Objective: The aim of this study was to determine (1) the effect of various scan delays on pixel intensity measurements and (2) the diagnostic accuracy of occlusal caries detection.
Methods: 72 non-cavitated extracted human permanent molar teeth with sound and carious occlusal surfaces were radiographed using the DIGORA® storage phosphor plates (SPPs) (Soredex Corporation, Helsinki, Finland). Plates were scanned immediately and 10 min, 30 min, 60 min and 120 min after exposure. Three different plates were used for each scan delay. An aluminium (Al) wedge was used to evaluate the change in mean grey values (MGVs) for scan delays. Diagnostic accuracy for occlusal caries detection was performed by scoring all the digital images. The true presence of caries was determined by stereomicroscopy. Comparison of MGVs of the Al wedge for immediate and delayed scans was made using repeated measures analysis of variance (ANOVA) and Tukey-Kramer tests. The accuracy was expressed as the area under the receiver operating characteristics curves (A(z)). A(z)s were compared using two-way ANOVA and t-tests. Kappa was used to measure inter and intraobserver agreement.
Results: The mean A(z)s of immediately scanned plates were higher than the A(z)s of plates scanned with four different delays (p < 0.05). The A(z) of immediately scanned images was significantly higher than the A(z)s of images with 60 min and 120 min scan delays (p < 0.05) but not for 10 min and 30 min delays (p > 0.05). MGVs of immediately scanned plates were significantly higher than the plates scanned with four different delays (p < 0.05). Agreement decreased with increasing scan delays.
Conclusion: Scanning of DIGORA SPPs should not be delayed beyond 30 min for accurate occlusal caries diagnosis.