Mandibular Growth, Remodeling, and Maturation During Infancy and Early Childhood

Although the craniofacial complex has been extensively studied,1–4 growth and development during infancy and early childhood remain poorly understood. It is, however, well established that the greatest postnatal rates of somatic growth occur during the first 5 postnatal years.5 US children, for example, undergo a marked deceleration of growth in recumbent length during the first 3 years (http://www.cdc.gov/growthcharts).6 Based on the close associations between somatic and craniofacial growth and development,7–9 greater rates of craniofacial growth might also be expected during the first few postnatal years. Craniometric and anthropometric studies support the notion of marked craniofacial growth changes during the first 5 years.10,11,12 Ohtshuki et al13 reported greater cranial base growth during the first 5 postnatal years, especially during the first 2–3 years, than during the remaining postnatal years.

While the studies are limited and the data fragmentary, mandibular growth also appears to be most rapid during infancy and early childhood. The ascending ramus dramatically changes its spatial relationship with the mandibular corpus during infancy, when the gonial angle decreases substantially.14 Mandibular ramus height and corpus length demonstrate higher growth velocities between 3 and 5 years than anytime thereafter.15 Based on a subsample of 32 subjects, Broadbent et al16 showed greatest growth rates for ramus height between 1 and 2 years of age, with rates decreasing thereafter. The greatest increases in bicondylar width have also been reported to occur during the first 3 years of life.17 

The purpose of this study was to describe mandibular growth, remodeling, and maturational changes that occur during infancy and early childhood. The maturation, growth at the condyle, and remodeling at specific sites of the mandible have not been previously quantified for children aged 0.4–5 years. The aims were to determine whether there are differences in growth and maturation associated with age, sex, or molar relationship.

Serial lateral cephalometric records of 48 normal, untreated, healthy Caucasians (24 males and 24 females) were drawn from the longitudinal Bolton-Brush Growth Study, which includes approximately 5400 individuals of mostly European ancestry.21 The sample included 24 subjects with Angle Class I and 24 subjects with Class II occlusions, as categorized by the Bolton Study. Subjects with growth anomalies, syndromes, or poor quality cephalograms were excluded.

The subjects were also chosen based on having good quality serial lateral cephalograms taken sometime during the first year of life (average age 0.4 ± 0.1 year), at approximately 1 year of age, and every year thereafter until approximately age 5. Each subject also had to have an adult cephalogram, taken at the minimum ages of 15 and 17 years for females and males, respectively; females were 16.9 ± 0.9 years old on average, and males, 18.6 ± 1.2 years on average.

A total of 336 cephalograms were traced and digitized using the Dentofacial Planner (Dentofacial Software, v. 7.02, Toronto, Canada) software. All cephalograms were traced on 0.003-inch matte acetate and digitized by the primary author using a Numonics Model A30BL (Numonics Corporation, Montgomeryville, Pa) digitizer. The cephalograms were taken at the minimum midsagittal plane-to-film distance, producing average magnifications ranging from 7.4% to 8.4%.16 Differences due to magnification were not corrected in the present study.

Two cranial base and nine mandibular landmarks were identified on each tracing (Figure 1; Table 1). Three linear and two angular measurements were calculated to provide overall representations of mandibular growth and shape changes, including

• Ramus height (Co-Go)

• Overall length (Co-Gn)

• Corpus length (Go-Gn)

• Condylion angle (Go-Co-Me)

• Gonial angle (Co-Go-Me)

Anterior and posterior reference landmarks registered on sella and oriented on sella-nasion plane minus 7° (SN−7°) were marked on the adult tracing and transferred to the other tracings following each superimposition. The mandibles were superimposed on naturally stable structures using the techniques described by Björk and Skieller.2 They were superimposed on 1) the anterior contour of the chin, 2) the inner contour of the cortical plate at the lower border of the symphysis, 3) any distinct trabecular structure in the symphysis, and 4) the contour of the mandibular canal. The tracing of the 5-year-old mandible was first superimposed on the adult tracing, the 4-year-old tracing was then superimposed on the 5-year-old tracing, and so on, with the 0.4-year tracing being the last to be superimposed. Rectangular coordinates were used to describe mandibular remodeling changes, defined as the changes in landmark position over time evaluated horizontally and perpendicularly to the adult SN−7° reference plane.

To determine the cephalometric errors associated with tracing and landmark identification, 28 randomly selected cephalograms were replicated. Systematic errors, ranging from 0.01 mm–0.09 mm and 0.02°–0.11°, were not statistically significant. Method errors (√∑deviatons²/2n) ranged between 0.06 mm and 0.27 mm and between 0.26° and 0.42°, with the gonial angle and menton horizontal showing the largest random measurement errors.

Descriptive and inferential statistics were calculated using SPSS version 15.0 (Social Sciences, SPSS Inc, Chicago, Ill). Annual growth velocities were calculated by dividing the measurement differences between occasions by the corresponding age differences. Because the data were normally distributed, two-way analyses of variance were used to evaluate the effects of sex and occlusion. Relative growth changes of each of the measures were calculated as percentages of each subject's adult size.

A two-way analysis of variance showed that there was no significant (P > .05) interaction between sex and class of occlusion, and no significant class-of-occlusion effect. There were statistically significant (P < .05) sex differences in mandibular size, growth changes, and remodeling changes.

Mandibular size increased 18.2 mm to 34.7 mm between 0.4 and 5.0 years of age (Table 2). Overall mandibular length (Co-Gn) increased the most. Males displayed greater growth increases for ramus height (Co-Go) than for corpus length (Go-Gn), and females showed similar changes. While the condylar angle did not change significantly between 6 months and 5 years, the gonial angle decreased 2.8° and 2.0° in males and females, respectively.

Although yearly increments were significantly larger for males than for females, they showed similar patterns of change between 0.4 and 5.0 years (Table 3; Figure 2). Growth changes were greatest during the first 6 months, and they decreased progressively thereafter. Growth rates in mandibular size were approximately 1.8–2.2 times greater during the first postnatal year of life than during the second year and 1.3–1.9 times greater during the second than during the third year for females and males, respectively (Table 3). The gonial angle also showed the greatest changes during the first year (Figure 2).

Condylion, posterior ramus, inferior ramus, gonion, posterior corpus, and point B remodeled superiorly and posteriorly. Gnathion and pogonion remodeled inferiorly and posteriorly (Figure 3; Tables 4 and 5). Between 0.4 and 5.0 years of age, the condyle and posterior ramus remodeled 18.3 mm–22.5 mm posteriorly and 25.0 mm–29.4 mm superiorly. Landmarks in the gonial region remodeled 15.2 mm–19.7 mm posteriorly and only 6.3 mm–11.7 mm superiorly. The ramus landmarks showed the greatest yearly growth velocities between 0.4–1.0 year, with velocities decelerating thereafter. With the exception of the superior remodeling of B-point, the symphyseal landmarks showed only limited remodeling changes.

At approximately age 0.4 year, male mandibles were significantly (P < .05) less mature than female mandibles. For both sexes, ramus height was the least mature, followed by corpus length, and overall size, respectively. During the first 5 years, ramus height matured by approximately 31%, corpus length matured 25.2%–27.4%, and overall mandibular size matured 24.0%–25.2%. All three measures showed the greatest maturity changes during the youngest ages (Figure 4; Table 6).

Yearly rates of ramal growth and remodeling were greatest during the first postnatal year, and they decreased regularly thereafter. In the only other study quantifying longitudinal mandibular growth between 1 and 5 years of age, Broadbent et al16 reported mean values indicating that ramus height (Ar-Go) increases 2.1 mm–2.2 mm between ages 1 and 2, and 1.8 mm–1.3 mm between 2 and 3 years for males and females, respectively. Yearly velocities for ramus height (Co-Go) in the present study decreased from 10.1 mm–11.5 mm per year between 0.4 and 1.0 year, to 4.6 mm–5.8 mm per year between 1 and 2 years, and then to 2.9 mm–3.0 mm per year between 2–3 years for males and females, respectively. This indicates that articulare underestimates actual growth rates. The rapid deceleration of growth that occurred during the first 2 postnatal years also occurs in general somatic growth; yearly growth velocities for recumbent length of boys decrease from approximately 25 cm per year during the first year, to approximately 12 cm per year during the second year (http://www.cdc.gov/growthcharts).6 The rapid growth observed immediately after birth reflects a continuation of the even more rapid rates of growth that occur prenatally.18

Between 2 and 5 years of age, mandibular growth rates continued to decrease; growth rates were greater between ages 2 and 3 than between 3 and 4, which were in turn greater than the changes occurring between 4 and 5. Tracy and Savara,15 who studied the mandibular growth of 50 girls between ages 3 and 16, reported that annual increments of overall mandibular length, ramus height, and corpus length also decreased between 3 and 5. Based on a mixed longitudinal sample of 51 subjects, Baume et al19 reported similar trends for mid- and lower facial growth between 4 and 8 years. Again, the deceleration of craniofacial growth observed reflects a general somatic growth trend that occurs during the first 5 postnatal years.5 

Condylion and posterior ramus remodeled superiorly only slightly more than they remodeled posteriorly, which is dramatically different than the pattern seen in older children and adolescents. In fact, the condyle grew only 1.1 times more superiorly than posteriorly. In contrast, Björk's implant studies of 100 children 5 to 22 years of age show substantially greater amounts of vertical than posterior condylar growth.1,2,20 Baumrind et al21 evaluated 31 patients who had metallic implants placed in their mandibles and reported that condylion grew nine times more superiorly than posteriorly between ages 8.5 and 15.5. Using the same SN−7° orientation as in the present study, Buschang and Gandini22 reported that condylion and posterior ramus grew approximately eight times more superiorly than posteriorly between 6 and 15 years. These differences imply a definitive change in condylar growth direction during later childhood and adolescence, which may be related to the transition from the primary to permanent dentitions.

In contrast to the condylar region, the gonial region showed approximately twice as much posterior than superior remodeling. More specifically, gonion remodeled approximately two times more posteriorly than superiorly. While this follows the pattern of resorption and deposition of bone along the inferior and posterior aspects of the gonial region reported during later childhood and adolescence,1,2 the relative amounts of horizontal and vertical growth are again substantially different. Baumrind et al21 reported that gonion remodeled 6.7 mm posteriorly and 6.0 mm superiorly between 8.5 and 15.5 years of age.

Buschang and Gandini22 found that gonion remodeled superiorly approximately the same amount or slightly more than it remodeled posteriorly between 6 and 15. This again indicates a dramatic difference in the remodeling patterns of younger and older children that has not been previously reported. The remodeling changes observed for the ramus do not support the notion that the form and proportions of the human mandible are determined at a very early age and, once established, do not change.23 

The gonial angle decreased 2.5° from 129° at age 0.4 year to 126.5° at 5.0 years. Broadbent et al16 reported somewhat larger decreases in the gonial angle, from 134.5° at 1 year to 130.5° at 5 years. Their use of articulare rather than condylion could explain the differences reported. Decreases of the gonial angle observed in the present study are also less than the 5°–10° decrease reported by Izard between birth and complete eruption of the primary dentition,24 and the 15° decrease reported by Stunz25 during the first 2 postnatal years. The differences might again be due to the different measurement used. Importantly, there was great individual variability within the sample in the changes that occurred in gonial angulation, suggesting that sample composition might also explain some of the differences observed.

With the exception of point B, the symphyseal landmarks showed only minimal remodeling changes. Gnathion remodeled somewhat inferiorly and posteriorly, pogonion remodeled slightly superiorly, whereas point B remodeled 4.6 mm–5.5 mm superiorly. Björk1 reported that the anterior aspect of the chin was extremely stable in the anteroposterior direction. Thickening of the symphysis is due to apposition on the posterior surface.1,2,22 During childhood and adolescence, the height of the symphysis increases primarily due to vertical dentoalveolar development and secondarily from apposition at the lower border.21,26,27 

In terms of relative growth, corpus length was consistently the most mature, followed by overall length, and then by ramus height. Buschang et al28 reported a similar pattern of maturity for children 4.5 years of age, with corpus length being 71% and 76% complete for males and females, respectively; ramus height was 67% complete in males and 71% complete in females.28 Most important, the present study showed that ramus height had attained only 38%–42% of its adult size at age 0.4 year. Ramus height, being the least mature, might be expected to show a greater response to environmental and epigenetic stress than overall mandibular length and corpus length.29

The mandibles of males were significantly larger, they showed greater growth increments, and they were significantly less mature than female mandibles. Females were 3.3%–3.9% more mature than males, with corpus length (Go-Gn) showing the greatest maturity differences. Sex differences in mandibular size favoring males have been previously reported for older children and adolescents.30 Loth and Henneberg, who studied 62 juvenile mandibles from birth to 19 years, also found sex differences in the shape of the inferior border of the symphysis and the outline of the mandibular body during the first few years of life.31 

Variability in yearly growth changes between subjects during the first 5 postnatal years was large. Given the relatively good reliabilities obtained for the cephalometric procedures, the variability observed must have been partially innate and partially due to patient positioning. The latter was probably an important factor because many of the subjects had their mouths open to varying extents when the radiographs were taken, which is why the study was restricted to the mandible. It is reasonable to assume that some of these young children may have been positioning their mandibles off to one side, which would introduce projection errors that would increase intersubject variability. However, these errors might be expected to be random and, as such, have little or no effect on the mean values reported. They might, however, be expected to affect the statistical power of the tests performed. The fact that the present study was able to identify statistically significant differences make power a nonissue with respect to sex and age effects. However, power could have played a role in our inability to identify significant differences between Angle Class Is and Class IIs.

The lack of significant differences in mandibular growth and maturation between Class Is and Class IIs younger than 5 years of age can be supported by 1) the lack of clear pattern of Angle class differences across the ages and 2) previous studies showing little or no difference due to class in younger subjects. It has previously been shown that actual growth differences between Class Is and IIs are small,32 and might require many years to accumulate and become statistically significant. Kerr and Hirst33 found that only the cranial base angle could be used to distinguish between Class Is and IIs when they were 5 years of age. Bishara et al34 reported that Class Is and IIs in the primary dentition differed in terms of arch width but not in mandibular size.

• During infancy and early childhood, the greatest increases occur during the first 6 months of life and decrease progressively thereafter.

• Overall mandibular length showed the greatest growth changes, followed by ramus height and corpus length.

• The gonial angle decreased 2.8° and 2.0° in males and females, respectively.

• The condylar region remodeled superiorly only slightly more than it remodeled posteriorly; the gonial region showed approximately twice as much posterior than superior remodeling.

• Male mandibles were significantly (P ≤ .05) larger, displayed greater growth rates, and were significantly less mature than female mandibles.

• Ramus height was less mature than overall mandibular length, which was in turn less mature than corpus length.

• There were no significant differences in mandibular growth or maturation between Class Is and Class IIs.