occur more frequently at initial changes being a breakdown in the blood nerve
night, be associated with paresthesia in the hand outside barrier, followed by subperineurial edema and fibrosis,
the median nerve territory, and be inversely correlated with subsequent localized then diffuse demyelination
with neurophysiological severity of median nerve dam- (Mackinnon et al., 1985; O’Brien et al., 1987;
age (Caliandro et al., 2006; Zanette et al., 2006, 2007). Mackinnon and Dellon, 1988). Histopathologic exami-
Motor symptoms including weakness of the abductor nation of median nerve samples from cadavers with
pollicis brevis muscle occur at a late stage of disease, and CTS during life demonstrate segmental demyelination
manifest as reduced manual dexterity or hand weakness. in large myelinated fibers, degeneration and regeneration
This can be found in up to 50% of patient with CTS of small unmyelinated fibers, and Wallerian degenera-
(Tamburin et al., 2008), though the degree of median tion (Mackinnon, 2002). Fibrosis within the subperineur-
nerve damage poorly correlates to the degree of hand ial space and thickening of the epineurium with Renaut
weakness or clumsiness (Tamburin et al., 2008; Nazari bodies has also been demonstrated (Mackinnon, 2002).
et al., 2017). Early compressive injury leads to venous flow obstruc-
tion causing the nerve to become hyperemic and edem-
atous, due to an increase in intrafunicular pressure
Pathophysiology
(Sunderland, 1976). Subsequent stasis of blood leads
The pathophysiology of CTS is multifactorial, and to obstruction of flow within the vasa nervorum, causing
though not fully elucidated, is thought to be a result of ischemia (Werner and Andary, 2002).
the interplay between increased pressure within the The natural history of carpal tunnel syndrome
carpal tunnel, mechanical trauma, and ultimately ische- accurately parallels this presumed model. As outlined
mic injury to the median nerve within the tunnel in Clinical Characteristics section, initial complaints
(Werner and Andary, 2002). are of intermittent paresthesia, suggesting intermittent
As with any anatomical canal, the carpal tunnel is a ischemic injury due to transient increases in carpal tunnel
fixed space sensitive to internal and external pressures. pressure impairing neuronal blood flow (Szabo and
Increased pressure plays an important role in the devel- Gelberman, 1984). With repeated exposure to compres-
opment of CTS; however, there are only a small number sion, there is segmental demyelination with more persis-
of studies in humans that address this issue. Normal pres- tent paresthesia and weakness (Szabo and Gelberman,
sure within the carpal tunnel is between 2 and 10 mmHg, 1984). Chronic prolonged compression is then associ-
which can increase 10-folds on wrist extension and ated with axonal injury and Wallerian degeneration
8-folds with wrist flexion (Werner et al., 1983; Werner manifesting as numbness and persistent weakness
and Armstrong, 1997). Movements in the volar, radial, (Mackinnon, 2002).
and ulnar planes are thought to result in smaller increases
in pressure as the range of motion in these directions Risk factors for carpal tunnel syndrome
is limited. Repetitive mechanical movement in flexion
Most cases of CTS are idiopathic; however, some risk
and extension can cause chronic exposure to high
factors and secondary causes are associated with a higher
intra-compartmental pressures, subsequently leading to
prevalence of the condition.
injury to the median nerve. In addition, pressure is also
exerted on the median nerve by the surrounding connec-
OCCUPATIONAL-RELATED RISK FACTORS
tive tissue, which itself can be affected by repetitive
movements resulting in local inflammation (Phalen, A landmark 1997 epidemiological study found strong
1966). Synovial tissue biopsy from patients with CTS evidence for occupational exposure to repetition, pos-
reveals increased expression of prostaglandin E2 and ture, and vibration associating with the development of
vascular endothelial growth factor, as well as increased CTS (Bernard, 1997). These findings are consistent
fibroblast density, collagen fiber size, and vascular with the high rates of CTS in occupations with repetitive
proliferation (Ettema et al., 2004; Hirata et al., 2004). manual exertion (Werner, 2006). Exposure to vibration
CARPAL TUNNEL SYNDROME 67
has been associated with a fivefold increase in CTS AGE
risk, while exposure to hand force confers a fourfold
Middle-aged and older patients (45 years and above)
increase (Barcenilla et al., 2012). A combination of
have an increased incidence of CTS compared to those
repetitive use or repetitive use and force was associated
who are younger (Atroshi et al., 1999; Guan et al.,
with a doubling of these risks (Barcenilla et al., 2012).
2018). Women have a peak incidence in the fifth decade
Occupations that might be exposed to these hazards
while men have a bimodal peak in the fifth and seventh
include quarry/rock drillers and chainsaw users in for-
decades (Mondelli et al., 2002). The mechanisms under-
estry (increased hand transmitted vibration), and assem-
lying an age-related risk are unclear, but have been
bly workers or food processors and packers (sustained/
shown to exist independently of other risk factors.
repeated wrist movements) (Palmer, 2011).
The association between computer use and the
development of CTS is less clear. Meta-analyses have TENOSYNOVITIS
reported either no significant association or a minor asso- Tenosynovitis occurs due to inflammation of the
ciation with modest computer use in office workers fluid-filled synovium within the tendon sheath causing
(Mediouni et al., 2014; Shiri et al., 2015). thickening. It can be secondary to overuse, infective
conditions (including S. aureus and other common skin
commensals), autoimmune conditions (see below), and
FEMALE SEX can be idiopathic (Blood et al., 2016; Flevas et al.,
2019). Regardless of the cause, the swelling of the tendon
Female sex has been repeatedly associated with an sheaths within the carpal tunnel exerts a compressive
increased risk of CTS (Mondelli et al., 2002; Mattioli effect on the closely apposed median nerve and can
et al., 2008; Roquelaure et al., 2017). This may be influ- present as acute or chronic carpal tunnel syndrome.
enced by a number of factors, including anthropometric
variables, hormonal factors, and pregnancy. Women CONNECTIVE TISSUE DISEASES
have been shown to have a smaller carpal tunnel area
and less free space within the tunnel (Bower et al., Rheumatoid arthritis has been strongly associated with
2006). Hormonal factors are not fully elucidated; how- the development of CTS, occurring in approximately
ever, upregulation of estrogen receptors has been demon- 23% of patients (Sofat et al., 2006). Rheumatic disease
strated in tenosynovial tissue in postmenopausal women causes flexor tenosynovitis and sonographic changes
(Kim et al., 2010), potentially implicating estrogen in the in finger flexor tendons and/or radio-carpal joints have
pathogenesis of CTS. The increased risk in pregnancy is been demonstrated more commonly in patients with
thought to relate to hormonal factors causing a fluctua- rheumatoid-related CTS (Smerilli et al., 2021). Other
tion in carpal tunnel contents and fluid retention, leading collagen disorders, such as systemic lupus erythemato-
to an overall increase in carpal tunnel pressure (Wand, sus, scleroderma, and Sjogren’s syndrome have also
1990; Turgut et al., 2001; Yazdanpanah et al., 2012). all been associated with an increased risk of CTS
Presentation is often during the third trimester, however, (Sidiq et al., 1972; Sriwong et al., 2018; Jaskólska
has been reported throughout pregnancy and postpartum et al., 2020), likely related to similar pathophysiologic
(Gould and Wissinger, 1978; Wand, 1990). Symptoms of mechanisms.
CTS remit spontaneously in 50% of women by 1 year
postpartum and 70% of women by 3 years; the mecha- POLYNEUROPATHY
nism for this is not fully understood (Padua et al., 2010). Polyneuropathy of any cause is thought to increase the
risk of entrapment and compressive neuropathy, likely
due to an increased sensitivity of the neural structures
INCREASED BMI to compression (Perkins et al., 2002). Diabetes is the
most common cause of polyneuropathy; those with
Obesity is an important modifiable risk factor for the diabetic polyneuropathy have an increased risk of devel-
development of CTS (Kurt et al., 2008; Shiri et al., oping CTS compared to patients who have diabetes
2015; Mansoor et al., 2017). A meta-analysis on the sub- without neuropathy (Moon et al., 2020).
ject showed a 7.4% increased risk of CTS for every
increase in BMI point (Shiri et al., 2015). The mechanism
AMYLOIDOSIS
for this is unclear but may be related to adipose tissue
deposition, fluid retention, or as a consequence of a Amyloidosis refers to a group of disorders that occur due
broader metabolic syndrome causing neuropathy or teno- to the aggregation of insoluble and misfolded proteins
synovitis (Bland, 2005; Callaghan and Feldman, 2013). eventually leading to organ dysfunction (Milandri
68 N. HARINESAN ET AL.
et al., 2020). Amyloidosis can be characterized as Type 1, et al., 2004). As many as 80% of patients have CTS
in which proteins are related to a monoclonal expansion symptoms and subsequent surgical release prior to the
of light chains; or Type 2, caused by the deposition ultimate diagnosis of acromegaly (Vouzouneraki et al.,
of transthyretin (Milandri et al., 2020). The rates of 2021). The development of CTS may relate to irrevers-
CTS are higher in Type 2 amyloidosis, occurring in ible narrowing of the carpal tunnel as a consequence of
29%–68% of patients; however, it can still occur in up growth hormone excess (Kameyama et al., 1993).
to 30% of patients with Type 1 amyloidosis (Dubrey Female gender has also been shown to be an independent
et al., 1998; Maurer et al., 2016; Nakagawa et al., risk factor for the development of CTS in acromegaly
2016). Furthermore, CTS can be the presenting symptom (Vouzouneraki et al., 2021).
in up to 16% of patients with systemic amyloidosis
(Bishop et al., 2018).
CARPAL CANAL DEFORMITY
Beta-2 microglobulin amyloidosis related to chronic
renal failure is a complication of long-term dialysis Carpal tunnel syndrome has been shown to complicate
(Scarpioni et al., 2016), and is also associated with the approximately 4.3% of distal radial fractures, with a
development of CTS. In this condition, CTS is associated female preponderance (Dyer et al., 2008; Leow et al.,
with time on dialysis and beta-2 microglobulin clearance 2021). Comminuted fractures are associated with a
(Hoshino et al., 2014). higher risk of developing acute CTS, as is malunion
A study of 91 older patients undergoing carpal tunnel due to dorsal displacement of the capitate (Watanabe
release surgery identified amyloid deposits in 10% and Ota, 2019; Leow et al., 2021). Increases in carpal
(Sperry et al., 2018), with all positive patients presenting tunnel pressure, particularly with volar flexion, has also
with bilateral symptoms or a history of prior contralateral been shown to occur after Colle’s fracture (Kongsholm
carpal tunnel release. Amyloid precursor deposits have and Olerud, 1986).
subsequently been demonstrated in the flexor retinacu-
lum, synovial tissue, flexor tendon sheath, fascia, vessel
GENETIC RISK FACTORS
walls, and the endoneurium, though it is unclear which
of these is the pathological entity (Bastian, 1974; Stein A positive family history is reported by 27%–39% of
et al., 1987). patients with CTS (Radecki, 1994) and a large UK twin
study identified a heritability of 0.46 in developing CTS,
suggesting a moderate genetic basis for its development
ENDOCRINE DISEASE
(Hakim et al., 2002). A positive family history has
Diabetes mellitus is associated with the development of also been reported to be more common in bilateral
CTS (Oktayoglu et al., 2015; Chen et al., 2015a; Rydberg CTS compared to unilateral (Alford et al., 2004). Vari-
et al., 2020). However, more recent data suggest that this ants in genes including BGN, ACAN, COL5A1, and
association may not be as strong as previously thought IL6R have been identified as being associated with an
(Coggon et al., 2013; Hendriks et al., 2014). Mechanis- increased risk of CTS (Burger et al., 2014, 2015a,b).
tically, compression causes endoneurial hypoxia, which A genome-wide association study identified 16 suscep-
can be aggravated by glycosylation-end products in the tibility loci with three candidate genes (ADAMTS17,
setting of hyperglycemia (Rydberg et al., 2020). Diabetic ADAMST10 and EFEMP1) (Wiberg et al., 2019). More
control may be a key factor to explain some of the recently, mutations in the COMP gene have been identi-
discrepancies in the data, and further studies are needed fied in two families with a dominant inheritance pattern
to evaluate this (Low et al., 2021). of bilateral CTS (Li et al., 2020a).
Hypothyroidism has also been associated with a Additionally, genetic disorders such as Charcot–
modest increased risk of CTS (Shiri, 2014). The preva- Marie-Tooth neuropathy or hereditary transthyretin
lence of hypothyroidism in CTS is higher than in matched amyloidosis-associated neuropathy may predispose to
controls, with a pooled odds ratio of 1.4 (1.0–2.0) (Van the development of CTS (Gossett and Chance, 1998;
Dijk et al., 2003). CTS in hypothyroidism may be Elstner et al., 2006).
related to deposition of pseudomucinous material on the
median nerve, while uncontrolled hypothyroidism can
OTHER CAUSES
also cause swelling and thickening of the synovial mem-
branes in the carpal tunnel (Golding, 1970; Schwarzer Gouty tophi are a rare but recognized cause of CTS,
et al., 2010). and can be deposited on the flexor tendons surrounding
Carpal tunnel syndrome is also associated with the median nerve, increasing carpal pressure, and also
acromegaly, with a purported prevalence of 18%–64% directly compressing the median nerve (Chen et al.,
(Baum et al., 1986; Kameyama et al., 1993; Mestron 2000; Kim, 2014).
CARPAL TUNNEL SYNDROME 69
Anatomical vascular variations are rare causes of practitioners and their role in management decisions
carpal tunnel syndrome. A persistent median artery is remains controversial due to a paucity of strong evidence
the most common anomaly and can cause compression in this area.
of the median nerve due to aneurysm formation or throm-
bosis. An aberrant radial artery arising from the main
radial artery and descending into the wrist radially to Clinical assessment
medially, ending at the superficial palmar arch (Olave A thorough case history is essential to diagnosing CTS.
et al., 1996; Gwynne-Jones and Hartnett, 2008), is The typical presenting features are outlined in Clinical
present in 3% and may also predispose to CTS. We have Characteristics section. The history should also focus
also encountered venous malformations as a rare cause of on provocative factors, such as hand positions or repeti-
acute CTS (Fig. 4.4). tive hand movements; activities performed during work,
Infective causes of CTS are exceedingly rare, though such as instrument use or the use of vibrating tools; alle-
several case reports document a wide variety of fungal, viating symptoms, such as hand shaking or position
bacterial, mycobacterial, viral, and parasitic pathogens changes; sports activities, and the presence of predispos-
causing CTS (Mascola and Rickman, 1991). ing factors, such as diabetes, adiposity, chronic arthritis,
CTS can also be caused by the presence of space- acromegaly, pregnancy, or myxedema.
occupying lesions within the carpal tunnel. These Tinel’s and Phalen’s tests are easy to perform at the
include occult ganglion, lipoma, hemangioma, sarcoma, bedside and further support the diagnosis. The tests are
fibroma, or other calcifying mass (Kang et al., 2009). deemed positive when symptoms are reproduced by
tapping the wrist overlying the carpal tunnel (Tinel’s test)
or by forced flexed wrist position causing compression of
DIAGNOSIS
the tunnel for 1 min (Phalen’s test). The reported sensitiv-
Methods for the diagnosis of CTS vary greatly between ity and specificity of these tests vary widely. Tinel’s test
specialty and location, though a thorough clinical assess- sensitivity is reported to range from 38% to 100% and
ment remains the cornerstone (Padua et al., 2016). specificity 54% to 98%, while Phalen’s test sensitivity
The diagnosis of CTS is further supported with the use is reportedly 42% to 85% with specificity 55% to
of electrodiagnostic testing (EDX), including nerve- 100% (Br€uske et al., 2002). In the authors’ experience,
conduction studies and needle electromyography, and it is not uncommon that these bedside tests are negative
imaging modalities, including nerve ultrasound and in patients with a typical presentation and supportive
MRI. These investigations may help to confirm a median investigations; thus, sensitivity is unlikely to be as high
neuropathy as well as provide information on its severity. as suggested in these previous reports.
Additionally, these adjunctive investigations may assist A variety of clinical scores have also been developed
with the exclusion of mimics. It should be noted that to aid in the diagnosis. A self-administered hand diagram
not all of these investigations are available to all is a tool developed to assist in the evaluation of CTS, with
Fig. 4.4. Arteriovenous malformation (AVM) causing median neuropathy. (A) Gadolinium-enhanced T1-weighted image dem-
onstrates heterogeneous enhancement of a lobulated mass consistent with AVM supplied by a branch of the radial artery.
(B) T2-weighted image shows close opposition of the AVM to the median nerve (arrow).
70 N. HARINESAN ET AL.
a sensitivity of 80% and specificity of 90% (Katz The combination of clinical findings and EDX has
and Stirrat, 1990). The Boston CTS questionnaire and been shown to be the most accurate method to establish
CTS-6 score are brief clinical surveys of symptoms a diagnosis (Rempel et al., 1998). In the correct clinical
developed to quantify the severity and frequency of context, NCS has a sensitivity of >85% and a specificity
nocturnal and daytime numbness, tingling, and pain, of 95% in the diagnosis of CTS (Jablecki et al., 2002).
and are useful in classifying the severity of CTS and It should be noted that approximately 10%–15% of
monitoring response to treatment (Levine et al., 1993; patients with typical symptoms have normal nerve con-
Atroshi et al., 2011b). duction results, which may represent acute and rapidly
reversible ischemia brought about by functional com-
pression in the wrist (Gelberman et al., 1981). It is also
Electrodiagnostic testing (EDX) common in clinical practice to demonstrate EDX find-
ings consistent with median neuropathy in individuals
The purpose of EDX in the evaluation of suspected car-
with no symptoms; the predictive value of these changes
pal tunnel syndrome is to confirm median nerve dysfunc-
is unclear (Nathan et al., 1998; Werner et al., 2001).
tion at the level of the wrist and quantify and qualify the
extent and type of nerve injury. EDX used in the diagno-
SENSORY STUDIES
sis of carpal tunnel syndrome includes nerve conduction
studies (NCS) and needle electromyography (EMG). Sensory studies can be performed with either an ortho-
These studies are useful in differentiating proximal dromic or antidromic setup. Antidromic studies, with
median neuropathy, which might be confused clinically stimulation at the wrist and recording from the index fin-
with CTS, and may also lead to the identification of ger produce larger sensory nerve action potential (SNAP)
an entirely different or superimposed condition, which amplitudes as compared with orthodromic studies, in
could have management implications. The American which stimulation is performed at the index finger and
Academy of Neuromuscular and Electrodiagnostic responses recorded from the wrist (Goddard et al.,
Medicine (AANEM) guideline recommends performing 1983) (Fig. 4.5A). The recommended distance from
median sensory and motor studies, with the addition of stimulation to recording is 13–14 cm. A prolonged
comparison studies and EMG in specific situations. latency or slowed conduction velocity is suggestive of
Fig. 4.5. Suggested neurophysiology setup for the assessment of median nerve function. (A) Median sensory study stimulating
digit 2 and recording over the wrist (orthodromic). (B) Median motor study stimulating at the wrist and recording over abductor
pollicis brevis muscle. (C) Median palmar study stimulating in the palm and recording over the wrist (orthodromic). (D) Ulnar
palmar study stimulating in the palm and recording over the wrist (orthodromic). Panels (C) and (D) make up the palmar com-
parison study. (E) Stimulating digit 1 and recording from the wrist at the median nerve. (F) Stimulating digit 1 and recording from
the wrist at the radial nerve. Panels (E) and (F) make up the median-radial latency difference study. (G) Stimulating the median
nerve and recording over the lumbrical muscle. (H) Stimulating the ulnar nerve and recording over the interosseous muscle. Note
that the recording electrodes have not moved between Panels (G) and (H), these panels make up the lumbrical/interossei latency
comparison study.
CARPAL TUNNEL SYNDROME 71
median nerve compression at the wrist, as early injury innervation, thus, comparing these two latencies can
causes demyelination (see Pathophysiology section). demonstrate median nerve dysfunction across the
Importantly, a prolonged latency may also be the result wrist. Additionally, the sensory fibers supplying the ring
of a focal lesion to a sensory branch distal to the carpal finger are grouped on the medial margin of the median
tunnel (Jablecki et al., 2002), which can be identified nerve anteriorly, directly underneath the transverse
by comparing the wrist to palm latency. That is, with carpal ligament (Uncini et al., 1989). This topography
recording electrodes on the digit and stimulating the can make these fibers particularly susceptible to early
nerve both proximal (wrist) and distal (palm) to the CTS. A difference in latency of >0.4 ms is significant.
carpal tunnel, a difference in latency will localize The median-radial latency difference to the thumb
the lesion to the tunnel. This can also assist in differen- compares median and radial nerve stimulation while
tiating CTS from peripheral neuropathy, as in CTS the recording over the thumb at 10 cm distance. The position
damage to the median nerve is quite focal, with relative of the thumb can alter these measurements, so it must be
preservation of segments outside the carpal tunnel, kept extended though relaxed, to eliminate muscle arti-
whereas in peripheral neuropathy the damage begins dis- fact (Fig. 4.5E and F). A latency difference of >0.5 ms
tally (Werner and Andary, 2011). This technique also is significant (Pease et al., 1989). This test is useful as
allows for the identification of conduction block, which the radial nerve is generally preserved in neuropathy
can be seen in acute carpal tunnel syndrome. An increase and can be an alternative comparison site when there is
in SNAP amplitude of 50% with palm stimulation as concomitant ulnar nerve injury.
compared with wrist stimulation is thought to indicate Notably, no one technique of those outlined above
its presence. clearly demonstrates superiority over the others. The
Additionally, a number of studies have investigated combined sensory index (CSI) was designed to increase
the placement of the recording electrode over other the sensitivity and specificity of these sensory compari-
median nerve-innervated digits, with contrasting results. sons by summarizing the latency differences. Specifi-
Specifically, one study demonstrated that the thumb was cally, the three comparison tests listed above are
more sensitive than the fingers in mild CTS; another con- performed and the latencies of each are added. A CSI
cluded that the ring finger was most sensitive; yet another of >0.9 ms gives a sensitivity of 83% and specificity
demonstrated that the middle finger was most sensitive of 95%, whereas a CSI >1.1 ms improves specificity
(Kothari et al., 1995; Terzis et al., 1998; Tada et al., to 100% without significantly impacting sensitivity
2022). The variability of these results may relate to dif- (82%) (Lew et al., 2000). It should be noted that this
ferences in technique between centers and reflect the index will rapidly exceed normal values if any one of
many reported variations in electrodiagnostic assessment the sensory comparisons on its own has a significantly
of mild carpal tunnel syndrome. increased latency difference, so its utility may primarily
be in mild median neuropathies.
SENSORY COMPARISON STUDIES
Comparison studies can be used to demonstrate slowing
MOTOR STUDIES
of transmission along the median nerve when compared
to the ulnar or radial nerves. This provides internal con- Motor nerve conduction studies provide information
trol for confounding factors such as age, gender, weight, about the compound muscle action potential (CMAP),
height, temperature, hand size, and the presence of a summated response from individual muscle fiber action
coexisting polyneuropathy. potentials. The median motor study is performed by
Orthodromic palmar stimulation is used to determine stimulating the median nerve at the wrist with the record-
the median-ulnar palmar latency difference. The median ing electrodes over the abductor pollicis brevis muscle,
and ulnar nerves are stimulated in the palm 8 cm distal keeping the distance between the recording electrode
to the wrist, with the recording electrode placed over and stimulator at 4–6 cm (Fig. 4.5B). A delay in the distal
the wrist (Fig. 4.5C and D). This generates a mixed motor latency supports a diagnosis of CTS, though other
nerve action potential, though the value recorded at the length-dependent neuropathies can also cause a loss
wrist primarily represents sensory nerve fiber conduction of large myelinated fibers resulting in an increase in
(Mills, 1985). A difference in latency of >0.4 ms is the distal motor latency. Stimulation is also performed
abnormal and suggests median nerve dysfunction at at the antecubital fossa; proximal slowing in the wrist
the wrist. to elbow segment can also occur in CTS due to retrograde
Antidromic stimulation of the median and ulnar axonal degeneration (Chang et al., 2002). Detecting
nerves at the wrist, recorded 14 cm distally on the ring forearm conduction velocity slowing should therefore
finger, is termed the ring finger latency difference. not exclude the diagnosis. In severe CTS, axonal loss
The ring finger receives both median and ulnar sensory results in reduction in CMAP amplitude.
72 N. HARINESAN ET AL.
MOTOR COMPARISON STUDIES characterize the nature of the neuropathic damage and
may help determine whether reduced CMAP amplitude
The lumbrical-interossei latency comparison aids in the
is the result of focal demyelination and conduction
diagnosis of severe CTS. The recording electrode is
block or axonal loss. EMG is essential in the work-up
placed 1 cm proximal and lateral to the midpoint of the
of other related conditions that can present like CTS,
3rd metacarpal, with the reference electrode placed
such as radiculopathies, thoracic outlet syndrome, plexo-
over the metacarpophalangeal joint of the index finger.
pathies, and motor neuron disease; thus, its utility might
The median and ulnar nerves are then stimulated at a
primarily be in the exclusion of other conditions.
point 10 cm from the recording electrode (Fig. 4.5G
and H). The correct placement of the recording electrode
allows for recording of both the median-innervated GRADING SEVERITY
lumbricals and the ulnar-innervated interossei with
A grading system based on neurophysiological parame-
median and ulnar stimulation, respectively. A difference
ters has been developed. Mild CTS is defined by pro-
of >0.4 ms between the median and ulnar latencies is
longed (relative or absolute) sensory latencies with
significant (Preston and Logigian, 1992). The utility of
normal median motor nerve conduction studies, with
this assessment in severe CTS, in which other motor
no evidence for axon loss (Werner and Andary, 2011).
responses may be severely reduced or absent, is due to
Moderate CTS is defined by abnormal median sensory
relative sparing of the median nerve axons supplying
latencies as noted for mild CTS, and relative or absolute
the second lumbrical (Boonyapisit et al., 2002).
prolongation of median motor distal latency. Again, there
The median to ulnar thenar latency difference is
is no evidence of axon loss, preserved motor and sensory
another comparison study to demonstrate median nerve
amplitudes (Werner and Andary, 2011). Severe CTS is
dysfunction across the wrist. The latency of the standard
defined by any of the aforementioned abnormalities with
median motor study, when stimulating the wrist, is com-
evidence of axon loss as defined by one of either low or
pared with the latency following ulnar nerve stimulation
absent SNAP amplitude, mixed nerve action potential
to ulnar-innervated muscles, such as adductor pollicis.
amplitude, or thenar CMAP amplitudes. Low amplitudes
The positions of the recording and reference electrodes
are generally defined as a decrease in amplitude of more
need not change as volume conduction will allow the
than 20% of the standard laboratory reference range.
CMAP of the ulnar-innervated thenar muscles to be
Needle electromyography demonstrating fibrillation
detected. Importantly, the ulnar CMAP will have an
potentials, sharp waves, or motor unit potential changes
initial positive deflection, and it is the start of this that
consistent with denervation-reinnervation including any
should be taken as the beginning of the CMAP. A latency
of large amplitude, long duration, or excessive polypha-
difference of >0.8 ms is significant (Sander et al., 1999);
sic motor unit action potentials, also reflect severe dis-
however, normal differences of up to 1.4 ms have been
ease (Werner and Andary, 2011). Despite the wide use
reported in patients without CTS.
of this grading system, its clinical utility has only been
shown in its ability to predict postsurgical outcomes,
MOTOR SHORT SEGMENT STUDY with severe patients having worse outcomes after surgery
Palmar motor stimulation can be used to obtain informa- (Bland, 2001). In addition, the correlation between the
tion about the presence or absence of conduction block. severity of EDX findings and symptoms is not well
The median nerve is stimulated in the palm and wrist and established; that is, patients with severe symptoms might
the APB CMAP is recorded. A reduction in the CMAP have only mild changes on neurophysiology, and those
amplitude when stimulating the wrist as compared with with mild symptoms might have more marked changes.
the palm is suggestive of conduction block. Caution must In the case of the former, a patient might not be referred
be taken, however, in the interpretation of the CMAP for surgical management that could prove helpful, and
amplitude with palmar stimulation. This can be impacted the latter might not see much benefit from surgical inter-
by volume conduction to the ulnar nerve, or by direct vention. Given these limitations, caution should be taken
muscle stimulation, resulting in a larger amplitude and when using EDX findings to make treatment decisions
area of the CMAP. Care should also be taken to avoid (Robinson and Kliot, 2008). Importantly, this grading
a positive deflection at the start of the CMAP, indicative system also establishes a pattern of evolution in EDX
of ulnar nerve stimulation. findings; a deviation from this pattern should alert the
clinician to consider an alternative or superimposed
condition. Based on the above, we have summarized
NEEDLE ELECTROMYOGRAPHY (EMG)
the commonly used techniques (Table 4.1) and severity
Current AANEM guidelines outline EMG as an optional grading scales (Table 4.2) in our institutes, which we
procedure when evaluating CTS. It is useful to recommend for assessment for carpal tunnel syndrome.
CARPAL TUNNEL SYNDROME 73
Table 4.1
Recommended nerve conduction studies for assessment of carpal tunnel syndrome.
Table 4.2
Grading severity of EDX findings.
Comparison Studies
Palmar/Digit 4/Radial Normal Delay > 0.4 ms Delay > 0.4 ms Delay > 0.4 ms Delay > 0.4 ms
Sensory Studies
Latency Normal Normal " " "/absent
Amplitude Normal Normal # # #/absent
Velocity Normal Normal # # #/absent
Motor Studies
Latency Normal Normal Normal " "
Amplitude Normal Normal Normal Normal #
Velocity Normal Normal Normal Normal #/normal
Fig. 4.6. Ultrasound of the wrist depicting the median nerve. (A) Cross-sectional assessment of the median nerve in a healthy
individual (arrow). (B) Cross-sectional assessment of the median nerve in carpal tunnel syndrome (arrow), note that the nerve
is enlarged. (C) Longitudinal assessment of the median nerve in a healthy individual (arrow). (D) Longitudinal assessment of
the median nerve in carpal tunnel syndrome, note that the nerve has an hourglass appearance with a point of compression (arrow).
(E) Cross-sectional assessment with color Doppler in a healthy individual demonstrates no surrounding vascularity.
(F) Cross-sectional assessment with color Doppler in carpal tunnel syndrome demonstrates increased vascularity (red).
CARPAL TUNNEL SYNDROME 75
CSA in the forearm (at the level of the pronator in healthy controls (Fig. 4.6E and F) (Ghasemi-Esfe
quadratus) is also performed. A wrist-to-forearm ratio et al., 2011). Additionally, Doppler US can assist in iden-
of 1.4 or more was shown in one trial to have a 100% tifying vascular anomalies that may contribute to CTS,
sensitivity for detecting CTS (Hobson-Webb et al., such as a persistent median artery.
2008), but these values are influenced by patient age,
height, and severity of median neuropathy. Several stud- US ELASTOGRAPHY
ies have also assessed the size of the median nerve at
the carpal tunnel outlet, i.e., at the level of the hamate This technique measures soft tissue stiffness by measur-
(Nakamichi and Tachibana, 2002; Wiesler et al., 2006; ing strain or shear wave elastography. Shear wave elas-
Csillik et al., 2016). An inlet:outlet ratio is calculated, tography measures the speed of propagation of a shear
with a cut-off value of 1.14 demonstrating a sensitivity wave in the target tissue, with higher speeds suggestive
of 77.9% and a specificity of 55.5% (Zhang et al., of stiffer tissue (Ophir et al., 1991). In the setting of nerve
2015). With lower predictive values, this ratio may not injury, the pathological change of damage to myelin and
be as useful as the more commonly performed wrist- subsequent replacement with less compliant connective
to-forearm ratio. tissue can be represented through increased stiffness
(Kantarci et al., 2014). A recent systematic review and
DYNAMIC CROSS-SECTIONAL US meta-analysis of the use of shear wave elastography
in CTS demonstrated increased stiffness of the median
Dynamic cross-sectional analysis was developed to nerve at the wrist (Lin et al., 2019; Wee and Simon,
clarify the morphological changes and movement of 2021). In addition, shear wave elastography may be able
the median nerve during flexor tendon sliding (Van to differentiate CTS of different severities, something
Doesburg et al., 2010, 2012). The subsynovial connec- which is not achievable by measuring cross-sectional
tive tissue provides a lattice within the carpal tunnel, area alone (Wee and Simon, 2020).
and in patients with CTS, it has been reported to
proliferate, with fibrosis and edema reducing the visco-
elasticity compared to healthy controls. During this MAGNETIC RESONANCE IMAGING (MRI)
technique, the subject is asked to extend then flex the
MRI can provide high-spatial resolution and soft tissue
fingertips into the palm. Target images are acquired at
contrast of the carpal tunnel and its contents, including
end flexion and extension, and subsequently evaluated
the median nerve along its entire course. Exact protocols
for median nerve morphology and displacement, as
may differ between institutions, though a minimum
well as structural integrity (Van Doesburg et al., 2010).
assessment should include axial sequences that are
Regarding morphological changes, the circularity and
T1-weighted and T2-weighted with fat suppression.
aspect ratio of the nerve decreases and increases, respec-
With these standard images, the median nerve is differ-
tively, in CTS on finger flexion, whereas healthy controls
entiated from the surrounding epineural fat, the latter
demonstrate an increase in circularity and decrease in
being hyperintense on T1 imaging and hypointense on
aspect ratio (the inverse of CTS) (Van Doesburg et al.,
fat-suppressed imaging (Fig. 4.7A). High-resolution
2010). Regarding displacement, the median nerve is
imaging and dedicated neurography sequences can
known to be mobile with finger flexion and extension,
depict the individual fascicles of the nerve (Felisaz
sliding toward the ulna upon flexion and the radius upon
et al., 2016). Loss of fascicular architecture as well as
extension in healthy controls (Kuo et al., 2016). This
high T2 signal of the nerve is abnormal, suggestive of
transverse sliding is reduced in patients with CTS; in
edema or inflammation (Fig. 4.7B). MRI parameters
particular, the degree of mobility has been shown to
used to assess CTS include the median nerve CSA,
correlate with the severity of the median neuropathy,
flattening ratio (FR), and signal intensity (SI) as well
so that mobility decreases with reducing median nerve
as palmar retinacular bowing (BR).
conduction velocity (Kuo et al., 2016).
The median nerve CSA proximal to the carpal tunnel
is the most frequently studied parameter and, like in
DOPPLER US
ultrasound assessments, has been shown to be signifi-
Doppler ultrasonography can be used to identify the flow cantly larger in patients with CTS compared with healthy
and the direction of blood within structures. The periph- controls (Kleindienst et al., 1998; Jarvik et al., 2000).
eral nerves are supplied by a dense anastomotic network A cut-off value of >15 mm2 in this location yields a sen-
of epineurial and endoneurial blood vessels and com- sitivity of 85.5% and a specificity of 100% (Ng et al.,
pression of the nerve can disrupt this blood supply. 2020). There has been limited research on measuring
Doppler US can detect intraneural blood flow within median nerve CSA in other locations by MRI. However,
the median nerve in the carpal tunnel, which is not visible one study took the largest CSA either proximal or distal
76 N. HARINESAN ET AL.
Fig. 4.7. MRI of the wrist depicting the median nerve and surrounding structures (axial fat-suppressed T2-weighted images).
(A) Healthy individual showing a normal appearance of the median nerve (arrow). (B) Patient with carpal tunnel syndrome,
the median nerve is enlarged and hyperintense (arrow).
to the tunnel and reported a sensitivity of 100% and which diffusivity is equal in all directions. Further,
a specificity of 94% for the diagnosis of CTS DTI-derived measures including axial (along the axons)
(Ng et al., 2020). and radial (across the axons) diffusivity provide more
The flattening ratio is calculated by measuring the specific information about axonal and myelin pathology.
major and minor axes of the median nerve at the level Mean diffusivity (also reported as apparent diffusion
of the distal radioulnar joint (DRUJ) and within the tun- coefficient, ADC) measures the extent of water diffusion
nel, usually at the level of the hamate. In one study, the and is a surrogate for the compactness of a structure
ratio was 3.4 0.4 in healthy controls and 5.1 0.5 in (Simon and Kliot, 2014). A recent meta-analysis demon-
CTS patients (Allmann et al., 1997). Additionally, the strated that when compared with healthy control sub-
ratio was shown to improve postsurgery, normalizing jects, patients with CTS have significantly lower FA
in up to 94% of patients (Allmann et al., 1997). and higher ADC, with a sensitivity and a specificity of
83% and 78%, respectively (Wang et al., 2016). There
SIGNAL INTENSITY is also a correlation between the severity of CTS and
the degree of reduction in FA and increase in ADC, with
The signal intensity of the median nerve is increased more severe cases demonstrating greater change (Wang
in 83.3%–100% of patients with CTS (Mesgarzadeh et al., 2016; Wafaie et al., 2018).
et al., 1989; Britz et al., 1995; Deryani et al., 2003),
and appears to correlate to median neuropathy severity.
The median nerve may appear to have a hyperintense MANAGEMENT
rim of high signal around an isointense center in early Management of CTS can be nonsurgical or surgical; the
CTS, followed by homogenously increased signal as choice is informed by the severity of the syndrome both
compression progresses. In severe chronic compression, on clinical assessment and EDX testing, as well as patient
a decrease in signal intensity may be seen, possibly as a preference. Surgical management tends to have long-
result of endoneurial fibrosis (Kleindienst et al., 1998). lasting effects; however, it is an invasive procedure with
The area of high signal can also expand proximally associated risks. Adequate patient education should be
toward the DRUJ and distally to the metacarpal bases a part of first-line management regardless of additional
(Mesgarzadeh et al., 1989). therapy offered and is now recommended in a recent
guideline on carpal tunnel management (Huisstede
DIFFUSION TENSOR IMAGING AND MR TRACTOGRAPHY et al., 2014).
Diffusion tensor imaging (DTI) measures the diffusion of
water molecules in tissues (Basser et al., 1994). In neural Non-surgical
structures, water diffuses freely along the long axis of
ERGONOMIC TOOLS
the nerve tract but is restricted in the perpendicular
plane by an intact myelin sheath. In this case, diffusion Ergonomic tools are used to modify the wrist position
is anisotropic, i.e., limited in one direction. Fractional during use so that it remains straight (neither flexed,
anisotropy estimates diffusivity along a scale from comp- extended, nor abducted/adducted) as well as minimizing
letely anisotropic (1) to completely isotropic (0), in repetitive movements. The straight position is expected
CARPAL TUNNEL SYNDROME 77
to place the least pressure on the median nerve as the et al., 1999; Armstrong et al., 2004). In addition, cortico-
carpal tunnel is the most spacious in this orientation. steroid injection with the use of ultrasound guidance
Specific examples of treatments include modified key- has been found to be more effective than blind adminis-
boards, insulating gloves to avoid exposure to vibration, tration in a meta-analysis of the technique (Chen et al.,
or forearm supports. A Cochrane review on this 2015b).
subject determined there was insufficient evidence to Oral corticosteroids and nonsteroidal anti-
determine whether ergonomic techniques were benefi- inflammatories have been considered in the management
cial or harmful in the treatment of CTS (O’Connor of CTS; however, the evidence is conflicting. A system-
et al., 2012); however, one of the two included studies atic review of oral corticosteroids prescribed for
demonstrated a statistically significant reduction in pain 2–4 weeks found improved patient reported outcomes
at 12 weeks when compared to no intervention (Rempel and symptoms in the short-term compared with placebo,
et al., 1999). but found no evidence supporting their use in the longer
term (up to 12 months) (Huisstede et al., 2018). This is
SPLINTS similar to comparisons between oral corticosteroids
and splinting, which found superior outcomes for oral
A splint is a device worn on the affected hand aimed at treatment after 2 weeks, but no difference after 3 weeks
immobilizing the wrist joint, while leaving the fingers (Huisstede et al., 2018). It is reasonable to consider a
and thumb to move freely, theoretically reducing the short course of oral corticosteroids in very symptomatic
fluctuations in carpal tunnel pressure. It can be worn at patients while considering long-term management
night or during activities causing wrist motion during options, such as surgical intervention. However, targeted
the day. The conclusion of a Cochrane review on wrist corticosteroid injection is preferred due to the lower rate
splinting based on limited evidence was that nocturnal of systemic side effects.
splinting was more effective than no treatment at all,
and did not find a difference between types of splints
Surgical intervention
(Page et al., 2012a).
Surgical treatment is considered the most effective
NERVE AND TENDON GLIDING method of altering the dynamics of the carpal tunnel to
relieve the pressure within it. The transverse carpal liga-
Gliding of the median nerve or flexor tendons is a ment is transected to release pressure on the contents of
mobilization technique aimed at decreasing carpal tunnel the tunnel. Surgery is generally performed as a day-only
pressure by elongating the structures that surround the procedure and can be performed using three techniques:
median nerve (the nerve bed) and reducing adhesions a traditional open technique, a minimally invasive
(Totten and Hunter, 1991). The median nerve is mobi- technique, or an endoscopic technique.
lized by putting the hand and wrist into prescribed posi-
tions, including maintaining a neutral wrist position
INDICATIONS
while flexing and extending the fingers; extending the
wrist, fingers, and thumb; and supinating the forearm The electrophysiological grading of CTS severity, and
and stretching the thumb (Totten and Hunter, 1991). response to previous therapy, are guides to surgical
Tendon gliding exercises involve moving through dis- intervention. Surgery is generally indicated, if there is
crete positions from an open hand to a fist. A Cochrane moderate–severe carpal tunnel syndrome with evidence
review concluded that there was limited and very of axonal loss (as exemplified by reduced CMAP ampli-
low-quality evidence of benefit when accounting for a tude) or denervation on EMG, or if there is a mild
diverse collection of mobilization techniques in the syndrome not responding to conservative therapy. The
management of CTS (Page et al., 2012b). rate of surgical intervention has been shown to vary quite
markedly across health services, and standardized
CORTICOSTEROIDS referral criteria are lacking (Maggard et al., 2010).