Identification of the temporomandibular joint and adjacent cephalometric landmarks using a dual sensitivity...

Clinical evaluation of a cassette with dual speed screens for cephalometric radiography.

Two lateral cephalometric radiographs were taken on 20 consenting subjects using the TMJ Orthoceph Slimline Cassette System (TOSCS), incorporating circular Trimax 12 screens in the area adjacent to the temporomandibular joint, and a control cassette (Trimax 8 screens). Ten pairs of radiographs with optimal image quality were randomly presented to 10 observers trained in cephalometric interpretation. Observers rated the overall diagnostic quality of each radiograph and of the TMJ region on an ordinal scale. They then located specific landmarks and traced the TMJ anatomy using acetate overlays. Overlays were digitized by a single operator who repeated tracing placements and digitizations to determine the error of recording method. Landmark variability was compared in the x- and y-axis by the Wilcoxon matched-pairs signed ranks test (p < 0.05). Six repeat tracings were performed and assessed by percentage of repeated observations above the maximum affordable error. Fossa space values were analyzed by the coefficient of variation (CV). The variability of the angular and linear values was also compared.

TOSCS image quality was perceived as significantly better than the control. Method error was 0.34 mm in the x-axis and 0.4 mm in the y-axis. Interobserver variability was 2 to 3 times greater than intraobserver. There was less variability with TOSCS for identification of basion (x-axis), center-of-rotation (x-axis) and condyle (posterior) (x-axis). However, this was clinically insignificant. Accurate determination of the fossa space was not possible as CV varied from 23 to 84%. No differences in the variability of angular or linear values variability were found.

While observers preferred TOSCS, no significant clinical differences could be demonstrated between the two systems.