OBJECTIVE: To assess the reproducibility of repeat bitewing radiographs. METHODS: Five right and left pairs of posterior bitewing radiographs were taken without the use of beam-aiming or film-holding devices in five dry skulls with complete dentitions. The amelocemental junction (ACJ) and alveolar crest (AC) of every interdental site was traced on projected radiographs and the linear distance between the ACF and AC determined. RESULTS: There was close concordance between the ability to read the ACJ on the approximal tooth surface and the AC. The overall readability of interdental sites was 71% although sites at the extremities of radiographs were often unreadable. Kappa analysis of the readability of alveolar bone height from repeat radiographs was 0.52. There were significant differences in the reproducibility of ACJ and AC measurements from repeated radiographs (p<0.001). Intra-examiner variability was significantly less (p<0.001) than that due to repeat radiography with no significant differences between sites. The standard deviation for all surfaces from repeat readings of the same radiographs was 0.12 mm compared with 0.51 mm from repeat radiography. CONCLUSIONS: In this study a freehand technique produced a degree of imprecision in readings of alveolar bone height from successive films of the same site. In clinical terms, a difference of more than 1.4 mm between pairs of serial measurements from the same site would have to occur in order to be certain that the difference was not measurement error but real bone loss.
OBJECTIVE: To assess the reproducibility of repeat bitewing radiographs. METHODS: Five right and left pairs of posterior bitewing radiographs were taken without the use of beam-aiming or film-holding devices in five dry skulls with complete dentitions. The amelocemental junction (ACJ) and alveolar crest (AC) of every interdental site was traced on projected radiographs and the linear distance between the ACF and AC determined. RESULTS: There was close concordance between the ability to read the ACJ on the approximal tooth surface and the AC. The overall readability of interdental sites was 71% although sites at the extremities of radiographs were often unreadable. Kappa analysis of the readability of alveolar bone height from repeat radiographs was 0.52. There were significant differences in the reproducibility of ACJ and AC measurements from repeated radiographs (p<0.001). Intra-examiner variability was significantly less (p<0.001) than that due to repeat radiography with no significant differences between sites. The standard deviation for all surfaces from repeat readings of the same radiographs was 0.12 mm compared with 0.51 mm from repeat radiography. CONCLUSIONS: In this study a freehand technique produced a degree of imprecision in readings of alveolar bone height from successive films of the same site. In clinical terms, a difference of more than 1.4 mm between pairs of serial measurements from the same site would have to occur in order to be certain that the difference was not measurement error but real bone loss.