A literature review disclosed that even under very strict caries preventive control, in young Swedish people many initial approximal lesions slowly progress. A study was performed to investigate if the same trend could be seen in young Danish adults. In 1996 a group of Danish military conscripts were investigated clinically and radiographically. Based on the radiographs, one lesion with radiolucency up to one third in the dentine was randomly selected from each conscript. Contact was established with these former conscripts in 2002 and radiographs taken around 2002 were used to investigate the state of the selected lesions in terms of regression, stabilization or progression. It was observed that about 60% of the selected lesions had progressed during the 5 to 6 year period. Therefore there was a need to look for other preventive measures than those traditionally used in a Danish context, such as instructions in flossing and local application of fluoride. Looking through the literature, it was realized that sealing of occlusal lesions was highly effective but this procedure was not used on approximal surfaces. The next step was to investigate if it was technically possible to seal approximal lesions and if so what material should be used. A laboratory study was conducted in which 6 dental materials were compared in relation to: their handling, presence at the surface of the lesion, and ability to penetrate into a lesion. Based on that study Gluma One Bond® (Heraeus Kulzer) was selected to be used in further studies. The final step was an 18-month clinical investigation to evaluate the effect of sealing approximal lesions. About 80 young adults from Denmark and Colombia participated in this investigation; each had at least 2 initial approximal lesions. One lesion was selected for sealing; the other one acted as control. Baseline radiographs were taken by means of special equipment so new radiographs could be taken in the same position after 18 months. The sealing procedure was performed in accordance with the guideline designed in the laboratory study. The patient was instructed to floss all sealed and non-sealed approximal lesions. Radiographs were used to assess caries progression in the test and control lesions in three ways: by comparing baseline with final scoring of the lesions on the individual conventional radiographs, by reading the conventional radiographs as pairs and finally using subtraction radiography of digitized images. The examiner was blind to lesion type (test or control) for all comparisons and also blind to whether radiographs were baseline or follow up in the case of examinations of the single radiographs and paired comparison. Repeat assessment of the radiographs was conducted on 20% of the radiographs and the subtraction images. For the paired comparison of radiographs and subtraction radiography assessment significantly more control lesions progressed (47.2%; 85%) than test lesions (22.2%; 43%) (P-values<0.05). For the independent examination of the radiographs a similar pattern was seen but the difference between the two groups was not statistically significant (P=0.06). In conclusion, many approximal lesions seemed to progress on young adults irrespective of the preventive regime they were advised to take. This study has developed and investigated a technique for sealing early approximal lesions which seems to be more effective at preventing lesion progression than conventional interventions such as brushing and flossing. There is a need however, for defining more precisely criteria at selecting lesions to be sealed; for developing better materials in terms of better penetration skill and that they stay there, and for clinical investigations which evaluate the long term effect of sealing the approximal lesions.