The aim of this study was to determine the reliability and the measuring error (by means of the smallest detectable error) of 11 angular and 4 linear measurements commonly used for cephalometric analysis.
Twenty-five digital lateral cephalograms were randomly selected and traced with Viewbox software (version 126.96.36.199, dHAL Software, Kifissia, Greece). This was repeated 3 times by 2 observers during 3 sessions. There was at least 1 week between each session. Differences were analyzed with a repeated measurement analysis of variance (ANOVA). Intraobserver and interobserver reliabilities were calculated with intraclass correlation coefficients (ICC) based on absolute agreement. Measurement error was determined by means of the smallest detectable difference.
The intraobserver agreement of the measurements was good (ICC >0.82). SNA, SNB, ANB, and ANS-Me had the smallest intraobserver errors for both observers (>1.86 mm or degrees). Except for SN-FH (ICC = 0.76), interobserver agreement was good (ICC >0.87).
Determining the appropriate measuring error of cephalometric measurements by means of the smallest detectable difference is necessary to find the true difference between the start and the end of active treatment. Depending on the magnitude of clinical significance, the measuring error was possibly clinically significant for all variables tested and, therefore, questions the use of these variables to detect the true treatment effect.
We orthodontists are, by necessity, very attentive to details. It is, therefore, only expected that we would be fussy about precision in cephalometrics, ascribing meaning even to the last significant digit that our computer reports. Fortunately, numerous studies have tempered this tendency by showing that cephalometric measurements have limited accuracy, repeatability, and even validity. This study further helps to put things in their proper perspective. Assume that mandibular length, as measured from condylion to gnathion (Co-Gn), has increased for a particular patient by 3 mm during treatment. Would anyone doubt that the patient has grown? The answer, according to the results of this research, might be surprising.
The authors used the concept of the smallest detectable difference (SDD) to present their findings. The SDD is the smallest difference between 2 measurements that gives us sufficient confidence to accept the change as actual and not just an accidental finding. It is obvious that a very small difference in Co-Gn could easily be attributed to measurement error. What is the threshold above which you would consider the change real?
It seems that the SDD for this specific measurement is greater than 3 mm and can exceed 4 mm if the pretreatment and posttreatment measurements are made by different observers. The errors inherent in cephalometric measurements, even if performed with the help of computer software, are so great that it makes no sense to claim evidence of a 2-mm mandibular growth change. Table IV of the article gives values for 14 commonly used measurements. When studying it, we should keep in mind that these SDDs apply to individual patients—ie, to measurements of 1 radiograph; they do not represent our limits when evaluating large samples of patients, when random errors tend to average out. Also, the SDD is the value that represents a 95% confidence limit, as is customary in statistics. It is up to you to decide whether you need such a high confidence, depending on the use of your data.
|Measurement||Unit||Observer 1||Observer 2||Mean observer 1 vs 2|