nerve regeneration through a combined autologous conduit (vein plus acellular muscle grafts)

Biomaterials 19 (1998) 173—181

Nerve regeneration through a combined autologous conduit(vein plus acellular muscle grafts)

Giovanni Di Benedetto!,*, Germano Zura!, Roberta Mazzucchelli",Alfredo Santinelli", Marina Scarpelli", Aldo Bertani!

! Clinica di Chirurgia Plastica e Ricostruttiva, Universita degli Studi di Ancona, Piazza, Cappelli, 1 I-60121 Ancona, Italy" Department of Pathology, University of Ancona, Ancona, Italy

Received 2 June 1997; accepted 10 October 1997

Abstract

The authors describe nerve regeneration obtained by using a combined autologous conduit, consisting of a vein plus acellularmuscle grafts. The right sciatic nerve of seven Sprague—Dawley rats was transected for a length of 2 cm and the gap was filled witha 2 cm long femoral vein conduit in which two autologous acellular muscle grafts had been previously inserted. Clinical andelectrophysiologic tests were carried out twelve weeks after the surgical procedure. The nerve was then removed and a morphologicalstudy, including histologic examination, immunohistochemistry and quantitative analysis, was performed. The left sciatic nerve was alsoremoved and used as a control. Regeneration was observed in the middle and distal parts of the conduit in 5 rats. Nerve conductionvelocity ranged between 0 and 14.9 ms~1. In the distal part the nerves were enclosed by a perineurium thicker than their normalcounterpart and in which groups of small axons were surrounded by thin myelin sheaths. Quantitative analysis showed that the operatednerve presented a wide variation of the area of the fascicle and the density of the fibres per area, while the diameter of the axons andmyelinated fibres showed only small variation, independent of the size of the fascicle. In conclusion, by using this technique, the authorssucceeded in obtaining regeneration of a well formed nerve fascicle. ( 1998 Published by Elsevier Science Ltd. All rights reserved.

Keywords: Nerve regeneration; Acellular muscle; Veins; Morphometry

1. Introduction

Peripheral nerve regeneration is a challenging field andthe topic has been studied by many authors, since the lastcentury [1—3]. Taking into account just the last twodecades, a number of papers in which various materialshave been used as an alternative to the nerve graft havebeen published. In particular, the possibility of bridgingnerve gaps either with synthetic or autologous materialshas been investigated. Silicon [4, 5], reabsorbable poly-mers [6, 7] or hyaluronic acid guides [8] were the syn-thetic materials most commonly used. Alternatively,veins [9, 10], pseudosynovial membranes around a sili-con implant [11, 12], collagen conduits [13], perineu-rium tubes [14], fresh nerve grafts in a vein conduit[15, 16] and acellular muscle grafts [17—19] are someexamples of autologous materials that have been used.

*Corresponding author. Tel: #39-71-5963481; fax: #39-71-5963481.

Pre-treated muscle has been described to be almostequivalent to fresh nerve autografts for peripheral nerveregeneration [17, 18]. Our own experiments [20], as wellas the results from others [19], however, do not com-pletely confirm this data.

In an attempt to improve our previous results weconducted the present experiment by using a new tech-nique based on the use of acellular muscle grafts insertedin a venous conduit.

The rationale for this approach was that the vein couldrepresent a guide for the regenerating axons better thanthe muscle alone.

2. Materials and methods

2.1. Surgical procedure

Seven male Sprague—Dawley rats, average weight250 g, identified from A1 to A7, were anaesthetised by an

0142-9612/98/$19.00 ( 1998 Published by Elsevier Science Ltd. All rights reserved.PII S 0 1 4 2 - 9 6 1 2 ( 9 7 ) 0 0 2 0 0 - 7

intraperitoneal injection of 12% chloral hydrate (0.1 mlper 100 g body weight).

Using a sterile technique, a 3 cm long skin incision onthe inner surface of the left thigh and leg was carried out;the femoral vein was isolated for a length of 2.5 cm, thedistal and proximal ends were ligated and the vein wastransected.

After closing the skin with 5/0 nylon, the rats wereturned on their ventral surface. Through a 3 cm longskin incision over the gluteal region, the right sci-atic nerve, from the sciatic notch to the trifurcation,was exposed and a 2 cm long segment was sharplytransected.

Autologous muscle graft strips were obtained at thetime of operation by excising the semimembranousmuscle along its entire length. The muscle, pinned toa 5 cm long cork block, underwent a freezing-thaw-ing procedure, consisting of placing the tissue alter-natively in liquid nitrogen (!196°C) and saline (37°C)for 5 min. The procedure was repeated three timesand resulted in a shrinkage of the muscle, averaging30%.

After this, the muscle was fashioned into rectangularblocks of about 0.3]0.1 cm and two of these wereinserted into the lumen of the previously transectedfemoral vein.

The vein graft was then interposed coaxially betweenthe proximal and distal ends of the transected rightsciatic nerve and secured by three to four 10/0 epineurialinterrupted sutures.

Muscle and skin layers were then closed primarily bymeans of 5/0 vicryl and nylon interrupted stitches, res-pectively. The animals were caged in separately andobserved weekly.

2.2. Neurophysiology

After 12 weeks, a walking-track test was performed[21]. The animals were then anaesthetised with intra-peritoneal 12% chloral hydrate and the nerve con-duction velocity (NCV) (Graph 1) was recorded fromthe sciatic nerves, both right and left, using a BasisP 300 (Epos). Electrical stimuli in the range 0.1—3.0 mAwere applied proximal to the graft using a bipolar elec-trode and NCV was recorded distal to it. The distancebetween the stimulating and the recording electrodeswas 2.2 cm. The same distance was maintained whenNCV was recorded from the left, non-operated, sciaticnerves. In each case at least three registrations weretaken and the mean value was calculated. The animalswere than euthanized and the right sciatic nerve wasremoved for a morphologic study that included his-tologic examination, immunohistochemistry and quant-itative analysis.

The left sciatic nerve was also removed and used asa control.

2.3. Morphologic investigation

Soon after removal, the specimens were put on a pieceof cork and allowed to adhere to the surface. Care wastaken to mark the proximal and distal ends. They werefixed in cold (4°C) neutral formol at pH 7.4 for 5 h. Afterfixation, serial sections perpendicular to the major axiswere cut, the starting point always being the proximalend. From the left sciatic nerves only two sections,approximately at the midpoint, were obtained. After this,the specimens were alternatively processed for paraffin orresin embedding. The aim was to obtain paraffin andresin embedded material so as to have good morphologi-cal detail (resin) without losing the possibility of applyingimmunohistochemistry (paraffin). This approach alsoallowed us to follow regeneration in a proximal-distaldirection. Specimens for paraffin passed through gradedalcohols, xylol and paraffin. Four-five micron thick sec-tions were cut and stained with Hematoxilin—Eosin (Hand E). From the same blocks serial sections were cut forimmunohistochemistry and lectin histochemistry. Forimmunohistochemistry the sections were treated with0.3% hydrogen peroxide in methanol to block the endo-genous peroxidase and then the primary antibody wasapplied (see Table 1 for details). After three washes withphosphate buffered saline, the sections were incubatedwith biotinylated secondary antibodies, washed in PBSand incubated with streptavidin peroxidase followed bystaining with 0.05% 3,3 diaminobenzidine-tetrahydroch-loride. For lectin histochemistry biotinylated lectins wereused. The slides were incubated at room temperature for30 min. After washing in PBS and incubation with strep-tavidin peroxidase, the peroxidase reaction product wasdeveloped by incubating sections with 0.05% 3,3 di-aminobenzidine. Normal and pathological human nervesprocessed in the same manner were used as positivecontrols.

The remaining material was post-fixed in OsO4, dehy-drated and embedded in epoxy-araldite resin. Semithinsections, approximately 1 lm thick, were stained withtoluidine blue.

Table 1List of primary antibodies and lectins used in the study, their sourceand dilutions

Supplier Working dilution

AntibodiesEMA DAKO 1 : 100CD68 DAKO 1 : 150

¸ectinsCon A VECTOR labs 1 : 250WGA VECTOR labs 1 : 166PNA VECTOR labs 1 : 166

174 G. Di Benedetto et al. / Biomaterials 19 (1998) 173—181

Graph 1. Values of mean NCV recorded from operated (right) and control (left) nerves.

Quantitative analysis was carried out on toluidine bluestained semithin sections using an IBAS Kontron imageanalyser and a semiautomatic approach. In the operatednerves, the section corresponding to the distal part of thegraft was analysed. The following features were meas-ured: transverse area of the fascicle, diameter of themyelinated fibres, diameter of the axons, and ratio be-tween the diameter of the axon and the diameter of themyelinated fiber (G factor). In practice, the first stepconsisted of measuring the area of the nerve fascicle byusing a low power objective (63]), the remainingmeasurements were performed in a second step by usinga 100] oil immersion objective. The slides were systema-tically scanned and at least 40 fields were counted in eachcase. The number of myelinated fibres per area (i.e., fiberdensity), was evaluated by using a grid inserted in theocular of the microscope (40] objective, area of the grid0.08 mm2). Repeated measurements showed an intra-observer variability inferior to 0.05.

Since the number of cases was small the data were notprocessed for statistical analysis. We considered thatvisual inspection of the graphs could give an idea of theresults.

3. Results

3.1. Clinical and electrophysiological findings

Three weeks after the operation, extension of the anklejoint, as well as extension and spreading of the toes, wasimpaired in all animals. During the first week, trophic

ulcers appeared in the heel region, persisting through the12th week in rats A1 and A4.

The results of the walking track test were consistentwith good motor recovery in five cases while in theremaining two (A1 and A4) a skidding of the limb wasdetected [19]. Nerve conduction velocity in the five caseswith good motor recovery ranged from 7.1 to 14.9 m s~1.In case A1 the value was close to 0, while in case A4 it was3.7. The values recorded from the left side were ratherhomogeneous, being between 18.0 and 19.5 m s~1 in allbut one animal. In fact, in rat A2, it was 28.9. Graph 1 isa representation of the NCV values recorded from theright (operated) and left (control) side of the five animalswith good motor recovery. The NCV values are reportedin Table 2.

3.2. Histopathological findings

Regeneration was achieved in five out of seven cases.In one case (A1) no regeneration at all was obtained,while in another case (A4) it was considered abortive aswe only observed a small structure consisting of connec-tive tissue and Schwann cells but completely devoid ofmyelinated fibres. In the remaining five cases, a regene-rated nerve trunk was observed. Regeneration could befollowed in a proximal-distal direction. In the proximalpart of the graft there were many small regeneratingaxons growing inside the muscle fibres, surrounded bytheir basal lamina (Fig. 1). The vein conduit was seenas a very thin lamina of connective tissue surroundingthis proliferation (Fig. 2). Distally, a nerve fascicle inwhich the central part was made up of connective tissue

G. Di Benedetto et al. / Biomaterials 19 (1998) 173—181 175

Table 2Values of the feature studied by quantitative analysis for the operated side (right) and in control nerves (left)

A2 A3 A5 A6 A7

Left Right Left Right Left Right Left Right Left Right

Area of the fascicle (mm2) 0.723 0.243 0.579 0.058 0.423 0.282 0.451 0.166 0.380 0.355(0.003) (0.001) (0.004) (0.002) (0.002) (0.002) (0.003) (0.002) (0.002) (0.005)

Fibre density (mm2) 1000 1430 944 453 1158 2109 1100 1899 1099 1328

Diameter of the fibres (lm) 6.27 3.71 7.13 2.58 6.70 3.16 6.90 2.80 6.96 2.82(2.00) (1.14) (2.4) (0.73) (1.97) (1.09) (1.32) (0.97) (2.32) (1.12)

Diameter of the axons (lm) 3.62 2.12 3.67 1.64 3.40 1.72 3.90 1.70 3.79 1.55(1.21) (0.94) (2.02) (0.82) (1.21) (0.95) (1.17) (0.81) (1.25) (1.07)

G factor 0.57 0.57 0.52 0.63 0.50 0.55 0.56 0.60 0.54 0.55(0.11) (0.12) (0.12) (0.14) (0.14) (0.12) (0.11) (0.10) (0.13) (0.14)

NCV (m s~1) 28.9 7.6 18.1 9.0 19.6 7.2 18.3 7.1 18.0 14.5

The values of the standard deviation are in brackets. The mean values of the nerve condution velocity (NCV) are also reported.

Fig. 1. Case A3. Small regenerating axons growing inside themuscle fibres and surrounded by their basal lamina (Toluidine blue25]).

and small myelinated axons (Fig. 3) were observed.Externally, there were proliferated, tightly packed, axons.Their density progressively decreased toward the exter-nal part of the specimen where a number of dilated, thin-walled blood vessels were also observed (Fig. 4). In thedistal part, the nerve appeared fully organised. It wasenclosed by a perineurium thicker than its normalcounterpart and groups of small axons surrounded bythin myelin sheaths were found (Fig. 5). Mast cells werealso present with an apparently random distribution.When visually compared with the controlateral normalnerve (Fig. 6), the regenerated sciatic nerve was charac-terised by numerous small myelinated axons associatedwith an increase in the number of Schwann cell nuclei.No lymphocytes or plasma cells were detected, except for

Fig. 2. Case A5. The muscle graft is surrounded by a thin connectivelamina representing the vein conduit. The muscle fibres on the right arestill devoid of axons while on the left there are muscle fibres surroundedby thin myelinated axons (Toluidine blue 16]).

a few giant multinucleated cells containing remnants ofsurgical material in their cytoplasm.

3.3. Immunohistochemical findings

Unfortunately, in our hands, the results of immunohis-tochemistry were rather unrewarding. The EMA anti-body did not stain the perineurium either in normal oroperated sciatic nerves while it beautifully decorated themyelin sheaths. The same antibody specifically stainedthe perineurium of normal human nerves used as a con-trol. We also applied a number of lectins since it has beenshown that some lectins specifically bind to human aswell as to rat perineurium. However, even by using this


Fig. 3. Case A6. Central part of the regenerating nerve fascicle madeup of connective tissue and small myelinated axons (Toluidine blue25]).

Fig. 4. Case A7. External part of the nerve fascicle surrounded byconnective tissue. (Toluidine blue 25]).

Fig. 5. Case A6. Fully organized nerve fascicle surrounded by athick perineurium in which lie small groups of axons (Toluidine blue25]).

Fig. 6. Case A6. Normal left sciatic nerve for comparison (Toluidineblue 40]).

approach we did not achieve satisfactory results. In prac-tice, we were not able to specifically stain the perineuriumwith any of the methods we employed, while good resultswere obtained using normal or pathological humannerves. No histiocytes were observed with the CD68antibody.

3.4. Quantitative analysis

As it was expected, the left sciatic nerves, used ascontrols, showed remarkably constant values regardingthe number of the fibres per area and the diameter ofthe myelinated fibres and their axons. The only fea-ture showing a certain variability was the area of thenerve fascicle. On the other hand, the operated nervesshowed a wide variation of the values of the area of thefascicle and the density of the fibres per area. In particu-lar, the latter feature had, in all but one case, valueshigher than its normal counterpart. The mean diametersof the myelinated fibres ranged from 2.80 to 3.71 lmwhile the values for the axons were between 1.55 and2.12. The G factor had values very close to those of thenormal nerves.

Values of the features measured are reported inTable 2 while Graphs 1—5 give a representation of thedistribution of the values in comparison to control nervesfor the single cases.

4. Discussion

Autologous skeletal muscle grafts have been provedeffective for the repair of peripheral nerves [16—18, 20].Most of the authors agree that, after the sarcoplasm andplasma membrane have been eliminated by macro-phages, the basal lamina of the muscle acts as a guide forthe regenerating axons [18, 22]. Good revascularisation


Graph 2. Area of the nerve fascicle in the operated (right) and control (left) nerves.

Graph 3. Diameter of the myelinated axons in the operated (right) and control (left) nerves.

of the graft is essential for a successful regeneration ofthe nerve [23]. Nerve regeneration seems to be im-proved by the association of the muscle graft with vascu-lar implantation [24]. The blood supply enhances themigration of Schwann cells and the diffusion of neurot-rophic factors but is affected by the length of the gap[25].

Vein grafts have been used for many years to repairperipheral nerves. It has been shown that, when the gapsin the nerve have a similar short length, the results usingvein grafts or nerve grafts are comparable [26, 27]. Infact, although after two or three months the vein graft hasa reinnervation percentage lower than the nerve graft, asshown by electrophysiological tests, the difference is no


Graph 4. Diameter of the axons in the operated (right) and control (left) nerves.

Graph 5. Density of the myelinated axons in the operated (right) and control (left) nerves.

longer detectable after six months [28, 29]. It has beentheorised that the lumen of the vein provides a favour-able microenvironment for nerve regeneration andthat coating the luminal epithelium with collagen gelimproves regeneration [30].

In the present experiment five out of seven ratsachieved regeneration as shown by the results of thewalking track test, measurement of nerve conduction

velocity and histological and quantitative analyses. Thecomplete lack of regeneration observed in case A1 couldhave been caused by technical problems at the time ofoperation. On the other hand, in case A4, regenerationappeared to be abortive rather than absent. We derivethis hypothesis from the morphological observation ofa small structure containing Schwann cells and connec-tive tissue, but devoid of myelinated axons, and from the


recording of a very slow NCV. Even in this case, technicalproblems could have accounted for the unsuccessful re-generation. In the remaining cases, the NCV we recordedindicated regeneration. The values were very close tothose reported by others using nerve grafts to bridgegaps of comparable lengths [25]. The histological ap-pearance was rather uniform when corresponding levelsof the grafts were compared. In particular, regenerationseemed to progress from the muscle implants to com-pletely fill the vein conduit. The latter was very thin andsince it became precociously involved in the regenera-tion, it was no longer recognisable in the distal part of thegraft. One problem that we faced in this study was theidentification of the perineurium in the regeneratednerves. In fact, it was difficult to establish the role of themuscle basal lamina, or the vein conduit, from a purelymorphologic point of view. In our attempt to demon-strate by immunohistochemistry and/or by lectin his-tochemistry, the early organisation of the perineuriumwas not successful as this structure failed to stain withany of the reagents we used.

The quantitative analysis showed basically that regene-rated nerves were smaller than their normal counter-parts, although with a rather wide range of values. In fact,in one case (A7) the value was close to that of thecontralateral normal nerve, while in another (A3) theregenerated nerve was more than ten times smaller. Onthe other hand, the feature ‘area of the fascicle’ is theone showing the greatest variability even in the groupof normal nerves. It is interesting to observe that theregenerating axons are small but they have myelinsheaths of proportional thickness giving G factor valuesin the range of normal nerves. The values obtained forthe diameters of the myelinated fibres and axons wereclose to that reported by others who used a vascularizednerve graft model [25].

The measurement of nerve conduction velocity gaveresults in line with morphologic and quantitative ana-lyses in most of the animals. In fact, the values wererather homogeneous and roughly half of those re-corded in the control nerves, corresponding to thereduced diameters of the myelinated fibres and theiraxons. Only in case A7 the NCV was high, beingin fact close to that of the corresponding left sciaticnerve. In this case, the only relevant quantitative valuethat could be related to this result was the area of thefascicle: it was the highest recorded among the operatednerves.

In conclusion, by using this technique, which in-volved the creation of a rather wide neural gap filledwith a vein conduit in which acellular muscle was in-serted, we succeeded in obtaining regeneration of awell formed nerve fascicle. The effect of the acellularmuscle graft in promoting regeneration was probablystrengthened by the presence of the vein conduit acting asa guide.

References

[1] Glueck T. Ueber Neuroplastic auf dem Wege der Transplanta-tion. Arch Klin Chir 1980;25:606—16.

[2] Millesi H, Berger A, Meissl G. Experimentelle Untersuchungenzur Heilung durchtrennter peripherer Nerven. Chir Plast 1972;1:174—206.

[3] Berger A, Millesi H. Nerve grafting. Clin Orthop 1978;133:49.[4] Lundborg G, Dahlin LB, Danielsen N et al. Nerve regeneration in

silicone chambers: influence of gap length and of distal stumpcomponents. Exp Neurol 1982;76:361—75.

[5] Varon S, Williams LD. Peripheral nerve regeneration in a siliconemodel chamber: cellular and molecular aspects. Periph Nerve RepReg 1986;1:9—25.

[6] Seckel BR, Chiu TH, Nyilas E, Sidman RL. Nerve regenerationthrough synthetic biodegradable nerve guides: regulation by thetarget organ. Plast Reconstr Surg 1984;74:173—81.

[7] Rice DH, Burstein FD, Newman A. Use of polytetrafluorinatedethylene compound in peripheral nerve grafting. Arch Otolaryn-gol 1985;111:259—61.

[8] Favaro G, Bortolami MC, Cereser S, Dona M, PastorelloA, Callegaro L, Fiori MG. Peripheral nerve regenerationthrough a novel bioresorbable nerve guide. Asaio Trans 1990;36:291—4.

[9] Chiu DTW, Janecka I, Krizek TJ et al. Autogenous vein graft asa conduitfor nerve regeneration. Surgery 1982;91:226—33.

[10] Rigoni G, Smahel J, Chiu DTW, Meyer VE. Venen interponat alsLeitbahn fuer die Regeneration peripherer Nerven. HandchirMikrochir Plast Chir 1983;15:227—31.

[11] Lundgorg G, Hansson HA. Nerve regeneration through pre-formed pseudosynovial tubes. J Hand Surg 1980;5:35—8.

[12] Lundborg G, Dahlin LB, Danielsen NP et al. Reorganization andorientation of regenerating nerve fibres, perineurium and epineu-rium in preformed mesothelial tubes - an experimental study onthe sciatic nerve of rats. J Neurosci Res 1981;6:265—83.

[13] Li ST, Archibald ST, Krarup C, Madison RD. Peripheral nerverepair with collagen conduits. Clin Mater 1992;9:195—200.

[14] Restrepo Y, Merle M, Micon J et al. Fascicular nerve graft usingan empty perineurial tube: an experimental study in the rabbit.Microsurgery 1983;4:105—12.

[15] Tang J. Vein conduits with interposition of nerve tissue for peri-pheral nerve defects. J. Reconstr Microsurg 1995;11:21—6.

[16] Fawcett JW, Keynes RJ. Muscle basal lamina: a new gratematerial for peripheral nerve repair. J Neurosurg 1986;65:354—63.

[17] Glasby MA, Gschmeissner S, Hitchcock RJI, Huang CLH. Rege-neration of the sciatic nerve in rats: the effect of muscle basementmembrane. J Bone Joint Surg 1986;68:829—33.

[18] Gattuso J, Glasby MA, Gschneissner S, Norris RW. A compari-son of immediate and delayed repair of peripheral nerves usingfreeze-thawed autologous skeletal muscle grafts in the rat. BrJ Plast Surg 1989;65:354—63.

[19] Mountain RE, Glasby MA, Sharp JF, Murray IAM. A morpho-logical comparison of interposed freeze-thawed skeletal muscleautografts and interposed nerve autografts in the repair of the ratfacial nerve. Clin Otolaryngol 1993;18:171—7.

[20] Di Benedetto G, Lassner F, Riccio M et al. The use of acellularautologous skeletal muscle grafts in peripheral nerve repair:a morphometrical-morphological study. Clinical Materials 1998;16:137—43.

[21] De Medinaceli L, Freed WJ, Wyatt RJ. An index of the functionalcondition of rat sciatic nerve based on measurements made fromwalking tracks. Exp Neurol 1982;77:634—42.

[22] Gattuso JM, Davies AH, Glasby MA, Gschmeissner SE, HuangCLH. Peripheral nerve repair using muscle autografts. J BoneJoint Surg 1988;70:530—3.


[23] Sunderland S. Axon regeneration and related changes. In Sunder-land S, editor. Nerves and Nerve injuries, Edimburgh: ChurchillLivingstone, 1968:91—113.

[24] Li Qiang G, Jia-Kai Z. Repair of different sized nerve defectsusing degenerated muscle grafts with vascular implantation; anexperimental study in the rat. J Reconstr Microsurg 1992;8:47—52.

[25] Best TJ, Mackinnon SE, Bain JR, Makino A, Evans PJ. Verifica-tion of a free vascularized nerve graft model in the rat withapplication to the peripheral nerve allograft. Plast Reconst Surg1993;92;516—25.

[26] Chiu DTW, Janecka I, Krizek TJ, Wolff M, Lovelace RE. Auto-genous vein graft as a conduit for nerve regeneration. Surg Forum1980;31:550—7.

[27] Chiu DTW, Janecka I, Krizek T, Wolff M, Lovelace RE. Auto-genous vein graft as a conduit for nerve regeneration. Surgery.1982;91:226—33.

[28] Chiu DTW, Lovelace RE, Yu LT, Wolff M et al. Comparativeelectrophysiological evaluation of nerve grafts and autogenousvein grafts as nerve conduits: an experimental study. J ReconstrMicrosurg 1988;4:303—9.

[29] Suematsu N, AtsutaY, Hirayama T. Vein graft for repair ofperipheral nerve gap. J Reconstr Microsurg 1988;4:313—8.

[30] Wang K-K, Costas PD, Jones DS, Miller RA, Seckel BR.Sleeve insertion and collagen coating improve nerve regen-eration through vein condiuts. J Reconstr Microsurg 1992;9:39—48.


nerve regeneration through a combined autologous conduit (vein plus acellular muscle grafts)

Documents