GP6a - Archial Skeletal Analysis is not a Significant Predictor of Adult OSA Syndrome Severity

Abstract:

Introduction: Archial analyses may be a useful tool in the diagnosis and treatment planning of adult OSA. The purpose of this study was to evaluate the association of vertical and horizontal archial analyses on predictors of the severity of OSA. The primary research question of this study was: “Does facial skeletal archial analysis predict severity of OSA?” The authors hypothesize that archial analyses will predict severity of adult OSA.

Subjects and methods: A retrospective cohort study of adult polysomnogram (PSG)-confirmed OSA subjects’ charts was designed from subjects who presented to Allegheny General Hospital Division of Oral and Maxillofacial Surgery between January 1, 1992, and December 31, 2017. Inclusion criteria included PSG-confirmed OSA (AHI > 5), diagnostic lateral cephalometric radiographs, and demographic data. Exclusion criteria included subjects with a history of maxillofacial trauma or tumor surgery, craniofacial syndromes, and inadequate chart records. Independent predictor variables were age, sex, body-mass index (BMI), anterior arc independent (continuous data), and combined horizontal and vertical analyses (categorical data). Skeletal facial types were categorized into 27 distinct facial types based on combined anterior horizontal and vertical arcs. Primary outcome variable was severity of OSA as defined by apnea-hypopnea index (AHI). Secondary outcome variables included Epworth sleepiness scale (ESS) and lowest oxygen saturation (nO2). Lateral cephalometric radiographs were analyzed with Dolphin software (v. 11.8) for Sassouni Plus archial analysis. Anterior arc analysis was recorded for maxillary incisor position to ANS (U1 - ANS arc), Point B to Point A arc (B - A arc), and pogonion to ANS arc (Pg - ANS arc).

Statistical analysis:

Standard descriptive analysis, multiple linear regression on continuous variables, and multinomial logistic regression on categorical and dummy variables were performed using XLSTAT. Stratified by sex, multilinear regression was performed on anterior arc analyses. Continuous variables (median values) and categorical data were transformed to dummy variables for multinomial logistic regression. Results were reported as mean, standard deviation, and odds ratio (OR). Standard error of measurement (SEM) was completed on repeated cephalometric analyses of 20 randomly selected radiographs. Statistical significance was set at P < .05 level.

Results: A total of 171 subject charts that met the inclusion criteria were included in this study. The cohort consisted of 124 females (72.5%). The cohort mean for age was 47 years (Standard deviation [SD] 11), for BMI 31 (6.5), for AHI 42 (30.5), for ESS 13 (4.9), and for nO2 81% (11). All SEM were greater than or equal to 0.90. Stratified by sex, and controlling for age and BMI, no statistically significant associations for anterior archial analyses (U1 - ANS arc, B - A arc, or Pg - ANS arc) and severity of adult OSAS were found. Multinomial logistic regression revealed that none of the archial skeletal classifications were significant predictors of the severity of OSAS (AHI OR = 1.0, P = .89; ESS OR = 0.99, P = .70; nO2 OR = 1.0, P = .81).

Conclusion: Within the limitations of this study, when controlling for age, sex, and BMI, anterior horizontal archial analyses were not significantly associated with, and archial skeletal facial types were not significant predictors for, OSAS severity as defined by AHI, ESS, or nO2. Additional research may be required to fully analyze the full diagnostic potential of skeletal archial analysis in diagnosis and treatment planning of subjects with OSAS.

1. Cillo JE, Thayer S, Dasheiff RM, Finn R: Relations between obstructive sleep apnea syndrome and specific cephalometric measurements, body-mass index, and apnea-hypopnea index. J Oral Maxillofac Surg 70(4):e278-e283, 2012
2. Sassouni V: A classification of skeletal facial types. Am J Orthod 55(2):109-23, 1969