oral and maxillofacial radiology

Editor:
ALLAN G. FARMAN, BDS, PhD (Odont), MBA
Department of Primary Patient Care
University of Louisville School of Dentistry
Louisville, Kentucky 40292

Cross-sectional tomography
A diagnostic technique for determining the buccolingual relationship of
impacted mandibular third molars and the inferior alveolar neurovascular
bundle

Craig S. Miller, DMD, MS,a Pirkka V. Nummikoski, DDS, A4Sb

Douglas A. Barnett, DMD,c and Robert P. Langlais, DDS, MS,d
Lexington, KY., San Antonio, Texas, and Portland, Ore.

UNIVERSITY OF KENTUCKY COLLEGE OF DENTISTRY, UNIVERSITY OF TEXAS SCHOOL OF

DENTISTRY, AND OREGON HEALTH SCIENCES UNIVERSITY

Twenty-two patients with 31 impacted mandibular third molars were examined with a new, precise,
cross-sectional tomographic technique to assess the radiographic size, shape, branching pattern, location,
and degree of cortication of the mandibular canal, and the inclination of impacted mandibular third molars
in the buccolingual plane. The mandibular canal, including bifid canals, was accurately identified in 30
cases (96.8%). The cross-sectional appearance of the canal was an uncorticated, or partially corticated,
radiolucent oval that measured on average (5 SD) 2.9 f 0.7 X 2.5 5 0.6 mm in diameter. It was located
more frequently (45.2%) on the buccal aspect of the impacted mandibular third molar. About 60% of the
mandibular canals notched the inner cortical plate of the mandible or the third molar root surface. Cystic
expansion and quantification of cortical bone destruction were readily assessed by this technique. It was
concluded that diagnostic information obtained from cross-sectional tomograms significantly aids the oral
and maxillofacial surgeon during the preoperative diagnostic workup and that the radiation risks are
comparable to those of other accepted localization techniques.
(ORAL SURC ORAL MED ORAL PATHOL 1990;70:791-7)

R emoval of mandibular third molars is one of the

most common surgical procedures performed in the
of the respondents had treated patients who had sus-
tained nerve damage after mandibular third-molar
United States, with estimates that more than 2 mil- odontectomy. Permanent nerve damage had occurred
lion third molars are extracted annually. A frequent in 418 cases, and 2399 cases of temporary nerve
and serious complication associated with the removal damage were reported. In a retrospective study, Kipp
of impacted mandibular third molars is injury to the et aLi5 found that mandibular nerve dysesthesia oc-
sensory nerve bundle. 2-13 In a 1973 survey of more curred in 4.4% of the cases, whereas Osborn et a1.13
than 1200 oral surgeons, Schwartz14 found that 66% reported neurologic complications in 0.9%. Overall,
the incidence of inferior alveolar nerve dysesthesia
ranges between 0.57% and 5.3%.16-20Etiologic factors
*Assistant Professor, Department of Oral Health Science, Univer- associated with nerve dysesthesia include relative
sity of Kentucky College of Dentistry. proximity of the nerve to the tooth, experience of the
bAssistant Professor, Department of Dental Diagnostic Science, surgeon, age of the patient, traumatic tissue manip-
University of Texas School of Dentistry at San Antonio. ulation, and postsurgical edema. 3, 21,22
CAssistant Professor, Department of Oral Radiology, School of To help prevent neurologic complications, localiz-
Dentistry, Oregon Health Sciences University.
dProfessor, Department of Dental Diagnostic Science, University ing the accurate position of the inferior alveolar nerve
of Texas School of Dentistry at San Antonio. in relation to the impacted third molar is a recom-
7/16/19992 mended diagnostic strategy. Historically, researchers
791
792 Miller et a/.
Fig. 1. Geometric calculations used to determine appro-
priate tomographic setting and cut.
using anatomic dissections and radiographic studies
have mapped the anteroposterior and inferosuperior
location of the mandibular canal as it courses through
the mandible.23e27 However, until recently the medio-
lateral position of the inferior alveolar neurovascular
bundle has not been well defined.
The study of Rajchel et a1.28 on 45 Asian adult
mandibles demonstrated that the mandibular canal,
when proximal to the third-molar region, is usually a
single large structure, 2.0 to 2.4 mm in diameter. It
courses approximately 2.0 mm from the inner lingual
cortex, 1.6 to 2.0 mm from the medial aspect of the
buccal plate, and about 10 mm from the inferior bor-
der of the mandible. Because these anatomic findings
suggest only general mappings of the mandibular ca-
nal and do not account for the specific location each
patient displays, individual radiographs are required
to assessthe accurate position of the canal in question.
The purpose of this study was to devise a tomo-
graphic technique for imaging impacted mandibular
third molars and the mandibular canal in cross-
section. By localizing the canal in this format and by
using a standard panoramic radiograph, a three-
dimensional composite is obtained that provides the
surgeon with diagnostic information in the buccolin-
gual plane that may be needed for accurate decision-
making when dealing with the difficult impaction.
Linear cross-sectional tomography was used to image
the principal structures of interest, because this tech-
nique is readily available and less expensive to oper-
ite than tomographic units with more complex mo-
tion patterns.
MATERIAL AND METHODS
Twenty-two third-molar surgery patients, referred
iom the department of oral and maxillofacial sur-
:ery, agreed to participate in this study. Sixteen pa-
ORAL SURG ORAL MED ORAL PATHOL
December 1990
tients were women; six were men. Their ages ranged
from 19 to 54 years, with a mean of 36.5 years. Pre-
operative panoramic radiographs were taken on all
patients. Impacted mandibular third molars that were
in close apposition to the mandibular canal, as evident
in the panoramic radiograph, were selected for tomo-
graphic examination.
Cross-sectional tomographic radiographs were
made with the use of the Quint Sectograph linear to-
mographic unit (Denar Corp., Anaheim, Calif). Ra-
diographic exposures were recorded on Kodak 8 x 10
inch T-Mat G film using Kodak Lanex Medium
intensifying screens (Eastman Kodak Co., Rochester,
N-Y.), Exposure factors were initially set at 80 kVp,
50 mA, and a 2.75-second exposure time. The first
cross-sectional film was developed as a scout view, and
the section location and film exposure were evaluated
individually. The final exposure setting varied from 78
to 82 kVp. No additional filtration was needed to
minimize burnout, because the radiation exposures
selected produced optimum density in the area of the
canal. The narrow tomographic layer setting (3 mm)
was used, and the beam was collimated to a 60 x 60
mm square on the film. Radiation dosage at the skin
surface was measured by an mdh 1015 x-ray monitor
(mdh Industries, Inc., Monrovia, Calif.).
Projection geometry was optimized by seating the
patient facing the film cassette, thereby placing the
principal structures closest to the film, and by using
a focus-to-object distance of 5 feet, which is inherent
to sectograph tomography. The beam was aimed from
behind tangentially to the mandibular ramus and the
cassette was set 15 cm from the zero line. The mag-
nification factor ranged from 1.07 to 1.10 depending
on the tomographic cut position.
The mandibular third molar was oriented within
the selected focal plane by laterally rotating the
cephalostat and the patients head 25 degrees away
from the midsagittal machine setting. The patients
mandible was tipped superiorly to orient the long axis
of the impacted tooth perpendicular to the central ray
and to visualize better the periapical region. The de-
gree of superior tilt was determined by the angle of the
impaction, as viewed on the panoramic film. Patients
with mesioangular impactions were instructed to tilt
the head vertically at an angle commensurate with the
degree of tooth tipping (usually about 30 degrees).
Patients with vertical impactions required no tipping,
and patients with horizontal impactions were tilted 45
degrees superiorly. Patients were not tilted beyond 45
degrees, to prevent superimposition of radiodense
structures of the basilar region of the skull over the
field of interest.
To position the impacted mandibular third molar in
the focal plane of the tomographic cut, its antero-
posterior position was determined from measure-
Volume 70
Number 6
ments derived from a submentovertex radiograph
taken on the Quint Sectograph unit and a panoramic
radiograph taken on the Versaview panoramic unit (J.
Morita Corp., Osaka, Japan) as follows:
Procedure. On the submentovertical radiograph, the in-
termeatal base line (ZBL) was drawn joining the ear rod
markers, and the midpoint or center of rotation (CR) was
identified (Fig. 1). A line tangent to the buccal surfaces of
the posterior mandibular teeth was drawn and labeled BCP.
A third line, termed the zero line, was drawn perpendicular
to BCP joining CR to P. The angle (0) between the zero line
and IBL represents the optimum lateral rotation of the pa-
tient, usually about 25 degrees.The cephalostat was rotated
to the calculated angle ((Y)subtracted from 90 degrees(e.g.,
90 - 25 = 65 degrees) and set at this position. The distance
(X) from P to the distal surface of the third molar was mea-
sured and recorded as the approximate distance of the zero
line to the third molar. Becausea vertical upward tilt of the
patients mandible was used to position the tooth properly,
an additional calculation was made to compensate for the
anterior movement of the third molar with respect to the
tomographic zero line during the vertical head tilt proce-
dure. The magnitude of this forward movement was ap-
proximated from the panoramic film by tracing the film,
then rotating the film upward (while maintaining the orig-
inal position of the trace paper) and measuring the relative
change in tooth position. The measurement derived from
the panoramic film was corrected for magnification and
added to the distance X. Then the fulcrum (cut position) of
the Quint Sectograph unit was adjusted to correspond with
this number, thereby placing the focal plane of the ma-
chine on the impacted mandibular third molar.
The first film was exposed at an average tomo-
graphic cut position of 24 mm, which approximated
the distal aspect of the impacted mandibular third
molar. After evaluating the distal tomogram for the
proper position and exposure, two additional tomo-
grams were exposed at 4 mm increments. A total dis-
tance of 12 mm recorded radiographically the entire
anteroposterior width of the tooth (Fig. 2).
Tomograms were examined by four viewers who
had advanced training in oral and maxillofacial radi-
ology. Viewing was conducted under ideal conditions,
which included the use of subdued light, film masking,
and magnifying lenses. After reviewing the pan-
oramic and tomographic radiographs, impacted man-
dibular third molars were classified according to the
depth and angulation of the impaction as mesioangu-
lar, distoangular, buccoangular, linguoangular, buc-
coversion, linguoversion, horizontal, vertical, or in-
verted. By consensus the mandibular canal was iden-
tified by first estimating on the panoramic radiograph
the inferosuperior position of the canal relative to the
impacted third molar; then the appropriate region of
the tomogram was visually scanned for evidence of a
corticated, oval radiolucency approximately 2.8 mm
in diameter. The inner cortical plates of the mandible
and the root surface of the impacted molar were
Cross-sectional tomography 793
Inferior Alveolar
Neurovascular Bundle
Fig. 2. Illustration of tomographic cuts used to visualize
impacted mandibular third molar and mandibular canal.
closely examined at the appropriate inferosuperior
position for a distinct concavity (notch or groove),
which indicated that the neurovascular bundle was in
contact with the mandibular cortical plate or root
surface.
Working as a group, the viewers were asked to lo-
calize the mandibular canal with respect toadjacent
anatomic structures by indicating the following: (1)
the position of the mandibular canal (lingual, infero-
lingual, inferior, inferobuccal, or buccal to the im-
pacted tooth), (2) the distance (in millimeters) from
the external root surface to the mandibular canal, (3)
the presence or absence of cortication of the mandib-
ular canal, (4) the presence of a concavity of the tooth
or root produced by the inferior alveolar neurovascu-
lar bundle, and (5) the presence of a concavity of the
internal cortical marginal of the buccal or lingual
cortical plate caused by the inferior alveolar neu-
rovascular bundle. Cortication of the mandibular ca-
nal evident on the cross-sectional tomogram was
compared with the presence of cortication of the
mandibular canal evident on the panoramic radio-
graph. The location of the mandibular canal was de-
termined by postoperative consultation with the oral
and maxillofacial surgeon who directly observed the
inferior alveolar neurovascular bundle or indirectly
assessed its location by examining the tooth and ex-
traction socket for indentations suggestive of the neu-
rovascular bundle.
RESULTS
Thirty-one impacted mandibular third molars were
radiographically recorded by cross-sectional tomog-
raphy. Using the American Academy of Oral and
Maxillofacial Surgery classification of impacted man-
dibular third molars, the authors found that 14 were
vertical impactions (45.2%), seven were mesioangular
impactions, (22.6%), seven were horizontal impac-
tions (22.6%), two were linguoangular impactions
(6.5%), and one was a linguoversion impaction (3.2%).
Thirty mandibular molars (96.8%) were complete
794 Miller et al.
Fig. 3. A, Panoramic radiograph of a 44-year-old woman
with impacted mandibular third molars and bilateral bifid
mandibular canals. B, Cross-sectional tomogram reveals
superior fork of left canal (arrow) as oval radiolucency
notching buccal surface of midroot of impacted molar, and
inferior fork (lined arrow) notching inner margin of lingual
cortical plate.
bony impactions, and one (3.2%) was a partial bony
impaction. The average surface radiation dose mea-
sured during a single cross-sectional tomogram with
a 60 X 60 mm collimation was 0.65 mGy (65 mrad)
in the peripheral region of 36 cm2 with a maximum
of 1.3 mGy (130 mrad) in the center of the collima-
tion.
The mandibular canal was identified in 30 cases
(96.8%). One case was indeterminate. There were 29
single mandibular canals (93.5%) and two bifurcated
canals (6.5%) (Fig. 3). According to criteria for
inclusion in the study, the 31 mandibular canals ap-
peared radiographically to contact the impacted third
molar crown or root surface.
The size and shape of the mandibular canal were
variable, depending on the relative position of the in-
ferior alveolar neurovascular bundle to the root
apexes of the impacted tooth and cortical plate. The
majority of the canals were oval with a mean ( f SD)
0R.M SURG OR&i. MEI> ORAL PATHOI.
December 1990
Fig. 4. A, Panoramic radiograph of a 25year-old woman
with impacted mandibular third molars. B, Cross-sectional
tomogram reveals left mandibular canal (arrow) as a radi-
olucent structure 3 mm in diameter located lingual to im-
pacted third molar and in contact with apical third of root
surface and inner lingual cortical plate. Inferior surface of
mandibular canal appears corticated.
vertical-horizontaldiameterof2.9 k 0.7 X 2.5 f 0.6
mm. The largest diameter of a mandibular canal was
4.0 mm, and the smallest was 1.2 mm. The latter oc-
curred when the canal passed between the roots of an
impacted mandibular third molar. Correcting for the
magnification inherent in this cross-sectional tomo-
graphic technique, the mean diameter of the mandib-
ular canal was calculated to be 2.6 + 0.6 X 2.2 2 0.5
mm, with a range from 1.1 to 3.6 mm.
The location of the mandibular canal varied and
could not be predicted with certainty from the
panoramic radiograph (Fig. 4). The locations of the
31 mandibular canals identified relative to the im-
pacted tooth were as follows: buccal, 11 (35.5%); lin-
gual, 9 (29.0%); inferior, 5 (16.1%); inferolingual, 3
(9.7%); and inferobuccal, 3 (9.7%).
The cortication of the mandibular canal evident in
the panoramic radiograph did not serve as a predic-
tor of cortication in the cross-sectional tomogram.
Only eight (25.8%) canals evident in tomograms ap-
peared completely corticated, whereas four (12.9%)
were partially corticated (Fig. 5). In the panoramic
film, 97% of the mandibular canals in the third-molar
region were corticated along the superior and inferior
Volume 70
Number 6
Fig. 5. A, Panoramic radiograph of 31-year-old man
with horizontally impacted mandibular third molars. B,
Cross-sectional tomogram reveals left mandibular canal
(corticated only along inferior margin) contacting impacted
molar root and lateral aspect of lingual cortical plate.
margin. Cortication of the mandibular canal on the
panoramic film, however, did serve as a predictor of
the proximity of the mandibular canal to the cortical
plates. Ninety percent of the mandibular canals in
contact with the mandibular cortical plate (buccal or
lingual) were corticated in the panoramic radiograph.
The inferior alveolar neurovascular bundle pro-
duced a concave depression in adjacent anatomic
structures (the inner cortical plate of the mandible or
impacted molar) in 20 (64%) of the cases. These con-
cavities were evident on 11 lingual cortical plates
(35.5%) 5 buccal cortical plates (16.1%), and 4 im-
pacted mandibular third molars (12.9%). The canal
appeared completely confined to the medullary bone
in 11 patients (35.5%).
Two dentigerous cysts associated with impacted
mandibular third molars were evident in the cross-
sectional tomograms. Cystic expansion, multilocula-
tions, and thinning of the lingual and buccal cortical
Cross-sectional tomography 795
Fig. 6. A, Panoramic radiograph of a 61-year-old man
with a follicular cyst associatedwith impacted left mandib-
ular third molar. B, Cross-sectional tomogram reveals cys-
tic expansion of buccal and lingual cortical plates.
plates were apparent (Fig. 6). The accurate location
of one mandibular canal could not be determined be-
cause of the size and location of one of the radiolucent
cysts.
DISCUSSION
Injury to the sensory nerve bundle from removal of
impacted teeth is a frequent and serious complication
of mandibular third-molar surgery.2 3 Retrospective
studies indicate that nerve dysfunction after mandib-
ular third-molar surgery is the third most common
complication after alveolar osteitis and postoperative
infections, with an approximate incidence of 1% to
5%.13-20 Although this condition usually disappears
within a few months, it can be distressing, especially
if it becomes a persistent condition.
A recommended preventive diagnostic strategy is
the appropriate use of dental radiographs. These films
should provide detailed information that permits the
clinician to localize and determine in three dimensions
the size, shape, branches of the mandibular canal, in-
clination of roots, and relationship of the tooth to im-
portant adjacent anatomic structures. Once the im-
pacted mandibular third molar has been classified and
the mandibular canal localized, the surgeon can pre-
cisely plan the operative approach and procedure.
796 Miller et al.
Because the dental radiograph is the primary means
by which the surgeon can adequately visualize the
impacted tooth, care should be exercised in selecting,
exposing, and processing the appropriate film.
The panoramic radiograph is the film most oral and
maxillofacial surgeons use to view impacted mandib-
ular third molars; however, it is lacking in diagnostic
information with respect to specific anatomic rela-
tionships and planning treatment of the difficult im-
paction. Panoramic radiographs cannot be used to
measure accurately the amount of bone overlying an
impacted tooth or to preplan with certainty the buc-
colingual or inferosuperior relationships of the im-
pacted molar and the mandibular cana1.29 By supple-
menting an additional view, a more precise demon-
stration of the actual anatomic relationships can be
achieved.30
Various radiographic techniques are available that
can be used to localize the mandibular canal and aid
the surgeon in the removal of impacted mandibular
third molars. These include periapical, occlusal, lat-
eral-oblique mandible, and panoramic radiographs,
together with specialized localization techniques such
as Clarks use31 of the horizontal tube shift, Dono-
vans use32 of the periapical film to record an occlusal
view, Millers right angle localization technique,33
which involves a periapical and occlusal film, and Ri-
chards vertical x-ray tube shift technique.34 Limita-
tions in these techniques have resulted in the develop-
ment of this tomographic method, which accurately
displays impacted mandibular third molars and the
mandibular canal in cross section on a single film.
In evaluating a new radiographic technique, one
needs to compare the radiation exposure with ac-
cepted techniques. Current localization techniques
use two or three direct exposure films such as intraoral
periapical films and occlusal films. The skin exposure
of one D-speed rectangular collimated intraoral film
is approximately 1.80 mGy (180 mrad), 1.O mGy
(100 mrad) for E-speed film, and 2.9 mGy (290 mrad)
for an occlusal film.35 The resultant skin exposure of
these localization techniques ranges from 2.0 mGy
(200 mrad) to 4.7 mGy (470 mrad). The average sur-
face dose of a single cross-sectional tomogram ranges
from 0.65 mGy (65 mrad) to 1.3 mGy (130 mrad).
The estimated skin dose for three exposures would
range from 1.95 mGy (195 mrad) to 3.90 mGy (390
mRad), which correlates well with the surface radi-
ation dose of two direct exposures with D- or E-speed
film.
The cross-sectional tomograms obtained in this
study revealed the mean +(SD) dimensions of the
mandibular canal to be 2.6 +- 0.6 X 2.2 it 0.5 mm,
which is slightly larger than the anatomic findings re-
ported by Rajchel. The discrepancy in the reported
dimensions may be due to differences in the racial
ORAL SURC ORAL MED ORAL PATHOL
December 1990
population groups selected for study or due to differ-
ences in measuring anatomic structures on radio-
graphs as opposed to dry skull specimens. The man-
dibular canal was widest in the vertical plane, which
corresponds with the quantum of medullary bone
available to conduct the neurovascular bundle be-
tween the impacted tooth and the mandibular corti-
cal plates. Compression of the inferior alveolar bun-
dle between the impacted molar and the cortical
plates reduced the size of the canal and often pro-
duced a groove in mandibular cortical plate and/or
the impacted mandibular third molar.
The buccolingual location of the mandibular canal
was variable in the cross-sectional tomograms. Al-
though it appeared more frequently on the buccal as-
pect of the impacted mandibular third molar, the an-
gle and depth of the impaction, the width of the man-
dible, and the location of the impacted tooth roots
were all factors in the canals anatomic position.
There was no relationship between cortication of
the mandibular canal on the panoramic radiographs
and cortication on the tomograms. The proximity of
the nerve to the mandibular cortical plate, however,
was correlated with panoramic cortication, and sev-
eral canals completely confined to the medullary bone
in cross section demonstrated panoramic cortication.
The lack of cortication demonstrated by the mandib-
ular canal in cross-sectional tomograms, despite its
cortication in panoramic radiographs raises the fol-
lowing question: Why do cross-sectional tomograms
show cortication of the mandibular canal less fre-
quently than panoramic radiographs? One answer
may relate to the tomographic exposures used, which
may have burned out the cortical margin. This
suggests that additional filtration may have been use-
ful in select cases. A second possibility could be that
tomograms lack sufficient clarity to resolve cortica-
tion. This is doubtful because some cross-sectional
tomograms (38.7%) did show either partial or com-
plete cortication of the mandibular canal. The third
and most likely possibility is that the radiographic
appearance of cortication of the mandibular canal is
dependent on three factors: (1) the geometry of the
radiographic projection used, (2) the proximity of the
canal to the cortical bone in the projected plane, and
(3) the variation in radiodensity (compactness and
thickness) of the closely apposed anatomic struc-
tures.36 Because the panoramic radiation beam is ap-
proximately perpendicular to the radiolucent inferior
alveolar neurovascular bundle and the radiodense
cortical plate, a smooth radiodense delineation or
cortication is produced on the film at the margin of the
canal, if the two structures are close together. In
cross-sectional tomography, however, the radiolucent
canal and the radiodense cortical plate are parallel in
the projected geometric plane, thus radiographic vi-
Volume 70
Number 6
sualization of the corticated margin of the canal is less
apparent in this view.
CONCLUSION
Surgical removal of impacted mandibular third
molars may result in vascular and neurologic compli-
cations. To prevent these complications and to provide
quality care, a well-planned surgical approach should
include a proper patient historical and clinical evalu-
ation, and selected dental radiographs that localize
anatomic structures involved in the surgical field.
Views that provide precise localization of the im-
pacted mandibular third molar and inferior alveolar
neurovascular bundle and provide minimal risk of ex-
posure to ionizing radiation should be selected.
Cross-sectional tomograms provide accurate infor-
mation regarding the size and shape of the mandib-
ular canal, the inclination of the impacted mandibu-
lar third molar, and the distance and location of the
mandibular canal to the impacted tooth in the bucco-
lingual and inferosuperior dimension. Diagnostic fea-
tures concerning the depth of bone, width of bone,
canal bifurcations, buccolingual cortical expansion,
and odontogenic and other pathoses can also be visu-
alized. At comparable radiation dosages by currently
available localization techniques, this information
enables the oral and maxillofacial surgeon to better
inform the patient of the surgical risks and possible
complications and to prevent or minimize damage to
the inferior alveolar neurovascular bundle during
mandibular third-molar surgery.
REFERENCES
1. Manpower survey of oral surgery. Part 1. J Oral Surg 1977;35
(special issue A):A55-A95.
2. Rose F. Paralysis of sensory nerves from extraction of a wis-
dom tooth. Dent Ret 188l;l:lOO.
3. Sewill H. Case of paralysis of the parts supplied by the left in-
ferior dental nerve. J Br Dent Assoc 1881;2:173.
4. Gabell DP. Sensory paralysis of the face due to extraction of
a wisdom tooth. Dent Ret 1902;22:565-6.
5. Robinson TO. Paresthesia: a postoperative complication fol-
lowing removal of the mandibular third molar. Am J Orthod
1940;26:278-84.
6. DeBats ML, Phillips WH. Nerve injuries incidental to dental
surgery. J Oral Surg 1944;2:13-8.
7. Beauchamp WS. Severance of mandibular nerve during ex-
traction. Br Dent J 1947;83:259-60.
8. Deighan W Jr. Anesthesia and paresthesia of lower lip follow-
ing oral surgical procedures. Tufts Dental Outlook 1947;21:
4-10.
9. Simpson HE. Injuries to the inferior dental and mental nerves.
J Oral Surg 1958;16:300-5.
10. Frank VH. Paresthesia: evaluation of 16 cases. J Oral Sure
1959;17:27-33.
11. Verunac JJ, Lindsay JS. Treatment of persistent paresthesia
after third molar odontectomy: report of case. J Am Dent As-
sot 1971;83:364-6.
12. Van Go01 AV, Ten Bosch JJ, Boering G. Clinical consequences
of complaints and complications after removal of the mandib-
ular third molar. Int J Oral Surg 1977;6:29-37.
13. Osborn TP, Frederickson G, Small IA, Torgerson TS. A pro-
Cross-sectional tomography 797
spective study of complications related to mandibular third
molar surgery. J Oral Maxillofac Surg 1985;43:767-9.
14. Schwartz LJ. Lingual anesthesia following mandibular odon-
tectomy. J Oral Surg 1973;31:918-20.
15. Kipp DP, Goldstein BH, Weiss WW. Dysesthesia after man-
dibular third molar surgery: a retrospective study and analysis
of 1377 surgical procedures. J Am Dent Assoc 198O;lOO:
185-92.
16. Howe GL, Poyton HG. Prevention of damage ot the inferior
dental nerve during extraction of mandibular third molars. Br
Dent J 1960;109:355-63.
17. Hochwald DA, Davis WH, Mortinoff JM. Modified distoligual
splitting technique for removal of impacted mandibular third
molars: incidence of postoperative sequela. Oral Surg 1983;
56:9-11.
18. Alling CC. Dysesthesia of the lingual and inferior alveolar
nerves following third molar surgery. J Oral Maxillofac Surg
1986;44:454-7.
19. Wofford DT, Miller RI. Prospective study of dysesthesia fol-
lowing odontectomy of impacted mandibular third molars. J
Oral Maxillofac Surg 1987;45:15-9.
20. Bruce RA. Frederickson GC, Small GS. Age of patients and
morbidity associated with mandibular third molar surgery. J
Am Dent Assoc 1980;101:240-5.
21. Rud J. Third molar surgery: relationship of root to mandibu-
lar canal and injuries to inferior dental nerve. Dan Dent J
1983;87:619-31.
22. Sisk AL, Hammer WB, Shelton DW, Joy ED Jr. Complica-
tions following removal of impacted third molars: the role of
the experience of the surgeon. J Oral Maxillofac Surg 1986;
44:855-9.
23. Starkie C, Stewart D. The intramandibular course of the in-
ferior dental nerve. J Anat 1931;65:319-23.
24. Carter RB, Keen EN. The intramandibular course of the in-
ferior alveolar nerve. J Anat 1971;108:433-40.
25. Niirtje CJ, Farman AG, Grotepass FW. Variations in the nor-
mal anatomy of the inferior dental canal: a retrospective study
of panoramic radiographs from 3612 routine dental patients.
Br J Oral Surg 1977;15:55-63.
26. Davis WM. Relationship of posterior mandibular teeth to
mandibular alveolar canal. Am J Orthod 1938;24:382-92.
27. Stockdale R. Relationship of the roots of mandibular third
molars to the inferior dental canal. ORAL SURG ORAL MED
ORAL PATHOL 1959;12:1061-72.
28. Rajchel J, Ellis E, Fonseca RJ. The anatomical location of the
mandibular canal: its relationship to the sagittal ramus osteot-
omy. Int J Adult Orthod Orthognath Surg 1986;1:37-47.
29. Chandler LP. Laskin DM. Accuracv of radioeraohs in classi-
fication of impacted third molar teeth. J Oral Maxillofac Surg
1988;46:656-60.
30. Westesson PL, Carlson LE. Anatomy of mandibular third
molars. ORAL SURG ORAL MED ORAL PATHOL 1980;49:90-4.
31. Clark CA. A method of ascertaining the relative position of
unerupted teeth by means of film radiographs. Proc R Sot Med
(Odont Set) 1910;3:87-90.
32. Donovan MH. Occlusal radiography of the mandibular third
molar. Dent Radiogr Photogr 1952;25:53-5.
33. Richards AG. Roentgenographic localization of the mandib-
ular canal. J Oral Surg 1952;10:325-9.
34. Richards AG. A technic for the roentgenographic examination
of impacted mandibular third molars. J Oral Surg 1952;10:
138-41.
35. Gibbs SJ, Pujol A, McDavid WD, Welander U, Tronje G. Pa-
tient risk from rotational panoramic radiography. Dentomax-
illofac Radio1 1988;17:25-32.
36. Bender IB. Factors influencing the radiographic appearance of
bony lesions. J Endodont 1982:8:161-70.
Reprint requests to:
Dr. Craig S. Miller
Department of Oral Health Science
University of Kentucky College of Dentistry
MN 288 Medical Center
Lexington, KY 40536-0084