The objective of this research was to evaluate the correlation between 3-dimensional (3D) lip vermilion (LV) morphology and skeletal patterns as well as incisor measurements in young Chinese adults.
In all, 240 young adults were enrolled; these included 80 patients each with skeletal Class I, Class II, and Class III malocclusions, respectively. Each sagittal skeletal pattern included 40 male and 40 female subjects. Twenty-two 3D LV measurements were obtained from 3D facial scans. Skeletal and incisor measurements were evaluated on lateral cephalograms. Correlation and regression analysis were performed between soft and hard tissue measurements.
Six of 22 LV measurements showed significant differences between male and female subjects. The 3D LV morphology showed significant differences with respect to different skeletal patterns and sex. Adults with skeletal Class III malocclusion tended to have thinner upper vermilion and fuller lower vermilion than subjects with skeletal Class II and III malocclusion. The mandibular plane angle negatively correlated with the upper-lower vermilion midsagittal curve length and surface area ratio in adults with skeletal Class I and II malocclusion, yet the vertical facial skeletal type showed no correlation in adults with skeletal Class III malocclusion. The vermilion angle, central bow angle, vermilion height, vermilion midsagittal curve length, vermilion height and width ratio, and vermilion surface area showed a significant correlation with incisor measurements. Regression analysis found that the ANB angle was an important factor affecting the upper and lower vermilion midsagittal curve length and surface area ratio. Further, the vermilion height and height and width ratio were closely correlated with the interincisal (U1/L1) angle, whereas the central bow angle was closely correlated with the maxillary incisor torque.
Most LV morphology variables were correlated to skeletal patterns and incisor measurements. Skeletal Class III malocclusion showed significant differences in vermilion morphology. Both the sagittal and vertical skeletal pattern have effects on vermilion proportion. The incisor torque was closely correlated to vermilion shape and central bow angle and might influence the vermilion esthetics. However, the proportion of the upper and lower vermilion was mainly affected by the ANB angle.
We obtained and compared lip vermilion (LV), skeletal, and incisor measurements.
3D LV morphology showed significant sex- and skeletal pattern–based differences.
Most LV morphology variables correlated with skeletal patterns and incisor measurements.
Adults with skeletal Class III malocclusion showed significant differences in LV morphology.
The proportion of upper and lower vermilion was mainly affected by the ANB angle.
Facial esthetics are a very important aspect of people’s everyday life. During interpersonal communication and interaction, the lip, among all other facial features, attracts people’s attention at first sight. According to previous studies, the lip is one of the determining factors of facial esthetics. , Therefore, both orthodontics practitioners and patients focus a lot of attention on lip changes during orthodontic treatment because the initial position and shape of the lip is a deciding factor that cannot be neglected while developing an orthodontic treatment plan.
Most previous relevant studies have mainly focused on the protrusion of the lip and judged the lip on the sagittal plane by using lateral cephalograms because of the limited technology. However, patients tend to observe their lip multidirectionally, and the bright-colored lip vermilion (LV) typically stands out. In clinical practice, we notice that 2 patients with almost identical lip protrusion might have distinctly different LV shape from the frontal view, which means that an ideal profile sometimes may not always guarantee a satisfactory appearance of the vermilion shape from the frontal view. Thus, it is still unclear whether the determining factor of LV morphology is influenced by the skeletal pattern, the teeth, or only determined by the inherent lip structure.
Using lateral cephalograms and 2-dimensional (2D) digital photograms, LV morphology was evaluated from the frontal and lateral view in previous studies. , Tanikawa et al outlined the LV profile on lateral facial photographs and showed its association with dental position and skeletal pattern. Furthermore, a significant difference in lip morphology has been reported among the 3 sagittal skeletal patterns. However, the information provided by 2D images has certain limitations. The actual border of the LV cannot be seen from the lateral view, neither on photographs nor in the cephalograms. Without color information, it would be tough to determine the boundary of the upper lip mucosa and skin on lateral cephalography based entirely on the profile silhouettes. Even if the lateral photographs captured the vermilion color, the arciform vermilion border was hidden from the lateral view, which means that lip bow morphology information was unavailable. Using the 2D frontal photographs, the boundary of the LV was obtainable, but the measurement of curve and surface was not. There are certain challenges in describing LV morphology with just linear and angular dimensions on a flat image without surface information. In addition, it is hard to ensure the measurement accuracy of 2D methods on account of the rather unstable head position.
The limitation of 2D methods suggests that maybe the LV needs to be studied in 3-dimension. With the development of stereo photography, the structured light scanning system can capture the 3-dimensional (3D) facial image in a few seconds and provide soft-tissue measurements efficiently. The measures and indicators of the lip extend to the curves, surfaces, and volume on 3D images. To assist the diagnosis and treatment of clinical orthodontics, some scholars have analyzed the standard values of the 3D indexes of LV in different age groups of patients. , To further evaluate the LV esthetics, Jang et al analyzed the 3D facial characteristics of adults with the preferred appearance and obtained the esthetic reference value of young adults.
Orthodontists often wonder whether they can improve facial appearance by improving lip esthetics through orthodontic treatment. Despite this, studies analyzing the relationship between hard tissue and 3D LV are scarce. Therefore, the ability of the orthodontic treatment to influence lip aesthetics is not clearly defined. Our study aimed to understand the relationship between the morphologic characteristics of LV and dental and skeletal factors in young Chinese adults by using 3D facial scanning techniques.
Material and methods
This study used a cross-sectional design. The study subjects were recruited from patients attending the Department of Orthodontics, Peking University, School and Hospital of Stomatology, Beijing, China.
The inclusion criteria were as follows: Chinese adults aged 18-30 years with overall good health and a body mass index in the range of 18-24 kg/m 2 . The exclusion criteria were as follows: patients with defective dentition, cleft lip and palate or other craniofacial syndromes, and obvious facial asymmetry, and those that had undergone prior facial surgery.
In all, 240 young Chinese adults were enrolled in this study and divided into 6 groups after obtaining written informed consent. The subjects were divided according to sex and sagittal skeletal patterns. The skeletal pattern was determined by the ANB angle: 0° ≤ ANB < 5° was classified as skeletal Class I malocclusion; ANB ≥5°, was classified as skeletal Class II malocclusion; and ANB <0°, was classified as skeletal Class III malocclusion. There were 80 subjects each in the skeletal Class I, II, and III groups. Each skeletal Class included 40 male and 40 female participants. The sample size was calculated using previous related articles. This study was reviewed and approved by the ethics committee of Peking University School and Hospital of Stomatology. (IRB no. PKUSSIRB-201631134a).
Each patient had complete treatment records, including cephalometric radiographs, photographs, and plaster models. Three-dimensional facial scans were acquired with the lips relaxed, using a 3D structured light scanning system (accuracy, ± 0.2 mm; 3Dshape, Erlangen, Germany) ( Fig 1 , A ). The scanning system consisted of a charge-coupled device camera, a projector, and 2 mirrors. This system can output 3D face models with real color texture and approximately 40,000 triangles. Three-dimensional evaluation of LV was completed on Rapidform (Inus Technology, Seoul, Korea, 2006) ( Fig 1 , B ).
The 3D coordinate system of facial soft tissue was first set after obtaining the 3D facial scan images. The steps of the method are as follows: (1) horizontal reference plane: create the shortest curve connecting the bilateral outer canthi along the curved surface, set the horizontal plane using bilateral outer canthi point and the midpoint of the curve mentioned above ( Fig 2 , A ); (2) midsagittal plane: connect the bilateral outer cathi point, build a perpendicular plane that passes through the pronasale point ( Fig 2 , B ); and (3) coronal reference plane: perpendicular to the sagittal and horizontal reference plane, and passes through the cross point of the midsagittal plane and the line connecting the bilateral outer cathi ( Fig 2 , C ).
Ten representative 3D facial landmarks with high repeatability were marked in the lip area, according to the definition in 3 previous 3D studies. , , The definition of landmarks are presented in Table I .
|Cheilion (chr, chl)||The merged point of the upper and lower vermilion arch|
|Labial superius (ls)||The superior border of the upper LV on the midsagittal plane|
|Labial inferius (li)||The inferior border of the upper LV on the midsagittal plane|
|Crista philtra (cphr, cphl)||The merged point of the philtrum ridges and LV|
|Soft-tissue-B point (B’)||The deepest point on the mentolabial sulcus on the midsagittal plane|
|Subnasale (sn)||The merged point of the nasal septum inferior border and upper cutaneous lip on the midsagittal plane|
|Stomion superius (stos)||The superior border of the oral fissure on the midsagittal plane|
|Stomion inferius (stoi)||The inferior border of the oral fissure on the midsagittal plane|
To evaluate the 3D vermilion morphology, we introduced 22 LV measurements in the present study, including 9 angular dimensions, 4 curve length dimensions, 3 straight-line dimensions, 2 surface area measurements, and 4 ratios representing the LV shape and proportion that were accordingly calculated ( Fig 3 ). The definitions of 3D measurements are presented in Table II .
|Right/left upper vermilion angle (°)||Angle formed by the ch(r,l)-cph(r,l) and chr-chl lines|
|Right/left lower vermilion angle (°)||Angle formed by the ch(r,l)-li and chr-chl lines|
|Upper vermilion base angle (°)||Angle formed by the chr-ls and chl-ls lines|
|Lower vermilion base angle (°)||Angle formed by the chr-li and chl-li lines|
|Right/left cupid’s bow angle (°)||Angle formed by the ch(r,l)-cph(r,l) and cph(r,l)-ls lines|
|Central bow angle (°)||Angle formed by the cphr-ls and cphl-ls lines|
|Upper cutaneous lip length (mm)||Distance between sn and ls on the midsagittal plane along the surface|
|Lower cutaneous lip length (mm)||Distance between li and B’ on the midsagittal plane along the surface|
|Vermilion height (mm)||Straight-line distance between ls and li|
|Mouth width (mm)||Straight-line distance between chr and chl|
|Cupid’s bow width (mm)||Straight-line distance between cphr and cphl|
|Upper LV midsagittal curve length (mm)||Distance between ls and stos on the midsagittal plane along the surface|
|Lower LV midsagittal curve length (mm)||Distance between li and stoi on the midsagittal plane along the surface|
|Upper vermilion surface area (mm 2 )||Superficial area of upper LV|
|Lower vermilion surface area (mm 2 )||Superficial area of lower LV|
|Vermilion height-width ratio||Vermilion height/mouth width|
|Cupid’s bow width ratio||Cupid’s bow width/mouth width|
|Vermilion upper lip/lower lip midsagittal curve ratio||Upper LV fullness/lower LV fullness|
|Surface area ratio||Upper vermilion surface area/lower vermilion surface area|
Lateral cephalograms of each subject were obtained with lips in the relaxed position. Skeletal and dental measurements were evaluated on lateral cephalograms. The skeletal measurements included 6 angular dimensions (SNA, SNB, ANB, Mp-SN, MP/FH, MP/PP), and the dental measurements included 7 angular dimensions (U1/NA, SN-U1, PP-U1, U1/FH, L1/MP, L1/NB, U1/L1) and 4 linear dimensions (U1-NA, L1-NB, U1-AP, L1-AP). The definitions of cephalometric measurements are presented in Supplementary Table .
All statistical analyses were performed with SPSS software (version 19.0; IBM Corp, Armonk, NY). All 3D measurements were obtained by a single operator (Z.Y.L.) twice, and the mean values were used. Consistency check and normality test were done first. Intraclass correlation coefficients of all the LV measurements were greater than 0.9. A t test was used for normally distributed variables, and a rank sum test was used for nonnormally distributed variables to compare the LV measurements on the basis of sex. One-way analysis of variance (for normally distributed variables) and rank sum test (for nonnormally distributed variables) were performed to compare the LV measurements between each skeletal Class pattern. Spearman correlation test and stepwise regression analysis were performed to evaluate the relationship between hard tissue profile and LV shape.
As shown in Table III , significant differences in 6 vermilion morphology measurements were found between male and female subjects in each skeletal pattern. The vermilion upper lip/lower lip midsagittal curve ratio was lower in male than female patients in all 3 classes. The surface area ratio was also lower in male subjects with skeletal Class II and III malocclusion than female subjects. The lower LV midsagittal curve length and lower vermilion surface area were significantly greater in male subjects with skeletal Class II and III malocclusion than in female subjects. Mouth width was significantly greater in males with skeletal Class II malocclusion than in female patients. The central bow angle was lower in males with skeletal Class III malocclusion than in female patients.
|Variables||Class I||Class II||Class III|
|Male||Female||P value||Male||Female||P value||Male||Female||P value|
|Right upper vermilion angle (°)||45.89 ± 2.71||46.67 ± 2.59||0.194||45.77 ± 3.02||46.59 ± 3.21||0.239||44.83 ± 2.08||44.5 ± 2.39||0.513|
|Left upper vermilion angle (°)||45.87 ± 2.66||46.43 ± 2.7||0.346||45.69 ± 2.87||46.5 ± 3.12||0.231||44.93 ± 2.12||44.68 ± 2.25||0.615|
|Right lower vermilion angle (°)||30.93 ± 3.02||30.69 ± 2.73||0.719||31.21 ± 2.95||30.87 ± 2.9||0.564||32.44 ± 2.45||32.05 ± 2.52||0.489|
|Left lower vermilion angle (°)||30.79 ± 3.13||31.17 ± 2.8||0.566||31.04 ± 2.78||30.72 ± 2.92||0.616||32.6 ± 2.47||32.18 ± 2.55||0.454|
|Upper vermilion base angle (°)||104.96 ± 5.39||103.76 ± 5.76||0.340||104.34 ± 6.15||102.7 ± 5.37||0.207||106.09 ± 5.91||107.46 ± 5.86||0.302|
|Lower vermilion base angle (°)||118.29 ± 6.02||118.14 ± 5.32||0.906||117.75 ± 5.51||118.41 ± 5.8||0.604||114.96 ± 4.89||115.77 ± 5.02||0.467|
|Right cupid’s bow angle (°)||126.34 ± 4.08||126.78 ± 4.9||0.338||127 ± 1.81||127.45 ± 1.84||0.275||128.48 ± 2.23||128.36 ± 2.67||0.838|
|Left cupid’s bow angle (°)||126.21 ± 3.22||126.74 ± 4.7||0.563||127.32 ± 1.62||127.26 ± 1.99||0.873||128.44 ± 2.13||128.63 ± 2.91||0.916|
|Central bow angle (°)||133.49 ± 6.53||134.57 ± 3.71||0.593||132.9 ± 4.31||133.99 ± 3.99||0.243||134.38 ± 3.24||136.22 ± 3.86||0.023∗|
|Upper cutaneous lip length (mm)||15.88 ± 1.54||15.3 ± 1.72||0.120||15.72 ± 1.89||15.44 ± 1.69||0.488||16.86 ± 1.92||16.68 ± 2.03||0.689|
|Lower cutaneous lip length (mm)||11.39 ± 1.44||10.8 ± 1.42||0.070||11.41 ± 1.22||11.23 ± 1.1||0.488||11.47 ± 1.44||11.29 ± 2.25||0.773|
|Vermilion height (mm)||18.05 ± 2.72||18.07 ± 2.68||0.974||18.26 ± 1.86||18.68 ± 1.98||0.329||16.9 ± 2.5||17.19 ± 2.9||0.633|
|Mouth width (mm)||50.23 ± 3.63||48.69 ± 3.84||0.071||49.52 ± 1.76||48.63 ± 1.93||0.034∗||51.46 ± 2.85||50.45 ± 3.05||0.129|
|Cupid’s bow width (mm)||11.88 ± 1.62||11.62 ± 1.49||0.590||11.91 ± 1.04||11.94 ± 0.9||0.939||11.85 ± 1.25||11.98 ± 1.21||0.645|
|Upper LV midsagittal curve length (mm)||9.54 ± 0.73||9.66 ± 0.88||0.620||9.87 ± 0.45||10.01 ± 0.49||0.203||8.83 ± 0.98||9.17 ± 0.92||0.123|
|Lower LV midsagittal curve length (mm)||10.18 ± 0.56||9.99 ± 0.76||0.266||9.83 ± 0.54||9.51 ± 0.66||0.018∗||11.02 ± 0.98||10.4 ± 1.14||0.006∗∗|
|Upper vermilion surface area (mm 2 )||494.13 ± 59.18||487.71 ± 86.11||0.699||500.45 ± 36.31||501.24 ± 32.09||0.918||478.63 ± 44.99||463.85 ± 46.71||0.153|
|Lower vermilion surface area (mm 2 )||548.24 ± 55.79||521.91 ± 83.49||0.169||513.87 ± 36.69||499.17 ± 32.99||0.046∗||588.83 ± 66.21||538.66 ± 56.54||0.001∗∗|
|Vermilion height-width ratio||0.36 ± 0.06||0.37 ± 0.06||0.353||0.37 ± 0.04||0.38 ± 0.04||0.075||0.33 ± 0.04||0.34 ± 0.06||0.259|
|Cupid’s bow width ratio||0.24 ± 0.03||0.24 ± 0.03||0.538||0.24 ± 0.02||0.25 ± 0.02||0.181||0.23 ± 0.02||0.24 ± 0.02||0.108|
|Vermilion upper lip and lower lip midsagittal curve ratio||0.94 ± 0.05||0.97 ± 0.05||0.015∗||1.01 ± 0.06||1.06 ± 0.07||0.001∗∗||0.8 ± 0.07||0.88 ± 0.05||<0.001∗∗|
|Surface area ratio||0.9 ± 0.09||0.94 ± 0.09||0.103||0.98 ± 0.07||1.01 ± 0.06||0.047∗||0.82 ± 0.05||0.86 ± 0.06||<0.001∗∗|
We used analysis of variance and post-hoc testing to compare the differences among the 3 skeletal classes in male and female patients. The 3D LV morphology showed significant differences in different skeletal classes. Male and female patients showed similar differences in vermilion morphology among the 3 classes ( Table IV ). As shown in Table IV , patients with skeletal Class I and II malocclusion showed slight differences. The vermilion midsagittal curve length and surface area ratio were greater in patients with skeletal Class II malocclusion than those with skeletal Class I malocclusion. The upper vermilion angle, upper vermilion midsagittal curve length, upper vermilion surface area, upper cutaneous vermilion lip, vermilion height, vermilion height-width ratio, vermilion midsagittal curve length ratio, and surface area ratio were significantly smaller in patients with skeletal Class III malocclusion than those with skeletal Class I and II malocclusion. Lower vermilion angle, lower vermilion midsagittal curve length and surface area, cupid’s bow angle, and central bow angle were all significantly greater in patients with skeletal Class III malocclusion than those with skeletal Class I and II malocclusion.
|P value||Post-hoc test||P value||Post-hoc test|
|Class I vs Class II||Class I vs Class III||Class II vs Class III||Class I vs Class II||Class I vs Class III||Class II vs Class III|
|Right upper vermilion angle (°)||0.150||0.830||0.075∗||0.116||0.001∗∗||0.900||0.001∗∗||0.001∗∗|
|Left upper vermilion angle (°)||0.228||0.758||0.106||0.189||0.004∗∗||0.919||0.005∗∗||0.003∗∗|
|Right lower vermilion angle (°)||0.042∗∗||0.651||0.018∗∗||0.054∗||0.057∗||0.768||0.028∗∗||0.055∗|
|Left lower vermilion angle (°)||0.009∗∗||0.688||0.005∗∗||0.014∗∗||0.057∗||0.466||0.104||0.019∗∗|
|Upper vermilion base angle (°)||0.397||0.635||0.386||0.181||0.001∗∗||0.403||0.004∗∗||<0.001∗∗|
|Lower vermilion base angle (°)||0.017∗∗||0.662||0.008∗∗||0.025∗∗||0.058∗||0.822||0.052∗||0.030∗∗|
|Right cupid’s bow angle (°)||0.004∗∗||0.309||0.001∗∗||0.024∗∗||0.043∗∗||0.617||0.051∗||0.231|
|Left cupid’s bow angle (°)||<0.001∗∗||0.042∗∗||<0.001∗∗||0.008∗∗||0.004∗∗||0.493||0.014∗∗||0.001∗∗|
|Central bow angle (°)||0.593||0.331||0.870||0.346||0.030∗∗||0.506||0.057∗||0.011∗∗|
|Upper cutaneous lip length (mm)||0.010∗∗||0.692||0.015∗∗||0.005∗∗||0.001∗∗||0.740||0.001∗∗||0.003∗∗|
|Lower cutaneous lip length (mm)||0.961||0.940||0.786||0.845||0.093∗||0.250||045∗∗||0.366|
|Vermilion height (mm)||0.026∗∗||0.702||0.032∗∗||0.012∗∗||0.035∗∗||0.289||0.123||0.010∗∗|
|Mouth width (mm)||0.010∗∗||0.273||0.055∗||0.003∗∗||0.022∗∗||0.931||0.011∗∗||0.007∗∗|
|Cupid’s bow width (mm)||0.728||0.413||0.715||0.861||0.357||0.248||0.189||0.872|
|Upper LV midsagittal curve length (mm)||<0.001∗∗||0.082∗||<0.001∗∗||<0.001∗∗||<0.001∗∗||0.033∗∗||0.040∗||<0.001∗∗|
|Lower LV midsagittal curve length (mm)||<0.001∗∗||0.009∗∗||<0.001∗∗||<0.001∗∗||0.001∗∗||0.012∗∗||0.039∗∗||<0.001∗∗|
|Upper vermilion surface area (mm 2 )||0.114||0.555||0.150||0.043∗||0.020∗∗||0.311||0.076∗||0.006∗∗|
|Lower vermilion surface area (mm 2 )||<0.001∗∗||0.005∗∗||0.001∗∗||<0.001∗∗||0.002∗∗||0.163||0.419||<0.001∗∗|
|Vermilion height-width ratio||<0.001∗∗||0.455||0.002∗∗||<0.001∗∗||0.004∗∗||0.348||0.009∗∗||<0.001∗∗|
|Cupid’s bow width ratio||0.032∗∗||0.102||0.501||0.040∗∗||0.212||0.224||0.768||0.132|
|Vermilion upper lip and lower lip midsagittal curve ratio||<0.001∗||<0.001∗∗||<0.001∗∗||<0.001∗∗||<0.001∗∗||<0.001∗∗||<0.001∗∗||<0.001∗∗|
|Surface area ratio||<0.001∗∗||<0.001∗∗||<0.001∗∗||<0.001∗∗||<0.001∗∗||<0.001∗∗∗||<0.001∗∗∗||<0.001∗∗∗|