An alternative to nerve grafts in peripheral nerve repair: Nerve guides seeded with adult Schwann cells

Introduction

Nerve grafts remain the most successful means of bridging large gaps in peripheral nerve following a lesion. This is despite the fact that their use requires the sacrifice of functional nerve to achieve the repair, that reinnervation may be unsatisfactory, and neuroma and scar tissue are common(1,2). Consequently, interest in the use of alternatives to conventional grafts remains high.

The promise of "nerve guides", tubes used to bridge the proximal and distal portions of a severed nerve, has been shown in a number of studies. In guides 3 to 8mm in length, relatively less aberrant axonal growth(3-6), fibrous scar tissue, and neuromas(6,7) have been reported in comparison to nerve grafts. This enhanced regeneration appears to be due in part to a reduction in fibrotic tissue, but also because the guide chamber provides a storage space for the trophic factors secreted by the distal stump(8-10) and a space for the axons to navigate, find and enter the appropriate distal Schwann tube(11).

Unfortunately, few reports use guide lengths >15mm, nor are the results comparable to autografts even when using supplementation with growth factors such as laminin(12), NGF(13), and FGF(14).

Why do longer nerve guides fail to support regeneration? Firstly, guides lack Schwann cells. Nerve grafts have Schwann cells, and it is well established that Schwann cells play a critical role in degeneration and regeneration. In response to injury, Schwann cells separate from the axons they ensheath, proliferate rapidly, and align to create bands of Büngner(2). The activated Schwann cells synthesise growth factors such as Insulin-like growth factor-I (IGF-I)(15), ciliary neurotrophic factor (CNTF)(16), Brain-derived neurotrophic factor (BDNF)(17), which guide and support the regenerating axons. In the absence of Schwann cells, the distal stump may be unable to produce enough trophic factors to reach the proximal stump.

Guides also lack the structural support provided by the basal lamina which is crucial to successful regeneration(18). Instead, a fibrin matrix bridge must form between the stumps, allowing Schwann cells to migrate into the cavity from both ends(19). Without Schwann cells, regenerating axons fail to penetrate into the guide(20,21). It has been suggested that guides longer than 10mm fail because of inadequate matrix formation in the early stage of regeneration(19).

The failings of nerve guides might be overcome, if, instead of leaving a guide empty, or filling the guide with neurotrophic factors (which in time run out), Schwann cells are inserted into the cavity, acting a a continuing source of trophic factors. It is also probable that seeding with Schwann cells accelerates the formation of a bridge giving support to growth cones seeking the distal stump. In this study, we have investigated this possibility using a circumferentially rigid, but longitudinally flexible guide with a collagen surface; a surface suited to axonal outgrowth(22), seeded with adult Schwann cells. Only adult cells would be compatible with future application of this method to humans.

Materials and Methods

Young adult inbred (syngeneic) Wistar rats of either sex were used as donors and recipients of the Schwann cells.

Cell cultures: Cultures of adult Schwann cells were prepared from the sciatic nerve of Wistar rats 3 - 6 months old. The method used to establish adult Schwann cell cultures has been described in the previous paper(23). Briefly, in each donor animal, a conditioning lesion was administered to one sciatic nerve. Eleven days after surgery, the animals were killed with an overdose of anaesthetic. The conditioned nerve was excised, stripped of epineural sheaths, chopped into small tissue blocks, then enzymatically dissociated. The cells were plated on poly-L-lysine/laminin coated glass coverslips and maintained in complete medium, supplemented with several additives. The cells were maintained in complete medium for 7 to 12 days. Prior to harvesting, the cultured cells were labelled with Hoechst dye (H33342, 8µmol/ml for 30 minutes, Calbiochem) and harvested using trypsin-EDTA solution (Sigma). A sample was used to estimate the total numbers of cells. The remaining cells were resuspended in 100µl of complete medium.

Guide insertion: Sterile 20mm collagen nerve guide (BioNova Neo Technics Pty, Ltd, Melbourne; internal diameter 1mm) manufactured as a vascular prosthesis(24), was implanted in a gap created in the left sciatic nerve of the recipient animal (n = 21), using microsurgical techniques. The stumps were inserted 1 mm into the guides, for a final gap of 18 mm, and held in place with three epineurial 10/0 microsutures. The edges were sealed with Vaseline. Before inserting the distal stump, the guides were filled with either a suspension of Schwann cells (n = 18), or with phosphate buffered saline (PBS, n = 3).

Behavioural functional assessment: The techniques used to accomplish the walking track and nerve conduction studies have been described in detail elsewhere(25). Briefly, motor function was assessed behaviourally using the toe spread reflex and the sciatic functional index (SFI), a non-invasive method of assessing the functional status of the sciatic nerve based on measurement of the animal's walking tracks, expressed in units of functional deficit(26).

Nerve conduction measurements: The pre-operative and post-operative nerve conduction velocity of the sciatic nerve of both hind legs was assessed in the anaesthetised experimental animals. Briefly, the sciatic nerve was stimulated at the sciatic notch and at the ankle using two pairs of sub-cutaneous stainless steel needle electrodes. Supra-maximal square-wave impulses 0.1 ms in duration were delivered at 1 Hz or 10 Hz using a microprocessor-based system (Medilec MS91). Evoked compound muscle action potentials were recorded from the interosseous muscles of the foot. The distance between the stimulating electrodes was measured on the skin to the nearest 1 mm. The latency was measured to the onset of the compound action potential, thus giving a measure of the conduction velocity in the fastest fibres.

Histology: At the end of the recovery period, the animals were deeply anaesthetised and perfused intra-cardially with 4% paraformaldehyde in phosphate buffer (pH 7.3). The nerve guides, proximal and distal stumps were excised. The guides were divided longitudinally to expose the guide contents. If a bridge had formed across the gap, one half was frozen in liquid nitrogen and 10 µm frozen sections were obtained for fluorescence microscopy. The other half was post fixed with 2% glutaraldehyde and 2% acrolein in phosphate buffer (pH 7.3) for a further 24-48h, then processed for electron microscopy.

For the proximal and distal stumps, thin sections were taken approximately 5 mm away from the guide insertion. Thin sections were obtained from the middle section of the guide. Thick and thin sections were examined using light and electron microscopy.

Morphometry: The total area of the cross-section of the nerve was calculated using the reconstruction of the whole transverse section at the light microscope level. For the preparations of less than 4 months regeneration, the number of myelinated axons was estimated by counting all the axons with myelin sheaths within all the grid squares of each section and the percentage area sampled was calculated. The number of non-myelinated fibres was estimated using a systematic random sampling technique applied to micrographs printed at a final magnification of 13x10³(25). The criteria used for distinguishing non-myelinated axons was that the axonal profile was surrounded by a basal lamina and the presence of neurofilaments. The systematic random sampling technique technique was used to assess the total number of regenerated myelinated axons and non-myelinated axons in long-term animals.

Results

The 3 control guides lacked a matrix bridge and failed to support regeneration across the gap. All the guides seeded with Schwann cells (n=21) showed varying numbers of regenerated axons, which correlated with the number of Schwann cells implanted into the guides. Those animals receiving guides seeded with less than 0..5×10⁶ Schwann cells are denoted Group S. Those animals receiving guides filled more than 0.5×10⁶ are denoted Group L. The limited amount of the experimental tubing used to form nerve guides meant once we determined this difference, we subsequently concentrated on guides seeded with more than 0.5×10⁶ cells. 6 animals formed Group S and 12 animals are in Group L. 6 guides were analysed during the first month post-operatively (2 S and 4 L); 3 (L) were analysed at 2 months, 5 were analysed at 3 months (3 L and 2 S); and 4 were analysed at 6+ months (2 L and 2 S). Where numbers warranted, statistical analysis was carried out using mean ± S.E. and the Mann-Whitney U-test for significance.

Functional evaluation

Sciatic functional index: The long-term animals showed a small improvement in the functional use of the operated leg 5 months after surgery compared to 7 months in the animals of Group S. This was seen as an improvement in the toe-spread of the two Group L animals. There was no difference between the two groups by the end of the study (Fig. 5A).

Conduction velocity: Compound muscle action potentials (CMAP) could be detected during the 4th month of recovery in the two long-term Group L animals (17 m/s and 5 m/s respectively), but not in either Group S animals. CMAPs were recorded at month 7 in the Group S animal still surviving (29 m/s). From then till the end of the study, there was no difference in the conduction velocity of the two animals (1L and 1S), which remained at approximately 60% of their control values (unoperated leg; 44.7 ± 1.32m/s).

Qualitative Histology

There was a clear division in the degree of regeneration observed histologically between Group S and Group L.

Group L: During the first month, the guides had a fibrin matrix bridge populated with non-myelinated axons, assessed by electron microscopy. The regenerated axons were surrounded by many Hoechst labelled cells (Fig. 1) indicating that implanted Schwann cells had survived and were associated with the regenerated axons. Wallerian degeneration was well advanced in the distal stump (not shown here). Small myelinated axons were observed within the guide and the distal stump, surrounded by numerous non-myelinated axons.
In the second month, the numbers of myelinated and non-myelinated axons increased within the guide and distal stump. Hoechst labelled Schwann cells were still found within the guide throughout this period indicating that the implanted Schwann cells had persisted.
In the third month, the most obvious feature was a decrease in the large groups of non-myelinated fibres and the spreading out of the axon-Schwann cell units. Hoechst labelled cells were still evident in the guides at this time. The myelinated fibres had normal morphology.
The regenerated axons in the guides and distal stumps of the long-term animals (6+ months) were comparable in morphology (Fig. 2C & D) to regenerated nerves which had been repaired microsurgically using nerve grafts(25).

Group S: These guides contained neither fibrin matrix nor labelled Schwann cells during the first month.
Guides assessed during the third month showed a protein matrix bridge populated with Schwann cells enveloping small groups of non-myelinated axons. Signs of Wallerian degeneration were still evident in the distal stumps, but non-myelinated and myelinated axons were also found (Fig. 2B). Labelled Schwann cells were not seen within the guides of any of these preparations. There was no difference in the morphology of the regenerated axons in the guides and distal stumps of the long-term animals when compared to Group L (Fig. 2A & B).

Quantitative Histology

Axon counts: Axon counts were carried out using the distal stumps for 2 reasons:

1. Axons that had reached the distal stump were an indicator of successful regeneration across the guide.
2. Whole cross-sectional areas were only available from distal stumps (see methods), therefore assessing axonal numbers was only possible in these preparations.

Group L: At the early stages of regeneration there were signs of prolific sprouting of axon collaterals in the guides and the distal stumps, seen as a 10× increase in the number of non-myelinated axons estimated when compared to normal, unoperated nerves. Peak numbers of non-myelinated axons were found at 2 months (Fig. 3B), followed by a sharp drop. The number of myelinated axons was small in the early stages of recovery, but increased substantially between the third and sixth month (see Fig. 3A) while non-myelinated axon numbers decreased. The long term animals still showed an increased number of myelinated axons (12.5×10³) when compared to normal nerves (approximately 7×10³)(25).

Group S: The number of non-myelinated axons peaked at 3 months post-operatively (see Fig. 3B), one month later than Group L, followed by a drop. Furthermore, the number of non-myelinated fibres was substantially less between 1 and 3 months when compared to Group L. Long term animals showed a similar number of non-myelinated axons to Group L which remained substantially larger than what is found in normal sciatic nerves (approximately 16×10³)(25).

The number of myelinated axons was small at the early recovery periods, but had increased to about half normal values in the long-term animals (see Fig. 3A).

Axon diameters: There was no significant difference in the mean myelinated axon diameter distribution between group S (1.84µm±0.08) and group L (1.87µm±0.07) in the long-term animals (Fig. 4). The distribution of regenerated fibres in both groups is consistent with observations in lesioned nerves repaired with conventional microsurgical techniques(27).

Myelin thickness: The g-ratio (axon diameter/total fibre diameter) is a parameter which is closely related to conduction velocity(28). At optimum conduction velocity, most myelinated axons have a g-ratio between 0..6 and 0.7(29). High g-values indicate the axon to be thinly myelinated and low g-values to be thickly myelinated. Most regenerated axons stood within normal g-values, but a few axons with diameters less than 1µm had low g-values which indicated over myelination in respect to their diameter, were found in both groups. The mean g-values (± s.e.) of Group S myelinated axons (0.61 ± 0.09) were significantly smaller than Group L axons (0.64± 0.11, p<0.001)) but the difference was still within the normal range.

Discussion

This is the first study reporting the use of guides seeded with adult Schwann cells to bridge gaps of more than 15mm in adult peripheral nerve. Previous studies have used nerve guides seeded with neonatal Schwann cells to bridge gaps of less than 10mm(30-32).

This study demonstrates that it is possible to culture adult Schwann cells to seed a nerve guide which results in the enhancement of regeneration of peripheral axons across an interstump gap of 18 mm; a length that normally blocks regeneration. Furthermore, the survival of the Hoechst-labelled cells within the guides suggests that the cells participated in the regenerative process for at least the first 3 months.

The use of cultured adult Schwann cells is an important factor, since it shows the clinical potential of such an application. Peripheral nerve repair in humans requires that autologous Schwann cells be used since allogenic Schwann cells would be rejected without immunosuppression(33) which is not a preferred option(34). The results also demonstrate that culturing the Schwann cells has not had an adverse effect on the functional potential of the cells. The resulting regenerated axons were functionally and morphologically comparable to regenerated axons in autographs.

The study also shows that the number of cells used to seed the guides is important specifically in the early stages of regeneration. Too few cells results in delayed and limited regeneration. This was evident in the Group S animals (guides seeded with less than 0.5×10⁶ Schwann cells). There was no evidence of the pre-labelled implanted cells, and the presence of regenerated axons in the guides was delayed by one month when compared to the Group L animals (guides seeded with more than 0.5×10⁶ Schwann cells). The seeded cells in Group S guides may have either proliferated to such an extent that the labelling was no longer resolvable, or they may have survived for a limited time only, after implantation. Nevertheless their presence in the initial recovery period must have had a stimulatory effect, since axons were found in substantial numbers in the distal stumps of animals 3 months or more after implantation, unlike the control guides (see Fig. 3).

It was evident however that having more than 0.5×10⁶ Schwann cells in the guides (Group L) had a greater beneficial effect. Axon counts in the distal stumps of Group L animals suggest that the implanted Schwann cells initiated extensive sprouting and branching of axons. Although the high numbers decreased with time, they were still above normal values in the long-term animals. As expected, the numbers of regenerated myelinated axons in the distal stump were low in the early stages of recovery, but increased with time reaching about half normal numbers in the Group S animals, and well above normal values in the Group L animals. This is where the difference between the two groups was most evident. While regeneration was successful in both groups, the use of more Schwann cells to seed the guides resulted in a greater number of regenerated myelinated fibres in the distal stump. Presumably this would increase the chances of functional recovery.

The quality of the regeneration is best evaluated by comparison with nerve grafts, the more traditional form of nerve repair. When assessing the distribution of axon calibre in the regenerated distal stump, from 15 to 25% of the total regenerated myelinated axons observed were less than 1µm in diameter. Most of the small axons were relatively over-myelinated for their calibre, which may have been the result of excessive branching. Increased numbers of the small myelinated axons are commonly found in the distal stumps of nerve repaired with either guides or nerve grafts(35-37). The overall reduction in axon calibre is a phenomenon observed in all cases of regenerated axons, and is thought to be partly associated with the deprivation of terminal connections during the regenerative process(38,39), but also by other factors such as the increased collagenation, endoneurial shrinkage and delayed effects on axonal somas(40).

Despite the preponderance of smaller myelinated axons, the mean g-ratio of the myelinated axons within the distal stumps of Schwann cell seeded guides was within the optimal range (0.6 to 0.7;(29)). This suggests that the environment found in guides seeded with Schwann cells is similar to the one provided by nerve grafts.

The return of function, evaluated with the onset of nerve conduction and SFI, as delayed by about 2 months in Group S, when compared to the Group L animals (see Fig. 5). Interpreting the sciatic functional index (SFI) is not without problems(41), but in conjunction with the nerve conduction studies, it does suggest that the use of more Schwann cells within the guides accelerates functional recovery. Once functional recovery had been established, the difference was no longer evident. It is to be noted that the long-term animals maintained a 40% deficit in nerve conduction velocity (NCV) in the operated leg with respect to the control (unoperated) leg, regardless of the number of seeded cells. Such deficits are not uncommon and have been reported in other cases of regeneration following conventional nerve repair, especially when long nerve grafts are used. The decreased NCV is one of the consequences of the loss of the large diameter fibres, a phenomenon reported in other traumatic nerve lesions, see for example(42-45).

In conclusion, the results of this study show that axons have the capacity of regenerating across a large gap given an appropriate environment, namely a biosynthetic nerve guide seeded with syngeneic Schwann cells. The regenerated motor axons demonstrated conduction properties comparable to that obtained with nerve grafts, and there was recovery of function.

The importance of the implanted Schwann cells is greatest in the earlier stages of recovery. The advantage of seeding nerve guides with larger numbers of Schwann cells was particularly noticeable in the decreased onset of the regeneration process, more than upon the final outcome.

That adult cells are able to promote and support this regeneration is important in considering the use of implanted Schwann cells in human nerve repair, where autologous cells would be preferred. Patients with severed nerves could donate a small portion of the damaged nerve from which Schwann cells could be extracted. While the patient's muscles and tendons were recovering from injury, the isolated Schwann cells could be cultured and made to proliferate. At a suitable time, the cells could then be seeded into a nerve guide implanted to bridge the nerve gap. Such a process would circumvent the need to sacrifice the sural nerve for nerve repair.

Acknowledgements

Part of this work was supported by the Microsearch Foundation of Australia. We are indebted to BioNova Neo Technics Pty, Ltd., Melbourne, for the generous supply of experimental nerve guides.

References

1. Sumner AJ: Aberrant reinnervation. Muscle Nerve 1990;13:801-803.
2. Sunderland S: Nerves and Nerve Injuries, 2nd edition. London, Churchill Livingstone, 1978.
3. Brushart TME, Seiler WA: Selective reinnervation of distal motor stumps by peripheral motor axons. Exp Neurol 1987;97:289-300.
4. Evans PJ, Bain JR, Mackinnon SE, Makino AP, Hunter DA: Selective reinnervation - A comparison of recovery following microsuture and conduit nerve repair. Brain Res 1991;559:315-321.
5. Pham HN, Padilla JA, Nguyen KD, Rosen JM: Comparison of nerve repair techniques: suture vs. avitene-polyglycolic acid tube. J Reconstr Microsurg 1991;7:31-36.
6. Seckel BR, Ryan SE, Gagne RG, Chiu TH, Watkins Jr. E: Target-specific nerve regeneration through a nerve guide in the rat. Plast Reconstr Surg 1986;78:793-798.
7. Fields RD, Le Beau JM, Longo FM, Ellisman MH: Nerve regeneration through artificial tubular implants. Prog Neurobiol 1989;33:87-134.
8. Longo FM, Manthorpe M, Skaper SD, Lundborg G, Varon S: Neuronotrophic activities accumulate in vivo within silicone nerve regeneration chambers. Brain Res 1983;261:109-116.
9. Longo FM, Skaper SD, Manthorpe M, Williams LR, Lundborg G, Varon S: Temporal changes of neuronotrophic activities accumulating in vivo within nerve regeneration chambers. Exp Neurol 1983;81:756-769.
10. Lundborg G, Longo FM, Varon S: Nerve regeneration model and trophic factors in vivo. Brain Res 1982;232:157-161.
11. Brushart TME: Motor axons preferentially reinnervate motor pathways. J Neurosci 1993;13:2730-2738.
12. Madison R, Da Silva CF, Dikkes P, Chiu TH, Sidman RL: Increased rate of peripheral nerve regeneration using bioresorbable nerve guides and a laminin-containing gel. Exp Neurol 1985;88:767-772.
13. Rich KM, Alexander TD, Pryor JC, Hollowell JP: Nerve growth factor enhances regeneration through silicone chambers. Exp Neurol 1989;105:162-170.
14. Aebischer P, Salessiotis AN, Winn SR: Basic fibroblast growth factor released from synthetic guidance channels facilitates peripheral nerve regeneration across long nerve gaps. J Neurosci Res 1989;23:282-289.
15. Hansson HA, Dahlin LB, Danielsen N, Fryklund L, Nachemson AK, Polleryd P, Rozell B, Skottner A, Stemme S, Lundborg G: Evidence indicating trophic importance of IGF-I in regenerating peripheral nerves. Acta Physiol Scand 1986;126:609-614.
16. Rende M, Muir D, Ruoslahti E, Hagg T, Varon S, Manthorpe M: Immunolocalization of ciliary neuronotrophic factor in adult rat sciatic nerve. Glia 1992;5:25-32.
17. Meyer M, Matsuoka I, Wetmore C, Olson L, Thoenen H: Enhanced synthesis of brain-derived neurotrophic factor in the lesioned peripheral nerve: different mechanisms are responsible for the regulation of BDNF and NGF mRNA. J Cell Biol 1992;119:45-54.
18. Ide C, Tohyama K, Yokota R, Nitatori T, Onodera S: Schwann cell basal lamina and nerve regeneration. Brain Res 1983;288:61-75.
19. Williams LR, Danielsen N, Müller H, Varon S: Exogenous matrix precursors promote functional nerve regeneration across a 15-mm gap within a silicone chamber in the rat. J Comp Neurol 1987;264:284-290.
20. Hall SM: The effect of inhibiting Schwann cell mitosis on the re-innervation of acellular autografts in the peripheral nervous system of the mouse. Neuropathol Appl Neurobiol 1986;12:401-414.
21. Hall SM: Regeneration in the peripheral nervous system. Neuropathol Appl Neurobiol 1989;15:513-529.
22. Archibald SJ, Krarup C, Shefner J, Li ST, Madison RD: A collagen-based nerve guide conduit for peripheral nerve repair - an electrophysiological study of nerve regeneration in rodents and nonhuman primates. J Comp Neurol 1991;306:685-696.
23. Ansselin AD, Corbeil SD, Davey DF: Successfully culturing Schwann cells from adult peripheral nerve. Acta Chururgica Austriaca 1998;30(Supplement 147):15-19.
24. Ramshaw JA, Peters DE, Werkmeister JA, Ketharanathan V: Collagen organization in mandrel-grown vascular grafts. J Biomed Mater Res 1989;23:649-660.
25. Ansselin AD, Davey DF: The regeneration of axons through normal and reversed peripheral nerve grafts. Restor Neurol Neurosci 1993;5:225-240.
26. de Medinaceli L, Freed WJ, Wyatt RJ: An index of the functional condition of rat sciatic nerve based on measurements made from walking tracks. Exp Neurol 1982;77:634-643.
27. Ansselin AD, Corbeil SD, Davey DF: Culture of Schwann cells from adult animals. In Vitro Cell Dev Biol 1995;31:253-254.
28. Hildebrand C, Bowe CM, Remahl IN: Myelination and myelin sheath remodelling in normal and pathological PNS nerve fibres. Prog Neurobiol 1994;43:85.
29. Smith RS, Koles ZJ: Myelinated nerve fibers: computed effect of myelin thickness on conduction velocity. Am J Physiol 1970;219:1256-1258.
30. Guénard V, Kleitman N, Morrissey TK, Bunge RP, Aebischer P: Syngeneic Schwann cells derived from adult nerves seeded in semipermeable guidance channels enhance peripheral nerve regeneration. J Neurosci 1992;12:3310-3320.
31. Keeley R, Atagi T, Sabelman E, Padilla J, Kadlcik P, Agras J, Eng L, Wiedman TW, Nguyen K, Sudekum A, Rosen J: Synthetic nerve graft containing collagen and synthetic Schwann cells improves functional, electrophysiological, and histological parameters of peripheral nerve regeneration. Restor Neurol Neurosci 1993;5:353-366.
32. Kim DH, Connolly SE, Kline DG, Voorhies RM, Smith A, Powell M, Yoes T, Daniloff JK: Labeled Schwann cell transplants versus sural nerve grafts in nerve repair. J Neurosurgery 1994;80:254-260.
33. Ansselin AD, Westland K, Pollard JD: Low dose, short term Cyclosporin A does not protect the Schwann cells of allogeneic nerve grafts. Neurosci Lett 1990;119:219-222.
34. Ansselin AD, Pollard JD, Davey DF: Immunosuppression in nerve grafting: Is it desirable?. J Neurol Sci 1992;112:160-169.
35. Schröder JM: Altered ratio between axon diameter and myelin sheath thickness in regenerated nerve fibers. Brain Res 1972;45:49-65.
36. Scherer SS, Easter Jr. SS: Degenerative and regenerative changes in the trochlear nerve of goldfish. J Neurocytol 1984;13:519-565.
37. Fields RD, Ellisman MH: Axons regenerated through silicone tube splices. II. Functional morphology. Exp Neurol 1986;92:61-74.
38. Sanders FK, Young JZ: The influence of peripheral connexion on the diameter of regenerating nerve fibres. J Exp Biol 1946;22:203-212.
39. Gutmann E, Sanders FK: Recovery of fibre numbers and diameters in the regeneration of peripheral nerves. Physiologist 1943;101:489-518.
40. Bowe CM, Hildebrand C, Kocsis JD, Waxman SG: Morphological and physiological properties of neurons after long-term axonal regeneration - observations on chronic and delayed sequelae of peripheral nerve injury. J Neurol Sci 1989;91:259-292.
41. Shenaq JM, Shenaq SM, Spira M: Reliability of sciatic function index in assessing nerve regeneration across a 1 cm gap. Microsurgery 1989;10:214-219.
42. Fried K, Erdelyi G: Inferior alveolar nerve regeneration and incisor pulpal reinnervation following intramandibular neurotomy in the cat. Brain Res 1982;244:259-268.
43. Hildebrand C, Kocsis JD, Berglund JD, Waxman SG: Remodelling of internodes in regenerated rat sciatic nerve. Brain Res 1985;358:163-170.
44. Beuche W, Friede RL: A new approach toward analyzing peripheral nerve fiber populations. II Foreshortening of regenerated internodes corresponds to reduced sheath thickness. J Neuropathol Exp Neurol 1985;44:73-84.
45. Fields RD, Ellisman MH: Axons regenerated through silicone tube splices. I. Conduction properties. Exp Neurol 1986;92:48-60.

Address for correspondence:
A/Prof. D.F. Davey
Department of Physiology (F-13)
University of Sydney
NSW 2006 Australia

Phone: +61 2 9351 4559
Fax: +61 2 9351 5182
Email: daved@physiol.usyd.edu.au