archives of oral biology 59 (2014) 324–335

Available online at www.sciencedirect.com

ScienceDirect journal homepage: http://www.elsevier.com/locate/aob

Applied anatomy of the lingual nerve: Relevance to dental anaesthesia Vui Leng Tan a, Alice Andrawos b, Mounir N. Ghabriel b, Grant C. Townsend a,* a b

School of Dentistry, The University of Adelaide, Adelaide, SA, Australia Discipline of Anatomy and Pathology, The University of Adelaide, Adelaide, SA, Australia

article info

abstract

Article history:

Objectives: (1) to classify the external morphology of the lingual nerve and investigate any

Received 10 November 2013

relationship between its external and internal morphology, (2) to explore the fascicular

Accepted 5 December 2013

structure, nerve tissue density and capillary density of the lingual nerve, and (3) to provide

Available online

an anatomical explanation as to why adverse clinical outcomes more commonly affect the

Keywords:

between the lingual and inferior alveolar nerves.

Lingual nerve

Materials and methods: The lingual and inferior alveolar nerves were examined in 23 hemi-

Inferior alveolar nerve

sectioned heads macroscopically and microscopically 2 mm above the lingula. The lingual

Nerve trauma

nerve was also examined in the regions of the third and second molars. Specimens under-

lingual nerve following local dental anaesthesia. Where possible, comparisons were made

Dental anaesthesia

went histological processing and staining with Haematoxylin & Eosin, Masson’s Trichrome, anti-GLUT-1 and anti-CD 34. Results: The lingual nerve became flatter as it traversed through the pterygomandibular space. There was an increase in the connective tissue and a decrease in nerve tissue density along the lingual nerve ( p < 0.001). At 2 mm above the lingula, the lingual nerve was unifascicular in 39% of cases, whilst the inferior alveolar nerve consistently had more fascicles ( p < 0.001). The lingual nerve fascicles had thicker perineurium but the endoneurial vascular density was not significantly different in the two nerves. Conclusions: The greater susceptibility of lingual nerve dysfunction during inferior alveolar nerve blocks may be due to its uni-fascicular structure and the thicker perineurium, leading to increased endoneurial pressure and involvement of all axons if oedema or haemorrhage occurs due to trauma. # 2013 Elsevier Ltd. All rights reserved.

1.

Introduction

Originating from the mandibular division of the trigeminal nerve (cranial nerve V) within the infratemporal fossa, the lingual nerve travels in an anterior direction towards the medial surface of the mandible. As it reaches the lingual plate in the vicinity of the mandibular third molar, it is covered only

by the gingival mucoperiosteum.1 The lingual nerve continues to travel anteriorly, crossing the lateral and inferior surfaces of the submandibular duct, before passing upwards to reach the ventral surface of the tongue where it divides into its terminal branches.2–5 The main functions of the lingual nerve are to provide general sensation to the mucosa and papillae of the anterior two thirds of the tongue, mandibular gingivae

* Corresponding author. Fax: +61 (0)8 8313 2016. E-mail address: [email protected] (G.C. Townsend). 0003–9969/$ – see front matter # 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.archoralbio.2013.12.002

archives of oral biology 59 (2014) 324–335

(lingual) and mucosa of the floor of the mouth. As it incorporates the chorda tympani nerve, it also carries taste sensation from the taste buds of the anterior two thirds of the tongue, and delivers parasympathetic secretomotor innervations to the submandibular and sublingual salivary glands.3–6 Whilst the lingual nerve tends to follow a defined path, its disposition towards anatomical variations adds a dimension of complexity which increases the risk of damage. The lingual nerve may be susceptible to physical and chemical damage during dental procedures, which may manifest as a temporary or permanent sensory disturbance. In order of frequency, the most commonly-reported dental causes of lingual and inferior alveolar nerve damage are mandibular third molar surgery, followed by inferior alveolar nerve blocks, and then endodontic and periodontal complications.7–12 For reasons unknown, dysaesthesia occurs at a higher frequency after injuries associated with inferior alveolar nerve blocks (34%) than after mandibular third molar surgery (8%).13 It has also been consistently reported that the lingual nerve suffers a greater frequency of permanent neural damage than the inferior alveolar nerve, particularly during inferior alveolar nerve blocks.14 The incidence of nerve damage as a result of inferior alveolar nerve blocks has been estimated to be in the range of 1:26,000 to 1:800,000, with the lingual nerve reported to sustain damage more frequently (70%), compared with the inferior alveolar nerve (30%).13,15,16 In a study by Garisto et al.17, the lingual nerve was affected in as many as 89% of cases involving nerve paraesthesia after inferior alveolar nerve blocks. Several hypotheses have been put forward to explain this phenomenon, including the tendency for the lingual nerve to be uni-fascicular at the level where inferior alveolar nerve blocks are administered.14 Other factors to consider may be the different locations of the nerves, with the lingual nerve being only 5–6 mm below the surface mucosa, or because of the lack of bony protection to the lingual nerve by the features on the medial aspect of the mandibular ramus, such as the lingula and the crista endocoronoidea.18,19 In addition, it has also been suggested that the lingual nerve has a tendency to be stretched taut during the administration of inferior alveolar nerve blocks, as the lingual nerve may be bound down by the interpterygoid fascia during the opening of the mandible, and consequently unable to move when it comes into contact with an injecting needle.20 This will consequentially result in needle penetration of the nerve, resulting in damage to nerve fibres. The consequences of neural damage following a traumatic incident may manifest as numerous subjective symptoms, ranging from sensory deficit such as anaesthesia (complete loss of feeling) or hypoaesthesia (diminished sensitivity to all forms of stimulation), to abnormal neuro-sensory disturbances such as paraesthesia (numb feeling, burning and prickling sensation), dysaesthesia (painful sensation), hyperaesthesia (increased sensitivity), and allodynia, where there is pain from stimulus that is not normally painful when applied elsewhere to the body. When the lingual nerve is involved, the chorda tympani branch of the facial nerve may also be affected, leading to dysgeusia (impaired sense of taste) and xerostomia (reduced salivation).9,12,21 Although extremely rare, altered sensation in the maxilla may also occur as a

325

consequence.22 Other problems associated with diminished function of the lingual and inferior alveolar nerves include difficulty with speech, gustatory impediment, hindrance to social interaction, problems with tooth brushing, mechanical (biting) and thermal trauma when there is paraesthesia of the tongue and lip, and sleep disturbances where allodynia is present.21,23 Injury to the lingual nerve has also been associated with changes in the epithelium of the tongue, with a decrease in the number of fungiform papillae.1,23 If there is localised fascicular damage, it is expected that only the skin, mucosal surfaces or sensory parameters supplied by that fascicle(s) are disrupted.7 Fortunately, most neural injuries caused by clinical procedures, such as mandibular third molar surgery and local anaesthetic injections, resolve without treatment within a few months.12,23,24 Although neural damage is rare and the prognosis generally favourable, it should still be of concern as its consequences can be distressing to the affected patient. It is therefore paramount that situations in which dental procedures may cause unintentional impairment to the lingual and inferior alveolar nerves are well understood, so that appropriate prevention and management can be implemented efficiently. This study aims to improve our understanding of the clinical anatomy of the lingual and inferior alveolar nerves to diminish the chances of iatrogenic damage. More specifically, our aims are: Aim 1: To investigate whether there is any relationship between the external and internal morphologies of the lingual nerve. We hypothesise that the lingual nerve becomes flatter as it descends through the pterygomandibular space towards the tongue. Aim 2: To explore the fascicular structure and nerve tissue density of the lingual nerve as it descends through the pterygomandibular space and oral cavity and, where possible, to make comparisons with the inferior alveolar nerve. We hypothesise that there is an increase in fascicle number and nerve tissue density along the length of the lingual nerve. Furthermore, we hypothesise that the lingual nerve has fewer fascicles when compared with the inferior alveolar nerve at the level of the lingula, and a lower nerve tissue density. Aim 3: To determine the capillary density in the lingual nerve compared with the inferior alveolar nerve. We hypothesise that the lingual nerve has a greater capillary density than the inferior alveolar nerve at the level of the lingula. Aim 4: To provide an anatomical explanation about why adverse clinical outcomes affect the lingual nerve more frequently following local anaesthetic use in dentistry.

2.

Materials and methods

This study used 23 hemi-sectioned heads from different cadavers, which had been preserved for 1–4 years, and ranged from 65 to 103 years of age at the time of donation. All donors had signed approval for the use of their mortal remains for research as part of the donor program, and approved by the Head of School. Of the 23 specimens, 14 were edentulous. Dissections were performed from the medial aspect to carefully expose the lingual and inferior alveolar nerves.

326

archives of oral biology 59 (2014) 324–335

Fig. 1 – Diagrammatic illustration of the regions of interest for the lingual nerve (LN) and the inferior alveolar nerve (IAN). A piece of the medial pterygoid muscle (M) was placed adjacent to the IAN. Fine silk sutures were placed through the epineurium and tied loosely around the nerves 2 mm proximal to the regions of interest. L, lingula; (A) 2 mm above the lingula; (B) M3 region, below the point () where the oblique line on the medial aspect of the mandible changes direction from vertical to horizontal; (C) M2 region is 1 cm anterior to the M3 region.

The lingual nerve was exposed from a point 5 mm above the lingula proximally to the lower second molar region distally. The inferior alveolar nerve was exposed from a point 5 mm above the lingula to where the nerve entered the inferior alveolar canal below the lingula. Photographs of the specimens were obtained at various stages of the dissection. The regions of interest were identified as follows: 2 mm above the tip of the lingula for both the lingual and inferior alveolar nerves; at the mandibular third molar (M3) region for the lingual nerve only; and at the second molar (M2) region for the lingual nerve only (Fig. 1). The M3 region was defined as the region immediately inferior to the point where the oblique line on the medial surface of the mandible changes its orientation from a vertical to a horizontal direction (Fig. 1). The M2 region was defined as 1 cm anterior to the M3 region. The nerves were subdivided into these regions of interest. To facilitate orientation during processing, fine surgical sutures were carefully placed in the epineurium 2 mm proximal to these regions of interest (Fig. 1). In the segments of lingual and inferior alveolar nerves obtained above the lingula, a longitudinal piece of muscle from the adjacent medial pterygoid was sutured close to the inferior alveolar nerve (Figs. 1 and 2) to differentiate between the two nerves during histological examination. The divided segments were immediately fixed in 10% formalin and processed in paraffin wax blocks. Transverse sections, 10 mm, were collected on silane-coated slides and adjacent sections were stained with Haematoxylin & Eosin, and Masson’s Trichrome. Other sections from of the lingual and inferior alveolar nerves, 2 mm above the lingula, underwent additional immunohistochemical staining for the glucose transporter protein GLUT-1 (Goat polyclonal primary antibody, Santa Cruz Biotechnology Company, CA, USA, 1:2000 dilution) and anti-CD34 (Mouse monoclonal primary antibody, Novocastra Leica Biosystems, NSW, Australia, 1:100 dilution).

GLUT-1 labels the perineurium of peripheral nerves, while anti-CD 34 labels endothelial cells. With the latter stain, TRS was used as an antigen retrieval agent. The appropriate secondary antibodies and peroxidase-conjugated streptavidin were used with 3,30 -diaminobenzidine tetrahydrochloride (DAB) as chromogen. Positive labelling was evident as a brown reaction product. All histological slides were then scanned to provide digital images using the NanoZoomer (Hamamatsu, Japan) and analyses were performed using the NanoZoomer Digital Pathology (NDP) program. To assess the nature and extent of errors in the methodology and to determine the degree of intra-examiner reliability, three specimens were randomly selected from the sample pool, and their scanned images re-analysed.

2.1.

External morphology

The shape of the cross section of the lingual nerve was studied 2 mm above the lingula, as well as at the M3 and M2 regions, whereas the inferior alveolar nerve was only studied 2 mm above the lingula. The widest and narrowest diameters of each nerve segment were measured using the NDP program. After calculating the ratio between the widest and narrowest diameters, the external morphology of the nerves was classified as ‘circular’, ‘oval’, ‘flat’ or ‘mixed’. A ratio of 1 to 1.5 was classified as ‘circular’, a ratio between 1.6 and 2 was classified as ‘oval’, and a ratio greater than 2 was classified as ‘flat’. In circumstances where there were multiple nerve fascicles giving an irregular outline, the shape was classified as ‘mixed’.

2.2.

Internal morphology

2.2.1.

Fascicular disposition

The number of fascicles was counted manually for the lingual nerve at three levels (as described above) and only at 2 mm

Fig. 2 – Illustration showing the lingual nerve (LN) and inferior alveolar nerve (IAN) maintaining their relationship by excising some of the surrounding connective tissue proximally. A longitudinal piece of medial pterygoid muscle was placed along the IAN and enclosed in a loosely applied suture. Histological sections showed the proximity of the IAN to the muscle, facilitating identification.

archives of oral biology 59 (2014) 324–335

327

Fig. 3 – Light microscope sections stained with Masson’s Trichrome showing the range of shapes of lingual and inferior alveolar nerves found in the regions of interest. (A) circular; (B) oval; (C) flat; (D) mixed.

above the lingula for the inferior alveolar nerve. The mean, median, mode and range of fascicles were calculated for each nerve for the 23 cadavers.

2.2.2.

Endoneurial vasculature

As a measure of vascular density, the area of brown stain representing CD34 labelling of endothelial cells was calculated as a proportion of the total fascicular areas for the lingual and inferior alveolar nerves at 2 mm above the lingula using the NIH program Fiji, and the results were compared between the two nerves.

2.2.3.

Perineurium thickness

The number of perineurial lamellae of each fascicle of the lingual and inferior alveolar nerves was counted at 2 mm above the lingula. Perineurial layers were labelled for GLUT-1 immunohistochemically. For each nerve, the fascicle with the largest number of perineurial layers was used for the comparison. Of the 23 cadaver specimens, five had to be excluded from this analysis because the perineurium had lifted off the slides during GLUT-1 immunolabelling, to a degree that made counting of the lamellae impossible.

2.2.4.

Nerve tissue density

Nerve tissue density was investigated for the lingual and inferior alveolar nerves 2 mm above the lingula, and at the M3 and the M2 regions for the lingual nerve only. The area of each fascicle, inclusive of the endoneurial compartment and the perineurium, was calculated using the NDP program as a proportion of the whole nerve cross section, including the epineurium. This analysis described how much of a nerve cross section was made up of nerve tissue within the perineurium compared to the epineurial connective tissue.

2.3.

Statistical approaches

Both categorical and quantitative data were generated in the current study and ratios were also computed. Preliminary tests indicated that the data obtained for many of the variables did not conform to a normal distribution, with the skewness and kurtosis values beyond 1 and +1, therefore nonparametric statistical tests were performed for most analyses. A Wilcoxon Signed Ranks test was used to compare quantitative measurements obtained from the same cadaver, e.g., when comparing the numbers of fascicles and the nerve tissue

density between lingual and inferior alveolar nerves 2 mm above the lingula. A Friedman’s test was performed to compare quantitative measurements at the three different regions of the lingual nerve, and then Wilcoxon Signed Ranks tests were performed if a statistically significant relationship was found. Overall, statistical significance was set at p < 0.05.

3.

Results

3.1.

Errors of method

There was an overall high degree of intra-examiner reliability in categorising and measuring the variables that were studied. The concordance percentage for correct classification of external morphology was 92%. Minimal differences were noted in the second analysis of perineurial layers, with differences ranging from 3 to +3 between the first and second observations, leading to the same conclusions. There was also a strong concordance in the analysis of nerve tissue density, with differences ranging from +3.1% to 1.8% between both observations. However, the degree of intraexaminer reliability decreased as the number of fascicles increased in the nerve segment. For example, whilst monofascicular nerves were recognisable on all occasions, there were more discrepancies in counting the number of fascicles in polyfascicular nerve segments, as in the case of M3 and M2 regions. Differences ranged from +4 to +16 more fascicles on the second count for these regions. However, it should be noted that the nerves in these segments were still consistently classified as being polyfascicular.

3.2.

External morphology

3.2.1.

Shape

The different types of shapes of nerves observed in the specimens are illustrated in Fig. 3. Of the twenty-three lingual nerve specimens studied at 2 mm above the lingula, eight were circular (35%), eight were oval (35%), seven were flat (30%), and none were mixed. At the M3 region, one was circular (4%), four were oval (17%), fourteen were flat (61%), and four were mixed (17%). Of the four that were mixed, the main larger branch was flat in two of the specimens, and oval in the other two. At the M2 region, none were circular, two were oval (9%), fifteen were flat (65%), and six were mixed (26%). Of the six that

328

archives of oral biology 59 (2014) 324–335

Table 1 – Number of fascicles of the lingual and inferior alveolar nerves 2 mm above the lingula. Nerve 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Mean Median Mode Std. deviation Minimum Maximum a

Lingual nerve

Inferior alveolar nerve

2 1 7 9 3 1 1 1 9 6 10 4 2 2 7 4 2 1 7 3 1 1 5 3.9 3 1 3.00 1 10

9 11 4 8 8 16 8 5 22 10 14 12 14 10 13 12 4 14 9 18 10 12 11 11.0 11 8a 4.27 4 22

Multiple modes exist. The smallest value is shown.

were mixed, the main larger branch was flat in four of the specimens, and oval in the other two. Of the twenty-three inferior alveolar nerve specimens studied 2 mm above the lingula, nine were circular (39%), six were oval (26%), eight were flat (35%), and none were mixed. There was overall a 47% concordance (11 of the 23 specimens) between the shapes of the lingual and inferior alveolar nerves 2 mm above the lingula

3.3.

Internal morphology

3.3.1.

Fascicular disposition

For the lingual nerve, 2 mm above the lingula, the number of fascicles ranged from 1 to 10, with a median value of three fascicles (Table 1). Interestingly, in nine of the 23 specimens (39%), the lingual nerves were uni-fascicular. For the inferior alveolar nerve at the same level, there was a range of four to 22 fascicles, with a median value of 11 fascicles (Table 1). There were significantly more fascicles in the inferior alveolar nerve than the lingual nerve at this level (Z = 4.05, p < 0.001, n = 23). Of the twenty-three lingual nerve specimens studied at the M3 region, there was a range of nine to sixty-six fascicles, with a median value of 45 fascicles. The vast majority of the specimens (96%) were polyfascicular at this level. In the M2 region, there was a range of 17 to 78 fascicles, with the median value of fascicles being 50. At this region, the lingual nerve was always polyfascicular.

Fig. 4 – Light microscope section of lingual nerve (Masson’s Trichrome stain) obtained from the M3 region showing fascicles undergoing division or union (arrows). Divisions and unions of fascicles have been suggested to account for the frequent changes in the number of fascicles observed at different levels along nerves.

There was a significant difference in the number of fascicles of the lingual nerve at the three different regions, i.e., 2 mm above the lingula (median = 3), at the M3 region (median = 45), and the M2 region (median = 50) (x2 = 34.7, p < 0.001). There were fewer fascicles at 2 mm above the lingula than at either the M3 region ( p < 0.001) or the M2 region ( p < 0.001). However, there was no significant difference in the number of fascicles between the M3 and M2 regions. An interesting anatomical observation was the division and union of fascicles that occurred throughout the nerve, as illustrated in Fig. 4. This has been suggested by some authors to account for the frequent changes in the number of fascicles observed at different levels. Additionally, ganglionic neuronal cell bodies were seen intermingled within the lingual nerve in the M3 and M2 regions, near the submandibular ganglion (Fig. 5).

Fig. 5 – Light microscope section of lingual nerve from M2 region stained with H&E showing ganglionic neuronal cell bodies (arrows) intermingling within the nerve fibres.

archives of oral biology 59 (2014) 324–335

Fig. 6 – Light microscope section of inferior alveolar nerve immunolabelled with anti-CD34 primary antibody. Positive labelling of endothelial cells appears as brown reaction product. Most vessels appear in cross section suggesting their longitudinal course, with many horizontal vessels indicating links between longitudinal vessels. The area of brown staining was calculated in relation to the area of the fascicles within the perineurium to obtain a measure of vascular density.

3.3.2.

Endoneurial vasculature

The density of endoneurial blood vessels was measured using immunolabelling of endothelial cells with anti-CD34 antibody. Positive immunolabelling appeared as brown reaction product (Fig. 6). Scanned slides were analysed using NIH Fiji software to provide an assessment of the amount of brown stain in relation to the area of the endoneurium located within the confines of the perineurium in each fascicle. Two hundred and twenty fascicles from 20 inferior alveolar nerves, and 73 fascicles from 18 lingual nerves were analysed. The results from fascicles were grouped and a mean was obtained for each nerve. For the 18 lingual nerves, the mean density of brown stain per endoneurial area was 4.21% (SD 2.99, range 1.12– 9.83%). For the 20 inferior alveolar nerves, the mean density of

Table 2 – Nerve tissue density of the lingual nerve 2 mm above lingula, and at M3 and M2 regions, relative to the total cross-sectional nerve area.

Mean Median Mode Std. deviation Minimum Maximum a

2 mm above the lingula (%)

M3 region (%)

M2 region (%)

45.5 43.8 26.4a 12.15 26.4 85.7

25.7 24.0 13.6a 7.03 13.6 41.1

24.6 23.5 11.6a 6.38 11.6 37.5

Multiple modes exist. The smallest value is shown.

329

Fig. 7 – Light microscope section immunolabelled for GLUT1 showing brown positive staining of the perineurium surrounding fascicles of the lingual (Single arrow) and inferior alveolar (double arrows) nerves 2 mm above the lingula. Note the large uni-fascicular lingual nerve, compared with the oligo-fascicular inferior alveolar nerve. The longitudinal piece of muscle (M) was attached along the inferior alveolar nerve to aid in identification. Inset ‘A’ shows positive labelling of red blood cells within a venule. Inset ‘B’ shows labelling of perineurial partitions within the fascicle.

brown stain per endoneurial area was 4.67% (SD 3.06, range 1.1–9.75%). The means were not significantly different ( p = 0.32).

3.3.3.

Nerve tissue density

The mean nerve tissue densities of the lingual nerve and the inferior alveolar nerve 2 mm above the lingula were 45.5% and 40.7%, respectively, but these differences were not statistically significant. The nerve tissue density of the lingual nerve at the M3 and M2 regions were 25.7% and 24.6%, respectively (Table 2). There was a significant change in nerve tissue density in the lingual nerve across the three levels investigated: 2 mm above the lingula, and at the M3 and M2 regions (x2 = 31.9, p < 0.001). The nerve tissue density was significantly greater in the lingual nerve 2 mm above the lingula, than at both the M3 ( p < 0.001) and M2 regions ( p < 0.001). However, the difference in lingual nerve tissue density between the M3 and M2 regions was not statistically significant. This suggests that there is a larger epineurial component in the lingual nerve in the posterior molar region compared with the lingula region.

3.3.4.

Perineurium

The perineurium was labelled with anti- GLUT-1 (Fig. 7) and positive labelling was confirmed by labelling of red blood cells, which are known to express Glut-1 (Fig. 7). Anti-Glut-1 also labelled perineurial partitions within the fascicles (Fig. 7). Immunolabelling of the perineurium allowed counting of perineurial lamellae (Fig. 8). Generally, the number of layers

330

archives of oral biology 59 (2014) 324–335

literature. A ‘mixed’ category was also added to the classification, due to the branched lingual nerves that were observed in the M3 and M2 regions. A rational and repeatable method was formulated for identifying the shape of the nerve, based on the ratio between its widest and narrowest diameters.

4.1.2.

Fig. 8 – Light microscope section of inferior alveolar nerve immunolabelled for GLUT-1 showing positive labelling of the perineurium, which shows 10 consecutive cellular layers (arrows).

observed in the perineurium ranged from one to 16 layers. In 83% of cases (15 of the 18 specimens) 2 mm above the lingula, the lingual nerve fascicle with the largest number of perineurial layers had more lamellae than in all the fascicles of the inferior alveolar nerve. Of the three remaining specimens, in one cadaver the lingual nerve had more layers than 94% of the inferior alveolar nerve fascicles, and in the second specimen the lingual nerve had more layers than 80% of the inferior alveolar nerve fascicles. In the third cadaver, the lingual nerve had more layers than only 44% of the inferior alveolar nerve fascicles at the same level.

3.4. Relationship between internal and external morphology For the inferior alveolar nerve 2 mm above the lingula, there was no significant difference in the number of fascicles, or nerve tissue density, between flat, circular and oval shapes. Similarly, for the lingual nerve at each of the three levels, there was no significant association between external and internal morphologies.

4.

Discussion

4.1.

External morphology

4.1.1.

Shape

Knowledge of the cross-sectional morphology of the lingual nerve is important for clinicians as it may enhance understanding of nerve grafting procedures, where the apposition of a donor nerve with similar shape, size and fascicular disposition would enhance the success of recovery following neural injury.25,26 In the current study, a revised version of the common classification of ‘circular’, ‘oval’, ‘flat’ or ‘ribbon-like’ shapes was used. The external morphologies of ‘flat’ and ‘ribbon-like’ shapes were considered to be similar after a review of the

2 mm above the lingula

The current study indicates that the lingual and inferior alveolar nerves may be circular, oval or flat 2 mm above the lingula, which is typically where inferior alveolar nerve blocks are administered. Neither of the nerves was found to have a ‘mixed’ disposition, indicating that both nerves remain as a singular entity in this region. It is evident that the lingual nerve divides into more branches as it approaches the M3 and M2 regions. In approximately 50% of the cases, there was a concordance between external morphology of the lingual nerve and inferior alveolar nerve in the lingula region. This may suggest that there is a similar degree of influence from local anatomical structures in determining the external morphological characteristics of these nerves. There do not appear to be any other studies that have explored the external morphology of the inferior alveolar nerve or lingual nerve in this particular region.

4.1.3.

M3 region

The current study found that the lingual nerve may be either circular, oval, flat or mixed in the M3 region. However, it was more likely for the lingual nerve to be flat in this region, and least likely for it to be circular. These findings contrast with the reported trend in the literature, which suggests the lingual nerve to be predominantly circular,2,27 or oval6,10,28 in the M3 region. The first documented study in the literature that investigated the shape of the lingual nerve in the M3 region was conducted by Kiesselbach and Chamberlain.2 Through the dissection of 34 cadaveric adult heads, the authors observed that the lingual nerve was generally circular, although sometimes ovoid or flat, in the posterior molar region where the lingual nerve was located between the medial surface of the mandibular ramus and the medial pterygoid muscle. In the M3 region specifically, a majority of the lingual nerves investigated were circular (61.7%; 21/34), with some being oval (17.6%; 6/34), and flat or ribbon-like (20.5%; 7/34). The results of the study also indicated that the shape of the lingual nerve in the M3 region was not correlated with the distance from the medial surface of the mandible. An interesting finding by Kiesselbach and Chamberlain2 was the observation of flat lingual nerves as thin as 0.5 mm, which is the same size as a 25-gauge injection needle. This was postulated by the authors to explain how physical trauma to the lingual nerve may result from contact with an anaesthetic needle during inferior alveolar nerve blocks. Similarly, Miloro et al.27 reported that a ‘round’ shape was the most common external morphology of the lingual nerve observed in the M3 region. Through the use of high-resolution magnetic resonance imaging (HR-MRI) and modern modifications in image resolution, the authors were able to study the course of the lingual nerve without physical disturbance to nearby structures, and also obtain clear and direct crosssectional images of the lingual nerve as it traversed the M3

archives of oral biology 59 (2014) 324–335

region. Of the 20 lingual nerves examined in 10 healthy volunteers, 45% (9 of 20) were round, 30% (6 of 20) were elliptical, and 25% (5 of 20) were kidney bean shaped. Conversely, three separate cadaveric studies in the literature indicated a higher likelihood for the lingual nerve to adopt an oval external morphology in the M3 region. In a study conducted on 68 specimens, Ho¨lzle and Wolff6 found the lingual nerve to be oval in half of the cases, with a lesser likelihood of being circular or flat. Kim et al.10 conducted a study on 32 specimens and also described the oval shape to be the most prominent. Garbedian28 further supported this claim in a study of seven specimens, in confirming that the lingual nerve may be oval in 50% of circumstances in the M3 region, with a lesser likelihood of having other external morphologies. While some of the results of previous studies may differ, they are in agreement that the lingual nerve is predominantly oval or circular in morphology in the M3 region. This is in contrast with the findings of the current study, which suggest a higher likelihood for the lingual nerve to be flat in the M3 region.

4.1.4.

M2 region

The current study indicates that the lingual nerve may be either oval, flat or mixed in the M2 region, but most likely to be flat; a finding similar to the M3 region. In fact, the lingual nerve was never observed to be circular in the M2 region. When observing the external morphology of the lingual nerve as it progressed from the region of the lingula to the M3 and M2 regions, there was a tendency for the lingual nerve to become flatter and to have more fascicles. This finding confirms the hypothesis relating to Aim 1 of the current study. Overall, this trend may be due to adaptation of the lingual nerve to its position between the mandibular ramus and the medial pterygoid muscle as it descends inferiorly towards the medial surface of the mandibular angle, as postulated by Garbedian.28 Similarly, an anatomical study by Barker and Davies29 also supports this observation, where the lingual nerve was noted to become flattened between the muscle and mandibular ramus. The findings of the current study of the lingual nerve in the M2 region were similar to those of Kim et al.10 who described the lingual nerve to be flat or ‘ribbon-like’ in 78% of the specimens that they studied. However, they contrast with those of Garbedian,28 who described the lingual nerve to be ‘ribbon-like’ in 29% of their specimens and to have the greatest likelihood of being oval in shape (60%). Although the lingual nerve was never found to be circular in the current study, Kim et al.10 and Garbedian28 found the lingual nerve to be circular in 3.1% and 11% of cases, respectively.

4.2. Possible reasons for the different results between the current study and previously published studies The external morphology of the lingual nerve in the retromolar, M3 and M2 regions has been extensively documented in the literature.2,6,10,27,28 However, it is evident that these studies provide differing results from each other and the current study. This may be due to several reasons, such as a lack of any standardised method for describing nerve morphology. Additionally, variable classifications were

331

employed by different authors, with poor definitions and guidelines provided. For example, whilst some studies considered ‘flat’ and ‘ribbon-like’ to be the same shape, others considered them to be different. Furthermore, some were conducted on cadaver specimens and others on live patients through magnetic resonance imaging. Cadaveric dissections may have the disadvantage of post-mortem distortion, compression and/or morphological change as a result of formalin fixation. Some of the specimens in the current study had been fixed up to 4 years before analysis was performed. Although Ho¨lzle and Wolff6 concluded that the influence of formalin on the shape of the nerve was minimal, further research is required to determine the effect of formalin fixation on neural internal morphology. Some studies also failed to explain the method of identifying their regions of interest—for example, how the M3 region was identified when mandibular third molars were absent in the specimen. Compared to previous studies, with the exception of Garbedian,28 the current study was the first to employ anatomical landmarks on the internal oblique ridge to identify the M3 region; a method originally described by Pogrel et al.7 The results of the current study of the M3 region were expected to be similar to those of Garbedian,28 due to the similarity in defining the location of the regions. However, this was found not to be the case. The discordance between results of the current study and those of Garbedian28 for the M2 region may not be surprising, as the area studied by the latter author was identified through a different method. Kim et al.10 further suggested that the discrepancies may also be attributed to racial differences, as their data were based on Korean cadavers, whilst most of the literature is based on subjects of European descent. However, due to reasons previously discussed, there is little evidence to support this view.

4.3.

Internal morphology

4.3.1.

Fascicular disposition

Thorough understanding of the fascicular disposition of a nerve should lead to a more conclusive explanation of why a nerve may be particularly prone to sensory disturbance after anaesthetic or surgical events. It may also enhance surgical success in neural repair through nerve grafting procedures, where the selection of an appropriate donor nerve with similar fascicular size and patterns would improve results, or through the alignment of nerve segments in an anastomosis.25 The current study has shown the inferior alveolar nerve to have significantly more fascicles than the lingual nerve 2 mm above the lingula, which is where inferior alveolar nerve blocks would be administered typically. Whilst the lingual nerve was found to be uni-fascicular in more than a third of cases, the inferior alveolar nerve was never observed to be uni-fascicular at the same level. This is consistent with the hypothesis proposed in relation to Aim 2. Our findings are in agreement with those of Pogrel et al.,14 where the lingual nerve was found to be uni-fascicular in 33% of specimens examined 2 mm above the lingula. Similarly, in their observations, the inferior alveolar nerve was never uni-fascicular at the same level. An earlier study by Girod et al.30 also examined the fascicular structure of the lingual nerve after its fusion with the chorda

332

archives of oral biology 59 (2014) 324–335

tympani, and found the lingual nerve to be uni-fascicular in half of the cases. Aiming to explain the reported higher number of injuries sustained by lingual nerves compared to inferior alveolar nerves during inferior alveolar nerve blocks, Pogrel et al.14 postulated that a uni-fascicular nerve may be more prone to sensory dysfunction. Should damage occur to a uni-fascicular lingual nerve at the level of the lingula, whether through direct trauma from the injecting needle, haematoma formation, neurotoxicity or injection trauma, there would be functional disturbance in the entire anatomical distribution of the nerve due to a lack of compensatory function from other undamaged fascicles. Considering the tendency for the lingual nerve to be uni-fascicular at the level of the lingula, this theory may explain why the lingual nerve experiences more frequent injury than the inferior alveolar nerve during inferior alveolar nerve block injections, and why permanent disability is more likely to ensue. Taking into consideration the debilitation that occurs as a consequence of lingual nerve injury, the need for careful technique during the administration of inferior alveolar blocks is emphasised. Where possible, alternative anaesthetic techniques, such as intraligamentary, intraosseous, and local infiltrations using Articaine, may be considered in order to prevent unintentional nerve injury. In all of the 23 specimens studied, the lingual nerve had significantly greater number of fascicles at the M3 and M2 regions than 2 mm above the lingula. There was, however, no significant difference in the number of fascicles between the M3 and M2 regions. This finding is supported by those of Pogrel et al.14 who found that the lingual nerve had fewer fascicles at the level of the lingula than in the M3 region. However, whilst Pogrel et al.14 reported a range of seven to 39 fascicles at the M3 region, the current study observed a larger range of nine to 66 fascicles. Girod et al.30 also noted the lingual nerve to be either oligofascicular or polyfascicular before its connection with the submandibular ganglion, which is in the vicinity of the posterior mandibular molars. Overall, it can be concluded that the lingual nerve is never uni-fascicular at the posterior molar region, but may be either oligofascicular or polyfascicular, indicating that the fascicles divide at some point between the lingula and the M3 region. This conclusion is in agreement with that reached by Pogrel et al.,14 and supports the hypothesis associated with Aim 2 of the current study. The higher number of fascicles in the lingual nerve at this particular region may decrease the severity of neurosensory dysfunction sustained during clinical procedures, such as mandibular third molar extractions, as there would be compensatory function from other undamaged fascicles, unless the nerve experiences neurotmesis as in the case of complete transection.

4.3.2.

Endoneurial vasculature

It is well understood that the metabolic need of a biological tissue is reflected in its capillary density.31 The peripheral nervous system has been described in the literature as having a relatively low metabolic rate. This is reflected in its lower overall surface area available for metabolite exchange.32,33 Previous reports of peripheral nerve blood vessels in mammals, including humans, showed that the density of vessels is much less than other tissues such as the brain and skeletal

muscles.32 It has been shown that microvessels in the endoneurium are widely spaced and of large calibre.32 In the current study immunolabelling of endothelial cells with anti-CD34 supports the above findings as labelled endothelial cells formed a small percentage of the endoneurial surface area in the lingual (4.21%) and inferior alveolar nerve (4.67%). Most vessels in the current study were cut horizontally suggesting their longitudinal distribution, with many horizontal vessels that may represent the transverse loops that link the longitudinal channels as previously demonstrated in human and animal nerves by Bell and Weddell32 and in the sciatic nerve of the rat using micro-corrosion casts of the vasa nervorum with scanning electron microscopy.34

4.3.3.

Perineurium

The observations of the current study indicate that the perineurium ranges from 1 to 16 layers, 2 mm above the lingula, with some variability between the lingual and inferior alveolar nerves. This is largely in agreement with the literature, where the perineurium has been documented to contain one to 15 layers, with frequent variations along the length of the nerve and between nerves.31 The findings of the current study further suggest that, in the majority of cases, the lingual nerve contains a greater number of perineurial lamellae than any of the fascicles in the inferior alveolar nerve. It may therefore be postulated that the lingual nerve generally has more perineurial layers than the inferior alveolar nerve at the level where inferior alveolar nerve blocks is administered. A thicker perineurium has been proposed to confer greater mechanical strength and stiffness to the associated nerve.31 This anatomical difference between the lingual and inferior alveolar nerves may provide further insight into why the lingual nerve suffers a higher incidence of injury after the administration of inferior alveolar nerve blocks. In the situation where there is a rise in intrafascicular tissue pressure, for example, through oedema, the endoneurial arterioles and capillaries, and especially the oblique perineurial vasculature, may collapse from compression, consequentially resulting in intrafascicular ischaemia.35 Considering the restricted compliance and limited diffusion that a thicker perineurium would confer, the changing pressure effects would be transmitted over a greater distance, resulting in remote arterial reactions.32 Given that the lingual nerve was shown to have a uni-fascicular tendency in the area where inferior alveolar nerve blocks are administered, it is possible that the lingual nerve suffers more severe consequences from changes in the endoneurial tissue pressure, with subsequent functional disturbance along its entire distribution. Such a change of pressure may be caused from the intrafascicular injection of a neurotoxic local anaesthetic, or through the formation of an intrafascicular haematoma. As the perineurium provides mechanical strength to the nerve, it is not always homogeneous along the whole length of the nerve, or between nerves.31 Generally, it is recognised that more layers exist in regions where there is greater nerve strain; for example, where nerves cross joints, such as at bifurcations and branches.31 The lingual nerve may have more perineurial lamellae than the inferior alveolar nerve because it experiences greater nerve stretching, for instance, during

archives of oral biology 59 (2014) 324–335

tongue movements or when it is stretched taut during full opening of the mouth, which is a position assumed by the mandible during the administration of inferior alveolar nerve blocks.20 Nerves with fewer fascicles are also associated with a greater number of perineurial layers. This may be because nerves consisting of more fascicles allow longitudinal and transverse movement in response to tensile and compressive forces on the nerves and, as such, do not require as much protection from the perineurium.31 Taking into consideration that the lingual nerve was shown to have significantly fewer fascicles than the inferior alveolar nerve at the level of the lingula, this may provide an explanation as to why the lingual nerve may have more perineurial lamellae. There is also an established trend where larger fascicles are associated with greater lamellae to help modulate compressive and tensile forces.31,36 Although the current study did not explore the difference in fascicular size between the lingual nerve and inferior alveolar nerve, it is likely that the lingual nerve contains larger fascicles, especially in uni-fascicular patterns. Other factors that may influence the number of layers within the perineurium include the vicinity of sensory and motor end organs, where a single layer of perineurial cells is frequently reported,35 and advancing age, where the number of layers has been described to increase.36

4.3.4.

Nerve tissue density

The current study revealed that nerve tissue density varies along the length of a peripheral nerve and may range from 12% to 63% of the cross-sectional area of the nerve. This suggests that the relative area of the epineurium may range from 37% to 88%. This is in agreement with descriptions of the epineurium in the literature, where it has been reported to constitute from 30% to 75% of the cross-sectional area of the nerve, with extremes of 22% and 88%.36 The current study indicated that there was no significant difference in the nerve tissue density between the lingual nerve and inferior alveolar nerve 2 mm above the lingula, as hypothesised in Aim 2. However, the lingual nerve had a significantly lower nerve tissue density in the posterior molar region compared with the level of the lingula, implying that the lingual nerve had a larger epineurial component in the posterior molar region. This is consistent with this study’s hypothesis and is not a surprising observation, as a higher quantity of epineurial connective tissue is often associated with greater number of fascicles, which is characteristic of the lingual nerve in the posterior molar region.31 The larger epineurial component of the lingual nerve in the posterior molar region may offer greater protection against potential trauma, as opposed to the lesser epineurial component of the lingual nerve at the level of the lingula. The epineurium is known to play a protective role in dispersing tensile and compressive forces.22 It permits transverse and longitudinal movement of nerve fascicles, allowing them to glide over one another—a characteristic which may prevent direct trauma to the nerve tissue.31 The epineurium also plays a supportive role in encouraging regeneration and regaining of function. A larger epineurial component may improve diffusion of pressure within the nerve; for example, from intraneural oedema or haematoma, thereby enabling quicker recovery from pressure-related injuries. Additionally, as the

333

lingual nerve is polyfascicular in the posterior molar region, it may also be associated with a thinner perineurium, which would further improve diffusion of intrafascicular pressure.31 Furthermore, diffusion within the lingual nerve at the level of the lingula could also be restricted due its tendency to be stretched taut during mandibular opening, further limiting compliance within the nerve.20 It may therefore be postulated that the lingual nerve is more able to recover from intraneural pressure-related injuries in the posterior molar region where M3 surgery is often performed, rather than at the level of the lingula where inferior alveolar nerve block injections are administered. Unfortunately, as the literature presents inconclusive data comparing the incidence of permanent lingual nerve damage after M3 surgery and inferior alveolar nerve blocks, it is difficult to find supporting evidence for this theory.

4.4. Relationship between internal and external morphology The findings of the current study indicated that there was no difference in the number of fascicles or nerve tissue density between flat, circular, oval or mixed shapes, for both lingual and inferior alveolar nerves in the regions of interest. This suggests that there is no correlation between the internal and external morphology of a peripheral nerve, so our hypothesis relating to Aim 1 is not supported. There do not appear to be any previous studies in the literature that have compared these two variables.

4.5.

Immunohistochemical staining

The cadaver specimens used in the current study had been preserved for 1–4 years prior to dissection and histological processing. Interestingly, although these specimens were not ‘fresh’, immunohistochemical staining for GLUT-1 and CD 34 was still able to provide satisfactory results. Thus, this study demonstrated successful immunolabelling of endothelial cells of the endoneurial vasculature and the perineurium in human material that was embalmed for routine dissection.

4.6.

Limitations of the study

Whilst the methodology employed to determine a nerve’s external morphology provided a logical and repeatable means of classification, it was not without limitations. For example, a ‘kidney-bean’ shape, as first described by Miloro et al.27 was encountered in two lingual nerve specimens; one 2 mm above the lingula, and another at the M3 region. However, they were classified as circular and flat shapes, respectively, after their diameters were measured. Additionally, triangular shapes were encountered several times, but these were classified as either ‘circular’ or ‘oval’, depending on the ratio of their widest and narrowest diameters. Regardless, the current study has emphasised the extensive variability of the lingual nerve’s external morphology, and should provide clinicians with some guidance about the shapes that might be expected in various regions of interest during surgical procedures. Some specimens were compromised, as parts of the neural tissue had lifted off from the glass slides. In such circumstances, it was usually the perineurium which was folded.

334

archives of oral biology 59 (2014) 324–335

Fortunately, most of the lifting was only mild and, as such, the sections could still be used for analysis; for example, in the investigation of the perineurium, where other areas of the perineurium around the fascicle were used. Specimens in which perineurial layers could not be counted were excluded from the study. This study demonstrated successful immunolabelling of endothelial cells of the endoneurial vasculature using antiCD34 antibody, and the brown reaction product indicative of positive labelling was assessed as a percentage of the endoneurial surface area. However, there was wide variation in the percentage of labelling within both the lingual and inferior alveolar nerves, thus no statistically significant differences were detected between the two nerves. Despite the successful immunolabelling, it is likely that the variable times in which different cadavers were stored prior to dissection may have contributed to variability in labelling.

5.

Conclusion

The lingual nerve may be more susceptible to trauma than the inferior alveolar nerve following inferior alveolar nerve blocks due to its tendency to be uni-fascicular and its thicker perineurium, which may contribute to axonal compression in cases of increased endoneurial pressure, from either oedema or haemorrhage. This would consequentially lead to a miniature closed-compartment syndrome, which may result in a disturbance in nerve function.37 Nerves with multifascicular composition have a greater epineurial connective tissue, which may allow trauma-induced oedema or haemorrhage to disperse with less pressure on the axons. Better training in the administration of inferior alveolar nerve blocks and the use of alternative anaesthetic techniques when possible are recommended in order to decrease the occurrence of unintentional lingual nerve trauma.

Funding This research was funded by the Australian Dental Research Foundation and the Craniofacial Biology Research Group, The University of Adelaide.

Competing interests None declared.

Ethical approval Not required.

Acknowledgements This research was funded by the Australian Dental Research Foundation and the Craniofacial Biology Research Group, The University of Adelaide. Sincere thanks to Wesley Fisk and staff

in the Raymond Last Laboratory for their assistance. Thanks also to Angela Kinnell for her statistical input and to Suzanna Mihailidis and Jason Khoury for their co-supervision of Vui Tan’s honours project.

references

1. Piagkou M, Demesticha T, Skandalakis P, Johnson EO. Functional anatomy of the mandibular nerve: consequences of nerve injury and entrapment. Clin Anat 2011;24(2): 143–50. 2. Kiesselbach JE, Chamberlain JG. Clinical and anatomic observations on the relationship of the lingual nerve to the mandibular third molar region. J Oral Maxillofac Surg 1984;42(9):565–7. 3. Moore KL. Clinically oriented anatomy. 3rd ed. Baltimore, MD: Williams & Wilkins; 1992. 4. Norton NS. Netter’s head and neck anatomy for dentistry. Philadelphia, PA: Saunders; 2007. 5. Zur KB, Liancai M, Sanders I. Distribution pattern of the human lingual nerve. Clin Anat 2004;17(2):88–92. 6. Ho¨lzle FW, Wolff KD. Anatomic position of the lingual nerve in the mandibular third molar region with special consideration of an atrophied mandibular crest: an anatomic study. Int J Oral Maxillofac Surg 2001;30(4):333–8. 7. Pogrel MA, Bryan J, Regezi J. Nerve damage associated with inferior alveolar nerve blocks. J Am Dent Assoc 1995;126(8):1150–5. 8. Pogrel MA, Thamby S. The etiology of altered sensation in the inferior alveolar, lingual, and mental nerves as a result of dental treatment. J Calif Dent Assoc 1999;27(7):531–8. 9. Graff-Radford SB, Evans RW. Lingual nerve injury. Headache 2003;43(9):975–83. 10. Kim SY, Hu KS, Chung IH, Lee EW, Kim HJ. Topographic anatomy of the lingual nerve and variations in communication pattern of the mandibular nerve branches. Surg Radiol Anat 2004;26(2):128–35. 11. Baqain ZH, Abukaraky A, Hassoneh Y, Sawair F. Lingual nerve morbidity and mandibular third molar surgery: a prospective study. Med Princ Pract 2010;19(1):28–32. 12. Chan HL, Leong DJM, Fu JH, Yeh CY, Tatarakis N, Wang HL. The significance of the lingual nerve during periodontal/ implant surgery. J Periodontol 2010;81(3):372–7. 13. Pogrel MA, Thamby S. Permanent nerve involvement resulting from inferior alveolar nerve block. J Am Dent Assoc 2000;131(7):901–7. 14. Pogrel MA, Schmidt BL, Sambajon V, Jordan RCK. Lingual nerve damage due to inferior alveolar nerve blocks: a possible explanation. J Am Dent Assoc 2003;134(2):195–9. 15. Merrill RG. Prevention, treatment and prognosis for nerve injury related to the difficult impaction. Dent Clin North Am 1979;23(3):471–88. 16. Haas D, Lennon D. A 21-year retrospective study of reports of paraesthesia following local anaesthetic administration. J Can Dent Assoc 1995;61(4):319–30. 17. Garisto GA, Gaffen AS, Lawrence HP, Tenenbaum HC, Haas DA. Occurrence of paresthesia after dental local anesthetic administration in the United States. J Am Dent Assoc 2010;141(7):836–44. 18. Khoury J, Mihailidis S, Ghabriel M, Townsend G. Anatomical relationships within the human pterygomandibular space: relevance to local anaesthesia. Clin Anat 2010;23(8):936–44. 19. Ogle OE, Mahjoubi G. Local anaesthesia: agents, techniques, and complications. Dent Clin N Am 2012;56(1):133–48. 20. Harn SD, Durham TM. Incidence of lingual nerve trauma and postinjection complications in conventional

archives of oral biology 59 (2014) 324–335

21.

22. 23.

24.

25.

26. 27.

28.

29. 30.

mandibular block anaesthesia. J Am Dent Assoc 1990;121(4):519–23. Blanton PL, Jeske AH. Avoiding complications in local anaesthesia induction: anatomical considerations. J Am Dent Assoc 2003;134(7):888–93. Smith MH, Lung KE. Nerve injuries after dental Injection: a review of the literature. J Can Dent Assoc 2006;72(6):559–64. Renton T, Adey-Viscuso D, Meechan JG, Yilmaz Z. Trigeminal nerve injuries in relation to the local anaesthesia in mandibular injections. Br Dent J 2010;209(9):E15. Pretterklieber ML, Skopakoff C, Mayr R. The human maxillary artery reinvestigated: I. Topographical relations in the infratemporal fossa. Acta Anat (Basel) 1991;142(4):281–7. Svane TJ, Wolford LM, Milam SB, Bass RK. Fascicular characteristics of the human inferior alveolar nerve. J Oral Maxillofac Surg 1986;44(6):431–4. Wolford LM, Stevao EL. Considerations in nerve repair. Bayl Univ Med Cent Proc 2003;16(2):152–6. Miloro M, Slone W, Chakeres D. Assessment of the lingual nerve in the third molar region using magnetic resonance imaging. J Oral Maxillofac Surg 1997;55(2):134–7. Garbedian JWJ. The relationship of the lingual nerve to the 3rd molar region: a three dimensional analysis. M.Sc. thesis. Toronto, ON: University of Toronto; 2010. Barker BCW, Davies PL. The applied anatomy of the pterygomandibular space. Br J Oral Surg 1972;10(1):43–55. Girod SC, Neukam FW, Girod B, Reumann K, Semrau H. The fascicular structure of the lingual nerve and the chorda

31.

32.

33.

34.

35.

36.

335

tympani: an anatomic study. J Oral Maxillofac Surg 1989;47(6):607–9. Topp KS, Boyd BS. Peripheral nerve: from the microscopic functional unit of the axons to the biomechanically l oaded macroscopic structure. J Hand Ther 2012;25(2): 142–51. Bell MA, Weddell AG. A descriptive study of the blood vessels of the sciatic nerve in the rat, man and other mammals. Brain 1984;107(Pt 3):871–98. Cameron NE, Mathias CJ. Structure and function of the nervous system. In: Gries FA, Cameron NE, Low PA, Ziegler D, editors. Textbook of diabetic neuropathy. Germany: Thieme; 2003. p. 40–63. Mackenzie ML, Allt G. The vasa nervorum: microcorrosion casts for scanning electron microscopy. Acta Anat (Basel) 1989;136(4):319–24. Weerasuriya A, Mizisin AP. The blood-nerve barrier: structure and functional significance. Methods Mol Biol 2011;686:149–73. Thomas PK, Olsson Y. Microscopic anatomy and function of the connective tissue components of peripheral nerve. Dyck PJ, Thomas PK, Lambert EH, Bunge R, editors. Peripheral neuropathy, II. Philadelphia, PA: Saunders Company; 1984.

p. 97. 37. Lundborg G. Intraneural microvascular pathophysiology as related to ischemia and nerve injury. In: Daniel RK, Terzis JK, editors. Reconstructive microsurgery. Boston, MA: Little Brown and Co; 1977. p. 334.