Annals of Human Biology

ISSN: 0301-4460 (Print) 1464-5033 (Online) Journal homepage: http://www.tandfonline.com/loi/iahb20

Third molar impaction and agenesis: influence on

anterior crowding

Temitope Esan & Lynne A. Schepartz

To cite this article: Temitope Esan & Lynne A. Schepartz (2016): Third molar impaction
and agenesis: influence on anterior crowding, Annals of Human Biology, DOI:
10.3109/03014460.2016.1151549

To link to this article: http://dx.doi.org/10.3109/03014460.2016.1151549

Accepted author version posted online: 08

Feb 2016.
Published online: 09 Mar 2016.

Submit your article to this journal

Article views: 65

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=iahb20

Download by: [The Library, University of Witwatersrand], [L A Schepartz] Date: 20 December 2016, At: 05:05
ANNALS OF HUMAN BIOLOGY, 2016
http://dx.doi.org/10.3109/03014460.2016.1151549

RESEARCH PAPER

Third molar impaction and agenesis: influence on anterior crowding

Temitope Esana and Lynne A. Schepartzb,c
a
Department of Restorative Dentistry, Faculty of Dentistry, College of Health Sciences, Obafemi Awolowo University, Ile-Ife, Osun, Nigeria;
b
School of Anatomical Sciences, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, ZA, South Africa; cUniversity of
Pennsylvania Museum of Archaeology and Anthropology, Philadelphia, PA, USA

ABSTRACT ARTICLE HISTORY

Background Third molar influence on anterior crowding is controversial, but they are assumed to play Received 10 March 2015
a major role in compromising dental arch space. Revised 7 December 2015
Aim To evaluate the relationship among impaction, agenesis and crowding in black South African Accepted 13 December 2015
males. Published online 7 March
2016
Subjects and method Mandibles and maxillae of 535 black South African males in the Raymond A.
Dart Collection of Human Skeletons, University of the Witwatersrand were examined for anterior crowd- KEYWORDS
ing and third molar agenesis and impaction. Dental crowding was determined using Little’s irregularity Morphometrics; dental
index. development; dental
Results Individuals with impaction showed more moderate-to-extreme crowding than those with agen- variability
esis. Bilateral third molar presence was more frequently associated with ideal-to-minimal crowding.
Weak positive but significant correlations between crowding and impaction were found (mandible,
q ¼ 0.154, p ¼ 0.000; maxilla q ¼ 0.130, p ¼ 0.000). The direction was the opposite for bilateral presence
of molars (mandible, q ¼ 0.135, p ¼ 0.02; maxilla, q ¼ 0.111, p ¼ 0.010). Odds of mandibular crowd-
ing were greatest in individuals with impaction (OR ¼ 3.22, CI ¼ 1.716–6.05, p < 0.001). Maxillary results
were similar.
Conclusion Third molar impaction plays a role in anterior crowding. Third molar presence was not asso-
ciated with anterior crowding, while agenesis did not explain absence of crowding.

Introduction Aetiology of late anterior crowding

Dental crowding, or a lack of space within the dental arcade,
is measured by determining the difference between the space
available for a tooth and the mesiodistal diameter of the
tooth crown (Lavelle, 1973). Dental crowding was relatively
rare among pre-historic humans (Corruccini, 1984; Rose &
Roblee, 2009); when it was present it was typically extreme
and occurred early with the emergence of the first permanent
teeth. This form of crowding is viewed as having a genetic
basis (Sheneman 1968) as the expression is often bilateral,
and familial or populational patterns are observed (Miyar
2012). A second form of dental crowding develops later in
the developmental sequence, beginning with the emergence
of the third molars. This is termed late anterior crowding. A
high prevalence of late anterior crowding is observed in
Western societies and populations undergoing urbanisation
(Buschang & Shulman, 2003). It is often difficult to distinguish
between early crowding and late anterior crowding in cadav-
eric or archeological specimens; however, early crowding in
our experience appears to have bilateral expressions of tooth
rotation or displacement that are less frequently seen in late
onset crowding. Our study focuses on this second form of
dental crowding and the possible role of third molar status
(agenesis, impaction or presence) in its development.
Many theories have been put forward as to the aetiology
of late anterior crowding. Several authors documented
secular changes in the craniofacial skeleton with concomi-
tant reduction in arch width and length (Jantz & Meadows
Jantz, 2000; Lavelle, 1973; Smith et al., 1986; Truesdell,
2005), but without corresponding decreases in tooth size
(Lavelle, 1973). This leads to tooth/jaw mismatch (tooth size
arch size discrepancy) with crowding in the anterior denti-
tion. Whether this is differentially expressed in males and
females is unclear and the condition may vary by popula-
tion. Janosević et al. (2006) and Mugonzibwa et al. (2008)
did not find any significant difference in anterior crowding
between males and females in several populations.
However, a recent review by Stanaityte_ and Gervickas
(2014) found crowding to be more frequent in females,
because males have significantly longer and wider dental
arches. Theories as to why there is a recent trend toward
lack of space in the dental arcade include differential
growth patterns of the mandibular condyle, reduced growth
of the mandible due to a softer diet and delayed eruption
or exfoliation of the deciduous teeth. None of these has
been shown to be mutually exclusive (Bjo €rk et al., 1956;
Farman, 2007; Richardson, 1977).

CONTACT Lynne A. Schepartz lynne.schepartz@wits.ac.za School of Anatomical Sciences, Faculty of Health Sciences, University of the Witwatersrand, 7 York
Road, Parktown 2193, Johannesburg, ZA, South Africa
ß 2016 Taylor & Francis
2 T. ESAN AND L. A. SCHEPARTZ
The process of mesial drift has been implicated in the aeti-
ology of anterior crowding. Mesial drift refers to dental migra-
tion within the alveolar bone toward the midline of the
arcade (Kaifu et al., 2003). This movement is a skeletal growth
process resulting from the mesial eruption of teeth, as well as
a result of the functional use of the dentition and the forces
generated during mastication. The migratory process occurs
both in juvenile and adult dentitions (Yilmaz et al., 1980;
Kaifu et al., 2003). The continuous movement of teeth toward
the midline of the dental arcade contributes to the tight con-
tacts between teeth and diminished interproximal spaces.
Pre-historic peoples, in particular hunter-gatherers, but also
earlier hominins with coarse diets, have high levels of inter-
proximal attrition (Wolpoff, 1971), but they rarely exhibit
interproximal spaces or late anterior crowding. The lack of
inter-proximal spaces in these populations is attributed to
mesial drift (Hinton, 1982; Nara et al., 1998). Movement of the
posterior teeth in particular seems to be affected, implying
that the biomechanical forces generated during the grinding
of food result in mesial drift. In good occlusion, mesial move-
ment of the teeth maintains a Class I bite between the max-
illa and mandible. However, in individuals with inadequate
dental arcade dimensions this migration leads to pressure on
the teeth that can cause anterior crowding and even impac-
tion of erupting teeth (Lombardi, 1982; Richardson 2002).
The roles of soft tissue in dental crowding have been con-
sidered in the general equilibrium theory of tooth movement,
where importance has been placed on differential mandibular
growth, but in conjunction with soft tissue forces. The
American Association of Orthodontists (1999) described this
theory as forces from the maxillary arch and labial soft tissues
causing uprighting of the mandibular incisors. This uprighting,
or lingual movement, results in crowding by forcing the inci-
sors to occupy a smaller arch perimeter. Such differential
mandibular growth is not uncommon in late adolescence and
may continue at a slow rate to adulthood. Although the dif-
ferential mandibular growth theory is supported by sound
research, the correlation found between growth changes and
incisor crowding is generally weak (American Association of
Orthodontists, 1999). Alternative theories propose that
changes in facial muscles or other soft tissues are responsible
for the altered equilibrium that causes crowding. These
explanations are plausible, but they seem to tell only part of
the story (Proffit, 1978).
The effect of skeletal pattern variations of the jaws (such
as Angle’s classification of skeletal relationships) on dental
crowding has also been studied. In randomised samples, cor-
relations between skeletal patterns and crowding have not
been noted. However, positive correlations were reported
when extreme cases of crowding and skeletal patterns were
analysed (Miethke & Behm-Menthel, 1988).
Isaacson et al. (1971) and Nasby et al. (1972) found anter-
ior crowding and anterior open bite is associated with a large
mandibular plane-sella nasion angle (MP-SN), whereas the
opposite was the case when MP-SN is small. Leighton and
Hunter (1982) compared the skeletal patterns of crowded and
uncrowded dentitions and found the crowded cases had an
increased mandibular plane-Frankfort plane angle.
Interestingly, despite the difference in the skeletal patterns of
American blacks and whites, crowding in blacks was not
found to differ in frequency from crowding in whites
(Altemus, 1959). Together, these studies suggest an overall
association between skeletal pattern and more extreme
crowding, but they also suggest that the relationship is not
simple and that many factors contribute to late anterior
crowding.
The role of third molars in anterior crowding
The varying statuses of third molars (present, impacted or
agenesis) have been argued to correlate with or cause anter-
ior crowding, but no clear relationships have been found.
One theory about compromised space in the dental arcade
space is that the erupting third molars push anterior teeth
forward and cause their crowding (Vego, 1962; Antanas &
Giedre, 2006). This assertion was further substantiated by
researchers who observed that the incidence of crowding
increases during the period of third molar eruption through
to total occlusion (Harradine et al., 1998; Richardson, 2002). A
comparative longitudinal study by Vego (1962) demonstrated
that the prevalence of late mandibular anterior crowding was
greater in individuals with third molars compared to those
with congenitally missing third molars. His conclusion was
that erupting third molars can exert a force on the neigh-
bouring teeth. Conversely, other studies found no correlation
between mandibular third molar presence and late incisor
crowding (Harper, 1994). Furthermore, extraction of third
molars when anterior crowding is already present does not
resolve the crowding. Lindqvist and Thilander (1982) found
that third molar removal on one side did not relieve anterior
crowding 3 years post-extraction. Similarly, Pirttiniemi et al.
(1996) did not find a significant change in the crowding of
the anterior teeth after extraction of the mandibular third
molars. Another contributing factor to late anterior crowding
may be third molar impaction, as it is argued that lack of
space in the arcade may be worsened by an impacted tooth
that is pushing other teeth mesially (Richardson 2002).
However, the relationship between impaction and late anter-
ior crowding has not been fully examined. This study aims to
re-evaluate the relationship between third molar presence,
impacted third molars, agenesis, and dental arch crowding
and to predict the level of influence of third molar status on
late anterior crowding.
Materials and methods
Mandibular and maxillary jaws of black South African males
in the Raymond A. Dart Collection of Human Skeletons at the
School of Anatomical Sciences, University of the
Witwatersrand, Johannesburg, South Africa were examined for
anterior crowding, presence or absence of third molars and
impacted third molars. The Dart Collection is primarily com-
posed of cadaveric specimens of known age and population
affinity.
Five hundred and thirty-five maxillae and mandibles of
black South African males born between the years 1880–1970
were selected. The age range was 19–70 years with a mean
 ANNALS OF HUMAN BIOLOGY 3

age of 43.66 6 12.00 years. Permission to use the collection using Student’s t-test. A Spearman rho correlation was done
for research was covered by the South African Human Tissues to assess the relationship between third molar bilateral pres-
Act of 1963 and subsequent legislation. ence, unilateral and bilateral agenesis, unilateral and bilateral
Dental crowding was determined using Little’s (1975) third molar impaction and anterior crowding. Binary logistic
irregularity index. This index represents the linear displace- regressions were run to predict the effect of third molar
ment of the anatomical contact points (as distinguished from impaction, third molar agenesis and bilateral presence of
the clinical contact points) of each incisor from the adjacent third molars on crowding status for each jaw. Age was
tooth’s anatomical contact point. The sum of these five dis- included in the model to determine if it was a confounder.
placements represents the relative degree of anterior irregu- The effect of collinearity was taken into account by dropping
larity. An index score of 0–0.9 mm is ideal, between 1–3.9 mm one of the variables that was collinear from each model.
is considered minimal, between 4–6.9 mm is moderate, from Statistical significance was set at p < 0.05.
7–9.9 mm is severe and a score of more than 10 mm is
extreme. The crowding index was dichotomised into two
Results
groups, with ideal to minimal (0–3.9 mm) representing no
crowding and moderate-to-extreme levels (>3.9 mm) repre- Crowding
senting crowding, as per standard orthodontic practice.
The prevalence of anterior crowding (calculated by dividing
The irregularity index was obtained by one examiner using
the number of individuals with moderate-to-severe crowding
a dental calliper calibrated to the nearest 0.01 mm. Twenty
by the total population sample) in the mandible was 15.4%,
specimens were randomly selected and re-measured on two
while it was only 2.1% in the maxilla. The mean crowding
separate occasions. The error of the measurement method
index was significantly lower (0.29 6 2.03 mm) for the maxilla
was investigated using analysis of variance. For an acceptable
compared to the mandible (1.82 6 4.56 mm) (t ¼ 7.00,
level of error, the variance ratio of the difference of the two
p ¼ 0.000).
measurements needed to be below 10% of the first set of
There was a statistically significant difference in the distri-
measurements.
bution of crowding severity. The mandible showed more
Agenesis of third molars was assessed visually by one of
severe and extreme levels of anterior crowding, while the
the authors and checked radiographically when there was
maxilla had more of the ideal level of crowding (Likelihood
uncertainty. Third molars were classified as missing when any
ratio v2 ¼ 75.434, df ¼ 4, p ¼ 0.000). The difference between
of the following conditions was observed: the tooth was
the dichotomized scores was also highly significant (v2 ¼
absent in the arch, the second molars were very close to the
57.148, df ¼ 1, p ¼ 0.000) (Table 1).
ascending rami and lacked distal contact facets, there was no
evidence of bone resorption due to extraction or pre-mortem
tooth loss and no sign of mineralisation of the crown could Agenesis
be identified on radiographs. Impaction was scored as present
or absent and the angulation of impaction was noted based Third molar agenesis occurred more frequently in the maxilla
on Winter’s (1926) classification of impaction with mesioangu- (n ¼ 49) than the mandible (n ¼ 39), although the difference
was not statistically significant (v2 ¼ 0.187, df ¼ 1, p ¼ 0.665).
lar, vertical or horizontal orientation recorded. Radiographs
This was also the case for unilateral vs bilateral expressions of
were not routinely used except in very doubtful cases. It may
agenesis (v2 ¼ 0.011, df ¼ 1, p ¼ 0.917) (Table 2).
be possible that some cases of agenesis or impaction were
incorrectly scored. However, Pedersen (1944, 1949) found that
only one tooth from their observational assessment of Impaction
Greenland Eskimos remains went undetected. This suggests
that there should be only negligible error in estimating agen- The mesioangular (n ¼ 58) and vertical (n ¼ 28) forms of
esis frequencies without radiographs if observations are made impaction were the prevailing types of third molar impaction
with caution. Moreover, impacted molars can be easily seen in the mandibles and maxillae, respectively. Impaction
detected in skeletal materials. For this study, 42 periapical occurred more frequently in the mandible (n ¼ 59) than the
radiographs comprising of 10 maxillary and 32 mandibular maxilla (n ¼ 28). However, the difference was not statistically
images of missing third molars were taken using a portable Table 1. Distribution of dental crowding by mandible and maxilla.
Aribex Nomad hand held X-Ray machine. Only one case from Mandible Maxilla
the mandible conflicted with our visual diagnosis. Crowding Range (mm) n % n %
Ideal 0–0.9 428 80.0 514 96.1
Minimal 1–3.9 25 4.7 9 1.7
Statistical analysis Moderate 4–6.9 22 4.1 4 0.7
Severe 7–9.9 19 3.6 4 0.7
Data were analysed using SPSS version 22 and Stata intercool Extreme 10 41 7.7 4 0.7
release 12. Descriptive statistics including frequencies, mean, Total 535 100.0 535 100.0
standard deviations and ranges were calculated. Differences Likelihood ratio v2 ¼ 75.434, df ¼ 4, p ¼ 0.000
Ideal-to-minimal 0–3.9 453 84.7 523 97.8
between categorical variables were tested for significance Moderate-to-extreme 3.9 82 15.3 12 2.2
using Chi-square tests, while differences between means of Total 535 100.0 535 100.0
crowding for the maxilla and the mandible were evaluated v2 ¼ 57.148, df ¼ 1, p ¼ 0.000
4 T. ESAN AND L. A. SCHEPARTZ

Table 2. Status of third molars. Table 4. Correlation analysis between status of third molar and crowding in
Mandible Maxilla both maxilla and the mandible.
Mandibular Maxillary
Third molar status n % n % Status of third molar crowding crowding
Bilateral M3 present 439 82.1 458 85.6 Bilateral third molar presence
Unilateral agenesis 23 4.3 31 5.8 q 0.135** 0.111*
Bilateral agenesis 14 2.6 18 3.4 p 0.002 0.010
Unilateral impaction 36 6.7 16 3.0 n 533 535
Bilateral impaction 21 3.9 12 2.2 Unilateral and bilateral agenesis
Agenesis and impaction 2 0.4 0 0.0 q 0.015 0.035
Total 535 100.0 535 100.0 p 0.723 0.422
(v2 ¼ 0.187, df ¼ 1, p ¼ 0.665) Mandibular vs maxillary agenesis. n 533 535
(v2 ¼ 0.011, df ¼ 1, p ¼ 0.917) Unilateral vs bilateral agenesis. Unilateral and bilateral impaction
(v2 ¼ 0.067, df ¼ 1, p ¼ 0.796) Mandibular vs maxillary impaction. q 0.154** 0.130**
(v2 ¼ 0.286, df ¼ 1, p ¼ 0.593) Unilateral vs bilateral impaction. p 0.000 0.003
n 533 535
Table 3. Third molar status by level of crowding in the mandible and maxilla. *significant at 0.05 level; **significant at 0.01 level.
Status of mandibular third molar
Unilateral and Unilateral and Table 5. Regression analysis of variables against crowding.
Bilateral bilateral bilateral Variables Odds ratio SE z p 95% CI
Crowding level presence agenesis impaction Total
Mandible
Ideal-to-minimal n 382 30 39 451 cons 0.404 0.207 1.76 0.078 0.148–1.105
% 87.02% 81.08% 68.42% 84.62% Age 1.003 0.010 0.37 0.709 0.016–0.024
Moderate-to-extreme n 57 7 18 82 Agenesis 0.500 0.255 1.35 0.175 0.183–.363
% 12.98% 18.92% 31.58% 15.38% Third molar presence 0.310 0.010 3.64 0.000 0.165–0.583
Total n 439 37 57 533*
cons 0.125 0.060 4.35 0.000 0.049–0.320
Status of mandibular third molar Age 0.004 0.010 0.37 0.709 0.984–1.024
Ideal-to-minimal n 451 48 24 523 Agenesis 1.610 0.717 1.07 0.285 0.672–3.856
% 98.47% 97.96% 85.71% 97.76% Third molar impaction 3.22 1.035 3.64 0.000 1.716–6.049
Moderate-to-extreme n 7 1 4 12 Maxilla
% 1.53% 2.04% 14.29% 2.24% cons 0.480 0.517 0.68 0.495 0.058–3.961
Total n 458 49 28 535 Age 0.972 0.025 1.10 0.273 0.924–1.023
*Two specimens were excluded from the analysis due to presence of agenesis Agenesis 0.149 0.173 1.64 0.100 0.015–1.442
on one side and impaction on the other side of the jaw. Third molar presence 0.108 0.073 3.31 0.001 0.029–0.404
2
cons 0.051 0.057 2.68 0.007 0.006–0.453
significant (v ¼ 0.067, df ¼ 1, p ¼ 0.796). There was no sig- Age 0.972 0.025 1.10 0.273 0.924–1.023
nificant difference in the frequency of unilateral or bilateral Agenesis 1.381 1.494 0.30 0.765 0.165–11.505
Third molar impaction 9.252 6.224 3.31 0.001 2.475–34.585
forms of impaction (v2 ¼ 0.286, df ¼ 1, p ¼ 0.593) (Table 2).
Third molar impaction and third molar presence, respectively, were dropped
Individuals with either unilateral or bilateral impaction of
from the models because of collinearity. Mandible: Pseudo R2 ¼ 0.0275;
third molars showed more moderate-to-extreme crowding Goodness-of-fit test ¼ 0.5735. Maxilla: Pseudo R2 ¼ 0.0947; Goodness-of-fit test
(mandible ¼ 31.58%, maxilla ¼ 14.29%) than those with any ¼ 0.4843.
form of agenesis (mandible ¼ 18.92%, maxilla ¼ 2.04%).
Bilaterally, presence of the third molars was more frequently p < 0.001). The results for the maxilla were similar. The odds
associated with ideal-to-minimal crowding in both jaws of maxillary crowding were greater in individuals with third
(Table 3). molar impaction than with agenesis (OR ¼ 9.25, CI ¼
The Spearman rho correlation analysis showed weak posi- 0.924–1.022, p < 0.05). Again, the odds of having anterior
tive but significant correlations between crowding of the
crowding were less for those with maxillary third molars pre-
anterior teeth and third molar impaction (mandible, q ¼ 0.154,
sent than those with agenesis (OR ¼ 0.108, CI ¼ 0.028–0.404,
p ¼ 0.000; maxilla, q ¼ 0.130, p ¼ 0.000). The direction was the
p < 0.05).
opposite for bilateral presence of third molars, which showed
weak negative but significant correlations with anterior
crowding (mandible, q ¼ 0.135, p ¼ 0.02; maxilla, q ¼ 0.111, Discussion
p ¼ 0.010) (Table 4).
Binary logistic regression was run to predict the effect of Late anterior dental crowding is the most prevalent type of
impaction, agenesis, and bilateral presence of third molars on malocclusion worldwide and it is increasing with urbanisation
crowding (Table 5). The analysis was age-adjusted and took and westernisation of diets (Buschang & Shulman, 2003). The
into account the collinearity found between third molar normal sequential eruption of teeth occurs whether or not
impaction and presence of third molars. The odds of man- there is adequate space to accommodate them. Inadequate
dibular anterior crowding were greater in individuals with space in the dental arch for erupting teeth (tooth size arch
impacted third molars compared with those who did not size discrepancy) is the ultimate common cause of most
have impaction (OR ¼ 3.22, CI ¼ 1.716–6.05, p < 0.001). anterior dental crowding (Alt & T€urp 1998). The cause of the
The presence of mandibular third molars was protective of reduced arch space in many contemporary populations has a
anterior crowding. The odds for crowding were much lower multi-factorial basis. Aside from the secular trends in facial
in individuals who had third molars present compared to and jaw reduction that result from the lessened masticatory
those who had agenesis (OR ¼ 0.31, CI ¼ 0.165–0.563, demands of modern diets, the third molars have been

implicated as a key factor in late anterior crowding. Our
results lend some clarification to the issue. We found that the
major determinant of the available space appears not to be
merely the presence of third molars, but most importantly
their impaction status. The available space is significantly
compromised by third molars that are impacted and this
leads to late anterior crowding. In the South African males of
this study, third molars typically erupted without any associ-
ated late anterior crowding. We also found eight individuals
with agenesis of the third molar had moderate-to-severe
crowding. Thus, presence of third molars is not the only rea-
son for anterior crowding. In some cases, compromised space
in the arch occurs even in the absence of the third molars.
The lack of space most often leads to impaction of the third
molar itself.
Crowding and impaction
Several authors (Graber & Kaineg, 1981; Richardson, 1979,
1989, 1996; Stephens, 1980) observed that anterior crowding
was found more frequently in patients with third molar
impaction than in subjects with correctly positioned third
molars. Our results are in agreement with those studies, as
we found individuals with third molar impaction showed
more severe and extreme levels of crowding as compared to
the ideal and minimal levels seen in those with third molar
bilateral presence. The effect of impaction on crowding was
greater for the maxilla than the mandible, even though, as
seen in other studies, mandibular anterior crowding was
more prevalent. The implication of our result is that, although
crowding is less frequently seen in the maxilla, when it does
occur it is significantly associated with impaction.
One explanation given for late anterior crowding is that
erupting third molars push anterior teeth forward and cause
their crowding (Lindauer et al., 2007; Vego, 1962). If this is the
case, it stands to reason that an erupting mesioangularly or
vertically impacted third molar with insufficient retromolar tri-
gone space (and potentially taking up even more arcade
space than a correctly oriented third molar) could generate
greater mesial force that would aggravate the anterior crowd-
ing. More quantitative research is needed to test this
assertion.
Conflicting clinical data on the role of third molars
Harradine et al. (1998) and Richardson (2002) linked the pres-
ence of third molars with anterior mandibular crowding in
modern populations and found that the relationship increases
over the individual lifespan. However, Harris (1997) suggested
that increased incisor crowding with age is predominately
caused by occlusal and physiological forces that alter the
arcade, regardless of third molar eruption. Bergstrom and
Jensen (1961), as well as Sheneman (1968), found that reten-
tion of third molars was associated with severe incisor crowd-
ing, which was not observed in their agenesis groups.
Similarly, Schwarze (1975) found a significantly greater for-
ward movement of first molars associated with increased
lower arch crowding in the non-extraction group compared
ANNALS OF HUMAN BIOLOGY 5
to a group with third molar germectomy. Other authors
found no significant influence of the third molars on anterior
crowding (Lindqvist and Thilander, 1982; Niedzielska, 2005;
Pirttiniemi et al., 1996; Shanley, 1962) and Bishara (1999) con-
cluded, similar to this study, that correctly positioned third
molars did not play a significant role in mandibular anterior
crowding. The differing results of these studies have led to
the current confusion over the role of third molars and the
lack of consensus over treatment. Based on our review of the
literature, there are substantial limitations to many studies
that consider the aetiology of crowding without specifically
evaluating the form of crowding (whether early or late), the
different third molar statuses, and the extent of population-
based variability in tooth size arch size discrepancies.
Despite negative associations between crowding and third
molar retention (Bjo €rk & Skieller, 1972; Little et al. 1981), it
cannot be denied that the third molar plays some role in
anterior crowding and overall tooth movement in certain
cases. Third molars appear to have the most effect on individ-
uals with tooth size arch size discrepancy and, in particular,
individuals with large tooth dimensions.
Changing levels of crowding and impaction: Implications
The overall prevalence of anterior crowding in black male
South Africans born between the years 1880–1970 was
16.07%. This prevalence is low, as is the prevalence of impac-
tion (13.27%). In general, studies of Asian, European, and US
populations have found much higher levels of crowding
(Perillo et al., 2010; Proffit et al., 1997; Soh et al., 2005) and
third molar impaction (Celikoglu et al., 2010; Kramer &
Williams, 1970; Quek et al., 2003). While the prevalence of
these conditions is substantially lower in our sample, recent
studies of 12 year old South African children showed a preva-
lence of crowding of more than 40% (Drummond, 2003;
Hlongwa & Du Plessis, 2005). We, therefore, predict that the
prevalence of impaction has increased to match this crowd-
ing rate and that levels of tooth size arch size discrepancy are
also changing in South Africa.
The presence of third molars and third molar impaction
are significant but very small contributors to anterior crowd-
ing status in our study population. A binary logistic regression
analysis was performed to determine the relationship
between crowding and third molar presence, impaction and
agenesis. Third molar impaction was found to significantly
predict the occurrence of crowding in this population,
whereas third molar presence was significantly associated
with no crowding. In contrast to other studies (Bergstrom &
Jensen, 1961; Sheneman, 1968), agenesis was not significantly
associated with the presence or absence of anterior segment
crowding. This means that absence of third molars may not
necessarily explain absence of anterior crowding. The R-
square showed that the predictor variables could only predict
2.8% and 9.5% of the changes in the outcome variable
(crowding) in the mandible and maxilla, respectively. Based
on our logistic regression result, there are other factors apart
from third molar impaction, such as tooth size and jaw
length, that may influence anterior crowding. Future studies
should identify those factors and explore the level of their
6 T. ESAN AND L. A. SCHEPARTZ
influence on late anterior crowding, along with changes to
diet and food preparation.
In the light of our results, is it justifiable to extract third
molars to prevent anterior crowding? Ades et al. (1990) and
Southard (1992) concluded that prophylactic extraction of
third molars to prevent anterior segment crowding is unwar-
ranted and unjustified. We agree that extractions of unim-
pacted third molars may not be scientifically defensible in
many populations, since other factors, apart from third molar
presence, influence late anterior crowding in the upper and
lower jaws.
Conclusions
The determinants of anterior crowding in South African black
males appear not to be merely the presence of third molars,
but also their impaction status. Presence of third molars was
not associated with anterior crowding. If the third molars
were fully erupted then there was sufficient space in the
jaws, illustrating that crowding is not a normal consequence
of their emergence in all populations. Third molar agenesis
did not fully explain absence of anterior crowding in this
study.
This research has relevance for anthropological studies of
past and present populations. First, it is important to distin-
guish between the two different forms of anterior tooth
crowding. Crowding is unusual in pre-historic populations and
early hominins that are generally characterised by heavy attri-
tional and inter-proximal wear, edge-to-edge occlusion and
infrequent tooth size arch size discrepancies. When crowding
is observed, it appears to be early in its onset and genetic in
origin. Individuals in recent populations, in contrast, exhibit
an entirely different form of crowding that is related to cra-
niofacial reductions that are in disharmony with tooth dimen-
sions. This results in third molar impaction, which is rising in
prevalence in conjunction with late onset anterior crowding.
The conflicting clinical literature on the relationship
between third molars and anterior crowding stems from a
lack of data on specific third molar status, limited attention to
population variation and confounders that are associated
with research based on clinical samples. As a result, the role
of the third molars in anterior crowding has been misunder-
stood and over-emphasised.
Acknowledgements
We thank Mr Brendon Billings, Curator in the School of Anatomical
Sciences, University of the Witwatersrand, for access to the Dart
Collection, along with Mr Daigo Shangase of the same institution.
Disclosure statement
The authors report no conflicts of interest. The authors alone are respon-
sible for the content and writing of the paper.
References
Ades AG, Joondeph DR, Little RM, Chapko MK. 1990. A long-term study of
the relationship of third molars to changes in the mandibular dental
arch. Am J Orthod Dent Ortho 97:323–335.
Alt KW, T€urp JC. 1998. Hereditary dental anomalies. In: Alt KW, R€osing FW,
Teschler-Nicola M, editors. Dental anthropology: fundamentals, limits
and prospects. Vienna: Springer. p 95–128.
Altemus LA. 1959. Frequency of the incidence of malocclusion in
American Negro children aged twelve to sixteen. Angle Orthod
29:189–200.
American Association of Orthodontists. 1999. Orthodontic Dialogue.
Volume 11. St. Louis, MO: American Association of Orthodontists.
Antanas S, Giedre T. 2006. Effect of the lower third molars on the lower
dental arch crowding. Stomatologija 8:8–84.
Bergstrom K, Jensen R. 1961. Responsibility of the third molar for second-
ary crowding. Dent Abstr 6:544.
Bishara S. 1999. Third molars: a dilemma! Or is it? Am J Orthod Dent
Orthop 115:628–633.
Bj€
ork A, Elli J, Mogens P. 1956. Mandibular growth and third molar impac-
tion. Acta Odont Scand 14:231–272.
Bj€
ork A, Skieller V. 1972. Facial development and tooth eruption. An
implant study at the age of puberty. Am J Orthod 62:339–383.
Buschang PH, Shulman JD. 2003. Incisor crowding in untreated persons
15-50 years of age: United States, 1988–1994. Angle Orthod
73:502–508.
Celikoglu M, Miloglu O, Kazanci F. 2010. Frequency of agenesis, impac-
tion, angulation, and related pathologic changes of third molar teeth
in orthodontic patients. J Oral Maxillofac Surg 68:990–995.
Corruccini RS. 1984. An epidemiologic transition in dental occlusion in
world populations. Am J Orthod 86:419–426.
Drummond R. 2003. Orthodontic status and treatment need of 12-year-
old children in South Africa: an epidemiological study using the Dental
Aesthetic Index. MSc Thesis, University of Pretoria.
Farman A. 2007. Tooth eruption and dental impaction. In: Farman A, edi-
tor. Panoramic radiology. Seminars on maxillofacial imaging and inter-
pretation. New York: Springer. p 73–82.
Graber TM, Kaineg TF. 1981. The mandibular third molar–its predictive
status and role in lower incisor crowding. Proc Finn Dent Soc
77:37–44.
Harper C. 1994. A comparison of medieval and modern dentitions. Eur J
Orthod 16:163–173.
Harradine NW, Pearson MH, Toth B. 1998. The effect of extraction of third
molars on late lower incisor crowding: a randomized controlled trial.
Br J Orthod 25:117–122.
Harris EF, Corruccini E. 2008. Quantification of dental occlusal variation: a
review of methods. Dent Anthropol 21:1–11.
Hinton RJ. 1982. differences in interproximal and Occlusal Tooth Wear
Among Prehistoric Tennessee Indians: implications for masticatory
function. Am J Phys Anthropol 57:103–115.
Hlongwa P, Du Plessis J. 2005. Malocclusion among 12-year-old school
children in Mankweng, Limpopo Province of South Africa. SADJ
60:455–457.
Harris EF. 1997. A longitudinal study of arch size and form in untreated
adults. Am J Ortho 111:419–427.
Isaacson JR, Isaacson RJ, Speidel TM, Worms FW. 1971. Extreme variation
in vertical facial growth and associated variation in skeletal and dental
relations. Angle Orthod 41:219–229.
Jantz RL, Meadows Jantz L. 2000. Secular change in craniofacial morph-
ology. Am J Hum Biol 12:327–338.
Kaifu Y, Kazutaka K, Townsend GC, Richards LC. 2003. Tooth wear and the
‘‘design’’ of the human dentition: a perspective from evolutionary
medicine. Am J Phys Anthropol 46:47–61.
Kramer R, Williams A. 1970. The incidence of impacted teeth. A survey at
Harlem Hospital. Oral Surg Oral Med Oral Pathol 29:237–244.
Lavelle CLB. 1973. Variation in the secular changes in the teeth and den-
tal arches. Angle Orthod 43:412–421.
Leighton BC, Hunter WS. 1982. Relationship between lower arch spacing/
crowding and facial height and depth. Am J Orthod 82:418–425.
Lindauer S, Laskin D, T€ ufekçi E, Taylor R, Cushing B, Best A. 2007.
Orthodontists’ and surgeons’ opinions on the role of third molars as a
cause of dental crowding. Am J Orthod Dent Orthop 132:43–48.
Lindqvist B, Thilander B. 1982. Extraction of third molars in cases of antici-
pated crowding in the lower jaw. Am J Orthod 81:130–139.

Little RM. 1975. The irregularity index: a quantitative score of mandibular
anterior alignment. Am J Orthod 68:554–563.
Little RM, Wallen TR, Riedel RA. 1981. Stability and relapse of mandibular
anterior alignment—first premolar extraction cases treated by trad-
itional edgewise orthodontics. Am J Ortho 80:349–365.
Lombardi AV. 1982. The adaptive value of dental crowding: a consider-
ation of the biologic basis of malocclusion. Am J Orthod Dent Orthop
81:38–42.
Miethke RR, Behm-Menthel A. 1988. Correlations between lower incisor
crowding and lower incisor position and lateral craniofacial morph-
ology. Am J Orthod and Dent Orthop 94:231–239.
Miyar KO. 2012. The origins of dental crowding in the Florida Archaic: an
anthropological investigation of malocclusions in Windover Pond
(8BR246). PhD Dissertation, Florida State University.
Mugonzibwa EA, Eskeli R, Laine-Alava MT, Kuijpers-Jagtman AM, Katsaros
C. 2008. Spacing and crowding among African and Caucasian children.
Orthod Cranio Res 11:82–89.
Nara T, Hanihara T, Dodo Y, Vandermeersch B. 1998. Influence of the
interproximal attrition of teeth on formation of Neanderthal retromolar
space. Anthropol Sci 106:397–309.
Nasby JA, Isaacson RJ, Worms FW, Speidel TM. 1972. Orthodontic extrac-
tions and the facial skeletal pattern. Angle Orthod 42:116–122.
Niedzielska I. 2005. Third molar influence on dental arch crowding. Eur J
Orthod 27:518–523.
Pedersen PO. 1944. Dental notes and a chapter on the dentition. In:
Br€oste K, Fisher-Møller K, editors. The Medieval Norsemen at Garder.
Anthropological investigation. Medd om Grønland 89:40–51.
Pedersen PO. 1949. The East Greenland Eskimo dentition. Medd om
Grønland 149:1–244.
Perillo L, Masucci C, Ferro F, Apicella D, Baccetti T. 2010. Prevalence of
orthodontic treatment need in southern Italian schoolchildren. Eur J
Orthod 32:49–53.
Pirttiniemi P, Oikarinen K, Raustia A. 1996. The effect of removal of all
third molars on the dental arches in the third decade of life. Cranio
12:23–27.
Proffit WR. 1978. Equilibrum revisited: factors influencing position of the
teeth. Angle Orthod 48:175–186.
Proffit W, Fields H, Jr, Moray L. 1997. Prevalence of malocclusion and
orthodontic treatment need in the United States: estimates from the
NHANES III survey. Int J Adult Orthodon Orthognath Surg 13:97–106.
Quek S, Lay C, Lay K, Loh S, Lim K. 2003. Pattern of third molar impaction
in a Singapore Chinese population: a retrospective radiographic study.
International J Oral Maxillofac Surg 32:548–552.
Richardson M. 1977. The etiology and prediction of mandibular third
molar impaction. Angle Orthod 47:165–172.
ANNALS OF HUMAN BIOLOGY 7
Richardson M. 1979. Late lower arch crowding facial growth or forward
drift? Eur J Orthod 1:219–225.
Richardson M. 1989. The role of the third molar as the cause of late lower
arch crowding: a review. Am J Orthod Dent Orthop 95:79–83.
Richardson M. 1996. Orthodontic implications of lower third molar devel-
opment. Dent Update 23:96–102.
Richardson ME. 2002. Late lower arch crowding: the aetiology reviewed.
Dent Update 29:234–238.
Rose JC, Roblee RD. 2009. Origins of dental crowding and malocclusions:
an anthropological perspective. Com Cont Edu Dent 30:292–300.
Scheuer L, Black SM. 2000. Developmental juvenile osteology. San Diego,
CA: Elsevier Academic Press.
Schwarze C. 1975. The influence of third molar germectomy- a compara-
tive long-term study. Transactions of the Third International
Orthodontic Congress, Held in London 13-18 August, 1973. Mosby. p
551–562.
Shanley L. 1962. The influence of mandibular third molars on mandibular
anterior teeth. Am J Orthod 48:786–787.
Sheneman J. 1968. Third molar teeth and their effect upon the lower
anterior teeth: a survey of forty-nine orthodontic cases five years after
band removal. MSD Thesis, St. Louis University.
Smith BH, Garn SM, Hunter WS. 1986. Secular trends in face size. Angle
Orthod 56:196–204.
Soh J, Sandham A, Chan YH. 2005. Occlusal status in Asian male adults:
prevalence and ethnic variation. Angle Orthod 75:814–820.
Southard T. 1992. Third molars and incisor crowding: when removal is
unwarranted. J Am Dent Assoc 123:75–79.
Stanaityte_ R, Gervickas GTA. 2014. Do wisdom teeth induce lower anterior
teeth crowding? A systematic literature review. Stomatologija
16:15–18.
Stephens C. 1980. The effects of third molar removal on the size of pre-
molar extraction spaces in the lower arch. Brit J Ortho 7:189–193.
Truesdell N. 2005. Secular change in the skull between American Blacks
and Whites. MA Thesis, Department of Geography and Anthropology,
Louisiana State University.
Vego L. 1962. A longitudinal study of mandibular arch perimeter. Angle
Orthod 32:187–192.
Winter GB. 1926. Principles of exodontia as applied to the impacted man-
dibular third molar: a complete treatise on the operative technic with
clinical diagnoses and radiographic interpretations. New York:
American Medical Book Company.
Wolpoff MH. 1971. Interstitial wear. Am J Phys Anthropol 34:205–228.
Yilmaz RS, Darling AI, Levers GH. 1980. Mesial drift of human teeth
assessed from ankylosed deciduous molars. Arch Oral Biol 25:127–131.