A longitudinal analysis from bite-wing radiographs of the rate of progression of approximal carious lesions through human dental enamel.

Four to ten years of serial bite-wing radiographs from over 700 children from five groups, three in Sweden and two in the U.S., were interpreted. By analysing changes in the depth of unfilled lesions over time, the mean time and probability distribution for the time a lesion remains in both the outer half and inner half of the enamel were estimated. The procedure incorporated information on filled lesions and non-progressing lesions and thus minimized bias that results in overestimation of the progression rate. In primary teeth, in both the U.S. and Swedish groups, it took on average 12 months for a lesion to progress through the outer half of the enamel and on average 10-12 months for a lesion to progress through the inner half. In newly-erupted first permanent molars, it took 21-23 months for a lesion to progress through the outer half of the enamel and between 19 (U.S. data) and 28 months (Swedish data) for progression through the inner half. In older adolescents in the two Swedish groups, progression was slower: 38-41 months through the outer-half and 47-56 months through the inner-half. In older U.S. adolescents, progression appeared to be more rapid: 16 months through the outer half of the enamel and 27 months through the inner half. The duration of time a lesion remains in different halves of the enamel could be approximated by a piecewise exponential or exponential probability distribution, which exhibits extreme variability. Assuming duration in each half of the enamel follows an exponential distribution with a mean of 2 yr, about 10 per cent of new lesions will progress through the enamel in one year and 25 per cent in two years. However, over 40 per cent of the lesions will not have progressed in 4 yr. There were no consistent differences in the rate of progression by sex, between upper and lower dentitions, for premolars versus molars, or between high and low-risk individuals.