Clinical dermatology Review article

CED
Clinical and Experimental Dermatology

CPD

Whats new in acne? An analysis of systematic reviews published in 20092010

E. V. Smith, D. J. C. Grindlay* and H. C. Williams*
University Hospital of Wales, Cardiff, UK; and *NHS Evidence-skin disorders, Centre of Evidence Based Dermatology, University of Nottingham, Nottingham, UK
doi:10.1111/j.1365-2230.2010.03921.x

Summary

This review highlights clinically important ndings about acne treatment identied in nine systematic reviews published or indexed in the period March 2009 to February 2010. A systematic review of dietary inuences on acne suggested that a possible role of dietary factors in acne cannot be dismissed, as the studies to date have not been sufciently large or robust. Another review looked at benzoyl peroxide, which may be enjoying a comeback because of increasing bacterial resistance to antibiotics, and suggested that there was a lack of evidence that stronger preparations were more effective than weaker ones. The same team also carried out a systematic review addressing the question of whether topical retinoids cause an initial worsening of acne. They found no evidence to suggest initial worsening of acne severity, although there was evidence of skin irritation that typically settled by 812 weeks. A review of oral isotretinoin and psychiatric side-effects reinforced a possible link between the two, although it pointed out that the better-quality primary studies were still inconclusive. An updated Cochrane Review conrmed the efcacy of combined oral contraceptives (COCs) in reducing acne lesion counts. It also found that the evidence to support COCs containing cyproterone acetate over others was very limited. Another Cochrane Review failed to show any benet of spironolactone for acne, based on limited studies. Three reviews examined laser and light therapies, and found some evidence of superiority only for blue or blue red light treatment over placebo light, but a general absence of comparisons against other acne treatments. Photodynamic therapy had consistent benets over placebo but was associated with signicant side-effects and was not shown to be better than topical adapalene.

Background
Correspondence: Professor Hywel Williams, Centre of Evidence Based Dermatology, University of Nottingham, Queens Medical Centre, Nottingham NG7 2UH, UK E-mail: hywel.williams@nottingham.ac.uk Conict of interest: EVS, DJCG and HW work in the UK National Health Service (NHS). NHS Evidence skin disorders is funded by the NHS. None of the authors has any nancial connections with any pharmaceutical company. A similar and more detailed review to the material published here appeared in the 2010 Annual Evidence Update on Acne published by NHS Evidence skin disorders in March 2010 (http://www.library.nhs.uk/skin/ViewResource. aspx?resID=343542&tabID=289&catID=8275) and explicit reference is given to that fuller version throughout. There are no copyright issues with using material from that source. Accepted for publication 24 May 2010

This review summarizes nine systematic reviews dealing with treatment and prevention of acne, which were indexed in bibliographic databases between March 2009 and February 2010 and were included in the 2010 Annual Evidence Update on Acne Vulgaris from NHS Evidence skin disorders. This review aims to pick out clinically important points with the busy clinician in mind. Readers are encouraged to view the full report and original papers cited in the 2010 Annual Evidence Update (http://www.library.nhs.uk/skin/ViewResource. aspx?resID=343542&tabID=289&catID=8275), where the methods and omitted citations are given. This review considers systematic reviews only, as they are generally

The Author(s) CED 2010 British Association of Dermatologists Clinical and Experimental Dermatology, 36, 119123

119

Whats new in acne? E. V. Smith et al.

considered to be the most reliable evidence source, whereas the results of single randomized controlled trials (RCTs) are often contradicted by subsequent trials.1 A paper summarizing the results of the 2008 and 2009 Annual Evidence Updates on Acne Vulgaris has previously been published in this journal.2

Associations
Diet

lesion counts in the rst fortnight.10 They did not specify study types, numbers or quality assessments, or explain their selective citation of studies. Eight studies found no evidence of worsening and one suggested slight early deterioration. Irritation was common, but normally settled by weeks 8 to 12. This review was conducted by a team whose research centre is supported by Galderma, the manufacturers of adapalene, and the topic seemed slightly contrived. However, it conrms that skin irritation is common with topical retinoids and that it takes 23 months before tolerance occurs.
Oral isotretinoin and depression

Spencer et al.3 undertook a systematic review of dietary inuences in acne that included 21 observational studies and 6 RCTs. These suggested that dairy products (especially milk) are associated with increased risk and greater severity of acne, and that a low glycaemic load diet might improve the condition. The question of whether chocolate worsens acne remained unanswered. The reviewed lacked a thorough assessment of study quality, and most included studies were observational in design with selfreported outcomes, which may be a signicant bias in this type of study. The glycaemic diet trial4 was published in duplicate5 and also as a third paper reporting a subgroup.6 The authors of the systematic review included two of these as separate trials, highlighting the problem of disproportionate effects of duplicate publications.7

Treatments
Benzoyl peroxide

Fakhouri et al.8 revisited benzoyl peroxide as a potential solution to the problem of antibiotic resistance in acne, looking at usage methods to increase efcacy and minimize irritancy. A PubMed search returned 900 reports. The authors concluded that efcacy was similar for 2.5%, 5% and 10% preparations of benzoyl peroxide, and that efcacy may be enhanced by vitamin E and tertiary amines, and by combining with retinoids. New delivery systems increase tolerability without compromising efcacy. The review was not performed to a high standard. The inclusion criteria for studies were unclear, and those studies included were not assessed for quality. No attempt was made to combine the studies (i.e. metaanalysis). The overall conclusion on equivalence was based on just one study that was probably underpowered to determine equivalence.9
Topical retinoids

A recent systematic review by a team of psychiatrists addressed the key question of whether oral isotretinoin is linked to depression.11 MEDLINE and EMBASE were searched, but no other methodology was specied. The authors found 24 case reports and series that apparently suggested a link, but such reports are very prone to publication bias.12 Some reports described clear symptom cessation when stopping isotretinoin, and recurrence on restarting. Two large database studies found no association, and two found slightly increased antidepressant use. A case crossover study of 30 000 people with acne found that those developing depression were 2.68 times more likely to have taken isotretinoin in the preceding 5 months.13 Only two small trials were controlled, with neither reporting increased psychiatric side-effects. Study selection and quality assessment within this review were not clear. Severe acne is itself associated with depression. The authors state the evidence strongly supports a link as a great number of reports support this. However, the better-quality studies included in their review were inconclusive, and publication bias is a concern. The review has added little to the debate, although it does include some interesting discussion about plausible mechanisms by which retinoids affect the central nervous system.
Oral contraceptives and antiandrogens

A review by the same team investigated whether initial use of topical retinoids paradoxically increases acne

An updated Cochrane Review examined 25 trials of combined oral contraceptive pills (COCs) in acne.14 Six compared COCs to placebo, and conrmed their superiority in reducing lesion counts. COCs containing cyproterone acetate (CPA) are traditionally used for acne, but evidence of superiority over other COCs was limited and inconclusive. Of 13 direct comparisons of different COCs, methodological diversity and conicting results prevented conclusions. One small study compared CPA with minocycline 50 mg, which produced

120

The Author(s) CED 2010 British Association of Dermatologists Clinical and Experimental Dermatology, 36, 119123

Whats new in acne? E. V. Smith et al.

similar self-assessed improvements in acne. The analysis was generally hampered by high dropout rates, weak design and poor reporting. A second updated Cochrane Review considered spironolactone for hirsutism and or acne,15 but only three studies were relevant for acne. Sebum excretion rate was not reduced in a study that compared spironolactone 3% and 5% cream against topical canrenoate (a metabolite) in 31 patients.16 A double-blinded cross-over RCT17 compared oral spironolactone 200 mg with placebo in 29 women. The authors of that report claimed signicant reductions in mean inamed lesion counts, but they did not perform an intention-to-treat analysis. The spironolactone group was more severe at baseline, and the imbalance was not adjusted in the analysis, which meant that those in the active group might have simply improved with time (regression to the mean). Another excluded study compared four doses of spironolactone against placebo.18 Those authors claimed that doses of 100 mg resulted in improvement, yet they presented no statistics. The small numbers (n = 36) and multiple groups make statistical signicance very unlikely.
Laser and light therapies and photodynamic treatment

Three new reviews have examined laser and light therapies for acne. The rst, by Hamilton et al.,19 a team supported by the Cochrane Collaboration, was well reported. This review searched eight databases and comprised 694 patients from 25 RCTs. Trials varied widely in design and quality, and meta-analysis was impossible. Ten RCTs evaluated light vs. placebo, and found that green, yellow and infrared spectrums either showed no difference or slight improvement. A red light trial claimed signicant improvement but was unblinded. Some evidence for superiority of blue or bluered light over placebo was found in three studies, with reductions in inammatory lesions of 4975% vs. 10 25% in the untreated patients, with minimal sideeffects. Three studies compared light therapies against other active topical comparators. Only one study found a signicant benet, with blue-red light reducing lesion counts to a greater degree than 5% benzoyl peroxide at week 8 (75% vs. 60%, P = 0.02). Studies comparing blue light with topical clindamycin, and intense pulsed laser to intense pulsed light plus benzoyl peroxide, found no signicant differences in outcomes. The review also included 12 small trials of light plus light-activated cream (photodynamic therapy; PDT), which showed more consistency, most suggesting benet over light alone. However, the one active comparator trial reported PDT to be less effective in reducing

inammatory lesions compared with 1% adapalene gel at 12 weeks. Many participants on PDT experienced side-effects such as pain and peeling that were sufciently severe to discontinue treatment. The two other reviews specically concerned PDT and acne. Riddle et al.20 added little, undertaking uncritical analysis of 8 trials and 13 case series from one database. All reported reduction in inammatory lesions of 25 88% and or signicant improvement in acne, with consistent superiority of PDT over light alone. Pain, oedema and erythema featured in all studies, and in some participants, long-term photosensitivity was described. An unpublished multicentre RCT failed to show a difference between blue light with aminolaevulinic acid (ALA) or vehicle. The other review on PDT attempted to answer practical questions on PDT use.21 A wider search found 5 randomized trials (4 were RCTs) and 16 other reports. Considering these, the authors favoured topical photosensitizers, shorter contact times, methyl aminolaevulinate over ALA, and lower light uences, because of more tolerable side-effects. They recommended treating inammatory and moderately severe acne and skin types IIII, using 24-week intervals to minimize side-effects. Although this was an interesting commentary, there was limited hard evidence to substantiate the guidance.

Clinical and research implications

The key learning points are summarized in the box below. The general quality of systematic reviews dealing with acne was poorer than those we have found on eczema and skin cancer, often limited by unclear study selection criteria and lack of critical appraisal of the quality of included studies. At best, the reviews have highlighted the need for further research, especially into comparing novel therapies such as light sources and laser against commonly used active comparators for a sustained period.

Learning points
It is possible that a low glycaemic diet may help acne and that chocolate worsens acne; goodquality prospective studies are needed to resolve such uncertainties. Wider use of benzoyl peroxide is one means of possible reduction of bacterial resistance due to prolonged use of antibiotics.

The Author(s) CED 2010 British Association of Dermatologists Clinical and Experimental Dermatology, 36, 119123

121

Whats new in acne? E. V. Smith et al.

Although use of topical retinoids will often result in skin irritation during the rst 812 weeks of treatment, they do not seem to worsen acne lesion counts during this period. Oral isotretinoin may be associated with depression, although the evidence to date is not entirely convincing. As better studies are awaited, it is prudent to continue to warn patients of a possible effect on depression and mood. There is good evidence that COCs are useful in reducing acne lesion counts. They should be given greater consideration for women with acne who need contraception. There is little evidence to support favouring COCs containing cyproterone acetate above other combined preparations for acne. There is no convincing evidence to support the use of topical or oral spironolactone for acne. Light and laser treatments have been shown to be of short-term benet if patients can tolerate some initial discomfort. Light and laser therapies have not been shown to be better than simple topical treatments. Longterm benets are unknown. Even though PDT is better than placebo for acne in the short term, it cannot be recommended at present for acne as a rst-line treatment because of its unacceptable local side-effects. One comparative trial has shown that PDT was less effective than 1% adapalene in the short-term reduction of inammatory lesions.

7 8

10 11

12

13

14

15

16

17

References
1 Ioannidis JPA. Contradicted and initially stronger effects in highly cited clinical research. JAMA 2005; 294: 21828. 2 Ingram JR, Grindlay DJ, Williams HC. Management of acne vulgaris: an evidence-based update. Clin Exp Dermatol 2009; 35: 3514. 3 Spencer EH, Ferdowsian HR, Barnard ND. Diet and acne: a review of the evidence. Int J Dermatol 2009; 48: 33947. 4 Smith RN, Mann NJ, Braue A et al. A low-glycemic-load diet improves symptoms in acne vulgaris patients: a randomized controlled trial. Am J Clin Nutr 2007; 86: 10715. 5 Smith RN, Mann NJ, Braue A et al. The effect of a highprotein, low glycemic-load diet versus a conventional, high

18

19

20

21

glycemic-load diet on biochemical parameters associated with acne vulgaris: a randomized, investigator-masked, controlled trial. J Am Acad Dermatol 2007; 57: 24756. Smith RN, Braue A, Varigos GA, Mann NJ. The effect of a low glycemic load diet on acne vulgaris and the fatty acid composition of skin surface triglycerides. J Dermatol Sci 2008; 50: 4152. Wilhelmus KR. Redundant publication of clinical trials on herpetic keratitis. Am J Ophthalmol 2007; 144: 2226. Fakhouri T, Yentzer BA, Feldman SR. Advancement in benzoyl peroxide-based acne treatment: methods to increase both efficacy and tolerability. J Drugs Dermatol 2009; 8: 65761. Mills OH Jr, Kligman AM, Pochi P, Comite H. Comparing 2.5%, 5%, and 10% benzoyl peroxide on inflammatory acne vulgaris. Int J Dermatol 1986; 25: 6647. Yentzer BA, McClain RW, Feldman SR. Do topical retinoids cause acne to flare? J Drugs Dermatol 2009; 8: 799801. Kontaxakis VP, Skourides D, Ferentinos P et al. Isotretinoin and psychopathology: a review. Ann Gen Psychiatry 2009; 8: 2. Albrecht J, Meves A, Bigby M. A survey of case reports and case series of therapeutic interventions in the Archives of Dermatology. Int J Dermatol 2009; 48: 5927. Azoulay L, Blais L, Berard A. Isotretinoin and the risk of depression in patients with acne: a case crossover study. Pharmacoepidemiol Drug Saf 2006; 15: S261. Arowojolu AO, Gallo MF, Lopez LM et al. Combined oral contraceptive pills for treatment of acne. Cochrane Database Syst Rev 2009 (3): CD004425. Brown J, Farquhar C, Lee O et al. Spironolactone versus placebo or in combination with steroids for hirsutism and or acne. Cochrane Database Syst Rev 2009 (2): CD000194. Walton S, Cunliffe WJ, Lookingbill P, Keczkes K. Lack of effect of topical spironolactone on sebum excretion. Br J Dermatol 1986; 114: 2614. Muhlemann MF, Carter GD, Cream JJ, Wise P. Oral spironolactone: an effective treatment for acne vulgaris in women. Br J Dermatol 1986; 115: 22732. Goodfellow A, Alaghband-Zadeh J, Carter G et al. Oral spironolactone improves acne vulgaris and reduces sebum excretion. Br J Dermatol 1984; 111: 20914. Hamilton FL, Car J, Lyons C et al. Laser and other light therapies for the treatment of acne vulgaris: systematic review. Br J Dermatol 2009; 160: 127385. Riddle CC, Terrell SN, Menser MB et al. A review of photodynamic therapy (PDT) for the treatment of acne vulgaris. J Drugs Dermatol 2009; 8: 101019. Taylor MN, Gonzalez ML. The practicalities of photodynamic therapy in acne vulgaris. Br J Dermatol 2009; 160: 11408.

122

The Author(s) CED 2010 British Association of Dermatologists Clinical and Experimental Dermatology, 36, 119123

Whats new in acne? E. V. Smith et al.

CPD questions
Learning objective

Question 4

The purpose of this activity is to review recent developments in dening the causes and in the treatment of acne, and to demonstrate up-to-date knowledge relating to the management of acne.
Question 1

In a recent case-crossover study, participants with depression were how many times more likely to have taken isotretinoin in the preceding ve months? a) 1.48 b) 2 c) 2.68 d) 3 e) 3.68
Question 5

Benzoyl peroxide treatment may help which of the following acne problems? a) Local skin irritation b) Post-inammatory skin pigmentation c) Antibiotic resistance d) Initial acne aring e) Incompliance
Question 2

What is the reason for not currently recommending photodynamic therapy (PDT) as a rst line treatment for acne? a) It is no better than placebo b) Unacceptable local side effects c) Laser is better than PDT d) The frequency of treatments required e) Lack of long term benet

Which of the following is recognised as a long-term side effect of laser treatment for acne? a) Pain b) Oedema c) Photosensitivity d) Erythema e) Desquamation
Question 3

Instructions for answering questions

This learning activity is freely available online at www.wileyblackwellcme.com. Users are encouraged to Read the article in print or online, paying particular attention to the learning points and any author conict of interest disclosures Reect on the article Register or login online at www.wileyblackwellcme. com and answer the CPD questions Complete the required evaluation component of the activity Once the test is passed, you will receive a certicate and the learning activity can be added to your RCP CPD diary as a self-certied entry.

Which of the following have been shown to be effective at reducing acne lesion counts? a) Milk exclusion diet b) Testosterone c) Topical spironolactone d) Oral spironolactone e) Combined oral contraceptives

The Author(s) CED 2010 British Association of Dermatologists Clinical and Experimental Dermatology, 36, 119123

123