Semester

Spring

Date of Graduation

2023

Document Type

Thesis

Degree Type

MS

College

School of Dentistry

Department

Orthodontics

Committee Chair

Peter Ngan

Committee Co-Chair

Lorri Ann Morford

Committee Member

Khaled Alsharif

Committee Member

Matt Harper

Abstract

Abstract

Comparison of Skeletal Class III Malocclusion Facial-Shape Subtypes Between

Caucasian, African American, and Hispanic Populations

Rachel K. Huffman, DMD

Introduction: The development of a skeletal Class III malocclusion (Cl III) is complex as it can arise from the overgrowth of the mandible, undergrowth of the maxilla, or a combination of the two. To understand the heterogeneity of this trait clinically, studies have been conducted using facial shape analysis (e.g., Geometric morphometrics (GeoMorph)) of Caucasian and Asian populations to learn more about the principal shapes. Little is known, however, about the predominant principal shape components of Cl III within African American and Hispanic populations. The aim of this retrospective research study was to examine the principal shape components of Cl III malocclusion in patients from three different ethnic/racial populations living in Central Kentucky: namely individuals of Caucasian, African American, and Hispanic descent. Methods: The study utilized de-identified (X, Y) coordinate data generated by investigators at the University of Kentucky College of Dentistry (UKCD) in Lexington, Kentucky on Cephalometric tracings from 168 patients who were diagnosed and/or treated for a Cl III malocclusion based on a Wits appraisal of < -3.0). The (X, Y) coordinate data, from both males and females, was analyzed in MorphoJ software [17,18] using Procrustes superimposition combined with a Principal Component Analysis (PCA) to sub-classify the Class III facial shapes by ethnic/racial background. The most defining principal components of each ethnic/racial population were then compared and/or contrasted to one another within a single study. Results: The measures of age, Wits appraisal, and ANB were not normally distributed according to the Shapiro-Wilk test. While age and Wits Appraisal measurements by racial/ethnic groups were not statistically different from one another using the Kruskal-Wallis rank-sum test (p=0.084 and p=0.93, respectively), there was a statistically significant difference among the three populations for ANB (p65% or more of variation in skeletal Cl III. In the Caucasian Cl III population, Principal Component 1 (PC1) made up 32.5% of the overall facial shape and was characterized by variation in mandibular plane angle (MPA; defined as SN-GoMe (degrees, o)) and anterior facial height (SN-NPg (o)) when compared to the average Caucasian facial shape. PC1 for the African American population identified a similar variation in the MPA and anterior face height when compared to the average African American facial shape; however, these features were combined with variation in Ramus height (Ar-Go and Co-Go) and posterior facial height (S-Go) leading to a larger MPA in the African American PC1 when compared to the Caucasian PC1. The African American PC1 also included variation in the Facial Plane Angle (SN-NPg (degrees, o)), which as not detected in the Caucasian PC1. PC1 for the African American population made up 24.4% of the overall facial shape variation. In the Hispanic population, PC1 differed from the Caucasian and African American with an increase in total mandibular body length and a decrease in MPA. PC1 in the Hispanic cohort made up 34.6% variation of the sample.

Caucasian PC1 and PC4 showed a more hyperdivergent average shape while PC 2 and 3 showed a hypodivergent average shape. In contrast, PC3 was the only average shape in the African American cohort that showed a hypodivergent shape, PC1 and PC4 showed hyper-divergence and PC2 showed an average MPA with a short retruded midface and increased mandibular length. PC2, PC3, and PC4 in the Hispanic cohort show a tendency for hyper-divergency with a decreased ramus height-(Ar-Go and Co-Go) in PC3 and PC4. The Hispanic PC1 showed increased ramus length and more hypodivergent average shape. Conclusions: Skeletal Cl III features were studied in males and females using GeoMorph for three different ethnic/racial populations living in Central Kentucky (Caucasians, African Americans, and Hispanics). While the first 4 PCs of each population explained > 65% of the skeletal C III variations for the population, the order and percentage contribution of each PC differed among the Caucasian (reference population), African American and Hispanic populations.

Embargo Reason

Publication Pending

Download

Available for download on Monday, April 27, 2026

Share

COinS