Assessment and Prediction of Dentofacial growth are perhaps the most essential, yet to a great extent the most subjective, aspects of clinical orthodontics. Although the percentage of adult patients has increased in recent years, the majority of treatment is still directed toward pre-adolescent and adolescent patients. These individuals are undergoing significant growth changes in their occlusions, facial skeletons, and profiles. The purpose of this paper is to review pertinent longitudinal facial growth data and some of the methods used to predict facial growth. The clinical implications of the available information is also discussed. The ability to forecast or predict growth lies at the very heart of contemporary clinical orthodontics. The orthodontist in formulating a treatment plan relies largely upon subjective criteria in conceiving the outcome of treatment. This intuitive perception is necessary, but the overall approach should be based on the available scientific information. There are several components to be dealt with in the prediction of craniofacial changes: the direction, the magnitude, the timing, the rate of change, and the effects of treatment.32 Orthodontists, in general, are well informed regarding the effects of orthodontic treatment on the patient, but are not yet able to accurately predict the direction, timing, and magnitude of the facial changes that occur with growth in any single individual.Abstract
INTRODUCTION
Assessment and prediction of dentofacial growth are perhaps the most essential, yet to a great extent the most subjective, aspects of clinical orthodontics. Although the percentage of adult patients has increased in recent years, the majority of treatment is still directed toward pre-adolescent and adolescent patients. These individuals are undergoing significant growth changes in their occlusions, facial skeletons, and profiles.
Such changes are complex1 because each person has a unique growth pattern influenced by their genetic make-up (ie, the biological or internal environment) as well as external environmental factors such as function, disease, habits, and orthodontic treatment.
Cephalometric superimpositions often demonstrate dramatic dental, skeletal, and soft tissue changes during orthodontic treatment. Many orthodontists give themselves full credit when they take advantage of the patient's favorable growth combined with a reasonable orthodontic treatment plan. These same clinicians, however, blame unfavorable growth and lack of patient cooperation when the treatment results are anything short of their expectations.
Since growth can be either a friend or a foe, it is important to determine the timing, magnitude, and direction of facial growth. Such an understanding enables orthodontists to better plan the treatment of skeletal discrepancies in their attempt to achieve a more stable and pleasing result. Therefore, it behooves all clinicians to discern between the science and the fiction in facial growth, or as Professor Koski once said “the facts and fallacies” in facial growth.2
The purpose of this paper is to review pertinent longitudinal facial growth data and some of the methods used to predict facial growth. The clinical implications of the available information is also discussed.
IOWA GROWTH STUDIES
The Iowa Growth Study is a set of data collected by Drs Meredith and Higley at the University of Iowa starting in 1946. Cephalograms, study casts, and other records were obtained on the participants from ages 4 through 18 years. Another set of records was taken at adulthood around 25 years of age and a final set was taken at 45 years of age.3 All subjects were Caucasians and 97% were of northern European ancestry. This population provided the material for the series of Iowa Facial Growth Studies published during the last 50 years.
OVERALL CHANGES BETWEEN 5 AND 25 YEARS OF AGE
The growth changes in standing height between 5 and 25 years as well as those for various craniofacial parameters were divided arbitrarily into 3 stages, specifically: 5–10, 10–15 and 15–25 years.6 This was an attempt to better understand what happens in the face in very broad terms at various stages of development that are of particular interest to the orthodontist.
Standing height is usually used as a standard or indicator of skeletal body maturation and was included in these evaluations. Approximately 40% of the total change in standing height occurred between 5 and 10 years, another 40% occurred between 10 and 15 years, and the balance occurred after 15 years (Table 1). When the females are compared to the males, the females have relatively greater growth increments between 5 and 10 years than between 10 and 15 years. These relative differences in the magnitude of the changes in standing height between males and females are also found in most linear facial dimensions such as face height and depth, but not as readily observed when the facial relationships were evaluated (Tables 1 and 2).
What is the magnitude and direction of the changes in the facial parameters?
The relative changes in maxillary length were approximately 40%, 40%, and 20% in males and 50%, 30%, and 20% in females (Table 1). On the other hand, the changes in maxillary relationship were: 11%, 78%, and 11% in males and −25, 50, and 75% in females. Two observations are worth emphasizing. The percentages are different in males and females. Also, the timing of the linear and positional changes of the maxillary complex does not closely correspond in the 3 periods. For example, maxillary length expressed 50% of its total increase, between 5 and 10 years of age in females. On the other hand, SNA angle actually decreased (ie, point A moved relatively further back in relation to the cranial base). This change could be the result of a relatively greater forward movement at nasion or a clockwise rotation of the maxillary complex.
In the mandible, the changes in mandibular length in males were 34%, 39%, and 27% in the 3 growth periods (Table 1). The corresponding changes in relationship (SNPog) were similar at 37%, 33%, and 30%. In females, the changes in mandibular length were 48%, 41%, and 11% with similar changes in mandibular relationship of 41%, 50%, and 9%.
Looking at the maxillary-mandibular relationship as described by the ANB angle, some interesting differences are apparent (Table 1). In males, the relative decrease in the ANB angle was 31%, 6%, and 63% in the 3 periods of growth, while in females there was an increase in the angle in the last period.
After evaluating other parameters in the dentofacial complex (Tables 1 and 2), the following conclusions were reached.
-
Significant changes occur in the periods between 5–10 and 10–15 years; these changes are, in general, significantly greater than the changes between 15 and 25 years.
-
The changes occur earlier in females than in males (Tables 1 and 2).
-
The changes in the different parts of the face are not necessarily similar in either their timing or their magnitude (Tables 1 and 2).
-
The timing of change between the linear and angular changes in a structure does not often occur at the same time or in the same direction (Table 1).
-
There are significant changes between 15–25 years in certain facial parameters, specifically the face height (Table 2), soft tissue facial convexity (Table 2), and ANB angle (Table 1).
Actually, most of the effective changes in the soft tissue profile occurred between 15 and 25 years of age (Table 2). This is an interesting observation because it indicates that most of the decrease in the convexity of the profile occurred in late adolescence.
CHANGES IN VARIOUS FACE TYPES WITH AGE
In a study7 that characterized face types as either long, average, or short, as determined from an evaluation of both the cant of the mandibular plane (MP:SN angle) and the ratio of posterior/anterior face heights (S-Go:NMe), several questions were addressed.
How often do people change their facial types between 5 and 25 years of age?
Most people (77%) were categorized as having the same facial type at 5 and 25 years of age. This means that as facial growth progressed with age, there was a strong tendency to maintain the overall facial pattern. The differences between the 3 facial types, particularly in the vertical relationships, were more pronounced at adulthood than in childhood. This indicated that the facial pattern became more expressed with age.7
This concept of predisposition has been illustrated previously in other parameters. For example, resemblance between the size of children and parents is small in infancy as compared to later in life. This is thought to be due to the ability of some genes to be more strongly expressed sometime after birth. In other words, children with a genetic predisposition to be large in size but were born small actually increased to a higher percentile of height and weight within the first 6 months of life. Similarly, children born large, but had a genetic predisposition to be small, took approximately 18 months to move downwards in the height and weight percentiles. It seems that size at birth reflects uterine conditions much more than the baby's genetic make-up.8
Facial growth data is supportive of this predisposition in the majority of cases since 77% of the faces in the study maintained their face type.7 However, there was a change in the categorization of facial type between 5 years and 25 years of age in 23% of the subjects. This “shifting” from one face type to another mostly occurred in cases with “borderline” characteristics between 2 facial types (ie, from average to long or from short to average). Whether these changes were a late genetic expression or whether they resulted from environmental influences or both is difficult to determine and creates a dilemma for the clinician planning the treatment of these cases.
How does the growth of the 3 facial types, long, average, and short, differ from each other?
When the growth curves between 5 and 25 years of age for the 3 facial types were evaluated, there was a lack of significant differences in the shape or slope of the absolute growth curves. This finding indicated that the curves were parallel for the 3 face types. As an example, the mean curves for SNA and SNPog that are illustrated in Figures 1 and 2, essentially describe similar trends.7
Citation: The Angle Orthodontist 70, 6; 10.1043/0003-3219(2000)070<0471:FADCIA>2.0.CO;2
This overall pattern of parallelism was observed in most of the curves for the 48 dentofacial parameters evaluated and indicated that, regardless of facial type, the curves demonstrate a parallel relationship showing similar growth behavior and similar growth direction. As interesting as this finding might seem in the context of facial growth, such a phenomenon parallels the findings in the well-publicized standards for standing height for tall, average, and short persons.10
This consistency in the shape of the curves (parallelism) was matched by a consistent presence of significant differences among the 3 facial types in the amount of growth expressed for various parameters.7 For example, the short face type expressed relatively larger curves for most of the antero-posterior dentofacial parameters evaluated such as SNA and SNPog. On the other hand, the magnitudes of the curves for the vertical facial dimensions were larger in the long face type.
What causes the differences between the various facial types (ie, what should we expect to see as clinicians)?
The outcome of facial growth in the various facial types is influenced, at least in part, by: (1) the original size and relationships of the different parts of the face, and (2) the differences in the magnitude of change between the successive ages (ie, the rate of change). For example, the comparisons of the curves for SNPog indicate that the 3 facial types have parallel growth curves. Yet, the overall magnitudes of both the absolute and incremental curves for the 3 facial types were different. These curves were smaller for the long face and larger for the short face types.
How much variation is there in the dentofacial characteristics within each facial type (ie, can they all be treated the same)?
Even within a relatively homogenous and small sample, each face type expressed a considerable amount of variation.7 Individuals within each facial type were neither of similar size nor had similar dentofacial relationships. In other words, within each facial type there is more than 1 combination in both the size and relationship of the different parts of the dentofacial complex. Therefore, subtle changes in various parameters can influence the overall direction of facial growth as well as the ultimate facial relationships. As a result, during treatment planning the clinician should evaluate both the overall face type and the individual facial characteristics of each patient and develop a unique treatment plan.
METHODS OF PREDICTING FACIAL GROWTH CHANGES
All the variations described represent the range of normal relationships within a normal population, but how predictable are these variations in the growth of the face? According to Bjork,11 growth prediction can be accomplished by 3 general methods: longitudinal, metric, and structural.
Longitudinal Approach
With this method, an individual may be evaluated over a period of time in order to determine the pattern of growth. This concept was clinically applied by Tweed on his growing patients.12 He advocated taking 2 lateral cephalograms 12–18 months apart to evaluate the skeletal facial changes. Consequently, the patient was placed into 1 of 3 categories that were used to predict future growth trends. In Type A, growth of the middle and lower face proceeded in unison with approximately equal changes in the vertical and horizontal dimensions. In Type B, the middle face grew downward and forward more rapidly than the lower face predominantly in a vertical direction. In Type C, the lower face developed at a faster rate than the middle face. Tweed's basic assumption was that the growth pattern would remain constant.
This concept of the constancy of the growth pattern was presented in the early 1950s by Brodie.13 Soon after, Moore14 and other investigators and clinicians concluded that this constancy could only be observed with population averages, but was not useful in predicting the changes occurring in any single individual. Often the pattern and rate of growth occurring in a given period was not similar to that occurring in a subsequent period in any given individual. The obvious limitation of the longitudinal approach is that it is accurate only when it is performed retrospectively, but not prospectively. Therefore, it can be concluded that the longitudinal approach is not an accurate method of predicting future dentofacial changes.
Metric Approach
The metric method of predicting growth consists of measuring different structures on a single radiograph and then relating these measurements to future growth changes. From a clinical perspective, this should be an ideal method of prediction because of its simplicity. How successful is this method of prediction within a facial structure, between the various facial structures, and between the facial structures and other body dimensions?
At this juncture it might be helpful to explain the scientific determination and clinical application of the strength of the relationships between any 2 variables. A correlation coefficient, symbolized by a small “r,” describes the association or the strength of the relationship between 2 variables. A correlation coefficient also gives the direction, positive or negative, of this relationship. Its use in prediction is derived from squaring the value of “r,” which is called the “coefficient of determination” or r2. This coefficient, describes the amount of variation of the second variable that can be eliminated if the first variable is known.15
According to Horowitz and Hixon,15 a correlation coefficient may be statistically significant at the .001 level of confidence, but is still of no clinical significance for prediction. As a rule, they suggested an “r” value of 0.8, to be the dividing line for use in clinical prediction since the coefficient of determination, or r2, is 0.64 (ie, 64% of the variation can be accounted for in the variable that is being predicted).15 It is with these facts in mind that the available data will be interpreted.
In independent studies by Bjork,16,17 Harvold,18 Lande,19 Solow,20 and others,21,22 correlation coefficients for facial dimensions, whether linear or angular, when related to future growth of that same dimension did not exceed an r-value of 0.4 or 0.5 (ie, they only explained 16% to 25% of the variation). Meredith earlier examined this concept in considerable detail on the Iowa growth data.23,24 He related the growth changes for a series of head, face, and other body dimensions and calculated correlations between: (1) the size at one age to the size of the same parameter at another age; (2) the size of the parameter at the first age to the amount of change at a subsequent age; and (3) the amount of change at one period to the changes at subsequent periods. These correlations were performed both within a variable and between different variables. Of hundreds of correlation coefficients calculated, 60% had an r-value of less than 0.4.23–25 The highest correlation found was between the growth in face width and the growth in shoulder width with an r-value of 0.65. This is still a fairly low correlation for clinical use.
In summary, the metric method of prediction has its own limitation in predicting facial changes at least from a clinical perspective.
Structural Approach
The structural method for predicting mandibular growth direction was developed by Bjork from superimpositions on metallic implants.26 The method consists of recognizing specific structural morphological features in the mandible that would indicate future growth trends.
When evaluating mandibular morphology, Bjork listed 7 areas on the cephalogram that should be evaluated: (1) The inclination of the condyle, as an indication of its growth direction whether vertically or sagitally (eg, with vertical condylar growth, the mandible rotates forward); (2) the curvature of the mandibular canal (ie, the more curved the canal is the more forward mandibular rotation will be); (3) inclination of the symphysis (if it is inclined lingually the mandible will rotate forward); (4) shape of the lower border of the mandible; (5) the interincisal angle, which is more acute in forward rotators; (6) the interpremolar or molar angles, which are more acute in forward rotators; and (7) the anterior lower face height. Bjork recommended evaluating all these structural features in order to help predict future mandibular growth direction.11,26
Bjork's comprehensive work on both maxillary and mandibular changes also demonstrated the wide range of variation in the growth of the nasomaxillary complex, condyles, and mandibular position within a given population.11,17,26,27 Such variation is to be expected since the future direction of mandibular growth is influenced by the changes in other parts of the craniofacial complex. Namely, the changes in the cranial base, the position of the glenoid fossa, the nasomaxillary complex, and growth of the alveolar processes as well as what should be considered as the great unknown—the effects of future changes in the environment and function on the growth of the face. These last 2 variables are very difficult to quantify by any prediction method because they are essentially unknown to both the clinician and the patient.
However, the question that needs to be answered has to do with the usefulness of the structural method of prediction. In a later study, Skieller, Bjork, and Linde-Hansen28 attempted to refine this prediction approach by quantifying it. They found that the combination of 4 variables gave the best prognostic estimate of future mandibular growth direction. The variables were: (1) mandibular plane inclination to the anterior cranial base (MP:SN angle) or the ratio of posterior/anterior face heights; (2) the intermolar angle; (3) shape of the lower border of the mandible measured as the angle between Go-Me and a tangent to the lower border of the mandible; and (4) the inclination of the symphysis measured as the angle between the tangent of the anterior surface of the symphysis and SN.
When the measurements of these variables were included in the multiple regression analysis, the R2 was calculated to be 0.8612. This is a fairly high R2 since it explains more than 86% of the variation in the direction of mandibular growth. However, unfortunately, there was a catch. According to Skieller et al,28 the high R2 value obtained might be related to the fact that the sample they evaluated contained a number of extremes whose growth usually proceeds in a more consistent and predictable direction. They suggested that if the extreme cases were eliminated, prediction of the direction of growth will be much less reliable.
From a clinical perspective, one can conclude that if a patient has a very steep MP with an obtuse gonial angle and an openbite tendency with a severely retrognathic or prognathic mandible, it becomes fairly obvious that there is a high probability that the future direction of mandibular growth will be unfavorable. This will occur regardless of the best efforts of the clinician.
Since the prediction of the growth for individuals with extreme facial types does not seem to be a great challenge to the average clinician, this discussion will focus on the various methods of predicting the changes that occur in the average adolescent population. This group comprises most of the patients that the clinician will manage on a day-to-day basis.
In summary, the longitudinal, metric and structural methods of prediction are of limited clinical value. As a result, more sophisticated approaches to the problem were attempted using the more recently acquired computer technology.
COMPUTERIZED PREDICTION METHODS
Computerization is essentially a tool of analysis and not a method of analysis. Computers are programmed to use equations based on either the longitudinal, metric, structural, or other methods of prediction. The biggest advantage of computer technology is that it facilitates testing and applying more complex formulas to growth prediction.
In the 1970s, Ricketts29–32 was among the first to realize such a potential existed and suggested that the clinician could be provided with a much more complete analysis of a cephalogram, including diagnosis, treatment planning, and short- and long-term growth forecasts of the dentofacial changes with and without treatment. Ricketts introduced his method of computer analysis based on the concept of the cubic root combined with a vast clinical experience.
Greenberg and Johnston33 tested this method of computer predictions and found that the computer forecasts were essentially no better than the assumption of average growth. In other words, there were no significant differences in accuracy between the predictions generated by the computer and those based on the simple addition of the average changes. Furthermore, neither method (ie, the computer predictions and adding average changes) accurately represented the actual changes that occurred in most of the facial parameters evaluated.
The different methodologies30,34–36 presented by the computerized techniques might appear complex, yet they have inherently the same limitations described earlier. In general, dental changes caused by treatment are more predictable because they are, to a great extent, dictated by the clinician.
In summary, the overall changes in the size and relationship of the human face in the 20-year period from childhood to adulthood are difficult to accurately predict due to the influence of the combined and complex effects of genetic and environmental factors. The situation is rendered even more complex because we are using a 2-dimensional image, the cephalogram, to predict a 3-dimensional multifunctional object, the face.
According to Hixon,37,38 “to err is human; to really foul things up requires a poorly programmed computer.” This is not intended to mean that clinicians should not use the various computer programs available, but such programs are useful for general patient education and average growth or treatment simulations—not for individualized predictions.
FACIAL CHANGES AND THE ADOLESCENCE GROWTH SPURT
The adolescent growth spurt in the dentofacial structures, specifically in the mandible, is one of the most frequently mentioned concepts in facial growth. Clinicians have been told to gear their treatment timing so that it will coincide with the adolescent growth spurt. As clinicians, we would like to know how applicable is this concept?
The most accurate and reliable data on this subject can be obtained from studies using metallic implants. In a 1963 study, Bjork,11 evaluated the growth of the condyles in 45 boys between 7 and 21 years of age. Of the 45 boys evaluated in the study, Bjork found that only 11 individuals (less than 25% of the sample) had what he was able to describe as a “discernible pubertal growth variation.”11
What is the magnitude of the spurt, when it occurs?
For these 11 subjects, Bjork11 described a slower growth rate around 12 years of age amounting to a mean of 1.5 mm, and a “spurt” 2.0 years later that averaged 5.5 mm and ranged between 4.0 and 8.0 mm (Table 3). For the rest of the 34 subjects in the study there was a more steady annual condylar growth averaging 3.0 mm during the same period. As for the timing of the spurt, the mean age for its occurrence was 14.0 years with a range between 12 and 15 years.11
Bjork's findings and conclusions indicate that there was a discernable, but not necessarily significant spurt in condylar growth in less than 25% of the sample, and the magnitude, duration and timing of the spurt varied widely even in this selected subsample of 11 subjects.
There was no relationship between the intensity of the growth and its direction. These facts need to be kept in mind particularly when planning the orthodontic treatment of patients with skeletal discrepancies.
How do other (nonimplant) longitudinal studies, describe the adolescent changes in mandibular growth?
A slightly different picture is presented by the Michigan4 and Broadbent-Bolton5 studies. In the Michigan data4 the change in mandibular length (Ar-Gn) was fairly gradual. Similarly, the change in the anteroposterior relationship of the mandible (SNPog) also described a gradual increase with age. The Bolton data5 on the same parameters also showed a small consistent increase in their magnitude in both males and females with no significant spurts evident.
So how did many orthodontists come to believe in the universal existence of a mandibular growth spurt? Probably, it has to do with the fact that in some studies, the sample was divided into persons who actually demonstrated a pubertal acceleration in mandibular dimensions and those who did not demonstrate such a change. Only the findings on the first group (ie, the exceptions) were highlighted.11 Another possible explanation is that, if the scale of any curve is sufficiently enlarged, a small and clinically insignificant acceleration may be made to look fairly impressive.39–41
So what do all these findings mean to the clinician, as far as treatment planning?
The presence of a significant pubertal acceleration in mandibular parameters may occur in less than 25% of the cases, but not in all or even most persons. To expect anything else is wishful thinking. As a result, to routinely postpone or delay treatment in cases with skeletal anteroposterior discrepancies in anticipation of a “spurt” is not scientifically justifiable. This should not be interpreted to mean that such accelerations do not occur in any single person. It does indicate that changes which could be described as clinically significant “spurts” do not occur in a consistent pattern in the majority of our patients.
In reality, the clinician faces 2 problems: (1) the formidable task of determining which of the subjects within a given population will experience an adolescent growth spurt, and (2) for those few who will experience a spurt, when and to what extent will it occur and in what direction? In the absence of a consistent and significant growth spurt in adolescence, how can we explain the fact that we are successfully treating most of our growing and cooperating patients who have mild to moderate skeletal discrepancies?
OVERALL GROWTH CHANGES DURING EARLY ADOLESCENCE
How much growth is occurring during adolescence in a normal population?
In a series of studies on the Iowa Growth sample, the subjects were grouped according to the timing of the greatest change in the rate of growth in any 2 consecutive years between 8 and 17 years of age regardless of the skeletal or chronologic ages.20,21 Such an arrangement, will accentuate the amount of change in the 2 years that were labeled the “maximum growth period,” and compared to the 2 years prior (premaximum) and the 2 years following this period (postmaximum).
As Table 4 indicates, the average changes in standing height for males in the 3 periods were: 14.1 cm, 12.2 cm, and 7.4 cm, respectively. The average changes in mandibular length were: 6.3 mm, 5.4 mm, and 3.7 mm, respectively. The changes in mandibular relationship were: 1.2°, 1.0°, and 0.7°, respectively. The results further indicated that no significant differences were present in the changes in mandibular relationships between the premaximum and maximum periods. The logical conclusions from these findings are that, in most individuals, significant facial growth occurs during adolescence for a relatively long period of time.
This is not meant to advocate a particular age for treatment (ie, “early” vs “late”) since the timing of treatment is dependent on a number of factors including: (1) the nature of the malocclusion and its etiology; (2) the severity of the skeletal discrepancy; (3) the presence of a functional shift; (4) the stage of dental development, including the relationship of the erupting permanent second molar to the roots of the first molars; (5) growth expectations; (6) the ability of the patient to cooperate; and (7) the need for tooth extractions.
PREDICTION OF FACIAL CHANGES FROM SKELETAL BODY CHANGES
Is there a relationship between the timing of changes in the facial parameters and the changes in standing height?
In the same series of studies21,22 a special type of analysis referred to as the autocorrelation analysis was used to determine the relationship between the timing of various growth events in the face and standing height. In other words, the analysis compares the growth profile of the various facial parameters to that of standing height between 8 and 15 years on a longitudinal basis.
The findings indicated that, with a single exception, all the correlations between the timing of the changes in height and those for the changes in mandibular, maxillary, and cranial base length and relationships were below 0.5. Only mandibular length in girls had a clinically significant correlation with the timing of changes in standing height (r = 0.83). So much for using standing height to predict facial growth changes!
What about the correlations between the timing of facial growth events and the general skeletal maturation, as determined from wrist x-rays?
In 1990, Moore et al42 assessed the relevance of hand-wrist radiographs to craniofacial growth and clinical orthodontics. They evaluated serial cephalometric and hand-wrist radiographs on 86 children between 11 and 16 years of age. They used 4 skeletal linear measurements that are known to have statistically significant increases during that period. Their results indicated that: (1) growth spurts could not be consistently observed on an individual basis, and (2) the correlations between adolescent growth acceleration and deceleration in the facial dimensions, with both standing height and skeletal maturity as determined from wrist x-rays, were not clinically significant for prediction.
Therefore, this brief review strongly suggests that there is a lack of clinically useful correlations between the timing of change in either standing height or wrist x-rays as indicators of skeletal maturation and the changes in the various dentofacial parameters that clinicians are interested in predicting. On the other hand, wrist x-rays are good predictors of skeletal body growth.
Is the change in mandibular length accompanied by a corresponding forward positioning of the mandible? What should we expect?
Clinicians are interested in finding out whether the increase in mandibular length is going to be translated as a forward positioning of pogonion. It can be readily illustrated from the graphs of the mean curves that, as mandibular length increases, we do not observe a corresponding forward position at the chin point (Figure 3).
Citation: The Angle Orthodontist 70, 6; 10.1043/0003-3219(2000)070<0471:FADCIA>2.0.CO;2
Individual curves further indicate that the changes in mandibular relationship are expressed as a relatively constant change between 8 and 17 years of age. Even in those cases that expressed a spurt in mandibular length, the change in mandibular relationship did not express a corresponding significant spurt in the forward positioning of the chin point (Figure 4).
Citation: The Angle Orthodontist 70, 6; 10.1043/0003-3219(2000)070<0471:FADCIA>2.0.CO;2
Therefore, the forward positioning of the chin is more dependent on forward mandibular rotation/translation, than on the simple increase in mandibular length. This has been previously illustrated by Bjork,27 who found that similar amounts and directions of growth at the condyles can be expressed differently at pogonion depending on whether the center of rotation of the mandible is at the condyles, the premolars, or the incisors.
Is there a difference in the treatment results if orthopedic/orthodontic changes are initiated at different stages of development?
In 1998, Tulloch et al43 reported on a prospective evaluation of the dentofacial structures in 166 children in the mixed dentition between 7 and 12 years of age. The children had an increased overjet ranging from 7 to 12 mm. Tulloch and her colleagues randomly assigned these patients to 1 of 3 groups: observation only (N = 61), bionator treatment (N = 53), and headgear treatment (N = 52).
In part of the study the effects of treatment on the dentofacial structures were related to the time treatment was initiated as it relates to chronological age, dental age, and skeletal age as determined from wrist radiographs. The overall changes in the dentofacial structures over 15 months of treatment were correlated to each of the 3 maturity indicators. Their findings indicated that these correlations were close to 0 (ie, there is a very low association between the magnitude of treatment changes and treatment timing). Keeling et al44 and Ghafari et al,45 in similar prospective studies, confirmed the results obtained by Tulloch et al.43
CLINICAL IMPLICATIONS
The conclusions from these studies strongly suggest that there is very little to be gained from timing growth modification treatment to a particular stage of maturation whether it is chronological, dental, or skeletal. Therefore, similar treatment results can be obtained over a wider age range in the child's development.43 These conclusions explain why we are successful in treating most of our patients in spite of our scientific and clinical biases.
In treatment planning cases with anteroposterior discrepancies in growing individuals, the clinician should consider the following points:
-
The timing of mandibular changes in both size and relationship are not closely correlated to each other as well as to the changes in standing height or wrist x-rays and, hence, not accurately predictable from these parameters.
-
There are significant mandibular changes in size and relationship between the ages of 8 and 17 years. The changes in mandibular relationship were not significantly different in the maximum and premaximum periods in either boys or girls. The magnitude of change in the postmaximum period tended to be smaller than in the other 2 periods.
-
Because the occurrence, magnitude, and timing of the spurt in a particular patient is highly unpredictable, the treatment of anteroposterior discrepancies should be initiated as soon as the orthodontist believes that treatment is indicated and not delayed for the pubertal “spurt”.
-
Patient cooperation in wearing extraoral or functional appliances does not seem to improve at the later stages of adolescence. If anything, most of these patients are involved in various activities that often distract them from wearing such appliances. Actually, a study by Tung and Kiyak46 found that there was greater cooperation in younger ages than in adolescence.
So what are we left with?
Instead of the clinician attempting a hit-and-miss approach to prediction, it might be wiser to use more straight-forward concepts that can be applied in planning treatment for growing patients with anteroposterior skeletal discrepancies.
For cases with severe skeletal discrepancies, in general, it can be assumed that the existing growth pattern will prevail in most of these cases. For example, if a patient has a very steep mandibular plane, open bite tendency, long anterior face, and a Class II malocclusion at age 10 years, there is a high probability that a vertical growth pattern will continue. As a result, the orthopedic correction should include the use of an extraoral high pull force to the molars or any other appropriate appliance that the clinician prefers to use.
In the majority of cases (ie, those with an average skeletal discrepancy in which future growth is less predictable), one should assume the “the worst case scenario”. In other words, for the milder version of the case described, the assumption is that growth is going to proceed in an unfavorable direction relative to the needed correction. The treatment mechanics should avoid extrusive forces whether extraoral or intraoral. As treatment progresses, 2 possible outcomes may occur: (1) if the case significantly improves as a result of favorable growth and treatment changes, the clinician can easily modify or adjust the mechanics accordingly, or (2) if the growth proceeds in an unfavorable direction, the mechanics are already designed with this in mind.
CAN ALVEOLAR GROWTH AFFECT THE PLACEMENT OF IMPLANTS?
In patients with either a missing central incisor due to trauma or a congenitally missing lateral incisor, the treatment options for replacing the lost tooth following orthodontic treatment may include: (1) adding a tooth to the Hawley retainer as a temporary solution; (2) acid etched or conventional prosthetic replacement; or, (3) placing a single tooth implant.
In this presentation, we briefly discuss the effects of alveolar growth on the timing of implant placement. In a 1996 study, Iseri and Solow47 evaluated cephalograms on patients from the original Bjork material with metallic implants placed in the maxilla and mandible. They found significant anterior alveolar growth that continued into late adolescence and early adulthood. As a result, clinicians should postpone placing an implant in younger patients until alveolar growth is completed. Such growth could cause the implant to become progressively in infraocclusion.
MOLAR RELATIONSHIP CHANGES BETWEEN THE PRIMARY AND PERMANENT DENTITIONS
Clinicians involved in early orthodontic treatment must have a good understanding of the anteroposterior occlusal changes that occur from the primary to the permanent dentitions. These changes were evaluated on 121 individuals (60 males and 61 females) from the Iowa Growth Study.48 None of these subjects had congenitally missing teeth, early loss of primary first or second molars, or had undergone orthodontic therapy.
What is the terminal plane relationship in the primary dentition and its clinical implications?
The terminal plane relationship of the second primary molars can be described as: flush, occurring in 29.4% of the individuals; mesial step, occurring in 61.1%; and distal step, occurring in 9.5% of the individuals.48
At the time of eruption of the permanent first molars, their initial occlusion is dependent on the terminal plane relationship of the primary second molars. Therefore, in about 30% of the population the first molars will erupt into a cusp-to-cusp relationship (ie, with the mesiobuccal cusp of the maxillary first permanent molar anterior to the buccal groove of the mandibular first permanent molar). The findings from the study further indicated that, if the initial occlusion in the permanent dentition has a distal step, the molar relationship in the permanent dentition develops into a Class II. This provides credence to a frequently repeated axiom: “once a Class II, always a Class II”.48 Since the disto-occlusion in the primary or mixed dentitions will not self-correct with growth, cases with distal steps in the primary dentition should be observed on a regular basis and treatment started as soon as the clinician and the patient are ready.
Individuals with a flush terminal plane relationship present a more challenging diagnostic question. Of these cases, 56% became Class I and 44% became Class II in the permanent dentition. In other words, slightly less than half of these cases will progress to a Class II or end-to-end occlusion.48 These findings imply that what was previously considered as “normal” occlusion in the primary or mixed dentitions, in reality, does not often result in a “normal” occlusion in the permanent dentition. Therefore, it is important for the clinician to closely observe these cases and to initiate treatment when needed at the appropriate time. In the 56.4% of individuals with a flush terminal plane, placing a lower lingual holding arch to maintain space may have an adverse effect on the developing dentition (ie, it may prevent more cases with a flush terminal plane from becoming Class I). This is important to remember because orthodontists, pediatric dentists, and general dentists often consider placing a lower lingual arch as a preventive or interceptive but noninvasive procedure. One can readily imagine how maintaining arch length and preventing the mesial drift of the mandibular permanent molars, can adversely affect the changes in the molar relationship from the mixed to the permanent dentitions. As a result, in cases with an end to end molar relationship in which a lingual arch is placed, the clinician should consider using headgear or other appliances to obtain a Class I occlusion. The parents should be made aware of such a possibility beforehand.
Of the cases with a 1 mm mesial step, 76% became Class I, 23% became Class II, and 1% became Class III. In cases with a mesial step of 2 mm or more in the primary dentition, 68% became Class I, 13% became Class II, and 19% became Class III. These findings indicated that the greater the mesial step, the greater the probability for the molar relationship to develop into a Class I occlusion. The findings also indicated that a more favorable molar relationship in the primary dentition would lessen the chance for a Class II occlusion developing in the permanent dentition. However, it will not prevent it! In other words, some cases with a mesial step may still develop into Class II molars in the permanent dentition. Such changes will occur in a smaller percentage of cases.48
Finally, the incidence of a Class III molar relationship in the permanent dentition increased as the magnitude of the mesial step increased in the primary dentition. The incidence was 1% with a 1 mm mesial step, and increased to 19% with a mesial step of 2 mm or more.48
Therefore, as clinicians we need to periodically evaluate the changes in the occlusal relationship in our young patients in order to inform the parents of any developing adverse relationships that might require future treatment.
SUMMARY AND CONCLUSIONS
The ability to forecast or predict growth lies at the very heart of contemporary clinical orthodontics. The orthodontist in formulating a treatment plan relies largely upon subjective criteria in conceiving the outcome of treatment. This intuitive perception is necessary, but the overall approach should be based on the available scientific information.
There are several components to be dealt with in the prediction of craniofacial changes: the direction, the magnitude, the timing, the rate of change, and the effects of treatment.32 Orthodontists, in general, are well informed regarding the effects of orthodontic treatment on the patient, but are not yet able to accurately predict the direction, timing, and magnitude of the facial changes that occur with growth in any single individual.
The growth changes in the face are both complex and highly variable as has been demonstrated from the low correlations between the changes in the various facial parameters. The same can be said about the relationship between the changes in the facial parameters to the various indices of skeletal maturation such as standing height and wrist x-rays. Contemporary methods are generally not capable of providing an efficient estimate of individual changes attributable to growth.
The adolescent growth spurt in the mandible occurs in less than 25% of the cases, but the presence, onset, duration, and magnitude of the pubertal growth spurt in facial dimensions cannot be accurately predicted for any single individual. There is substantial mandibular growth occurring during adolescence over a number of years. Therefore, in the presence of significant skeletal discrepancies treatment should not be postponed in anticipation of the elusive “spurt” particularly if treatment is indicated at an earlier age.
The simplest method of predicting changes in facial dimensions is to start with the patient's facial type and add the average growth changes expected for that facial type. Obviously this method has its limitations regarding the prediction of individual changes, but it is as good or as bad as any other more complex method. In regard to the future changes in facial relationships, treatment planning should be based on a “worst case scenario.” In other words, for individuals with unfavorable skeletal relationships, it is wiser to design a treatment plan with the assumption that the same facial growth pattern will be maintained during the treatment period. Favorable growth changes, if they occur, will make the treatment objectives easier to accomplish. Orthodontists should be familiar with the effects of the mechanics used on the facial and dental structures. Therefore, growth projections will require careful attention to the mechanics used.
Citation: The Angle Orthodontist 70, 6; 10.1043/0003-3219(2000)070<0471:FADCIA>2.0.CO;2
Absolute changes in SNA between 5 and 25 years of age for males (A) and females (B) for 3 different facial types: average (AFT), short (SFT), and long (LFT).
Mean incremental changes in: standing height, mandibular length (ArPog), and mandibular relationship (SNPog) for 20 males (A) and 15 females (B) illustrating the yearly changes between 8 and 17 years of age
(A and B) Individual incremental curves for 2 male subjects illustrating the variation in the yearly changes of various parameters between 8 and 17 years of age
Absolute changes in SNPog between 5 and 25 years of age for males (A) and females (B) for 3 different facial types: average (AFT), short (SFT), and long (LFT)