Purpose: To compare the level of agreement between carious lesion assessments according to the visual clinical examination and the colour photograph methods. Materials and Methods: Data on the presence of enamel/dentin carious lesions in previously sealed occlusal surfaces in first molars were obtained by two trained and calibrated examiners through visual clinical examination and from colour photographs 4 years after sealing. Kappa statistics were applied to calculate agreement between assessment methods. Data analysis was performed using sign, Bowker symmetry and McNemar's tests. Results: The prevalence of dentin carious lesions was very low. The kappa coefficients for detecting enamel/dentin carious lesions using the two assessment methods were 0.65 (CI: 0.56-0.74) for examiner 1 and 0.70 (CI: 0.62-0.78) for examiner 2. Examiner 2 observed more enamel/dentin carious lesions on colour photographs than did examiner 1 (p = 0.008). Sensitivity analyses did not confirm this outcome. Conclusions: There was no difference in the detection of enamel/dentin carious lesions in previously sealed occlusal surfaces using colour photographs vs visual clinical examination. The colour photograph method is therefore equivalent to the visual clinical examination in detecting enamel/dentin carious lesions. More studies are required. Keywords: carious lesion detection, dental caries, intraoral photography, oral examination, sealants