(27) Reliability and Reproducibility in Cephalometric Tracing

Abstract: Since the development of cephalometry by Dr. Broadbent in 1930[1], discovery and appointment of new cephalometric landmarks gave rise to various standardized analytic measurements for successful orthognathic surgery planning. To this day, Steiner+Wits analysis remains as one of the most commonly used cephalometric analysis because it appropriately includes parameters that diagnose and analyze skeletal and dental structures[2]. However, there has not been any study studying the reliability and reproducibility of lateral cephalometric tracing in different skeletal classifications. The purpose of this study was to explore reproducibility and reliability in lateral cephalometric tracing of Class 2, Class 3, Pseudo-class 3 (midface deficiency), Anterior open-bite, and Asymmetric mandible in addition to observing whether there were patterns amongst trainee residents.
5 lateral cephalograms with a different skeletal deformity were obtained (Image1,2,3,4,5), and 18 OMFS residents at Thomas Jefferson University Hospital were divided into two groups based on their PGY experience: Junior or Senior. Both groups were given an introductory lecture on how to conduct lateral cephalometric tracing with Steiner+Wits analysis using Dolphin Imaging&Management Solutions®11.95 (Chatsworth, CA, USA). Measurements were collected from each resident and analyzed using R-Statistical Software (v4.2.0; RCoreTeam.2021). To determine the agreement between the groups, mean concordance correlation coefficient (MCCC) was calculated. The senior group was set as the gold standard with MCCC < 0.5 chosen to distinguish poor reproducibility. To check reliability, Two-sample T-test/Rank-sum test–depending on the normality of each dataset–was calculated. P< 0.05 was considered to have a significant difference between the groups, correlating to poor reliability. Statistically significant values with poor reproducibility and reliability values were then analyzed by observing traced lateral cephalograms and the absolute mean difference between the groups.
For reproducibility, 6 measurements had poor MCCC ( < 0.5) (table1): U1-NA in Class2 (0.29), Pog-NB in Open-bite (0.33), ANB in Class2 (0.36), L1-NB in Class3 (0.38), Lower-lip-to-E-plane in Class2 (0.38), and IMPA in Asymmetry (0.41). Despite poor MCCC, following measurements did not have statistical significance (p>0.05): U1-NA in Class2, ANB in Class2, and L1-NB in Class3. Instead, there were 3 measurements with p< 0.05 which correlated with poor MCCC ( < 0.5) (table2): POG-NB in Open-bite (mean difference:0.63mm), Lower-lip-to-E-plane in Class2 (mean difference:1.14mm), and IMPA in Asymmetry (mean difference:6.3-degrees).
Based on cephalogram analysis of those with poor MCCC and p< 0.05, following erroneous patterns in the Junior group were observed that explain discrepancies between the groups: 1. Inferiorly placed nasion due to large frontal sinus and radiolucent discrepancy over nasal bridge, and incorrectly placed pogonion’ (soft-tissue Pog) in Class2 cephalogram resulted in narrower ANB angle, wider U1-NA distance, and wider Lower-lip-to-E-plane distance; 2. More labially placed mandibular incisors due to interference from orthodontic brackets in Class3 cephalogram reflected in wider L1-NB distance; 3. Not placing pogonion at the most anterior point on the contour of the bony chin in Open-bite cephalogram reflects shorter Pog-NB in the Junior group; 4. Variation in choice of MP in Asymmetry cephalogram resulted in wider IMPA angle.
Most significant discrepancies especially in Class2 and Class3 were due to discrepancy between anatomical markers such as nasion placement, orthodontic brackets and pogonion that were not associated with skeletal deformity. Asymmetry was the most unreliable and difficult to reproduce between the groups due to two distinct MP. In conclusion, skeletal deformity adds complexity to reliability and reproducibility in lateral cephalogram tracing to untrained and inexperienced eyes as exemplified by the Junior group. In order to properly diagnose and treatment plan appropriately, experience and training are required.

References
[1] HB, Broadbent. “A New X-Ray Technique and Its Application to Orthodontia.” Angle Orthod, 1931;1(2):45–66.
[2] Steiner, Cecil C. “Cephalometrics for You and Me.” American Journal of Orthodontics, 1953;39(10):729–55, doi:10.1016/0002-9416(53)90082-7

Mean Concordance Correlation Coefficient (*: MCCC < 0.5)
T-test/Rank-sum Test (*: p < 0.05)

Lateral cephalograms (Image 1, 2, 3, 4, 5)