| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1Department of Mechanical Engineering, Speed School of Engineering, 2Laboratory for Biomechanics and Reconstructive Surgery of the Hand, Price Institute for Surgical Research, Department of Surgery, and 3School of Medicine, University of Louisville; and 4Louisville Arm and Hand Center, Norton Hospital, Louisville, Kentucky
Submitted 26 June 2007; accepted in final form 5 May 2008
 |
ABSTRACT |
|---|
|
|
|
INTRODUCTION |
|---|
|
, 1993). Being superficial, the juncturae have been described many times (el-Badawi et al. 1995; Hirai et al. 2001; Kitano et al. 1996; Seradge et al. 1999; von Schroeder and Botte 1993, 1995, 2001; von Schroeder et al. 1990; Wehbe 1992; Zilber and Oberlin 2004). Far less attention has been given to the structure of the ED muscle body in the forearm. Commonly it is assumed that because of the tendon displacement limitations from the juncturae, the independence of ED parts to the different fingers is also limited. Anatomically, this is emphasized by the fact that the index and little finger have an extra, independent extensor muscle—the extensor indicis (EI) and extensor digiti minimi (EDM), respectively. However, also in the middle and ring fingers, more-or-less independent finger extension is possible, so that some independence must exist in the corresponding ED muscle parts.
|
) and detailed needle electrode studies on human and primate finger flexors, extensors, and intrinsics (Reilly and Schieber 2003; Schieber 1995; Schieber et al. 2001) were conducted as well as cross-talk analyses of forearm muscles (Mogk and Keir 2003). Surface electromyography (EMG) of individuated activity of ED parts to the fingers has to our knowledge not been explored. Clinically, in upper limb paralysis, it is attempted to restore gross hand function by neuroprostheses through muscle stimulation by surgically implanted electrodes (Carroll et al. 2000; Hausman and Masters 2002; Hobby et al. 2001; Taylor et al. 2002). In such treatment, optimal anatomic electrode positioning is of prime importance. The aim of this study was to determine the anatomic basis for individuated ED finger extension, the feasibility and optimal electrode locations for individuated ED surface EMG, and optimal neuroprostheses finger extensor electrode locations for hand opening in grasp.
Morphologically, a consistent gross individuation of the ED parts to index, middle, and ring fingers was found, correlating with distinct locations at the ED origin tendons. Concepts on the relationship between origin tendons and muscle morphology were presented in (Leijnse 1997) and are summarized in METHODS to facilitate the anatomical description.
|
|
METHODS |
|---|
|
The further anatomic descriptions may benefit from the following model of muscle origin tendons (OTs). As a model basis for gross description, it is assumed that muscles are assemblies of parallel elementary units, consisting of an individual tendon fiber, a muscle fiber, and a tendon fiber in series, spanning between bone attachments. The individual tendon fibers of the individual muscle fibers assemble in tendon fiber sheets, one of origin, one of insertion (Fig. 2 A). The tendon sheets lose thickness or gain thickness as tendon fibers terminate by giving rise to their muscle fibers or are added as muscle fibers insert with their tendon fibers (Fig. 2A). The OT sheets, while themselves arising from small areas or lines on bone, vastly expand the skeletal surface area for muscle attachment and allow muscles, such as the extensor digitorum and other forearm muscles, to develop their muscle bodies far from the actual OT bone attachment (Leijnse 1997).
|
FASCIAS. OT sheets must be contrasted with fascias. Fascias are here defined as sheets of connective tissue separating muscles in compartments without containing tendon fibers of muscle origin or insertion. However, tendon fibers of muscle origin can be embedded in, or covered by fasciatic tissues, which may continue as true fascias after all tendon fibers have ended in muscle fibers (Fig. 2A). The preceding concepts are of relevance in anatomical ED description. For instance, the posterior fascia antebrachii that covers the ED in the forearm is distally a true fascia (containing no tendon fibers of muscle origin) but proximally contains a deep layer of OT fibers covered by fascia tissue.
Specimens
Fifteen forearms of lightly embalmed bodies were amputated 10 cm proximal to the elbow. Light embalming preserves refrigerated specimens for 
3 wk while not significantly affecting tendon tissue stiffness (Anderson 2006), allowing detailed and realistic observations. In the first five specimens, the extensor muscles were qualitatively dissected and photographed in detail (Nikon D100 camera with Nikon 60-mm lens, bilateral illumination by custom-built fluorescent lights, standard black background, scale in view). When it became clear that the muscle parts of the ED had a constant individuated morphology, the ED and all extensor muscles of the forearm in 10 additional specimens were quantitatively documented (Table 1).
|
Skin was removed and the posterior fascia antebrachii cleaned from fat. The outline of the extensor muscles through the opaque fascia antebrachii was photographed in five specimens. In 10 other specimens, the surface width of the documented muscles was measured (Table 1). The fascia antebrachii was removed up to the distal edge of the superficial OT sheets of the superficial muscles. The superficial borders of the ED parts to index (ED2) and ring finger (ED4) were identified and their width was measured. The ED2–ED4 muscle parts were gently eased apart, from distal to proximal, without cutting muscle fibers. To facilitate this, the superficial ED OT sheet was incised parallel to the tendon fibers up to the lateral epicondyle. The ED5 end tendon(s) (ET) were identified and its muscle body dissected from ED4 from distal to proximal. Thereafter all other superficial muscles [extensor digiti minimi (EDM), extensor carpi radialis brevis and longus (ECRB and ECRL), and extensor carpi ulnaris (ECU)] were measured. Their ET were severed for full access to the deep extensors [extensor indicis (EI), extensor pollicis longus (EPL), and abductor pollicis longis/extensor pollicis brevis (APL/EPB)], which were similarly photographed or measured. In selected specimens, the neurovascularization was dissected and photographed in detail.
Measurements
All measurements were performed by a 200-mm range calipers. Normalization: radius length was measured between the humerus-radius joint (HRJ) and the distal radius edge at the radius-lunate joint (RLJ). The maximum width of radius-ulna at the wrist was measured (Table 1). Location and length of origins and muscle bodies (Fig. 3): for each muscle, three length measurements were taken, relative to the HRJ as marked by a pin. 1) The length between the HRJ and the proximal origin edge; 2) between the HRJ and the distal origin edge, and 3) between the HRJ and the distal insertion edge of the muscle belly in the end tendon. The latter was measured with the end tendons pulled taut and fingers and wrist in neutral position of extension (which was not always possible as some finger joints were rigid). If a distance exceeded calipers length (200 mm), it was measured relative to the RLJ (marked by a pin) and recalculated relative to the HRJ using the radius length. Surface widths: the surface width of the ED, EDM, ECRB, ECRL, and ECU was measured with the fascia antebrachii intact. With the fascia removed, but before any dissection, the border between ED2 and ED4 can be identified by a longitudinal line of areolar tissue (Fig. 4 B). ED2 and ED4 surface widths were measured where the proximal edge of the APL crosses the radial edge of the ED (Fig. 4C) (as will be shown, this site is the proposed optimal anatomic location for ED2 and ED4 surface electrodes). After separating ED2 and ED4 from ED3, the surface width of ED3 was measured.
|
|
Quantitative anatomy of finger and wrist extensors
QUANTITATIVE DATA FOUND IN TABLES 1 AND 2 AND FIG. 9. SUPERFICIAL FINGER EXTENSORS (ED AND EDM). V-shaped ED OT compartment. The ECRB, ED, EDM, and ECU originate from an extensive complex of OT sheets, which arises from the lateral epicondyle and extends posterior to the ulna. At the epicondyle, the OT sheets arise as a thick tendon band, which unfolds immediately to form distinct OT compartments. Central is the ED compartment, which is proximally cone shaped with relatively short superficial and deep tendon fibers, superficially shorter radial than ulnar. However, the radial and ulnar sides of the ED OT compartment continue distal far into the forearm, giving the ED OT a narrow V-shape with long legs (Fig. 5 C). The radial ED OT reaches to about mid-forearm [49 ± 7% of radius length (RL)]. The ulnar ED OT reaches up to or beyond the distal third of the forearm (65 ± 9% RL).
|
|
|
The ED3 origin completely occupies the proximal tip of the ED OT compartment (Fig. 5, A–C). The most proximal ED3 muscle fibers arise well proximal to the HRJ by an average 9.2 ± 3.4 mm or 4.0 ± 1.5% RL. Distally, the origin area of the ED3 is delimited by the origins of ED2 and ED4. The distal ED3 origin edge and the proximal ED2 and ED4 origin edges may to some degree overlap as can be seen in Fig. 5, A and C. The total ED3 origin reaches 22.6 ± 5.2% RL (maximum of edges on radial and ulnar ED OT), slightly more distal radial than ulnar (19.1 ± 5.4 vs. 17.8 ± 7.7% RL, respectively), as is the case in Fig. 5, A and B. From these origins, the ED3 muscle belly runs deep to the ED2 and ED4 muscle bellies, reaching to 47.3 ± 5.5% RL.
The ED2 origin occupies the radial ED OT distal to the ED3, with the proximal muscle fibers arising at 14.5 ± 4.4% RL. The proximal ED2 origin at the OT is to some degree superficial to and overlaps to a variable degree with the distal ED3 origin (Fig. 5A). The most distal ED2 muscle fibers arise at 49.4 ± 7.3% RL. The muscle belly reaches 74.9 ± 7.9% RL. The ED2 tendon runs ulnar to the ED2 muscle body near the ED midline (as in Fig. 3).
ED4 arises from the ulnar ED OT distal to ED3 at 13.7 ± 8.3% RL, somewhat proximal but with greater variability than the ED2. The ED4 origin reaches 38.8 ± 5.5% RL, the muscle belly 64.4 ± 6.2% RL. The proximal ED4 origin may interdigitate with the distal ED3 origin, resulting in a three-dimensional interface that may prevent neat separation of these muscle parts near the OT. The ED4 OT envelops the ulnar side of the distal ED3 muscle body so that the ED4 muscle body covers the distal ED3 muscle belly ulnarly. The ED4 tendon runs central in the ED, adjacent and ulnar to the ED2 tendon. In the mid forearm, the ED4 and ED2 muscle bellies are superficially adjacent, entirely covering the ED3 muscle belly and tendon.
The ED has an additional muscle part further called ED5. Its ET runs ulnar to the ED4 ET, typically inserts with a V shape in both the ring and little finger and may practically be considered common to both fingers (Fig. 1). The ED5 ET may be strongly connected to or even fused with the ED4 ET at the dorsum of the hand (Fig. 1B) but in any case, gives a strong junctura to the little finger. The ED5 ET may consist of two or even more tendon strands (Fig. 1, A and D), which at the dorsum of the hand may be strongly interconnected but proximal to the wrist may be individuated from their connective tissues and followed up proximal to sublayers in the ED5 muscle belly. Occasionally, the ED5 and ED4 muscle bellies can be readily separated pointing to some independent action (specimen in Figs. 5, A, C–D, and 6A, and specimen in Fig. 6 D). In other cases, the ED4 and ED5 muscle bellies are strongly interconnected by connective tissue, suggesting common action (specimens in Fig. 6, B and C). The ED5 origin starts proximal at 37.8 ± 5.9% RL and reaches to 65.4 ± 9.1% RL, which coincides with the distal edge of the ulnar ED OT. The muscle belly reaches to 85.3 ± 8.8% RL.
|
Extensor digiti minimi. The slender EDM (Fig. 6, A–D) arose in all cases from within its own OT compartment, sandwiched superficially between the ED and ECU. The ulnar EDM compartment side is the radial surface of the radial ECU OT. The radial EDM compartment side is the bipennate ulnar ED OT from which on the radial side the ED4 and ED5 arise, and on the ulnar side the EDM muscle fibers (Fig. 5D). From these OT, the EDM muscle fibers converge to a central tendon. The EDM OT compartment develops proximally at 14.7 ± 5.3% RL. The origin from the radial ECU OT was generally much shorter than from the ulnar ED OT, reaching to 31.6 ± 2.2% RL as compared with 63.9 ± 8.1% RL for the latter (Fig. 6, B and D). Distally, the bipennate ulnar ED OT sheet occasionally separated in two individual unipennate OT slips for the ED5 and EDM, respectively, as modeled in Fig. 2B and shown in Fig. 6A. Figure 7 shows the ulnar ED OT after removal of all ED4, ED5, and EDM muscle fibers. The EDM origin reached distally to 63.9 ± 8.1% RL and the muscle belly to 85.4 ± 6.0% RL.
|
WRIST EXTENSORS. Extensor carpi radialis longus. The ECRL arises from short OT fibers from the lateral humerus crest at the lateral epidondyle, starting at 59 ± 11.0 mm or 25.4 ± 4.0% RL proximal to the HRJ. The ECRL origin continues for a short length of 10.5 ± 9.6 mm or 4.5 ± 3.9% RL distal to the HRJ at the OT sheet of the ECRB, distally delimited by the origin of the ECRB. The ECRL muscle belly reaches 34.8 ± 3.8% RL.
Extensor carpi radialis brevis. The ECRB arises radial to the ED from within a large half open OT compartment, bluntly L-shaped in transverse cross-section. The ulnar wall of the compartment is the radial ED OT sheet, which is bipennate with the ED arising ulnarly and the ECRB radially. This OT sheet has a large deep unipennate radial expansion which forms the floor of the L shaped ECRB compartment (Fig. 7A). The most proximal muscle fibers of the ECRB arise 4.2 ± 7.0 mm or 1.7 ± 3.0% RL distal to the HRJ. The origin reaches to 44.5 ± 3.4% RL and muscle belly to 67.2 ± 5.5% RL.
Extensor carpi ulnaris. The ECU arises in a closed flat OT compartment that originates at the lateral epicondyle. The most proximal ECU muscle fibers start at 5.7 ± 3.7 mm or 2.4 ± 1.6% RL. Proximally, the radial (deep) side of the ECU OT compartment is common with the ulnar ED OT until separated by the EDM compartment. The ECU OT compartment is asymmetric, being much longer radial than ulnar (Fig. 8). The radial ECU OT reaches 71.2 ± 4.0% RL with its muscle fibers inserting at the radial side of the ECU ET up to 88.5 ± 2.7% RL. The ulnar (superficial) ECU OT, covered by the posterior fascia antebrachii, reaches barely halfway the forearm at 49.5 ± 4.1% RL with its muscle fibers inserting at the ulnar side of the ECU ET up to 67.3 ± 2.1% RL. More distally the superficial ECU ET surface is bare of muscle fibers as modeled in Fig. 8 and illustrated in Fig. 4, C and D.
|
Extensor indicis.
The EI arises from the radial surface of the ulna as well as from short OT fibers from the dorsal ulnar crest and from OT fibers in the interosseus membrane. The EI origin starts proximal at 48.7 ± 3.7% and reaches 77.0 ± 2.8% RL. The EI muscle body reaches the most distal of all extensors to 95.5 ± 3.2% RL up to or underneath the extensor retinaculum of the wrist (Fig. 7, A and B). Anatomical variations in the EI are not rare—in 3 of 10 arms (2 bilateral in the same subject) accessory muscle bellies/tendons were found. The bilateral accessory tendons inserted ulnar to the ED3 tendon in the extensor assembly of the medius, a variation also described by von Schroeder and Botte (1991). In the other specimen, the accessory tendon was continuous with the thickened distal-radial edge of the sheath enveloping the EI-EPL tendons at the hand dorsum.
Extensor pollicis longus. The EPL arises adjacent, proximal, and radial to the EI from the radial surface of the ulna, short OT fibers from the dorsal ulnar crest (Fig. 7B) and from OT fibers in the membrana interossea. The EPL crosses deep to the ED tendons in a radial-distal course and becomes superficial at entering the third extensor compartment at the distal radius. The EPL ET curves at the ulnar side around the tubercle of Lister and runs then distal-radially to the thumb extension side. The EPL origin starts at 31.0 ± 5.1% and reaches to 73.0 ± 2.6% RL. The muscle belly reaches slightly less distal than the EI at 91.6 ± 2.4% RL.
Extensor pollicis brevis and abductor pollicis longus. The EPB and APL origins (which we consider together) occupy the entire space between the supinator and EPL at the membrana interossea and the ulnar surface of the radius; include OT fibers from radius and ulna, and reach proximal near the ulna even somewhat underneath the supinator. The APL/EPB origin starts at 24.4 ± 2.1% RL and reaches to 73.5 ± 5.4% RL, while the muscle bellies reach 92.5 ± 2.3% RL.
NEUROVASCULARIZATION. The motor branch of the radial nerve emerges distal from beneath the supinator together with the posterior interosseus artery to form a neurovascular plexus of forward and recurrent branches innervating all ED parts, EDM, ECU, APL, EPB, EPL, and EI (Fig. 10). The ED2–ED5 muscle bellies receive distinct nerve branches entering the ED compartment from deep at the center of the V-shape of the ED OT. Additional vascularization is provided by more distal arteries perforating the interosseus membrane from palmar to dorsal.
|
|
ED2.
ED2 surface electrode locations are determined by three constraints (Fig. 11, A–D). Electrodes must be placed proximal to where the proximal APL edge crosses the radial ED edge, to minimize cross-talk from APL, which is a powerful muscle; near the distal ED3 muscle belly edge, so that no ED3 activity can be picked up; and as radial as possible to avoid cross-talk from the adjacent ED4/5. This determines a narrow range, basically, a single location, at 45% RL. The ED2 and ED4/5 muscle bellies are adjacent over their entire length so that mutual cross-talk is unavoidable. Note that because ED2 is narrower than ED4 (Table 2), ED2 electrodes placed as in Fig. 11 will be closer to ED4 than ED4 electrodes to ED2. ED2 electrodes will thus likely pick up more cross-talk from ED4 than inversely. Minimizing ED4 cross-talk by placing ED2 electrodes at the radial ED2 edge can be envisaged when only ED muscle parts need to be assessed, as in individuated finger tapping tasks with forearm and hand in rest and no wrist extensor activity (J. N. Leijnse, N. H. Campbell-Kyureghyan, D. Spektor and P. M. Quesada, unpublished data). With active wrist extensors, radially placed ED2 electrodes will receive increased ECRB cross-talk. From deep, ED2 electrodes may receive APL cross-talk.
ED4. The ED4 is surrounded by radially ED2; proximal, radial, and deep ED3; and ulnar EDM. ED3 cross-talk is avoided by placing ED4 electrodes near 45% RL where the ED3 muscle belly is ending. ED2 cross-talk can be minimized by placing the ED4 electrodes near the ulnar ED4 edge, but this increases EDM cross-talk. With a very distal arising EDM compartment, which can sometimes be palpated, ED4 electrodes may be put somewhat more proximal to minimize EDM cross-talk, but this will increase ED3 cross-talk. With a proximal arising EDM compartment, EDM cross-talk is unavoidable. Cross-talk to ED4 electrodes might also arise from the deep APL/EPB, of which the origins reach ulnarly beneath the ED4 muscle belly (Fig. 11, C and D).
ED5. The ED5 muscle belly is narrow and anatomically not systematically independent from ED4. Therefore we do not consider it a candidate for valuable independent surface EMG assessment. ED4 electrodes placed as in Fig. 11 will measure some ED5 activity too.
EDM. EDM width is 9.0 ± 2.1 mm, just enough for accurately placed 4-mm cup electrodes. EDM electrodes can be placed about halfway up the forearm, which is near the proximal origin edge of the underlying EI. This will minimize EI cross-talk, but EPL cross-talk remains possible. Placing EDM electrodes more distal will increase EI cross-talk. Given the small EDM width, ED4 and ECU cross-talk is unavoidable.
EI. The EI muscle belly consistently reaches the most distal of all extensors. EI electrodes may be put as distal and ulnar as just radial to the ulnar head (Fig. 11D). However, the EDM and ED5 muscle bellies, while reaching on average less distal than EI, are highly variable in length and may reach the EI electrodes (Fig. 9). Such distal EDM and ED5 muscle bellies lay then between EI and its electrodes, so that significant EDM and ED5 cross-talk may result, especially when EI is short, so that its electrodes must be placed more proximally.
EPL.
The EPL reaches only slightly less distal than EI. The most distal EPL electrode might be placed 10–15 mm proximal to Listers' tuberculum at the radial EPL side to minimize EI cross-talk. ED5 cross-talk is statistically to be expected as the EPL crosses beneath the ED tendons and their distal reaching muscle bellies. As Fig. 11D shows, EPB cross-talk can also be expected.
APL/EPB.
The APL/EPB can be assessed, without great expected cross-talk from any other muscle, radial and lateral-dorsal to the radius at 75% RL in the forearm.
ECRB. The ECRB tapers considerably at origin and insertion, and ECRB electrodes are most isolated from adjacent muscles when placed at the center of the muscle belly at 40% RL.
ECRL. The ECRL origin reaches proximal on the humerus 59 ± 11.0 mm or 25.4 ± 4.0% RL from the HRJ. ECRL electrodes may thus be placed high on the epicondyle and close to the lateral humerus crest as far from the ED3 electrodes as possible. Even so, with flexed elbow the ECRL electrodes align closer to the ED3 and may well pick up ED3 cross-talk.
ECU.
The ECU is proximally adjacent to powerful muscles: ulnar the anconeus, deep the supinator, and radial the ED. Distal to the anconeus, the radial ECU surface lays against, from deep to superficial, the ulnar-dorsal ulna surface, the ulnar borders of EPL and EI (Fig. 11D), and the EDM. The thick ECU ET becomes ulnarly free of muscle fibers at 67.3 ± 2.1% RL, although muscle fibers thin out well proximal to that point (Fig. 8). Therefore no ECU electrodes should be placed distal to 60% of RL. It follows that despite the great ECU muscle belly length, optimal ECU electrode locations are actually limited to 50% of RL to avoid cross-talk from the large proximal muscles or placing ECU electrodes over the bare distal ET surface.
Optimal locations for neuroprosthesis finger extensor electrodes
In neuroprosthetic upper extremity palsy restoration, finger extensor electrodes serve to restore hand opening in grasp. Ideally the fingers are extended evenly, independent of the thumb, allowing thumb opposition/abduction by other means in preparation of subsequent grasping hand closure against an opposed thumb. Optimal neuroprosthesis electrode locations may differ from optimal EMG assessment locations. An EMG signal is after acquisition normalized by maximum voluntary contraction EMG, meaning that signal strength may be traded for lesser cross-talk by positioning electrodes eccentric at the muscle. In contrast, neuroprosthesis electrodes aim to stimulate a maximum of muscle bulk, requiring electrodes central at the muscle belly. From the dissection data, the following can be proposed.
SINGLE NEUROPROSTHESIS ELECTRODE CANNOT LIKELY EVENLY ACTIVATE ALL FINGER EXTENSORS. As shown in Fig. 12, a single electrode at position e1 will activate almost exclusively the ED3 muscle belly as it does not cover the other ED parts. At position e2, it will strongly activate ED4 but weakly activate ED2, ED3 and ED5 and almost none of EDM, being strongly eccentric at the latter muscle bellies. At position e3, it will not activate the ED3 muscle belly at all. In conclusion, in all preceding cases, a single electrode will result in greatly different extension forces in the individual fingers.
|
THREE OR FOUR ELECTRODES MAY EVENLY ACTIVATE ALL FINGER EXTENSORS. Three electrodes placed as in Fig. 13, A–C, would cover ED3 (e1), ED2 and partly ED4/5 (e2), and EDM and partly ED4/5 (e3). Four electrodes placed as in Fig. 13, D–F, might provide even more homogenous stimulation, with e1 covering ED3, e2 mainly ED4, e3 covering ED2 and partly ED5, and e4 covering EDM. However, electrode e2 is closer to the RN plexus, although when placed superficial at ED4, it is still shielded from the nerve plexus by the underlying ED3 muscle belly and its ET. Because no individuated finger extensor function is intended in neuroprosthetic hand opening in grasp, all electrodes can be connected to the same activation channel.
|
|
|
DISCUSSION |
|---|
|
The ED is generally regarded as a common finger extensor, hence its name. However, the dissection results show that by means of extensive origin tendons, the ED parts to the different fingers are spaced out in a way naturally conducive to individuation. The ED3 arises from the proximal part of the V-shaped OT, ED2 from the radial ED OT, and ED4 and ED5 sequentially from the long ulnar ED OT. No exceptions were found in this pattern in 15 specimens. While the ED2–ED4 were found reasonably well separable, occasionally tendon strands were found crossing between ED2 and ED3 and between ED3 and ED4 but, in our specimens, not between ED2 and ED4. The ED4 and ED5 could not be consistently separated—such inconsistent individuation can also be observed in their end tendons.
Individuated EMG assessment in ED and EDM
The extreme proximal-distal spacing of the ED parts allows placing bipolar surface electrodes in lengthwise spaced locations that partially compensate for the small muscle belly widths. Because all muscle parts are superficial, reproducible individuated EMG should be obtainable at the locations proposed in Fig. 11. Needle electrodes should measure individuated EMG with little cross-talk. With surface electrodes, the expected cross-talk on basis of anatomic muscle part proximity was analyzed in the results. ED3 activity should be measurable without significant cross-talk from or to ED2 and ED4. ED2 and ED4, being adjacent, should be independently measurable but with more mutual cross-talk. ED4 and EDM will have significant mutual cross-talk. Between ED5/EDM and the deep extensors EI/EPL, cross-talk will likely vary with the anatomically variable distal reach of their muscle bellies. Extensor surface EMG assessment in finger tapping tasks closely confirmed the above anatomical projections (Leijnse et al., unpublished data). Concerning cross-talk filtering, it has been pointed out that a source of cross-talk is nonpropagating waves when action potentials reach muscle fiber ends at their tendon attachment (which happens at OT and ET alike). Cross-talk reductions were proposed by spatial filtering combining outputs of more than two electrodes to accentuate propagating waves from the muscle of interest while attenuating nonpropagating waves from other muscles (Farina et al. 2004; van Vugt and van Dijk 2001). In the finger extensors, the application of such techniques or spatial filter choice may not be straightforward. The long OT and ET interlace so that standing waves will be generated all along the muscle of interest. In closed OT compartments such as ECU and EDM, nonpropagating waves will within the same muscle equally arise from muscle fibers inserting at opposite OT tendon surfaces of the compartment as well as at both surfaces of the bipennate ET. It follows that minimizing cross-talk by electrode placement based on least anatomic proximity remains of prime relevance.
Optimal ED electrode locations with neuroprostheses
Usefulness of hand opening in grasp is determined by the fingers with the smallest extension range. Figure 12 suggests that a single electrode cannot likely evenly activate all ED parts and EDM. Therefore in neuroposthetic hand opening with a single extensor electrode, some fingers will likely trail others. This may result in the need to overstimulate well-simulated ED parts to obtain sufficient stimulation of understimulated ED parts to achieve the desired extension range in all fingers. As Fig. 13 show, three or four electrodes connected to a single control channel may significantly improve homogenous finger extensor stimulation. The consistent anatomical location of the ED parts and the fact that they can be easily superficially stimulated per-operatively through the fascia antebrachii should allow accurate multiple electrode placement without requiring extensive surgical dissection. A point of attention remains the radial motor nerve plexus distal to the supinator. An electrode placed near this nerve plexus may cause unwanted activation of all finger and thumb extensors, including ECU, among others resulting in compromised thumb opposition with extensor activation in hand opening for grasp.
|
|
GRANTS |
|---|
|
|
|
ACKNOWLEDGMENTS |
|---|
|
|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: J.N.A.L. Leijnse, Laboratory for Biomechanics and Reconstructive Surgery of the Hand, Dept. of Mechanical Engineering, Speed School of Engineering, 200 Sackett Hall, University of Louisville, Louisville, KY 40292 (E-mail: jnleij01{at}louisville.edu)
|
|
REFERENCES |
|---|
|
Carroll S, Cooper C, Brown D, Sormann G, Flood S, Denison M. Australian experience with the Freehand System for restoring grasp in quadriplegia [see comment]. Aust N Z J Surg 70: 563–568, 2000.[CrossRef][Web of Science][Medline]
el-Badawi MG, Butt MM, al-Zuhair AG, Fadel RA. Extensor tendons of the fingers: arrangement and variations-II. Clin Anat 8: 391–398, 1995.[CrossRef][Medline]
Farina D, Merletti R, Indino B, Graven-Nielsen T. Surface EMG cross-talk evaluated from experimental recordings and simulated signals. Reflections on cross-talk interpretation, quantification and reduction. Methods Inform Med 43: 30–35, 2004.[Web of Science][Medline]
Hausman MR, Masters JE. Percutaneous Freehand system intramuscular electrode placement. J Hand Surg 27: 465–469, 2002.[Medline]
Hirai Y, Yoshida K, Yamanaka K, Inoue A, Yamaki K, Yoshizuka M. An anatomic study of the extensor tendons of the human hand. J Hand Surg 26: 1009–1015, 2001.[Medline]
Hobby J, Taylor PN, Esnouf J. Restoration of tetraplegic hand function by use of the neurocontrol freehand system. J Hand Surg 26: 459–464, 2001.[Medline]
Keen DA, Fuglevand AJ. Distribution of motor unit force in human extensor digitorum assessed by spike-triggered averaging and intraneural microstimulation. J Neurophysiol 91: 2515–2523, 2004.
Kitano K, Tada K, Shibata T, Yoshida T. Independent index extension after indicis proprius transfer: excision of juncturae tendinum. J Hand Surg 21: 992–996, 1996.[CrossRef][Medline]
Leijnse JN. A generic morphological model of a muscle group–application to the muscles of the forearm. Acta Anat 160: 100–111, 1997.[Web of Science][Medline]
Leijnse JN, Bonte JE, Landsmeer JM, Kalker JJ, Van der Meulen JC, Snijders CJ. Biomechanics of the finger with anatomical restrictions–the significance for the exercising hand of the musician. J Biomechan 25: 1253–1264, 1992.[CrossRef][Web of Science][Medline]
Leijnse JN, Snijders CJ, Bonte JE, Landsmeer JM, Kalker JJ, Van der Meulen JC, Sonneveld GJ, Hovius SE. The hand of the musician: the kinematics of the bidigital finger system with anatomical restrictions. J Biomechan 26: 1169–1179, 1993.[CrossRef][Web of Science][Medline]
Mogk JPM, Keir PJ. Cross-talk in surface electromyography of the proximal forearm during gripping tasks. J Electromyogr Kinesiol 13: 63–71, 2003.[CrossRef][Web of Science][Medline]
Reilly KT, Schieber MH. Incomplete functional subdivision of the human multitendoned finger muscle flexor digitorum profundus: an electromyographic study. J Neurophysiol 90: 2560–2570, 2003.
Schieber MH. Muscular production of individuated finger movements: the roles of extrinsic finger muscles. J Neurosci 15: 284–297, 1995.[Abstract]
Schieber MH, Gardinier J, Liu J. Tension distribution to the five digits of the hand by neuromuscular compartments in the macaque flexor digitorum profundus. J Neurosci 21: 2150–2158, 2001.
Seradge H, Tian W, Baer C. Anatomic variation of the extensor tendons to the ring and little fingers: a cadaver dissection study. Am J Orthop 28: 399–401, 1999.[Medline]
Taylor P, Esnouf J, Hobby J. The functional impact of the Freehand System on tetraplegic hand function. Clinical results. Spinal Cord 40: 560–566, 2002.[CrossRef][Web of Science][Medline]
van Vugt JP, van Dijk JG. A convenient method to reduce cross-talk in surface EMG. Clin Neurophysiol 112: 583–592, 2001.[CrossRef][Web of Science][Medline]
von Schroeder HP, Botte MJ. The extensor medii proprius and anomalous extensor tendons to the long finger. J Hand Surg 16: 1141–1145, 1991.
von Schroeder HP, Botte MJ. The functional significance of the long extensors and juncturae tendinum in finger extension. J Hand Surg 18: 641–647, 1993.[CrossRef][Medline]
von Schroeder HP, Botte MJ. Anatomy of the extensor tendons of the fingers: variations and multiplicity. J Hand Surg 20: 27–34, 1995.[CrossRef][Medline]
von Schroeder HP, Botte MJ. Anatomy and functional significance of the long extensors to the fingers and thumb. Clin Orthop 74–83, 2001.
von Schroeder HP, Botte MJ, Gellman H. Anatomy of the juncturae tendinum of the hand. J Hand Surg 15: 595–602, 1990.[Medline]
Wehbe MA. Junctura anatomy. J Hand Surg 17: 1124–1129, 1992.
Zilber S, Oberlin C. Anatomical variations of the extensor tendons to the fingers over the dorsum of the hand: a study of 50 hands and a review of the literature. Plast Reconstr Surg 113: 214–221, 2004.[CrossRef][Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  H. van Duinen, W. S. Yu, and S. C. Gandevia Limited ability to extend the digits of the human hand independently with extensor digitorum J. Physiol., October 15, 2009; 587(20): 4799 - 4810. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.N.A.L. Leijnse, N. H. Campbell-Kyureghyan, D. Spektor, and P. M. Quesada Assessment of Individual Finger Muscle Activity in the Extensor Digitorum Communis by Surface EMG J Neurophysiol, December 1, 2008; 100(6): 3225 - 3235. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
TOP
ABSTRACT
INTRODUCTION
) and continue as independent unipennate OT sheets.
