Photograph of the 3-dimensional digital photographic system. This system is composed of 6 digital cameras, 3 on each side of the patient.
Three-dimensional image of the measurement used for the interalar width (IAW), corresponding to the widest distance from ala to ala seen on frontal view.
Three-dimensional image of the internostril width (INW) measurement, which measured the widest distance from the lateral portion of each nostril.
Measurement of nasal tip projection (NTP), which measured the distance from the subnasale (Sn) to the most anterior midline surface point at the tip of the nose, and columellar length (CL), which corresponded to the most anterior point of the nostril rim.
Preoperative 3-dimensional photograph of a patient with class 3 occlusion (patient 7).
Postoperative 3-dimensional photograph of patient 7 taken 6 months after surgery. Note the slight widening of the nasal ala.
Soft tissue differences after surgery for patient 7.
Color histogram of patient 7.
Color-based soft tissue differences showing both positive and negative changes of the facial soft tissues that occurred after surgery.
Color histogram showing the changes in the patient in Figure 9.
Honrado CP, Lee S, Bloomquist DS, Larrabee WF. Quantitative Assessment of Nasal Changes After Maxillomandibular Surgery Using a 3-Dimensional Digital Imaging System. Arch Facial Plast Surg. 2006;8(1):26-35. doi:10.1001/archfaci.8.1.26
Author Affiliations: Department of Otolaryngology–Head and Neck Surgery, New York Medical College, Valhalla (Dr Honrado); Department of Otolaryngology–Head and Neck Surgery, University of Washington, Seattle (Dr Lee); Department of Oral and Maxillofacial Surgery, Swedish Medical Center, Seattle (Dr Bloomquist); and Larrabee Center for Facial Plastic Surgery, Seattle (Dr Larrabee).
Correspondence: Carlo P. Honrado, MD, ENT Faculty Practice LLP, 1055 Saw Mill River Rd, Suite 101, Ardsley, NY 10502 (firstname.lastname@example.org).
Objective To evaluate nasal changes after maxillomandibular surgery by means of images taken with a 3-dimensional digital camera.
Design Thirty-two patients (26 female and 6 male) with preoperative and postoperative 3-dimensional photographs were studied. The patients underwent maxillary movement with impaction (upward rotation), maxillary movement with lengthening (downward rotation), or maxillary movement without rotation. With the 3-dimensional imaging software, preoperative and postoperative calculations were performed for interalar width, internostril width, nasal tip projection, and columellar length from the 3-dimensional digital images. The nasolabial angle was also measured.
Results Postoperative interalar and internostril widening was significant (P<.05) for all 3 categories of maxillary movement. However, there was no statistically significant change in nasal tip projection and columellar length. Interestingly, movement of the maxilla with upward rotation did show a statistically significant decrease in the nasolabial angle.
Conclusions Changes to the nose clearly occur after orthognathic surgery. There was a statistically significant increase in postoperative interalar width and internostril width with maxillary movement. However, no clear correlation could be determined between amount of change and maxillary movement. Interestingly, maxillary advancement did not show any significant change in nasal tip projection or columellar length, with data showing both increases and decreases in measurements. The nasolabial angle in patients who underwent maxillary advancement with impaction (upward rotation) was the only measurement that showed a statistically significant increase.
Surgical correction of dentofacial anomalies such as class 2 and 3 abnormalities aims to improve both the function and the aesthetic appearance of the patient. Both aspects are equally important in achieving optimal results.
When planning surgery, the surgeon must understand the possible effects of maxillomandibular manipulation not only on the soft tissues that overlie the maxilla and mandible but also on other structures that play a significant role in the overall aesthetic balance of the face, such as the nose. The patient needs to be informed of these possible changes and may need to be aware that future nasal surgery may be necessary to create a cosmetically pleasing result.
Widening of the nasal ala is often associated with maxillary advancement. This effect would be ideal in a patient with narrow nostrils but not in someone who already has a wide nasal base. To combat this effect, the placement of an alar base or alar cinching suture has been used to decrease the amount of this movement.1-2 Maxillary repositioning also affects nasal projection and the nasolabial angle but appears to have less predictable results.3 Some studies report an increase in projection and angle measurements with maxillary advancement, while others report no significant change.1, 4 Clearly, predicting these soft tissue changes has proven to be challenging.
Currently, measuring and analyzing photographs and radiographs have largely determined our present understanding of facial aesthetics and proportion. Numerous articles have described measurements or ratios obtained from either the profile or the frontal view that indicate aesthetically pleasing proportions of the face.5 Analysis of radiographs in the form of lateral cephalograms has also been used to assess hard and soft tissue changes. However, these pictures evaluate the head only in the midsagittal plane, thereby neglecting most of the soft tissues of the face. The shortcomings offered by conventional photographs and radiographs lie in the fact that they represent objects in 2 dimensions. The human body is a 3-dimensional (3D) object, and any changes, whether from movement during facial expression or from surgery, take place in 3 dimensions.
In the surgical realm, many computer programs for 2-dimensional surgical treatment planning that combine skeletal and soft tissue analysis are available. However, flaws exist in representing a 3D structure in 2-dimensional form: facial depth and shape are not accounted for.6 Recent advances in technology have generated a variety of 3D techniques to capture facial topography and overcome the deficiencies of conventional photographic and radiographic methods. They have included 3D cephalometry,7 morphoanalysis,8 moire topography,9 3D computed tomography,10 3D magnetic resonance imaging, 3D ultrasonography,11 laser scanning,5, 12 and digital stereophotogrammetry.9, 13
In general, few studies discuss the nasal changes that occur with orthognathic surgery, and those that do focus only on the effects of maxillary advancement by using conventional imaging techniques. Among published 3D studies, only one has focused on the nasal changes that occur after surgery, but these were in patients with cleft lip and palate,14 while others have discussed soft tissue changes.15
The purpose of this study was to evaluate nasal changes in a non–cleft lip and palate population after maxillomandibular surgery using images taken with a 3D digital camera (3dMDface System; 3dMD, Atlanta, Ga). This camera arrangement is a fast and accurate 3D modality imaging system that we used to examine not only the effects of maxillary advancement on the nose but also the effects on nasal anatomy from maxillary movement with impaction (upward rotation) and maxillary movement with lengthening (downward rotation). The soft tissue changes on a select number of patients will also be discussed.
Between January 1, 2000, and June 30, 2004, a total of 54 patients (39 female and 15 male) were identified as having undergone maxillomandibular surgery and having had preoperative and postoperative photographs taken with the 3D digital camera system. Of these patients, 32 (26 female and 6 male) had data and operative notes documenting the dentofacial abnormality in addition to the distance and direction of the movement that was performed. Mean patient age was 26.3 years (range, 17-61 years). Of the 32 patients, 23 had class 3 occlusion, 5 patients had class 2 occlusion, and 4 patients had an open bite deformity (Table 1).
All patient photographs were taken with the 3D camera system (Figure 1). This system consists of 6 digital cameras, 3 on each side of the patient. The patient is seated in a chair placed at a set distance from the cameras and positioned with a slight elevation of the chin above the Frankfort horizontal plane to gain maximum light exposure of the face. A random light pattern is projected on the subject, and an image is captured by the precisely synchronized digital cameras set at various angles in an optimum configuration. Picture accuracy is within 1 mm, with a resolution up to 40 000 polygons per square inch, and texture image is in 24-bit color. Image capture takes only 2 milliseconds, creating little distortion and making it very useful in children. Once the picture is taken, the 6 separate images are automatically merged to produce a single 3D polygon surface mesh. This is followed by layering soft tissue color and features over the wire frame, which results in a 3D image that can be rotated and manipulated to be viewed from any desired angle. Distances, angles, and volumetric data can subsequently be calculated by the computer software, in addition to facial shape, texture, and skin tone in 3 dimensions.16
A single point measurement on the image results in 3 numbers corresponding to its position on the x, y, and z planes. When point-to-point measurements are performed with the software, both linear distances from point to point and distance over the surface topography are calculated and displayed. In our study, only linear distances were recorded.
The following measurements were taken from the digital images: (1) interalar width (IAW), corresponding to the widest distance from ala to ala seen on frontal view (Figure 2); (2) internostril width (INW), which measured the widest distance from the lateral portion of each nostril (Figure 3); (3) nasal tip projection (NTP), which measured the distance from the subnasale, or base of the nose, to the pronasale, the most anterior midline surface point at the tip of the nose (Figure 4); and (4) columellar length (CL), which corresponds to the distance from the subnasale to the top point of the columella, which is the most anterior point of the nostril rim (Figure 4). Three separate point-to-point measurements were calculated, and the average of these 3 numbers was used as the final distance.
In addition, the midline profile images were extracted from the 3D digital picture. By means of imaging software (Adobe Photoshop; Adobe Systems Inc, San Jose, Calif), the nasolabial angle was calculated from intersecting lines drawn from the top point of the columella to the subnasale and from the subnasale to the labrale superius, which is the most anterior midline surface point on the upper lip.
Preoperative (Figure 5) and postoperative (Figure 6) photographs were taken approximately 3 to 6 months after the procedure. Preoperative and postoperative values were compared and statistical significance was determined by a paired t test.
Histograms of color-based soft tissue differences based on the preoperative and postoperative digital images were also evaluated for a select group of patients undergoing orthognathic surgery (Figures 7, 8, 9, and 10). This is done by performing a registration, which is the alignment of 2 (or more) surfaces so that they overlay each other in 3D space. These surfaces can then be compared by measuring the distance between them, and the results are easily displayed on a histogram (Figure 8 and Figure 10). This is very useful in determining the changes that occur over time, eg, preoperatively and postoperatively. Registration can be performed in 1 of 2 ways: surface-based registration or registration via selected points. The former can be performed if 2 surfaces are relatively close together and can be applied to the whole surface or a selected area. For example, if postoperative changes after rhinoplasty surgery are evaluated, areas of the face that have not moved, such as the eyes or mouth, would be selected. Registration via selected points is useful for comparing surfaces based on specific landmarks. Landmarks on each surface, such as the medial and lateral canthi, would be selected and used as comparison points for registration.
After nasotracheal intubation, 0.5% bupivacaine hydrochloride with 1:100 000 units of epinephrine was infiltrated into the mucobuccal fold of the maxilla and mandible. Attention was first turned toward the maxilla, where an incision was made in the upper gingivobuccal sulcus down to the bone. The periosteum was elevated, exposing the nasal spine, pyriform aperture, and the face of the maxilla. After the infraorbital nerve was identified to avoid injury, a maxillary Le Fort I osteotomy was performed and down-fractured, which allowed this segment to be manipulated forward, backward, upward, and/or downward. If maxillary impaction was being performed, a midsagittal groove was fashioned to accommodate the vomer and septum. If maxillary expansion was necessary, a 2-piece osteotomy was performed, with a palatal splint used to establish the planned width. The mobile maxillary segment was subsequently placed at its predetermined position and secured with the use of a mini–Luhr plating system. A 3-0 polyglactin suture was used to cinch the alar base. A midline V-Y extension was also performed with a 0.5-cm vertical limb. The mucosa was subsequently closed with 3-0 polyglactin sutures.
A gingivobuccal incision was then made in the mandible, exposing the bone. A sagittal split osteotomy was performed bilaterally. Care was taken to identify and preserve the inferior alveolar neurovascular bundle. The mandibular segment was advanced, set back, and/or rotated into position, aided by the use of a preformed acrylic splint and secured with the Luhr screw system. If a setback was performed, it was occasionally necessary to remove a fragment from the anterior portion of the proximal fragment. The wound was irrigated and closed with 3-0 polyglactin sutures. The patient was subsequently placed in temporary fixation with the use of elastics.
Among the 23 patients with class 3 abnormalities, 5 underwent maxillary advancement without any rotation. All of these patients showed an increase in the measured IAWand INW (Table 2). Measured differences ranged from 0.219 mm to 6.128 mm and 0.822 mm to 4.736 mm, respectively. It should be noted that increases in IAW did not directly correlate with increases in INW. Paired t tests showed P = .03 and .02, respectively. Nine patients had maxillary advancement with upward rotation. Interestingly, 2 of these (patients 6 and 9) actually showed a decrease in the measured IAW but not in INW. Measured changes ranged from −0.293 to 4.938 for IAW (P = .005) and 0.803 to 4.438 for INW (P < .001).
Maxillary advancement with downward maxillary movement was also performed in 9 patients. All of these patients displayed increases in IAW (range, 0.100-4.424 mm; P = .01). However, 2 patients (patients 15 and 19) showed decreases in INW (range, –0.526 to 6.57 mm). The difference was statistically significant (P = .04) (Table 2).
Five patients had class 2 occlusion. All underwent advancement with upward maxillary movement. The IAW increased in all patients (range, 0.263-2.474 mm; P = .04). One patient (patient 5) had a decrease in INW (range, −0.207 to 2.022 mm; P = .05) (Table 2). Four patients had open bite deformities with maxillary advancement. All 4 of these patients showed increases in IAW and INW (Table 2).
Of the 5 patients with class 3 occlusion who underwent maxillary advancement without rotation, 4 had an increase in their projection. Only 1 patient (patient 2) had a decrease in NTP as well as a decrease in CL. Another patient (patient 4) also showed a decrease in CL but not in NP. These changes were not statistically significant (P = .9 for NTP and P = .4 for CL) (Table 3).
Six patients with upward rotation and advancement of the maxilla showed decreases in NTP (P = .15). Six also showed decreases in CL (P = .30). Downward rotation also resulted in 6 patients having a decrease in NTP (P = .49), but only 4 patients had decreases in CL (P = .58) (Table 3).
Of the 5 patients with class 2 abnormalities, 2 had a decrease in NTP whereas 4 showed a decrease in CL. These trends were not statistically significant (P = .90 and P = .81, respectively). In the open bite group, NTP decreased in half of the patients, with all 4 displaying a decrease in CL (Table 3).
The nasolabial angle showed both increases and decreases in the group without maxillary rotation (P = .48). Downward movement of the maxilla showed a trend toward a decrease in the nasolabial angle, but this was not statistically significant. Only patients with class 3 abnormalities who underwent upward rotation of the maxilla with advancement had a statistically significant decrease in the measured nasolabial angle (P = .004) (Table 4).
Although there was no significance, all patients in the class 2 group showed increases in the nasolabial angle (Table 4). Statistical differences for patients in the open bite group were not determined (Table 4).
Soft tissue changes of the nasomaxillary region invariably occur on manipulation of the maxillomandibular area. Our ability to accurately plan postoperative changes relies on being able to correlate the soft tissue response of the nose and upper lip area to the underlying osseous movement.
Traditionally, 2-dimensional photographs and radiographs have been used to document these changes. However, the human body is a 3D object, and any changes, whether from movement during facial expression or from surgery, are in 3 dimensions. The use of 3D imaging has wide application in facial plastic and reconstructive surgery. Various techniques have been developed to overcome the shortcomings of conventional 2-dimensional imaging. These include 3D cephalometry, morphoanalysis, moire topography, 3D computed tomography, 3D magnetic resonance imaging, 3D ultrasonography, laser scanning, and digital stereophotogrammetry.
In our study, we used a system composed of 6 digital cameras, with 3 on each side of the patient. It is a safe and noninvasive technique that captures superior-quality “external surface” images in less than 2 milliseconds per frame, which makes it ideal in children. The data are processed, creating an accurate digital model of the patient that is ready for clinical use. Other advantages over 2-dimensional techniques include the ability to calculate area, volume, and the skin texture of specific regions by using the imaging software.
The 3D image captured relies on the principle of stereo photogrammetry, which has been in existence for many years. It uses stereo triangulation, whereby (at least) 2 cameras are positioned as a stereo pair to identify unique external surface features. By calculating the distance between the 2 cameras and their focal length in relation to the object, the shape of that object can be formulated.17 Additionally, instead of using the person’s natural skin patterns, this approach incorporates projection of a unique random light pattern that is used as the foundation for triangulating the geometry in 3 dimensions. This active stereo technique tends to be more resilient to variances in lighting conditions and enables the use of a wider range of camera sensors (in this case, 6 cameras) because the controlled random texture is momentarily projected onto the surface of the subject. After the image is taken, only 1 image is required because the software allows the picture to be manipulated and viewed from any angle.
The disadvantages of the system are that carefully controlled lighting conditions are required; expensive high-resolution cameras are also necessary, making it currently a nonportable system; and the cost of the setup can reach $70 000 or more. In addition, as is inherent in any new technological advance, the accuracy of these newly developed 3D imaging systems in recording facial morphologic features must first be assessed. Several recent studies have focused on determining the reproducibility of identifying landmarks by using various 3D modalities.18-21 Accuracy values range from 0.2 to 1 mm, meaning that these imaging techniques are extremely reliable in recording facial morphologic features and could be used to acquire normal data in 3 dimensions and to measure and calculate changes after surgery.
With the use of the 3D digital camera system, our data clearly demonstrate that statistical changes in the IAW and INW occur with movement of the maxilla. Widening of these 2 entities occurs with maxillary movement anteriorly without rotation and with rotation upward and downward. These changes occur even after the use of an alar cinching suture, which has been thought to control and decrease the amount of alar base flaring that is normally associated with maxillary advancement.1-2 It would be interesting to compare these findings with those of patients who did not have the alar base suture placed; however, this procedure is almost routinely performed.
As a result of the alar widening, the nostrils display a clinically visible flaring from a narrow taper to a more ovoid shape. Erbe et al22 studied patients who underwent maxillary advancement with maxillary impaction and showed that, despite superior maxillary movement, there is a significant increase in the external nasal valve area. Interestingly, they also evaluated the internal nasal valve area by using acoustic rhinometry, where a decrease in the area was seen; however, no effect on total nasal airflow was seen. On the other hand, Gotzfried and Masing23 showed that patients with cleft lip or palate who underwent midfacial osteotomies with advancement and downward movement had an increase in nasal airflow determined by rhinomanometric measurements. Unfortunately, we were not able to correlate how much the nose would widen with respect to the distance of maxillary movement.
Our results with maxillary advancement show equivocal results with NTP and CL. Only in patients who had maxillary movement with upward rotation was there a non–statistically significant trend toward a decrease in NTP and CL.
Other studies that discussed NTP with respect to orthognathic surgery also show varied results. Schendel and Williamson24 observed an average of 2.4 mm of tip elevation in patients who underwent an average of 6.4 mm of maxillary advancement and intrusion. Betts et al25 showed a 65% increase in nasal width and a 59% decrease in tip height in 26 patients who underwent advancement. On the other hand, NTP remained unchanged in other studies.1, 14
The fact that there is no change in NTP on maxillary advancement can be understood in light of the work of Freihofer.4 He analyzed 25 patients without cleft and 25 with cleft lip or palate who underwent Le Fort I maxillary advancement and observed that the nasal base was advanced by a ratio of 4:7 with the maxilla, while the nasal tip was advanced by a ratio of 1:3 with the maxilla. Clinically, this results in a nose that becomes flatter as the nasal base is advanced twice the amount of the tip. Anatomically, anterior-posterior compression of the cartilaginous nasal skeleton occurs, resulting in a reduction in NTP and a compensatory increase in nasal width.
Westermark et al1 described the use of the alar cinch suture to affect the IAW and noted that it can also cause an increase in the nasolabial angle. Presumably, this can be attributed to the fact that the suture crosses the midline circumferentially and compresses the soft tissue in the nasolabial region, thereby functioning like a plumping graft. To support this, our results did show a statistically significant change in the nasolabial angle when maxillary advancement was combined with impaction. However, our results also clearly demonstrated cases, especially those that had advancement with downward movement, in which a decrease in the nasolabial angle was calculated (although it was not statistically significant).
In a preliminary examination with the 3D technology, we studied soft tissue changes after orthognathic surgery. After the preoperative and postoperative photographs were aligned, the 2 pictures were registered, which resulted in a color histogram showing positive, negative, and neutral changes to the image. Clearly, advancement of the maxilla resulted in red coloration of the upper lip, nasolabial area, and lower cheek, which indicated that a positive postoperative change occurred in these areas. Blue areas represented negative soft tissue changes seen mostly around the mental area, consistent with posterior movement of the mandible. On the other hand, patients with class 2 occlusion who underwent mandibular advancement also showed positive changes in the lower third of the face.
This project opens many potential areas for analysis. Studies in 3D measurements and in the characterization of surface anatomy are fundamental in objectively analyzing facial anatomy. The various techniques available can be used to make accurate surface measurements, assess volumetric changes, and capture the facial shape, texture, and skin tone and evaluate them in 3 dimensions.
Recently, 3dMD launched its 4-dimensional technology–based 3dMDface Dynamic System. This system combines 3D imaging technology with the element of time. By capturing high-quality 3D geometry and texture data of a subject at high frame rates, the next level of detail and data from 2-dimensional imaging can be acquired, such as determining and quantifying emotional expressions and subtle nuances that happen during speech.
The ideal 3D imaging system is one that is accurate, able to display fast capture times and images, reliable in archiving and storing data, easy to use, and cost-effective. As the availability of these systems increases, they will present new opportunities for the facial plastic surgeon and will revolutionize planning, executing, and assessing outcomes in patients undergoing surgery of the head and neck. Their potential is endless.
As was demonstrated with 3D digital imaging techniques, changes to the nose clearly occur after orthognathic surgery. A statistically significant increase was noted in postoperative IAW and INW with maxillary movement. Unfortunately, no predictable correlation could be demonstrated between degree of maxillary movement and measured changes in rim width, nostril width, NTP, CL, and nasolabial angle. Interestingly, maxillary advancement did not show any significant change with NTP or CL, with data showing both increases and decreases in measurements. However, the nasolabial angle in patients who underwent maxillary advancement with impaction (upward rotation) was the only measurement that showed a statistically significant increase. The result of these findings are important because patients must be properly informed of these possible changes that can affect the overall aesthetics of the face, keeping in mind that further surgery may need to be performed. These findings also demonstrate that this digital imaging system can be an invaluable diagnostic tool in analyzing the head and neck region in 3 dimensions.
Accepted for Publication: June 21, 2005.