ORIGINAL ARTICLE

Analysis of efficacy of functional appliances

on mandibular growth
Jean Y. Chen, DMD,a Leslie A. Will, DMD, MSD,b and Richard Niederman, DMDc
San Francisco, Calif, and Boston, Mass

The purpose of this study was to examine the hypothesis that functional appliances enhance mandibular
growth in the treatment of skeletal Class II malocclusions. We systematically reviewed previously published
randomized controlled trials. A MEDLINE search strategy for the years 1966 to 1999 was developed and
implemented. Six articles meeting validity standards were evaluated for 12 clinical measures. The following
linear measures were assessed: condylion-pogonion (Co-Pg), articulare-pogonion (Ar-Pg), condylion-
gnathion (Co-Gn), articulare-gnathion (Ar-Gn), sella-gonion (S-Go), articulare-gonion (Ar-Go), and condylion-
gonion (Co-Go). Two angular parameters, sella-nasion-B point (SNB) and lower incisal angle (LIA), were also
measured. Three horizontal measurements were also variably used in the studies examined: gonion-menton
(Go-Me), pogonion to N (Pg to N), and gonion-pogonion (Go-Pg). For Co-Pg, Co-Gn, SNB, LIA, and other
horizontal measurements, we found no significant difference between the untreated control group and the
group treated with functional appliances. However, for Ar-Pg and Ar-Gn, there was a significant difference
between the control and the treated groups. Although these appliances can be used for other purposes,
these results suggest the need to reevaluate functional appliance use for mandibular growth enhancement.
These results complement those of quasi-experimental studies with discriminant analysis but differ from
nonsystematic reviews that provide qualitative summaries. (Am J Orthod Dentofacial Orthop 2002;122:
470-6)

C
lass II malocclusions can result from many
contributing components, both dental and skel-
etal. Although maxillary protrusion and man-
dibular retrusion are both found to be possible causative
factors, it has been reported that the most common
component in a Class II sample population is mandib-
ular retrusion.1 For Class II patients in whom the
mandible is retrognathic, the ideal means of correction
is to target the source and try to alter the amount or
direction of growth in that jaw. The primary treatment
for this purpose is functional appliance therapy. How-
ever, because previous studies have shown varying
degrees of success in the treatment outcomes, func-
tional appliance use remains controversial.2-12 A treat-
ment outcome that has been particularly questioned is
the enhancement of mandibular growth.
It was not until the 1970s that the use of functional
a
Graduate student in orthodontics, University of California, San Francisco.
b
Associate Professor and Chair, Department of Growth and Development,
Harvard School of Dental Medicine.
c
Senior scientist and director, Center for Evidence-Based Dentistry, Forsyth
Institute.
Reprint requests to: Leslie A. Will, DMD, MSD, Associate Professor and
Chair, Department of Growth and Development, Harvard School of Dental
Medicine, 188 Longwood Ave, Boston, MA 02115; e-mail, Lwill@hsdm.
harvard.edu.
Submitted, March 2001; revised and accepted, April 2002.
Copyright 2002 by the American Association of Orthodontists.
0889-5406/2002/$35.00 0 8/1/126730
doi:10.1067/mod.2002.126730
470
appliances became widespread in the United States.
This was, in part, the result of landmark studies in
animals that demonstrated that skeletal changes could
be produced by posturing the mandible forward.13,14
The initial studies seemed to validate the concept that
soft tissue stretching can promote bone growth. Many
studies followed, but later studies performed on hu-
mans were more equivocal and showed less impressive
results.3,7 Therefore, the controversy remains regarding
the efficacy of functional appliances to correct Class II
malocclusions.
It is currently difficult to obtain definitive answers
about appliance efficacy from the literature because of
many inconsistencies in measuring treatment outcome
variables.2 Some investigators use condylion (Co) as
the posterior end point in measuring the overall man-
dibular length, whereas others use articulare (Ar). In
addition, durations of treatment vary, as do the lengths
of follow-up; and treatment groups are sometimes
compared with untreated control groups or with groups
undergoing other forms of treatment, such as head-
gear.2
Therefore, the purpose of this study was to examine
the hypothesis that functional appliances enhance man-
dibular growth for skeletal Class II malocclusions. We
performed a systematic review and analysis of the
reported randomized controlled trial literature, and we
American Journal of Orthodontics and Dentofacial Orthopedics Chen, Will, and Niederman 471
Volume 122, Number 5

Table I. MEDLINE* search strategy

No. Search history Results

1 exp orthodontics/ or exp orthodontics. corrective/ or exp orthodontics, 23393

interceptive/ or exp orthodontics, preventive/ or orthodontics. mp
2 exp orthodontic appliances/ or orthodontic appliances. mp 10502
3 exp malocclusion. angle class ii/ or class ii malocclusions. mp 2120
4 1 and 2 and 3 1064
5 limit 4 to (human and [meta analysis or randomized controlled trial]) 17

*1966 to May 1999 (week 4).

present the clinical and statistical significance of these
findings.
MATERIAL AND METHODS
A MEDLINE search strategy was developed and
used to identify articles that address the effects of
functional appliances on mandibular growth and length
(Table I). Ovid was the search engine for MEDLINE
from 1966 to 1999. We used randomized control trials
and meta analyses as limiting criteria. Items retrieved
from the search were critically appraised for their
validity according to Sackett et al15 (Table II). Articles
were selected for inclusion and analysis if they met 4 of
7 validity standards and then met all of the following
criteria: (1) pertained to functional appliance use in the
early treatment of Class II malocclusions, (2) included
a randomized study, and (3) included measurable man-
dibular cephalometric values. Articles were excluded if
they did not have cephalometric results on mandibular
treatment outcomes or if they did not relate to the
present topic (Table II).
The selected articles examined the efficacy of
various functional appliances. Cephalometric values
were used to assess mandibular growth in the horizontal
and the vertical dimensions. Reference points are
shown and defined in Figure 1. The points were defined
similarly to those described by Nelson et al16 unless
otherwise noted. Growth in the horizontal direction was
evaluated by the following measurements: sella-na-
sion-B point (SNB), condylion-pogonion (Co-Pg), ar-
ticulare-pogonion (Ar-Pg), condylion-gnathion (Co-
Gn), articulare-gnathion (Ar-Gn), gonion-menton
(Go-Me), gonion-pogonion (Go-Po), and horizontal
perpendicular from pogonion to nasion (Pg to N perp
[mm]). Growth in the vertical direction was evalu-
ated by the measurements of sella-gonion (S-Go),
articulare-gonion (Ar-Go), and condylion-gonion
(Co-Go). The change in the lower incisal angle (LIA)
was also examined. The annualized changes in both
the treatment and the control groups were determined
by:

Table II. Randomized control trials and meta analyses identified

Relates to treatment of Contains measureable Mn Selected

References Appliance used Class II malocclusions Randomized study cephalometric values for study

Johnson PD et al, 1998 Bionator Yes Yes No No

Illing HM et al, 1998 Bass, Bionator, Yes Yes Yes Yes
Twin-block
Reukers EA et al, 1998 Fixed appliances Yes Yes No No
Tulloch JF et al, 1998 Bionator Yes Yes No No
Ghafari J et al, 1998 Frankel Yes Yes Yes Yes
Keeling SD et al, 1998 Bionator Yes Yes No No
Cura et al, 1997 Bass Yes Yes Yes Yes
Tulloch et al, 1997 Bionator Yes No No No
Tulloch et al, 1997 Bionator Yes Yes Yes Yes
Courtney et al, 1996 Yes No No
Webster et al, 1996 Frankel & Harvold Yes Yes Yes Yes
Keeling et al, 1995 Bionator No
Dann et al, 1995 Yes Yes No No
Almeida et al, 1995 No No
Ghafari et al, 1994 Yes No
Nelson et al, 1993 Frankel & Harvold Yes Yes Yes Yes
Mills JR, 1991 Yes No No No
472 Chen, Will, and Niederman American Journal of Orthodontics and Dentofacial Orthopedics
November 2002

the inclusion and validity criteria and were selected for

analysis16-21 (Table II).
The articles selected for further analysis are given
in Table III. They are organized according to the
appliances used, the age group studied, the duration of
treatment, and the duration of daily appliance wear. The
age group ranged from 7 to 13 years, with most articles
reporting a mean age of 11 years. Treatment duration
ranged from 6 to 24 months. Duration of daily appli-
ance wear was greater than 14 hours per day in all
cases.
Annualized changes (linear or angular) in specific
cephalometric data were pooled from the selected
articles and then compared and plotted (Figures 2 and
3). Figure 2 and Table IV show the annual linear
mandibular change for both the control and treated
groups. The cephalometric values shown are Co-Pg,
Ar-Pg, S-Go, Ar-Go, Co-Go, Ar-Gn, Co-Gn, and var-
ious other measurements of horizontal growth, such as
Go-Me, Pg to N, and gonion-pogonion (Go-Pg). Of
these, only Ar-Gn (P .003) and Ar-Gn (P .028)
appear to show significant differences between the
control and the treated groups. Figure 3 shows the
annual angular changes for SNB and LIA. The control
Fig 1. Cephalometric reference points. and the treated groups do not appear to be different in
either case.
Ending value Starting value
12
Number of months DISCUSSION

For the annualized changes of each cephalometric
value of the control and treated groups, statistical
significance was tested through an analysis of variance
(ANOVA), the Student t test for paired data, and 95%
confidence intervals with InStat software (GraphPad,
San Diego, Calif).
RESULTS
The MEDLINE search identified 17 randomized
control trial or meta analysis studies. Six of them met
For this systematic review, we limited our MED-
LINE search of the orthodontic literature from 1966 to
1999 to studies performed on humans and written in
English. The search was stratified for randomized
control trials and meta analyses, which are viewed as
providing the highest level of evidence quality.22 Ran-
domized control trials have been recommended as the
standard for comparing alternative treatment approach-
es.20 The randomization incorporated in these studies is
the best approach to eliminating confounding or bias

Table III. Articles selected for inclusion in study

Age group Duration of Duration of daily

Selected references Appliance used Control group studied (y) treatment appliance wear

Illing et al, 1998 Bass, Bionator, Twin-block Untreated 8-15 9 mo Instructed for full-time wear
Ghafari et al, 1998 Frankel Headgear 7-12.5/13 2y 16 h/d
Cura et al, 1997 Bass Untreated Females: 11.86 6 mo 20-22 h/d
Males: 12.1
Tulloch et al, 1997 Bionator Untreated & headgear Mean: 9.4 1.0 15 mo
Webster et al, 1996 Frankel & Harvold Untreated 10-13 18 mo
Mean: 11.6
Nelson et al, 1993 Frankel & Harvold Untreated 10-12.9 18 mo Up to 14 hr by 1 m
Mean: 11.6
American Journal of Orthodontics and Dentofacial Orthopedics Chen, Will, and Niederman 473
Volume 122, Number 5

Fig 2. Box plots comparing annualized rates of mandibular change (mm). Mean values are
represented by square within box plot. 75th percentile (top quartile), 25th percentile (bottom
quartile), and 50th percentile (median) are represented by top line, bottom line, and line through
middle of box, respectively. Graphics suggest, and statistical analysis indicate, that controls and
experimental groups are different only for Ar-Pg and Ar-Gn.

factors, thus allowing more equality in comparing
different treatment groups.20
This approach to clinical trials is very different
from other approaches, such as discriminant analysis
for sample selection.23 It has been argued that random-
ized controlled trials allow bias. Thus, discriminant
analysis might be an alternative method of choosing
comparable cohorts.
Interestingly, randomized controlled trials and dis-
criminant analysis produce similar results. Functional
appliances appear to have little clinical effect on
mandibular length. Obtaining similar results from en-
tirely different experimental approaches is comforting.
In our search of 23,393 orthodontic articles written
in the past 33 years, 155 articles were categorized as
randomized control trials or meta analyses. Although it
is evident that more controlled clinical trials are
needed, the difficulties associated with conducting
474 Chen, Will, and Niederman
Fig 3. Box plots comparing annualized rate of change of SNB and LIA (degrees). Mean values are
represented by square within box plot. Graphics suggest, and statistical analysis verifies, that for
both groups control subjects are not different from experimental subjects.
these studies are recognized. Withholding possible
effective treatment from the control group presents an
ethical problem, especially if growth modification is
limited by time.
When conducting our study, we chose specific
cephalometric measurements to represent mandibular
length. Horizontal and vertical dimensions were con-
sidered. Past articles looked at these dimensions of
growth separately24,25 and reported that mandibular
growth from functional appliance use occurs mainly in
the vertical ramus and not in the horizontal ramus.2
Considering previous studies, we determined that
growth in both dimensions should be examined when
assessing mandibular growth enhancement in response
to appliance use. The values used to represent horizon-
tal growth were Co-Pg, Ar-Pg, Co-Gn, Ar-Gn, Go-Me,
Go-Gn, and Pg to N perp (mm). The values represent-
ing vertical growth were S-Go, Ar-Go, and Co-Go.
After pooling the annualized changes for the vari-
ous cephalometric measurements, statistical analysis
via the paired t test showed a significant difference for
Ar-Pg and Ar-Gn. No other horizontal or vertical
measurements were statistically significant. Mandibular
length can be measured in several ways: from Co to Pg,
Ar to Pg, Co to Gn, or Ar to Gn. It is reported that
values with Ar as an end point, such as Ar-Pg or Ar-Gn,
might give significantly longer measurements, without
a corresponding increase in Co-Pg or Co-Gn.16,26 The
difference between the values with Ar and those with
Co can be the result of the downward and forward
positioning of the condyle after treatment or of any
changes that take place in the glenoid fossa.16,26 The Ar
is a structure determined by the pharyngeal surface of
the cranial base and the posterior part of the condyle;
therefore, its position is determined by the mandible.
After functional appliance therapy, during which the
mandible is positioned down and forward, the Ar can
American Journal of Orthodontics and Dentofacial Orthopedics
November 2002
be found more posteriorly and superiorly on the con-
dyle. This can alter and increase any measurement
taken from this point; therefore, it has been suggested
that Co is the more accurate end point landmark.16 We
can conclude from the statistical analysis that despite
the significant difference of Ar-Pg and Ar-Gn between
the control and treated groups, there is no difference in
overall mandibular change in the horizontal and vertical
direction.
The ANOVA test was performed to ensure that the
appliance type did not play a role in the previous
statistical outcome. The control and treated groups
annualized values were categorized according to appli-
ance type. In both groups, a comparison was made
between appliance types to determine whether there
was a significant difference in annualized changes
(measured in millimeters or degrees). No significant
differences were identified. Therefore, the appliance
type did not affect the outcome of the study.
In Figure 3, SNB and LIA values are plotted.
Statistical analysis showed no difference between the
control and the treated groups for each value. This was
unexpected and might be because all appliances were
analyzed as a group. If they had been examined
separately or categorized according to appliance type, a
difference might have been seen in at least 1 appliance.
This might be examined in a future study.
Past studies have examined whether early-phase
treatment with functional appliances improves the pa-
tients final treatment outcome.27,28 Some have ques-
tioned appliance efficacy and any considerable differ-
ence in treatment outcome when compared with
controls.27,28 Tulloch et al27 concluded that early treat-
ment with a functional appliance followed by compre-
hensive treatment with the second phase of fixed
appliances did not produce a significant difference in
jaw relationships or dental occlusions. It was also
American Journal of Orthodontics and Dentofacial Orthopedics Chen, Will, and Niederman 475
Volume 122, Number 5

Table IV. Statistical analysis of annualized changes reported from selected articles

Control mean Treated mean P value 95% confidence Significant (Y/N)

Co-Pg 2.80 3.34 .130 1.32 to 0.25 N

Ar-Pg 1.66 3.47 .003 2.69 to 0.92 Y
S-Go 2.20 4.26 .237 6.52 to 2.39 N
Ar-Go 0.99 3.94 .414 15.3 to 9.45 N
Co-Go 2.79 2.38 .441 3.91 to 4.73 N
Ar-Gn 1.33 4.09 .028 4.78 to 0.74 Y
Co-Gn 1.73 4.12 .124 6.40 to 1.61 N
Various horizontal values 1.12 2.65 .173 4.28 to 1.21 N
SNB 1.04 1.02 .962 0.93 to 0.97 N
LIA 0.47 1.78 .075 4.80 to 0.30 N

Y, yes; N, no.

determined that although patients spent less time in
fixed appliances after phase-1 treatment, the total treat-
ment time of both phases was longer than if phase 2
were the only treatment provided.27 Therefore, the cost
and the time benefit of functional appliances are also
questionable.
Some variation existed among the subject groups
studied in the articles we selected. The age groups
ranged from age 7 to 13 years (Table III). Although
there was overlap in the ages studied, the age differ-
ences produced some problems when comparing stud-
ies. Growth does not occur at a constant rate, especially
in young children. Even children of the same chrono-
logic age might not have equivalent skeletal maturity or
growth potentials. Therefore, when studies such as
these do not have skeletal age as a common factor, it is
difficult to produce a conclusive statement regarding
the amount of growth modification that might or might
not occur.
In addition, treatment duration differed widely
among the studies, ranging from 6 to 24 months. We
standardized the reported changes by annualizing them
(Table IV). However, if change does not occur uni-
formly during the entire treatment time, this process
can skew the analysis of treatment outcome.
One option, mentioned by Johnston29 in a discus-
sion of Class II growth, is to estimate the growth for
each year from areas under male and female growth
curves. A coefficient for a given year can be derived
from these areas, and the coefficient can be used to take
starting age and treatment time into account when
analyzing treatment effects.
In summary, it is currently difficult to obtain defin-
itive answers about appliance efficacy. For example, it
is recognized that there are many inconsistencies in
measuring treatment outcome variables.2 Some inves-
tigators use Co-Pg as end points in measuring anatomic
features such as overall mandibular length, whereas
others use Ar-Pg. In addition, among studies, treatment
durations vary, different cases are followed for varying
amounts of time, and treatment groups are compared
either with untreated control groups or with those
undergoing other forms of treatment, such as head-
gear.2
In the last 10 years, there have been an increasing
number of studies of Class II treatment.20 However,
there is still a need to conduct more randomized control
trials to reduce the methodologic limitations. Once
confounding variables such as discrepancies in age,
treatment durations, lack of control-matched groups,
patient compliance, and patient accountability are min-
imized, we can identify which patients might or might
not benefit from specific treatment options.
REFERENCES
1. McNamara JA. Components of Class II malocclusion in children
8-10 years of age. Angle Orthod 1981;51:177-202.
2. Mills JRE. The effect of functional appliances on the skeletal
pattern. Br J Orthod 1991;18:267-75.
3. Wieslander L, Lagerstrom L. The effect of activator treatment on
Class II malocclusions. Am J Orthod 1979;75:20-6.
4. Creekmore TD, Rodney LJ. Frankel appliance therapy: orthope-
dic or orthodontic? Am J Orthod 1983;83:89-108.
5. Pancherz H. A cephalometric analysis of skeletal and dental
changes contributing to Class II correction in activator treatment.
Am J Orthod 1984;85:125-34.
6. McNamara JA, Bookstein FL, Shaughnessy TG. Skeletal and
dental changes following functional regulator therapy on Class II
patients. Am J Orthod 1985;88:91-110.
7. Vargervik K, Harvold EP. Response to activator treatment in
Class II malocclusions. Am J Orthod 1985;88:242-51.
8. Pancherz H, Malmgren O, Hagg U, Omblus J, Hansen K. Class
II correction in Herbst and Bass therapy. Eur J Orthod 1989;11:
17-30.
9. McNamara JA, Howe RP, Dischinger TG. A comparison of the
Herbst and Frankel appliances in the treatment of Class II
malocclusion. Am J Orthod Dentofacial Orthop 1990;98:134-44.
10. Hansen K, Pancherz H. Long-term effects of Herbst treatment in
relation to normal growth development: a cephalometric study.
Eur J Orthod 1992;14:285-95.
476 Chen, Will, and Niederman American Journal of Orthodontics and Dentofacial Orthopedics
November 2002

11. Cope JB, Buschang PH, Cope DD, Parker J, Blackwood HO.
Quantitative evaluation of craniofacial changes with Jasper
Jumper therapy. Angle Orthod 1994;64:113-22.
12. Cura N, Sarac M, Osturk Y, Surmeli N. Orthodontic and
orthopedic effects of activator, activator-HG combination, and
Bass appliances: a comparative study. Am J Orthod Dentofacial
Orthop 1996;110:36-45.
13. Proffit WR, Fields HW. Contemporary orthodontics. St Louis:
Mosby; 1993.
14. McNamara JA, Graber LW. Neuromuscular skeletal adaptations
to altered function in the orofacial regions. Am J Orthod
1973;64:578-606.
15. Sackett DL, Richardson NS, Rosenberg W, Haynes RB. Evi-
dence-based medicine. How to practice and teach EBM. New
York: Churchill Livingstone; 1997.
16. Nelson C, Harkness M, Herbison P. Mandibular changes during
functional appliance treatment. Am J Orthod Dentofacial Orthop
1993;104:153-61.
17. Illing HM, Morris DO, Lee RT. A prospective evaluation of
Bass, Bionator and Twin Block appliances. Part IThe hard
tissues. Eur J Orthod 1998;20:501-16.
18. Ghafari J, Shofer FS, Jacobsson-Hunt U, Markowitz DL, Laster
LL. Headgear versus function regulator in the early treatment of
Class II, Division 1 malocclusion: a randomized clinical trial.
Am J Orthod Dentofacial Orthop 1998;113:51-61.
19. Cura N, Sarac M. The effect of treatment with the Bass appliance
on skeletal Class II malocclusions: a cephalometric investigation.
Eur J Orthod 1997;19:691-702.
20. Tulloch JF, Phillips C, Koch G, Proffit WR. The effect of early
intervention on skeletal pattern in class II malocclusions: a
randomized clinical trial. Am J Orthod Dentofacial Orthop
1997;111:391-400.
21. Webster T, Harkness M, Herbison P. Associations between
changes in selected facial dimensions and the outcome of orthodon-
tic treatment. Am J Orthod Dentofacial Orthop 1996;110:46-53.
22. US Department of Health and Human Services, Public Health
Service, Agency for Health Care Policy and Research. Acute
pain management: operative or medical procedure and trauma
(AHCPR Publication 92-0038). Rockville (Md): Agency for
Health Care Policy and Research Publications; 1992.
23. Livieratos FA, Johnston LE. A comparison of one-stage and
two-stage nonextraction alternatives in matched Class II sam-
ples. Am J Orthod Dentofacial Orthop 1995;108:118-31.
24. Derringer K. A cephalometric study to compare the effects of
cervical traction and Andresen therapy in the treatment of Class
II Division 1 malocclusions. Part I-skeletal changes. Br J Orthod
1990;17:33-46.
25. McNamara JA, Bookstein FL, Shaughnessy TG. Skeletal and
dental changes following function regulator therapy in Class II
patients. Am J Orthod 1985;88:91-110.
26. Jakobsson S. Cephalometric evaluation of treatment effect on
Class II, Division 1 malocclusions. Am J Orthod 1967;53:446-
56.
27. Tulloch JFC, Phillips C, Proffit WR. Benefit of early Class II
treatment: progress report of a two-phase randomized clinical
trial. Am J Orthod Dentofacial Orthop 1998;113:63-72.
28. Tulloch JFC, Proffit WR, Phillips C. Influences on the outcome
of early treatment for Class II malocclusion. Am J Orthod
Dentofacial Orthop 1997;111:533-42.
29. Johnston LE. Growth and the Class II patient: rendering unto
Caesar. Semin Orthod 1998;4:58-62.

BOUND VOLUMES AVAILABLE TO SUBSCRIBERS

Bound volumes of the American Journal of Orthodontics and Dentofacial Orthopedics are
available to subscribers (only) for the 2002 issues from the Publisher, at a cost of $96.00
($115.56 Canada and $108.00 international) for Vol. 121 (January-June) and Vol. 122
(July-December). Shipping charges are included. Each bound volume contains subject and
author indexes, and all advertising is removed. The binding is durable buckram, with the
journal name, volume number, and year stamped in gold on the spine. Payment must
accompany all orders. Contact Mosby, Subscription Customer Service, 6277 Sea Harbor Dr,
Orlando, FL 32887; phone 800-654-2452 or 407-345-4000.
Subscriptions must be in force to qualify. Bound volumes are not available in place
of a regular Journal subscription.