associations between changes in selected facial dimensions and the outcome of orthodontic treatment

Associations between changes in selected facial dimensions and the outcome of orthodontic treatment

Trevor Webster, MDS, Michael Harkness, BDS, MSc, PhD, DOrth, and Peter Herbison, MSc" Dunedin, New Zealand

The purpose of this study was to determine, in children with Class II, Division I malocclusion who were treated with functional appliances, the strength of the associations between the changes over 18 months in selected facial dimensions and the success of orthodontic treatment as determined by the weighted Peer Assessment Rating (PAR). Forty-two children, between 10 and 13 years of age (mean age 11.6 years), were randomly assigned to either an untreated group (control) or a group treated with either a Fr&nket function regulator or Harvold activator (treatment). The outcome of treatment was assessed on study models and the craniofacial changes were measured on lateral cephalometric radiographs. Correlation coefficients were then calculated between the differences in the cephalometric variables over 18 months and the differences in the PAR scores. In the treatment group, the effects of normal growth were held constant by partial correlation. The partial used was the change in both stature and weight. Significant positive partial correlations were found between the increases in total anterior face height, posterior face height, S-Pg, and treatment success. Significant negative partial correlations were found between downward movement of the maxilla and mandibular body and lower anterior face height and treatment success. It is postulated that these associations occurred mainly in response to the bite opening by the appliances. Treatment success was also significantly associated with maxillary restriction, an increase in the SNB angle and a reduction in the ANB angle. Change s in B point due to proclination of the mandibular incisors were considered to be responsible for the two latter significant associations. Although mandibular length increased significantly in the treatment group, as compared with the control group, it was not significantly associated with treatment success. (Am J Orthod Dentofac Orthop 1996;110:46-53.)

D e s p i t e the interest shown by orthodontists in the outcome of orthodontic treatment and facial growth, very few studies have correlated treatment outcome with the changes in certain facial dimensions during treatment. In the case of functional appliances, the most recent work has focused on whether these appliances affect facial growth anteroposteriorly. 1-7 However, some authors have pointed out that vertical, rather than horizontal changes, occur in patients with Class II, Division 1 malocclusions treated "successfully" with these appliances. 8-m The extent to which the facial changes, measured from lateral cephalometric radiographs, are due to normal growth and/or the appliances is rarely considered.

If associations between appliance-induced changes in certain facial dimensions and the outcome of treatment are to be determined, a satisfactory method of assessing the latter is essential. Recently, Richmond et

From the Department of Orthodontics, School of Dentistry, University of Otago. Supported by the Medical Research Council of New Zealand. aDepartment of Preventive and Social Medicine, University of Otago. Reprint requests to: Dr. E.M. Harkness, School of Dentistry, P.O. Box 647, Dunedin, New Zealand. Copyright © 1996 by the American Association of Orthodontists. 0889-5406/96/$5.00 + 0 811160741

al, 11'12 developed and validated a method for scoring various occlusal traits on subjects' study casts. The individual scores of each trait are summed, weighted, and the difference between the pretreatment and post- treatment scores is used to indicate the degree of change. Although the index does not take into account the patients' attitudes, facial appearance, or the incli- nation of teeth, it has proved to be a reliable and valid method for quantifying the dental features that make up a malocclusion.

The aim of this study is to determine, in a group of children with Class II, Division 1 malocclusions and treated with functional appliances, the strength of associations between changes in the Peer Assessment Rating (PAR) and changes in certain facial dimensions. Because any changes in facial dimensions may be due to the appliances and/or normal growth, the effects of the latter will be held constant by partial correlation.

MATERIALS AND METHODS Subjects

The subjects were 42 Dunedin schoolchildren with Class II, Division 1 malocclusion 13 who participated in a random- ized control Ilia] 14 of two functional appliances. At the start of the study, the children, who had a mean age of 11.6 years (range 10.0 to 12.8 years), were randomly assigned to one of

46

American Journal of Orthodontics and Dentofacial Orthopedics Webster, Harkness, and Herbison 47 Volume 110, No. 1

Table I. Number, age and gender of the subjects at the start

Mean age

Gender (yr) Range Control FFR HA

(n) (n) (n) Total (n)

Boys 11.3 10.0-i2.8 11 7 7 25

Girls 11.9 10.8-12.8 6 6 5 17

Both 11.6 10.0-12.8 17 I3 12 42

FFR, Friinkel function regulator. HA, Harvold activator,

three groups. Thirteen children (6 boys, 7 girls) were treated with Fr~inkel function regulators/5~16 12 children (7 boys, 5 girls) were treated with Harvold activators, 17 and the remain- ing 17 children (11 boys, 6 girls) were the untreated control group. All the children were observed for 18 months. Addi- tional details of the subjects and appliances have been given by Nelson et al. 1~

There were no statistically significant differences between the boys and the girls at the start, except for age, weight, and maxillary length, and no statistically significant differences between the appliance groups at the start. There- fore, for this investigation, the children in the two appliance groups were combined into a single treatment group and the boys and girls were combined in both the treatment group and the control group.

The experimental design is given in Fig. 1 and details of the subjects in Table I.

Methods

The mean bite opening in the treatment group was 8.45 mm (range 3 to 13 ram) and the mean mandibular advancement was 5.90 mm (range 3 to 10 ram). Lateral cephalometric radiographs, height and weight measurements, and study models were taken at the beginning of the study and at 6 monthly intervals to 18 months. Only measurements taken from the initial and 18-month radiographs and study models are used in this report.

Cephalometric measurements

The reference points and planes shown in Fig. 2 were transferred with the aid of individual templates from the initial and 18-month radiographs to a single sheet of mylar film. The reference planes were based on the anatomically stable structures in the cranial base, maxilla, and mandible. 3'~8-z~

The coordinates of each point on the tracings were digitized three times with a reflex metrograph, 22 and the means converted to the linear and angular measurements given in Table II. Perpendicular measurements to the left of the vertical reference planes and below the horizontal reference planes are reported as negative values. All the radiographs were retraced and redigitized by the same examiner approximately 4 weeks later.

Peer Assessment Rating

The Peer Assessment Rating (PAR) index was used to provide a single summary score for the overall alignment and occlusion of the teeth on the initial and 18-month study

42 Children with Class II, Division I maloeclusions

Control FrSnkel function Harvold activator regulator

• , , /

Ii ] Post-trial comparisons

II

I I Pearson product-moment correlation coefficients

Facial dimensions - Outcome

Partial correlations (stature and weight)

Fig. 1. Experimental design.

models." The difference between the initial and 18-month scores indicated the change in tooth position and occlusion in the control and treatment groups. The PAR index measures the following components: contact point displacements of the teeth in the upper and lower anterior segments, including impacted and ectopic teeth (xl); left and right buccal occlusion in all three planes of space (xl); overjet (x6); overbite (x2), the difference between the maxillary and mandibular dental midlines (x4). The weightings, given in brackets, were determined statistically. 11 The initial and 18-month study models of each subject, with the name and date covered, were set out in a random order and scored by an examiner with the translucent ruler designed for this purpose, tl The assessments were repeated by the same examiner approximately 3 weeks later.

Error of the method

To determine the errors in the radiographic method, the measurements taken on the two occasions were compared with Dahlberg's formula. 23 With this method, the position of the maxillary first molar (U6MCpt horizontal), on the 18- month radiographs, had the largest error (1.28 mm) relative to the overall mean of the two determinations, and the total anterior face height (N-Me), measured on the initial radiographs, had the smallest error (0.30 ram). The reliability of the cephalometric measurements in this study is comparable to those reported by others. ~4'24-27

There were no statistically significant differences be-

48 Webster, Harkness, and Herbison American Journal of Orthodontics and Dentofacial Orthopedics July 1996

Fig. 2. Reference points and planes.

tween the PAR scores obtained on the two occasions or between differences (initial minus 18 months) in the mean PAR scores. The errors in the PAR scores were also determined with Dahlberg's formula and found to be 2.52 (initial PAR score) and 3.09 (18-month PAR score).

To improve the reliability of the cephalometric measurements, the means of both sets of measurements were used. The mean PAR scores were also calculated by averaging the scores obtained on the two occasions, and the differences in the PAR scores were obtained by subtracting the mean 18-month scores from the mean initial scores.

Statistical analysis

The t test for unpaired data was used to compare the measurements in the boys and girls and the measurements in the control and treatment groups. Associations between the PAR differences and the changes in stature, weight, and the cephalometric measurements (18-month minus initial) for both groups were investigated by Pearson product-moment correlation coefficients. Because the relationships between appliance-induced changes in the craniofaciai skeleton and treatment outcome may be wholly or partly because of normal growth, partial correlation was used to control the effects of the latter. The partial used was statural and weight change together.

RESULTS

The results are given in Tables III to VI and Fig. 3.

Initial observations

There were three statistically significant male-female differences. The girls were slightly older (mean

difference 0.61 year, p < 0.024) and heavier (mean difference 4.84 kg, p < 0.010) than the boys, and maxillary length (ANS-PNS) was 2.58 mm longer in the boys than in the girls (p < 0.001). The cephalometric measurements from the boys and girls were combined and the control and treatment groups compared (Table III). There were no statistically significant differences between the control and treatment groups at the beginning of the study.

Treatment comparisons

Twelve statistically significant differences (Co-Pg, N-Me, Me/ANS-PNS, S-Go, S-Pg, Fid(1) vertical, Fid(2) vertical, overjet, overbite, UIA, L6DCT vertical, and the PAR difference) were found between the control and treatment groups after 18 months (Table IV). Mandibular length (Co-Pg) and the total anterior face height (N-Me) increased to a greater extent in the treatment group than in the control group. Almost 2.00 mm of the increase in face height occurred in the lower face (Me/ANS-PNS). In the control group, the mean PAR difference was -1.21 (SD 2.62) and in the treatment group 11.34 (SD 10.41) (Fig. 3).

Associations. Pearson product-moment correlation coefficients and probability values were calculated between the PAR differences and the changes in stature, weight, and 20 craniofacial measurements (Table V). Note that perpendicular measurements to the left of the vertical reference planes and below the horizontal reference planes are reported as negative values. In the control group, no statistically significant associations were found between the PAR difference and the changes in stature, weight, and the cephalometric variables. In the treatment group, however, significant negative associations were found between the PAR difference and the ANB angle, IZ(2) horizontal, Me/ANS-PNS, IZ(1) vertical, IZ(2) vertical, and Fid(1) vertical. There were significant positive correlations between the PAR difference and maxillary length (ANS-PNS) and the total anterior face height (N-Me).

In the treatment group, partial correlations were calculated between the PAR differences and the differences in the cephalometric variables over 18 months after controlling for the changes in stature and weight together (Table VI). Statistically significant partial correlations were found between the PAR difference and 13 variables (SNB, ANB, ANS-PNS, IZ(1) horizontal, IZ(2) horizontal, N-Me, Me/ANS-PNS, S-Go, S-Pg, IZ(1) vertical, IZ(2) vertical, Fid(1) vertical, and Fid(2) vertical). Highly significant associations (p < 0.01) were found for IZ(2) vertical and Fid(2) vertical.

DISCUSSION

The conventional view is that successful treatment of Class II, Division 1 malocclusion in growing chil

American Journal of Orthodontics and Dentofacial Orthopedics Webste~ Harkness, and Herbison 49 Volume 110, No. 1

Table Ih Cephalometric measurements

Anteroposterior SNA SNB ANB ANS-PNS Co-Pg IZ(1) horizontal Fid(1) horizontal Ba horizontal

Vertical N-Me N/ANS-PNS Me/ANS-PNS S-Go S-Pg IZ(1) vertical Fid(i) vertical Ba vertical

Dentoalveolar Over jet Overbite UIA LIA U6MCpt horizontal L6MCpt horizontal U6DCT vertical L6CDT vertical

Reference planes S-SE

IZ(1)-IZ(2)

Fid(1)-Fid(2)

The angle between i, 3 and 5. The angle between 1, 3 and 22. The angle between 5, 3 and 22. Maxillary length. The distance between 4 and 8. Mandibular length. The distance between 20 and 21. The perpendicular distance from 6 to the perpendicular to the cranial base reference plane through 1. The perpendicular distance from 17 to the perpendicular to the cranial base reference plane through 1. The perpendicular distance from 16 to the perpendicular to the cranial base reference plane through l.

Total anterior face height. The distance between 3 and 15. Upper anterior face height. The perpendicular distance from 3 to 4-8. Lower anterior face height. The perpendicular distance from 15 to 4-8. Posterior face height. The distance between 1 mad i9. The distance between 1 and 21. The perpendicular distance from 6 to I-2. The perpendicular distance from 17 to 1-2. The perpendicular distance from 16 to 1-2.

The distance between 10 and 27. The distance between 10 and 26. The internal angle between 9-10 and 6-7. The internal angle between 23-24 and 17-18. The perpendicular distance from 11 to the perpendicular to the maxillary reference plane through 6. The perpendicular distance from 13 to the perpendicular to the mandibular reference plane through 17. The perpendicular distance from 12 to 6-7. The perpendicular distance from 14 to 17-18.

Cranial base reference plane. The line joining sella (1) and the intersection of the sphenoidal plane and the averaged outline of the greater sphenoid wings (2).

Maxillary reference plane. The line joining the lowest point on the average of the right and left outlines of the zygomatic processes (6) and a contructed point (7) two centimeters anterior to 6 along the line parallel to the floor of the nose through 6.

Mandibular reference plane. The line joining a constructed point (17) located within the mandibular symphysis and a second con- slructed point (18) located in the posterior body of the mandible about the region of the developing third molar and/or the supe- rior surface of the mandibular canal.

NOTE: Points 26 and 27 were used to facilitate the calculation of overbite and ovel~et.

dren is primarily associated with the correction of the anteroposterior discrepancy. This is achieved by the differential growth between the maxilla and the mandible and the retraction of the upper incisors behind the lower lip.

Most studies have methodologic weaknesses such as nonrandom allocation of subjects to treatment and control groups or have failed to adjust for the different ages, sex, or treatment times of the participants. 3'2s The experiment reported here endeavored to meet these criticisms by being a prospective study, by random allocation of children to the treatment and nontreat- ment groups and by observing the children for the same length of time. Changes in tooth position and occlusion were assessed with the PAR index and changes in craniofacial dimensions were measured from standardized lateral cephalometric radiographs with structural methods of superimposition. The treatment group comprised two subgroups and although the children in each of these subgroups were treated with different functional appliances (either a Frfinkel function regulator or a Harvold activator), both appliances share the principal objective of functional appliances, namely, to correct the anteroposterior maxillomandibu-

lar relationship by opening the bite and advancing the mandible.

The PAR scores at the start of the study were similar in both treatment and control groups and the values were spread over a slightly smaller range (21 to 48) than those reported by Richmond et al. 12 and Fox 29 for British children (<10 to >50) with Class I, II, and III maloclusions. Although the girls were, on average, older and heavier than the boys at the outset, these differences are comparable to those reported by Tan- ner, Whitehouse, and Takaishi 3° for British children. The male-female differences in maxillary length are also comparable to the data given by Riolo et al. 31 for North American children. There were no significant differences between the treatment and control groups at the outset, which eliminated a source of bias found in many retrospective studies (Table III). 28

In comparison to Frohlich's longitudinal observations 32 of children with untreated Class II malocclusion, there was little change in the alignment and occlusion of the teeth in the subjects in the control group. In the treatment group, however, the majority of subjects improved, although in seven of these subjects, there was little or no change in the alignment and


Table III. Comparison of the control and treatment groups at the start

Control (n = 17) Treatment (n = 25)

Variables Mean SD Mean SD P

Age (yr) 11.47 0.93 11.57 0.84 0.737 Stature (cm) 148.00 6,91 146.14 6.00 0,377 Weight (Kg) 37.22 6.08 37.35 6.24 0,948 Anteroposterior

SNA (°) 80.16 2.88 81.13 3.01 0.303 SNB (°) 74.87 3.19 75.36 3.08 0.625 ANB (°) 5.29 1.44 5.77 2.38 0,458 ANS-PNS (mm) 58.07 2.14 58.33 2.86 0.749 Co-Pg (ram) 109.73 4.43 108,76 4.18 0.477 IZ(1) horizontal (mm) 35.72 5.16 35,87 4.34 0.918 Ba horizontal (ram) -31.78 3.50 -31,45 3.28 0.758

Vertical N-Me (ram) 114.44 6.06 112,95 5.36 0.406 N/ANS-PNS (ram) 51.88 3.68 51.77 2.45 0.907 Me/ANS-PNS (ram) -60.65 3.56 -59.67 4.59 0.463 S-Go (ram) 72.94 3.94 72.64 3.45 0.798 S-Pg (ram) 114,02 4.70 113.34 4.24 0.625 IZ(1) vertical (ram) -48.03 3.16 -45.94 3.51 0.056 Ba vertical (mm) -32.40 5.13 -32.43 3.65 0.984

Dentoalveolar Overjet (ram) 9.33 1.87 9.28 2.18 0.94i Overbite (ram) 3.97 1.77 3.86 3.35 0.907 UIA (°) 114.13 4.64 113.70 6.89 0.823 LIA (°) 95.25 7.94 97.08 6.02 0.402 U6MCpt horizontal (ram) -2.98 3.17 -2.88 2.16 0.908 L6MCpt horizontal (ram) -23.31 2.48 -23.05 2.15 0.718 U6DCT vertical (ram) -20.13 2.43 -20.43 1.84 0.654 L6DCT vertical (ram) 15.38 1.52 15.03 1.34 0.435

Study models PAR score 30.38 4.96 33.36 6.41 0.115

occlusion of the teeth (Fig. 3). Mandibular length (Co-Pg) increased significantly during treatment with functional appliances, and this is in agreement with the findings of Marschner and Harris, ~ McNamara et al., 2 and Righellis 33 (Table IV). Hamilton et al., 4 on the other hand, failed to find any increase in mandibular length after treatment with the Fr~inkel function regulators. Nelson et al., ~4 who used the same subjects as this study, and Harvold and Vargervik 34 also reported that treatment with the Harvold activators had no effect on mandibular length. In this study, however, the larger sample obtained by combining the Fr~inkel function regulator and Harvold activator groups, showed a significantly greater increase in mandibular length (Co- Pg).

Face height increased and the mandibular molars developed vertically to a greater degree in the treatment group than in the control group. Similar findings have been reported by others, z34-36 These increases and the reduction in the overbite are attributed to the bite opening obtained with the appliances.

In agreement with others, 4'~6-4~ the increased overjet in the treatment group was reduced by tipping the

Table IV. Comparison of the control and treatment groups using the differences between measurements at the start of the study and 18 months later

Control Treatment (n = 17) (n = 25)

Mean Variables Mean SD Mean SD difference P

Stature (cm) 8.68 2.70 10.14 3.25 1.46 0.134 Weight (Kg) 6.11 2.70 7.60 3.52 1.49 0.149 Anteroposterior

SNA (°) 0.35 0.73 -0.16 0.77 0.51 0,061 SNB (°) -0.64 0.48 -0.78 0.87 0.14 0.556 ANB (°) -0.36 0.71 -0.88 0.93 0.52 0.056 ANS-PNS (ram) 1.75 1.71 2.06 1.43 0.31 0.525 Co-Pg (ram) 3.53 2.14 5.47 2.47 1.94 0.012 IZ(1) horizontal (ram) 0.80 0.59 0.39 0.79 0.41 0.080 IZ(2) horizontal (ram) 0.84 0.59 0.40 0.82 0.44 0.064 Fid(1) horizontal (ram) 1.33 1.18 0.57 2.00 0.76 0.165 FJd(2) horizontal (ram) 0.75 0.81 0.33 1.41 0.42 0.277 Ba horizontal (mm) -0.72 0.76 -1.25 0.98 0.53 0.070

Vertical

N-Me (ram) 3.30 1.43 5.60 2.47 2.30 0.001 N/ANS-PNS (mm) 1.68 0.89 1.94 1.03 0.26 0.411 Me/ANS-PNS (ram) -1.91 0.99 -3.90 1.94 1.99 0,000 S-Go (ram) 3.19 1.51 4.42 1.63 1.23 0,018 S-Pg (ram) 3.70 1.19 5.87 2.30 2.17 0.001 IZ(I) vertical (ram) -1.10 0.94 -1.29 0.98 0.19 0.537 IZ(2) vertical (ram) -0.91 0.88 -1.31 1.08 0.40 0.209 Fid(1) vertical (mm) -3.57 1.36 -6.04 2.49 2.47 0.001 Fid(2) vertical (ram) -4.29 1.61 -6.40 2.67 2.11 0.006 Ba vertical (ram) -0.89 0.91 -0.96 1.00 0.07 0.825

Dentoalveolar Overjet (mm) 0.21 1.21 -3.65 3.34 3.86 0.000 Overbite (mm) 0.34 0.94 -0.82 2.16 1.16 0,049 UIA (°) 0.94 2.49 -5.23 5.62 6.24 0.000 LIA (°) 0.20 2.69 2.39 4.65 2.19 0.088 U6MCpt horizontal (nun) 0.80 1.09 0.73 0.95 0.15 0.597 L6MCpthorizontal(mm) -0.64 0.69 -0.52 1.02 0.12 0.692 U6DCT vertical (ram) -2.03 1.19 -2.42 1.42 0.39 0,362 L6DCT vertical (ram) 1.03 0.90 2.53 1.50 1.50 0,001

Study models PAR difference -1.21 2.62 11.34 10.41 12.55 0,000

Significant values are in bold type.

maxillary incisors palatally. Luder, 4° Chang e t al . , 36

and Nelson et al. 14 have reported that the mandibular incisors are proclined during treatment with certain functional appliances but, in this study, no statistically significant differences were found.

There were no statistically significant associations between changes in the occlusion, as determined by the PAR differences, and the changes in stature, weight, and the facial dimensions in the control group. The only measurement to approach statistical significance was the horizontal movement of the mandibular symphysis, relative to the cranial reference plane (Table V). In the treatment group, increases in face height (N-Me, Me/ANS-PNS, IZ(1) vertical, IZ(2) vertical, Fid(1) vertical), reduction in the ANB angle, increase in the length of the maxilla (ANS-PNS), and horizontal restriction of the maxilla (IZ(2) horizontal) were asso-


Z ji -i0

Control

1'0 20 3'0 4b

Z :f io 0 10 20

PAR difference

Treatment

3'0 4b

Fig. 3. PAR differences for control and treatment groups.

Table V. Correlation coefficients expressing the relation between the changes (18 months - initial) in selected craniofacial dimensions and treatment outcome, as determined by the differences in PAR scores (initial - 18 months)

n = 25)

Table Vl. Treatment group (n = 25). Partial correlations between the changes (18 months- initial) in selected craniofacial dimensions and the PAR difference (init ial- 18 months). Controlling for changes in stature and weight together

Variables p Variable

Stature (cm) 0.375 0.I38 0.128 0.541 Weight (Kg) 0.182 0.483 0.080 0.704 Anteroposterior

SNA (°) -0.116 0.657 -0 . I94 0.352 SNB (°) 0.174 0.504 0.376 0.064 ANB (o) -0.236 0.363 -0 .404 0,045 ANS-PNS (ram) -0.067 0.798 0.463 0,020 Co-Pg (ram) 0.101 0.700 0.177 0,397 IZ(1) horizontal (mm) -0.088 0.737 -0.353 0,083 IZ(2) horizontal (ram) -0.084 0.750 -0 .436 0,029 Fid(1) horizontal (n-an) 0.465 0.060 0.163 0.436 Fid(2) horizontal (ram) 0.330 0.196 0.201 0.330 Ba horizontal (ram) -0.240 0.353 0.105 0.618

Vertical N-Me (ram) 0.15I 0.562 0.415 0.039 N/ANS-PNS (man) -0.100 0.701 0.292 0.157 Me/ANS-PNS (ram) -0.224 0.388 -0 .434 0,030 S-Go (mm) 0.321 0.209 0.335 0.101 S-Pg (ram) 0.304 0.235 0.392 0.053 IZ(1) vertical (mm) 0.154 0.555 -0 .403 0.046 IZ(2) vertical (nma) 0.243 0.347 -0 .545 0.005 Fid(1) vertical (mm) -0.100 0.703 -0 .438 0.028 Fid(2) vertical (ram) -0.260 0.313 -0.384 0.058 Ba vertical (ram) -0.206 0.427 -0.132 0.528

Significant values are in bold type.

ciated significantly with improved treatment outcomes, although the coefficients were generally low (r = -0 .40 to r = - 0 . 5 4 ) . Because changes in the craniofacial variables may be due to the appliances and to normal

Statural and weight change (df = 21)

Anteroposterior SNA (°) -0.198 0.182 SNB (°) 0.385 0.035 ANB (°) -0 .389 0.038 ANS-PNS (ram) 0.470 0.012 Co-Pg (ram) 0.138 0.265 IZ(1) horizontal (nun) -0 .377 0.038 IZ(2) horizontal (ram) -0 .470 0.012 Fid(1) horizontal (ram) 0.180 0.205 Fid(2) horizontal (mm) 0.240 0.135 Ba horizontal (ram) 0.256 0.119

Vertical N-Me (ram) 0,421 0.023 N/ANS-PNS (ram) 0.272 0.104 Me/ANS-PNS (mm) -0 .438 0.018 S-Go (ram) 0.421 0.023 S-Pg (ram) 0.441 0.018 IZ(1) vertical (ram) -0 .385 0.035 IZ(2) vertical (turn) -0 .558 0.003 Fid(1) vertical (ram -0 .471 0.012 Fid(2) vertical (ram) -0 .499 0.008 Ba vertical (ram) -0.115 0.300

df= Degrees of freedom. Significant values are in bold type.

growth, we attempted to hold the effects of the latter constant by partial correlation. Statural and weight changes together were used as the partials because various investigators have shown that changes in stature and weight are closely related to the changes in certain craniofacial dimensions in untreated chil-


dren . 42-48 For example, Brown et al.46 reported highly significant correlations between the age of the peak growth velocity in stature and the age of the peak growth in mandibular length (Ar-Pg: boys, r = 0.78; girls, r = 0.79) and the peak growth velocity in total anterior face height (N-Gn: boys, r=0.83; girls, r = 0.84). Pike 45 also found correlations between the growth rates in stature and mandibular length (Co-Pg: boys, r=0.58; girls, r=0 .80) and stature and total anterior face height (N-Me: boys, r=0.57; girls, r = 0.21).

When statural and weight changes together were controlled, the coefficients for each variable were generally of similar strengths (Table VI). The majority of the significant associations between the changes in craniofacial measurements and the success of treatment were in the vertical rather than the horizontal dimension. This is similar to the results of Hixon and Klein 49 who reported extremely low correlations between anteroposterior jaw growth and the change in molar relationship or the change in overjet (r = 0.1 to 0.3). Steiner's concept ~° that good jaw growth contrib- utes to the successful outcome of treatment is therefore in some doubt, especially in relationship to the anteroposterior changes. Changes in mandibular length were not significantly associated with treatment success, despite the significant increase in mandibular length in the treatment group, as compared with the control group. This is consistent with Nielsen 1° who reported that in patients treated with Frfinkel appliances, the improvement in the occlusion horizontally was due more to vertical rather than horizontal growth changes. Ahlgren and Laurin 8 examined 50 consecutively treated patients comparing the dentofacial form between successfully and unsuccessfully treated cases. Their findings indicate that activators work primarily at the dentoalveolar level by retroclining the maxillary incisors and proclining the mandibular incisors. This is assisted by the retardation of forward growth in the maxilla and by a significant increase in the lower face height. This last point is also highlighted by Cohen 9 who found that those cases showing more complete overjet reduction also tended to have a greater increase in facial height and showed a faster rate of growth in facial height, when compared with the less successfully treated cases.

Improved treatment outcomes were also significantly associated with appliance-induced restriction of the maxilla horizontally. Johnston 3 reported that approximately 0.5 mm of the total molar correction in Class II, Division I malocclusions treated with functional appliances was due to the restriction of the maxilla in a horizontal direction. In this study, treatment was more successful when the SNB angle in-

creased and the ANB angle reduced. Bearing in mind the size of the ANB difference (less than 1 °) and the magnitude of any errors in the method of measurement these findings, which have been shown to result from the changes in the position of B point, 51 are attributed primarily to the proclination of the mandibular incisors and secondarily to the increase in mandibular length.

The significant positive association between the outcome of treatment and the changes in maxillary length (ANS-PNS) indicates that treatment is more successful in subjects with long maxillary dental bases. One explanation for this finding is that subjects with spaced maxillary dental arches before treatment had less incisor crowding after retroclination of the maxillary incisors.

The significant findings of this study hinge on the validity of the measurement of treatment success. The PAR index was created as a measure to test the success of orthodontic treatment in the British National Health system. Although it has shown validity and reliability, there remains an area of doubt as to the degree that the individual components contributing to the summed score are weighted. In a separate, unpublished, study, the models used in this study were scored by six orthodontists who used a visual analogue scale and correlated with the PAR scores used in this study. The strength of the correlation coefficients obtained (between 0.72 and 0.86) confirmed that the British weightings could be used in this study. Finally, it should be emphasized that the significant associations between the craniofacial variables and changes in the PAR index could have been caused by some unknown third factor. It should not be assumed that because two variables are correlated, there is a causal relationship between them.

This study provides evidence that after treatment of Class II, Division I malocclusion with functional appliances, improvement in the occlusion is principally related to vertical changes in the facial skeleton rather than an increase in mandibular length. Successful treatment was also associated with an increase in the SNB angle, a reduction in the ANB angle, and restriction of the forward movement of the maxilla by the appliances.

SUMMARY AND CONCLUSIONS

The aim of this study was to determine the strength of the associations between the success of orthodontic treatment, as determined by the weighted Peer Assessment Rat- ing, and changes in selected craniofacial dimensions after 18 months of treatment with functional appliances.

Partial correlations were calculated between treatment success and 20 craniofacial variables after controlling for both statural and weight changes.

Significant partial correlations were found between the


increases in total anterior face height, lower anterior face height, posterior face height, S-Pg, downward movement of the maxilla and mandibular body, and treatment success. Treatment success was also significantly associated with maxillary restriction, reduction in the ANB angle, maxillary length but not mandibular length. The latter increased significantly in the treatment gronp as compared with the control group.

It is postulated that these associations occurred mainly in response to the bite opening by the appliances. Mandibular advancement and the use of the mandibular arch for anchor- age were considered to be responsible for the significant associations between treatment success and maxillary restriction and changes in B point because of proclination of the mandibular incisors.

REFERENCES

1. Marschner JE Harris JE. Mandibular growth and Class II treatment. Angle Orthod I966;36:89-93.

2. McNamara JA, Bookstein FL, Shaugbnessy TG. Skeletal and dental changes following functional regulator therapy on Class II patients. Am J Orthod 1985;88:91-110.

3. Johnston LE. A comparative analysis of Class II treatments. In: Vig PS, Ribbens KA. Science and clinical judgement in orthodontics. Monograph 19. Craniofacial Growth Series. Ann Arbor: Center for Human Growth and Development, University of Michigan, 1986:103-48.

4. Hamilton SD, Sinclair PM, Hamilton RH. A cephalometric, tomographic, and dental cast evaluation of Frankei therapy. Am J Orthod Dentofac Orthop 1987;92:427-34.

5. Falck F, Friinkel R. Clinical reievance of step-by-step mandibular advancement in the treatment of mandibular retmsion usiug the Frankel appliance. Am J Orthod Dentofac Orthop 1989;96:333-41.

6. Mamandras AH, Allen LE Mandibular response to orthodontic treatment with the bionator appliance. Am J Orthod Dentofac Orthop 1990;97:113-20.

7. Mills JRE. The effect of functional appliances on the skeletal panern. Br J Orthod 1991;18:267-75.

8. Ahlgren J, Laurin C. Late results of activator treatment: a cephalometric study. Br J Orthod 1976;3:181-7.

9. Cohen AM. A study of Class II division 1 malocchisions treated by the Andresen appliance. Br J Orthod 1981 ;8:159-63.

10. Nielsen IL. Facial growth during treatment with the function regulator appliance. Am J Orthod 1984;85:401-10.

11. Richmond S, Shaw WC, O'Brien KD, et al. The development of the PAR Index (Peer Assessment Rating): reliability and validity. Er r J Orthod 1992; l 4:125-39.

12. Richmond S, Shaw WC, Roberts CT, Andrews M. The PAR Index (Peer Assessment Rating): methods to determine outcome of orthodontic treatment in terms of improvement and standards. Eur J Orthod 1992;14:180-7.

13. British Standards Institution 1983 BS4492, British Standard Glossary of Terms Relating to Dentistry. London: British Standards Institution, 1983.

I4. Nelson C, Harkness M, Herbison E Mandibular changes during functional appliance treatment. Am J Orthod Dentofac Orthop 1993;104:153-61.

15. Eirew L, McDowelI E Phillips JG. The Fraenkel appliance-avoidance of lower incisor proclination. Br J Orthod I981;8:189-91.

16. McNamara JA, Huge SA. The FrankeI appliance (FR-2): model preparation and appliance construction. Am J Orthod 1981;80:478-95.

17. l-Iarvold EP. The activator in interceptive orthodontics. St Louis: CV Mosby, 1974. 18. Bjork A. Prediction of mandibular growth rotation. Am J Orthod 1969;55:585-99. 19. Me/sen B. The cranial base: the postnatal development of the cranial base studied

histologically on human autopsy material. Acta Odont Scand 1974;32(Suppl 62):1-i26. 20. Bjork A, Skieller V. Growth of the maxilla in three dimensions as revealed radiographi-

cally by the implant method. Br J Orthod 1977;4:53-64. 21. Bjork A, Skieller V. Normal and abnormal growth of the mandible: a synthesis of

longitudinal cephalometric implant studies over a period of 25 years. Eur J Orthod

1983;5:1-46. 22. Scott PJ. The reflex pIotters: measurement without photographs. Photogram Rec

1981;10:435-46. 23. Dahlberg G. Statistical methods for medical and biological students. New York:

Interscience Publications, 1940. 24. Jakobsson SO. Cephalometric evaluation of treatment effect on Class U, Division 1

malocchisions. Am J Orthod 1967;53:446-57. 25. Cook PA, Gravely JE Tracing error with Bjork's mandibular structures. Angle Orthod

1988;58:169-78. 26. Derringer K. A cephalomettic study to compare the effects of cervical traction and

Andresan therapy in the treatment of Class lI division 1 malocclusion. Part 1--skeletal changes. Br J Orthod 1990;17:33-46.

27. Jakobsson SO, Paulin G. The influence of activator treatment on skeletal growth in Angle Class II: 1 cases: a roentgenocephalometric study. Eur J Or'hod 1990;I2:174- 84.

28. Tulloch JFC, Medland W, Tuncay OC. Methods used to evaluate growth modification in Class II malocchision. Am J Orthod Dentofac Orthop 1990;98:340-7.

29. Fox NA. The first 100 cases: a personal audit of orthodontic treatment assessed by the PAR (peer assessment rating) index. Br Dent J 1993;174:290-7.

30. Tanner JM, Whitehouse RH, Takaishi M. Standards from birth to maturity for height, weight, height velocity, and weight velocity: British children, 1965. Part II. Arch Dis Child 1966;41:613-35.

31. Riolo ML, Moyers RE, McNamara JA, Hunter WS. An atlas of craniofacial growth. Ann Arbor: Center for Human Growth and Development, University of Michigan, 1974.

32. Frrhlicb FJ. A longitudinal study of untreated Class II type rnalocchisions. Trans Eur Orthod Soc 1961;1-23.

33. Righellis EG. Treatment effects of Frankel, activator and extraoral traction appliances. Angle Orthod 1983;53:107-21.

34. Harvold EP, Vargervik K. Morphogenetic response to activator treatment. Am J Orthod 1971;60:478-90.

35. Vargervik K, Harvdid EP. Response to activator treatment in Class II malocchisions. Am J Orthod 1985;88:242-51.

36. Chang H, Wu K, Cheu K, Chang M. Effects of activator treatment on ciass II division 1 malocclusion. J Clin Orthod 1989;23:560-3.

37. Trayfoot J, Richardson A. Angle Class II division 1 malocchisions treated by the Andresen method. Br Dent J 1968;124:516-9.

38. Reey RW, Eastwood A. The passive activator: case selection, treatment response, and corrective mechanics. Am J Orthod 1978;73:378-409.

39. Wieslander L, Lagerstrom L. The effect of activator treatment on Class n mahicchi- situs. Am J Orthod 1979;75:20-6.

40. Luder HU. Effects of activator treatment--evidence for the occurrence of two different types of reaction. Eur J Orthod 1981;3:205-22.

41. Derringer K. A cephalometric study to compare the effects of cervical traction and Andresen therapy in the treatment of Class II division 1 malocclusion: part 2 ~ dentoalveolar changes. Br J Orthod 1990;17:89-99.

42. Nanda RS. The rates of growth of several facial components measured from serial cephalometfic roentgenograms. Am J Orthod 1955;41:658-73.

43. Singh IJ, Savara BS, Miller PA. Interrelations of selected measurements of the face and body in pre-adolescent and adolescent girls. Growth 1967;31:119-31.

44. Fukuhara T, Matsumoto M. A longitudinal study of facial growth in relation to general body height during adolescence. Bull Tokyo Med Dent Univ 1968;15:161-70.

45. Pike JB. A serial investigation of facial and statural growth in seven to twelve year old children. Angle Orthod I968;38:63-73.

46. Brown T, Barrett MJ, Grave KC. Facial growth and skeletal maturation at adolescence Tandlaegebladet 1971;75:1211-22.

47. Bishara SE, Jamison JE, Puterson LC, DeKock WH. Longitudinal changes in standing height and mandibular parameters between the ages of 8 and 17 years. Am J Orthod 1981 ;80:115-35.

48. Lewis AB, Roche AF, Wagner B. Pubertal spurts in cranial base and mandible: comparisons within individuals. Angle Orthod 1985;55:17-30.

49. Hixon E, Klein E Simplified mechanics: a means of treatment based on available scientific information. Am J Orthod 1972;62:113-41.

50. Steiner CC. The use of cephalometrias as an aid to planning and assessing orthodontic treatment. Am J Orthod 1960;46:721-35.

51. Huldaway RA. Changes in relationship of points A and B during orthodontic treatment. Am J Orthod 1956;42:176-93.

associations between changes in selected facial dimensions and the outcome of orthodontic treatment

Documents