Median_ Nerve Displacement Throughthe Carpal Canal

Robert M. Szabo, MD, Brian K.. Bay, PhD, Nell A. Sharkey, PhD,Chris Gaut, MD, Davis, CA

We determined the direct relationships between wrist position and displacement of the mediannerve during active contraction of the flexor tendons at the wrist with an intact, transectedtransverse carpal ligament (TCL). Nine fresh cadavers were mounted in an apparatus to allow

variable wrist position. Excursions of the tendons and displacement of the median nerve weremeasured by tracking markers with a video camera. Each limb was tested at 0°, 30°, and 60°

of wrist extension before and after release of the TCL. Excursion of the flexor tendons required

for full finger flexion ranged from 2.3 to 3.I cm (mean, 3 cm). Median nerve displacementranged from 0.9 to I .4 cm (mean, I cm). The relationship between median nerve and flexortendon excursion was consistently linear. Finger motion alone allows for median nerve dis-

placement after surgery in the carpal tunnel. (J Hand Surg 1994~ 19A:901-906,

path of the median nerve lies anteriorcenterg\of rotation of the joints of the hand,of the ,fist and fingers will shorten the bed

nerve, it to side proximally. Simi-of the wrist and fingers lengthens

the median nerve in the hand, causing~slide distally. Distal-proximal sliding of the me-

"re at the wrist in response to hand motion.)bserved in live subjects1 and in cadavers.:"xtent that the normal sliding of the mediart

restricted, for example, as a result of adhe--or compressive entrapments, lengthening the

in response to joint motion must be ac-by local elongation of the nerve. Nerw~

or traction of this type can result in tern-

or permanent disruption of action potentialation, causing impairment of sensory and

hypothesize that altered kine-

Department of Orthopaedics, University of Califor-School of Medicine, Davis, CA.for publication Dec. 15, 1993; accepted in revise, d

II, 1994.any form have been received or will be receiw,~d

party related directly or indirectly to the sub-this article.

uests: Robert M. Szabo, MD, Dep~.rtment of Ortho-University of California. Davis, 2230 Stockton Biwl.,

CA 95817.

matics may be an etiologic element in nerve entrap-

ment syndromes, such as carpal tunnel syndrome.The purpose of this study is to describe the normal

motion of the median nerve at the wrist in relationto stimulated active finger flexion, with the wristimmobilized in a neutral position or in extension.

Methods

Nine fresh cadaver forearms disarticulated at theelbow were used. Each was rigidly mounted-in a jigby means of three Steinmann pins, one each in theradius and ulna proximally and one distal, traversingboth radius and ulna (Fig. 1). Two 3-ram externalfixation half-pins in the index metacarpal were at-tached to the jig by a pivoting clamp that allowedimmobilization of the wrist in any desired position offlexion or extension. Each finger was independentlyloaded with a 75-gram counterweight attached to thefinger tip with a heavy suture to extend the finger.

The forearm was superficially dissected to exposethe median nerve, transverse carpal ligament, andflexor digitorum superficialis (FDS) tendons. Theflexor digitorum profundus (FDP) tendons were ex-posed by deeper dissection in the mid-forearm. Carewas taken not to disrupt the natural tendon-nervepositional relationship, and all fascia was left intact..Stainless steel sutures were attached to the FDS andFDP tendons at the muscle-tendon junctiom

The Journal of Hand Surgery

.902 S~zabo et al. / Median Nerve Displacement Through the Carpal Canal

VideoCamera

FixationPins

Wrist SupportLoadCell

Stepper Motor

2. Exampledisplacemenand throughing markers z

ne e/tendon :

FingerCounterweights

Figure 1. Experimentalmodel and data collection system.

proximally along the muscle bellies, and attachedto a load cell in series with a computer-controlledstepper motor. The stepper motor simulated slowmuscle contraction by producing 12.5 mm/min offlexor displacement and was positioned to maintainthe normal line of action of the flexor muscles. Eachkinematic study was initiated with the fingers fullyextended, and the stepper motor was used to flexthe fingers until they reached the palm.

Excursion of the FDS tendons and displacementof the median nerve relative to the transverse carpalligament was measured by tracking the positions ofmarkers attached to each structure (Fig. 2). Steelsphere marke~s 0.5 mm in diameter were placed inthe substance of the.tendon of the flexor digitorumsuperficialis of the ring finger and within the sub-stance of the median nerve. With this placement,markers moved with the nerve or tendon, and notwith the overlying fascia, Reference markers wereplaced in the transverse carpal ligament. As the fin-gers were actively flexed, positions of the markerswere recorded with a Pulniz CDC video camera(Pulnix America, Sunnyvale, CA) placed directlyabove the wrist. Images were recorded with the fin-gers extended and at each 5-ram increment of flexortendon motion proximally, as determined by dis-placement at the stepper motor. Marker positionswere measured in each captured frame usingIMAGE software (Wayne Rasband, Research Ser-vices Branch, National Institute of Mental Health),

and displacements of the flexor tendons andnerve at the wrist were calculated.

Each limb was tested in three positions:30° wrist ex~:ension, 60° wrist extension. Th.positions selected reflect a sequential incretension that parallel increasing carpal tunnel)sures noted ’with increasing wrist extensionple with and without carpal tunnel sylowing these tests, the transverse carpal liwas surgicalb) sectioned to simulate a release~to treat carpal tunnel syndrome, and the kineexperiments were repeated.

For each specimen, the FDS tendonwas plotted against displacement of the cortesing median nerve, and this relationshiplyzed by linear regression. A three-fa~of variance was then conducted with theline slopes to ,determine the significance ofsvariation, wrist position, and surgicaltransverse carpal ligament. This wasa two-factorial blocked design analysis(ANOVA) with specimen as blockingwrist position and surgical release as the extal factors.

ResultsExcursion of the FDS tendons required

finger flexion ranged from 2.3 to 3.1 cmcm). Corresponding displacement of the

ranged fro~betv,

median nervregardlessverse carpa

of r2 for netvaried fr

,f the con.’and tendc

;sion was u:dis

,we3. Ptot e#" :" :tigame: :

The Journal of Hand Surgery / Vol. 19A No. 6 November 1994 903

2. Example of video frames used to measure median nerve (MN) and flexor digitorum supevficialis (FDS-ringdisplacement. Measurements are references to a line drawn parallel to the edge of the transverse carpal ligament:,~,~,d through a marker placed within the substance of the ligament. At the beginning of loading (A), the MN and

-ri:= markers are approximately equidistant from the TCL. At the end of the test (B), both have moved proximallynerve/tendon ratio of approximately 0.4 . Palmaris longus (PL) is labelled for reference.

ranged from 0.9 to 1.4 cm (mean, 1 cm). The~ between excursion of the FDS tendons

median nerve displacement was consistently lin-regardless of wrist position or whether the

asverse carpal ligament was intact (Fig. 3). Val-or r2 for nerve displacement versus tendon ex-

varied from 0.893 to 0.998 (mean, 0.96). Be-of the consistently linear relationship betweenand tendon displacement, slope of the linear.sion was used to characterize each test. These

.placement ratios were used for sta-

tistical comparison of the experimental groups (Fig.4).

The ANOVA revealed significant differences be-tween specimens (p = .002 I), but differences owingto surgical release of the transverse carpal ligament(p = .142) and wrist position (p = .349) were significant. The effect of ligament release and wristposition remained insignificant and had similar p-values when eliminating the effect of specimen vari-ability With the: two-way blocked ANOVA.

Load versus motor displacement data were ap-

1.5

slope = 0.43

R2 = 0.992

1.0

0.0 0.5 1.0 1.5 ,,.0 2.5 3.0Tendon Displacement (cm)

Plot of nerve displacement versus tendon displacement. This sample i:s in the neutral position with the transverseligament intact. Range of nerve and tendon motion and the linear relationship are typical.

904 S.zabo et al. / Median Nerve Displacement Through the Carpal Canal

0olm

Neutral 30° Extensiono

60 Extension

~ Pre Release [~ Post ReleaseFigure 4. The nerve/tendon displacement ratio (slope of the nerve versus tendon displacement) for the three(wrist in neutral, 30° extension, and 60° extension) and the two conditions (transverse carpal ligament intact and

proximately linear; r 2 = 0.95 by regression analysison compiled data from all experiments. Loads neverexceeded 45 N. No slipping at the tendon-suture wasnoted.

Passive flexion of the fingers at each wrist posi-tion resulted in bunching up of the nerve and tendondistal to the transverse carpal ligament, with littleevidence of sliding.

DiscussionCarpal tunnel iyndrome (CTS) has classically

been considered a nerve compression injury; how-ever, it is clearly multifactorial in origin. 5 CTS isknown to be aggravated by activities requiring pro-longed wrist flexion or repeated wrist or finger flex-ion and power grip. 6 Using wick catheter pressuremeasurements, we determined that hydrostaticpressure increases in the carpal canal when the wristis held in either flexion or extension.7 Furthermore,repetitive wrist flexion and extension cause a sus-tained increase in intracarpal pressure.8 Recentmagnetic resonance imaging studies have demon-strated that in flexion the finger flexor tendons areshifted anteriorly against the transverse carpal liga-ment, while in extension they are aligned more dot-.

sally. 9 In its interposed position between theriot row of tendons and the transverse carpalment, the median nerve is subject to shearincompression as the fingers are flexed. Thewhich these shear forces result in tensilethe mediarr’nerve has not been investigated.

A related problem concerns "spot-weldinthe me, dian nerve caused by postoperativesions. Postoperative splinting regimens andtherapy programs have been recommended tovent this problem without any knowledge ofnerve mechanics. For instance, a lonebate concerns whether splinting the wristpal tunnel surgery to prevent bowstringingflexor tendons is harmful because of limitingnerve displacement by adhesions.

Wilgis and Murphy2 observed longitudinalof the median nerve at the wrist in response tosive motion of the hand in cadaver limbs.we are concerned that passive motion of aarm may not accurately reflect nervean active person, l° we simulated active fingerion in cadaver preparations by pulling onflexor tendons with a motor. Wedian nerve displacement with power grip inthe FDP and FDS act in unison. The s

architect’~and flexorto justify t~

single motorof the medi.

such as typ}

range of.iaat we cfinger ticand Mu

ion/extensiwas note

who edeflections

In additi

:nt of the me~.9.~2 and rm

-~otion b,and ma’:_

One of the fcin re,~

ernal elastiis clean

cm, indic~being ~

:asured onnerve atd length

si~z~ is strai.tension. At.

motion ofuntil eq

In additionthe ne :

motion,con~

the rateacent tern

shear tWe k~

forces:prodm

ests that ’membran

fie,"of the:

Goldstein e~tendonsligamer

loadsexist b.

transversefinger fic

The Journal of Hand Surgery / Vol. 19A No. 6 November 1994 905

architecture of the flexor digitorum superfi-flexor digitorum profundus muscles ap-

ustify tendodesis of their tendons and usele motor for finger flexion. ~ The displace-the median nerve may be different in activi-

as typing that require in-dependent FDS/.range of longitudinal sliding of the median

tln::: we observed (9-14 mm) in response finger flexion is similar to that observed by

is and Murphy2 in response to passive wristm/extension in cadavers. A similar range of

was noted in live subjects by McLelland and~ who estimated nerve sliding by observing

:flections of microelectrodes inserted in theIn addition to longitudinal sliding, displace-

:of the median nerve in the palmar-dorsal direc-~.~-" and radial-ulnar direction~3 in response tom, cion have been observed in dynamic ultra-and magnetic resonance imaging studies.of the forces causing the nerve to slide longi-

in response to hand motion is the nerve’s:rnal elastic tension. Commonly if the median

is cleanly transected, the ends will retract 1indicating that the nerve was under tension

being cut. Kwan et al.,~4 in a rabbit study,on 11% shortening in segments of periph-

nerve after explanation. If we take the ex-iength as the gauge length, then the nerve

~s strained 1 I%, and therefore is continuallyAny imbalance in tensile forces causedof joints would result in sliding of the

until equilibrium is reestablished.addition to internal tension, shear forces be-

the nerve and adjacent tissues may affectmotion. Our finding that the flexor digitorum

consistently translocate at two to threethe rate of the median nerve indicates that the

tendons exert proximally directed, fric-shear forces on the nerve as the fingers are

verse carpal ligament, we may expect that similarshear forces will affect the median nerve; thesewouM be aligned opposite to the shear forces result-ing from friction with the proximally moving ten-dons. We speculate that increased pressure in thecarpal tunnel might amplify the relative role of shearforces in median nerve motion. Such a pressure in-crease could result from abnormal anatomy5 or sim-ply from forceful contraction of the flexor digitorumprofundus muscle.~° Similarly, any adhesions be-tween nerve and tendon or between nerve and trans-verse carpal ligament or entrapment would be ex-pected to alter the normal nerve kinematics.

The failure of surgical release of the transversecarpal ligament to affect nerve and tendon excursionin this study is not surprising. Disruption of normal:hydrostatic relations in a cadaver hand would beexpected to alter carpal tunnel pressure dynamics;in addition, there was no evidence that any of thesecadavers suffered from carpal tunnel syndromewhen alive. Increasing wrist extension caused atrend toward diminished sliding of the mediannerve. Wrist flexion would be more likely to elicitkinematic abnormalities;~5 however, technical prob-lems with the wrist-flexed preparation prevented usfrom collecting data in this position.

The observation that both flexor tendons and me-dian nerve exhibit substantial translocation at thewrist in response to finger motion when the wrist isrigidly immobitzed is significant with respect to carefollowing surgical repair of nerve or tendon injuries.It appears that active finger motion alone would pro-wide :sufficient motion of the median nerve andflexor tendons in the vicinity of the wrist to preventadhesf~n formation even if the wrist is immobilized.Conw~rsely, if one desires complete immobilizationof the: median nerve or finger flexor tendons afterre.pair, digital motion must be prevented. --

ReferencesWe know of no published measurements ofi~’fix’~

~se forces; our observation that passive finger ~ cLelland DL, Swash M. Longitudinal sliding of theproduced little motion of the median nerve ~’t----~"median nerve during movements of the upper limb. J

that they may be important. In addition,membranous connections between the median

flexor tendons may contribute to the syn-of their movement.

et al. ~ measured strain in the digitaltendons proximal and distal to the transverseligament in response to stepwise increasing

loads and concluded that significant shearexist between the finger flexor tendons and

se carpal ligament. Wrist flexion and ac-finger flexion increased these traction effects.

median nerve passes adjacent to the ~rans-

d:’~eurol Neurosurg Psych 1977;39:566-70...~gilgis EFS, Murphy The significance of longitudi-R." hal excursion in peripheral nerves. Hand Clin 1986;

2:761-6.

9/~/Brown R, Pedowitz R, Rydevik B, et al. Effects ofacute graded strain on efferent conduction propertiesin the rabbit tibial nerve. Clin Orthop 1993;296:

,-~,~28:~-94.Q~lSunderland S. Stretch-compression neuropathy. Clin~.~

Exp Neurol 1981;18:1-13.v/~i Szabo RM, Madison M. Carpal tunnel syndrome. Or-

thop Clin North Am 1992;23:103-9.6. Masear VR, Hayes JM, Hyde AG. An industrial cause

906 Szabo et al. / Median Nerve Displacement Through the Carpal Canal

of carpal tunnel syndrome. J Hand Surg 1986;11A:222-7.

7. Gelberman RH, Szabo RM, Mortensen WW. Carpaltunnel pressures and wrist position in patients withColles’ fracture. J Trauma 1984;24:747-9.

8. Szabo RM, Chidgey LK. Stress carpal tunnel pres-sures in carpal tunnel patien’rs’ and normal patients. JHand Surg 1989;14A:624-7.

9. Skie M, Zeiss J, Ebraheim NA, Jackson WT. Carpaltunnel changes and median nerve compression duringwrist flexion and extension seen by magnetic reso-nance imaging. J Hand Surg 1990;15A:934-9.

I0. Smith EM, Sonstegard DA, Anderson WH. Carpaltunnel syndrome: contribution of flexor tendons.Arch Phys Med Rehabil 1977;58:379-85.

11. Brand PW, Beach RB, Thompson DE.sion and potential excursion of musctes inand hand. J Hand Surg 1981;6:209-16. ,

12. Zeiss J, Skie M, Ebraheim N, Jacksonrelations between median nerve and flexor.the carpal tunnel. Am J Roentgenol 1989:

13. Nakamichi K, Tachibana S. Transverse s]median nerve beneath the flexor retinacul

~. uwrg 1992;17B:213-6.an MK, Wall E J, Massie J, Garfin SR~ stress, and stretch of peripheral nerve. Ac

Scand 1992;63:262-72.15. Goldstein SA, Armstrong TJ, Chaffin DB,

LS. Analysis of cumulative strain in tendonszdon sheaths. J Biomech 1987;20:1-6.