Available online at www.sciencedirect.

com

British Journal of Oral and Maxillofacial Surgery 46 (2008) 567572

Face bow and articulator for planning orthognathic surgery: 1 face bow
Fraser Walker a , Ashraf F. Ayoub b, , Khursheed F. Moos b , Joseph Barbenel c
a b c

Southern General Hospital, Department of Oral and Maxillofacial Surgery, Glasgow, UK Glasgow University Dental Hospital and School, Oral and Maxillofacial Surgery Department, Glasgow, UK University of Strathclyde, Glasgow, UK

Accepted 21 February 2008 Available online 7 May 2008

Abstract Orthognathic surgery that involves movement of the maxilla relative to the skull is usually planned using casts mounted on an articulator. Accurate positioning of the maxilla relative to the skull is essential for reliable planning, but current methods of mounting casts on articulators are inaccurate and unreliable. We propose that the casts should be mounted using the relation between the horizontal plane and the resting head position to dene the position of the skull. A photographic study of 10 subjects conrmed the reproducibility of the head position and its relation to the horizontal plane. A face bow incorporating a circular spirit level was used to transfer the relation between the horizontal and the maxillary dentition to a semiadjustable articulator. The angle between the horizontal and maxillary occlusal planes was measured from six lateral cephalograms and compared with those of casts mounted on a semiadjustable articulator using a face bow with either an orbital pointer or a spirit level. The face bow with a spirit level produced considerably more accurate results. 2008 The British Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
Keywords: Orthognathic; Head position; Face bow; Articulator

Introduction The use of orthognathic surgery for the correction of facial deformity has increased with the development of many techniques for the correction of increasingly complex congenital and acquired deformities. Such surgery commonly involves altering the position of the maxilla relative to the skull. The use of dental casts mounted on an articulator is a crucial part of the planning, as is the production of inter-occlusal positioning wafers as templates to guide surgeons perioperatively in positioning the jawbones. The casts are generally mounted on semiadjustable articulators and those designed for dental prosthetics have two major limitations when used for orthognathic purposes. Accurate positioning of casts of the upper

teeth, indicating the position of the maxilla relative to the skull, is essential for reliable planning; in current articulators the casts are mounted using a face bow which produces inaccurate and unreliable relations between the casts and the skull. Improvement of the mounting method is the subject of this rst paper in a series of publications. Current articulators have been designed to replicate a few interdental occlusal relations, not to reproduce the positions of important features relating to the maxilla and to the skull.1 The limitations of current articulators and the development of an improved orthognathic articulator will be described in a further paper.

The face bow Many articulators and face bows that are used for mounting casts are designed on the assumption that the Frankfort

Corresponding author. E-mail address: a.ayoub@dental.gla.ac.uk (A.F. Ayoub).

0266-4356/$ see front matter 2008 The British Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.

doi:10.1016/j.bjoms.2008.02.015

568

F. Walker et al. / British Journal of Oral and Maxillofacial Surgery 46 (2008) 567572

plane (also known as the Frankfort horizontal plane) and the upper cross member of the articulator are both parallel and horizontal.2 Several studies have shown that the planes dened on the articulator do not accurately correspond to anatomical planes,1,35 which produces errors in planning that are transferred to the interocclusal positioning wafers constructed on the articulated dental casts and used perioperatively to position the jaw bones. This results in a postoperative outcome different from that predicted. The aim of this investigation was to establish a reproducible and accurate method of relating the natural position of the head, and hence the position of the skull, to an absolute horizontal plane, and to design and construct a face bow that could accurately transfer this relation to the articulator when mounting maxillary casts. For this method it is essential that the natural position of the head be reproducible.

Position of the head The variability of natural position of the head was evaluated in a photographic study of 10 volunteers; each volunteer was photographed under controlled studio conditions on three separate occasions with an interval of not less than one hour between them. A digital camera was mounted on a tripod with the base horizontal; this allowed a horizontal line to be placed on each photograph, so the upper edge of the image was therefore horizontal. A line was drawn automatically across the image parallel to the edge to act as the horizontal reference. The exposure on the camera remained constant during the sessions; only the focus was altered to ensure sharpness of the nal images. A chair was positioned at a distance of two metres from a full length mirror, which had a line of 3mm thickness running down its full length; the height of the chair was set for each subject and remained constant during all three sessions. Each volunteer sat upright on the chair with the back straight and then looked into their own eyes in the mirror, positioning the vertical line on the mirror centrally on the reected image. This technique complied with the protocol described by Moorrees and Kean6 for adopting the natural position of the head. The right condylar region and the right side of the tip of the nose of each volunteer were marked. The marks were joined by a line on the printed photograph (Fig. 1). The angle between this line and the absolute horizontal line on the photograph was measured to 0.5 using a protractor. The same marks were used for all three sessions.

Fig. 1. Natural head position showing the head position angle.

The angles of the position of the head (Table 1) showed considerable variation among the different subjects and smaller differences between measurements made on different occasions. The Friedman non-parametric test was used to make a two-way analysis of the results. The difference between subjects was highly signicant (p < 0.001); the difference between measurements made on different occasions was not signicant (p > 0.05) and were considered as replicates. The variability of the within replicate measurements was quantied by calculating the difference between the largest and smallest value for each subject. The median difference was 1.75 and the 95% condence interval (CI) of the median using the Hodges Lehmann estimate was 1.25 , showing that the natural position of the head was sufciently
Table 1 Value of the head position angle of 10 volunteers measured on three occasions T1, T2 and T3 Case No 1 1 2 3 4 5 6 7 8 9 10 0.0 +8.5 +9.0 +7.0 +2.0 5.5 4.0 +6.0 +5.0 +8.5 Time 2 +2.0 +4.0 +9.0 +7.0 +2.0 6.0 3.5 +7.0 +5.0 +5.0 3 +6.0 1.0 +3.5 +7.0 +10.0 +2.0 6.0 5.0 +7.0 +6.0

Results Statistics There is a small number of cases with no evidence of the data being normally distributed, so non-parametric statistics were used to compare results.

F. Walker et al. / British Journal of Oral and Maxillofacial Surgery 46 (2008) 567572

569

Fig. 3. Dentatus average value face bow with orbital pointer.

Fig. 2. Spirit level attached to the face bow.

reproducible to be used for mounting a maxillary cast on an articulator. The signicant differences among subjects conrm the importance of using an adjustable, rather than a mean value, articulator for surgical planning.

Modication of the face bow The orbital pointer of a Dentatus face bow was replaced by a circular spirit level to record the horizontal plane and transfer it to an articulator to make use of the relation between natural position of the head and the horizontal plane for mounting the casts. The spirit level was attached to the clamp, which retained the orbital pin and was adjustable around two orthogonal axes to centralise the bubble (Fig. 2). The spirit level could be locked in place, which recorded the horizontal plane.

Six patients who required assessment for orthognathic surgery were recruited for this study; none had serious facial asymmetry. Each patient had a lateral cephalometric radiograph taken as part of their assessment, and two face bow recordings with the head in the natural position. The condylar rods on the face bow were positioned over the condylar heads. In one recording we used a conventional orbital pointer to record the position of the left orbitale; the second used the spirit level face bow to record the horizontal plane (Figs. 3 and 4). Upper and lower irreversible hydrocolloid impressions of the dental arches were taken and cast in dental stone. The casts were trimmed and duplicated using a reversible hydrocolloid. The casts were mounted on the articulator, and the condylar spheres locked into the centric position within the condylar track assemblies. The upper cross member of the articulator was set horizontal using the markings on the incisal guidance pin and a mounting plate attached. The face bow condylar rods were then positioned symmetrically on the condylar elements of the articulator. When we used the conventional technique, the orbital pointer of the face bow was adjusted to contact the underside of the orbital plane indicator that was attached to the upper cross member of the articulator. The maxillary cast was

The accuracy of mounting occlusal casts The spirit level face bow was compared with the conventional method of mounting casts on a Dentatus ARH semiadjustable articulator, which is often used for planning orthognathic surgery. The conventional face bow uses the position of the condylar heads and the lowest point of the left orbit-orbitale as the third reference point to dene the plane used for mounting the maxillary cast.

Fig. 4. Dentatus average value face bow with attached circular spirit level.

570

F. Walker et al. / British Journal of Oral and Maxillofacial Surgery 46 (2008) 567572

Fig. 5. Cast mounted on an average value Dentatus face bow using the orbital pin and orbital plane guide.

Fig. 7. Graph showing angle between horizontal and maxillary occlusal planes measured by three methods.

placed in the bite fork and attached to the mounting plate with plaster of Paris (Fig. 5). The duplicate cast was mounted using the spirit level face bow. The face bow was positioned on the articulator, and rotated about the condylar rods by raising or lowering the anterior bar of the face bow frame to centre the spirit level bubble. The maxillary cast was then mounted as described above (Fig. 6).

Measurement of the angle between the horizontal and maxillary occlusal planes (Fig. 7) The angle between the maxillary occlusal plane and the horizontal plane was measured on the lateral cephalogram and compared with that of the mounted duplicate maxillary casts. A horizontal line was drawn on each patients lateral cephalogram parallel to the horizontal edge of the nasion rest, which is an integral component of the radiographic apparatus set by the manufacturer in a horizontal plane. A second line was drawn on the radiograph from the incisal tip of the central incisor to the lowest cusp tip of a maxillary molar tooth, normally the mesiobuccal cusp of an upper rst molar. The patients maxillary dental cast was used to ensure that the appropriate molar cusp was identied on the radiograph. Both lines were extended posteriorly and the angle between them measured using a protractor. The

Fig. 8. Maxillary occlusal plane angle of a cast mounted on Dentatus articulator using average value face bow.

measurements were repeated to evaluate the accuracy of the technique. The angle between the horizontal cross member of the articulator and the maxillary occlusal plane of the casts mounted by the two face bow methods were then measured (Figs. 8 and 9). A protractor was adapted to allow the vertical adjustment of its position. A at plate was placed across the occlusal plane of the maxillary teeth of the mounted cast and the angle measured on the protractor.

Results Table 2 shows good reproducibility of the duplicate measurements taken from the lateral cephalogram; the median difference between the rst and second values was 0.5 and the Hodges Lehmann 95% CI of the median was 0.5 . There was, however, considerable intersubject variability, with Table 2 showing a different rank order for each method. The discrepancy in order was greatest between the cephalographic measurements and the casts mounted using

Fig. 6. Cast mounted using spirit level face bow.

F. Walker et al. / British Journal of Oral and Maxillofacial Surgery 46 (2008) 567572

571

Table 2 Angle between horizontal and maxillary occlusal planes (occlusal plane angle) measured by three methods: lateral cephalogram, orbital pointer face bow, and spirit level face bow Case No 1 Lateral cephalogram 1st measurement 2nd measurement Spirit level face bow Orbital pointer face bow Differences in occlusal plane angle between lateral cephalogram and spirit level face bow Differences in occlusal plane angle between lateral cephalogram and orbital pointer face bow 5.5 4 10 26 5.2 21.2 2 8 9 6 1 2.5 2.5 3 5 3 4 5 0 21 4 2 3 1 13 1.5 10.5 5 14 16.5 11 17 4.2 1.7 6 10 7 8 20 0.5 11.5

Fig. 9. Maxillary occlusal plane angle of a cast mounted on Dentatus articulator using spirit level orthognathic face bow.

the orbitale; the discrepancy was so large that the Spearman rank correlation coefcient between the values was negative. There were also obvious differences between the magnitudes of the angles measured using the different methods. The difference between the mean values of the angles measured from the cephalograms and the casts mounted using the spirit level was 1.0 with a Hodges Lehmann 95% CI of 1.25 . The values for casts mounted using the orbital pointer were 10.75 and 11.5 . The differences between the three methods of obtaining the horizontal-occlusal plane angle was highly signicant (Friedman test, p < 0.001). There was a signicant difference between the cephalographic values and those for the casts mounted using the orbital pointer (Nemenyis test, p < 0.005), but not for the casts mounted using the spirit level (p > 0.05).

Discussion The results of the photographic study showed that the subjects could repeatedly assume the same head posture under the same conditions. The relation between the horizontal plane and the natural position of the head was applied as a

new method of mounting casts on an articulator for planning operations. The lateral cephalograms provided the most direct method of measuring the values of the angle between horizontal and maxillary occlusal planes. The duplicate measurements from the cephalogram showed it to be an accurate and reliable method, with good agreement between the duplicate measurements, and it provided the standard against which the other methods could be evaluated. Both the photographic and cephalometric measurements showed considerable inter-subject variation in the angle between the horizontal and the occlusal plane. Some of this variation will be incorporated into the orbitale registration, but the unjustied assumption of a xed angle between the axis-orbital plane and the Frankfort and horizontal planes can introduce serious errors. There were large signicant systematic differences between the occlusal angle measured from the cephalograms and casts mounted using the orbital pointer which conrmed previous criticisms.2,3,79 The number of cases studied was small and further studies are required, but results of greater signicance have been obtained. The spirit level face bow produced a considerably more accurate method of mounting casts without producing increased clinical complexity. The method of using the spirit level face bow is not greatly different from the method currently used. The mounted casts replicated more accurately the angle between the occlusal plane and the horizontal in the anteroposterior direction, producing an appearance of the teeth on the articulator similar to that seen in the patient. This is essential for planning orthognathic treatment, and for the production of dentures and crowns. It is, however, impossible to reproduce the lateral angle between the horizontal and the occlusal plane using current articulators because there is no mechanism for such adjustment. This is usually of minor importance in mounting the casts of subjects with facial symmetry, but may produce a major distortion if the patient has condylar asymmetry, as occurs in some patients who require orthognathic surgery. Current articulators have a xed condylar height and intercondylar axis, and mounting casts of patients with condylar asymmetry leads to rotation of the cast about an anteroposterior axis, which produces tilting of the cast and an appearance that differs from reality. This distortion is a problem not only

572

F. Walker et al. / British Journal of Oral and Maxillofacial Surgery 46 (2008) 567572 3. Ellis III E, Tharanon W, Gambrell K. Accuracy of face-bow transfer: effect on surgical prediction and postsurgical result. J Oral Maxillofac Surg 1992;50:5627. 4. Bailey JOJR, Nowlin TP. Evaluation of the third point of reference for mounting maxillary casts on the Hanau articulator. J Prosthet Dent 1984;51:199201. 5. Ferrario VF, Sforza C, Serrao G, Schmitz JH. Three-dimensional assessment of the reliability of a postural face-bow transfer. J Prosthet Dent 2002;87:2105. 6. Moorrees CFA, Kean R. Natural head position: a basic consideration in the interpretation of cephalometric radiographs. Am J Phys Anthropol 1958;16:21334. 7. Gonzales JB, Kingery RH. Evaluation of planes of references for orienting maxillary casts on articulators. J Am Dent Assoc 1968;76: 32936. 8. Gold BR, Setchell DJ. An investigation of the reproducibility of face-bow transfers. J Oral Rehabil 1983;10:495503. 9. Schallhorn RG. A study of the arbitrary center and kinematic center of rotation for face bow mountings. J Prosthet Dent 1957;7:1629.

for surgical planning, but also for the production of dentures and crowns. The orbital pointer can lead to misalignment of the mounted cast when there is orbital asymmetry. The lateral angle can be recorded on the spirit level face bow but this cannot be used for mounting the cast on current articulators. An improved more adjustable articulator is required to realise the potential of the spirit level face bow. The design, construction, and evaluation of such an improved articulator are reported elsewhere.

References
1. OMalley AM, Milosevic A. Comparison of three face bow/semiadjustable articulator systems for planning orthognathic surgery. Br J Oral Maxillofac Surg 2000;38:18590. 2. Pitchford JH. A re-evaluation of the axis-orbital plane and the use of orbitale in a facebow transfer record. J Prosthet Dent 1991;66: 34955.