Eleven studies, with 462 participants, were included in the review. Nine were RCTs and two were controlled trials. Five RCTs were at a low risk of bias, and four were judged to be at a moderate risk; the two controlled trials were judged to be at a high risk of bias. Eight of the RCTs reported blinding the assessors to allocation.
Immediate effect: Seven trials used visual measures of the immediate colour change. There was significant heterogeneity between the trials, which were analysed by hydrogen peroxide concentration, and the two non-randomised trials were removed. There was no evidence of a difference between light-activated and non-light activated systems with a high concentration of hydrogen peroxide (25% to 35%; three trials). There was a significant difference in favour of light-activated systems with low concentrations of hydrogen peroxide (15% to 20%), with a mean difference of -1.78 (95% CI -2.30 to -1.26; two trials).
Short-term effect: Four trials used visual scales and three trials used instrumental measures. There were mixed results and substantial heterogeneity. Subgroup analyses found no significant difference in three high-concentration trials, but one low-concentration trial reported a significant effect.
Medium-term effect: Two trials reported visual measures; one also reported an instrumental measure. One trial used low-concentration hydrogen peroxide and reported a significant difference with light activation, while the other used a high concentration and found no differences in outcomes.
Tooth sensitivity: Four trials reported on the presence or absence of tooth sensitivity. Meta-analysis showed a significantly higher risk of sensitivity with light activation (OR 3.53, 95% CI 1.37 to 9.10). Three trials reported the degree of tooth sensitivity, and meta-analysis indicated that non-light activated systems were associated with less sensitivity (MD 0.57, 95% CI 0.21 to 0.92). No significant heterogeneity was noted for either analysis.