An Introduction to Orthodontics, 2nd Edition

6. Cephalometrics


Cephalometry is the analysis and interpretation of standardized radiographs of the facial bones. In practice, cephalometrics has come to be associated with a true lateral view (Fig. 6.1). An anteroposterior radiograph can also be taken in the cephalostat, but this view is difficult to interpret and is usually only employed in cases with a skeletal asymmetry.


In order to be able to compare the cephalometric radiographs of one patient taken on different occasions, or those of different individuals, some standardization is necessary. To achieve this aim the cephalostat was developed by B. Holly Broadbent in the period after the First World War (Fig. 6.2). The cephalostat consists of an X-ray machine which is at a fixed distance from a set of ear posts designed to fit into the patient's external auditory meatus. Thus the central beam of the machine is directed towards the ear posts, which also serve to stabilize the patient's head. The position of the head in the vertical axis is standardized by ensuring that the patient's Frankfort plane (for definition see below) is horizontal. This can be done by manually positioning the subject or, alternatively, by placing a mirror some distance away level with the patient's head and asking him or her to look into their own eyes. This is termed the natural head position, and some orthodontists claim that it is more consistent than a manual approach. It is normal practice to cone down the area exposed so that the skull vault is not routinely included in the X-ray beam.

Fig. 6.1. A lateral cephalometric radiograph. Note the scale on the upper right-hand side.

Unfortunately, attempts to standardize the distances from the tube to the patient (usually between 5 and 6 feet (1.5 to 1.8 m)) and from the patient to the film (usually around 1 foot (around 30 cm)) have not been entirely successful as the values in parenthesis would suggest. Some magnification, usually of the order of 78 per cent, is inevitable with a lateral cephalometric film. In order to be able to check the magnification and thus the comparability of different films, it is helpful if a scale is included in the view (see Fig. 6.1).

To give a better definition of the soft tissue outline of the face, either a thin layer of barium paste can be placed down the central axis of the face or an aluminium wedge positioned so as to attenuate the beam in that area.


An increasing awareness of the risks associated with X-rays has led clinicians to re-evaluate the indications for taking a cephalometric radiograph. The following are considered valid.

Fig. 6.2. A cephalostat.

6.2.1. An aid to diagnosis

It is possible to carry out successful orthodontic treatment without taking a cephalometric radiograph, particularly in Class I malocclusions. However, the information that cephalometric analysis yields is helpful in assessing the probable aetiology of a malocclusion and in planning treatment. The benefit to the patient in terms of the additional information gained must be weighed against the X-ray dosage. Therefore a lateral cephalometric radiograph is best limited to patients with a skeletal discrepancy and/or where anteroposterior movement of the incisors is planned. In a small proportion of patients it may be helpful to monitor growth to aid the planning and timing of treatment by taking serial cephalometric radiographs, although again the dosage to the patient must be justifiable.

In addition, a lateral view is often helpful in the accurate localization of unerupted displaced teeth and other pathology.

6.2.2. A pretreatment record

A lateral cephalometric radiograph is useful in providing a baseline record prior to the placement of appliances, particularly where movement of the upper and lower incisors is planned.

6.2.3. Monitoring the progress of treatment

In the management of severe malocclusions, where tooth movement is occurring in all three planes of space (for example treatments involving functional appliances, or upper and lower fixed appliances), it is common practice to take a lateral cephalometric radiograph during treatment to monitor anchorage requirements and incisor inclinations. A lateral cephalometric radiograph may also be useful in monitoring the movement of unerupted teeth and is the most accurate view for assessing upper incisor root resorption if this occurs during treatment.

6.2.4. Research purposes

A great deal of information has been obtained about growth and development by longitudinal studies which involved taking serial cephalometric radiographs from birth to the late teens or beyond. While the data provided by previous investigations are still used for reference purposes, it is no longer ethically possible to repeat this type of study. However, those views taken routinely during the course of orthodontic diagnosis and treatment can be used to study the effects of growth and treatment.


Before starting a tracing it is important to examine the radiograph for any abnormalities or pathology. For example, a pituitary tumour could result in an increase in the size of the sella turcica. Shortening of the roots of the incisors is often seen more clearly on a lateral cephalometric radiograph. This view is also helpful in assessing the patency of the airway, as enlarged adenoids can be easily seen.

In order to be able to derive meaningful information from a lateral cephalo-metric tracing, an accurate and systematic approach is required which also involves selecting the right conditions and equipment for the task.

·     The tracing should be carried out in a darkened room on a light viewing box. All but the area being traced should be shielded to block out any extraneous light.

·     Although it is possible to use tracing paper, proprietary acetate sheets are more transparent and give a more professional result.

·     A sharp pencil should be used. The author recommends a 0.3 mm leaded propelling pencil (as this saves hours searching for pencil sharpeners). Some orthodontists with very steady hands use a fine ink stylus, but this is not advocated for the novice.

·     The tracing paper or acetate sheet should be secured onto the film with masking tape, which does not leave a sticky residue when removed. The tracing should be oriented in the same position as the patient was when the radiograph was taken, i.e. with the Frankfort plane horizontal.

·     Some orthodontists use stencils to obtain a neat outline of the incisor and molar teeth. However, too much artistic licence can lead to inaccuracies, particularly if the crown root angle of a tooth is not average.

·     For landmarks which are bilateral (unless they are directly superimposed) an average of the two should be taken.

·     With a careful technique tracing errors should be of the order of + 0.5 mm for linear measurements and + 0.5° for angular measurements.

·     It is a valuable learning experience to trace the same radiograph on two separate occasions and compare the tracings. This helps to reduce the temptation to place undue emphasis upon small variations from normal cephalometric values.

An example of a tracing is shown in Fig. 6.3 (see also Fig. 6.4). Definitions of the various points and reference planes are given in Section 6.5.

Fig. 6.3. A cephalometric tracing: patient LH (male) aged 14 years.

6.3.1. Digitizers

A digitizer comprises an illuminated radiographic viewing screen which is connected to a computer. Information from a lateral cephalometric radiograph is entered into the computer by means of a cursor which records the horizontal and vertical (x, y) co-ordinates of cephalometric points or bony or soft tissue outlines. Specialized software can then be employed to utilize the information entered to produce a tracing and/or the analysis of choice. Studies have shown digitizers to be as accurate as tracing a radiograph by hand. Certainly, this approach is particularly useful for research as any number of radiographs can be entered, superimposed, and/or compared statistically.


The orthodontic literature is replete with different cephalometric analyses, which in itself suggests that no single method is sufficient for all purposes and that all have their drawbacks. In a book of this size it is more appropriate to consider one analysis in depth. Therefore one of the approaches used commonly in the UK will be considered (Table 6.1). For details of other analyses the reader is referred to the publications cited in the section on further reading.

Table 6.1 Cephalometric norms for Caucasians (Eastman Standard).


Mean value

Standard Deviation






UInc to MxPl


LInc to MnPl


Inter-incisal angle





Facial Proportion



LInc to APog line

+1 mm

2 mm

SN to MxPl

For definitions see Section 6.5.
*Or 120°
MMPA (see Section 6.8).

Cephalometric analyses are often based upon comparing the values obtained for certain measurements for a particular individual (or group of individuals) with the average values for their population (e.g. Caucasians). An indication of the significance of any difference between the actual measurement for an individual and the average value can be obtained from the standard deviation. The range given by one standard deviation around the mean will include 66 per cent of the population and two standard deviations will include 97 per cent.

Cephalometric analysis is also of value in identifying the component parts of a malocclusion and probable aetiological factors it is useful when a tracing is finished to reflect why that individual has that particular malocclusion. However, it is important not to fall into the trap of giving more credence to cephalometric analysis than it actually merits; it should always be remembered that it is an adjunctive tool to clinical diagnosis, and differences of cephalometric values from the average are not in themselves an indication for treatment, particularly as variations from normal in a specific value may be compensated for elsewhere in the facial skeleton or cranial base. In addition, cephalometric errors can occur owing to incorrect positioning of the patient and incorrect identification of landmarks.

A lateral cephalometric radiograph is a slightly magnified, two-dimensional representation of a three-dimensional object (the patient). For this reason angular measurements are generally to be preferred to linear measurements as the element of magnification is less important.


The points and reference lines are shown in Fig. 6.4.

A point (A) this is the point of deepest concavity on the anterior profile of the maxilla. It is also called subspinale. This point is taken to represent the anterior limit of the maxilla and is often tricky to locate accurately. However, tracing the outline of the root of the upper central incisor first and shielding all extraneous light often aids identification. The A point is located on alveolar bone and is liable to changes in position with tooth movement and growth.

Anterior nasal spine (ANS) this is the tip of the anterior process of the maxilla and is situated at the lower margin of the nasal aperture.

B point (B) the point of deepest concavity on the anterior surface of the mandibular symphysis. The B point is also sited on alveolar bone and can alter with tooth movement and growth.

Gonion (Go) the most posterior inferior point on the angle of the symphysis. This point can be guesstimated, or determined more accurately by bisecting the angle formed by the tangents from the posterior border of the ramus and the inferior border of the mandible (Fig. 6.5).

Menton (Me) the lowest point on the mandibular symphysis.

Nasion (N) the most anterior point on the frontonasal suture. When difficulty is experienced locating nasion, the point of deepest concavity at the intersection of the frontal and nasal bones can be used instead.

Orbitale (Or) the most inferior anterior point on the margin of the orbit. By definition, the left orbital margin should be used to locate this point. However, this can be a little tricky to determine radiographically, and so an average of the two images of left and right is usually taken.

Pogonion (Pog) the most anterior point on the mandibular symphysis.

Porion (Po) the uppermost outermost point on the bony external auditory meatus. This landmark can be obscured by the ear posts of the cephalostat, and some advocate tracing these instead. However, this is not recommended as they do not approximate to the position of the external auditory meatus. The uppermost surface of the condylar head is at the same level, and this can be used as a guide where difficulty is experienced in determining porion.

Posterior nasal spine (PNS) this is the tip of the posterior nasal spine of the maxilla. This point is often obscured by the developing third molars, but lies directly below the pterygomaxillary fissure.

Sella (S) the midpoint of the sella turcica.

SN line  this line, connecting the midpoint of sella turcica with nasion, is taken to represent the cranial base.

Frankfort plane  this is the line joining porion and orbitale. This plane is difficult to define accurately because of the problems inherent in determining orbitale and porion.

Mandibular plane  The line joining gonion and menton. This is only one of several definitions of the mandibular plane, but is probably the most widely used. Other definitions can be found in the publications listed in the section on further reading.

Maxillary plane  the line joining anterior nasal spine with posterior nasal spine. Where it is difficult to determine ANS and PNS accurately, a line parallel to the nasal floor can be used instead.

Functional occlusal plane  a line drawn between the cusp tips of the permanent molars and premolars (or deciduous molars in mixed dentition). It can be difficult to decide where to draw this line, particularly if there is an increased curve of Spee, or only the first permanent molars are in occlusion during the transition from mixed to permanent dentition. The functional plane can change orientation with growth and/or treatment, and so is not particularly reliable for longitudinal comparisons.

Fig. 6.4. Commmonly used cephalometric points and planes.

Fig. 6.5. Construction of Gonion (Go): (a) draw tangents to posterior and inferior borders; (b) bisect the angle formed by the tangents and mark where it crosses the angle of the mandible; (c) repeat for the other outline (if one is visible). Go is located midway between the two points.





6.6.1. Angle ANB (Fig. 6.6)

In order to be able to compare the position of the maxilla and mandible, it is necessary to have a fixed point or plane. The skeletal pattern is often determined cephalometrically by comparing the relationship of the maxilla and mandible with the cranial base by means of angles SNA and SNB. The difference between these two measurements, angle ANB, is classified broadly as follows:

Fig. 6.6. Assessment of skeletal pattern using angles SNA and SNB: patient LH (male) aged 14 years.



ANB < 2°

Class III

2° < ANB < 4°

Class I

ANB > 4°

Class II

However, this approach assumes (incorrectly in some cases) that the cranial base, as indicated by the line SN, is a reliable basis for comparison and that points A and B are indicative of maxillary and mandibular basal bone. Variations in the position of nasion can also affect angles SNA and SNB and thus the difference ANB (Fig. 6.7); however, variations in the position of sella do not. If SNA is increased or reduced from the average value, this could be due to either a discrepancy in the position of the maxilla (as indicated by point A) or nasion. The following (rather crude) modification is often used in order to make allowance for this:

Provided the angle between the maxillary plane and the sellanasion line is within 5°11°:

·     if SNA is increased, for every degree that SNA is greater than 81°, subtract 0.5° from ANB;

·     if SNA is reduced, for every degree that SNA is less than 81°, add 0.5° to ANB.

If the angle between the maxillary plane and the sellanasion line is not within 5°11°, this correction is not applicable.

Alternatively, an approach which avoids the cranial base (e.g. the Ballard conversion or the Wits analysis) can be used to supplement the above analysis, particularly where the cephalometric findings are at variance with the clinical assessment.

Fig. 6.7. Effect of variations in the position of nasion on angles SNA, SNB, and ANB:



6.6.2. Ballard conversion (Fig. 6.8)

This analysis uses the incisors as indicators of the relative position of the maxilla and mandible. It is easy to confuse a Ballard conversion and a prognosis tracing (see Fig. 6.12), but in the former the aim is to tilt the teeth to their normal angles (thus eliminating any dento-alveolar compensation) with the result that the residual overjet will indicate the relationship of the maxilla to the mandible.

Fig. 6.8. Ballard conversion: average upper incisor angle to maxillary plane, 109°; lower incisor angle to mandibular plane, 120°31.5° = 88.5°.

The method is as follows.

1. Trace on a separate piece of tracing paper the outline of the maxilla, the mandibular symphysis, the incisors, and the maxillary and mandibular planes.

2. Mark the rotation points of the incisors one-third of the root length away from the root apex.

3. By rotating around the point marked, reposition the upper incisor at an angle of 109° to the maxillary plane. Repeat for the lower incisor (allowing for the maxillary mandibular planes angle of 31.5° in this case).

4. The residual overjet reflects the underlying skeletal pattern. In this case the Ballard conversion indicates a mild Class III skeletal pattern as the repositioned incisors are nearly edge to edge.

6.6.3. Wits analysis (Fig. 6.9)

This analysis compares the relationships of the maxilla and mandible with the occlusal plane. There are several definitions of the occlusal plane, but for the purposes of the Wits analysis it is taken to be a line drawn between the cusp tips of the molars and premolars (or deciduous molars), which is known as the functional occlusal plane. Perpendicular lines from both point A and point B are dropped to the functional occlusal plane to give points AO and BO. The distance between AO and BO is then measured. The mean values are 1 mm (SD + 1.9 mm) for males and 0 mm (SD + 1.77 mm) for females.

Fig. 6.9. Wits analysis: LH (male) aged 14 years. The method is as follows.

1. Draw in the functional occlusal plane (FOP).

2. Drop perpendiculars from point A and point B to the FOP to give points AO and BO.

3. Measure the distance between AO and BO.

The average value is +1 mm (ą 1.9 mm) for males and 0 mm (ą 1.77 mm) for females. The distance from AO to BO for LH (male) is +2 mm, suggesting a mild Class III skeletal pattern.

The drawback to this approach is that the functional occlusal plane is not easy to locate, which obviously affects the accuracy and reproducibility of the Wits analysis. A slight difference in the angulation of the functional occlusal plane can have a marked effect on the relative positions of AO and BO.


Again there are many different ways of assessing vertical skeletal proportions. The more commonly used include the following.

·     The MaxillaryMandibular Planes Angle (Fig. 6.10). The average angle between the maxillary plane and the mandibular plane (MMPA) is 27° + 4°. Some analyses measure the angle between the Frankfort and the mandibular planes (average 28° + 4°). However, the maxillary plane is easier to locate accurately and therefore the MMPA is preferred.

·     The Facial Proportion (Fig. 6.11). This is the ratio of the lower facial height to the total anterior facial height measured perpendicularly from the maxillary plane, calculated as a percentage:

If there appears to be a discrepancy between the results for these two measurements of vertical relationship, it should be remembered that the MMPA reflects both posterior lower facial height and anterior lower facial height. Therefore in the case of patient LH who has an increased MMPA but an average facial proportion it would appear that the posterior lower facial height is reduced (as opposed to an increased anterior lower facial height).

Fig. 6.10. Assessment of vertical skeletal pattern using the MMPA and FMPA: LH (male) aged 14 years.



Both the MMPA and the Frankfort mandibular planes angle are increased. This may be due to either an increased lower anterior face height or a reduced lower posterior face height.

Fig. 6.11. Calculating the facial proportion: LH (male) aged 14 years.




The average value for the angle formed between the upper incisor and the maxillary plane is 109°. The normal value for lower incisor angle given in Table 6.1 is for an individual with an average MMPA of 27°. However, there is a relationship between the MMPA and the lower incisor angle: as the MMPA increases, the lower incisors become more retroclined. As the sum of the average MMPA (27°) and the average lower incisor angle (93°) equals 120°, an alternative way of deriving the average lower incisor angulation for an individual is to subtract the MMPA from 120°:

lower incisor angle = 120° MMPA.

6.8.1. Prognosis tracing

Sometimes it is helpful to be able to determine the type and amount of incisor movement required to correct an increased or reverse overjet. Although the skeletal pattern will give an indication, on occasion compensatory proclination or retroclination (known as dento-alveolar compensation) of the incisors can confuse the issue. When planning treatment in such a case it may be helpful to carry out a prognosis tracing. This involves moving the incisor(s) to mimic the movements achievable with different types of appliance therapy to help determine the best course of action for that patient. An example is shown in Fig. 6.12, where it can be seen that bodily retraction of the upper incisors would result in their being retracted out of the palatal bone obviously not a practical treatment proposition.

A quick method of calculating the final upper incisor angle following tipping movements is to assume that for 2.5° of angular movement (about a point of rotation one-third of the way down the root from the apex) the upper incisor edge will translate approximately 1 mm. However, it should be stressed that both methods provide only a rough guide to the tooth movements required.

Fig. 6.12. Prognosis tracing: CP (female) aged 18 years. From this diagram it can be seen that bodily movement of the upper incisors to reduce this patient's overjet would not be feasible. Therefore a surgical aproach was recommended.

6.8.2. A-Pogonion line (APog)

Raleigh Williams noted when he analysed the lateral cephalometric radiographs of individuals with pleasing facial appearances that one feature which they all had in common was that the tip of their lower incisor lay on or just in front of the line connecting point A with pogonion. He advocated using this position of the lower incisor as a treatment goal to help ensure a good facial profile. While this line may be useful when planning orthodontic treatment, it must be remembered that it is only a guideline to good facial aesthetics, and not an indicator of stability. If the lower incisors are moved from their pretreatment position of labiolingual balance, whatever the rationale, there is a likelihood of relapse following removal of appliances. This topic is discussed in more detail in Chapters 7 and 10.


The major role of analysis of the soft tissues is in diagnosis and planning prior to orthognathic surgery (Chapter 20). As with other elements of cephalometric analysis, there are a large number of different analyses of varying complexity. The following are some of the more commonly used:

The Holdaway line

This is a line from soft tissue chin to the upper lip. In a well-proportioned face this line, if extended, should bisect the nose (Fig. 6.13).

Fig. 6.13. Soft tissue analysis.

Rickett's E-plane

This line joins soft tissue chin and the tip of the nose. In a balanced face the lower lip should lie 2 mm (ą 2 mm) anterior to this line with the upper lip positioned a little further posteriorly to the line (Fig. 6.13).

Facial plane

The facial plane is a line between the soft tissue nasion and the soft tissue chin. In a well-balanced face the Frankfort plane should bisect the facial plane at an angle of about 86° and point A should lie on it (Fig. 6.13).

As with other aspects of cephalometrics, but perhaps more pertinently, these analyses should be supplementary to a clinical examination, and it should also be remembered that beauty is in the eye of the beholder.


The advantage of standardizing lateral cephalometric radiographs is that it is then possible to compare radiographs either of groups of patients for research purposes or of the same patient over time to evaluate growth and treatment changes. In some cases it may be helpful to monitor growth of a patient over time before deciding upon a treatment plan, particularly if unfavourable growth would result in a malocclusion that could not be treated by orthodontics alone. During treatment it can be helpful to determine the contributions that tooth movements and/or growth have made to the correction and to help ensure that, where possible, a stable result is achieved. For example, in a Class II division 1 malocclusion, correction of an increased overjet can occur by retroclination of the upper incisors and/or proclination of the lower incisors and/or forward growth of the mandible and/or restraint of forward growth of the maxilla. If the major part of the correction is due to proclination of the lower incisors there is an increased likelihood of relapse of the overjet following cessation of appliance therapy owing to soft tissue pressures. If this is determined before appliances are removed, it may be possible to take steps to rectify the situation.

However, in order to be able to compare radiographs accurately it is necessary to have a fixed point or reference line which does not change with time or growth. Unfortunately this poses a dilemma, as there are no natural fixed points or planes within the face and skull. This should be borne in mind when interpreting the differences seen using any of the superimpositions discussed below.

6.10.1 Cranial base

The SN line is taken in cephalometrics as approximating to the cranial base. However, growth does occur at nasion, and therefore superimpositions on this line for the purpose of evaluating changes over time should be based at sella. Unfortunately, growth at nasion does not always conveniently occur along the SN line if nasion moves upwards or downwards with growth, this will of course introduce a rotational error in comparisons of tracings superimposed on SN. It is more accurate to use the outline of the cranial base (called de Coster's line) as little change occurs in the anterior cranial base after 7 years of age (see Chapter 4). However, a clear radiograph and a good knowledge of anatomy is required to do this reliably.

6.10.2. The maxilla

Growth of the maxilla occurs on all surfaces by periosteal remodelling. For the purpose of interpretation of growth and/or treatment changes the least affected surface is the anterior surface of the palatal vault, although the maxilla is commonly superimposed on the maxillary plane at PNS.

6.10.3. The mandible

It was noted above that there are no natural stable reference points within the face and skull. Bjork overcame this problem by inserting metal markers in the facial skeleton. Whilst this approach is obviously not applicable in the management of patients, it did provide considerable information on patterns of facial growth, indicating that in the mandible the landmarks which change least with growth are as follows (in order of usefulness):

·     the innermost surface of the cortical bone of the symphysis;

·     the tip of the chin;

·     the outline of the inferior dental canal;

·     the crypt of the developing third permanent molars from the time of commencement of mineralization until root formation begins.


Brown, M. (1981). Eight methods of analysing a cephalogram to establish anteroposterior skeletal discrepancy. British Journal of Orthodontics8, 13946.

This paper admirably illustrates the pitfalls and problems with cephalometric analysis, whilst also briefly presenting some alternative analyses.

Ferguson, J. W., Evans, R. I. W., and Cheng, L. H. H. (1992). Diagnostic accuracy and observer performance in the diagnosis of abnormalities in the anterior maxilla: a comparison of panoramic with intra-oral radiography. British Dental Journal173, 26571.

Gravely, J. F. and Murray Benzies, P. (1974). The clinical significance of tracing error in cephalometry. British Journal of Orthodontics1, 95101.

A classical paper on tracing errors.

Guidelines for the use of radiographs in clinical orthodontics. British Orthodontic Society, London, 1994.

Houston, W. J. B. (1979). The current status of facial growth prediction. British Journal of Orthodontics6, 1117.

Houston, W. J. B. (1986). Sources of error in measurements from cephalometric radiographs. European Journal of Orthodontics8, 14951.

Jacobson, A. (1995). Radiographic cephalometry: from basics to videoimaging. Quintessence Publishing, USA.

An authoratative book. Includes a very good section on how to trace a cephalometric radiograph with actual copy films and overlays to aid landmark identification.

Lewis, D. H. (1981). Basic tracing for lateral skull radiographs. Dental Update8, 4551.

Lewis, D. H. (1981). Lateral skull radiographs: growth and treatment changes. Dental Update8, 19399.

These two papers by Lewis give an introduction to cephalometrics which is easy to follow.

Sandham, A. (1988). Repeatability of head posture recordings from lateral cephalometric radiographs. British Journal of Orthodontics15, 15762.

If you find an error or have any questions, please email us at Thank you!