INTEROBSERVER AND INTRAOBSERVER VARIABILITY IN RADIOGRAPHIC BONE LEVEL ASSESSMENT AT BRANEMARK FIXTURES

The aim was to determine inter- and intraobserver variability in radio graphic bone level assessments at Branemark fixtures and to study the influence of various factors (radiographic density; projection geometr y; jaw in which the fixtures were inserted; degree of bone loss; time after fixture loading; and number of radiographs of each fixture) on t he variability. Intraoral radiographs from bridge connection and 1- an d 3-year check-ups from 15 upper and 15 lower jaws (172 fixtures) were assessed by 6 observers. Measurements were taken from a reference poi nt on the fixture to the marginal bone level, and some were repeated b y all observers after 1 month. Results showed a small interobserver va riation (0.14 mm) with the intraobserver variation (0.08 mm) as its la rgest component. The radiographic density and the degree of bone loss showed the strongest influence on the interobserver variation. The onl y variable with a significant effect on the intraobserver variation wa s the number of radiographs of each fixture. Calculated confidence val ues showed that measurement reliability can be improved by letting one observer or preferably more make several, independent readings, allow ing for the demonstration of minor differences in bone height over tim e or between implant systems.