Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=ioto20

Acta Oto-Laryngologica

ISSN: 0001-6489 (Print) 1651-2251 (Online) Journal homepage: http://www.tandfonline.com/loi/ioto20

Facilitation of facial nerve regeneration usingchitosan-β-glycerophosphate-nerve growth factorhydrogel

Xiuhua Chao, Lei Xu, Jianfeng Li, Yuechen Han, Xiaofei Li, YanYan Mao,Haiqiong Shang, Zhaomin Fan & Haibo Wang

To cite this article: Xiuhua Chao, Lei Xu, Jianfeng Li, Yuechen Han, Xiaofei Li, YanYan Mao,Haiqiong Shang, Zhaomin Fan & Haibo Wang (2016) Facilitation of facial nerve regenerationusing chitosan-β-glycerophosphate-nerve growth factor hydrogel, Acta Oto-Laryngologica, 136:6,585-591, DOI: 10.3109/00016489.2015.1136432

To link to this article: https://doi.org/10.3109/00016489.2015.1136432

Published online: 16 Feb 2016.

Submit your article to this journal

Article views: 190

View Crossmark data

Citing articles: 4 View citing articles

ACTA OTO-LARYNGOLOGICA, 2016VOL. 136, NO. 6, 585–591http://dx.doi.org/10.3109/00016489.2015.1136432

RESEARCH ARTICLE

Facilitation of facial nerve regeneration using chitosan-b-glycerophosphate-nervegrowth factor hydrogel

Xiuhua Chaoa,b, Lei Xua,b, Jianfeng Lia,b, Yuechen Hana, Xiaofei Lia, YanYan Maoa, Haiqiong Shanga,b, Zhaomin Fana,b

and Haibo Wanga,b

aDepartment of Otolaryngology Head and Neck Surgery, Shandong Provincial Hospital Affiliated to Shandong University, Jinan, PR China;bShandong Institute of Otolaryngology, Jinan, PR China

ABSTRACTConclusion C/GP hydrogel was demonstrated to be an ideal drug delivery vehicle and scaffold inthe vein conduit. Combined use autologous vein and NGF continuously delivered by C/GP-NGFhydrogel can improve the recovery of facial nerve defects. Objective This study investigated theeffects of chitosan-�-glycerophosphate-nerve growth factor (C/GP-NGF) hydrogel combined withautologous vein conduit on the recovery of damaged facial nerve in a rat model. Methods A 5 mmgap in the buccal branch of a rat facial nerve was reconstructed with an autologous vein. Next,C/GP-NGF hydrogel was injected into the vein conduit. In negative control groups, NGF solution orphosphate-buffered saline (PBS) was injected into the vein conduits, respectively. Autologousimplantation was used as a positive control group. Vibrissae movement, electrophysiologicalassessment, and morphological analysis of regenerated nerves were performed to assess nerveregeneration. Results NGF continuously released from C/GP-NGF hydrogel in vitro. The recoveryrate of vibrissae movement and the compound muscle action potentials of regenerated facialnerve in the C/GP-NGF group were similar to those in the Auto group, and significantly better thanthose in the NGF group. Furthermore, larger regenerated axons and thicker myelin sheaths wereobtained in the C/GP-NGF group than those in the NGF group.

ARTICLE HISTORY

Received 4 November 2015Revised 9 December 2015Accepted 15 December 2015Published online12 February 2016

KEYWORDS

Chitosan-�-glyceropho-sphate hydrogel; nerveregeneration; NGF; nerveconduit; autologous vein

Introduction

Peripheral facial paralysis is a common but devastating diseasethat not only causes functional defects, but also has significantaesthetic and psychological impacts on the patient. Althoughmost peripheral facial paralysis, especially Bell’s palsy, canreach sufficient recovery [1], facial nerve defects caused bytrauma remain difficult to treat on clinics. Several types offacial reanimation surgery have been designed to reconstructfacial nerve defects [2,3], but satisfactory recovery of facialfunction has been elusive. Recently, various nerve conduits(NCs) have been investigated as an alternative approach toreconstruct peripheral nerve defects. NCs can provide an idealmicroenvironment for injured nerves, which is critical for theregenerating nerves. Among the various types of NCs,autologous veins, which are rich in extracellular matrix andcollagen that are beneficial for regenerating cells, are stilldeemed to be an ideal NC [4]. In addition, autologous veins arebiodegradable and present no side-effects compared with someartificial NCs, such as polycaprolactone [5,6]. However,isolated veins are prone to collapse. In previous studies,muscles or adipose tissues had been introduced into veinconduits to avoid luminal collapse [7]. However, these tissuesdegraded so slowly that they impeded the regeneration ofperipheral nerves [8]. In the past few decades, chitosan hasbeen widely used as a scaffold to improve the regeneration of

bones, and has achieved successful results [9]. In addition, ithas been confirmed that after chitosan is combined with�-glycerophosphate (GP) it becomes a pH-dependent andthermosensitive gel, which makes chitosan/�-glycerophosphate(C/GP) hydrogel an ideal injectable hydrogel [10]. Thus, oneaim of this study was to assess the use of C/GP hydrogel as ascaffold to maintain the luminal condition of isolated veinconduits.

Successful regeneration of peripheral nerves also requiresthe use of neurotrophins. Among neurotrophins, nerve growthfactor (NGF), which plays an important role in controlling thesurvival, proliferation, and migration of various cells involvedin nerve regeneration, is widely used in clinical applications[11]. However, NGF solutions are not stable under physio-logical conditions. In order to enhance the efficacy and stabilityof NGF, it has been added into various NCs [12]. To date, theseNGF seeded NCs have been complicated to generate. However,because the C/GP hydrogel has the ability to release drugs overa course of several days [13], it is possible that this hydrogelmay provide a means of delivering NGF to the nerve injurysite. In our previous study, we had achieved success in usingchitosan as a drug delivery vehicle to locally administermethylprednisolone to treat facial nerve crush injury [14].Therefore, a second aim of this study was to evaluate the effectsof locally administrate NGF by the C/GP hydrogel on nerveregeneration in a rat model.

CONTACT Haibo Wang whbotologic797@163.com; Zhaomin Fan fanent@126.com No.4, West Duanxing Road, Jinan 250022, PR China

� 2016 Taylor & Francis

In summary, the purpose of this study was to test thehypothesis that the structural support properties of C/GPcombined with its potential for targeted delivery of NGF makeit an optimal product to facilitate successful repair of damagednerves using an autologous vein conduit. The hypothesis wastested by assessing the benefit of C/GP-NGF hydrogel in therepair of facial nerve lesions using a rat model.

Materials and methods

Preparation of C/GP-NGF hydrogel

The chitosan hydrogel was generated by dissolving 20 mg ofchitosan (with medium viscosity and a degree of deacetylation ofmore than 90%, NovaMatrix�, Sandvika, Norway) in 600mL ofdouble-distilled water (ddH2O). The GP solution was made bydissolving 56 mg of GP in 100mL of ddH2O. Both the chitosanhydrogel and the GP solution were kept at 4 �C for 30 min. TheGP solution was then added to the chitosan hydrogel withstirring for 30 min on ice. The C/GP-NGF hydrogel wasobtained by adding NGF (Sigma-Aldrich, St. Louis, MO)solution (100 ng/mL) to the freshly prepared C/GP hydrogel.The pH of the prepared C/GP-NGF hydrogel was � 7.5.

Measurement of the in vitro release of NGF

In the in vitro experiment, 0.01 mL of C/GP-NGF hydrogel wasadded into an eppendorf tube and placed in a 37 �C incubatorfor 1 h. After the C/GP-NGF hydrogel turned to a gel, 1 mLphosphate-buffered saline (PBS, 0.01 M, pH 7.3) was addedinto the tube. Then the tube was maintained in an incubator at37 �C with a shaking rate of 50 rpm. At each time point (12 h,1 day, 2 days, 3 days, 5 days, 7 days, 9 days, and 12 days), therelease solution was removed from the tube, and an additional1 mL pre-warmed fresh PBS was added to the tube. Thecollected solution was stored at�80 �C until analysis. The NGFconcentration was determined by Enzyme-LinkedImmunosorbent Assay kit (ELISA, Chemicon International,Temecula, CA) using a detection wavelength for NGF of450 nm. The NGF concentration was re-tested another twotimes using newly made C/GP-NGF hydrogel each time. Thenthe average NGF concentration was calculated.

Animals and groups

Thirty female Wistar rats (220–250 g, provided by theLaboratory Animal Center of ShanDong University, Jinan,

PR China) were randomly divided into five groups (n¼ 6 ineach group). Group A was the autologous implantation group(Auto group); Group B was the C/GP-NGF group; Group Cwas the NGF group; Group D was the PBS group. Six rats wereused as the Normal control group. All animals were main-tained under veterinarian supervision, kept in a temperature-controlled room with a 12-h light/dark cycle, and providedwith standard rodent diet and water.

Facial nerve transection and reconstruction

All rats were anesthetized by an intraperitoneal injection of10% chloral hydrate mixture (4 mL/kg body weight, producedby ShanDong Provincial Hospital, China). A vertical 2 cmincision near the middle of the neck was made, and theexternal jugular vein was exposed (Figure 1A). Then, about10 mm of the left external jugular vein was excised from thesurrounding tissues, and immediately put into sterile 0.9%normal saline.

After harvesting the segment of external jugular vein, theshaved right facial skin was opened with a preauricularincision, and the buccal branch and the marginal mandibularbranch of the facial nerve were exposed under an operatingmicroscope (Leica, M525 F40, Germany). A 4 mm longsegment of the buccal branch nerve was removed 1 cm distallyto the bifurcation of the main nerve trunk near thestylomastoid foramen. After retraction, a 5 mm gap remainedat the buccal branch (Figure 1B). In group A, the excised nervestem was reversed and re-implanted with 11-0 nylon sutures.In groups B–D, the transected nerves were repaired using theobtained veins, and 0.01 mL of C/GP-NGF (containing 500 ngNGF, group B) gel, 0.01 mL of NGF solution (group C), and0.01 mL PBS (group D) were, respectively, injected into thevein conduits using a sterile syringe (Figures 1C and D). Inaddition, to assess the vibrissae movement, the marginalmandibular branch was also resected in all rats, and theproximal segment was ligated. Finally, all wounds were closedwith 3-0 nylon silk sutures. After the rats awoke they werereturned to their cages.

Vibrissae movement evaluation

Vibrissae movement was graded to assess the whiskingbehavior of the rats. The vibrissae movement grade can bedivided into five different categories, with scores ranging from0–4 (0 ¼ no whisker movement; 1 ¼ slight whisker movement;

Figure 1. Surgical photographs of reconstructing facial nerve with vein conduit. (A) The left external jugular vein. (B) The 5-mm gap in the right buccal branch. (C) Thenerve stumps at each end were inserted 0.5 mm into the vein conduit, and the epineurium of each nerve stump was anchored into the vein. (D) A total of 0.01 mL C/GP hydrogel was injected into the vein conduit.

586 X. CHAO ET AL.

2 ¼ slow movement; 3 ¼ rapid movement, but distinguishablefrom the contralateral normal side; and 4 ¼ symmetricmovement compared to the normal side). The vibrissaemovement evaluation was performed at the 4th, 8th, and12th weeks after the operation. To avoid subjective bias, theassay was performed in a double-blind manner.

Electrophysiological assessment

At the 12th week after the operation, after rats were completelyanesthetized, the repaired buccal branches of facial nerves wereexposed. To test the compound muscle action potentials(CMAPs) of regenerated facial nerves, two stimulation elec-trodes were placed next to the buccal branch just at theproximal segment of the repaired nerve, and two recordingelectrodes were placed at the middle of the vibrissae row. Also,the ground electrode was inserted at the front of the head. Thefacial nerve was stimulated at a maximum of 0.3 mA, and themaximal CMAPs were recorded. Ten CMAP waves wereobtained for each rat, and the peak amplitude and latency wererecorded. Normative values of the CMAPs were obtained forthe normal control group.

Histological observation for nerve regeneration

After the electrophysiological examination, the rats weresacrificed under overdose anesthesia. Then they were trans-cardiac perfused with saline followed with 4% (v/v) parafor-maldehyde in 0.1 M PBS (pH 7.2). The regenerated facialnerves were removed from each of the animals. Normal facialnerves were also obtained as control.

The distal parts of the regenerated nerves were embedded inparaffin. Then these nerves were transversely cut into 4mmthick sections. For modified trichrome staining, the sectionswere first stained with Harris haematoxylin for 5 min, andwashed with water. Then, the sections were soaked intrichrome solution for 30 min, then washed with water anddehydrated with gradient ethylalcohol. The slides were viewedunder a light microscope (Leica) after being covered withneutral gum.

The proximal halves of the regenerated nerve segments wereembedded in epoxy resin. Transverse semi-thin sections(2 mm) were cut with an ultra-microtome (Leica) followed by1% toluidine blue staining. The prepared sections wereexamined under a light microscope (Leica), and photographswere taken from each of the semi-thin sections. The diameterof myelinated fibers, the myelination thickness, and thequantity of regenerated axons were then calculated with theImage J software. In addition, transverse ultrathin sections(0.5 mm) were also cut and stained with lead citrate and uranylacetate followed by examination under a transmission electronmicroscope (JEOL, Japan).

Statistics

The data were expressed as means� standard error of themean (SEM). Statistical analysis was performed with one-way

ANOVA and Scheffe’s post-hoc test. Values of p50.05 wereconsidered statistically significant.

Results

NGF release curve measured by ELISA kit

The amount of NGF released from the C/GP-NGF hydrogelwas analyzed by an ELISA kit (Figure 2). Over the first 12 h,� 60% of the NGF was released from the hydrogel. From Day1 to Day 3, the release rate slowed, and �13 ng NGF wasreleased every day. After 3 days, the release rate slowed againand NGF was steadily released each day. By day 12, nearly 80%of the NGF had been released from the C/GP hydrogel.

Functional recovery of vibrissae movement

Figure 3 shows the time course of vibrissae movement recovery.After the buccal and marginal mandibular branches weretransected, all rats lost their whisking ability. Then the functionof vibrissae movement recovered gradually with the regener-ation of buccal branch. Better recovery of vibrissae movementwas obtained in the Auto group and the C/GP-NGF group, and

Figure 2. NGF release curve. The NGF released from the C/GP-NGF hydrogel inPBS was measured by ELISA kit over 12 days. Each point represents the meanvalue� SEM (n¼ 3).

Figure 3. Vibrissae movement grade. The vibrissae movement was assessed atthe 4th, 8th, and 12th weeks after operation. The recovery rate of vibrissaemovement in the Auto and C/GP-NGF groups was faster than in the NGF and PBSgroups at all-time points (p50.05). �p50.05 compared to the NGF group, and*p50.05 compared to the NGF group.

ACTA OTO-LARYNGOLOGICA 587

no significant differences were seen between these two groups.The recovery rates of the vibrissae movement in the Auto andC/GP-NGF groups were significantly faster than the rats in theNGF and PBS groups (p50.05). No significant differences wereseen between the NGF and PBS groups (p40.05).

Improvement in the amplitude and latency of CMAPs

The normal latency and amplitude of CMAPs in rats’ facialnerves were 1.38� 0.01 ms and 18.5� 0.05 mV, respectively.The normal facial nerve’s nerve conduction velocity (NCV)calculated from the latency was 18.2 m/s (Figure 4). Twelveweeks after operation, the fastest recovery of CMAPs wereobtained in group A, but the NCV was still significantly slowerthan in the Normal control group (p50.05). Although theCMAPs in the Auto group appeared to be better than in theC/GP-NGF groups, there was no significant difference(p40.05). The NCV and amplitude in the C/GP-NGF groupwere significantly better than those in the NGF and PBSgroups (p50.05). The CMAPs in the NGF group were similarto those in the PBS group.

Modified trichrome staining of regenerated facialnerves

Figure 5 shows the regenerated facial nerves stained bymodified trichrome staining at the 12th week after injury. In

normal facial nerves, myelinated axons were neatly displayed,and the axons and myelin were stained into blue and red,respectively. Twelve weeks after repair, all vein conduits werepartially degraded and integrated into the surrounding tissueswithout serious side-effects. Relatively mature myelinatednerves could be seen in all experimental groups. In the Autogroup, myelinated nerves were neatly and regularly ordered. Inthe other groups, regenerated nerves were a little disordered,and abundant proliferated fibrous tissue and inflammatorycells were also observed, especially in NGF and PBS group. Itwas shown in the sections that the areas of the regeneratednerves in the Auto and C/GP-NGF group were larger thanthose in the NGF and PBS group. No chitosan gel could beseen in sections of C/GP-NGF group.

Morphological evaluation of regenerated nerves

Figure 6 shows the micrographs of regenerated facial nervesand the statistical analyses of the regenerated fibers at the 12thweek after the operation. In normal facial nerves, myelinatedfibers were clearly and neatly displayed. Regenerated fiberscould be seen in all experimental groups, but the diameter offibers and the thickness of myelin sheaths were significantlyless than those in the normal group. Additionally, someSchwann cells and newly forming myelin sheaths were alsoobserved. The diameter of fibers and the thickness of myelinsheaths in the C/GP-NGF group were significantly larger

Figure 4. Electrophysiological assessment of facial nerve function. The left column shows representative CMAPs traces for all groups at the 12th week after theoperation. The right column shows the analysis of amplitude and NCV of the CMAPs. *p50.05 compared with the Normal group; mp50.05. (A) Auto group; (B) C/GP-NGF group; (C) NGF group; (D) PBS group; (N) Normal group.

588 X. CHAO ET AL.

compared to the NGF group (p50.05), but still smaller thanthe Auto group (p50.05). Due to larger myelinated axons, thedensity of regenerated fibers in the C/GP-NGF and Autogroups was not significantly superior to in the NGF and PBSgroups (p40.05). In addition, the number of axons in thewhole regenerated buccal branch was calculated. Comparedwith the NGF and PBS groups, the quantity of myelinatedaxons in the regenerated buccal branch was significantlysmaller to the C/GP-NGF and Auto groups (p50.05). Themyelinated fibers in the NGF group were not superior to thePBS group.

Discussion

Over the past few decades, several informative experimentshave been conducted to investigate optimal modalities withwhich to repair peripheral nerve injuries. Up to now, however,the repair of transected peripheral nerves, especially the nervedefects, remains a formidable challenge in terms of basicresearch and clinical practice because of the lack of safe andefficient materials [15]. In the present study, we have for thefirst time fabricated a novel vehicle composed of threeelements: (1) autologous vein, which served as a nerve

Figure 6. Microstructure of the regenerated facial nerves. The first column shows the morphology of regenerated nerves by light microscopy; the second columnshows the morphology of regenerated nerves by transmission electron microscopy; and the third column displays the statistical analysis of the myelin sheaththickness, the myelinated axon diameter, the regenerated axons density, and myelinated axon number in buccal branch (n¼ 6). (A) Auto group; (B) C/GP-NGF group;(C) NGF group; (D) PBS group; (N) Normal group. *p50.05 compared with the Normal group; mp50.05. The scale bar is 5 mm in the semi-section micrograph, and is20mm in the ultra-thin micrograph.

Figure 5. Modified trichrome staining of regenerated facial nerves. In normal facial nerves, axons were stained into blue and myelin were stained into red by themodified trichrome staining (N1 and N2). (A) Auto group; (B) C/GP-NGF group; (C) NGF group; (D) PBS group; (N) Normal control group. A2–D2 and N2 are highermagnifications of the square areas in A1–D1 and N1. The scale bar is 20mm in N1, and 5mm in N2. (Yellow arrows: regenerated axons; red arrows: inflammatory cells).

ACTA OTO-LARYNGOLOGICA 589

guidance channel; (2) C/GP hydrogel, which provided asupportive matrix inside the vein; and (3) NGF, which actedas a promoter for nerve proliferation. This combined veinconduit integrated the advantages of each element, which wespeculated would provide an appropriate microenvironmentfor the regeneration of injured nerves.

As the C/GP-NGF hydrogel mixture was first used for therefinement of autologous vein conduit, we initially performedan in vitro experiment to evaluate the structural and functionalcharacteristics of this matrix. First, the quantitative analysis ofNGF concentration showed that NGF was steadily releasedfrom the C/GP-NGF hydrogel over several days. This indicatesthat the C/GP hydrogel might possess the ability to maintainNGF at an efficient and stable concentration in the veinconduit. This controlled drug release characteristic of C/GPhydrogel is in accordance with previous studies [16].Moreover, the C/GP-NGF hydrogel made in this study canalso turn into a semi-solid state after being incubated at bodytemperature, which has been described in previous studies[17]. This character may allow the C/GP-NGF hydrogel to staywithin the vein conduit and prevent luminal collapse of theisolated veins in vivo. Actually, during the surgery, we foundthat after C/GP-NGF hydrogel was injected into the autologousveins, the veins maintained their shape. Taken together, theseproperties showed that C/GP-NGF hydrogel was an efficaciousmaterial for combining a support matrix with a drug deliveryvehicle. Our results laid a feasible foundation for oursubsequent experiments in vivo.

Next, the vein conduit filled with C/GP-NGF hydrogel wasassessed for efficacy via grafting to bridge a 5 mm long gap inthe facial nerve of rats. In our study, all vein conduits weresuccessfully integrated into the host tissue without serious side-effects. In addition, the light microscopy results showed thatthe nerve conduit was full of regenerated fibers, and there wereno chitosan hydrogels in it. This indicates that the C/GPhydrogel can be biodegrade at an appropriate rate withoutblocking the regenerating axons. This property of C/GPhydrogel is superior to the skeletal muscles or adipose tissueswhich blocked the vein conduits in reconstructing theperipheral nerves [18,19]. Furthermore, the CMAPs ofregenerated facial nerves obtained in the C/GP-NGF groupwere similar to those in the Auto group, and superior to thosein the NGF group. This indicates that NGF delivered by C/GP-NGF hydrogel is superior to NGF solution alone for restoringthe function of the damaged nerves in the vein conduits. Ourresults reaffirmed the in vitro findings that C/GP hydrogelpossessed a unique sustained release characteristic after theinjection into the vein, i.e. distributing NGF in a time-dependent manner.

Now that grafting to bridge the nerve defect with thisvehicle leads to functional improvement of the nerves in rat, itis conceivable that the functional recovery of injured nervesdepends on the axonal regeneration and the reinnervation oftarget organs [20]. Indeed, larger axons, thicker myelinsheaths, and larger quantity of regenerated fibers wereobtained in the C/GP-NGF group than those in the NGFgroup in this study. This remodeling might contribute to betterfunctional recovery in the C/GP-NGF group than in the NGFgroup. Despite the maturity of regenerated axons in the

C/GP-NGF group were still inferior to those in the Auto group,large quantity of axons might also contribute to the functionalrecovery. These morphological improvements in the C/GP-NGF group might be due to the sustained release of NGF fromthe C/GP-NGF hydrogel, which further demonstrates thebenefit of continuously released NGF for the regeneration offacial nerves. As NGF solution is not stable under physiologicalconditions, it would flow away after being injected into thevein conduit. In this study, we didn’t see any functional ormorphological improvement in the NGF group compared tothe PBS group. Together, these results indicate that usingautologous veins combined with C/GP-NGF hydrogel canimprove the regeneration of damaged facial nerves.

Conclusion

In this study, C/GP-NGF hydrogel was injected into anautologous vein conduit to reconstruct facial nerve defects in arat model. Although they failed to provide equivalent mor-phological recovery to nerve autografts, satisfying functionalrecovery of regenerated facial nerves was achieved. The C/GP-NGF hydrogel can not only continuously release NGF at localsites but also act as a scaffold in the vein conduit. Thefunctional and morphological recovery of the damaged facialnerve was improved by locally administration of C/GP-NGFhydrogel instead of NGF solution. Overall, combined use of theC/GP-NGF hydrogel and autologous vein conduit may repre-sent a promising strategy for alternative therapy for facialnerve defects in the future.

Acknowledgements

This work was supported by grants from the National Natural ScienceFoundation of China (no.81100711 and no.81170907), the 12th Five-YearNational Key Technologies R&D Program (no. 2012BAI12B01), the MajorState Basic Research Development Program of China (973 Program, no.2015CB965000), and Youth Fund Project of Shandong provincial Medicaltechnology development plan (2011WSB02002).

Disclosure statement

The authors report no conflicts of interest. The authors alone areresponsible for the content and writing of the paper.

References

[1] Ikeda M, Abiko Y, Kukimoto N, Omori H, Nakazato H, Ikeda K.Clinical factors that influence the prognosis of facial nerve paralysisand the magnitudes of influence. Laryngoscope 2005;115:855–60.

[2] Beutner D, Luers JC, Grosheva M. Hypoglossal-facial-jump-anas-tomosis without an interposition nerve graft. Laryngoscope2013;123:2392–6.

[3] Slattery WH, III, Cassis AM, Wilkinson EP, Santos F, Berliner K.Side-to-end hypoglossal to facial anastomosis with transposition ofthe intratemporal facial nerve. Otol Neurotol 2014;35:509–13.

[4] Acar M, Karacalar A, Ayyildiz M, Unal B, Canan S, Agar E, et al.The effect of autogenous vein grafts on nerve repair with sizediscrepancy in rats: an electrophysiological and stereologicalanalysis. Brain Res 2008;1198:171–81.

[5] Tos P, Battiston B, Ciclamini D, Geuna S, Artiaco S. Primary repairof crush nerve injuries by means of biological tubulization withmuscle-vein-combined grafts. Microsurgery 2012;32:358–63.

590 X. CHAO ET AL.

[6] Hsu SH, Kuo WC, Chen YT, Yen CT, Chen YF, Chen KS, et al. Newnerve regeneration strategy combining laminin-coated chitosanconduits and stem cell therapy. Acta Biomater 2013;9:6606–15.

[7] Brunelli GA, Battiston B, Vigasio A, Brunelli G, Marocolo D.Bridging nerve defects with combined skeletal muscle and veinconduits. Microsurgery 1993;14:247–51.

[8] Dornseifer U, Fichter AM, Leichtle S, Wilson A, Rupp A,Rodenacker K, et al. Peripheral nerve reconstruction with collagentubes filled with denatured autologous muscle tissue in the ratmodel. Microsurgery 2011;31:632–41.

[9] Sajesh KM, Jayakumar R, Nair SV, Chennazhi KP. Biocompatibleconducting chitosan/polypyrrole-alginate composite scaffold forbone tissue engineering. Int J Biol Macromol 2013;62:465–71.

[10] Crompton KE, Goud JD, Bellamkonda RV, Gengenbach TR,Finkelstein DI, Horne MK, et al. Polylysine-functionalised thermo-responsive chitosan hydrogel for neural tissue engineering.Biomaterials 2007;28:441–9.

[11] Pfister LA, Papaloizos M, Merkle HP, Gander B. Nerve conduitsand growth factor delivery in peripheral nerve repair. J PeripherNerv Syst 2007;12:65–82.

[12] Xu H, Yan Y, Li S. PDLLA/chondroitin sulfate/chitosan/NGFconduits for peripheral nerve regeneration. Biomaterials2011;32:4506–16.

[13] Sinha VR, Singla AK, Wadhawan S, Kaushik R, Kumria R, Bansal K,et al. Chitosan microspheres as a potential carrier for drugs. Int JPharm 2004;274:1–33.

[14] Chao X, Fan Z, Han Y, Wang Y, Li J, Chai R, et al. Effects oflocal application of methylprednisolone delivered by the C/GP-hydrogel on the recovery of facial nerves. Acta Otolaryngol2015;135:1178–84.

[15] Konofaos P, Ver Halen JP. Nerve repair by means of tubulization:past, present, future. J Reconstr Microsurg 2013;29:149–64.

[16] Hsiao MH, Larsson M, Larsson A, Evenbratt H, Chen YY, Liu DM.Design and characterization of a novel amphiphilic chitosannanocapsule-based thermo-gelling biogel with sustained in vivorelease of the hydrophilic anti-epilepsy drug ethosuximide. JControl Release 2012;161:942–8.

[17] Ruel-Gariepy E, Chenite A, Chaput C, Guirguis S, Leroux J.Characterization of thermosensitive chitosan gels for the sustaineddelivery of drugs. Int J Pharm 2000;203:89–98.

[18] Papalia I, Raimondo S, Ronchi G, Magaudda L, Giacobini-RobecchiMG, Geuna S. Repairing nerve gaps by vein conduits filled withlipoaspirate-derived entire adipose tissue hinders nerve regener-ation. Ann Anat 2013;195:225–30.

[19] Raimondo S, Nicolino S, Tos P, Battiston B, Giacobini-RobecchiMG, Perroteau I, et al. Schwann cell behavior after nerve repair bymeans of tissue-engineered muscle-vein combined guides. J CompNeurol 2005;489:249–59.

[20] Krarup C, Archibald SJ, Madison RD. Factors thatinfluence peripheral nerve regeneration: an electrophysio-logical study of the monkey median nerve. Ann Neurol2002;51:69–81.

ACTA OTO-LARYNGOLOGICA 591

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具