Methods \& Materials: This clinical trial was approved by the Scientific Review Committee and by the Committee for the Protection of Human Subjects of the local university (CAAE 50511415.1.0000.5546).
Trial design: This study will be a randomized-controlled, single-blind clinical trial with a parallel design. The patients included will be submitted to two in-office bleaching sessions with a 7-day interval receiving 35% hydrogen peroxide (control) or 37% carbamide peroxide in a single application of 45 minutes; while the same treatments will be allocated for each session. The study will be conducted at the clinic of the School of Dentistry of the local university from May 2016 to December 2016.
Participants: Patients included in this clinical trial will be at least 18 years old with good oral health. Patients with any of the six upper anterior teeth with caries, restoration, severe discoloration (e.g., stains caused by tetracycline), enamel hypoplasia, gingival recession, dentin exposure, pulpitis, or endodontics will be excluded. Participants submitted to previous bleaching procedures, presenting prior tooth sensitivity, pregnant or breastfeeding will be also excluded. Only patients presenting all six upper anterior teeth with shade mismatching with 2.5 M2 (Vita Bleach guide 3D-Master scale, Vita-Zahnfabrik, Bad Sackingen, Germany) or darker will be included.
Sample Size Calculation:
The sample calculation will be based on the primary continuous outcome for superiority control. Power of the test will be set at 80%, considering a type I error of 0.05; for a minimum detectable difference between treatments 25%, based on a prior study using similar bleaching agent, while a coefficient of variation of 30% of control treatment will be expected. The calculation resulted in fifty patients, through the formula used n = f(α/2, β) × \[p1 × (100 - p1) + p2 × (100 - p2)\] / (p2 - p1)2. The calculation resulted in fifty patients.
Randomization:
A randomized list will be computer-generated by a person not involved in intervention or evaluation. The participants were defined as blocks in the randomization process, where the sequence of treatment (treatment or control) will be randomly set for each block by using computer-generated tables (www.sealedenvelope.com). The sequence will be inserted into sealed envelopes numbered from 1 to 50 that were opened by the operator only at the moment of the intervention. The patients were numbered according to the sequence of enrollment. Neither the participant nor the operator knew the group allocation determining blinding to the protocol.
Baseline evaluation:
Prior to bleaching procedure, the teeth will be cleaned using rubber cups associated to pumice and water. The shades of canine upper teeth will be assessed on a baseline with a spectrophotometer Vita EasyShade (Vita-Zahnfabrik, Bad Säckinge, Germany) using the CIE L \* a \* b \*. 3 measurements on each tooth will be conducted, with the end result the average of 3 values (L \*, a \* and b \*).
Intervention:
Control treatment: A light-cured resin dam will be applied (Top Dam, FGM, Joinville, SC, Brazil) on the gingival tissue corresponding to the teeth to be bleached. A 35% hydrogen peroxide based bleaching agent (Whiteness HP Maxx, FGM, Joinville, SC, Brazil) will be mixed and applied on the vestibular surface of teeth, in the ratio of 1 drop every 3 to the thickener and peroxide, remaining for 45 minutes. After this time, the bleaching agent will be removed. A second session will be as carried out after 1 week following the same procedures.
Experimental Treatment: A 37% carbamide peroxide based bleaching agent (Power bleaching, BM4, Palhoça, SC, Brazil) will be applied on the vestibular surface of teeth with a relative isolation only using cotton roller, remaining for 45 minutes. After this time, the bleaching agent will be removed. A second session will be as carried out after 1 week following the same procedures.
Evaluations:
The tooth sensitivity reported by patients will be assessed using a visual analog scale (VAS) and verbal rating scale (VRS). The VAS consisted of a scale 10 cm in length ranging from 0 (absence of pain) to 10 (unsupportable pain). The patient set his or her level of sensitivity by pointing on the scale corresponding to this level, while the distance from this point to the 0 border will be recorded. For VRS, the patient reported his or her level of sensitivity based on scores: 0 = none, 1 = light, 2 = moderate, 3 = considerable and 4 = severe. The tooth sensitivity will be assessed during the bleaching, immediately after bleaching agent removal and and up to 24 hours after each session. One week after each session and 30 days after the last session the tooth color will be evaluated again using the CIE L \* a \* b \* as previously described.
Statistical Analysis:
For color analysis, the parameter delta data L \*, a \* and b \*, and ΔE and Δ00, will be individually subjected to two-way repeated measures ANOVA test to compare the two bleaching techniques at each evaluation time. Based on presence of any tooth sensitivity (scores different from 0 for VRS), the absolute risk, odds ratio and relative risk will be calculated regarding the treatments for each moment of evaluation/session of bleaching, as well as its confidence intervals (95%). For each moment, differences on presence/absence ratios will be analyzed by Fisher's exact test. For the overall risk related to each treatment, the odds ratio will be adjusted to independent variable "session of bleaching" using the Mantel-Haenszel statistic. The homogeneity of odds ratios will be analyzed by the Breslow-Day and Tarone's tests. Following, the odds ratio estimated will be converted to relative risk and the overall presence/absence ratios will be analyzed by the McNemar's test, considering the study design (parallel).
For VRS, the data from scores observed in each time of evaluation/session of bleaching will be submitted to the Mann-Whitney rank sum test. Despite the measurement of tooth sensitivity using VAS giving a continuous outcome, data assessed with this scale did not show a normal distribution (Shapiro-Wilk's test). Thus, data from VAS will be also analyzed by the Mann-Whitney rank sum test, while one test per time of evaluation will be carried out. All statistical analyses will be performed considering a 95% significance level (α = 0.05).
