Reliability of cranial base measurements on lateral skull radiographs

Abstract

To explore the reliability of identification of anatomic landmarks on lateral skull radiographs of young unaffected individuals that has conventionally been used to diagnose pathologic relationships in the craniovertebral junction. From the Helsinki longitudinal growth study, 20 randomly selected lateral radiographs were analyzed and re-analyzed by two examiners. Both located seven cephalometric landmarks based on which five measurements were calculated. The differences of results were compared. With similar method three radiographs were analysed by 11 examiners and results were compared. Some anatomic landmarks were easier to locate than others on lateral skull radiographs leading to differences in measurements based on them. We found the magnitude of the difference to be dependent on the landmark serving as reference. Inter- and intra-examiner errors were of similar magnitude, although intra-examiner error declined in the repeated landmark identification. Variation in a single landmark location had in general little effect on the measurement value. Variations in landmark location lead to differences in numeric evaluation of the anatomic relationships in the skull base area. These differences were, however, shown to have little clinical significance. Hence, the documented methods are applicable for screening of basilar pathology.