Successful intramuscular neurotization is dependent on the denervation period. A histomorphological study of the gracilis muscle in rats

ABSTRACT: To characterize the extent to which reinnervation potentialdepends on the duration of denervation, intramuscular neurotization of thegracilis muscle was performed either immediately or 2, 4, 6, and 8 weeksafter transection of the obturator nerve. For neurotization, the sciatic nervewas split into three fascicle groups and fixed intramuscularly. Muscle mor-phology after 6 weeks of regeneration was identified with anti-myosin im-munohistochemistry and NADH staining. Newly formed motor endplateswere characterized using acetylcholinesterase staining and electron micros-copy. Wet muscle weight ratio indicated the functional state of synapses.Depending on the denervation period, three levels of regenerative outcomewere evident. Best results were seen after immediate neurotization or after2 weeks of denervation. Regeneration, although at a significantly lower level,also occurred after denervation periods of 4 and 6 weeks. Regenerationfollowing neurotization after 8 weeks of denervation was negligible. Quantityand quality of motor endplate formation depended on the denervation pe-riod. Thus, in special clinical situations intramuscular neurotization within adistinct time window provides a good reconstructive option.

Muscle Nerve 31: 221–228, 2005

SUCCESSFUL INTRAMUSCULAR NEUROTIZATIONIS DEPENDENT ON THE DENERVATION PERIOD.A HISTOMORPHOLOGICAL STUDY OF THEGRACILIS MUSCLE IN RATS

GERBURG KEILHOFF, PhD,1 and HISHAM FANSA, MD, PhD2

1Institute for Medical Neurobiology, Otto-von-Guericke University, Leipziger Strasse 44,D-39120 Magdeburg, Germany2Department of Plastic, Reconstructive and Aesthetic Surgery, Staedtische Kliniken – Mitte,Bielefeld, Germany

Accepted 20 October 2004

Direct nerve-to-muscle neurotization or intramuscu-lar neurotization is a special technique to insert anerve into a denervated muscle. The procedure ofinserting peripheral nerves directly into a muscle is(with the exception of free neurovascular muscletransplantation) the only reconstructive optionwhen loss of motor nerve function has occurred at adistal level, for example, after direct trauma to themuscle or surgical en bloc removal of tumors. How-ever, the functional value of this procedure remainscontroversial, and it is not the therapy of first choice.

The technique of reinnervating skeletal musclesby insertion within it of a donor nerve was developed

at the beginning of the 20th century during WorldWar I, when injuries and poliomyelitis producedparalytics for whom no therapy was available. Only afew reports of clinically successful reinnervation ex-ist, and clinical application of the method during thefollowing decades was limited.

However, animal experiments investigating di-rect muscular neurotization have shown acceptableresults. In a dog model, Sorbie and Porter grafted amotor nerve fascicle of the flexor carpi ulnaris intothe muscle belly of the flexor carpi radialis muscle.22

This resulted in variable and unpredictable out-comes, with regeneration ranging from 44% to 94%of the original muscular force. Macroscopically thereinnervated muscles still had degenerated fibersand histologically motor endplates were concen-trated around the inserted stump of the donornerve. In contrast, Sakellarides and coworkersachieved a highly consistent 60% to 75% of originalmuscle function in the dog by dividing the donornerve into two or three fascicles prior to implanta-tion.21 This resulted in the regenerated muscles hav-

Abbreviations: AChE, acetylcholinesterase; DAB, diaminobenzidine;NADH, nicotineamide adenine dehydrogenase; NBT, nitroblue tetrazoliumchloride; PBS, phosphate-buffered salineKey words: gracilis muscle; neurotization; obturator nerve; sciatic nerve;synaptogenesisCorrespondence to: G. Keilhoff; e-mail: [email protected]

© 2004 Wiley Periodicals, Inc.Published online 22 December 2004 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/mus.20260

Intramuscular Neurotization MUSCLE & NERVE February 2005 221

ing a macroscopic appearance barely distinguishablefrom normal muscles.

This procedure was clinically refined by severalgroups.2,4,14 The distal end of each graft segment ismicrosurgically dissected to separate the fascicles ina proximal direction, thus creating neural branch-ing, and then each fascicle is inserted into the de-nervated muscle. Utilizing this technique, wide-spread distribution of motor axons throughout thedenervated muscle is ensured. Other studies withtechnical modifications showed that it is possible toreinnervate a muscle by introducing a nerve deepinto the muscle.5 There were no specific differencesin the reinnervation outcome between original andforeign nerve neurotization.18 The formation of newmotor endplates occurred also in zones in whichthey are not normally found. The quantity and qual-ity of newly formed motor endplates depended onthe distance of the muscle field from the nerveimplants.1 Time between denervation and reinner-vation was a crucial determinant of functional re-sults, but the precise time span that allows successfulreinnervation is unclear.8 Mersa and coworkers re-stored orbicularis oculi muscle function using thehypoglossal nerve even after prolonged denervation(up to 12 weeks), whereas other authors showed thatimmediate neurotization was superior to nerve im-plantation performed at different intervals after de-nervation.17,25

To further expand the clinical use of direct neu-rotization, meticulous experimental examination isrequired. We therefore developed an experimentalmodel in the rat to examine the reliability and timedependence of direct neurotization.

MATERIAL AND METHODS

Animal Model. A total of 60 adult Wistar female rats(Charles River, Sulzfeld, Germany; 250–300 g) wereused. All animals were maintained in accordance tothe guidelines of the German Animal Welfare Act.The experimental protocol was approved by a reviewcommittee of the state of Sachsen-Anhalt, Germany.The animals were housed under temperature-con-trolled conditions at 21 � 1°C, with a 12-h light/darkcycle, with free access to standard rat chow (Al-tromin 1324, Altromin GmbH, Lage, Germany) andwater.

The experimental design consisted of five neuro-tization (each with 8 animals) and four denervationgroups (without re-neurotization, each with 5 ani-mals). The groups were (1) immediate neurotiza-tion; (2) denervation for 2 weeks with or withoutsubsequent neurotization; (3) denervation for 4

weeks with or without subsequent neurotization; (4)denervation for 6 weeks with or without subsequentneurotization, and (5) denervation for 8 weeks withor without subsequent neurotization.

The surgical protocol for denervation of the gra-cilis muscle was identical for all groups. The rightgracilis muscle was exposed through a dorsal inci-sion under general anesthesia with pentobarbital (60mg/kg, intraperitoneally) and aseptic conditions.The branches of the obturator nerve were com-pletely transected 0.2 cm proximal to their entry intothe muscle belly. The nerve was traced cranially andresected for about 1 cm. Its proximal stump and thedistal stump at the muscle were coagulated usingbipolar forceps. This technique is routinely appliedin denervation operations. The muscle was com-pletely freed from the surrounding fascia and theadjacent muscles; only the entering blood vesselsand original attachments of the muscle were leftintact. With the exception of the first experimentalgroup, the wound was closed. Once the respectivedenervation time was reached (or, for the first ex-perimental group, immediately after denervation),the right sciatic nerve was exposed, split into 3 fas-cicles, and implanted into the gracilis muscle. Thefascicles were spread over the entire muscle belly,assuring widespread implantation. Whereas the ob-turator nerve enters the muscle in the proximalthird, the implanted fascicles have contact with themuscle in the proximal, middle, and distal third.The intact epineurium was fixed at the muscle fasciaby a 10-0 monofilament nylon suture with the aid ofan operating microscope (Moller, Wedel, Germany),and wounds were irrigated with normal saline solu-tion and closed. A 6-week period of recovery wasallowed.

Evaluation. The gracilis muscles were harvestedthrough the original incision (under general anes-thesia). The unexposed contralateral side served ascontrol. The wet muscle weight was determined aspercentage of control and mean values are ex-pressed with standard deviations. Then the muscleswere prepared for histological evaluation. Musclesfrom 5 animals of the neurotization groups werefixed for 48 h in 4% buffered paraformaldehyde(Merck, Darmstadt, Germany, pH 7.4) at 4°C. Forimmuno-enzyme histochemistry, muscle sampleswere cryoprotected for 2 days in a solution of 30%sucrose (Merck) in 0.4% buffered paraformalde-hyde (pH 7.4) and rapidly frozen to �20°C using2-methylbutane (Roth, Karlsruhe, Germany). Serialsections (20-�m thick) were cut on a cryostat (JungFrigocut 2800 E, Leica, Bensheim, Germany). Every

222 Intramuscular Neurotization MUSCLE & NERVE February 2005

second of the free-floating sections (a total of 100per muscle) was then incubated with the specificantimyosin antibody for fast skeletal muscle fibers(monoclonal; Sigma-Aldrich, Deisenhofen, Germany;1:400) diluted in phosphate-buffered saline (PBS).Following incubation with the primary antibody,slices were washed in PBS (3 � 5 min) and processedusing the peroxidase–antiperoxidase technique. Vi-sualization was performed with a DAB-H2O2 mix-ture. The specificity of the immunoreactions wascontrolled by the application of buffer instead ofprimary antiserum. All control sections were free ofany immunostaining.

For light-microscopy acetylcholinesterase (AChE)staining, every first, third, and fifth of the free-float-ing sections (300 per muscle) were incubated inKarnovsky medium, containing acetylcholine iodideas a substrate (pH 6.0, 6°C, 4 h).13 Every fourthsection (100 per muscle) was used for nicoti-namide adenine dehydrogenase (NADH) tetrazo-lium reductase histochemistry. They were incubatedin PBS containing 0.2 mg/ml nitroblue tetrazoliumchloride (NBT, Sigma), 1 mg/ml �-NADH (Sigma),and 0.3% Triton X-100 (pH 7.4, 37°C, 1 h), respec-tively. Reactions were terminated by washing withPBS.

For electron microscopy, muscles from the re-maining 3 animals from each group were fixed in 0.2M cacodylate buffer containing 2.5% glutaralde-hyde, osmicated (2% OsO4), dehydrated, en-blocstained with 7% uranyl acetate, flat-embedded inDurcupan (ACM, Fluka/Sigma, Seelze, Germany),and cut ultrathin (50–70 nm). Ultrathin sectionswere mounted on Formvar-coated slot grids and ex-amined with a transmission electron microscope (E900 Zeiss, Oberkochen, Germany). Light and trans-mission electron micrographs of 20 neuromuscularjunctions were chosen randomly from the musclespecimens from each of the different test and con-trol groups for morphological characterization. Forquantification, the number of motor endplates inevery third of the 300 AChE-stained sections permuscle was counted manually by two independentinvestigators blinded to the treatment group un-der �40 magnification, and the total number wascalculated (values are expressed as means with theirstandard deviations).

Statistical Analysis. Statistical analysis was per-formed with one-way analysis of variance (ANOVA)and nonparametric Kruskal–Wallis test. The Stu-dent’s t-test was used as posthoc test. Statistical sig-nificance was set at P � 0.05.

RESULTS

Clinical and Macroscopic Findings. There was no selfmutilation, and all wounds healed primarily. Nosigns of discomfort were observed. The denervatedmuscles appeared grossly atrophic with thin bulk,pale color, and fibrosis with prolonged denervationtime. After neurotization, muscular atrophy de-pended on the interval between denervation andneurotization, indicating an increase in adipose andconnective tissue.

Muscle Weights. Weights of the reinnervated graci-lis muscles were lower than those of the untreatedside but higher than those of denervated muscles.The values depended on the timing of nerve implan-tation. Immediate reinnervation and neurotizationafter 2 weeks gave the best results. The mean gracilisweights were 85.1% and 85.7% of control, respec-tively. After 2 weeks the denervated muscles showeda greater muscle loss, with muscle weights of 75% ofcontrol. Four weeks of denervation reduced theweights of the untreated muscle significantly to 48%of the control side. This remained constant over theevaluation time. Neurotization after 4 weeks led to astatistical significant difference between reinner-vated and control muscles. The mean weight was71% of the control in the 4-week group and 69% inthe 6-week group (Fig. 1). After 8 weeks neurotiza-tion failed, showing the same massive loss of musclemass as the untreated denervated group (47% ofcontrol, Fig.1).

Muscle Histology. The control gracilis muscle hadtightly packed muscle fibers and peripherally posi-

FIGURE 1. Wet muscle weights of the gracilis muscle expressedas percentage of control, evaluated after a regeneration time of 6weeks, with respective denervation periods. Data are mean val-ues � SD. Significant differences relative to the control: �P �0.05, �P � 0.005.


tioned myonuclei. The perimysal septa showed noadipose tissue (Fig. 2A). The NADH staining (Fig.2C) and antimyosin immunolabeling (Fig. 2D) re-vealed the typical random distribution of differentfiber types (checkerboard pattern). There was a nu-

merical dominance of type IIb fibers (fast, NAHD�,antimyosin�). This pattern was seriously disturbed inthe denervated and reinnervated muscles. In thedenervated muscle the most striking feature was fi-brosis. The degree of fibrosis correlated with the

FIGURE 2. (A) Normal rat gracilis muscle with peripherally positioned myonuclei (magnification, �300). (B) Re-innervated musclefollowing neurotization after denercation for 4 weeks, with central location of myonuclei (HE staining; magnification, �300). (C) NADHreaction with typical checkerboard pattern in control muscles. The darkest fibers are the most oxidative type I fibers (magnification, �240).(D) The same section with antimyosin immunolabeling and labeling that indicates type II fibers (magnification, �240). (E–G) Denervatedmuscles with increased fibrosis (stars) and a reduced number of normal-sized fibers (open arrows) among atrophic muscle tissue,depending on the denervation period (HE staining; magnification, �200). (H–J) Type grouping of muscle fibers and fibrosis in thereinnervated groups. Arrows point to more intensely stained fibers, indicating an increase in NADH activity (NADH staining; H, immediateneurotization; I, after 2 weeks; J, after 4 weeks; magnification, �240).


length of denervation. An increase of intrafascicularfat cells correlated with the amount of fibrosis. Up tothe second week after denervation, groups of nor-mal-sized fibers were encountered frequently amongatrophic muscle fibers (Fig. 2E). With increasingdenervation time, the number of such normal-sizedfibers decreased (Fig. 2F,G). The reinnervatedgroups showed fiber-type grouping, and adipose andconnective tissue was located between the fibers, butto a lesser extent than for denervated muscles (Fig.2H–J). After neurotization, normal-sized myofibersoutnumbered atrophic fibers up to a denervationinterval of 6 weeks. Atrophic changes affected bothfiber types. In some cases, NADH tetrazolium reduc-tase activity was increased (Fig. 2J). In both dener-vation and neurotization groups, some myonucleiwere centrally localized (Fig. 2B).

Neuromuscular Endplates. The AChE-stained sec-tions showed that, in normal control muscles, motorendplates were properly labeled and localized instraight lines (Fig. 3A,B). In all neurotizationgroups, sciatic nerve branches were in contact withmuscle fibers and had formed new motor endplatesthat were scattered (Fig. 3C,D) primarily near theimplantation area (Fig. 3E). Only rare separatenerve fascicles and endplates were seen among mus-cle fibers far from the implanted nerve stump, indi-cating formation of new nerve pathways (Fig. 3F).New endplates demonstrated a high degree of poly-morphism. Endplates situated transversely to the fi-ber were observed (Fig. 3G,H), with compact orindefinite shapes (Fig. 3I). Some were pale in color,indicating reduced AChE reactivity (Fig. 3J). Com-pared to controls, newly formed endplates after neu-

FIGURE 3. AChE-stained neuromuscular junctions. (A, B) Distribution of endplates as bands on the muscle surface in the control group(magnification: A, �40; B, �100). Note the compact shapes and labeling. (C–E) Scattered distribution of endplates in reinnervatedmuscles (C, neurotization after 4 weeks; magnification, �40. D, immediate neurotization; magnification, �45. E, neurotization after 2weeks; magnification, �40). After shorter periods of denervation the endplates are packed more densely. (F–I) High degree ofpolymorphism in the newly formed endplates. Transverse positions in relation to the fibers (F, neurotization after 2 weeks; G, neurotizationafter 4 weeks), indefinite shapes (H, neurotization after 4 weeks) and pale color (I, neurotization after 6 weeks; F–I, magnification, � 100).


rotization were increased in number after 2 and 4weeks denervation and less after 6 and 8 weeks de-nervation. They were packed more densely in thefirst groups (Fig. 3C,D). With increased denervationtime, more endplates were shaped irregularly.

At the electron microscope level, control neuro-muscular junctions appeared as well-defined struc-tures with the axon terminal capped by a Schwanncell contacting the muscle-fiber membrane whichwas infolded, forming secondary synaptic clefts (Fig.4A,B). Endplates in the 2 and 4 weeks denervatedgroups were characterized by degenerated nerve ter-minals surrounded by apparently normal junctionalfolds (Fig. 4C); after longer denervation, both end-plates compartments were degenerated (Fig. 4D). Inthe reinnervation groups 1–4, a mixture of well-developed and immature endplates was found. Thelatter were identified by the muscle membrane,which had started to fold to a variable extent tobecome the muscle part of the motor endplate (Fig.4E). A thin layer of Schwann cell cytoplasm coveredthe nerve terminal that contained less mitochondria.Some mitochondria were swollen or disrupted. Divi-sion of the terminal axon inside the synaptic gutteroccurred in varying degrees (Fig. 4F,G). Some ter-

minals seemed to contain only vesicles (Fig. 4F,G)and sometimes large areas of Schwann cells con-tacted the muscle surface (Fig. 4H). Between thestructural elements of the new endplates, wider cleftswere evident. Schwann cells contained a large num-ber of vacuoles, indicating phagocytic activity (Fig.4F,H). With increasing denervation interval, thesigns of endplate malformation were more impres-sive. After 8 weeks of denervation, few, if any, well-developed endplates were seen.

DISCUSSION

We found that (1) the technique of widespread in-sertion of nerve endings into a muscle offers suffi-cient contact to establish reinnervation, and (2) in-tramuscular neurotization leads to acceptablereinnervation after denervation periods of up to 6weeks, although ideally this interval should not ex-ceed 2 weeks.

All muscles undergoing neurotization were rein-nervated. Muscle weight, a strong indicator for func-tion, indicated no significant differences betweenimmediate and 2-week delayed neurotization com-pared to control.9 However, our neurotization

FIGURE 4. Electron micrographs showing endplates of the gracilis muscle. (A, B) In control animals the nerve terminal (NT) is cappedby terminal Schwann cells (SC), has intact mitochondria and synaptic vesicles, and is directly opposite the postsynaptic specializations(junctional folds, JF, magnification, �12,000) of muscles (M). (C) A fully degenerated individual nerve terminal bouton surrounded by arelatively intact junctional fold after a denervation period of 4 weeks (magnification, �12,000). (D) After prolonged denervation (6 weeks),terminal Schwann cells occupy large areas of the junctional folds, phagocytosing the fragmented nerve terminals (magnification,�12,000). (E–H) Newly developed endplates following a denervation time of 4 weeks. Postsynaptic specializations are underdeveloped(arrows, E, magnification �25,500). Division of terminal axons inside the synaptic gutter occur. Some terminals are filled with vesicles(arrowheads, F, magnification �12,000; G, magnification �25,500). Schwann cells contact the muscle surface (H, magnification�12,000). They contain a mass of vacuoles demonstrating their phagocytic activity.


model offered a large number of fibers for reinner-vation compared to the original innervation by theobturator nerve. Further, using the contralateral gra-cilis as control may be problematic because musclefunction of the contralateral limb is changed by anoperation on one hindlimb. Again, although thenerve stumps were coagulated, autoreinnervation bycollateral sprouting may occur. The stagnancy ofmuscle atrophy after 6 weeks even in the denervatedgroups indicates that collateral sprouting may play arole in muscle reinnervation.

In the neurotization groups, a decline in func-tional recovery was evident after a longer denerva-tion period.3,10,12 Features of degeneration are notfully reversible in muscles after prolonged denerva-tion. The comparison of our model, in which syn-apses were newly installed, with the reinnervationachieved after nerve repair, in which existing syn-apses are reactivated, leads to the assumption thathooking up of original synapses allows better regen-eration even after longer periods of denervation.3,12

In patients, however, in whom traditional repairby direct nerve suture or nerve graft is impossibledue to traumatic or surgical reasons, muscle reinner-vation may be achieved by direct implantation of anerve into a muscle, despite the limitations that re-sult from the formation of new synapses.

The reinnervated muscles show a typical cluster-ing of fibers (fiber-type grouping), the result of suc-cessful muscle reinnervation, as it occurs with reinner-vation of muscles by end-to-end nerve coaptation.20

Thus, this phenomenon is not related to the type ofnerve repair. Furthermore, the decline in functionaloutcome in the groups with a longer denervationtime correlates with an increase in adipose and con-nective tissue and a reduction of normal-sized fibersof the muscle. Fibrosis is one of the main factorshampering successful recovery from muscle dener-vation.24

The number of endplates is a further importantparameter.19 None of the groups, except the 8-weekdenervation group, showed statistically significantdifferences compared to the control group. Morpho-logically, the endplates of the neurotization musclesdiffer from those of normal muscle.11,15 They aredensely packed after 2 and 4 weeks of denervation,and have a high degree of polymorphism in allgroups. New endplates form in the vicinity of ingrow-ing axons, if the muscle is denervated.6 The in-creased sensitivity of denervated muscles to acetyl-choline is a possible contributory factor to thisphenomenon, which is not seen if a motor nerve isimplanted into an innervated muscle.18,19

Our results from the rat are not directly transfer-able to humans. There are, however, clinical studiesindicating that the time dependency of successfulregeneration after nerve repair and intramuscularneurotization is similar.5,7,16,23 The time interval be-tween trauma and reconstruction is a crucial deter-minant of functional result. In addition, the func-tional outcome depends on the quality of the donornerve, age of the patient, size of the nerve defect(regenerative distance), and quantity and quality ofremaining muscle mass.8

This work was supported by a grant from the Deutsche For-schungsgemeinschaft (DFG). The authors thank Leona Buck andKarla Klingenberg for technical assistance. Catherine Munro re-vised the manuscript, and we appreciate her linguistic and scien-tific help.

REFERENCES

1. Askar I, Sabuncuoglu BT, Yormuk E, Saray A. The fate ofneurotization techniques on reinnervation after denervationof the gastrocnemius muscle: an experimental study. J Recon-str Microsurg 2001;17:347–355.

2. Askar I, Sabuncuoglu BT. Superficial or deep implantation ofmotor nerve after denervation: an experimental study—-su-perficial or deep implantation of motor nerve. Microsurgery2002;22:242–248.

3. Bain JR, Veltri KL, Chamberlein D, Fahnestock M. Improvedfunctional recovery of denervated skeletal muscle after tempo-rary sensory nerve innervation. Neuroscience 2001;103:503–510.

4. Becker M, Lassner F, Fansa H, Mawrin C, Pallua N. Refine-ments in nerve to muscle neurotization. Muscle Nerve 2002;26:362–366.

5. Brunelli GA, Brunelli GR. Direct muscle neurotization. J Re-constr Microsurg 1993;9:81–90.

6. Brunelli GA, Payne SH, Korneliussen H, Sommerschild H.Ultrastructure of the new neuromuscular junctions formedduring reinnervation of rat soleus muscle by a foreign nerve.Cell Tissue Res 1976;167:439–452.

7. Constantinidis J, Akbarian A, Steinhart H, Iro H, Mautes A.Effects of immediate and delayed facial–facial nerve suture onrat facial muscle. Acta Otolaryngol 2003;123:998–1003.

8. Frey M. Avulsion injuries to the brachial plexus and the valueof motor reinnervation by ipsilateral nerve transfer. J HandSurg [Br] 2000;25:323–324.

9. Frykman GK, McMillan PJ, Yegge S. A review of experimentalmethods measuring peripheral nerve regeneration in ani-mals. Orthop Clin North Am 1988;19:209–219.

10. Fu SY, Gordon T. Contributing factors to poor functionalrecovery after delayed nerve repair: prolonged denervation.J Neurosci 1995;15:3886–3895.

11. Hua J, Kumar P, Tay SSW, Pereira BP. Microscopic changes atthe neuromuscular junction in free muscle transfer. ClinOrthop Res 2003;411:325–333.

12. Irintchev A, Draguhn A, Wernig A. Reinnervation and recov-ery of mouse soleus muscle after long term denervation.Neuroscience 1990;39:231–243.

13. Karnovsky MJ, Roots L. A direct-coloring thiocholine Methodfor choline esterase. J Histochem Cytochem 1964;12:219–221.

14. Lassner F, Becker M, Fansa H, Mawrin C, Pallua N. Recon-struction of defects at the neuromuscular junction. HandchirMikrochir Plast Chir 2003;35:127–131.

15. Lubischer JL, Thompson WJ. Neonatal partial denervationresults in nodal but not terminal sprouting and a decrease in


efficacy of remaining neuromuscular junctions in rat soleusmuscle. J Neurosci 1999;19:8931–8944.

16. Lutz BS, Chuang SS, Cunang JC, Hsu JC, Ma SF, Wei FC.Nerve transfer to the median nerve using parts of the ulnarand radial nerves in the rabbit — effects on motor recovery ofthe median nerve and donor nerve morbidity. J Hand Surg[Br] 2000;25:329–335.

17. Mersa B, Tiangco DA, Terzis JK. Efficacy of the baby-sitterprocedure after prolonged denervation. J Reconstr Microsurg2000;16:27–35.

18. Park DM, Shon SK, Kim YJ. Direct muscle neurotization in ratsoleus muscle. J Reconstr Microsurg 2000;16:379–383.

19. Payne SH Jr, Brushart TM. Neurotization of the rat soleusmuscle: a quantitative analysis of reinnervation. J Hand Surg[Am] 1997;22:640–643.

20. Rab M, Koller W, Haslik W, Kamholz LP, Beck H,Meggeneder J, Frey M. The influence of timing on the func-tional and morphological result after nerve grafting: an ex-perimental study in rabbits. Br J Plast Surg 2002;55:628–634.

21. Sakellarides HT, Sorbie C, James L. Reinnervation of dener-vated muscles by nerve transplantation. Clin Orthop 1972;83:195–201.

22. Sorbie C, Porter TL. Reinnervation of paralysed muscles bydirect motor nerve implantation. An experimental study inthe dog. J Bone Joint Surg Br 1969;51:156–164.

23. Sungpet A, Suphachatwong C, Kawinwonggowit V. Transferof a single fascicle from the ulnar nerve to the biceps muscleafter avulsion of upper roots of the brachial plexus. J HandSurg [Br] 2000;25:325–328.

24. Tews DS, Goebel HH, Schneider I, Gunkel A, Stennert E,Neiss WF. Morphology of experimentally denervated andreinnervated rat fascial muscle. I. Histochemical and his-tological findings. Eur Arch Otorhinolaryngol 1994;251:36 – 40.

25. Zheng H, Zhou S, Chen S, Li Z, Cuan Y. An experimentalcomparison of different kinds of laryngeal muscle reinnerva-tion. Otolaryngol Head Neck Surg 1998;119:540–547.


Documents

Successful intramuscular neurotization is dependent on the denervation period. A histomorphological study of the gracilis muscle in rats