Recommendations for treatment are based on international guidelines [ICSI, 2006; Strauss et al, 2007] and narrative reviews [Webster, 2002; James, 2005; Ravenscroft, 2005; Zaenglein and Thiboutot, 2006; Acne Working Group, 2008], and reflect published data from randomized controlled trials (RCTs) where available, and otherwise, expert opinion.

Topical benzoyl peroxide

• Benzoyl peroxide is a potent bactericide and significantly reduces the population of Propionibacterium acnes in the sebaceous follicle [Thiboutot, 2000]. There is good evidence from placebo-controlled trials that show benzoyl peroxide reduces both inflammatory and non-inflammatory lesions [Purdy and DeBerker, 2008].

Topical retinoids

• Topical retinoids normalize follicular keratinization, promote drainage of comedones, and inhibit new comedone formation. Although they have been used historically mainly to treat comedones, they are also effective at treating inflammatory lesions (if used in the longer term) by inhibiting microcomedone formation [Thiboutot, 2000; Wolf, 2002]. This is supported by good evidence from placebo-controlled trials [Purdy and DeBerker, 2008].

Topical antibiotics

• Topical antibiotics reduce colonization of sebaceous follicles by P. acnes and may also have an anti-inflammatory effect. There is evidence for effectiveness from placebo-controlled trials [Purdy and DeBerker, 2008], and they are probably most useful against papules and pustules which are not impractically widespread [ICSI, 2006].

Azelaic acid

• Azelaic acid is a second-line option that should be considered if other treatments are unsuitable [Acne Working Group, 2008]. There is limited evidence from two small placebo-controlled trials that azelaic acid is effective in the treatment of acne, but clinical experience with its use has been reported to be disappointing [Brown and Shalita, 1998; James, 2005]. However, azelaic acid may cause less adverse effects than benzol peroxide or topical retinoids [Ravenscroft, 2005].

Combining topical treatments

• Combining topical treatments is recommended by experts for most people with moderate acne [Acne Working Group, 2008].
• Benzoyl peroxide combined with a topical antibiotic is usually the preferred choice, as it is an effective and well-tolerated regimen. The efficacy of this combination has been shown by evidence from three large RCTs, and the addition of benzoyl peroxide to a topical antibiotic has been shown to prevent the development of bacterial resistance [Strauss et al, 2007].
• A topical retinoid combined with a topical antibiotic is a useful option in people who cannot tolerate benzoyl peroxide. In theory this combination should demonstrate synergy, with both inflammatory and non-inflammatory lesions being directly treated, although limited evidence from an RCT suggests this combination is not as effective as benzoyl peroxide combined with a topical antibiotic, and there are concerns that bacterial resistance could develop.
• Benzoyl peroxide combined with a topical retinoid has been reported as being a 'very effective' treatment [Thiboutot, 2000], although CKS identified no good-quality RCTs to support this. This combination may cause an unacceptable rate of adverse effects.
• CKS identified no information on when it may be appropriate to combine azelaic acid with another topical treatment.

Oral antibiotics

• Oral antibiotics are universally recommended by experts for the treatment of severe acne, or extensive acne that would be difficult to treat with a topical drug [Dreno et al, 2004; ICSI, 2006; Strauss et al, 2007].
• Oral tetracyclines are recommended first-line. There is good evidence from placebo-controlled trials that tetracycline is effective at reducing lesion counts and severity. Although there is a lack of placebo-controlled trials to verify the efficacy of the other standard tetracyclines, there is evidence from comparative trials that there is likely to be a class effect.
• Oral erythromycin should be reserved for use when tetracyclines are contraindicated.
• There is a lack of evidence from placebo-controlled trials to verify the efficacy of erythromycin, although evidence from comparative trials indicate it is probably as effective as tetracyclines.
• However, there is evidence from observational and controlled studies that there are particular problems with the development of bacterial resistance to erythromycin.
• Minocycline is increasingly not recommended for the treatment of acne vulgaris as other tetracyclines are regarded as being as effective, and less expensive with better safety profiles [DTB, 2006; DTB, 2009].

Combined oral contraceptives (COCs)

• COCs are recommended as a first-line adjunctive treatment for women who have acne [ICSI, 2006; Strauss et al, 2007]. There is good evidence from placebo-controlled trials that COCs are effective in reducing lesion count, acne severity, and the woman's perception of the condition [Arowojolu et al, 2007].