- Email: [email protected]
Avoidance of transection of the palmar cutaneous branch of the median nerve in carpal tunnel release
Avoidance of Transection of the Palmar Cutaneous Branch of the Median Nerve in Carpal Tunnel Release Greg P. Watchmaker, MD, Dean Weber, MD, Susan E. Mackinnon, MD, St. Louis, MO The course of the palmar cutaneous branch of the median nerve (PCBMN) was studied in 25 fresh cadaveric upper extremities in order to identify its relation to local structures and commonly used incisions for carpal tunnel release. The PCBMN was found to closely underlie the tbenar crease (average, 0-2 mm radia] to crea.~.; range, 6 mm ulnar to fi mm radial to thenar crease), suggesting that an incision fashioned in the thenar crease would lead to fi'equent PCBMN iniury. The PCBMN was also found to cross the axis of the ring finger' when the axis was determined with the finger flexed inlo the palm. The axis of the ring finger, as drawn with the ring finger extended, projecled in a more ulnar direction. The PCBMN was an average of 9 mm radial [o Ihis projection (range, ~.-16 ram). An analysis of 100 human volunteer hands demonstrated (}~at [he deepest point between the thenar and hypothenar eminencies was a constant landmark in.the proximal palm (interthenar depression). The PCBMN traveled an average of 5 mm radial to the interthenar depression (range, 0-12 mm radial). Thenar crease anatomy and ring finger projection were highly variable both in absolute location and configuralion, providing a poor basis for incision placement. An incision placed approximately 5 mm urnar to the interthenar depression, extending in the direction of the third web space, will decrease the incidence ~3f injury to the PCBMN. (J Hand Surg 1996;21A:644.-650.)
T h e pallmtr cutaneous branch of the median nerve ( P C B M N ) is a constant branch that arises in the distal forearm from the radial aspect of the median nerve and travels distally in close association with the flexor carpi radialis tendon. It pierces the superficial palmar fascia in its own tunnel, sending sensory branches to the palmar skin. Although its origin and course in the dist~d forearm rind hand have been previotlsly described, its relationship to c o m m o n l y used
From the Division of PlasOc Surgery,Department el=Sm,gc13;,Wash inglon Utfiversi!ySchooltff Medicine, St. If~n~s,Me. Recollectf~rcpublic;aion April 7, 1995;accepted in l~vised foim Dec. 5, t 995. No beneiit~ in any iunn have been received or wJlJ be receivedfrom a commercialpartyrclateddirectlyor Lnxlh-ectlyto the subject o{"I:hi~article. Reprint requests: Susan E. Mackinnon, MD, One B;~_nmsHospital Pizza, Suite 17424, St. [~mN,Me 63111). 644
The Journal o{ Hand Surgery
landmarks in the palm has not been thoroughly investigated. Surgical approaches to the carpal tunnel release have been based on soft tissue creases a n d bony landmarks present in the hand. The pahnar creases have often been used to conceal a palmar incision. There are various scientific and astrologic terms used tbr the p a h n a r creases I (Fig. 1). Although the thenar crease is raosl often cited as an incision site, the more ulnar interthenar crease is also mentioned2 The axis of the ring finger has also been proposed as a suitable landmark for incision. 3 T h e term axis is s{~mewhat confusing, however, ,as this ctmld signify the axis with the digit fully flexed or extended, or the course of the metacarpal into the palm. This study describes the detailed course of the P C B M N in relation to the overlying creases and previously proposed landmarks in order to provide a
The Journal of Hand Surgery/ '4ol. 21A No. 4 July 1996
Figure 1. The palmar creases. DWC, distal wrist crease; TC, thenar crease (life line); DTC, distal transverse crease (heart lirtc); PTC, proximal transverse crease (head line); ITC, interthenar crease (destiny tint).
scientific basis for avoiding transection of the PCBMN.
Materials and Methods Dissections Twenty-five flesh human cadaveric upper extremities were ob 'tamed for dissection. The thenar crease was first traced with black ink (Fig. 2A). If the thenar crease did not extend to the distal wrist crease, a perpendicular line was drawn from the wrist crease to the most proximal point on the thenar crease. The flexion axis of the ring fitlger was then determined by flexing the metac~poplu 'dangeal and proximal interpha'langeal joints, then marking the point where the fingertip touched the palm (Fig. 2D). The extension axis of the ring linger was determined by marking the middie of the ring finger at both the proximal and distal digital flexion creases. A straight line was then drawn from these two points proximally into the palm while the finger was held in extension (Fig. 2B). Finally, the wrist was extended and the deepest point between the thenar and hypothenar eminences was marked (Fig. 2C). This point was then extended in line with the third web space. Distances between these lines were measured at the distal wrist cre~e, then at 1, 2, 3, ,and 4 cm distal to the distal wrist crease.
645
Disscction using 2.5• ]oupe magnification was begun in the tbrearm by exposing the median nerve between the muscle bellies of the superficial and deep flexors. The incision was then extended distally into the palm along the thenar crease. The PCBMN was found in "all specimens, and its origin was measured with respect to the distal wrist crease, The PCBMN was then traced distally into the palm, noting any branches and their, level of origin (Fig. 3). The nerve was left attached to the underlying tissue so as to not distort its position. The relationship of the P C B M N to the overlying thenar c r e a ~ was then measured at the level of the distal wrist crease. Positive values were assigned to measurements where the PCBMN coursed radial to the thenar crease, and negative values were assigned where the nerve ran ulnar to the crease. These measurements were repeated at 1, 2, and 3 cm distal to the crease. The distance between the nerve and interthenar crease and ring finger axes was also lneasure, d.
Clinical Analysis of Crease Markings One hundred human hands (50 pairs) were analyzed. Subjects who had had neither fractures nor surgery on the hand were chosen. The palmar creases were highlighted with ink, with minor creases identified by a broken line. The fingers were then flexed at the metacarpophalangeal and proximal interphalangeal joints. The point where the ring fingertip touched the palm was marked (Fig. 21)). The deepe~st point between the thenar and hypothenar eminences (interthenar depression) was 'also marked (Fig. 2C). Once the markings dried, the hand was placed palm down on a photocopier and a life-size reproduction was made. The photocopies were then used to analyze the palmar crease patterns and to provide measurements for further analysis. The axis of the ring finger in extension was drawn using the midpoint of the proximal and distal digital flexion cre~ses (Fig. 2B). q'nis line was extended proximaUy to the level of the wrist crease. The degree of radial or ulnar projection of" this line was measured with respect to the interthenar depression. The distance from the most proximal visible portion of the thenar crease to the distal wrist crease was 'also measured on cach hand.
Results A n a t o m i c Dissections
The P C B M N was found in all dissected specimens. The nerve arose an average of 41 m m proxi-
646
e[ al. / Avoiding TransecIion of t h e PCBMN
Watchmaker
}..
~i84
.~ii.!
9 ,,
,:,
~ii
:..!~:.~:. ..........
q,;.....
.....
~..:b
:~: -
:.
t,
,';;; ,,
ii i ':?~::?:
~!!i,::: ~::~ ~:iii~:,:~,~;'' ~i,~.. ....-..-...... 9
, i::. ::. -
~-:..-.
~,.......
,
,:::.
:...
~:. :.: 9 N~
"
9
i!
!);/. .-i:
-~i.i-~:I
'~'" "!
!::...........
, ~i ii!~.2:
~:.~;:?....
~i ' ":: ~:::. ~.: .. [f.i.:.
~:: ~:~:..
Figure 2, (A) Thenar crease highlighted. (B) Ring finger axis as drawn with digit in extension, (C) Interthenar line drawn from inter~enar depression proximally to third web space distally. (D) Digits flexed at lhe palmar cutaneous branch of the median nerve and proximal interphalangea/joinks, with mmk made adjacent to ring fingertip, (E) Ring finger flexion axis as drawn from base of ring finger to previously m~ked polm.
real to the distal wrist crease (range, 27-63 mm proximal). The nerve then coursed toward the flexor carpi radMis and entered the palm deep to the splayed tendon of the palmaris longus, when present, The nerve then pierced the superficial palmar fascia m lie in the subcutaneous fat of the palm. At the level of the distal wrisl crease, the P C B M N was an average of 2 m m radial to the thenar crease, with a range
of 6 m m radial to 6 m m ulnar to the thenar crease (.Fig. 4). As the nerve was traced more distally, it came to lie ctircctIy beneath the thenar crease. The axis of the ring finger was significantly different in flexion and extension. The PCBMN was an average of 9 m m radial to the extension axis of the ring finger at all levels (range, 1-16 ram). By contrast, the flexion axis projected more radially, with
The Journal of Hand Surgery/Vol, 21A No. 4Jury t996 :..l.,':.:.'.". I'. .." .:'.':. .,'." ~:, . . ." ". ..... :.:. :'....:
i~
i~
........ , .'".:.'.: / ~ : .., :::~:~?:~z.':'W."..'. ,,::<:
: : :::~!
:~'~.'~.~.~<"
..........
" ' .!~'::~.'~3
i 9 .. ,,.:..,.~ ~i%~."~.,.~.,.',.'-:,',,,.:,
i~'
~
."
.. ~r ,.~.~,:
x~:.::!
...,......
, .
f
"i
\
~
~(' :2) ::
~
!~ .~:..
:
~:~, i : ;:".: 2?""'"."
:; ....... .......
i:. . . . . . . o 9 "."
9,,....c~ : ..~x~:.
-,.?j:,
~ ',
'
647
~-~::
!~%;.~:::~
,~" .' , ..: ~- .~
::::::~i.~:~::::
.......
~;3:i)
..... :N ~:,:,::Z:;: .~i
"~,~ ~ ~
~
~=~ .....
~- ~"i,
"
..:...,2.1
~-~?~ '~, ~t:~i ~):~9"::
~
-
2 . ~
~.~:!~ ~" 9 ,2. ~ " ' "~: " ~< '. :. : ..5.:
~: ~ : - ~ - %
)
" -~' ' " ~t~.
'~ i ~
,~
Figure 3. Left hand in which the median nerve and its palmar cutaneous branch have been dissected and highlighted with ink. Note the course of the pallnar cutaneous branch directly beneath the thenar crease in th{s specimen. Several smaller branches arbore radially and ulnm'ly in ~e palm.
the P C B M N coursing ulnar to this axis. The P C B M N was found an average of 4-5.5 m m radial to the intcrthenar depression (range, 0 - 1 2 mm). In no case did the P C B M N travel ulnar to the interthenar depression. In two specimens, the P C B M N did pass beneath the depression. The branching pattern of the P C B M N was highly variable (Fig. 5). There was no consistent pattern, as branches projected in both radial and ulnar directions, arising at various levels in the distal forearm and palm. No cormnunicadng branches between the PCBMN and other cutaneous nerves were noted. In a single specimen, a small continuation of the medial antebrachial cutaneous nerve was found to cross superficial to the PCBMN, traveling in an ulnar to radial direction. Using loupe magnification, the P C B M N could be traced an average of 3.5-4 cm distal to the distal wrist crease.
Clinical Analysis of Crease Markings Palmar crease anatomy was variable among the hands studied. In no hand was the thenar crease found to extend as far proximally as the distal wrist crease. On average, the thenar crease began I8 m m distal to the wrist crease (range, 6-35 ram). Twelve of 100
Figure 4. Cumulative data from cadavefic dissections demonstrating the average course (dark line) and range (shaded area) of the palmar cutaneous branch of the median nerve in the palm with respect m overlying landmarks. Top left: course at nerve with respect to ring finger flexion axis. Note that nerve is found uNar to this line. Top fight: course with respect to ring finger extension axis. Bottom left: cnurse with respect to interthenar line. Bottom fight: course with respect to thenar crease9 Note that the palmar cutaneous branch of the median nerve travels directly beneafl~ fl~ethenar crease.
hands demonstrated a broken thenar crease consisting of two or more discontinuous creases. Forty-four hands demonstrated an interthenar crease that traveled parallel and uh~ar to the thenar crcase. In these hands (with the interthenar crease), the interthenar depression exactly coincided with the interthenar crease. The projection of the ring finger axis in flexion was 12 ~ more radial than in extension (range, 0~176 The axis of the ring finger in flexion crossed the distal wrist crease an average of 18 m m more radial than the axis of the ring finger when extended (range, 0 - 2 9 mm).
Discussion The P C B M N is vulnerable to injury in the distal forearm and palm. The procedure that c o m m o n l y places this branch at risk is carpal tunnel release9 Many approaches to the carpal tunnel, including thenar crease incisiun, longitudinal incision ulnar to the axis of the ring-finger ray, longitudinal incision "in line" with the axis o f the ring-finger ray,
648
Watchmaker et al. / Avoiding Transection of the PCBMN
Figure 5. Branching pattern of six of the dissected cadaverie hands (all hands diagrammed~s right hand for clarity).
interthenar incision, and small transverse incisions for use with or without endoscopic assistance, have been advocated. This study quantitatcs the course and variability of the PCBMN in this region in relation to these commonly used surgical approaches. In 12 dissections, Taleisnik describes the PCBMN as coursing immediately ulnar to the flexor carpi radialis tendon in the distal forearm, then penetrating into the palm in its own tunnel roughly in line with the axis of the ray of the ring finger.3 He comments that this anatomy was qualitatively si tailor in 9 of the 12 dissected specimens. Description of the actual method to determine the ring finger axis was not mentioned. Taleisnik concludes that a well-planned incision should be placed on the ulnar side of the axis of the ray of the ring finger at the level of the heel of the hand. His article does not describe how the ring finger ray axis is determined. Owing to the mobile nature of the ulnar side of the hand, the axis of the ring finger is significantly different in flexion and extension. Use of the ring finger metacarpal shaft as a landmark is difficult, owing to the bulk of overlying palmar soft tissue. Incision uluar to the ring-finger axis raises concern about transecting terminal branches supplying the hypothenar skin. Analysis of the palmar cutaneous branch of the ulnar nerve found this branch to be present in 3 of 21 dissected hands2 The ulnar palmar cutaneous nerve emerged ulnar to the ring-finger ray and extended radially to the ring-finger ray axis. This led to the recommendation that an incision be placed "in line" with the ring-finger ray axis.
Caffolt and Green describe the PCBMN as originating 2-3 in. above the wrist crease, running parallel to the median nerve, then crossing the base of the thenar eminence directly over the prominence of the tubercle of the scaphoid. 2 No data is given to support the statement that an "inter4henar" incision will avoid injury to the nerve. Dowdy et at. dissected 52 wrists to determine the relationship of the PCBMN and the palmaris longus tendon. ~ In two specimens, the PCBMN was found to pass through the palmaris Iongus tendon 1 cm and 1.5 cm proximal to its insertion into the palmar aponeurosis. This study identified another possible kvzafion in which the PCBMN could be injured. Naff et at. further described a separate tunnel for the PCBMN but did not comment on its relations to palmar landmarks.6 Bezerra et al. studied 50 adult fixed specimens and demonstrated the PCBMN emerging in the palm at .76 cm (SD + .32 cm) distal to the wrist crease] This study detailed the branching pattern of the nerve in the palm, with the most common configuration (29 specimens) consisting of one lateral and one inter mediate branch. This study also identified a communication with the superficial branch of the radial nerve in two specimens. Although this study concisely quantitated the branching pattern of the PCBMN no correlation was made either to overlying landmarks or to surgical approaches in the hand. Recently, Hobbs et at. described thc relationship of the PCBMN to a reference line described by the axis of the long finger.8 Their dissection of 25 fresh cadavefic specimens demonstrated that the PCBMN does not project more than 1 cm ulnar to this axis. Their conclusion, however, is that the axial fine of the ring finger may be consistently found 1 cm ulnar to the axial line of the tong finger and provides a safe line for incision. Analysis of 100 volunteer hands in our study, however, demonstrates significant variability in the ring finger axis (Fig. 6), providing a poor guideline for incision placement. Dissection of cadaveric upper extremities as well as analysis of ring finger projection and thenar crease anatomy in our study supports several conclusions. First, the thenar crease does not extend proximally to cross the distal wrist crease. The crease begins an average of 18 nun distal to the wrist crease, which creates ambiguity regarding the proper course of incision ff the thenar crease is c h u r n . Often, the thenar crease turns radia~y at its most proximal point, where it may bifurcate. The PCBMN directly underlies or crosses underneath the thenar
The Journal of Hand Surgery/VoL 21A No. 4 JLdy 1996
649
ous work, which describes a relation betweo~ the ring finger axis and the course of the PCBMN. Third, the depression between the thcnar and hypothenar eminences (interthenar depression) provides a constant landmark in the proximal palm. This depression is best visualized by placing the wrist in full extension and looking in line with the plane of the palm. The nerve courses an average o f 4-5.5 mm radial to this depression, ha no specimen was the PCBMN ulnar to the depression. In two specimens, a branch of the P C B M N passed beneath the depression; thus an incision placed several milli meters ulnar to the depression will provide a safe territory for surgical incision. C l i n i c a l Relevance
Figure 6. Variability of axis of ring linger in extension among 100 volunteer hands. Light shading represents range of extension line. Note that extremes extend from theuar to liypothcnar eminences.
crease, making this crease a poor choice for incision and exposure in the palm. Second, the "axis" of the ring finger ray is ambiguous and variable. 11 is significantly different depending on whether the finger is positioned in flexion or extension. Previous publications have not clarified whether the axis of the ring finger ray should be marked with the digit in ltexion or extension. Neither may be an appropriate choice, given the current data. The flexion axis usu',dly crosses tJnto the thenar eminence to lie radial to the tbenar crease. Our cadaveric dissections delnonstrated that the PCBMN would bc lrtmsected by an incision made "in line" or just ulnar to the flexion axis of the ring finger. An incision along the axis of the ring finger in extension, by contrast, would pre vent transection of the PCBMN in all of our specimens. This axis, however, was highly variable. In some specimens, this axis traveled over the interthenar dep~,ession in the middle of the palm, but in most, it projected well over the hypothenar eminence (Fig. 6). The axis of the ring finger in extension varies significantly, depending on whether the fingers are held together in adduction, spread apart in abduction, or somewhere in between. This variability provides a poor basis for the planning of a surgical incision. In addition, our data do not support the findings of previ-
We have used such an incision for carpal tmme] release since 1982. A recent review of 30 hands in 23 patients was performed by an independent evaluator in the form of a standardized telephone survey. All patients agreed to participate. The group consisted of 7 patients (10 hands) involved in workers' compensation claims and I6 patients (20 hands) non-workers' compensation. Fifteen of the nonworkers' compensation patients stated that dleir incision was completely Dfin-ii'ee within 3 weeks of surgery and remained pain-free at the 1-year followup examination. The remaining one patient had persistent pain rated as 2-3 out of 10 in both hands. Four of the workers' compensation patients werc completely pain-free al the 1-year follow-up examination. Five of the compensation patients returned to their preoperative employment (average return to work lime, 8 weeks). No patients described symptoms o f pillar pain. We recognize that many incisions are used to successfTttUy effect carp',d tunnel release. We offer thc incision described in this article and summarized below as an alternative for surgeons to consider :for the reasons outlined. Neither palmar creases nor ring finger axes provide reliable landmarks for the planning of incisions in the palm. The depression between the thenar aud hypothenar e.t~fincne.cs was a constant landmark in tt~e prnximal palm. An incision placed ulnar to the iuterthenar depression will avoid transection of the PCBMN. In order to avoid small arborizations of the PCBMN, wc recommend and use an incision approximately 5 mm ulnar to the intcrthenar depression, extending dista~y in line with the third web space (Fig. 7). Appropriate planning of the cm-p'al lnnnd
650
Watchmaker et a l . / A v o i d i n g Transection of Ihe ~'CHMN incision can decrease the incidence of transection of t h e P C B M N a n d t h e s u b s e q u e n t morbidity.
~i...
Refe f e n c e s
~
:)i:i'
~ : ~ .... !,x.:?
-'~ 5f::.
.
N
~.-~::
F i g u r e 7. Use o f an incision located approximately 5 Ilml ~flnar to interthenar depression can help avoid transection of the palmar cutaneous branch of the median nerve. Note daat the proximal portion is seldom incised.
t. Alter M. Vmiation in palmar creases, Am I Dis Child 1970; 120:424-431. 2. Canoll RE, G~een DP. The signilJcance of the pahuar cuta neous ner~,e at the wrist. Clin Ortholu 1972;83:24-28, 3. Talcisnik J, The. palmar cutmlcous branch of the medium nerve ~ld thc approach to the carpM t.tnmc]: an anato~nic study, J Bone Joint Surg 1973 ;55A:1212-1217. 4. Engber WD, Gmeiner JG. Pulm.'.Lr cuC,'.me~ms brarlch of lhe u]nar nerve. J Hand Surg 1980;5:2(~29~ 5, Dowdy PA, Richards RS, McHarkme I),M. The palmar cutaneous branch ef the median nerve and the palmaris longus tendon: a cadaveric study. J Hand Surg t994;19A:199-202. 6, Nafl"N, Dellon AL, Mackhmon SE. '['he anatomical course of ihe pahnar culaneous Nanch of the median netw-e, including a desm'iplion of its own unique runnel. J Hand S u ~ ]993;1813:316-317, 7. Bezerra AJ, Carvalho VC, Nucd A. An mlatomical study of Ihe palm~w cutmleous branch ef the m e d h n nerve, Surg Radiol Anat 1986;8:183-188. 8. Hobbs RA, Magnussen PA, Tonkia MA. Palmar cutaneous branch of the median ncrvc. J Hand Surg 1990;I 5A-38~43.
T h e pallmtr cutaneous branch of the median nerve ( P C B M N ) is a constant branch that arises in the distal forearm from the radial aspect of the median nerve and travels distally in close association with the flexor carpi radialis tendon. It pierces the superficial palmar fascia in its own tunnel, sending sensory branches to the palmar skin. Although its origin and course in the dist~d forearm rind hand have been previotlsly described, its relationship to c o m m o n l y used
From the Division of PlasOc Surgery,Department el=Sm,gc13;,Wash inglon Utfiversi!ySchooltff Medicine, St. If~n~s,Me. Recollectf~rcpublic;aion April 7, 1995;accepted in l~vised foim Dec. 5, t 995. No beneiit~ in any iunn have been received or wJlJ be receivedfrom a commercialpartyrclateddirectlyor Lnxlh-ectlyto the subject o{"I:hi~article. Reprint requests: Susan E. Mackinnon, MD, One B;~_nmsHospital Pizza, Suite 17424, St. [~mN,Me 63111). 644
The Journal o{ Hand Surgery
landmarks in the palm has not been thoroughly investigated. Surgical approaches to the carpal tunnel release have been based on soft tissue creases a n d bony landmarks present in the hand. The pahnar creases have often been used to conceal a palmar incision. There are various scientific and astrologic terms used tbr the p a h n a r creases I (Fig. 1). Although the thenar crease is raosl often cited as an incision site, the more ulnar interthenar crease is also mentioned2 The axis of the ring finger has also been proposed as a suitable landmark for incision. 3 T h e term axis is s{~mewhat confusing, however, ,as this ctmld signify the axis with the digit fully flexed or extended, or the course of the metacarpal into the palm. This study describes the detailed course of the P C B M N in relation to the overlying creases and previously proposed landmarks in order to provide a
The Journal of Hand Surgery/ '4ol. 21A No. 4 July 1996
Figure 1. The palmar creases. DWC, distal wrist crease; TC, thenar crease (life line); DTC, distal transverse crease (heart lirtc); PTC, proximal transverse crease (head line); ITC, interthenar crease (destiny tint).
scientific basis for avoiding transection of the PCBMN.
Materials and Methods Dissections Twenty-five flesh human cadaveric upper extremities were ob 'tamed for dissection. The thenar crease was first traced with black ink (Fig. 2A). If the thenar crease did not extend to the distal wrist crease, a perpendicular line was drawn from the wrist crease to the most proximal point on the thenar crease. The flexion axis of the ring fitlger was then determined by flexing the metac~poplu 'dangeal and proximal interpha'langeal joints, then marking the point where the fingertip touched the palm (Fig. 2D). The extension axis of the ring linger was determined by marking the middie of the ring finger at both the proximal and distal digital flexion creases. A straight line was then drawn from these two points proximally into the palm while the finger was held in extension (Fig. 2B). Finally, the wrist was extended and the deepest point between the thenar and hypothenar eminences was marked (Fig. 2C). This point was then extended in line with the third web space. Distances between these lines were measured at the distal wrist cre~e, then at 1, 2, 3, ,and 4 cm distal to the distal wrist crease.
645
Disscction using 2.5• ]oupe magnification was begun in the tbrearm by exposing the median nerve between the muscle bellies of the superficial and deep flexors. The incision was then extended distally into the palm along the thenar crease. The PCBMN was found in "all specimens, and its origin was measured with respect to the distal wrist crease, The PCBMN was then traced distally into the palm, noting any branches and their, level of origin (Fig. 3). The nerve was left attached to the underlying tissue so as to not distort its position. The relationship of the P C B M N to the overlying thenar c r e a ~ was then measured at the level of the distal wrist crease. Positive values were assigned to measurements where the PCBMN coursed radial to the thenar crease, and negative values were assigned where the nerve ran ulnar to the crease. These measurements were repeated at 1, 2, and 3 cm distal to the crease. The distance between the nerve and interthenar crease and ring finger axes was also lneasure, d.
Clinical Analysis of Crease Markings One hundred human hands (50 pairs) were analyzed. Subjects who had had neither fractures nor surgery on the hand were chosen. The palmar creases were highlighted with ink, with minor creases identified by a broken line. The fingers were then flexed at the metacarpophalangeal and proximal interphalangeal joints. The point where the ring fingertip touched the palm was marked (Fig. 21)). The deepe~st point between the thenar and hypothenar eminences (interthenar depression) was 'also marked (Fig. 2C). Once the markings dried, the hand was placed palm down on a photocopier and a life-size reproduction was made. The photocopies were then used to analyze the palmar crease patterns and to provide measurements for further analysis. The axis of the ring finger in extension was drawn using the midpoint of the proximal and distal digital flexion cre~ses (Fig. 2B). q'nis line was extended proximaUy to the level of the wrist crease. The degree of radial or ulnar projection of" this line was measured with respect to the interthenar depression. The distance from the most proximal visible portion of the thenar crease to the distal wrist crease was 'also measured on cach hand.
Results A n a t o m i c Dissections
The P C B M N was found in all dissected specimens. The nerve arose an average of 41 m m proxi-
646
e[ al. / Avoiding TransecIion of t h e PCBMN
Watchmaker
}..
~i84
.~ii.!
9 ,,
,:,
~ii
:..!~:.~:. ..........
q,;.....
.....
~..:b
:~: -
:.
t,
,';;; ,,
ii i ':?~::?:
~!!i,::: ~::~ ~:iii~:,:~,~;'' ~i,~.. ....-..-...... 9
, i::. ::. -
~-:..-.
~,.......
,
,:::.
:...
~:. :.: 9 N~
"
9
i!
!);/. .-i:
-~i.i-~:I
'~'" "!
!::...........
, ~i ii!~.2:
~:.~;:?....
~i ' ":: ~:::. ~.: .. [f.i.:.
~:: ~:~:..
Figure 2, (A) Thenar crease highlighted. (B) Ring finger axis as drawn with digit in extension, (C) Interthenar line drawn from inter~enar depression proximally to third web space distally. (D) Digits flexed at lhe palmar cutaneous branch of the median nerve and proximal interphalangea/joinks, with mmk made adjacent to ring fingertip, (E) Ring finger flexion axis as drawn from base of ring finger to previously m~ked polm.
real to the distal wrist crease (range, 27-63 mm proximal). The nerve then coursed toward the flexor carpi radMis and entered the palm deep to the splayed tendon of the palmaris longus, when present, The nerve then pierced the superficial palmar fascia m lie in the subcutaneous fat of the palm. At the level of the distal wrisl crease, the P C B M N was an average of 2 m m radial to the thenar crease, with a range
of 6 m m radial to 6 m m ulnar to the thenar crease (.Fig. 4). As the nerve was traced more distally, it came to lie ctircctIy beneath the thenar crease. The axis of the ring finger was significantly different in flexion and extension. The PCBMN was an average of 9 m m radial to the extension axis of the ring finger at all levels (range, 1-16 ram). By contrast, the flexion axis projected more radially, with
The Journal of Hand Surgery/Vol, 21A No. 4Jury t996 :..l.,':.:.'.". I'. .." .:'.':. .,'." ~:, . . ." ". ..... :.:. :'....:
i~
i~
........ , .'".:.'.: / ~ : .., :::~:~?:~z.':'W."..'. ,,::<:
: : :::~!
:~'~.'~.~.~<"
..........
" ' .!~'::~.'~3
i 9 .. ,,.:..,.~ ~i%~."~.,.~.,.',.'-:,',,,.:,
i~'
~
."
.. ~r ,.~.~,:
x~:.::!
...,......
, .
f
"i
\
~
~(' :2) ::
~
!~ .~:..
:
~:~, i : ;:".: 2?""'"."
:; ....... .......
i:. . . . . . . o 9 "."
9,,....c~ : ..~x~:.
-,.?j:,
~ ',
'
647
~-~::
!~%;.~:::~
,~" .' , ..: ~- .~
::::::~i.~:~::::
.......
~;3:i)
..... :N ~:,:,::Z:;: .~i
"~,~ ~ ~
~
~=~ .....
~- ~"i,
"
..:...,2.1
~-~?~ '~, ~t:~i ~):~9"::
~
-
2 . ~
~.~:!~ ~" 9 ,2. ~ " ' "~: " ~< '. :. : ..5.:
~: ~ : - ~ - %
)
" -~' ' " ~t~.
'~ i ~
,~
Figure 3. Left hand in which the median nerve and its palmar cutaneous branch have been dissected and highlighted with ink. Note the course of the pallnar cutaneous branch directly beneath the thenar crease in th{s specimen. Several smaller branches arbore radially and ulnm'ly in ~e palm.
the P C B M N coursing ulnar to this axis. The P C B M N was found an average of 4-5.5 m m radial to the intcrthenar depression (range, 0 - 1 2 mm). In no case did the P C B M N travel ulnar to the interthenar depression. In two specimens, the P C B M N did pass beneath the depression. The branching pattern of the P C B M N was highly variable (Fig. 5). There was no consistent pattern, as branches projected in both radial and ulnar directions, arising at various levels in the distal forearm and palm. No cormnunicadng branches between the PCBMN and other cutaneous nerves were noted. In a single specimen, a small continuation of the medial antebrachial cutaneous nerve was found to cross superficial to the PCBMN, traveling in an ulnar to radial direction. Using loupe magnification, the P C B M N could be traced an average of 3.5-4 cm distal to the distal wrist crease.
Clinical Analysis of Crease Markings Palmar crease anatomy was variable among the hands studied. In no hand was the thenar crease found to extend as far proximally as the distal wrist crease. On average, the thenar crease began I8 m m distal to the wrist crease (range, 6-35 ram). Twelve of 100
Figure 4. Cumulative data from cadavefic dissections demonstrating the average course (dark line) and range (shaded area) of the palmar cutaneous branch of the median nerve in the palm with respect m overlying landmarks. Top left: course at nerve with respect to ring finger flexion axis. Note that nerve is found uNar to this line. Top fight: course with respect to ring finger extension axis. Bottom left: cnurse with respect to interthenar line. Bottom fight: course with respect to thenar crease9 Note that the palmar cutaneous branch of the median nerve travels directly beneafl~ fl~ethenar crease.
hands demonstrated a broken thenar crease consisting of two or more discontinuous creases. Forty-four hands demonstrated an interthenar crease that traveled parallel and uh~ar to the thenar crcase. In these hands (with the interthenar crease), the interthenar depression exactly coincided with the interthenar crease. The projection of the ring finger axis in flexion was 12 ~ more radial than in extension (range, 0~176 The axis of the ring finger in flexion crossed the distal wrist crease an average of 18 m m more radial than the axis of the ring finger when extended (range, 0 - 2 9 mm).
Discussion The P C B M N is vulnerable to injury in the distal forearm and palm. The procedure that c o m m o n l y places this branch at risk is carpal tunnel release9 Many approaches to the carpal tunnel, including thenar crease incisiun, longitudinal incision ulnar to the axis of the ring-finger ray, longitudinal incision "in line" with the axis o f the ring-finger ray,
648
Watchmaker et al. / Avoiding Transection of the PCBMN
Figure 5. Branching pattern of six of the dissected cadaverie hands (all hands diagrammed~s right hand for clarity).
interthenar incision, and small transverse incisions for use with or without endoscopic assistance, have been advocated. This study quantitatcs the course and variability of the PCBMN in this region in relation to these commonly used surgical approaches. In 12 dissections, Taleisnik describes the PCBMN as coursing immediately ulnar to the flexor carpi radialis tendon in the distal forearm, then penetrating into the palm in its own tunnel roughly in line with the axis of the ray of the ring finger.3 He comments that this anatomy was qualitatively si tailor in 9 of the 12 dissected specimens. Description of the actual method to determine the ring finger axis was not mentioned. Taleisnik concludes that a well-planned incision should be placed on the ulnar side of the axis of the ray of the ring finger at the level of the heel of the hand. His article does not describe how the ring finger ray axis is determined. Owing to the mobile nature of the ulnar side of the hand, the axis of the ring finger is significantly different in flexion and extension. Use of the ring finger metacarpal shaft as a landmark is difficult, owing to the bulk of overlying palmar soft tissue. Incision uluar to the ring-finger axis raises concern about transecting terminal branches supplying the hypothenar skin. Analysis of the palmar cutaneous branch of the ulnar nerve found this branch to be present in 3 of 21 dissected hands2 The ulnar palmar cutaneous nerve emerged ulnar to the ring-finger ray and extended radially to the ring-finger ray axis. This led to the recommendation that an incision be placed "in line" with the ring-finger ray axis.
Caffolt and Green describe the PCBMN as originating 2-3 in. above the wrist crease, running parallel to the median nerve, then crossing the base of the thenar eminence directly over the prominence of the tubercle of the scaphoid. 2 No data is given to support the statement that an "inter4henar" incision will avoid injury to the nerve. Dowdy et at. dissected 52 wrists to determine the relationship of the PCBMN and the palmaris longus tendon. ~ In two specimens, the PCBMN was found to pass through the palmaris Iongus tendon 1 cm and 1.5 cm proximal to its insertion into the palmar aponeurosis. This study identified another possible kvzafion in which the PCBMN could be injured. Naff et at. further described a separate tunnel for the PCBMN but did not comment on its relations to palmar landmarks.6 Bezerra et al. studied 50 adult fixed specimens and demonstrated the PCBMN emerging in the palm at .76 cm (SD + .32 cm) distal to the wrist crease] This study detailed the branching pattern of the nerve in the palm, with the most common configuration (29 specimens) consisting of one lateral and one inter mediate branch. This study also identified a communication with the superficial branch of the radial nerve in two specimens. Although this study concisely quantitated the branching pattern of the PCBMN no correlation was made either to overlying landmarks or to surgical approaches in the hand. Recently, Hobbs et at. described thc relationship of the PCBMN to a reference line described by the axis of the long finger.8 Their dissection of 25 fresh cadavefic specimens demonstrated that the PCBMN does not project more than 1 cm ulnar to this axis. Their conclusion, however, is that the axial fine of the ring finger may be consistently found 1 cm ulnar to the axial line of the tong finger and provides a safe line for incision. Analysis of 100 volunteer hands in our study, however, demonstrates significant variability in the ring finger axis (Fig. 6), providing a poor guideline for incision placement. Dissection of cadaveric upper extremities as well as analysis of ring finger projection and thenar crease anatomy in our study supports several conclusions. First, the thenar crease does not extend proximally to cross the distal wrist crease. The crease begins an average of 18 nun distal to the wrist crease, which creates ambiguity regarding the proper course of incision ff the thenar crease is c h u r n . Often, the thenar crease turns radia~y at its most proximal point, where it may bifurcate. The PCBMN directly underlies or crosses underneath the thenar
The Journal of Hand Surgery/VoL 21A No. 4 JLdy 1996
649
ous work, which describes a relation betweo~ the ring finger axis and the course of the PCBMN. Third, the depression between the thcnar and hypothenar eminences (interthenar depression) provides a constant landmark in the proximal palm. This depression is best visualized by placing the wrist in full extension and looking in line with the plane of the palm. The nerve courses an average o f 4-5.5 mm radial to this depression, ha no specimen was the PCBMN ulnar to the depression. In two specimens, a branch of the P C B M N passed beneath the depression; thus an incision placed several milli meters ulnar to the depression will provide a safe territory for surgical incision. C l i n i c a l Relevance
Figure 6. Variability of axis of ring linger in extension among 100 volunteer hands. Light shading represents range of extension line. Note that extremes extend from theuar to liypothcnar eminences.
crease, making this crease a poor choice for incision and exposure in the palm. Second, the "axis" of the ring finger ray is ambiguous and variable. 11 is significantly different depending on whether the finger is positioned in flexion or extension. Previous publications have not clarified whether the axis of the ring finger ray should be marked with the digit in ltexion or extension. Neither may be an appropriate choice, given the current data. The flexion axis usu',dly crosses tJnto the thenar eminence to lie radial to the tbenar crease. Our cadaveric dissections delnonstrated that the PCBMN would bc lrtmsected by an incision made "in line" or just ulnar to the flexion axis of the ring finger. An incision along the axis of the ring finger in extension, by contrast, would pre vent transection of the PCBMN in all of our specimens. This axis, however, was highly variable. In some specimens, this axis traveled over the interthenar dep~,ession in the middle of the palm, but in most, it projected well over the hypothenar eminence (Fig. 6). The axis of the ring finger in extension varies significantly, depending on whether the fingers are held together in adduction, spread apart in abduction, or somewhere in between. This variability provides a poor basis for the planning of a surgical incision. In addition, our data do not support the findings of previ-
We have used such an incision for carpal tmme] release since 1982. A recent review of 30 hands in 23 patients was performed by an independent evaluator in the form of a standardized telephone survey. All patients agreed to participate. The group consisted of 7 patients (10 hands) involved in workers' compensation claims and I6 patients (20 hands) non-workers' compensation. Fifteen of the nonworkers' compensation patients stated that dleir incision was completely Dfin-ii'ee within 3 weeks of surgery and remained pain-free at the 1-year followup examination. The remaining one patient had persistent pain rated as 2-3 out of 10 in both hands. Four of the workers' compensation patients werc completely pain-free al the 1-year follow-up examination. Five of the compensation patients returned to their preoperative employment (average return to work lime, 8 weeks). No patients described symptoms o f pillar pain. We recognize that many incisions are used to successfTttUy effect carp',d tunnel release. We offer thc incision described in this article and summarized below as an alternative for surgeons to consider :for the reasons outlined. Neither palmar creases nor ring finger axes provide reliable landmarks for the planning of incisions in the palm. The depression between the thenar aud hypothenar e.t~fincne.cs was a constant landmark in tt~e prnximal palm. An incision placed ulnar to the iuterthenar depression will avoid transection of the PCBMN. In order to avoid small arborizations of the PCBMN, wc recommend and use an incision approximately 5 mm ulnar to the intcrthenar depression, extending dista~y in line with the third web space (Fig. 7). Appropriate planning of the cm-p'al lnnnd
650
Watchmaker et a l . / A v o i d i n g Transection of Ihe ~'CHMN incision can decrease the incidence of transection of t h e P C B M N a n d t h e s u b s e q u e n t morbidity.
~i...
Refe f e n c e s
~
:)i:i'
~ : ~ .... !,x.:?
-'~ 5f::.
.
N
~.-~::
F i g u r e 7. Use o f an incision located approximately 5 Ilml ~flnar to interthenar depression can help avoid transection of the palmar cutaneous branch of the median nerve. Note daat the proximal portion is seldom incised.
t. Alter M. Vmiation in palmar creases, Am I Dis Child 1970; 120:424-431. 2. Canoll RE, G~een DP. The signilJcance of the pahuar cuta neous ner~,e at the wrist. Clin Ortholu 1972;83:24-28, 3. Talcisnik J, The. palmar cutmlcous branch of the medium nerve ~ld thc approach to the carpM t.tnmc]: an anato~nic study, J Bone Joint Surg 1973 ;55A:1212-1217. 4. Engber WD, Gmeiner JG. Pulm.'.Lr cuC,'.me~ms brarlch of lhe u]nar nerve. J Hand Surg 1980;5:2(~29~ 5, Dowdy PA, Richards RS, McHarkme I),M. The palmar cutaneous branch ef the median nerve and the palmaris longus tendon: a cadaveric study. J Hand Surg t994;19A:199-202. 6, Nafl"N, Dellon AL, Mackhmon SE. '['he anatomical course of ihe pahnar culaneous Nanch of the median netw-e, including a desm'iplion of its own unique runnel. J Hand S u ~ ]993;1813:316-317, 7. Bezerra AJ, Carvalho VC, Nucd A. An mlatomical study of Ihe palm~w cutmleous branch ef the m e d h n nerve, Surg Radiol Anat 1986;8:183-188. 8. Hobbs RA, Magnussen PA, Tonkia MA. Palmar cutaneous branch of the median ncrvc. J Hand Surg 1990;I 5A-38~43.