VIEWPOINT

Angle classification revisited 2: A modified

Angle classification
Morton I. Katz, DDS*
Washhlgton, D.C.

Edward Angle, in his classification of malocclusions, appears to have made Class I a range of
abnormality, not a point of ideal occlusion. Current goals of orthodontic treatment, however, strive for
the designation "Class I occlusion" to be synonymous with the point of ideal intermeshing and not a
broad range. If contemporary orthodontists are to continue to use Class I as a goal, then it is
appropriate that Dr. Angle's century-old classification, be modified to be more precise. (AM J ORTHOD
DENTOFAC ORTHOP 1992;102:277-84.)

H a v i n g observed inconsistency in the ap- everyone "speak the same language." No one has dif-
plication of the Angle classification of malocclusion as ficulty classifying the extremes of a full Class II and a
a postgraduate orthodontic teacher, I developed a ques- full Class III. However, the gray area of Class I is
tionnaire to test whether the observed inconsistency is subject to much confusion and inconsistency. Second,
merely anecdotal or whether orthodontists across the classification has a significant effect on patient treat-
spectrum of ages, training, and geographic location ment. Once a patient is classified, the practitioner will
truly have problems applying the Angle classification. tend, almost unconsciously, to apply treatment me-
The questionnaire presented 10 buccal views of study chanics appropriate to that classification. Even though
models: three ideal interdigitations and seven with vary- model analysis, especially buccal interdigitation, is but
ing degrees o f Class II tendency (two of which were a small part of the complete orthodontic diagnosis, the
mutilated). A sample was selected of 347 orthodontists decision to use Class II or Class III mechanics invari-
(approximately 5% of active practitioners listed in the ably is strongly influenced by the perceived Angle clas-
1987 American Association of Orthodontists list of sification of the patient. Third, if one of the goals in
members) selected from every state and distributed the treatment of a malocclusion is to achieve Class I,
among orthodontists trained in the 1960s and before, there must be a consensus among orthodontists as to
the 1970s, and the 1980s, as well as all chairpersons what constitutes ideal occlusion, and Class I must be
of departments of dental school orthodontic programs. redefined to agree with the prototype standard. If every
The 77.8% response rate on the single mailing dem- orthodontist has a different idea of what ideal buccal
onstrated an intense interest in the subject of classifi- interdigitation means, then the dental specialty of ortho-
cation reliability. Unfortunately, the statistics showed dontics has no standardized method by which to eval-
a n extremely p o o r reliability of Angle classification uate successful or unsuccessful treatment. Angle de-
when applied by contemporary orthodontists. This veloped his classification 100 years ago to eliminate
study illustrated that a problem exists with the current the anarchy that existed in the specialty. Considering
use of Angle's valuable classification. ~ I have further the current level of disagreement as to what constitutes
attempted to supplement the Angle method to address Class I, and what describes ideal occlusion, is the Angle
the weaknesses observed, while maintaining the classification inherently faulty, or are contemporary
strengths of the Angle method. orthodontists not properly applying it, as designed by
Why is the classification of malocclusion important? Dr. Angle? Rather than supplanting Angle, supple-
First, classification is an essential communication tool menting the Angle classification might make it more
between dental school professor and student, between descriptive and precise.
practitioners, and between practitioner and insurance In an era of piezo-electric and magnet-accelerated
company or government bureaucracy. It is essential that tooth movement, titanium alloy wires, esthetic ortho-
dontic appliances, state of the art magnetic and x-ray
imaging, and diagnosis and treatment planning by com-
puters, it seems anachronistic to communicate with a
*Associate Professor, Postgraduate Clinical Director, Department of Ortho-
dontics. Howard University College of Dentistry, Washington. D.C.
classification scheme that has not been significantly
8/1/29405 modified or upgraded in 83 years. Our respect for tra-
277
278 Katz Am. J. Orthod. Dentofac. Orthop.
September 1992

Fig. 1. Angle's prototype ideal occlusion. All teeth (except lower centrals and upper third molars) have
two antagonists. Note proper mesial tilt of the upper first molar.

occlusion every tooth (except the lower centrals and

upper third molars) should have t w o antagonists (Fig.
1). In other words, even if a patient has the mesiobuccal
cusp of the upper first molar fitting perfectly into the
lower molar buccal groove (Fig. 2), the patient does
n o t possess p r o p e r occlusion according to Angle, 24 and

}- later Strang, 5 Stoller 6 and Andrews, 7 unless the upper

first molar also has a mesial crown tilt that allows the
distal incline of the distal cusp of the upper first molar
to occlude with the mesial incline of the mesial cusp
of the lower second molar.
The practitioner must, therefore, look at more than
Fig. 2. Traditional Angle Class I molar (upper mesiobuccal cusp first molars. Proper cuspal incline contacts of a l l teeth
in lower molar groove). However, teeth have only o n e antag- should be noted. Angle emphasized the importance of
onist. Not ideal Angle intermeshing because upper molar is too each premolar and canine contacting two occluding
upright, forcing an anterior displacement of the upper premolars, teeth. An occlusion where the first molars classically
canines, and incisors. fit the criteria of the upper mesiobuccal cusp to lower
molar groove, but the premolars and canine contact only
one opponent tooth each (Fig. 2), would be considered
dition must be balanced with the need to define more Class I by Angle (because Class I is a premolar-width
precisely our patients' occlusal problems. range of abnormality). However, Angle would not have
considered the occlusion as having met his standards
DISCUSSION for "ideal" occlusion of a well-treated case. Therefore
It is all too easy to dismiss the Angle classification all "ideal" occlusions are Class I, but not all Class I
as an archaic remnant of orthodontic past and develop occlusions are "ideal."
a complex, multifactorial classification to represent the Orthodontists are goal oriented. Can contemporary
Cutting edge of current thinking. If we can accept that orthodontists make attainment of Class I a goal of treat-
Dr. Angle, one of the specialty's most seminal thinkers, ment? The answer is no, after one reviews Angle's
put as much effort into his classification as he did his definitions. The original classification by Angle, 2"3 had
other innovations, then it would be appropriate to re- Class II as a full premolar-width distoclusion and Class
read his writings before judging him too harshly. III as a full premolar-width mesioclusion. Assuming an
Every dental student learns the Angle "mesiobuccal average premolar width of 7.5 mm, then Class I ranged
cusp of the upper first molar fits into the buccal groove from 7 mm mesioclusion to 7 mm distoclusion, for a
of the lower first molar" truism by rote. However, since total range of Class I of 14 mm. This range was far
few ever open the now yellowed pages of Angle's too broad, and so in 1907, Angle 4 revised his definition,
articles and books, many miss his complete "occlusion making Class II more than half of a cusp distoclusion
concepts. Angle 24 described in minute detail each con- and Class III more than half of a cusp mesioclusion.
tacting cusp incline to prove his point that in ideal Angle's modification reduced the range from 14 mm to
Voh,me102 --U- i e w n or ; n t - 279
Number 3

a 7 mm range. However, 7 mm is still too broad a range cepted). Modern orthodontists advance molars in ex-
to act as a treatment goal if an orthodontist is to treat traction treatments or distalize molars in nonextraction
with precision. treatments with little concern for the immutable rela-
It is my contention that Angle never intended his tionship of the upper first molar to the bony landmarks,
classification to depict Class I as a treatment goal or such as the key ridge, as promulgated by Angle.
"ideal," but as a range of abnormality between the In addition, Friel 9 and later Arya et al. ~~ demon-
extremes of full Class II and Class III. Angle would strated the dynamic nature of the position of the first
not have described, in the most minute detail, the pre- molar in the mixed dentition, changing as the occlusion
cise intermeshing of every cuspal incline and then matures into the permanent dentition because of jaw
hypocritically imply that teeth can be arranged almost growth and second deciduous molar leeway space con-
cusp tip to cusp tip and be acceptable. Class I is not a siderations. This natural "adjustment" makes a molar-
goal or an "ideal," as Angle envisioned it, but current defined classification awkward in young patients.
orthodontists speak with misguided pride of having
achieved the treatment goal of Class I. Contemporary Using CANINES to evaluate ideal buccal occlusion
orthodontists should become educated as to Angle's Maxillary canines are among the most stable of
complete occlusion criteria, in addition to first molar dental units because they are the longest rooted of all
relation. Also, some supplemental information should teeth and therefore very well anchored to the alveolar
be added to Angle's classification to better quantify bone. The canine is the "keystone" tooth in the dental
deviation from Angle's description of proper occlusal arch, and like the keystone of a stone archway, it pro-
intermeshing. vides a buttressing support for the incisors, as well as
the posterior teeth. Also, canines provide a vital pro-
EXPLORING THE POSSIBILITIES tective function in lateral excursive movements.
Using MOLARS to evaluate ideal buccal occlusion Historically, Simon 'L'2 attempted a canine-focused
Angle 24 considered the upper first molar as the most classification. His Law of the Canine considered the
reliable point of reference from which to compare other orbital plane (a line drawn from orbitale perpendicular
teeth because of its constancy in taking a correct po- to Frankfort horizontal) as coincident with the distal
sition relative to the bony skeleton's jugal buttress (also third of the maxillary canine in ideal occlusion. While
known as the key ridge). The key ridge is a ridge of modern orthodontists no longer consider Simon's law
bone descending downward and forward from th e zy- valid, the strategic position occupied by the canine
goma, which is the anterior edge of the jugal or molar makes it a favored tooth to reference for classification.
buttress of the maxillary bone and which marks the However, the principal objection to a canine-derived
union of the anterior and the posterior buccal walls of classification relates to tooth anatomy. The maxillary
the antrum of Highmore. Contemporary orthodontists, canine exhibits a mesial incisal ridge that is shorter and
however, do not consider the anatomic interrelationship less severely sloped than its distal incisal ridge. As a
of the upper molar to the cranium as significant. result, the central axis of the maxillary canine does not
Angle 8 acknowledged that the first molar might bisect the cusp tip. ~3 Tooth sizes and shapes vary, but
erupt in an altered position when influenced by the the cusp tip averages 1 to 1.5 mm mesial to the center
malpositions of other teeth or the loss or nondevelop- axis. Therefore the cusp tip of the maxillary canine
ment of deciduous and permanent teeth anterior to the does not directly fit into the embrasure formed b y t h e
first molar. Therefore Angle recommended visualizing mandibular canine and the first premolar, but rides up
the upper first molar into its proper position relative to on the distal slope of the mandibular canine (Fig. 3).
the jugal buttress before classifying the malocclusion. Also, the cusp tip of the maxillary canine does not work
There are two problems with this concept. First, vi- well as a landmark because occlusal wear frequently
sualizing the "correct" position of the upper first molar alters the cusp tip from a point to a flat facet, and the
to the jugal buttress and lining up the remaining dental modified architecture of its incisal edge obscures the
units relative to it is a very subjective pursuit. It is quite true cuspal form. Although not ideal, one could use the
probable that no two orthodontists would exactly vi- imaginary center axis of the maxillary canine as a ref-
sualize the same "correct" position. And second, mod- erence point, since it lines up with the mandibular ca-
em orthodontists are more concerned with the proper nine-first premolar embrasure (Fig. 3).
position of the incisors relative to the profile for esthe!!c Another objection to relying on the canine for clas-
and stability concerns and are willing to adjust first sification rests with the time of its eruption. The max-
molar position and even sacrifice teeth to better align illary canine is one of the last teeth to erupt (other than
the incisors (concepts Angle would never have ac- third molars). This holds up classification efforts until
280 Katz Am. J. Orthod. Dentofac. Orthop.
September 1992

y 7
Fig.4. One upper premolar (t3) opposing two lower premolars
Fig. 3. Central axis of upper canines bisects distal embrasure (B) in modified Class I. Note traditional Angle only classifies
of lower canines in ideal Class I. The cusp tip of upper canine, molars and would call this excellent occlusion a Class IH
however, is mesial to the axis.

the patient is 12 years, or older in slowly erupting to redefine Class I for the purpose of this modified Angle
patients. The deciduous canine offers little assistance classification as the pohu of ideal intermeshing.
with classification since it is smaller in mesiodistal Angle 2'3 described in detail the "ideal" occlusal rela-
width than its permanent successor, resulting in a center tionship, and his conception of ideal occlusion has with-
axis that is not coincident with the center axis of its stood the test of time and must be retained. However,
future permanent replacement. the large 7 mm range of Class I has been discarded in
this modified version, and all the teeth visible from a
Using PREMOLARS to evaluate ideal buccal buccal view (except the upper third molars) must oc-
occlusion clude with two antagonists as Angle demanded for an
Favorably, premolars usually present a sharply de- ideal occlusion (Fig. 1). By defining Class I more nar-
fined cusp tip, which is centered on the central axis of rowly, the designation "Class I" will match the occlu-
the premolar crown and which fits precisely into the sion goal modem orthodontists aspire to achieve.
opposing embrasure. Also, the cuspal inclines are
steeper and deeper than molar cusps, which makes a A premolar-derived classification
more positive fit. After considering the advantages and disadvantages
From the negative perspective, orthodontists tra- of the possible choices, o n l y the premolar occlusion
ditionally have not had high regard for premolars as offers sufficient precision of intermeshing to be a valid
functional dental units and have selected premolars standard.
most often of all tooth types for sacrifice in an extraction Class I, in the modified Angle classification is de-
treatment. Also, premolars may have anomalous tooth fined: The most anterior upper premolar fits exactly
size or shape. Furthermore, some judgment is required into the embrasure created by the distal contact of the
when less than a full complement of premolars are most anterior lower premolar. This definition applies
present. However, orthodontists have always been whether a full complement of premolars are present,
forced to contend with the frequently missing first mo- whether one upper premolar opposes two lower pre-
lar, so this problem is not unique to premolars. molars (Fig. 4), whether two upper premolars oppose
one lower premolar (Fig. 5), or whether only one pre-
A MODIFIED ANGLE CLASSIFICATION molar is present in each quadrant (Fig. 6).
A goal-directed classification When this relationship is achieved, the canines will
The Class I designation was used by Angle to de- also relate correctly, as will the incisors. We are un-
scribe a range of abnormality (malocclusions in the concerned about molar relation. The occlusion func-
middle ground between Class II and Class III). Unfor- tions and intermeshes properly and is considered Class
tunately, many modem orthodontists use the ~ 1 ~ : I I, even though when one upper premolar correctly op-
designation as a goal of successful treatment. There- poses two lower premolars, the molars are full Class
fore, to more closely match current use, it is necessary II (Fig. 4), and when two upper premolars oppose one
tblu,ne 102 Viewno;nt
r I ~ e 281
Number 3

Fig, 7. Mixed dentition: Central axis of upper first deciduous

molar bisects embrasure between two lower deciduous molars
in modified Class I.
Fig. 5. Two upper premolars (t3) opposing one lower premolar
(B) in modified Class I. Note traditional Angle only classifies
molars and would call this excellent occlusion a Class II1!

Fig. 8. Mixed dentition, with early loss or extraction of upper

Iirst deciduous molar: Central axis of upper edentulous space
bisects embrasure between two lower deciduous molars in mod-
ified Class I,

precision. However, in the event that an upper first

Fig. 6. One premolar (13) in each quadrant. deciduous molar is prematurely lost, a line drawn
through the center axis of the edentulous space should
bisect the embrasure between the two lower deciduous
lower premolar, the molars are full Class 1II (Fig. 5), molars (Fig. 8). Since orthodontists are currently treat-
according to Angle. Modem orthodontists do extract ing patients earlier, when dentitions are mixed, it is
teeth, whereas Angle did not, and we need to have an essential that a classification be used which eliminates
occlusion classification that works when teeth are ex- the first molar focus, as Friel 8 and Arya et el. 9 dem-
tracted in only one arch. In the rare instance where no onstrated, the relationship between the upper and lower
premolar exists in a quadrant, then the center axis of first molars changes considerably through life.
the upper canine crown (not the cusp tip) should be Just as Angle required the orthodontist to mentally
used as a reference to the distal contact of the lower reposition a first molar that had drifted mesially because
canine. of early tooth loss or ectopic tooth eruption, with this
proposed modified Angle classification, the orthodon-
Deciduous and mixed dentition classification tist may be confronted with an unusual situation re-
The deciduous or mixed dentition was not directly quiring some imaginative solution to classify the patient
addressed by Angle in his classification. In the modified accurately. Given the large diversity of the human den-
classification, the center axis of the upper first decid- tition, no simplistic classification system will always
uous molar should split the embrasure between both apply. This supplemental classification does not elim-
lower deciduous molars (Fig. 7). Although it is possible inate the need for professional judgment in the special
to use the center axis of the upper second deciduous" situation, but it is hoped that this classification approach
molars, they are less accurate than using first deciduous will prove to be more consistent with treatment goals
molars because of the leeway space, which adds im- and the mechanics necessary to obtain these goals.
282 Katz Am. J. Orthod. Dentofac. Orthop.
September 1992

R L

Fig. 9. Right side is ideal intermeshing = 0 deviation. Left side is 2 mm Class II = + 2 deviation.

R L
Fig. 10. Right side is half cusp Class II = + 4 deviation. Left side is full Class II = + 8 deviation.

Quantifying the classification meshing on the right side, but a 2 mm Class II tendency
Another major concern regarding the Angle clas- on the left side, then the modified classification would
sification is the lack of a numerical quantification of read ( 0 , + 2 ) (Fig. 9). However, if another patient pre-
the degree of Class II or Class III. Techniques that use sents half a cusp Class II (cusp-to-cusp occlusion) on
millimeter measurements to quantify Class II or Class the right side and a full cusp Class II on the left side,
III tendency have been used by Ricketts and others t3 then the modified classification would read ( + 4, + 8)
for Rocky Mountain Data Systems (with a cephalo- (Fig. 10). A third patient who is 1.5 mm Class II on
metric film to evaluate canines and molars), and the right and 3.5 mm Class III on the left side would
by Elsasser 14 in a complicated, multifactorial be classified ( + 1 . 5 , - 3 . 5 ) (Fig. I1).
analysis. Note that any occlusion numbers ranging from
This proposed modified classification designates + 3.5 to - 3 . 5 would have fallen within the range of
ideal cusp-embrasure occlusion (as described by Angle) a Class I malocclusion in the traditional Angle classi-
as zero (0). A plus sign ( + ) designates Class II direction fication. And so, the occlusion in (Fig. 11), which
and a minus sign ( - ) designates Class III tendency. In requires both Class II and Class III mechanics to correct
this article the fight side is evaluated first, then the left to ideal, would be considered Class I on both sides by
side. Ideal occlusion on both right and left side~'i~ Angle's definition. However, in the proposed modified
therefore, (0,0). Angle classification, only (0,0) would meet the require-
For example, if a patient presents with ideal inter- ments of the ideal Class I, and could be considered to
Volume 102 V;ewnol.t 283
Number 3

R L

Fig. 11. Right side is 1.5 rnm Class II = + 1.5 deviation. Left side is 3.5 mm Class III = - 3.5 deviation.
Note traditional Angle can not classify a patient with both Class II and Class III sides.

satisfy the goals of excellent occlusion described by the teeth in maximum intercuspation (centric occlu-
Angle. sion). Maximum intercuspation frequently places the
This simple modified Angle classification has sev- jaws in quite a different relationship than the ortho-
eral advantages over other classification efforts. First, dontist would see if the condyles were properly located
this new system establishes a treatment goal that is a and if the neuromuscular system was in balance and
specific cusp-embrasure point rather than a range of comfort. Therefore the ideal time to classify a patient,
7 mm (half of a cusp each way) as in the traditional either before treatment or during treatment, would be
Angle classification,of malocclusion. Second, this sup- after the patient has been "deprogrammed" so as not to
plemental classification quantifies the degree of occlu- see the tooth-deflected false bite of maximum interdi-
sion error of a malocclusion precisely in millimeters gitation. Optimally, classification can be done on mod-
and for each side separately. Third, this method allows els properly mounted on a fully adjustable articulator
the orthodontist to classify those patients who have with a correct bite registration. This modified Angle
Class II tendency on one side and Class III tendency classification works especially well on articulated study
on the other side. Angle 2-~'8 mentioned that this possi- models. Ideal Class I has not been achieved if the patient
bility existed, but he did not include a provision for slides on tooth prematurities into a nice "fit" that looks
this situation in his classification because he said it was good but is not in harmony with the comfort position
a rare occurrence. Although it is extremely rare to find of the condyles and the relaxed, unstrained postural rest
a full Class II coexisting with a full Class III, partial position of the muscles.
Class II does occur on one side and partial Class III on In conclusion, it is hoped that this effort to modify
the other side with some frequency. This new approach classification technique, without discarding the im-
covers that eventuality. Fourth, this numerical system pressive contributions of Angle and others, will make
allows computer input to the facilitated and the severity communication regarding malocclusions more precise
of the malocclusion to be easily rated and compared in teaching, research, and treatment.
for statistical and research purposes. Fifth, this new
system addresses the need for deciduous and mixed REFERENCES
dentition classification to assist early treatment analysis. 1. Katz Ml. Angle classificationrevisited 1: is current usage reli-
Because dentists are familiar with the term Class I, able? AM J ORTHOD DENTOFACORTHOP [in press].
2. Angle EH. Classification of malocclusion. Dent Cosmos
it has been retained in the modified Angle classification, 1899;41:248-64, 350-7.
but it has been more narrowly redefined to coincide 3. Angle EH. Treatmentof malocclusionof the teeth and fractures
with Angle's description of the "ideal" occlusion, al- of the maxillae. Angle system. 6th ed. Philadelphia: SS White
lowing Class I to be used as a goal of treatment in Manufacturing, 1900:6-8, 37-44.
keeping with contemporary usage. 4. Angle EH. Treatmentof malocclusionof the teeth and fractures
of the maxillae. Angle system. 7th ed. Philadelphia: SS White
Another significant issue is the long-standing tra- Manufacturing, 1907:44-59.
dition of orthodontists to evaluate the occlusion in the 5. Strang RHW. A text-bookof orthodontia, 3rd ed. Philadelphia:
mouth, as well as to trim plaster study models, with Lea & Febiger, 1950:107-34.
284 Katz Am. J. Orthod. Dentofac. Orthop.
September 1992

6. Stoller AE. The normal position of the maxillary first permanent 13. Kraus B, Jordan R, Abrams L. Dental anatomy and occlusion.
molar. A,',t J OR'IttOD 1954;40:259-71. Baltimore: Williams & Wilkins, 1969:33-41.
7. Andrews LF. Six keys to normal occlusion. AM J ORTIXOD 14. Ricketts RM, Roth RH,Chaconas SJ, Schulhof RJ, Engel GA.
1972;62:296-309. Orthodontic diagnosis and planning. Vol. 1. Denver: Rocky
8. Angle Ell. The latest and best in orthodontic mechanisms. Dent Mountain/Orthodontics, 1982:127.
Cosmos 1928;70:1143-58. 15. Elsasser WA. Orthodontic assessment by the numbers. J Clin
9. Friel S. Occlusion. Observations on its development from in- Orthod 1978;12:116-22.
fancy to old age. l:,,rrJ ORTHOD 1927;13:322-34.
Reprint requests to:
10. Arya BS, Sarava BS, Thomas DR. Prediction of the first molar
Dr. Morton I. Katz
occlusion. AM J OR'rHOD 1973;63:610-21.
3435 Philips Dr.
11. Simon PW. On gnathostatic diagnosis in orthodontics. INI J
Baltimore, MD 21208
OwnloD 1924;10:755-8.
12. Simon PW. The simplified gnathostatic method. I,'r J OR'rHOD
1932;18:1081-7.

AAO MEETING CALENDAR

1993mToronto, Canada, May 15 to 19, Metropolitan Toronto Convention Center
1994--Orlando, Fla., May 1 to 4, Orange County Convent!on and Civic Center
1995mSan Francisco, Calif., May 7 to 10, Moscone Convention Center
(International Orthodontic Congress)
1996--Denver, Colo., May 12 to 16, Colorado Convention Center
1997--Philadelphia, Pa., May.3to .7, Philadelphia Convention Center
1998--Dallas, Texas, May 16 to 20, Dallas Convention Center
1999--San Diego, Calif., May 15 to 19, San Diego Convention Center