Page 11 - SPEMD_60-3

rev port estomatol med dent cir maxilofac . 2019;60(3):96-103 99

®
in the x-axis direction, the free residual space between the Statistical analysis was performed with IBM SPSS 23.0
slices was filled with the light-body silicone Aquasil Ultra XLV software (SPSS; Chicago, IL, USA). The normality of data was
(Dentsply, DeTrey, Konstanz, Germany). For each tooth, the assessed with the Kolmogorov-Smirnov test and visual inspec-
tops of adjacent sticks were identified with two colors. Half of tion of the histograms. Homoscedasticity was assessed with
the peripheral and central sticks were used to measure mi- Levene’s test. Since these two assumptions were not verified,
crotensile bond strength (μTBS) after 7-day storage and the the Kruskal-Wallis test was run to determine the effect of the
others to determine μTBS after 4-year storage in water at 37ºC adhesion strategy and aging on bond strength, followed by
under the same protocol. During these 4 years, the medium all-pairwise comparisons with Bonferroni corrections. The chi-
was replaced every seven days to avoid contamination, follow- square test was used to compare the distribution of failure
22
ing the ISO/TS 11405:2015 recommendations. All sticks were modes between groups. For all analyses, the significance level
checked on an optical microscope (M300, Leica, Switzerland) was set at α=0.05.
at 40x magnification to exclude faulty specimens.
Each stick was attached to a microtensile sample holder
with cyanoacrylate adhesive (CE10Flex , Ce Chem Limited, Der- Results
®
byshire, UK) and then fixed on the microtensile device (Od04-
Plus; Odeme Dental Research, Luzerna, Brazil). Specimens were The overall mean μTBS, standard deviations, number of spec-
fractured in tensile mode using a universal testing machine imens (n) and multiple comparison statistical analyses of all
(Model AG-I, Shimadzu Corporation, Kyoto, Japan) at a cross- experimental groups are detailed in Table 2 and represented
head speed of 0.5 mm/min. The maximum load was recorded in Figure 2.
in Newtons, and microtensile bond strength was calculated in Data assessed with the Kolmogorov-Smirnov test failed to
MPa according to the following equation: μTBS = F/A, where F prove the normality of bond-strength values. The Kruskal-Wal-
is the load at fracture (N) and A is the bonded area (mm ). lis test determined that the distribution of bond-strength values
2
Figure 1 shows a schematic diagram of tooth preparation, res- across groups was not similar (p<0.001). All pair-wise compari-
toration, specimen sectioning and bond-strength testing. sons showed no statistically significant differences in μTBS
The failure mode was analyzed under an optical micro- between adhesive strategies in both periods (p=0.504 for imme-
scope (Leica CLS 150 MR, Switzerland) at 40x magnification. diate and p=0.408 for 4-year aging). Both adhesive strategies
The fracture pattern was classified as follows: adhesive, if the showed a significant decrease in bond strength between the
failure occurred entirely within the adhesive interface; cohe- immediate and the 4-year evaluation (p<0.001). Both evaluation
sive, if it occurred completely in the composite resin (cohesive periods showed superior microtensile bond strength when the
in the resin) or the dentin (cohesive in the dentin); and mixed, universal adhesive was coated with an extra hydrophobic resin
when both adhesive and cohesive failure occurred. layer, although the difference was not statistically significant.

Figure 1. Schematic diagram of specimen preparation for microtensile bond-strength testing.