Subject-Level Meta-analysis of 10% Hydrogen Peroxide Whitening Strip Clinical Response

Method: The analysis was inclusive of subject-level responses from 21 clinical trials at 8 different sites over 7 years.  Each subject had been randomly assigned whitening strips with a very thin 10% hydrogen peroxide gel for 7-days use (30 min BID) unsupervised at-home.  For efficacy, standard digital images were collected on the maxillary anterior facial surfaces, and image analysis was used to determine change in yellowness (Δb*) and lightness (ΔL*).  For safety, all subjects received comprehensive clinical examinations, plus interviews to ascertain the presence/absence of tooth sensitivity or oral irritation during treatment. 

Result: There were 673 subjects in the analysis ranging from 18-79 years of age.  Gender and age were significant (p<0.001) effects in the baseline tooth color model, and the age-color regression slope (SE) was 0.0457 (0.005) b* units per year (p<0.001).  Overall, 100% of subjects exhibited a measured improvement in tooth color (–Δb* or +ΔL*), while 94% had 2-parameter improvement (–Δb* and +ΔL*).  Transient tooth sensitivity (16.3%) and oral irritation (15.8%) were the most common adverse events, but overall, these events had no adverse impact on whitening response.  For tooth sensitivity, adjusted means (SE) for Δb* & ΔL* were –1.9 (0.12) & 1.8 (0.11) with sensitivity, compared to –1.9 (0.10) & 1.8 (0.08) without sensitivity.  For oral irritation, adjusted means (SE) for Δb* & ΔL* were –2.0 (0.12) & 1.9 (0.11) with irritation, compared to –1.8 (0.10) & 1.7 (0.08) without irritation, the latter of which differed significantly (p<0.01) favoring subjects with occurrence of irritation. 

Conclusion: This inclusive meta-analysis demonstrates that 10% hydrogen peroxide whitening strips are safe and effective, that starting color and age significantly affect whitening response, and that minor sensitivity or irritation occurrence do not limit effectiveness.