The influence of the ttigeminal neme on

facial growth and development Dr. Behrents

Ft. G. Behrents, D.D.S., MS.,* and L. E. Johnston, Jr., D.D.S., Ph.D.**

Cleveland, Ohio, and St. Louis, MO.

For the past twenty years it has been argued that the growth of the face is governed by the growth of a variety of
contiguous soft tissues and functioning spaces (functional matrices). Recently, extensive data have been
marshalled to suggest that the growth of the functional matrices is controlled by “trophic” influences from nerves.
Accordingly, the present study sought to evaluate the role of the trigeminal nerve in the regulation of facial growth
by placing unilateral stereotaxic lesions in the trigeminal nerve (root, ganglion, or major sensory branch) of the
lo- to 20-day-old rat. The unoperated side and a number of sham animals served as controls. In all, 323
operations were performed, with 119 animals surviving until the age of 50 days. The location and extent of the
lesions were controlled histologically. Cephalometric analysis revealed subtle but statisGcally significant distortions
of the facial skeleton; however, the magnitude of the effect was quite small. Thus, although a modest case can
be made for the contention that the trigeminal nerve exerts a trophic influence on the craniofacial complex, the
conclusion that trophic effects constitute a major factor in the regulation of craniofacial growth cannot be
supported.

Key words: Neurotrophism, craniofacial, growth, nerves, skeleton

A s with most scientific disciplines, progress

in craniofacial biology is built on a solid descriptive
foundation. From work with implants, vital dyes, ra-
dioactive tracers, and the like, contemporary inves-
tigators know, at least to a first approximation, when,
where, and how the bones of the head grow; unfortu-
nately, they do not yet know why. Indeed, contempo-
rary hypotheses concerning the regulation of transfor-
mative and translatory bone growth involve, for the
most part, code words borrowed from other fields:
genes, “function, ” cyclic AMP, and bioelectric po-
tentials, to name but a few.
Currently, one of the more obscure phenomena in-
voked to “explain” facial growth is neurotrophism a growth is now capable of being conceived of as a
term which, freely translated, means “nourishment
from the nervous system. ” The nature of this “nour-
ishment” has, for several decades, been the object of
intensive study in a number of laboratories throughout
the world. Currently, it is known that a wide variety of
structures (for example, taste buds, epithelia, salivary
glands) require innervation for their normal formation,
This article is based on research submitted to the Graduate School of Case
Western Reserve University in partial fulfillment of the requirements for the
Master of Science degree, Department of Orthodontics, 1975. Recipient of the
Milo Hellman Research Award of the American Association of Orthodontists.
*Associate Rofessor and Director of Clinics, Department of Orthodontics,
Case Western Reserve University, Cleveland, Ohio.
**Professor and Chairman, Department of Orthodontics, St. Louis University.
growth, and maintenance. It is thought that this effect is
mediated by one or more substances produced by indi-
vidual neurons and transported to the periphery by
axoplasmic flow .*
Nor are neurotrophic effects limited to isolated tis-
sues and organs. It has been shown that a common
growth model, limb regeneration, is also nerve-
dependent .3 In fact, by augmenting the peripheral in-
nervation, regeneration can be achieved in such nor-
mally nonregenerating animals as the frog4 and the
opossum.5
Recently, MOSS+~has suggested that facial growth
may also be neurotrophically regulated: ‘ ‘Oro-facial
homeostatically controlled series of processes in which
the neural center regulates the peripheral tissues and the
periphery, in turn, regulates the center.‘17
Certainly, from the standpoint of the various ver-
sions of the functional matrix hypothesis, the well-
documented relationship between nerves and practi-
cally all of the tissues that constitute the ‘ ‘periosteal”
and “capsular” matrices makes this a provocative,
even plausible, conjecture. In addition to an extensive
literature, the neurotrophic hypothesis has yet another
virtue-it is specific enough to generate predictions
*For an extensive review of neruotmphism mechanisms and effects, see Beh-
rents,’ 1975, and Behrents,* 1976.

199
200 Behrents and Johnston
Fig. 1. Dorsal view of the right trigeminal
histoloaicallv after the exoerimental period.
the first-order neurons within the ganglion.
that can be tested experimentally. Specifically, in a
paper entitled “Neurotrophic Regulation of Craniofa-
cial Growth, ’ ‘g Moss has stated (p. 41, his italics):
“Now if such epithelial growth is dependent upon ap-
propriate and operational afferent neuroepithelial tro-
phic processes, then if such process were absent or
deficient, we should reasonably anticipate oro-facial
hypoplasia, malgrowth, or possible malformation. This
is precisely what we find. ”
It is the purpose of the present study to examine the
impact of denervation on facial development in the
growing rat and to evaluate the outcome in light of
contemporary neurobiologic thought.
MATERIALS AND METHODS
A total of 382 Long-Evans rats from forty-seven
litters were used-323 to examine the effects of uni-
lateral section of the trigeminal nerve and 33 to test the
effects of bilateral destruction.
Each animal underwent surgery when it was be-
tween 10 and 20 days of age. In order to minimize
extraneous damage to the various “functional ma-
trices, ’ ’ the nerves were approached intracranially via
standard stereotaxic techniques. The nerves were
coagulated with radiofrequency electrical current (60 to
Am. J. Orthod.
March 1984
nerve demonstrating lesion locations which were verified
The stippled area denotes the location of the ceil bodies of
80 mA for 30 to 60 seconds) produced by a standard
lesion generator* and delivered through varnished
0.020-inch stainless steel electrodes. For each animal,
trigeminal lesions were placed in one of three locations:
in the root, in the semilunar ganglion, and in the major
divisions anterior to the ganglion (Fig. 1).
The animals were kept until they were 50 days old,
at which time a total of 119 had survived. All animals
with bilateral lesions died. The survivors were exam-
ined for gross facial asymmetry, loss of the blink re-
flex, presence and distribution of so-called “trophic”
ulcers, and sensory loss according to the method of
DeLahunta.g The rats were then killed and prepared for
gross, roentgenographic, photographic, and histologic
examination.
Dorsoventral cephalograms were obtained with a
specially designed cephalostat featuring a source-film
distance of 60 inches. The cephalometric analysis con-
sisted of a combination of direct and projected dimen-
sions (Fig. 2). All measurements were taken directly
from punched films and were executed to the nearest
0.05 mm. Statistical analysis was performed on the
difference between sides (lesioned and intact) for all
‘Grass LM-4, Grass Instrument Company. Quincy, Mass.
volume 85 influence of trigeminal nerve on facial growth and development 201
Number 3

Na

Fig. 2. Cephalometric analysis (all measurements were taken bilaterally). A, Orientation utilizing a
machine porion-midline axis. 6, Orientation utilizing the nasal septum. C, Orientation utilizing a man-
dibular orientation. D, Direct measures between anatomic landmarks.

202 Behrents and Johnston
Fig. 3. Dry skull and radiograph of a rat with a lesion
strated one of the more dramatic effects seen.
measures. Given the working hypothesis of this study,
the differences were computed in such a way that
asymmetry (“hypoplasia, malgrowth or possible mal-
formation”) consistent with a loss of trophic substances
would be signaled by a negative sign.
Based on Bodian-stained, 10 PM paraffin sections,
the location of each lesion was verified and these data
were then employed to group the animals for statistical
analysis: (1) anterior lesions-53 animals; (2) gan-
glionic lesions-48 animals; (3) posterior lesions- 12
animals; (4) sham lesions [6] and unoperated controls
[26]-32 animals (“control”).
Differences among these four categories were
evaluated by analysis of variance and, for purposes of
discussion, a single supplemental contrast (Ho : anterior
and ganglion = posterior and control) was performed
using the t statistic. In the presence of significant F
ratios, this test was designed to contrast the two groups
with axosomatic continuity with those whose cell
Am. J. Orthod.
March 1984
of the left trigeminal nerve. This animal demon-
bodies were either destroyed or separated from their
peripheral neurites. Theoretically, the former should
show neurotrophic effects, while the latter should
not. All tests were conducted at the 95% level of
confidence.
RESULTS
At the end of the experimental period, fifty-nine of
the 119 surviving experimental animals were free of
any obvious pathoses. The other sixty presented a va-
riety of cutaneous lesions ranging in severity from
small, isolated sores to large, bleeding ulcers. In addi-
tion, many of these animals showed cornea1 dryness,
opacity, and occasionally even complete orbital exen-
teration. All cutaneous and ocular lesions were con-
fined to the ipsilateral trigeminal field and were accom-
panied by varying degrees of sensory loss. Moreover,
twenty-five of these sixty showed such signs of motor
deficits as impaired attrition, overeruption, and mal-
Volume 85 Injfuence of trigeminal nerve on facial growth and development 203
Number 3

Table I. Descriptive and inferential statistics for histologic groupings (in millimeters)
Mean difference between conrrol 2 test
and experimental sides Ho: (Anterior and
Measure ANOVA ganglion) =
(see Fig. 2) Characteristic Anterior Ganglion Posterior Control F (Posterior and control)

Perpendicular distance from machine transporionic axis

1 Maxilla -0.06 -0.14 -0.08 0.02 1.82 -
2 Molar -0.17 -0.07 0.01 0.04 5.85** 3.63**
3 Anterior zygomatic arch -0.06 -0.14 -0.06 0.02 2.91* 2.19*
4 Posterior zygomatic arch 0.01 -0.19 0.03 -0.03 4.29** -1.43
Perpendicular distance from machine midline
5 Molar width -0.14 -0.47 -0.18 0.05 4.02* 2.49*
6 Palatal width 0.07 -0.43 -0.26 -0.08 7.39** 0.47
I Maxillary width -0.08 -0.67 -0.55 0.02 7.53** 2.05*
8 Zygomatic width -0.07 -0.30 -0.26 0.03 3.52* 1.70
9 Snout deviation -0.24 -0.54 -0.26 -0.06 3.39* 2.43*
IO Cranial base width -0.03 -0.25 -0.13 -0.01 0.87
11 Cranial base div. 0.02 -0.13 -0.10 -0.02 2.40 -
12 Mandibular midline -0.21 -0.31 -0.61 -0.01 3.63* 0.90
Perpendicular distance from septal midline
13 Mandibular midline -0.19 -0.02 -0.30 0.01 2.19
14 Maxillary width -0.04 -0.02 0.14 -0.04 0.95
15 Molar width 0.02 0.06 0.17 -0.05 1.13
16 Palatal width -0.01 0.01 0.18 -0.01 I .73
17 Zygomatic width 0.08 -0.10 0.10 -0.10 4.71** 0.71
18 Cranial base 0.26 -0.17 -0.11 0.03 0.91
(petrous) deviation
Perpendicular distance from mandibular midline
19 Condylar width 0.14 -0.05 0.27 -0.05 1.76
20 Ramal width 0.21 -0.03 0.33 0.04 2.21
21 Coronoid width 0.00 -0.08 0.17 -0.03 0.61
Anatomic measures
22 Maxillary div. -0.15 -0.06 0.06 -0.07 I .05
23 Anterior zygomatic root width -0.03 -0.10 0.03 0.03 2.43 -
24 Zygomatic length -0.07 -0.05 -0.13 0.04 1.06
25 Zygomatic advance -0.01 -0.26 0.06 -0.03 6.04** 1.91
26 Mandibular length -0.07 -0.12 -0.33 -0.01 2.49
21 Mandibular length 0.05 -0.05 -0.10 -0.03 0.67
28 Mandibular length (coronoid) -0.13 -0.09 -0.20 0.10 2.44

*Significant at 95% level.

**Significant at 99% level

position of the incisors. For these animals, body weight
was only about half that of their littermates. In addi-
tion, four animals had a generalized milky-white dis-
coloration of the ipsilateral incisors.
Although visual examination of dried skulls re-
vealed no obvious skeletal deformities (Fig. 3), a vari-
ety of subtle changes were apparent from an analysis
of the cephalometric data. The descriptive and infer-
ential statistics are summarized in Table I. For empha-
sis, significant (P < 0.05) mean deviations from zero
(asymmetries) are denoted, either by italics for changes
that seem consistent with the neurotrophic hypothesis
or by bold type for those that apparently contradict it.
Obviously, such a large number of tests inflates the
experimentwise type I error; however, they are not a
part of the formal inferential analysis and are intended
only to direct attention to those dimensions which may
show a trophic effect.
DISCUSSION
The purpose of this investigation was to charac-
terize, at least to a first approximation, the role of
nerves in facial development. On the basis of Moss’s
recent suggestions, 6~8 it was assumed that unilateral
destruction of the trigeminal nerve would lead to a
marked ipsilateral reduction in the titer of “trophic
substances” and thereby produce a ponderable impact
on the growth of the facial skeleton.
Given both the histologic evidence of extensive
nerve damage and the dramatic clinical appearance of
204 Behrents and Johnston Am. J. Orthod.
March 1984

Table II. Descriptive and inferential statistics for clinical groupings (in millimeters)
Mean dt@erence between control
and experimental sides
Measure ANOVA t test
(see Fig. 2) Characteristic Motor and sensory Sensory Asymptomatic Control F Ho: Sensory = Control

Perpendicular distance from machine transporionic axis

1 (As in Table I) 0.07 -0.16 -0.12 0.01 4.27** -2.82**
2 -0.13 -0.15 -0.07 0.04 3.50* -3.16**
3 0.02 -0.14 -0.11 0.03 4.16** 3.23**
4 0.08 -0.07 -0.I.5 0.01 4.03** 0.99
Perpendicular distance from machine midline
5 (As in Table I) -0.30 -0.19 -0.30 0.05 1.68
6 -0.21 -0.02 -0.26 -0.07 1.48 -
7 -0.24 -0.33 -0.44 0.04 2.21
8 -0.09 -0.14 -0.24 0.03 1.88 -
9 -0.04 -0.56 -0.36 -0.04 4.59** 3.51**
10 -0.08 -0.07 0.01 -0.02 1.37 -
11 -0.16 0.02 -0.05 -0.01 2.00 -
12 -0.72 -0.27 -0.10 0.00 9.45** 2.65**
Perpendicular distance from septal midline
13 (As in Table I) -0.62 -0.10 0.07 0.01 14.95** 1.76
14 0.14 -0.09 -0.03 -0.03 2.39 -
15 0.37 -0.06 -0.03 -0.04 9.42** 0.36
16 0.19 -0.01 -0.04 0.08 4.83** -0.31
17 0.07 0.07 -0.08 -0.08 2.78* 1.87
18 0.17 -0.65 -0.07 0.01 7.43** 3.86**
Perpendicular distance from mandibular midline
19 (As in Table I) -0.34 0.16 -0.10 -0.08 4.29** 1.92
20 0.42 0.23 -0.06 0.02 5.25** 1.84
21 0.14 0.00 -0.06 -0.08 0.78 -
Anatomic measures
22 (As in Table I) 0.09 -0.24 -0.07 -0.06 3.64* -2.21*
23 0.00 -0.06 -0.05 0.02 1 .oo
24 -0.03 -0.14 -0.02 0.00 1.17
2.5 -0.07 -0.12 -0.12 -0.02 0.62
26 -0.30 -0.11 -0.02 -0.04 4.21** -0.87
27 -0.20 0.05 0.05 -0.07 2.64* 1.43
28 -0.24 -0.12 -0.06 0.11 2.83* 2.83*
Number 25 35 53 32

many of the animals, it would seem that the stereotaxic
surgery achieved the general goals of the experimental
design. As judged by the F ratios, a variety of statisti-
cally significant asymmetries were produced. In gen-
eral, these asymmetries were associated with machine
coordinates and were most pronounced in those animals
with ganglionic lesions. The supplemental contrasts,
however, did not show a consistent relationship be-
tween facial asymmetry and the assumed trophic func-
tion of nerves; in only five of the twenty-eight dimen-
sions was there a significant difference between groups
that might be assumed to have an intact trophic mech-
anism (posterior and control) and those that did not
(anterior and ganglion). Moreover, the changes seen
here must be interpreted with care: for all dimensions,
they were quite small and were often seen in conjunc-
tion with a variety of dramatic stigmata (for example,
complete orbital exenteration) that might, in and of
themselves, be expected to produce changes in the un-
derlying facial skeleton.
For example, ulcers of the type seen here, although
once thought to have a “trophic” origin, are now most
often interpreted as a self-inflicted response to profound
anesthesia. More to the point, these anesthetic excoria-
tions were the most prominent result of the stereotaxic
surgery and could well have led to some of the
cephalometric changes, as could the motor impairment
suffered by at least twenty-five of the animals. In any
event, it would seem that any changes resulting from a
loss of neurotrophic substances would be confounded
with the effects of anesthesia and dysfunction. Despite
these obvious problems, it is possible to make a modest
case for a neutrophic influence on facial growth.
It may be recalled that both efference and afference
Volume 85 Irtjluence of trigeminal
Number 3
require intact neurons and functional central connec-
tions. By way of contrast, trophic effects presumably
need only a cell body and a peripherally directed neu-
rite (Fig. 1). Retrogasserian section, therefore, would be
expected to produce a conduction loss and, at the same
time, leave a portion of the trophic mechanism intact
(assuming, of course, that a significant proportion of
the primary cell bodies escape retrograde degenera-
tion). Conversely, ganglionic lesions (and, to a lesser
extent, anterior lesions) would be expected to eliminate
all neural functions. Thus, if trophic substances be im-
portant in facial development, ganglionic lesions, be-
cause they produce both conduction deficits and a loss
of neurotrophism, should cause a consistently greater
experimental effect than retrogasserian lesions. It may
be seen from Table I that, in the majority of the
twenty-eight skeletal dimensions, ganglionic lesions
did, in fact, produce a greater distortion; however, the
difference proved statistically significant for only three
(measures 4, 17, and 25; Ho: ganglionic 5 posterior),
all of which involve the zygomatic arch (and, perhaps,
the temporalis and masseter muscles). Although this
after-the-fact analysis may suggest a limited neuro-
trophic influence, it hardly seems indicative of a major
regulator of the totality of facial skeletal growth.
In an additional post hoc attempt to isolate and
identify neurotrophic effects, the data were regrouped
according to clinical appearance (Table II). As with the
histologic groupings, there were a number of small but
statistically significant asymmetries. In this scheme,
one category of animals ( “asymptomatic ’ ‘) was free of
clinical signs, even though each rat had suffered a his-
tologically verified trigeminal lesion. Obviously, for
these fifty-three animals, skeletal deformation could
hardly be ascribed to ulcers, inflammation, or paralysis
and might, therefore, be indicative of some sort of
‘ ‘aneurotrophia. ” As a result, it is interesting to note
that for this group, eleven of twenty-eight measures of
skeletal symmetry differed significantly from zero (t
test; Ho:deviation = 0)-the three zygomatic arch
contrasts that proved significant in the histologic group-
ings and eight additional measures that describe the
position of the head relative to the sag&al and trans-
porionic planes of the head holder. It would seem,
then, that although a reduction in the titer of trophic
substances produces little change in the absolute size of
any given anatomic dimension (for example, measures
13 to 18), it may possibly produce a variety of subtle
distortions, the sum of which can be detected only with
reference to planes defined by the head holder. It might
also be argued that the significant differences noted for
the histologic grouping were the result of functional
disturbances resulting from destruction of efferent fi-
nerve on facial growth and development 205
bers along with the sensory fibers. On the basis of t
tests-planned a priori-it is apparent that those ani-
mals with sensory defects occasionally differed from
the controls and to some extent exhibited effects differ-
ent from those seen in animals with overt functional
disturbances.
In the present study, only about one third of the
sample survived. As a result, it could be argued that a
major trophic effect was confined, by definition, to the
263 animals that died prematurely. In order to examine
this possibility, the skulls of these animals were recov-
ered, cleaned, and compared with a series of normal
specimens. No obvious abnormalities or asymmetries
were seen. Indeed, a majority of the deaths seemed to
be related to brain edema or to feeding difficulties
rather than to some sort of facial “hypoplasia” or
“malgrowth. ”
In view of the traditional assumptions concerning
the relationship between muscle pull and bone and,
more recently, the suggestion that nerves per se regu-
late skeletal growth, the results of the present study
may seem surprising. It should be noted, however, that
a number of workers have begun to question the pro-
priety of the so-called “pressure-tension” hypothesis
and have argued that the osseous changes produced by
muscle resection could well be an incidental by-product
of a loss of blood supply. (See Boyd, Heulke and Cas-
telli.‘O) In response to this argument, Moss” has re-
ported that the results of experiments such as those of
Washburn could be duplicated “by leaving the tem-
poral muscle intact and by cutting the branch of the
third division of the fifth nerve which supplies it. ” This
claim apparently has not yet been documented and
could not be substantiated here. Despite the fact that we
can lend only minimal support to much that is written
about the role of nerves and muscles, our results agree
quite well with those of several recent studies12-‘4 (al-
though the interpretations may differ14).
Pimenidis and Gianelly12 placed unilateral elec-
trolytic lesions in the trigeminal ganglia of sixteen al-
bino rats. Although the surgery ‘ ‘destroyed most of the
mandibular sensory neurons,” the resulting changes
were minimal (p. 95, their italics): “There was ~to
change in the length and height of the mandible. How-
ever, the shape of the mandible of most experimental
animals appeared different in certain respects. ”
They also reported a kind of functional Class III
malocclusion, a result seen on occasion in our material
but not examined in detail.
Similarly, Sarnat and associates13have reported that
unilateral section of the trigeminal motor root in four
adult monkeys failed to produce a significant change in
mandibular morphology.
206 Behrents und Johnston Am. J. Orrhod.
March 1984

On balance, these recent findings, as well as those Mizell M, Isaacs JJ: Induced regeneration of hindlimbs in the
of the present study, would seem to argue for a more newborn opossum. Am Zoologist 10: 141-155, 1970.
Moss ML: Neurotrophic processes in orofacial growth. J Dent
circumspect attitude toward the role of nerves and Res SO: 1492-1494, 1971.
muscles in facial development. Although the early de- Moss ML: An introduction to the neurobiology of orofacial
struction of the trigeminal nerve produced a host of growth. Acta Biotheor 21: 236-259, 1972.
interesting effects, some of which may be trophic in 8. Moss ML: Neurotrophic regulation of craniofacial growth. In
nature, the size and distribution of the skeletal sequelae McNamara JW Jr (editor): Control mechanisms in craniofacial
growth, Monograph No. 6, Craniofacial Growth Series, Ann
suggest only a limited regulatory role for trophic sub- Arbor, 1975, Center for Human Growth and Development, Uni-
stances. Obviously, a more definitive characterization versity of Michigan.
probably will require the isolation, identification, and 9. DeLahunta A: Small animal neurological examination. Vet Clin
purification of the chemicals responsible for the effects North Am 1: 191-206, 1971.
IO. Boyd TG, Castelli WA, Huelke DF: Removal of the temporalis
seen in other systems. In the meantime, craniofacial
muscle from its origin: effects on the size and shape of the
biologists would probably find it more profitable to coronoid process. J Dent Res 46: 997-1001, 1967.
look elsewhere for the major regulators of skeletal 11. Moss ML: Ontogenetic aspects of craniofacial growth. In
growth. Moyers RE, Krogman Wm (editors): Craniofacial growth in
man, New York, 1971, Pergamon Press, p. 120.
The authors would like to thank Dr. Joel Wisotzky and 12. Pimenidis MZ, Gianelly AA: Class III malocclusion produced
Dr. Marcus Singer for their contributions to this work. by oral facial sensory deprivation in the rat. AM J ORTHOD 71:
94-102, 1977.
REFERENCES 13. Samat BG, Feigenbaum JA, Krogman WM: Adult monkey
1. Behrents RG: The influence of the trigeminal nerve on facial coronoid process after resection of trigeminal nerve motor root.
growth and development. Master’s Thesis, Department of Ortho- Am J Anat 150: 129-137, 1977.
dontics, Case Western Reserve University, Cleveland, 1975. 14. Gardner DE, Luschei ES, Joondeph DR: Alterations in the facial
2. Behrents RG: Deja vu: Neurotrophism and the regulation of skeleton of the guinea pig following a lesion of the trigeminal
craniofacial’growth. In McNamara JW Jr (editor): Factors affect- motor nucleus. AM J ORTHOD 78: 66-80, 1980.
ing the growth of the midface, Monograph No. 6, Craniofacial Reprint requests fo:
Growth Series, Ann Arbor, 1976, Center for Human Growth and Dr. Rolf G. Behrents
Development, University of Michigan. Case Western Reserve University
3. Singer M: Nervous control of the regrowth of body parts in School of Dentistry
vertebrates. In Gutmann E, Hnik P (editors): The effect of use Department of Orthodontics
and disuse on neuromuscular functions, Amsterdam, 1963. 2123 Abington Rd.
Elsevier Publishing Company, pp. 83-94. Cleveland, OH 44106
4. Singer M: Induction of regeneration of the forelimb of the post-
metamorphic frog by augmentation of the nerve supply. J Exp
Zoo1 126: 419-472, 1954.