IADR Abstract Archives
Examiner Reliability in Assessing Caries Activity Using ICDAS and QLF-D
Objectives: Caries activity assessment is critical for proper diagnosis and treatment planning. This study aims to investigate the reliability of examiners assessing caries activity using the International Caries Detection and Assessment System (ICDAS) visual criteria and quantitative light-induced fluorescence digital (QLF-D; Inspektor Research Systems, The Netherlands) technology.
Methods: Extracted permanent human teeth with sound surfaces and non-cavitated carious lesions (ICDAS 1-3) on smooth (n=60) and root (n=60) surfaces were selected. Three calibrated examiners evaluated the teeth using ICDAS as sound, active, or inactive, repeating the exam one week later. The same examiners used the Plaque Patch function of QA2 image analysis software (Inspektor Research Systems) on QLF-D images of the teeth. QA2 White Spot Analysis was used to determine lesion presence.
Results: Inter-examiner reliability for ICDAS for caries lesion detection was Fleiss kappa 0.83 for smooth and 0.84 for root surfaces, while intra-examiner results ranged from 0.69-1.00 for smooth surfaces and 0.97 for root surfaces. When assessing activity using ICDAS, inter-examiner agreement was 0.64 for smooth and 0.59 for root surfaces, while intra-examiner ranged from 0.70-0.78 for smooth and 0.69-0.75 for root surfaces. Inter-examiner reliability with QLF-D ΔRmax values using ICC was 0.80 for smooth and 0.60 for root surfaces. Compared against each other, the agreement between White Spot 95Q Analysis and ICDAS demonstrated a Rank Biserial Correlation of 0.66 for smooth and 0.81 for root surfaces.
Conclusions: ICDAS criteria demonstrated “almost perfect” agreement when simply identifying lesion presence. However, intra- and inter-examiner reliability decreased when using ICDAS for activity assessment. QLF-D ΔRmax presents an off-purpose possibility to increase reliability of activity assessment. White Spot Analysis demonstrated near perfect agreement in confirming the presence of root lesions. Studies with clinical validations are warranted to support these uses, and these results emphasize the need for a reliable means of activity assessment.
Methods: Extracted permanent human teeth with sound surfaces and non-cavitated carious lesions (ICDAS 1-3) on smooth (n=60) and root (n=60) surfaces were selected. Three calibrated examiners evaluated the teeth using ICDAS as sound, active, or inactive, repeating the exam one week later. The same examiners used the Plaque Patch function of QA2 image analysis software (Inspektor Research Systems) on QLF-D images of the teeth. QA2 White Spot Analysis was used to determine lesion presence.
Results: Inter-examiner reliability for ICDAS for caries lesion detection was Fleiss kappa 0.83 for smooth and 0.84 for root surfaces, while intra-examiner results ranged from 0.69-1.00 for smooth surfaces and 0.97 for root surfaces. When assessing activity using ICDAS, inter-examiner agreement was 0.64 for smooth and 0.59 for root surfaces, while intra-examiner ranged from 0.70-0.78 for smooth and 0.69-0.75 for root surfaces. Inter-examiner reliability with QLF-D ΔRmax values using ICC was 0.80 for smooth and 0.60 for root surfaces. Compared against each other, the agreement between White Spot 95Q Analysis and ICDAS demonstrated a Rank Biserial Correlation of 0.66 for smooth and 0.81 for root surfaces.
Conclusions: ICDAS criteria demonstrated “almost perfect” agreement when simply identifying lesion presence. However, intra- and inter-examiner reliability decreased when using ICDAS for activity assessment. QLF-D ΔRmax presents an off-purpose possibility to increase reliability of activity assessment. White Spot Analysis demonstrated near perfect agreement in confirming the presence of root lesions. Studies with clinical validations are warranted to support these uses, and these results emphasize the need for a reliable means of activity assessment.