Changes in the inferior alveolar nerve following mandibular lengthening in the dog using distraction osteogenesis

J Oral Ma xillo fac Surg 51:652-660. 1993

Changes in the Inferior Alveolar Nerve Following Mandibular Lengthening in the Dog Using Distraction Osteogenesis MICHAEL S. BLOCK, DMD,* JOHN DAIRE, DDS,t JOHN STOVER, DDS,:f: AND ';v1URRAY MATTHEWS, PHD§ Distraction osteogenesis as per lIizarov was used to lengthen the canine mandible. In this study, physiological and ultrastructural examination of the inferior nerve was performed. Mandibular body corticotomies were performed, and the mandible was distracted 7 mm. The animals were killed 4 weeks after the distraction was completed. Bone formed within the distraction gap in all dogs. There was no statistically significant difference in the jaw-jerk voltage between control and experimental sides. There was a significant difference between the distracted and control nerves in only one area of one nerve.

Distraction osteogenesis (DO) is a technique of bone lengthening and remodeling. During the past 35 years, Gavriel A. I1izarov, a Russian physician, has been involved with the comprehensive use of this technique to regenerate bone and soft tissues.P DO involves the application of a transcutaneous endosseous appliance to short, misshapen, or discontinuous bones combined with cortical osteotomies, followed by the use of distraction forces to lengthen or reposition existing bone and soft tissues. According to I1izarov, this technique allows for a reduction in disability, number of procedures, and length of treatment.' Initially, the DO technique was used mainly for the bones of the upper and lower extremities , but recently there have been several attempts to use variants of this technique in the mandible.l -'? Previous work used the application of extraoral endosseous transcutaneous , pins. ·3,4 Disadvantages of extraoral pin placement include extraoral scars and facial nerve and inferior alveolar nerve injury. Michieli and Miottf have ad-

dressed these concerns by the use of a specially fabricated intraoral tooth-borne appliance to provide the necessary distraction forces. This technique, however, may result in unwanted tooth movements and the unavoidable creation of spaces between teeth iri the location of the distraction. We report the use of distraction osteogenesis to lengthen the dog mandible. Because of the known sensory nerve complications using sagittal split osteotomies, particular attention has been given to analysis of the inferior nerve after lengthening. Materials and Methods

Four mongrel dogs were used for this study. The contralateral, right, nonoperated side of the mandible was used as a control to evaluate the nerve without distr action. Under general anesthesia, the left mandibular premolars and first molar were extracted. Small dimples were placed in the mandibular left canine and second molar for clinical measurements. After 12 weeks, general anesthesia was induced for placement of the distraction device, creation of a corticotomy, and initiation of distraction osteogenesis of the mandibular body. Before and I week after placement of the devices, and then biweekly, the dogs were anesthetized with intravenous pentobarbital for radiographs, clinical measurements, and nerve testing, Occlusal and lateral mandibular radiographs were made to evaluate bone healing qualitatively. Clinical measurements were

Received from the Louisiana State University; School of Dentistry, New Orleans , LA. * Associate Professor, Department of Oral and Maxillofacial Surgery, t Former Resident, Department of Orthodontics. t PhD Cand idate and Resident , Department of Oral and Maxillofacial Surgery, § Professor, Department of Anatomy. Address correspondence and reprint requests to Dr Block: LSU School of Dentistry, lIDO Florida Ave, New Orleans , LA 70119.

© 1993 American Association of Oral and Maxillofacial Surgeons 0278-2391/93/5106-0009$3.00/0

652

653

BLOCK ET AL

reflex testing. Radiographs were taken immediately before the initial distraction, at day 4, day 10,and again at 5 weeks (Fig 3). Because unilateral distraction occurred, a crossbite slowly developed during this time. After 7 days of distraction, the device was stabilized with acrylic resin. NEUROSENSORY TESTING

FIGURE I. Intraoral view of the corticotomy created through the lingual and buccal cortical bone (arrOlI'). An extraoral distraction device has been placed in this dog.

made from the pins as well as from the dimples made in the canine and first molar teeth. Nerve testing was performed on the operated and nonoperated (control) sides using the jaw-jerk reflex.6- 10 During extraoral placement of the distraction device on the first dog, the inferior alveolar nerve was lacerated. The remaining three dogs had intraoral placement of the pins into the alveolus superior to the neurovascular canal followed by intraoral placement ofa shortened mini-lengthener (Howmedica, New Brunswick, NJ). To accomplish this, a crestal incision was made over the ridge and the periosteum was reflected to expose the buccal and lingual cortical bone. Four 2-mm diameter threaded pins were placed, with the two medial pins 15 mm apart. A mini-lengthener was adjusted to fit passively over the pins and then was removed. Cortical osteotomies were created with a tapered fissure bur under copious irrigation. A narrow osteotome was then used to complete the osteotomy through the inferior border of the mandible gently avoiding and encroachment on the cancellous bone and marrow cavity and damage to the inferior alveolar nerve. The DO device was then secured to the pins to stabilize and reapproximate the iatrogenically created fracture (Figs 1,2). A postoperative radiograph was taken to evaluate the position of the pins and the location of the osteotomy (Fig 3A). . Following the Ilizarov protocol, the iatrogenic fracture was stabilized without distraction with the minilengthener device for 7 days to allow for healing of the periosteum. After 7 days, the distraction was begun. The device was lengthened 0.5 mm in the morning and 0.5 mm in the evening, for a total of 1.0 mm distraction each day for 7 days.'? The activations were performed with gentle restraint of the dog without causing discomfort. For the first distraction, pentobarbital anesthesia was used to perform baseline jaw-jerk

The integrity ofthe mandibular nerve was evaluated by eliciting ajaw-jerk reflex."!" This was accomplished by delivering an increasing electric current to the mental nerves via transgingival electrodes until the dog's digastric muscles contracted as a reflex to the noxious stimulus. A Grass SDP stimulating machine was used to deliver the electric pulse (6-millisecond duration, 4~ millisecond delay, 0.8 pulses/s), The voltage required to elicit the reflexwas recorded. The jaw-jerk reflex test was performed bilaterally so that the nonoperated side served as a control. KILLING PROTOCOL

After 4 weeks of stabilization of the 7-mm distraction, the dogs were placed under general anesthesia with pentobarbital. Photographs, radiographs, and jawjerk reflex testing were performed. Heparin 500,000 U, was administered through a carotid artery canula and the neck veins were sectioned and the blood cleared with lactated Ringer's solution. The common carotid arteries were perfused with a freshly depolymerized solution of 4% paraformaldehyde and 0.5% glutaraldehyde in phosphate buffer at pH 7.35. Perfusion continued until the head and face were firm and cool to touch. The region of the mandible involved with the DO (10 mm on either side of the original osteotomy)

FlGURE 2. View of the intraoral distraction device. Note the anteriorly positioned hexagonal head that wil1 be rotated to lengthen the device.

654

DISTRACTION OSTEOGENESIS TO LENGTHEN MANDIBLE

:

FIGURE 3. Radiographs demonstrating the distraction and healing of the mandible. A, Time of distraction and device placement before distraction. B, Seven days after distraction showing a 7-mm gap. C. Four weeks later; note the bone consolidation within the gap.

was then removed en bloc. The DO suprastructure was removed and the specimen was decalcified and sagittal sections cut so that three.sections I mm apart could be stained with hematoxylin-eosin (general histology), trichrome (soft tissue), and Heidenhain's aniline blue (hard tissue). Sections from the unoperated side of the mandible were processed in a similar manner. NERVE PROCESSING

After postfixing the head overnight in paraformaldehyde/glutaraldehyde, segments of the mental nerves exiting from the mental foramina were carefully dissected and removed. Each segment of the nerve was pinned on pieces oflabeled cardboard under slight tension and placed in the fixative solution overnight. The former was done to prevent contraction of the segments, which would give false information with regard to fiber size and count in subsequent evaluations. The nerve segments attached to the cardboard were washed overnight in a 0.15 mol/L phosphate buffer solution and then placed in phosphate-buffered saline. The segments were dehydrated in graded concentrations of ethanol (70% to 100%),transferred to propylene oxide, and embedded in Epon/Araldite, Semithin sections (l

Jim) cut from pretrimmed blocks of tissue were stained with toluidine blue and mounted on slides for analysis at the light microscopic level. These sections were used to orient the block for thin sectioning. Thin sections were cut, stained with lead citrate and uranyl acetate, and examined in a Philips CM-1Oelectron microscope. Semithin sections were analyzed using the Cue-2 image analysis software (Olympus Corp, Lake Success, NY). Only analysis of myelinated fibers was possible because unmyelinated fibers are not visible at the light microscopic level. Twenty-five fields of 40 X 40 JIm 2 from fivesections ofthe experimental side mental nerve were morphometrically compared to 25 like fields from sections of control mental nerve of similar diameter. A qualitative description of the distribution of axon sizes in each section was recorded, including any.variance between central and peripheral areas of the nerve bundle. Any areas oflocalized ischemia or degeneration also were noted. Quantitative analysis included counting the total number ofaxons within the standard area, determining the average diameter of the axons, and noting the percentage of large (> 6 Jim) and small « 6 Jim) axons that exhibited degeneration. Data from each side were averaged and compared. Percentage differences were calculated for all of the data and evaluated by analysis of variance.

655

BLOCK ET AL

The semithin sections were used to determine the necessity of transmission electron microscopic (TEM) examination. Those sections that showed signs of degeneration or differences in axonal patterns (when compared with similar locations in the contralateral nerve) were processed for thin-section examination. This helped limit the total number of sections, which was restricted by the project's budget. Electron microscopic examination included a descriptive comparison of the experimental and control mental nerves. Documentation also was made of the regional distribution and relative density of unmyelinated fibers and compared with the contralateral control side. CLINICAL MEASUREMENTS

The distance between the dimples on the tooth cusps on either side of the distraction segment, as well as the distance between the medial pins, were measured before distraction, immediately after distraction, and at the time the dogs were killed. These distances were compared with each other and with the linear distraction distance measured on the DO device.

Table 2.

Jaw-Jerk Reflex Testing Jaw-Jerk Reflex Voltage (mean ± SO)

Time Interval

Distraction Side

Unoperated Side

Preoperative Immediately before first distraction Immediately after first distraction Day 4 of distraction Day 7 of distraction 2 Weeks of completion of distraction 4 Weeks after completion of distraction

10.0 ± 4.6

6.5 ± 0.7

8.0 ± 3.5

7.5 ± 4.9

10.0 ± 5.3 9.2 ± 2.7 8.0 ± 4.6

4.5 ± 2.1 4.7 ± 0.4

4.3 ± 0.6

8.3 ± 2.9

4.3 ± 1.0

7.6 ± 3.2

Gross examination indicated a bony union across the distraction gap; no bone defect could be palpated. The soft tissues were continuous and normal in appearance. A small callus was palpable along the inferior border of the mandible in two dogs. RADIOGRAPHIC EVALUATION

Results CLINICAL EVALUATION

The mandible was lengthened in all animals an average of 5.5 ± 0.8 mm, as determined by the distance between the most proximal pins, and 6.5 ± 2.2 mm when measured from the dimples made on the teeth (Table I). The canine on the operated side was advanced to a class III relationship and the mandible was deviated toward the unoperated side in all animals. Measurements taken at the 4-week postfixation period were 1.7 ± 0.8 mm less than the distance measured immediately after completion of the distraction. On direct examination, it was noted that the pins had bent slightly during the healing period.

Table 1. Clinical Measurements of Distraction Distance

Time period Before distraction Day 4 of distraction Day 7 of distraction 2 Weeks after completion of distraction 4 Weeks after completion of distraction

Distance Between Innennost Pins (mm)

Distance Between Canine and 1st Molar Cusp Dimples (mm)

16.2 ± 2.0 18.8 ± 3.4 22.7 ± 1.7

59.3 ± 4.7 63.3 ± 6.4 65.8 ± 5.9

21.8 ± 0.4

64.7 ± 5.3

21.0 ± 0.7

64.2 ± 5.3

The radiographs confirmed the distraction of the mandibular segments (Fig 3). Mineralized tissue was seen bridging the defect as early as 2 weeks after the mandible was fixed in its distracted position. After 4 weeks, more radiopaque tissue was observed within the DO site in all four animals. NEUROSENSORY EVALUATION

The results of the jaw-jerk neurosensory evaluation can be seen in Table 2. In the first animal with extraoral pin placement, there was complete obliteration ofsensory function. Closer examination showed that the pins had severed the neurovascular bundle. The subsequent three animals had their pins placed intraorally, with direct visualization of the mental foramen as a guide to pin placement, trauma to the nerve was avoided. These three animals are described below. The voltage resulting in ajaw-jerk reflexranged from 3.5 to 16 V. Both averaged values and percentage value changes from baseline indicated no statistically significant difference between experimental and control sides. The testing showed large standard deviations in voltage recordings both for the experimental and control sides, possibly representing differences in electrode placement throughout the experiment. HISTOLOGIC EVALUATION

Bone was seen traversing the distraction gap. The trabeculae were oriented in the sagittal plane along the

656

DISTRACTION OSTEOGENESIS TO LENGTHEN MANDIBLE

FIGURE 4. Photomicrographs of the distraction sites. A. Heidenhain-stained section (I: I) of the mandible with the distraction gap clearly visible. The inferior border of the mandible has healed with denser bone than the crest. Note the axial orientation of the trabeculae bone within the DO gap. B. A higher power of the edge of the DO gap. Note the increase in mineralization toward the center ofthe distraction site (Heidenhain stain, original magnification X 10). C. Note bone formation with multiple Haversian systems both axial and transverse to the distraction within the distraction gap (Heidenhain stain, original magnification X10). D. Trichrome stain showing mineralized tissue apposed to the underlying connective tissue (Heidenhain stain, original magnification X40).

line ofdistraction. As seen clearly in the sections stained with hematoxylin-eosin, the most mature lamellar bone was along the inferior border; the greenstick fracture along the inferior border had healed with denser bone than that seen at the alveolar crest or within the marrow space. Bone density seemed to be greatest along the edges ofthe original bone, with lessdensity at the center of the DO gap. The bone that filled the distraction gap at the alveolar crest was woven in appearance. The inferior alveolar nerve spanned the distraction gap without a continuity defect (Fig 4). SEMI-THIN SECTION ANALYSIS

Semi-thin sections revealed gross changes in the distal portion of the mental nerve in the one animal with direct trauma to the nerve. Analysis of the mental nerves of the three remaining dogs with intact jaw-jerk reflexes showed that 1.2% (4 of 337) of the large fibers

and none (0 of 94) of the small fibers on the control side exhibited evidence ofwallerian degeneration. On the experimental side, 9.1% (29 of 318) of the large and 15.9% (10 of 63) of the small fibers showed signs of degeneration. Such signs included demyelination, axonal swelling, and axoplasmic darkening. In general, degenerating axons were uniformly distributed throughout the cross-sectional area of nerve sections in this chronic tension neuropathy model (Fig SA). They were not preferentially located toward the nerve periphery, as is characteristically seen in classic models ofacute or chronic compression neuropathy. However, a fiber degeneration pattern consistent with compression-related ischemia was observed on the experimental side mental nerve of one of the three animals with an intact jaw-jerk reflex (Fig 5B-D). Within the ischemic region, there was a peripheral zone of severe demyelination and axoplasmic darkening (Fig 50) and a more central transition zone of moderate degeneration (Fig sci

BLOCK ET

657

AL

*

,

o FIGURE 5. A. Semithin section ofthe experimental side mental nerve from the one animal that showed evidence ofa compression neuropathy. There is a severe loss of large myelinated fibers at one edge of the fascicle and intrafascicular fibrosis consistent with wallerian degeneration (*) (toluidine blue stain, original magnification X200). B. Higher magnification of the same section seen in Fig SA showing typical appearance of the experimental side mental nerve . There are occasional fibers exhibiting varying degrees of myel in disruption (toluidine blue stain, original magnification X2,500). C. View of the transition zone adjacent to the area of severe degeneration seen in Fig 6A (toluidine blue stain, original magnification X2,500). D. High-magnification photomicrograph taken from the same section seen in Fig 5A showing remnants of severe wallerian degeneration. Medium diameter, thinly myelinated fibers persist in the collagen matrix (toluidine blue stain, original magnification X2,SOO). TRANSMISSION ELECfRON MICROSCOPIC ANALYSIS TEM analysis of distraction and control side sections from the three dogs with intact jaw-jerk reflexes demonstrated no overall difference in small-diameter unmyelinated (C fiber) or thinly myelinated (A-delta) fiber frequency or distribution in two animals (Fig 6A,B). The dog that had the compression-type neuropathy, however, despite a large-scale loss of medium and largediameter myelinated fibers (A-alpha and A-beta) and severe intrafascicular fibrosis, had a minimal loss or alteration in distribution of C fibers (Fig 6C-F). The C fibers, which mediate slow nociceptive impulses, are more resistant to compression trauma than are larger, myelinated fibers. II Discussion The sagittal split ramus osteotomy (SSO) used to correct mandibular retrognathism has been reported

to involve a postsurgical neurosensory complication rate ranging from 0% to 54%:Z Traction injury of the nerve is apparently more frequent with mandibular advancements of greater lengths. This injury may be due to mechanical tearing ofaxons within the nerve bundle or to ischemia caused by compression of the vasa nervorum.'? Neuronal ischemia, studied primarily using chronic compression models, leads to a neuropraxialike injury exhibiting wallerian axonal degeneration and demyelination. 14 Sunderland proposed that with compression, obstruction of the venous return leads to local nerve edema and chronically elevated intraneural pressure, which secondarily cause an obstruction of the blood supply. The resulting anoxia causes further pathology. IS In nerve compression models, most demyelination occurs at the interface between the compressed and noncompressed segments of the nerve. Therefore, demyelination may be minimal in segments of gently stretched nerves subjected to equal pressures through-

658

DISTRACTION OSTEOGENESIS TO LENGTHEN MANDIBLE

c

out. However, studies addressing nerve stretch injuries primarily examine the effects ofacute, forceful trauma resulting in avulsion. Information gained from compression models cannot be extrapolated to explain anatomic and physiologic changes observed in response to gentle traction injuries. Slow nerve stretching, as in Ilizarov distraction osteogenesis, may result in minimal

injury because less axonal tearing and less complete compression of the vasa nervorum occurs. Our findings suggest that nerve injury due to slow traction associated with distraction osteogenesis is mild. Morphologic changes in the mental nerve I month after slow stretch of the inferior alveolar nerve, induding myelin attenuation, axoplasmic swelling, and dark-

659

BLOCK ET AL

-.'

c

FIGURE 6. A. Photomicrograph from an experimental side mental nerve. Note the essentially normal distribution of large-diarneter (A-a), small-diameter (A-a), and small unmyelinated (C) fibers (original magnification X5,950). B, Higher-power photomicrograph of a normalappearing experimental side mental nerve showing healthy unmyelinated C fibers within a normal-appearing collagen matrix. A portion of a large myelinated fiber (A-a) is shown (original magnification X20,400). C. Photomicrograph taken from the experimental side mental nerve of the animal with an intact jaw-jerk reflex, showing a histologic appearance consistent with a compression neuropathy (Fig 5). Seen here are multiple large- and small-diameter myelinated fibers displaying myelin disruption (solid arrowlieadsi and healthy-appearing, small, unmyelinated fibers (curvedsolid arrow) (original magnification X5,950). D, Photomicrograph of an area from the same nerve as in Figs 5 and 6C showing moderate degeneration. Demyelination is evident (solid arrowhead), as are intact C fibers tcurved solid arrow). Curvedopenarrow indicates a fiber with disrupted axoplasm (original magnification X15,300). E, Photomicrograph from the same mental nerve shown in Figs 5 and 6C-D showing severe degeneration. Note the absence of large myelinated fibers, the persistence of small-diameter, lightly myelinated fibers and larger than normal collections of collagen (open arrowhead). Such collagen accumulations, called Renaut bodies, are consistent with fibroblastic proliferation and intrafascicular fibrosisseen characteristically after chronic nerve compression (original magnification X5,950). F. Photomicrograph (same as Figs 5 and 6C-E) showing a loss oflarge myelinated fibers, a diminished presence of small-diameter myelinated (A-a) fibers, a persistence of healthy-appearing small unmyelinated (c) fibers and excessive amounts of collagen matrix (open arrowhead). Abundant throughout the region are large macrophages (open curved arrow) containing numerous lipid lobules taken up after demyelination (original magnification X5,950).

ening, were observed with a uniform and relatively infrequent distribution throughout the fascicle. These changes were most likely due to a combination of mild mechanical deformation and impairment of venous blood flow. The earliest changes to occur after slow nerve stretch are probably due to mechanical deformation because significant alterations in conduction properties result after very small amounts of strain, in the range of 6% beyond the in situ strain," yet venous

blood flow within the vasa nervorum of the rabbit tibial nerve was not impaired until after a gradual stretch to 8% of its in vivo length. 17 The mild, uniformly distributed pattern of traction neuropathy associated with slow mandibular distraction was distinct from the compression-type neuropathy seen in one dog. The severe, focal degeneration seen in this instance may have been due to a slight malalignment of the bony segments undergoing distraction and the resultant compression

660 of the inferior alveolar nerve against the wall of the canal. Previous studies with distraction osteogenesis in the mandible have analyzed the morphology of the inferior alveolar nerve, but they report inconsistent results. Costantino et al' intentionally transected the entire mandible and its neurovascular contents in their investigation, and they report that the inferior alveolar artery recanalized in five of the six dogs tested with a 25-mm distraction distance, but none ofthese animals demonstrated regeneration of the inferior alveolar nerve in 8 weeks. Conversely, Karp et al 4 attempted preservation of the inferior alveolar neurovascular bundle in their experiment, but they also found an absence of myelinated fibers on the operated side of all six of the dogs tested with a 20-mm distraction distance after 8 weeks. Michieli and Miotti," on the other hand, report that there was no conspicuous histologic alteration in nerve fibers in either of the two dogs tested (5- and i5-mm distraction). Some of these conflicting results might be explained by the shorter advancement distances used in the last investigation and the fact that these authors also avoided the use of transcutaneous endosseous pins by using an intraoral tooth-borne distraction appliance. This would minimize the chance of inferior alveolar nerve injury. The results ofthis study show that DO can be applied to the mandible of the dog with minimal nerve effects. However, larger samples should be studied to confirm these results before applying these findings to the clinical situation.

DISTRACfION OSTEOGENESIS TO LENGTHEN MANDIBLE

2. 3. 4. 5. 6.

7. 8. 9. 10.

II.

12. 13.

14. 15. 16.

References 17. I. Costantino PD, Shybut G, Friedman CD, et al: Segmental mandibular regeneration by distraction osteogenesis. An experi-

mental study. Arch Otolaryngol Head Neck Surg 116:535, 1990 lIizarov GA: The principals of the Ilizarov method. Bull Hasp J Dis Orthop Inst 48:1, 1988 Snyder CC, Levine GA, Swanson HM, et al: Mandibular lengthening by gradual distraction (preliminary report). Plast Reconstr Surg 5:506, 1973 Karp NS, Thome CH, McCarthy JG, et al: Bone lengthening in the craniofacial skeleton. Ann Plast Surg 24:231, 1990 Michieli S, Miotti B:Controlled gradual lengthening of the mandible after osteotomy. Minerva StomatoI25:77, 1976 Widner S, Block MS, Liles S: A comparison of nerve grafting within and lateral to bone grafts in dogs. American Academy of Dental Radiology Annual Meeting, Montreal, March 913, J Dent Res 67:252, 1988 (abstr 1112) Matthews B, Baxter J, Watts S: Sensory and reflex responses to tooth pulp stimulation in man. Brain Res 113:83, 1976 Yemm R: Reflex jaw opening following electrical stimulation of oral mucous membrane in man. Arch Oral Bioi 17:513, 1972 Yamada Y, Ash MM: An electromyographic study of jaw opening and closing reflexes in man. Arch Oral Bioi 27:13, 1982 Hugon M: Methodology of the Hoffman reflex in man, ill Desmedt JE (cd): New Developments in Electromyography and Clinical Neurophysiology, vol 3. Basel, Switzerland, Karger, 1973, pp 277-293 Dahlin LB, Shyu BC, Danielsen N, et al: Effects of nerve compression or ischemia on conduction properties of myelinated and non-myelinated nerve fibers. An experimental study in the rabbit peroneal nerve. Acta Physiol Scand 136: 97,1989 Karus P, Nester D, Boyd C, et al: Recovery of neurosensory function following orthognathic surgery. J Oral Maxillofac Surg 48:124, 1990 Ochoa J: Histopathology of common mononeuropathies, ill Jewett DL, McCarroll HR Jr (eds): Nerve Repair and Regeneration: Its Clinical and Experimental Basis. St Louis, MO, Mosby, 1980, pp 36-52 Eames RA, Lange LS: Clinical and pathological study of ischemic neuropathy. J Neurol Neurosurg Psychiatry 30:215, 1967 Sunderland S: Nerve lesion in the carpal tunnel syndrome. J Neurol Neurosurg Psychiatry 39:615, 1976 Kwan MK, Savio L-YW: Biomechanicai properties of peripheral nerve, ill Gelberman RH (ed): Operative Nerve Repair and Reconstruction. Philadelphia, PA, Lippincott, 1991, pp 4754 Lundborg G, Rydevik B: Effects of stretching the tibial nerve of the rabbit: A preliminary study of the intraneural circulation and the barrier function of the perineurium. J Bone Joint Surg 55:390, 1973