M Beerens

DGGE FOR CARIES RISK ASSESSMENT IN ORTHODONTICS 3 49 separatedbasedonDGGE pattern. Nomanual editingwas applied. Some bands were not marked and detected using automated band detection in Gelcompar II. This shows the threshold of the program setting. No manual editing was applied. The number of bands detected in the DGGE pattern of subjects with or without WSL manually and at optimum ‘absolute’ (4%; minimum area 0.1%) and ‘relative’ (15%, minimum area 1%) minimum profiling settings for automated band detection are given in Table 1. For manual and automated band detection at the minimum profiling setting of 4% absolute to the most intense band, the number of bands in the WSL group was not significantly different from that in the group without WSL. For automated band detection at the minimum profiling setting of 15% relative to the most intense band the number of bands detected in the WSL group was significantly smaller than for the group without WSL (see Table 1). TABLE 1 . Comparison between numbers of bands detected using different profiling settings for subjects with or without WSL Median number of bands detected (quartiles) Group N Manual Automated band detection at 15% relative minimum profiling and 1% minimum area Automated band detection at 4% absolute minimum profiling and 0.1% minimum area WSL 28 20 - (16 - 25) 13 - ( 9 - 17) 20 - (14 - 23) No WSL 9 25 - (22 - 27) 18 - (16 - 22) 20 - (16 - 23) p-value (Mann Whitney U) 0.111 0.007 0.845 Results microbiological data Of the 37 included patients, collected microbiology data showed that of 3 samples not enough plaque was collected for cultivation. Data of one sample in the WSL group and 2 samples in the non-WSL group showed this data loss. The microbiology data, presented in Table 2, showed no significant difference between the two groups was observed in total CFU counts (p= 0.79), percentages of aciduric flora (p= 0.21), S. mutans (p= 0.19) and Lactobacillus spp. (0.49) or C. albicans (p= 0.36), although a trend towards a higher portion of the aciduric flora and S. mutans was observed in the WSL group. No correlation was found with the number of DGGE bands of a sample and the microbiological data, except for total CFU counts, for which an inverse relation was found (Spearman rho= - 0.42, p= 0.012).