This in vitro study assessed the effects of in-office bleaching gels with different pH values (neutral vs. acidic) and bracket conditions (without brackets or brackets) bonded using self-etch or etch-and-rinse adhesive strategies on hydrogen peroxide (HP) penetration into the pulp chamber and colour change.

Seventy premolars were randomly allocated into seven groups ( n = 10) according to bleaching gel (Whiteness HP Automixx 35% [acidic pH]; Whiteness HP Automixx Plus 35% [neutral/alkaline pH]), bracket conditions and adhesive strategies. Ultrapurified water was used as the negative control. Gel pH was measured with a digital pH meter. Colour was assessed spectrophotometrically at baseline and one week after bleaching (ΔE _ab , ΔE ₀₀ , ΔWI _D ). HP penetration into the pulp chamber was quantified using UV–Vis spectrophotometry. Data were analysed using repeated-measures, one-way, and two-way ANOVA followed by Tukey's and Dunnett's tests (α = 0.05).

Bleaching gels showed different pH values at all evaluated time points ( P = 0.001). All experimental groups showed higher HP penetration than the control group (95% CI: –0.001 to 0.001 μg/mL; P < 0.000001), with greater values observed for the acidic gel (95% CI: 0.029–0.057 μg/mL; P = 0.0001). Adhesive strategy also influenced HP penetration, with higher values observed for the etch-and-rinse approach when the acidic gel was used (95% CI: –0.020 to 0.108 μg/mL; P = 0.0001). Colour change did not differ significantly between bleaching gels ( P > 0.24). However, teeth without brackets showed greater colour change than those with brackets bonded using the etch-and-rinse strategy ( P < 0.001), with ΔE _ab ranging from 5.33–9.70 versus 3.25–6.80, ΔE ₀₀ from 3.61–6.63 versus 2.37–4.74, and ΔWI _D from 4.50–8.67 versus 2.97–5.70 (95% CI), respectively.

Neutral/alkaline stable-pH bleaching gels combined with self-etch bonding strategies were associated with reduced HP penetration and favourable bleaching outcomes, despite slight attenuation of colour change in the presence of brackets. These findings require further clinical validation.