Surg Radiol Anat DOI 10.1007/s00276-014-1371-x
ORIGINAL ARTICLE
Anatomical study of the musculocutaneous nerve branching pattern: application for selective neurectomy in the treatment of elbow flexors spasticity Adeline Cambon-Binder • Caroline Leclercq
Received: 10 February 2014 / Accepted: 30 August 2014 Ó Springer-Verlag France 2014
Abstract Purpose Spastic flexion deformity of the elbow is mainly mediated by the biceps brachii and the brachialis muscles, innervated by the musculocutaneous nerve. Selective neurectomy of the musculocutaneous nerve showed promising results to relieve excessive spasticity in the long term but lacks of a consensual surgical strategy. The aim of the study was to describe the distal branching pattern of the motor branches of the musculocutaneous nerve in an attempt to develop guidelines for surgery. Methods Sixteen arms of fresh cadaver specimen were dissected. We recorded the site of each primary and terminal motor branch as a percentage of the distance from the coracoid process to the lateral epicondyle. Results The biceps muscle was innervated by one to five primary motor branches. The first branch emerged from the nerve at an average of 37.1 % of the arm length, and the most distal terminal branch at 55.7 %. The brachialis muscle received one to three primary branches. The first branch exited the nerve at an average of 51.7 % of the arm length and the last terminal branch at 69.3 %. The average number of terminal branches dedicated to the biceps and the brachialis muscles were, respectively, 7.9 and 6.5. Conclusions According to our findings, we recommend to dissect the musculocutaneous nerve between 18 and 75 % of the distance between the coracoid process and the lateral epicondyle to identify the motor terminal branches to the A. Cambon-Binder (&) Hand and Peripheral Nerve Surgery Unit, European Georges Pompidou Hospital, 20 rue Leblanc, 75015 Paris, France e-mail: [email protected] C. Leclercq Institut de la Main, Clinique Jouvenet, 6 square Jouvenet, 75016 Paris, France
biceps brachii and the brachialis muscle, sparing sensory branches. Keywords Musculocutaneous nerve Selective neurectomy Spasticity Selective neurotomy
Introduction Spasticity has been defined as an increase in muscle tone as a result of the hyper-excitability of the muscle fibers and is characterized by a velocity-dependent increase in tonic stretch reflexes [18]. Spasticity is one of the clinical features of the upper motor neuron syndrome resulting from various etiologies such as cerebrovascular injury, traumatic brain injury, multiple sclerosis, and cerebral palsy [7]. In the upper limb, spasticity can lead to pain, impairing deformities interfering with activities of daily living and hygiene, and poor cosmesis. Inhibition of antagonist muscles may mask their residual motor function and induce functional impairment [7, 11]. Common manifestations of spastic disorders involving the upper limb include internal rotation and adduction of the shoulder, flexion of the elbow, pronation of the forearm, flexion and ulnar deviation of the wrist, flexion of the fingers and flexion-adduction of the thumb [11]. Spasticity of the elbow is mediated mainly by the biceps brachii and the brachialis muscles, innervated by the musculocutaneous nerve (MCN). The MCN arises from the lateral cord, from the fifth to seventh cervical roots, at the level of the distal border of the pectoralis major muscle. It passes through the coracobrachialis muscle [26] and travels distally in the lateral aspect of the arm between the biceps brachii and brachialis muscles [8]. Below the elbow and lateral to the tendon of
123
Surg Radiol Anat
the biceps brachii, it pierces the deep fascia and continues as the lateral antebrachial cutaneous nerve. The MCN provides motor innervation to the flexor compartment of the arm and sensory innervation to the radial aspect of the forearm. Many nonsurgical treatments were developed for elbow flexors spasticity but are characterized by their short-term effect. Intrathecal injection of baclofen administered via a programmable pump may provide a continuous spasmreducing effect in cases of regional spasticity [31]. Phenol nerve blocks [13, 15] provide temporary relief of painful contractures and allow for a period of spontaneous neurologic recovery of up to 6 months. Botulinum toxin injected into the muscle body provides a three to four months relief [9, 38]. In an attempt to relieve spastic symptoms in a permanent way, surgical procedures have been developed at the level of peripheral nerves, spinal roots, spinal cord, and the dorsal root entry zone [39]. Microsurgical selective peripheral neurectomy—formerly termed hyponeurotization [2]—consists in a partial and selective excision of the peripheral nerves. It has proven to be effective for some localized forms of spasticity [4, 21, 24, 34] involving muscles innervated by the musculocutaneous, the median, and the ulnar nerves. However, there is controversy in the technical strategy regarding the exact site and amount of nerve resection: main nerve approach and selection of fascicles using electrostimulation, or dissection of the motor branches till their entrance in the muscle and resection of a proportion of the terminal branches. This latter procedure requires a thorough knowledge of the branching pattern of the MCN for the biceps brachii and the brachialis muscles. As far as we know, previous anatomy articles have focused on the number of motor trunks [5, 14, 17, 25, 28, 45], or on the center of the location where the motor branches enter the muscle belly [3, 19, 30], but the distal division pattern of the nerve has not been studied. The purpose of the study was to analyze the distal branching pattern of the motor trunks of the MCN in an attempt to develop guidelines for hyperselective neurectomy.
incision on the anterior aspect of the arm from the pectoralis major tendon to the elbow crease along the bicipital sulcus. The skin, subcutaneous fat, and fascia were retracted to expose the MCN. The nerve was dissected using magnification loupes (92.3 HeineÒ) and microsurgery instruments. Each branch coming from the nerve was dissected with all its subdivisions until it reached the muscle surface, without entering within the muscle fibers. Sensory branches were dissected till their ending in the deep aspect of the dermis. The number of primary branches (we will term them trunks) for the biceps brachii and the brachialis was recorded, for each trunk the number of branches, for each branch the number of terminal division branches (after one or several division levels). We recorded the distance from the coracoid process to each trunk, to the most distal terminal branch for each muscle and to the sensitive branches (lateral antebrachial cutaneous nerve and any accessory cutaneous sensory branch), using a metallic ruler accurate to 1 mm along the axis coracoid-lateral epicondyle. The level of branching was measured with the elbow flexed at 45°, the shoulder abducted 80° and in neutral rotation. The mean length of a trunk before division was noted. It was considered as ‘‘0’’ if the trunk was not dividing and entered the muscle directly. Besides measurements, photographs, and drawings were performed for each case. To make the dissections easier, the vascular pedicle of the biceps brachii was ligated in all but two cases. In two cases, we performed a dissection of the fascicles of the MCN under microscope, from the emergence of the biceps brachii trunks to the lateral antebrachial cutaneous nerve.
Results The average arm length was 315.3 mm (from 280 to 345 mm). The nerve descended between the biceps brachii and the brachialis muscles. Then it travelled distally along the lateral border of the biceps brachii muscle as the lateral antebrachial cutaneous nerve. The motor branches to the biceps brachii muscle (Table 1)
Materials and methods Sixteen fresh cadaver arms were dissected (one limb per cadaver). There were nine females and seven males. None of the specimen cadaver had signs of previous trauma or operation on the arms. The mean age of the cadavers was 75 years old (range from 65 to 90). The arm length was measured from the tip of the coracoid process to the tip of the lateral epicondyle. The limbs were dissected through a longitudinal
123
They run laterally from the MCN. In nine cases (56.2 %) only one trunk exited from the main nerve trunk. In three subjects, two trunks were found (18.8 %); in two cases there were three trunks (12.5 %), in one case four trunks (6.3 %) (Fig. 1) and in one case five trunks (6.3 %). In the four arms with more than two trunks dedicated to the biceps brachii, the last trunk was usually just a thin fiber. This fiber continued with only one terminal branch in three
69.3 % (59.7 to 75 %) 62 % (59.4–70.3 %) 51.7 % (35.3–62.5 %) 55.7 % (29–64.3 %) Average distance from coracoid process to emergence of nerves as a percentage of coracoid process-lateral epicondyle distance
37.1 % (17.9–45.3 %)
52 % (42.2–59.7 %)
Last trunk in case of multiple trunks First trunk or single trunk Last terminal branch Last trunk in case of multiple trunks First trunk or single trunk
Motor nerves to brachialis Motor nerves to biceps brachii
Table 1 Location of the motor branches of the musculocutaneous nerve to the biceps brachii and brachialis muscles in the present study
Last terminal branch
Surg Radiol Anat
cases. The average distance of origin of the first trunk was 37.1 % of the total length of the arm (range 17.9–45.3 %). It then split into two or three secondary branches to supply each area of the muscle. In cases with multiple trunks, the mean distance between the first and the last trunk was 34.6 mm (from 10 to 68 mm), and the average distance of origin of the last trunk was 52 % of the total length. In four cases (25 % of patients), the last trunk exited the nerve more distally than the motor branches to the brachialis muscle (Fig. 2). These distal trunks exited the nerve at a mean length of 56.5 % of the total arm length (range 52.2–61.4 %). They continued in only one terminal branch in three cases and two terminal branches in one case. The mean length of the first trunk before division was 12.7 mm (range 5–30 mm). The most distal terminal branch to exit from the secondary branches of division of the last trunk was located at an average distance of 55.7 % of the total length (range 29–64.3 %). It exited the nerve at 54.9 % of the arm length in case of unique trunk, 55.9 % in case of two trunks, 57.8, 56.3, or 58.1 %, respectively, in case of three, four or five trunks. The mean number of terminal branches for the biceps brachii was 7.9 (range 2–13). The motor branches to the brachialis muscle (Table 1) They exited from the medial aspect of the MCN in ten cases, from its lateral aspect in five cases and on both sides in one case. A single trunk was dedicated to the brachialis in 12 cases (75 %). In two cases, there were two trunks (12.5 %) and in two there were three trunks (12.5 %) (Fig. 1). The mean length of the first trunk before division in several branches was 20 mm (range 0–75 mm). The first trunk arose from the nerve at a mean length of 51.7 % of the total length of the arm (range 35.3–62.5 %). In cases with several trunks dedicated to the brachialis, the last trunk arose at a mean length of 62 % of the arm length (range 59.4–70.3 %). The most distal terminal branch exited at 69.3 % (range 59.7–75 %). It exited at 69 % in cases of one trunk, 68 % in cases of two trunks, and 72.7 % in cases of three trunks. The mean number of terminal branches for the brachialis was 6.5 (range 3–13). The sensory branch The terminal sensory branch was identified at a mean distance of 55.7 % of the total length (range 43.5–69.4 %). An accessory sensory branch dedicated to the skin of the lateral aspect of the elbow arose from the sensory branch in 12 cases (75 % of cadavers), at a mean distance of 75.8 % (range 56.3–91.3 %) (Figs. 1, 2).
123
Surg Radiol Anat
Fig. 1 Anterior view of a right arm. Four trunks emerge from the musculocutaneous nerve toward the biceps brachii muscle and divide into 11 terminal branches (yellow stars). The brachialis muscle is innervated by three trunks and nine terminal branches (orange stars).
Note an accessory cutaneous sensory branch (white arrow) that exits the nerve at 82 % of the arm length (coracoid process-lateral epicondyle distance). MCN musculocutaneous nerve, LACN lateral antebrachial cutaneous nerve (color figure online)
Fig. 2 Anterior view of a left arm after arterial and veinous latex injection. Two trunks are dedicated to the biceps brachii, the second one exiting the musculocutaneous nerve 6.8 cm distally and downstream to the trunk of the brachialis muscle. Biceps brachii and brachialis muscles are innervated respectively by twelve (yellow
stars) and seven terminal branches (orange stars). An accessory cutaneous sensory branch (white arrow) leaves the lateral antebrachial cutaneous nerve at 69 % of the arm length. MCN musculocutaneous nerve, LACN lateral antebrachial cutaneous nerve (color figure online)
Anatomic variations with the median nerve
Discussion
Four limbs showed a communication between the MCN and the median nerve. In one of these, the median nerve gave a branch for the MCN downstream the branches for the brachialis. In two cases, the lateral root of the median nerve ran in the MCN and leaved it distally to the coracobrachialis muscle, respectively, at 47 and 52 % of the brachial segment length to join the median nerve. In one case, there was no MCN and the trunks for the biceps brachii, the brachialis and the lateral antebrachial cutaneous nerve exited directly from the median nerve (Fig. 3).
Our study showed that the distal motor branching pattern of the MCN was highly variable. In more than 50 % of cases, the biceps brachii muscle was innervated by only one trunk. But we observed several limbs with four or even five trunks dedicated to the biceps brachii, thus showing a difference with the existing relevant literature. Chiarapattanakom et al. [5] never found more than three trunks in 112 dissections of embalmed upper limbs. The branching pattern of the biceps brachii muscle is usually described using the classification of Yang [45]. Type I, with only one motor branch for the biceps brachii is the most common configuration, accounting for 55–90 % of dissections in the literature [5, 14, 17, 19, 25, 28, 45]. In the type II, two separate branches supply each head of the muscle. In the type III, there are also two separate trunks, one that splits into two branches with one for each head, and a more distal
Intraneural topography of the MCN The MCN main trunk showed a plexus-like organization at the middle third of the arm (Fig. 4) with interconnections between sensory and motor fascicles.
123
Surg Radiol Anat
Fig. 3 Anterior view of a right arm showing absence of musculocutaneous nerve. The trunk for the biceps brachii, and a trunk brachialis/ lateral antebrachial cutaneous nerve exit directly from the median
nerve. MCN musculocutaneous nerve. White arrow median nerve, yellow star terminal branch for the biceps brachii muscle, orange star terminal branch for the brachialis muscle (color figure online)
Fig. 4 Dissection of fascicles of the musculocutaneous nerve at the level of the middle third of the arm (right arm). MCN musculocutaneous nerve
one for the common belly. For the sake of comparison of studies concerning the number of trunks, we gathered the cases described as type II and III as both of them exhibit two trunks (Table 2). In the present study, we did not analyze the motor supply of each head, as we did not individualize each head of the biceps brachii during the dissections. The first reason is because the boundary of the short head, the long head, and the common belly is less visible in a fresh cadaver than in embalmed specimen, due to the absence of shrinkage of intermuscular fatty tissue [29]. Secondly, we used only the approach that would be performed in a clinical situation, when the denervation is function of the number of branches and not of their location. In the same way, we found three trunks innervating the brachialis muscle in two cases, when no study recorded more than two trunks. Interestingly, these cases showed also a complex innervation pattern of the biceps brachii with three or four trunks. In almost all our specimens, the brachialis muscle received only one trunk from the MCN, which then divided in several secondary branches on both sides. The main pattern of innervation of the brachialis was
shown in several publications to be only one trunk [5, 14, 17, 19, 28, 45] with 55.4–100 % of cases. In all the other cases, the muscle received two trunks. One explanation of the high number of trunks we observed in several cadavers could be that we made a precise dissection on fresh adult subjects and not fetuses or embalmed cadavers. These original cases are most probably very rare occurrences, and possibly more frequent in Caucasian than in Asian populations as in Chiarapattanakom study [5]. The small amount of arms we dissected makes it impossible to calculate the prevalence of anatomical variations. Another rare variation we observed was that the last trunk for the biceps brachii exited the main nerve distal to the trunk to the brachialis in four cases, which was never observed in the Chiarapattanakom study [5], but had been earlier described in 2/20 cadavers by Oberlin et al. [25]. The brachialis muscle is also known to receive nerve supply from the radial nerve at its inferolateral region, up in 81.6 % of cases [22] with 11 % of relative innervation [1], but we did not dissect the radial nerve, as this would probably lead to a considerably higher risk of iatrogenicity of the procedure for little potential benefit.
123
Surg Radiol Anat Table 2 Branching patterns of the biceps brachii and the brachialis muscles according to the literature References
Number of dissections
Status of cadavers
Motor trunks for the biceps brachii
Motor trunks for the brachialis
One (%)
Two (%)
Three (%)
Four or five (%)
One (%)
Two (%)
Three (%) 0
Chiarapattanakom et al. [5]
112
Embalmed
62
33
5
0
92
8
Yang et al. [45]
24
Fresh-frozen
83.4
16.6
0
0
95.8
4.2
0
Kervancioglu et al. [14]
20
Formalin-fixed foetuses (12–36 weeks of gestation)
80
20
0
0
80
20
0
Kwolczak-Mc Grath et al. [17]
40
Formalin-fixed foetuses (16–27 weeks of gestation)
90
10
0
0
100
0
0
Oberlin et al. [25]
20
Fresh
55
45
0
0
Pacha Vicente et al. [28] Lee et al. [19]
46 56
24 fresh-frozen, 22 embalmed Fresh
60.5 57.1
39.5 42.9
0 0
0 0
72.1 55.4
27.9 44.6
0 0
Present study
16
Fresh
56.2
18.8
12.5
12.5
75
12.5
12.5
We decided to study only one limb of each cadaver because Chiarapattanakom showed that the branching pattern of the MCN was similar in the two limbs of the same subject in 71 % of cases [5]. We chose an intermediary position of elbow flexion [28], with the shoulder abducted 80° in neutral rotation to do our measurements, as this is the standard position during surgery, allowing a good relaxation of biceps brachii muscles fibers. We checked that the distance between the coracoid process and the motor trunks did not vary more than 5 mm during elbow flexion in two cadavers. All the distances were reported as a percentage of the total length of the arm to eliminate any inter-individual variation. An accessory sensory nerve branch, that had never been described, was observed in 25 % of cases. The most proximal one originated from the main trunk at 56.3 % of the arm length. It could have been mistaken for a motor branch for the brachialis or the biceps brachii if the dissection had not been conducted as far as the subcutaneous tissue. In case of high accessory sensory branch, a nerve transfer to the first trunk exiting from the main nerve distally to the trunks to the biceps brachii would lead to a failure. Many authors described the communications between the MCN and the median nerve [6, 12, 16, 20, 27]. Their incidence has been reported to vary from 5 to 46.4 %, with an average of 33 % [6]. This is in accordance with the present study, in which communicating branches between the MCN and the median nerve were identified in 25 % of subjects (4/16). Many different communicating patterns have been described. Le Minor et al. [20] has classified them in five types: in type I, there is no communicating fiber between the MCN and the median nerve; in type II some medial root fibers run in the MCN and leave it after a
123
distance to join the median nerve; in type III, the lateral root of the median nerve runs in the MCN and leaves it after a distance to join the median nerve; in type IV, the fibers of the MCN are united with the lateral root of the median nerve and, after passing some distance into the upper arm, the MCN arises from the median nerve in the upper arm; and in type V the fibers of the MCN run within the median nerve along its course in the normal path of the median nerve. Of our five communicating branches, two were under type III, and one under type V (Fig. 3). In the fourth one, some lateral root fibers running in the median nerve were leaving it after a distance to join the MCN, which was not described in Le Minor’s classification. This variation was also observed in five limbs in Chiarapattanakom’s series [5] and 3/44 in Krishnamurthy’s series [16]. The absence of an independent MCN with nerve fibers emerging directly from the median nerve was rarely reported in literature (1.5–15 %) [32, 44]. In our case, the motor and sensory branches of the MCN shared the same conjonctive sheath as the median nerve (Fig. 3), as also reported by Fregnani et al. [10]. Most of the previous related anatomical studies focused on the number of main trunks exiting from the nerve and not on the distal branching pattern. Their results are most useful for nerve transfers, in which a normal (donor) nerve is sutured to the distal end of a damaged (recipient) nerve, as near as possible to the effectors, but before the final division of the trunk in the muscle. For instance, a fascicle of the ulnar nerve is sutured to a trunk dedicated to the biceps brachii muscle [25] and a fascicule from the median nerve is sutured to a trunk to the brachialis muscle [35, 43]. It is noteworthy, however, that none of the previous articles describe precisely how the recipient trunk is chosen in case of multiple recipient trunks. In view of our findings, some
Surg Radiol Anat
insufficient clinical results could be explained by the reinnervation of only a part of the target muscle. Other studies assessed only the localization of the intramuscular motor points of the biceps brachii and the brachialis to develop guidelines for percutaneous Botulinum toxin block injections [3, 19, 30]. Lee et al. observed that the intramuscular motor points of the biceps brachii were located between 44.6 and 76 % of the distance from the coracoid process to the medial epicondyle [19]. The motor points of the brachialis were located between 64 and 86.6 % of the same distance. Moreover, Lee noticed that the location of the entry points of the motor branches was independent of the branching pattern according to Yang classification [45]. Selective neurectomy (also termed ‘‘selective neurotomy’’ by some authors [4, 21, 24, 40, 41] consists in partial and selective resection of the motor branch(es) of a nerve to reduce the spastic tone of the target muscle(s). Some promising results have been reported with this procedure, including for the motor branches of the MCN to the biceps brachii and the brachialis muscle [4, 21, 34]. However, there are two levels of controversy regarding this technique. The first discussion involves the technique itself. One method consists in approaching only the main trunk of the nerve, and with intra-operative electrostimulation and intra-neuro-dissection identifying the motor branches and selecting the amount of fascicles which will be removed [33, 34, 37]. This technique, which avoids extensive dissection, has its limitations, partly because of the plexiform intra-neural anatomy (observed in our study and similar to Sunderland’s findings [42]), leading to a rather empirical choice of fascicles to resect. This may lead to a very heterogenous distribution of denervated muscle fibers, leaving in place large untouched areas of the target muscle(s) with insufficient results, and to sensory problems, as reported by several authors [21, 23, 36]. The second method, more invasive and more demanding technically, follows each motor trunk and each of its division all the way to its entry point into the involved muscle. Partial neurectomy, which we have termed ‘‘hyperselective’’, is then performed on each terminal branch at this level, by resecting a few millimetres of the required amount of fascicles. Sindou [40] recommends coagulating the proximal stump to prevent regrowth of the fibers. This technique allows making a harmonious, extremely selective, partial denervation of each area of the target muscle(s), while preserving the sensory branches. Magnifying loops and intra-operative nerve stimulation are both recommended for this technique. Perioperative electromyography [41] may also be applied for identifying the branches and particularly for differentiating between the motor and sensory nerves. The second problem concerns the amount of nerves fibers that should be resected. In his princeps paper, Brunelli [2] advised to
resect at least 50 % of the fascicles of each motor branch. But he later described the ‘‘adoption’’ (or ‘‘take-over’’) phenomenon, whereby the ‘‘orphan’’ muscle fibers are adopted by adjacent nerve branches, leading to some recurrence, and then advised to resect two-thirds of the fibers. On the other hand, excessive resection of the motor branches may lead to permanent paresis of the target muscles. Direct visualization of each and all of the terminal motor branches allows a much more precise quantification of the neurectomy. The present study showed that there is a mean distance of 57 mm between the origin of the first trunk and the last terminal branch dedicated to the biceps brachii, and an average of 56 mm between the first trunk to the brachialis muscle and the origin of the last terminal branch. The most efficient approach for this procedure of hyperselective neurectomy, requiring identification of all motor branches of the MCN to the biceps brachii and brachialis muscles, is longitudinal, over the medial bicipital groove. The superficial fascia and the biceps brachii perimysium are opened and the dissection proceeds directly down to the MCN, running longitudinally between the biceps, above, and the brachialis, underneath. According to our findings, we recommend to conduct the nerve dissection between 18 % (proximal) and 75 % (distal) of the distance between the coracoid process and the lateral epicondyle. This should allow proper identification of all the motor trunks and terminal branches dedicated to the biceps brachii and to the brachialis muscles, while sparing both sensory branches and communicating branches with the median nerve. Acknowledgments We thank all the staff of the anatomy laboratory of the Ecole de Chirurgie du Fer a` Moulin (Paris) for its precious help. Conflict of interest The authors have no conflicts of interest or financial ties to disclose.
References 1. Bendersky M, Bianchi HF (2012) Double innervation of the brachialis muscle: anatomic-physiological study. Surg Radiol Anat 34:865–870 2. Brunelli G, Brunelli F (1983) Partial selective denervation in spastic palsies (hyponeurotization). Microsurgery 4:221–224 3. Buchanan TS, Erickson JC (1996) Selective block of the brachialis motor point. An anatomic investigation of musculocutaneous nerve branching. Reg Anesth 21:89–92 4. Buffenoir K, Rigoard P, Ferrand-Sorbets S, Lapierre F (2009) Etude re´trospective du re´sultat a` long terme des neurotomies se´lectives pe´riphe´riques dans le traitement du membre supe´rieur spastique. Neurochirurgie 55(Suppl 1):50–60 5. Chiarapattanakom P, Leechavengvongs S, Witoonchart K, Uerpairojkit C, Thuvasethakul P (1998) Anatomy and internal topography of the musculocutaneous nerve: the nerves to the biceps and brachialis muscle. J Hand Surg Am 23:250–255
123
Surg Radiol Anat 6. Choi D, Rodrı´guez-Niedenfu¨hr M, Va´zquez T, Parkin I, San˜udo JR (2002) Patterns of connections between the musculocutaneous and median nerves in the axilla and arm. Clin Anat 15:11–17 7. Decq P, Cuny E, Filipetti P, Fe`ve A, Ke´ravel Y (1998) Les neurotomies pe´riphe´riques dans le traitement de la spasticite´. Indications, technique et re´sultats aux membres infe´rieurs. Neurochirurgie 44:175–182 8. Eglseder WA Jr, Goldman M (1997) Anatomic variations of the musculocutaneous nerve in the arm. Am J Orthop 26:777–780 9. Fe`ve A, Decq P, Filipetti P, Ke´ravel Y (1998) Traitement de la spasticite´ par injections de toxine botulique. Neurochirurgie 44(3):192–196 10. Fregnani JHTG, Mace´a MIM, Pereira CSB, Barros MD, Mace´a JR (2008) Absence of the musculocutaneous nerve: a rare anatomical variation with possible clinical-surgical implications. Sao Paulo Med J 126:288–290 11. Garland DE, Thompson R, Waters RL (1980) Musculocutaneous neurectomy for spastic elbow flexion in non-functional upper extremities in adults. J Bone Joint Surg Am 62:108–112 12. Guerri-Guttenberg RA, Ingolotti M (2009) Classifying musculocutaneous nerve variations. Clin Anat 22:671–683 13. Keenan MA, Tomas ES, Stone L, Gerste´n LM (1990) Percutaneous phenol block of the musculocutaneous nerve to control elbow flexor spasticity. J Hand Surg Am 15:340–346 14. Kervancioglu P, Orhan M, Kilinc N (2011) Patterns of motor branching of the musculocutaneous nerve in human fetuses and clinical significance. Clin Anat 24:168–178 15. Kong KH, Chua KS (1999) Neurolysis of the musculocutaneous nerve with alcohol to treat poststroke elbow flexor spasticity. Arch Phys Med Rehabil 80:1234–1236 16. Krishnamurthy A, Nayak SR, Venkatraya Prabhu L, Hegde RP, Surendran S, Kumar M, Pai MM (2007) The branching pattern and communications of the musculocutaneous nerve. J Hand Surg Eur 32:560–562 17. Kwolczak-Mc Grath A, Kolesnik A, Ciszek B (2008) Anatomy of branches of the musculocutaneous nerve to the biceps and brachialis in human fetuses. Clin Anat 21:142–146 18. Lance JW (1980) Symposium synopsis. In: Feldman RG, Young RR, Koella WP (eds) Spasticity: disorder of motor control. Year Book Medical Publishers, Chicago, pp 485–494 19. Lee JH, Kim HW, Im S, An X, Lee MS, Lee UY, Han SH (2010) Localization of motor entry points and terminal intramuscular nerve endings of the musculocutaneous nerve to biceps and brachialis muscles. Surg Radiol Anat 32:213–220 20. Le Minor JM (1990) A rare variation of the median and musculocutaneous nerves in man. Arch Anat Histol Embryol 73:33–42 21. Maarrawi J, Mertens P, Luaute J, Vial C, Chardonnet N, Cosson M, Sindou M (2006) Long-term functional results of selective peripheral neurotomy for the treatment of spastic upper limb: prospective study in 31 patients. J Neurosurg 104:215–225 22. Mahakkanukrauh P, Somsarp V (2002) Dual innervation of the brachialis muscle. Clin Anat 15:206–209 23. Matsuo T, Tada S, Hajime T (1986) Insufficiency of the hip adductor after anterior obturator neurectomy in 42 children with cerebral palsy. J Pediatr Orthop 6:686–692 24. Msaddi AK, Mazroue AR, Shahwan S, al Amri N, Dubayan N, Livingston D, Moutaery KR (1997) Microsurgical selective peripheral neurotomy in the treatment of spasticity in cerebralpalsy children. Stereotact Funct Neurosurg 69:251–258 25. Oberlin C, Be´al D, Leechavengvongs S, Salon A, Dauge MC, Sarcy JJ (1994) Nerve transfer to biceps muscle using a part of ulnar nerve for C5-C6 avulsion of the brachial plexus: anatomical study and report of four cases. J Hand Surg Am 19:232–237
123
26. Ozturk A, Bayraktar B, Taskara N, Kale AC, Kutlu C, Cecen A (2005) Morphometric study of the nerves entering into the coracobrachialis muscle. Surg Radiol Anat 27:308–311 ¨ ztu¨rk NC, Uzmansel D, O ¨ ztu¨rk H (2010) An unreported pattern 27. O of musculocutaneous and median nerve communication with multiple variations of biceps brachii: a case report. Surg Radiol Anat 32:887–890 28. Pacha Vicente D, Forcada Calvet P, Carrera Burgaya A, Llusa´ Pe´rez M (2005) Innervation of biceps brachii and brachialis: anatomical and surgical approach. Clin Anat 18:186–194 29. Paik DJ, Shin SY (2009) An anatomical study of the inferior oblique muscle: the embalmed cadaver vs the fresh cadaver. Am J Ophthalmol 147:544–549 30. Park BK, Shin YB, Ko HY, Park JH, Baek SY (2007) Anatomic motor point localization of the biceps brachii brachii and brachialis muscles. J Korean Med Sci 22:459–462 31. Penn RD, Kroin JS (1985) Continuous intrathecal baclofen for severe spasticity. Lancet 2:125–127 32. Prasada Rao PC, Chaudhary SC (2001) Absence of musculocutaneous nerve: two case reports. Clin Anat 14:31–35 33. Puligopu AK, Purohit AK (2011) Outcome of selective motor fasciculotomy in the treatment of upper limb spasticity. J Pediatr Neurosci 6(Suppl 1):118–125 34. Purohit AK, Raju BS, Kumar KS, Mallikarjun KD (1998) Selective musculocutaneous fasciculotomy for spastic elbow in cerebral palsy: a preliminary study. Acta Neurochir 140:473–478 35. Ray WZ, Pet MA, Yee A, Mackinnon SE (2011) Double fascicular nerve transfer to the biceps and brachialis muscles after brachial plexus injury: clinical outcomes in a series of 29 cases. J Neurosurg 114:1520–1528 36. Rousseaux M, Buisset N, Daveluy W, Kozlowski O, Blond S (2009) Long-term effect of tibial nerve neurotomy in stroke patients with lower limb spasticity. J Neurol Sci 278:71–76 37. Shin DK, Jung YJ, Hong JC, Kim MS, Kim SH (2010) Selective musculocutaneous neurectomy for spastic elbow. J Korean Neurosurg Soc 48:236–239 38. Simpson DM, Alexander DN, O’Brien CF, Tagliati M, Aswad AS, Leon JM (1996) Botulinum toxin type A in the treatment of upper extremity spasticity: a randomized, double blind, placebocontrolled trial. Neurology 46:1306–1310 39. Sindou MP (2005) History of the neurosurgical treatment for spasticity. Oper Tech Neurosurg 7:96–99 40. Sindou MP, Simon F, Mertens P, Decq P (2007) Selective peripheral neurotomy (SPN) for spasticity in childhood. Child Nerv Syst 23:957–970 41. Sitthinamsuwan B, Chanvanitkulchai K, Phonwijit L, Ploypetch T, Kumthornthip W, Nunta-Aree S (2013) Utilization of intraoperative electromyography for selecting targeted fascicles and determining the degree of fascicular resection in selective tibial neurotomy for ankle spasticity. Acta Neurochir 155:1143–1149 42. Sunderland S (1945) The intraneural topography of the radial, median, and ulnar nerves. Brain 68:243–299 43. Teboul F, Kakkar R, Ameur N, Beaulieu JY, Oberlin C (2004) Transfer of fascicles from the ulnar nerve to the nerve to the biceps in the treatment of upper brachial plexus palsy. J Bone Joint Surg Am 86-A:1485–1490 44. Uzel AP, Bulla A, Steinmann G, LaurentJoye M, Caix P (2011) Absence du nerf musculocutane´ et distribution a` partir du nerf median: a` propos de deux cas et revue de la litte´rature. Morphologie 95:146–150 45. Yang ZX, Pho RW, Kour AK, Pereira BP (1995) The musculocutaneous nerve and its branches to the biceps and brachialis muscles. J Hand Surg Am 20:671–675
ORIGINAL ARTICLE
Anatomical study of the musculocutaneous nerve branching pattern: application for selective neurectomy in the treatment of elbow flexors spasticity Adeline Cambon-Binder • Caroline Leclercq
Received: 10 February 2014 / Accepted: 30 August 2014 Ó Springer-Verlag France 2014
Abstract Purpose Spastic flexion deformity of the elbow is mainly mediated by the biceps brachii and the brachialis muscles, innervated by the musculocutaneous nerve. Selective neurectomy of the musculocutaneous nerve showed promising results to relieve excessive spasticity in the long term but lacks of a consensual surgical strategy. The aim of the study was to describe the distal branching pattern of the motor branches of the musculocutaneous nerve in an attempt to develop guidelines for surgery. Methods Sixteen arms of fresh cadaver specimen were dissected. We recorded the site of each primary and terminal motor branch as a percentage of the distance from the coracoid process to the lateral epicondyle. Results The biceps muscle was innervated by one to five primary motor branches. The first branch emerged from the nerve at an average of 37.1 % of the arm length, and the most distal terminal branch at 55.7 %. The brachialis muscle received one to three primary branches. The first branch exited the nerve at an average of 51.7 % of the arm length and the last terminal branch at 69.3 %. The average number of terminal branches dedicated to the biceps and the brachialis muscles were, respectively, 7.9 and 6.5. Conclusions According to our findings, we recommend to dissect the musculocutaneous nerve between 18 and 75 % of the distance between the coracoid process and the lateral epicondyle to identify the motor terminal branches to the A. Cambon-Binder (&) Hand and Peripheral Nerve Surgery Unit, European Georges Pompidou Hospital, 20 rue Leblanc, 75015 Paris, France e-mail: [email protected] C. Leclercq Institut de la Main, Clinique Jouvenet, 6 square Jouvenet, 75016 Paris, France
biceps brachii and the brachialis muscle, sparing sensory branches. Keywords Musculocutaneous nerve Selective neurectomy Spasticity Selective neurotomy
Introduction Spasticity has been defined as an increase in muscle tone as a result of the hyper-excitability of the muscle fibers and is characterized by a velocity-dependent increase in tonic stretch reflexes [18]. Spasticity is one of the clinical features of the upper motor neuron syndrome resulting from various etiologies such as cerebrovascular injury, traumatic brain injury, multiple sclerosis, and cerebral palsy [7]. In the upper limb, spasticity can lead to pain, impairing deformities interfering with activities of daily living and hygiene, and poor cosmesis. Inhibition of antagonist muscles may mask their residual motor function and induce functional impairment [7, 11]. Common manifestations of spastic disorders involving the upper limb include internal rotation and adduction of the shoulder, flexion of the elbow, pronation of the forearm, flexion and ulnar deviation of the wrist, flexion of the fingers and flexion-adduction of the thumb [11]. Spasticity of the elbow is mediated mainly by the biceps brachii and the brachialis muscles, innervated by the musculocutaneous nerve (MCN). The MCN arises from the lateral cord, from the fifth to seventh cervical roots, at the level of the distal border of the pectoralis major muscle. It passes through the coracobrachialis muscle [26] and travels distally in the lateral aspect of the arm between the biceps brachii and brachialis muscles [8]. Below the elbow and lateral to the tendon of
123
Surg Radiol Anat
the biceps brachii, it pierces the deep fascia and continues as the lateral antebrachial cutaneous nerve. The MCN provides motor innervation to the flexor compartment of the arm and sensory innervation to the radial aspect of the forearm. Many nonsurgical treatments were developed for elbow flexors spasticity but are characterized by their short-term effect. Intrathecal injection of baclofen administered via a programmable pump may provide a continuous spasmreducing effect in cases of regional spasticity [31]. Phenol nerve blocks [13, 15] provide temporary relief of painful contractures and allow for a period of spontaneous neurologic recovery of up to 6 months. Botulinum toxin injected into the muscle body provides a three to four months relief [9, 38]. In an attempt to relieve spastic symptoms in a permanent way, surgical procedures have been developed at the level of peripheral nerves, spinal roots, spinal cord, and the dorsal root entry zone [39]. Microsurgical selective peripheral neurectomy—formerly termed hyponeurotization [2]—consists in a partial and selective excision of the peripheral nerves. It has proven to be effective for some localized forms of spasticity [4, 21, 24, 34] involving muscles innervated by the musculocutaneous, the median, and the ulnar nerves. However, there is controversy in the technical strategy regarding the exact site and amount of nerve resection: main nerve approach and selection of fascicles using electrostimulation, or dissection of the motor branches till their entrance in the muscle and resection of a proportion of the terminal branches. This latter procedure requires a thorough knowledge of the branching pattern of the MCN for the biceps brachii and the brachialis muscles. As far as we know, previous anatomy articles have focused on the number of motor trunks [5, 14, 17, 25, 28, 45], or on the center of the location where the motor branches enter the muscle belly [3, 19, 30], but the distal division pattern of the nerve has not been studied. The purpose of the study was to analyze the distal branching pattern of the motor trunks of the MCN in an attempt to develop guidelines for hyperselective neurectomy.
incision on the anterior aspect of the arm from the pectoralis major tendon to the elbow crease along the bicipital sulcus. The skin, subcutaneous fat, and fascia were retracted to expose the MCN. The nerve was dissected using magnification loupes (92.3 HeineÒ) and microsurgery instruments. Each branch coming from the nerve was dissected with all its subdivisions until it reached the muscle surface, without entering within the muscle fibers. Sensory branches were dissected till their ending in the deep aspect of the dermis. The number of primary branches (we will term them trunks) for the biceps brachii and the brachialis was recorded, for each trunk the number of branches, for each branch the number of terminal division branches (after one or several division levels). We recorded the distance from the coracoid process to each trunk, to the most distal terminal branch for each muscle and to the sensitive branches (lateral antebrachial cutaneous nerve and any accessory cutaneous sensory branch), using a metallic ruler accurate to 1 mm along the axis coracoid-lateral epicondyle. The level of branching was measured with the elbow flexed at 45°, the shoulder abducted 80° and in neutral rotation. The mean length of a trunk before division was noted. It was considered as ‘‘0’’ if the trunk was not dividing and entered the muscle directly. Besides measurements, photographs, and drawings were performed for each case. To make the dissections easier, the vascular pedicle of the biceps brachii was ligated in all but two cases. In two cases, we performed a dissection of the fascicles of the MCN under microscope, from the emergence of the biceps brachii trunks to the lateral antebrachial cutaneous nerve.
Results The average arm length was 315.3 mm (from 280 to 345 mm). The nerve descended between the biceps brachii and the brachialis muscles. Then it travelled distally along the lateral border of the biceps brachii muscle as the lateral antebrachial cutaneous nerve. The motor branches to the biceps brachii muscle (Table 1)
Materials and methods Sixteen fresh cadaver arms were dissected (one limb per cadaver). There were nine females and seven males. None of the specimen cadaver had signs of previous trauma or operation on the arms. The mean age of the cadavers was 75 years old (range from 65 to 90). The arm length was measured from the tip of the coracoid process to the tip of the lateral epicondyle. The limbs were dissected through a longitudinal
123
They run laterally from the MCN. In nine cases (56.2 %) only one trunk exited from the main nerve trunk. In three subjects, two trunks were found (18.8 %); in two cases there were three trunks (12.5 %), in one case four trunks (6.3 %) (Fig. 1) and in one case five trunks (6.3 %). In the four arms with more than two trunks dedicated to the biceps brachii, the last trunk was usually just a thin fiber. This fiber continued with only one terminal branch in three
69.3 % (59.7 to 75 %) 62 % (59.4–70.3 %) 51.7 % (35.3–62.5 %) 55.7 % (29–64.3 %) Average distance from coracoid process to emergence of nerves as a percentage of coracoid process-lateral epicondyle distance
37.1 % (17.9–45.3 %)
52 % (42.2–59.7 %)
Last trunk in case of multiple trunks First trunk or single trunk Last terminal branch Last trunk in case of multiple trunks First trunk or single trunk
Motor nerves to brachialis Motor nerves to biceps brachii
Table 1 Location of the motor branches of the musculocutaneous nerve to the biceps brachii and brachialis muscles in the present study
Last terminal branch
Surg Radiol Anat
cases. The average distance of origin of the first trunk was 37.1 % of the total length of the arm (range 17.9–45.3 %). It then split into two or three secondary branches to supply each area of the muscle. In cases with multiple trunks, the mean distance between the first and the last trunk was 34.6 mm (from 10 to 68 mm), and the average distance of origin of the last trunk was 52 % of the total length. In four cases (25 % of patients), the last trunk exited the nerve more distally than the motor branches to the brachialis muscle (Fig. 2). These distal trunks exited the nerve at a mean length of 56.5 % of the total arm length (range 52.2–61.4 %). They continued in only one terminal branch in three cases and two terminal branches in one case. The mean length of the first trunk before division was 12.7 mm (range 5–30 mm). The most distal terminal branch to exit from the secondary branches of division of the last trunk was located at an average distance of 55.7 % of the total length (range 29–64.3 %). It exited the nerve at 54.9 % of the arm length in case of unique trunk, 55.9 % in case of two trunks, 57.8, 56.3, or 58.1 %, respectively, in case of three, four or five trunks. The mean number of terminal branches for the biceps brachii was 7.9 (range 2–13). The motor branches to the brachialis muscle (Table 1) They exited from the medial aspect of the MCN in ten cases, from its lateral aspect in five cases and on both sides in one case. A single trunk was dedicated to the brachialis in 12 cases (75 %). In two cases, there were two trunks (12.5 %) and in two there were three trunks (12.5 %) (Fig. 1). The mean length of the first trunk before division in several branches was 20 mm (range 0–75 mm). The first trunk arose from the nerve at a mean length of 51.7 % of the total length of the arm (range 35.3–62.5 %). In cases with several trunks dedicated to the brachialis, the last trunk arose at a mean length of 62 % of the arm length (range 59.4–70.3 %). The most distal terminal branch exited at 69.3 % (range 59.7–75 %). It exited at 69 % in cases of one trunk, 68 % in cases of two trunks, and 72.7 % in cases of three trunks. The mean number of terminal branches for the brachialis was 6.5 (range 3–13). The sensory branch The terminal sensory branch was identified at a mean distance of 55.7 % of the total length (range 43.5–69.4 %). An accessory sensory branch dedicated to the skin of the lateral aspect of the elbow arose from the sensory branch in 12 cases (75 % of cadavers), at a mean distance of 75.8 % (range 56.3–91.3 %) (Figs. 1, 2).
123
Surg Radiol Anat
Fig. 1 Anterior view of a right arm. Four trunks emerge from the musculocutaneous nerve toward the biceps brachii muscle and divide into 11 terminal branches (yellow stars). The brachialis muscle is innervated by three trunks and nine terminal branches (orange stars).
Note an accessory cutaneous sensory branch (white arrow) that exits the nerve at 82 % of the arm length (coracoid process-lateral epicondyle distance). MCN musculocutaneous nerve, LACN lateral antebrachial cutaneous nerve (color figure online)
Fig. 2 Anterior view of a left arm after arterial and veinous latex injection. Two trunks are dedicated to the biceps brachii, the second one exiting the musculocutaneous nerve 6.8 cm distally and downstream to the trunk of the brachialis muscle. Biceps brachii and brachialis muscles are innervated respectively by twelve (yellow
stars) and seven terminal branches (orange stars). An accessory cutaneous sensory branch (white arrow) leaves the lateral antebrachial cutaneous nerve at 69 % of the arm length. MCN musculocutaneous nerve, LACN lateral antebrachial cutaneous nerve (color figure online)
Anatomic variations with the median nerve
Discussion
Four limbs showed a communication between the MCN and the median nerve. In one of these, the median nerve gave a branch for the MCN downstream the branches for the brachialis. In two cases, the lateral root of the median nerve ran in the MCN and leaved it distally to the coracobrachialis muscle, respectively, at 47 and 52 % of the brachial segment length to join the median nerve. In one case, there was no MCN and the trunks for the biceps brachii, the brachialis and the lateral antebrachial cutaneous nerve exited directly from the median nerve (Fig. 3).
Our study showed that the distal motor branching pattern of the MCN was highly variable. In more than 50 % of cases, the biceps brachii muscle was innervated by only one trunk. But we observed several limbs with four or even five trunks dedicated to the biceps brachii, thus showing a difference with the existing relevant literature. Chiarapattanakom et al. [5] never found more than three trunks in 112 dissections of embalmed upper limbs. The branching pattern of the biceps brachii muscle is usually described using the classification of Yang [45]. Type I, with only one motor branch for the biceps brachii is the most common configuration, accounting for 55–90 % of dissections in the literature [5, 14, 17, 19, 25, 28, 45]. In the type II, two separate branches supply each head of the muscle. In the type III, there are also two separate trunks, one that splits into two branches with one for each head, and a more distal
Intraneural topography of the MCN The MCN main trunk showed a plexus-like organization at the middle third of the arm (Fig. 4) with interconnections between sensory and motor fascicles.
123
Surg Radiol Anat
Fig. 3 Anterior view of a right arm showing absence of musculocutaneous nerve. The trunk for the biceps brachii, and a trunk brachialis/ lateral antebrachial cutaneous nerve exit directly from the median
nerve. MCN musculocutaneous nerve. White arrow median nerve, yellow star terminal branch for the biceps brachii muscle, orange star terminal branch for the brachialis muscle (color figure online)
Fig. 4 Dissection of fascicles of the musculocutaneous nerve at the level of the middle third of the arm (right arm). MCN musculocutaneous nerve
one for the common belly. For the sake of comparison of studies concerning the number of trunks, we gathered the cases described as type II and III as both of them exhibit two trunks (Table 2). In the present study, we did not analyze the motor supply of each head, as we did not individualize each head of the biceps brachii during the dissections. The first reason is because the boundary of the short head, the long head, and the common belly is less visible in a fresh cadaver than in embalmed specimen, due to the absence of shrinkage of intermuscular fatty tissue [29]. Secondly, we used only the approach that would be performed in a clinical situation, when the denervation is function of the number of branches and not of their location. In the same way, we found three trunks innervating the brachialis muscle in two cases, when no study recorded more than two trunks. Interestingly, these cases showed also a complex innervation pattern of the biceps brachii with three or four trunks. In almost all our specimens, the brachialis muscle received only one trunk from the MCN, which then divided in several secondary branches on both sides. The main pattern of innervation of the brachialis was
shown in several publications to be only one trunk [5, 14, 17, 19, 28, 45] with 55.4–100 % of cases. In all the other cases, the muscle received two trunks. One explanation of the high number of trunks we observed in several cadavers could be that we made a precise dissection on fresh adult subjects and not fetuses or embalmed cadavers. These original cases are most probably very rare occurrences, and possibly more frequent in Caucasian than in Asian populations as in Chiarapattanakom study [5]. The small amount of arms we dissected makes it impossible to calculate the prevalence of anatomical variations. Another rare variation we observed was that the last trunk for the biceps brachii exited the main nerve distal to the trunk to the brachialis in four cases, which was never observed in the Chiarapattanakom study [5], but had been earlier described in 2/20 cadavers by Oberlin et al. [25]. The brachialis muscle is also known to receive nerve supply from the radial nerve at its inferolateral region, up in 81.6 % of cases [22] with 11 % of relative innervation [1], but we did not dissect the radial nerve, as this would probably lead to a considerably higher risk of iatrogenicity of the procedure for little potential benefit.
123
Surg Radiol Anat Table 2 Branching patterns of the biceps brachii and the brachialis muscles according to the literature References
Number of dissections
Status of cadavers
Motor trunks for the biceps brachii
Motor trunks for the brachialis
One (%)
Two (%)
Three (%)
Four or five (%)
One (%)
Two (%)
Three (%) 0
Chiarapattanakom et al. [5]
112
Embalmed
62
33
5
0
92
8
Yang et al. [45]
24
Fresh-frozen
83.4
16.6
0
0
95.8
4.2
0
Kervancioglu et al. [14]
20
Formalin-fixed foetuses (12–36 weeks of gestation)
80
20
0
0
80
20
0
Kwolczak-Mc Grath et al. [17]
40
Formalin-fixed foetuses (16–27 weeks of gestation)
90
10
0
0
100
0
0
Oberlin et al. [25]
20
Fresh
55
45
0
0
Pacha Vicente et al. [28] Lee et al. [19]
46 56
24 fresh-frozen, 22 embalmed Fresh
60.5 57.1
39.5 42.9
0 0
0 0
72.1 55.4
27.9 44.6
0 0
Present study
16
Fresh
56.2
18.8
12.5
12.5
75
12.5
12.5
We decided to study only one limb of each cadaver because Chiarapattanakom showed that the branching pattern of the MCN was similar in the two limbs of the same subject in 71 % of cases [5]. We chose an intermediary position of elbow flexion [28], with the shoulder abducted 80° in neutral rotation to do our measurements, as this is the standard position during surgery, allowing a good relaxation of biceps brachii muscles fibers. We checked that the distance between the coracoid process and the motor trunks did not vary more than 5 mm during elbow flexion in two cadavers. All the distances were reported as a percentage of the total length of the arm to eliminate any inter-individual variation. An accessory sensory nerve branch, that had never been described, was observed in 25 % of cases. The most proximal one originated from the main trunk at 56.3 % of the arm length. It could have been mistaken for a motor branch for the brachialis or the biceps brachii if the dissection had not been conducted as far as the subcutaneous tissue. In case of high accessory sensory branch, a nerve transfer to the first trunk exiting from the main nerve distally to the trunks to the biceps brachii would lead to a failure. Many authors described the communications between the MCN and the median nerve [6, 12, 16, 20, 27]. Their incidence has been reported to vary from 5 to 46.4 %, with an average of 33 % [6]. This is in accordance with the present study, in which communicating branches between the MCN and the median nerve were identified in 25 % of subjects (4/16). Many different communicating patterns have been described. Le Minor et al. [20] has classified them in five types: in type I, there is no communicating fiber between the MCN and the median nerve; in type II some medial root fibers run in the MCN and leave it after a
123
distance to join the median nerve; in type III, the lateral root of the median nerve runs in the MCN and leaves it after a distance to join the median nerve; in type IV, the fibers of the MCN are united with the lateral root of the median nerve and, after passing some distance into the upper arm, the MCN arises from the median nerve in the upper arm; and in type V the fibers of the MCN run within the median nerve along its course in the normal path of the median nerve. Of our five communicating branches, two were under type III, and one under type V (Fig. 3). In the fourth one, some lateral root fibers running in the median nerve were leaving it after a distance to join the MCN, which was not described in Le Minor’s classification. This variation was also observed in five limbs in Chiarapattanakom’s series [5] and 3/44 in Krishnamurthy’s series [16]. The absence of an independent MCN with nerve fibers emerging directly from the median nerve was rarely reported in literature (1.5–15 %) [32, 44]. In our case, the motor and sensory branches of the MCN shared the same conjonctive sheath as the median nerve (Fig. 3), as also reported by Fregnani et al. [10]. Most of the previous related anatomical studies focused on the number of main trunks exiting from the nerve and not on the distal branching pattern. Their results are most useful for nerve transfers, in which a normal (donor) nerve is sutured to the distal end of a damaged (recipient) nerve, as near as possible to the effectors, but before the final division of the trunk in the muscle. For instance, a fascicle of the ulnar nerve is sutured to a trunk dedicated to the biceps brachii muscle [25] and a fascicule from the median nerve is sutured to a trunk to the brachialis muscle [35, 43]. It is noteworthy, however, that none of the previous articles describe precisely how the recipient trunk is chosen in case of multiple recipient trunks. In view of our findings, some
Surg Radiol Anat
insufficient clinical results could be explained by the reinnervation of only a part of the target muscle. Other studies assessed only the localization of the intramuscular motor points of the biceps brachii and the brachialis to develop guidelines for percutaneous Botulinum toxin block injections [3, 19, 30]. Lee et al. observed that the intramuscular motor points of the biceps brachii were located between 44.6 and 76 % of the distance from the coracoid process to the medial epicondyle [19]. The motor points of the brachialis were located between 64 and 86.6 % of the same distance. Moreover, Lee noticed that the location of the entry points of the motor branches was independent of the branching pattern according to Yang classification [45]. Selective neurectomy (also termed ‘‘selective neurotomy’’ by some authors [4, 21, 24, 40, 41] consists in partial and selective resection of the motor branch(es) of a nerve to reduce the spastic tone of the target muscle(s). Some promising results have been reported with this procedure, including for the motor branches of the MCN to the biceps brachii and the brachialis muscle [4, 21, 34]. However, there are two levels of controversy regarding this technique. The first discussion involves the technique itself. One method consists in approaching only the main trunk of the nerve, and with intra-operative electrostimulation and intra-neuro-dissection identifying the motor branches and selecting the amount of fascicles which will be removed [33, 34, 37]. This technique, which avoids extensive dissection, has its limitations, partly because of the plexiform intra-neural anatomy (observed in our study and similar to Sunderland’s findings [42]), leading to a rather empirical choice of fascicles to resect. This may lead to a very heterogenous distribution of denervated muscle fibers, leaving in place large untouched areas of the target muscle(s) with insufficient results, and to sensory problems, as reported by several authors [21, 23, 36]. The second method, more invasive and more demanding technically, follows each motor trunk and each of its division all the way to its entry point into the involved muscle. Partial neurectomy, which we have termed ‘‘hyperselective’’, is then performed on each terminal branch at this level, by resecting a few millimetres of the required amount of fascicles. Sindou [40] recommends coagulating the proximal stump to prevent regrowth of the fibers. This technique allows making a harmonious, extremely selective, partial denervation of each area of the target muscle(s), while preserving the sensory branches. Magnifying loops and intra-operative nerve stimulation are both recommended for this technique. Perioperative electromyography [41] may also be applied for identifying the branches and particularly for differentiating between the motor and sensory nerves. The second problem concerns the amount of nerves fibers that should be resected. In his princeps paper, Brunelli [2] advised to
resect at least 50 % of the fascicles of each motor branch. But he later described the ‘‘adoption’’ (or ‘‘take-over’’) phenomenon, whereby the ‘‘orphan’’ muscle fibers are adopted by adjacent nerve branches, leading to some recurrence, and then advised to resect two-thirds of the fibers. On the other hand, excessive resection of the motor branches may lead to permanent paresis of the target muscles. Direct visualization of each and all of the terminal motor branches allows a much more precise quantification of the neurectomy. The present study showed that there is a mean distance of 57 mm between the origin of the first trunk and the last terminal branch dedicated to the biceps brachii, and an average of 56 mm between the first trunk to the brachialis muscle and the origin of the last terminal branch. The most efficient approach for this procedure of hyperselective neurectomy, requiring identification of all motor branches of the MCN to the biceps brachii and brachialis muscles, is longitudinal, over the medial bicipital groove. The superficial fascia and the biceps brachii perimysium are opened and the dissection proceeds directly down to the MCN, running longitudinally between the biceps, above, and the brachialis, underneath. According to our findings, we recommend to conduct the nerve dissection between 18 % (proximal) and 75 % (distal) of the distance between the coracoid process and the lateral epicondyle. This should allow proper identification of all the motor trunks and terminal branches dedicated to the biceps brachii and to the brachialis muscles, while sparing both sensory branches and communicating branches with the median nerve. Acknowledgments We thank all the staff of the anatomy laboratory of the Ecole de Chirurgie du Fer a` Moulin (Paris) for its precious help. Conflict of interest The authors have no conflicts of interest or financial ties to disclose.
References 1. Bendersky M, Bianchi HF (2012) Double innervation of the brachialis muscle: anatomic-physiological study. Surg Radiol Anat 34:865–870 2. Brunelli G, Brunelli F (1983) Partial selective denervation in spastic palsies (hyponeurotization). Microsurgery 4:221–224 3. Buchanan TS, Erickson JC (1996) Selective block of the brachialis motor point. An anatomic investigation of musculocutaneous nerve branching. Reg Anesth 21:89–92 4. Buffenoir K, Rigoard P, Ferrand-Sorbets S, Lapierre F (2009) Etude re´trospective du re´sultat a` long terme des neurotomies se´lectives pe´riphe´riques dans le traitement du membre supe´rieur spastique. Neurochirurgie 55(Suppl 1):50–60 5. Chiarapattanakom P, Leechavengvongs S, Witoonchart K, Uerpairojkit C, Thuvasethakul P (1998) Anatomy and internal topography of the musculocutaneous nerve: the nerves to the biceps and brachialis muscle. J Hand Surg Am 23:250–255
123
Surg Radiol Anat 6. Choi D, Rodrı´guez-Niedenfu¨hr M, Va´zquez T, Parkin I, San˜udo JR (2002) Patterns of connections between the musculocutaneous and median nerves in the axilla and arm. Clin Anat 15:11–17 7. Decq P, Cuny E, Filipetti P, Fe`ve A, Ke´ravel Y (1998) Les neurotomies pe´riphe´riques dans le traitement de la spasticite´. Indications, technique et re´sultats aux membres infe´rieurs. Neurochirurgie 44:175–182 8. Eglseder WA Jr, Goldman M (1997) Anatomic variations of the musculocutaneous nerve in the arm. Am J Orthop 26:777–780 9. Fe`ve A, Decq P, Filipetti P, Ke´ravel Y (1998) Traitement de la spasticite´ par injections de toxine botulique. Neurochirurgie 44(3):192–196 10. Fregnani JHTG, Mace´a MIM, Pereira CSB, Barros MD, Mace´a JR (2008) Absence of the musculocutaneous nerve: a rare anatomical variation with possible clinical-surgical implications. Sao Paulo Med J 126:288–290 11. Garland DE, Thompson R, Waters RL (1980) Musculocutaneous neurectomy for spastic elbow flexion in non-functional upper extremities in adults. J Bone Joint Surg Am 62:108–112 12. Guerri-Guttenberg RA, Ingolotti M (2009) Classifying musculocutaneous nerve variations. Clin Anat 22:671–683 13. Keenan MA, Tomas ES, Stone L, Gerste´n LM (1990) Percutaneous phenol block of the musculocutaneous nerve to control elbow flexor spasticity. J Hand Surg Am 15:340–346 14. Kervancioglu P, Orhan M, Kilinc N (2011) Patterns of motor branching of the musculocutaneous nerve in human fetuses and clinical significance. Clin Anat 24:168–178 15. Kong KH, Chua KS (1999) Neurolysis of the musculocutaneous nerve with alcohol to treat poststroke elbow flexor spasticity. Arch Phys Med Rehabil 80:1234–1236 16. Krishnamurthy A, Nayak SR, Venkatraya Prabhu L, Hegde RP, Surendran S, Kumar M, Pai MM (2007) The branching pattern and communications of the musculocutaneous nerve. J Hand Surg Eur 32:560–562 17. Kwolczak-Mc Grath A, Kolesnik A, Ciszek B (2008) Anatomy of branches of the musculocutaneous nerve to the biceps and brachialis in human fetuses. Clin Anat 21:142–146 18. Lance JW (1980) Symposium synopsis. In: Feldman RG, Young RR, Koella WP (eds) Spasticity: disorder of motor control. Year Book Medical Publishers, Chicago, pp 485–494 19. Lee JH, Kim HW, Im S, An X, Lee MS, Lee UY, Han SH (2010) Localization of motor entry points and terminal intramuscular nerve endings of the musculocutaneous nerve to biceps and brachialis muscles. Surg Radiol Anat 32:213–220 20. Le Minor JM (1990) A rare variation of the median and musculocutaneous nerves in man. Arch Anat Histol Embryol 73:33–42 21. Maarrawi J, Mertens P, Luaute J, Vial C, Chardonnet N, Cosson M, Sindou M (2006) Long-term functional results of selective peripheral neurotomy for the treatment of spastic upper limb: prospective study in 31 patients. J Neurosurg 104:215–225 22. Mahakkanukrauh P, Somsarp V (2002) Dual innervation of the brachialis muscle. Clin Anat 15:206–209 23. Matsuo T, Tada S, Hajime T (1986) Insufficiency of the hip adductor after anterior obturator neurectomy in 42 children with cerebral palsy. J Pediatr Orthop 6:686–692 24. Msaddi AK, Mazroue AR, Shahwan S, al Amri N, Dubayan N, Livingston D, Moutaery KR (1997) Microsurgical selective peripheral neurotomy in the treatment of spasticity in cerebralpalsy children. Stereotact Funct Neurosurg 69:251–258 25. Oberlin C, Be´al D, Leechavengvongs S, Salon A, Dauge MC, Sarcy JJ (1994) Nerve transfer to biceps muscle using a part of ulnar nerve for C5-C6 avulsion of the brachial plexus: anatomical study and report of four cases. J Hand Surg Am 19:232–237
123
26. Ozturk A, Bayraktar B, Taskara N, Kale AC, Kutlu C, Cecen A (2005) Morphometric study of the nerves entering into the coracobrachialis muscle. Surg Radiol Anat 27:308–311 ¨ ztu¨rk NC, Uzmansel D, O ¨ ztu¨rk H (2010) An unreported pattern 27. O of musculocutaneous and median nerve communication with multiple variations of biceps brachii: a case report. Surg Radiol Anat 32:887–890 28. Pacha Vicente D, Forcada Calvet P, Carrera Burgaya A, Llusa´ Pe´rez M (2005) Innervation of biceps brachii and brachialis: anatomical and surgical approach. Clin Anat 18:186–194 29. Paik DJ, Shin SY (2009) An anatomical study of the inferior oblique muscle: the embalmed cadaver vs the fresh cadaver. Am J Ophthalmol 147:544–549 30. Park BK, Shin YB, Ko HY, Park JH, Baek SY (2007) Anatomic motor point localization of the biceps brachii brachii and brachialis muscles. J Korean Med Sci 22:459–462 31. Penn RD, Kroin JS (1985) Continuous intrathecal baclofen for severe spasticity. Lancet 2:125–127 32. Prasada Rao PC, Chaudhary SC (2001) Absence of musculocutaneous nerve: two case reports. Clin Anat 14:31–35 33. Puligopu AK, Purohit AK (2011) Outcome of selective motor fasciculotomy in the treatment of upper limb spasticity. J Pediatr Neurosci 6(Suppl 1):118–125 34. Purohit AK, Raju BS, Kumar KS, Mallikarjun KD (1998) Selective musculocutaneous fasciculotomy for spastic elbow in cerebral palsy: a preliminary study. Acta Neurochir 140:473–478 35. Ray WZ, Pet MA, Yee A, Mackinnon SE (2011) Double fascicular nerve transfer to the biceps and brachialis muscles after brachial plexus injury: clinical outcomes in a series of 29 cases. J Neurosurg 114:1520–1528 36. Rousseaux M, Buisset N, Daveluy W, Kozlowski O, Blond S (2009) Long-term effect of tibial nerve neurotomy in stroke patients with lower limb spasticity. J Neurol Sci 278:71–76 37. Shin DK, Jung YJ, Hong JC, Kim MS, Kim SH (2010) Selective musculocutaneous neurectomy for spastic elbow. J Korean Neurosurg Soc 48:236–239 38. Simpson DM, Alexander DN, O’Brien CF, Tagliati M, Aswad AS, Leon JM (1996) Botulinum toxin type A in the treatment of upper extremity spasticity: a randomized, double blind, placebocontrolled trial. Neurology 46:1306–1310 39. Sindou MP (2005) History of the neurosurgical treatment for spasticity. Oper Tech Neurosurg 7:96–99 40. Sindou MP, Simon F, Mertens P, Decq P (2007) Selective peripheral neurotomy (SPN) for spasticity in childhood. Child Nerv Syst 23:957–970 41. Sitthinamsuwan B, Chanvanitkulchai K, Phonwijit L, Ploypetch T, Kumthornthip W, Nunta-Aree S (2013) Utilization of intraoperative electromyography for selecting targeted fascicles and determining the degree of fascicular resection in selective tibial neurotomy for ankle spasticity. Acta Neurochir 155:1143–1149 42. Sunderland S (1945) The intraneural topography of the radial, median, and ulnar nerves. Brain 68:243–299 43. Teboul F, Kakkar R, Ameur N, Beaulieu JY, Oberlin C (2004) Transfer of fascicles from the ulnar nerve to the nerve to the biceps in the treatment of upper brachial plexus palsy. J Bone Joint Surg Am 86-A:1485–1490 44. Uzel AP, Bulla A, Steinmann G, LaurentJoye M, Caix P (2011) Absence du nerf musculocutane´ et distribution a` partir du nerf median: a` propos de deux cas et revue de la litte´rature. Morphologie 95:146–150 45. Yang ZX, Pho RW, Kour AK, Pereira BP (1995) The musculocutaneous nerve and its branches to the biceps and brachialis muscles. J Hand Surg Am 20:671–675
