Twenty-five trials (the authors reported the number of patients to be 694, but this could not be calculated from the tables) were included. Ttrials were of variable quality and included a small number of patients (sample size ranged from eight to 46). Allocation concealment and method of randomisation was unclear in most studies.
Light therapy compared with placebo (10 RCTs, n=322): Moderate to large improvements were reported by trials that used blue lights (405nm to 420nm; two trials) and blue-red lights (415nm plus 600nm laser; one trial). Green (532nm pulsed laser; two trials) and yellow lights (585nm pulsed laser; two trials) showed either small to moderate or no improvements. A 1,320nm neodymium:yttrium-aluminium-garnet laser showed a small transient improvement in comedones, but not in inflammatory acne. Infrared radiation therapy (1,450nm diode laser) reduced the mean lesion count compared to control, with reductions in inflammatory lesions after a third treatment of 75.1% for the lower intensity side and 70.6% for the higher intensity side, maintained at 12 months. There were no significant differences between different light intensities.
Light therapy versus conventional topical treatments (three trials, n=119): Blue-red light was found to be effective compared to 5% benzoyl peroxide (p=0.006, one trial). No statistically significant improvements were found in trials that compared blue light to 1% clindamycin (one trial) or intense pulsed light with a 530nm to 575nm filter to benzoyl peroxide (one trial).
In the short term, blue-red light was more effective compared to blue light alone, but was no longer effective at 12 weeks (one trial). Both light therapy alone and light therapy with microdermabrasion significantly reduced inflammatory lesions over 12 weeks, but there were significant differences between these treatments (one trial).
Photodynamic therapy showed a statistically significant reduction in mean percentage inflammatory lesion count at four weeks (WMD 13.7%, 95% CI 11.4% to 15.9%; three trials) and 12 weeks (WMD 20.8%, 95% CI 15.6% to 26%) post treatment. Trials that compared photodynamic therapy using different creams (one trial) or that compared photodynamic therapy using different light sources for activation (one trial) showed no statistically significant differences. Two trials (n=44) that compared photodynamic therapy with placebo/untreated patients showed a significant reduction of inflammatory lesions in the photodynamic therapy group.
Side effects were reported in the review. Some patients discontinued photodynamic therapy due to unacceptable side effects.