Seminar
`
`Lancet 2012; 379: 361–72
`Published Online
`August 30, 2011
`DOI:10.1016/S0140-
`6736(11)60321-8
`This publication has
`been corrected.
`The corrected version fi rst
`appeared at thelancet.com
`on January 27, 2011
`Centre of Evidence-Based
`Dermatology, Nottingham
`University Hospitals NHS
`Trust, Nottingham, UK
`(Prof H C Williams PhD);
`Department of Dermatology,
`University of Colorado Denver,
`School of Medicine, Aurora and
`VA Eastern Colorado Health
`Care System, Denver, CO, USA
`(R P Dellavalle MD); Center for
`the Evaluation of Value and
`Risk in Health, Tufts Medical
`Centre, Boston, MA, USA
`(S Garner PhD), and The
`Commonwealth Fund,
`New York, NY, USA (S Garner)
`Correspondence to:
`Prof Hywel C Williams, Centre of
`Evidence-Based Dermatology,
`Nottingham University
`Hospitals NHS Trust,
`Nottingham NG7 2UH, UK
`hywel.williams@nottingham.
`ac.uk
`
`Acne vulgaris
`
`Hywel C Williams, Robert P Dellavalle, Sarah Garner
`
`Acne is a chronic infl ammatory disease of the pilosebaceous unit resulting from androgen-induced increased sebum
`production, altered keratinisation, infl ammation, and bacterial colonisation of hair follicles on the face, neck, chest,
`and back by Propionibacterium acnes. Although early colonisation with P acnes and family history might have important
`roles in the disease, exactly what triggers acne and how treatment aff ects the course of the disease remain unclear.
`Other factors such as diet have been implicated, but not proven. Facial scarring due to acne aff ects up to 20% of
`teenagers. Acne can persist into adulthood, with detrimental eff ects on self-esteem. There is no ideal treatment for
`acne, although a suitable regimen for reducing lesions can be found for most patients. Good quality evidence on
`comparative eff ectiveness of common topical and systemic acne therapies is scarce. Topical therapies including
`benzoyl peroxide, retinoids, and antibiotics when used in combination usually improve control of mild to moderate
`acne. Treatment with combined oral contraceptives can help women with acne. Patients with more severe infl ammatory
`acne usually need oral antibiotics combined with topical benzoyl peroxide to decrease antibiotic-resistant organisms.
`Oral isotretinoin is the most eff ective therapy and is used early in severe disease, although its use is limited by
`teratogenicity and other side-eff ects. Availability, adverse eff ects, and cost, limit the use of photodynamic therapy. New
`research is needed into the therapeutic comparative eff ectiveness and safety of the many products available, and to
`better understand the natural history, subtypes, and triggers of acne.
`
`Introduction
`Acne is a disease of the pilosebaceous unit—hair follicles
`in the skin that are associated with an oil gland (fi gure 1).2
`The clinical features of acne include seborrhoea (excess
`grease), non-infl ammatory lesions (open and closed
`comedones), infl ammatory lesions (papules and pustules),
`and various degrees of scarring. The distribution of acne
`corresponds to the highest density of pilosebaceous units
`(face, neck, upper chest, shoulders, and back). Nodules
`and cysts comprise severe nodulocystic acne. This Seminar
`summarises information relating to the clinical aspects of
`common acne
`(acne vulgaris). Acne classifi cation,
`scarring, acne rosacea, chloracne, acne associated with
`polycystic ovary syndrome, infantile acne, acne inversa,
`and drug-induced acne have been reviewed elsewhere.3–10
`
`Prevalence and natural history
`Some degree of acne aff ects almost all people aged 15 to
`17 years,11–13 and is moderate to severe in about 15–20%.8,12,14
`Prevalence estimates are diffi cult to compare because
`defi nitions of acne and acne severity have diff ered so
`much between studies, and because estimates are
`confounded by
`the availability and use of acne
`treatments.15 Surveys of self-reported acne have proven
`unreliable.16 Although perceived as a teenage disease,
`acne often persists into adulthood.17,18 One population
`study in Germany found that 64% of those aged 20 to
`29 years and 43% of those aged 30 to 39 years had visible
`acne.19 Another study of more than 2000 adults showed
`that 3% of men and 5% of women still had defi nite mild
`acne at the age of 40 to 49 years.20
`Acne typically starts in early puberty with increased
`facial grease production, and mid-facial comedones8
`followed by infl ammatory lesions. Early-onset acne
`(before the age of 12 years) is usually more comedonal
`than infl ammatory, possibly because such individuals
`have not yet begun to produce enough sebum to support
`
`large numbers of Proprionibacterium acnes.21 One
`prospective study of 133 children aged 5·5 to 12 years,
`followed up for an average of 2·5 years, found
`asynchronous facial sebum production initially, with
`increasing numbers of glands switching on sebum
`production over time.22 Subsequent expansion of the
`propionibacterial skin fl ora (in the nares and then facial
`skin) occurred earlier in children who developed acne
`than in children of the same age and pubertal status who
`did not, suggesting that postponement of sebum
`production or expansion of propionibacterial skin fl ora
`until after puberty could prevent acne or minimise
`disease severity. Predictors of acne severity include early
`onset of comedonal acne,8 and increasing number of
`family members with acne history.14 Factors that can
`cause acne to fl are include the menstrual cycle, picking,
`and emotional stress.23,24 Beliefs about external factors
`aff ecting acne vary according to ethnic group.25 Acne
`vulgaris is a chronic disease that often persists for many
`years.26 There is little research about what factors might
`predict whether acne will last into adulthood.27 We could
`not fi nd any good quality cohort studies summarising
`the natural history of acne. Sequential prevalence surveys
`of diff erent populations showing a gradual decrease in
`
`Search strategy and selection criteria
`
`Our main sources of evidence included all systematic reviews
`on acne published since 1999 which have been mapped by
`NHS Evidence—skin disorders annual evidence updates,1
`supplemented by specifi c searches on Medline for articles
`published between January, 2003, and Jan 16, 2011, using the
`search terms “acne”, “comedones”, “vulgaris”, and
`“aetiology”, “causes”, “natural history”, “pathophysiology”,
`“treatment”, “management”, and “guidelines”. We also
`scrutinised citation lists from retrieved articles.
`
`www.thelancet.com Vol 379 January 28, 2012
`
`361
`
`1
`
`AMN1025
`
`

`

`Seminar
`
`A
`
`B
`
`C
`
`Figure 1: Normal sebaceous follicle (A) and comedo (B), and infl ammatory acne lesion with rupture of follicular wall and secondary infl ammation (C)
`Reproduced, with permission, from reference 2.
`
`acne prevalence after the age of 20 years weakly underpin
`our current understanding of the natural history of acne.
`Mild infl ammatory acne declines or disappears in a large
`proportion of those with acne in their teens. Cytokines
`that induce comedogenic changes at the follicular
`infundibulum might also inhibit lipid secretion from the
`sebaceous gland, resulting in remission of individual
`lesions.28 However, seborrhoea persists throughout adult
`life, long after infl ammatory lesions have resolved.29
`Adult acne related to circulating androgens goes by
`several names, including post-adolescent or late-onset
`acne, and occurs most commonly in women beyond the
`age of 25 years.30
`
`Cause
`Risk factors and genes associated with acne prognosis
`and treatment are unclear.31,32 Twin studies have pointed
`to the importance of genetic factors for more severe
`scarring acne.33 A positive family history of acne doubled
`the risk of signifi cant acne in a study of 1002 Iranian
`16-year-olds,14 and the heritability of acne was 78% in fi rst-
`degree relatives of those with acne in a large study of
`Chinese undergraduates.34 Acne appears earlier in girls,
`but more boys are aff ected during the mid-teenage years.35
`Acne can occur at a younger age and be more comedonal
`in black children than in white children, probably from
`earlier onset of puberty.36 A study of 1394 Ghanaian
`schoolchildren found that acne was less common in rural
`locations, but the reasons for this are unclear.37
`Although earlier observational studies suggested an
`inverse association between smoking and acne,38 sub-
`sequent studies have shown that severe acne increases
`with smoking.19,39 Increased insulin resistance and high
`serum dehydroepiandrosterone might explain
`the
`presence of acne in polycystic ovary syndrome.40,41
`Occlusion of the skin surface with greasy products
`(pomade acne),42 clothing, and sweating can worsen acne.
`Drugs such as anti-epileptics
`typically produce a
`
`monomorphic acne, and acneiform eruptions have been
`associated with anti-cancer drugs such as gefi tinib.10 The
`use of anabolic steroids for increasing muscle bulk might
`be underestimated, and can give rise to severe forms of
`acne.43 Tropical acne can occur in military personnel
`assigned to hot, humid conditions.44 Dioxin exposure can
`result in severe comedonal acne (chloracne), but it is not
`associated with common acne.
`Diet, sunlight, and skin hygiene have all been
`implicated in acne,45 but little evidence supports or
`refutes such beliefs.46 One systematic review suggested
`that dairy products (especially milk) increase acne risk,
`but all the included observational studies had signifi -
`cant shortcomings.47
`Previous studies of giving young people large quantities
`of chocolate to try and provoke acne were too small and
`too short to claim no eff ect.48 The apparent absence of
`acne in native non-Westernised people in Papua New
`Guinea and Paraguay49 has led to the proposal that high
`glycaemic loads in Western diet could have a role in acne,
`perhaps through hyperinsulinaemia leading to increased
`androgens, increased insulin-like growth-factor 1, and
`altered retinoid signalling.50,51 A randomised controlled
`trial showing that a low glycaemic load diet might
`improve acne provides preliminary support for this
`theory.52 Although acne has been associated with
`increasing body mass,53 no evidence suggests that putting
`people on restrictive diets reduces acne.
`
`Disease mechanisms
`Four processes have a pivotal role in the formation of
`acne lesions: infl ammatory mediators released into the
`skin; alteration of the keratinisation process leading to
`comedones; increased and altered sebum production
`under androgen control (or increased androgen receptor
`sensitivity); and follicular colonisation by P acnes.27 The
`exact sequence of events and how they and other factors
`interact remains unclear.
`
`362
`
`www.thelancet.com Vol 379 January 28, 2012
`
`2
`
`

`

`Seminar
`
`infl ammatory processes might
`Immune-mediated
`involve CD4+
`lymphocytes and macrophages
`that
`stimulate the pilosebaceous vasculature precede follicular
`hyperkeratinisation.54 Defective terminal keratinocyte
`diff erentiation leads to comedo formation under the
`infl uence of androgens and qualitative changes in the
`sebum lipids that induce interleukin 1 (IL1) secretion.55
`Sebaceous glands are an important part of the innate
`immune system, producing a variety of antimicrobial
`peptides, neuropeptides, and antibacterial lipids such as
`sapienic acid. Each sebaceous gland functions like an
`independent endocrine organ infl uenced by corticotropin-
`releasing hormone, which might mediate the link
`between stress and acne exacerbations.56 Vitamin D also
`regulates sebum production, and insulin-like growth-
`factor 1 might increase sebum through sterol-response-
`element-binding proteins.57 Oxidised lipids such as
`squalene can stimulate keratinocyte proliferation and
`other
`infl ammatory
`responses mediated by
`the
`proinfl ammatory
`leukotriene B4.58 Matrix metallo-
`proteinases in sebum have an important role in
`infl ammation, cell proliferation, degradation of the
`dermal matrix, and treatment responsiveness.59
`Sebaceous follicles containing a microcomedone
`provide an anaerobic and lipid-rich environment in
`which P acnes fl ourishes.60 Lipogenesis is directly
`augmented by P acnes.61 Colonisation of facial follicles
`with P acnes follows the asynchronous initiation of
`sebum production,22 which might explain why treatment
`with isotretinoin treatment too early can need to be
`followed up with subsequent courses, as new previously
`P acnes-naive follicles become colonised and infl amed.
`Unique P acnes strains with diff erent bacterial resistance
`profi les colonise diff erent pilosebaceous units and
`induce infl ammation by the activation of toll-like
`receptors in keratinocytes and macrophages.62 In-vitro
`work suggests that P acnes could behave like a biofi lm
`within follicles,
`leading to decreased response to
`antimicrobial agents.63 P acnes resistance to commonly
`used oral antibiotics for acne aff ects treatment response,
`suggesting that direct antimicrobial eff ects might be
`important in addition to the anti-infl ammatory actions
`of antibiotics.64
`
`How does acne aff ect people?
`Acne results in physical symptoms such as soreness,
`itching, and pain, but its main eff ects are on quality of life.
`Psychological morbidity is not a trivial problem,65 and it is
`compounded by multiple factors: acne aff ects highly
`visible skin—a vital organ of social display; popular culture
`and societal pressures dictate blemishless skin; acne can
`be dismissed by health-care professionals as a trivial self-
`limiting condition; and acne peaks in teenage years, a
`time crucial for building confi dence and self-esteem.
`Case-control and cross-sectional studies assessing the
`eff ect of acne on psychological health found a range of
`abnormalities including depression, suicidal ideation,
`
`anxiety, psychosomatic symptoms, shame, embarrass-
`ment, and social inhibition,66 which improve with eff ective
`treatment.67 Anger inversely correlates with quality of life
`in acne and satisfaction with acne treatment.68 Patients
`might not volunteer depressive symptoms and need
`prompting during consultation. UK teenagers with acne
`twice as often scored in the borderline or abnormal range
`on an age-appropriate validated questionnaire of emotional
`wellbeing than did those who did not have acne, and had
`higher levels of behavioural diffi culties.69 The presence of
`acne was associated with unemployment in a case-control
`study of young men and women.70 One community study
`of 14–17-year-old Australian students reported no
`association between acne and subsequent psychological
`or psychiatric morbidity, a surprising fi nding perhaps
`explained by eff ective treatments or personality traits.71
`Acne severity and degree of psychological impairment
`do not necessarily correspond—mild disease in one
`person can cause high degrees of psychological disability,
`whereas another with more severe disease can seem less
`bothered by
`their acne.12 Most studies assessing
`psychological morbidity in acne have been cross-sectional,
`and therefore unable to establish causal direction. Few
`studies report the direct and indirect costs of acne.72,73
`
`How can acne be managed?
`Skin hygiene
`There is no good evidence that acne is caused or cured by
`washing.46 Antibacterial skin cleansers might benefi t
`mild acne, and acidic cleansing bars are probably better
`than standard alkaline soaps. However, excessive washing
`and scrubbing removes oil from the skin surface, drying
`it and stimulating more oil production. Antibacterial skin
`cleansers provide no additional benefi t to patients already
`using other, potentially irritating topical treatments.46
`
`Counselling and support
`Spending time dispelling myths and explaining that
`most treatments will not cure is worthwhile and might
`improve adherence.74 Because acne treatments work by
`preventing new lesions rather than treating existing ones,
`an initial response might not appear for some weeks.
`Most eff ective treatments can require months to work.75
`Health-care providers should assess loss of self-esteem,
`lack of confi dence, and symptoms of depression including
`suicidal thoughts. Acne’s emotional eff ect might not be
`immediately evident or volunteered, but even mild acne
`can cause signifi cant distress. Patients should also be
`told that online acne information, including from some
`support groups, varies in quality and can refl ect sponsor
`bias, and clinicians have a role in guiding them to
`trustworthy resources.
`
`Treatment guidelines
`The many over-the-counter and prescription treatments
`for acne allow for a large number of potential combin-
`ation treatments. A comprehensive systematic review
`
`www.thelancet.com Vol 379 January 28, 2012
`
`363
`
`3
`
`

`

`Seminar
`
`Sebum excretion
`
`Keratinisation
`
`Follicular
`Proprionibacterium
`acnes
`
`Infl ammation
`
`Benzoyl peroxide
`Retinoids
`Clindamycin
`Antiandrogens
`Azelaic acid
`Tetracyclines
`Erythromycin
`Isotretinoin
`
`–
`–
`–
`++
`–
`–
`–
`+++
`
`(+)
`++
`(+)
`+
`++
`–
`–
`++
`
`+++
`(+)
`++
`–
`++
`++
`++
`(++)
`
`(+)
`+
`–
`–
`+
`+
`–
`++
`
`+++=very strong eff ect. ++=strong eff ect. +=moderate eff ect. (+)=indirect/weak eff ect. –=no eff ect.
`
`Table: Targets of acne treatments
`
`in 1999 identifi ed 274 trials of 140 treatments in
`250 combin ations.76 Most were placebo-controlled studies
`of me-too products, and the authors found no basis from
`controlled trials to judge the effi cacy of any treatment in
`relation to others, nor in the sequence of therapy. The
`table shows how diff erent treatment medications target
`diff erent aspects of acne pathology. The large number of
`products and product combinations, and the scarcity of
`comparative studies, has led to disparate guidelines with
`few recommendations being evidence-based. Recent
`acne guidelines include those from the Global Alliance
`to Improve Outcomes in Acne,77 the American Academy
`of Dermatology/American Academy of Dermatology
`Association,78 and the European expert group on oral
`antibiotics in acne.79 Because of the paucity of evidence,
`these guidelines rely on the opinions of experts, many of
`whom declare signifi cant potential confl icts of interest.
`Practical advice on how to manage acne based on a
`systematic search of evidence by an independent team is
`available in an online UK Clinical Knowledge Summary.75
`All of these guidelines illustrate similar approaches on
`which initial therapies should be based—ie, acne severity
`and whether the acne is predominantly non-infl ammatory
`or infl ammatory. We propose an algorithm for treating
`acne in fi gure 2 on the basis of our interpretation of the
`clinical evidence. This interpretation diff ers slightly
`from
`the Global Alliance
`recommendations by
`suggesting slightly more initial use of topical benzoyl
`peroxide than topical retinoids on the grounds of cost
`and on a longer track record of effi cacy and safety.
`Assessment of treatment response in such a polymorphic
`condition can be diffi cult and should include an
`assessment of reduction of infl ammatory and non-
`infl ammatory lesions in relation to baseline photographs,
`plus an assessment of psychological wellbeing.
`
`Topical treatments
`Topical agents when used alone or in combination
`eff ectively treat mild acne consisting of open and closed
`comedones with a few infl ammatory lesions.77 The many
`treatment options off er diff erent modes of action. Although
`
`all are more eff ective than placebo, establishing the most
`appropriate strategy for initial and maintenance treatment
`requires further research.77,80 Topical treatments only work
`where applied. Because topical therapies reduce new lesion
`development they require application to the whole aff ected
`areas, rather than individual spots. Most cause initial skin
`irritation, and some people stop using them because of
`this. The irritation can be minimised by starting with lower
`strength preparations and gradually increasing frequency
`or dose. Where irritation persists, a change in formulation
`from alcoholic solutions to washes or gels to more
`moisturising creams or lotions might help.
`
`Benzoyl peroxide
`Benzoyl peroxide is a safe and eff ective81 over-the-counter
`preparation that has several mechanisms of action, and
`should be applied to all the aff ected area.82 Single-agent
`benzoyl peroxide works as well as oral antibiotics or a
`topical antibiotic combination that included benzoyl
`peroxide for people with mild-to-moderate facial acne.64 It
`has greater activity than topical (iso)tretinoin against
`infl ammatory lesions;83,84 the results of two further
`underpowered trials were equivocal.85,86 Further studies
`are needed, especially as combination therapy might be
`better.86 Benzoyl peroxide causes initial local irritation.
`Patients need to be counselled to expect irritation but
`discontinue treatment if it becomes severe. Irritation will
`decrease in most cases, especially if patients start applying
`it every other day and then increase the frequency. Low
`strength (2·5% or 5%) benzoyl peroxide is recommended,
`since it is less irritating and there is no clear evidence that
`stronger preparations are more eff ective.87
`
`Topical retinoids
`Treatment with tretinoin, adapalene, and isotretinoin
`require medical prescriptions. Tazarotene is not licensed
`in the UK for acne. All retinoids are contraindicated in
`pregnancy, and women of childbearing age must use
`eff ective contraception. Topical retinoids act on abnormal
`keratinisation and are also anti-infl ammatory, so they
`work for both comedonal and infl ammatory acne. Many
`placebo-controlled or non-inferiority studies citing better
`tolerability exist, but few trials guide practice. More trials
`comparing retinoids against each other and against other
`therapies are needed. Randomised controlled trials
`(RCTs) have shown that higher-strength preparations
`might have greater activity than lower-strength ones, but
`at the expense of more irritation. All topical retinoids
`induce local reactions, and should be discontinued if
`severe. They do not seem to cause temporary worsening
`of acne lesions,88 but can increase the sensitivity of skin
`to ultraviolet light.
`
`Topical antibiotics
`How topical antibiotics improve acne has not been
`clearly defi ned, but they seem to act directly on P acnes
`and reduce infl ammation. Topical antibiotics have less
`
`364
`
`www.thelancet.com Vol 379 January 28, 2012
`
`4
`
`

`

`Seminar
`
`Mild acne
`
`Mainly red spots
`
`Mainly blackheads
`
`Mixture of lesions
`
`Take baseline photographs if possible to help assess subsequent treatment
`
`Start 2·5% BP
`
`Start topical retinoids
`
`Start topical combination*
`
`Build up over weeks until tolerance to irritation develops. If not enough benefit when assessed at 6–8 weeks:
`
`Add in topical retinoid
`or topical antibiotic
`or azelaic acid
`
`Add in topical antibiotic
`or topical BP if tolerable
`or azelaic acid
`
`Try different combination
`product
`or azelaic acid
`
`Continue with topical therapy as long as benefit continues. If not, progress to oral treatments as for moderate acne
`
`Moderate acne
`(or back acne or mild acne that fails to respond to topical therapy)
`
`Severe acne
`(or moderate acne not responding to oral therapy)
`
`Women or older teenagers for whom
`contraception needed or is acceptable
`
`Women or older teenagers who do
`not need or want contraception,
`and men
`
`If results are not good or are not sustained with the above treatments—
`eg, two 8-week courses of different oral antibiotics without significant benefit:
`
`Start combined oral contraceptive
`
`Try topical combination product
`
`Proceed to oral isotretinoin early before scarring occurs (avoid wasting time
`with several prolonged courses of oral antibiotics if ineffective)
`
`Assess at 6 weeks. If no improvement then proceed to oral antibiotics plus
`topical BP or retinoid (but not topical antibiotic)
`
`Counsel for adverse effects and ensure adequate contraception for women of
`child-bearing potential
`
`Assess at 6 weeks. If no improvement, try a different oral antibiotic plus
`a topical
`
`If there is a beneficial response, carry
`on for 4–6 months. Then stop and
`use a 2·5% BP cream washout for
`2 weeks to eradicate resistant
`Propionbacterium acnes. Then try
`further topicals as for mild acne as a
`maintenance treatment, or if acne
`relapses return to oral antibiotics.
`
`If initial benefit is lost within the
`2–6-month period, stop oral
`antibiotics and try another oral
`antibiotic after a BP washout.
`
`Figure 2: Suggested algorithm for treatment of mild, moderate, and severe acne based on our appraisal of current clinical evidence and uncertainties
`Figures reproduced with permission from DermNet NZ. BP=benzoyl peroxide. *Topical combination could be benzoyl peroxide plus topical antibiotic, or topical
`benzoyl peroxide plus topical retinoid.
`
`activity than other agents against non-infl amed lesions.
`For more severe acne, topical antibiotics are usually
`combined with other products such as topical retinoids
`
`or benzoyl peroxide. Patients with back acne might
`respond better to oral antibiotic therapy because of the
`diffi culties of applying treatments to large areas that are
`
`www.thelancet.com Vol 379 January 28, 2012
`
`365
`
`5
`
`

`

`Seminar
`
`diffi cult to reach. Topical antibiotics include clindamycin,
`erythromycin, and tetracycline. Topical antibiotics are
`also available in combination with benzoyl peroxide and
`zinc acetate. Alcohol-based preparations are more
`drying, and therefore more suitable for oilier skins. The
`effi cacy of erythromycin might be declining because of
`bacterial resistance.89
`
`Other topical therapies
`Salicylic acid is an exfoliant and is a component of many
`over-the-counter preparations. No studies support routine
`use of salicylic acid in preference to other topical therapies.
`The American Guidelines state that data from peer-
`reviewed literature regarding the effi cacy of sulphur,
`resorcinol, sodium sulfacetamide, aluminium chloride,
`and zinc are limited.78 Similarly there is no reliable
`evidence to support the use of nicotinamide or combination
`triethyl citrate and ethyl linoleate.90 Despite recent interest
`in topical dapsone91 and taurine bromamine,92 neither is
`licensed in the UK, and current comparative evidence
`does not support a change in practice. A new vehicle,
`emollient foam, containing sodium sulfacetaminde 10%
`and sulphur 5% is now available for acne treatment in the
`USA.93 Azelaic acid has both antimicrobial and anti-
`comedonal properties but can cause hypopigmentation,
`and darker-skinned patients should therefore be monitored
`for signs. Anecdotal reports have suggested that azelaic
`acid might reduce post-infl ammatory hyperpigmentation,
`which is possibly attributable to its activity on abnormal
`melanocytes. The American Guidelines note that its
`clinical use, compared to other agents, has limited effi cacy
`according to experts.78
`
`Combination topicals
`There is accumulating information that combinations of
`topical treatments with diff erent mechanisms of action
`work better than single agents.94 Few combinations have
`been tested properly against the relevant monotherapy.
`The trials tend to be methodologically fl awed by factors
`such as suboptimal dose or frequency of monotherapy.82
`Compliance can be increased with once-daily combination
`products because of their convenience and faster speed
`of onset,95,96 although individual generic preparations
`used concomitantly might be more cost-eff ective.64
`Benzoyl peroxide inactivates tretinoin, and the two agents
`should not therefore be applied simultaneously; if used
`in combination one should be applied in the morning
`and one at night.
`
`Oral treatments
`Oral antibiotics
`Oral antibiotics are usually reserved for more severe
`acne, acne predominantly on the trunk, acne unresponsive
`to topical therapy, and in patients at greater risk of
`scarring. Although antibiotics have shown eff ectiveness
`in terms of reducing the number of infl ammatory lesions,
`none clear acne completely. Most patients seek acne
`
`clearance rather than reduction in lesion counts. There is
`no conclusive evidence that one antibiotic is more
`eff ective than another (including fi rst and second
`generation tetracyclines) or that oral antibiotics are more
`eff ective than topical preparations for mild-to-moderate
`facial acne.64 There is no evidence that higher doses are
`more eff ective than lower doses or that controlled-release
`preparations are necessary.64,76,82,97
`The choice of antibiotic should therefore be based on
`the patient’s preference, the side-eff ect profi le, and cost.
`The tetracyclines (tetracycline, oxytetracycline, doxy-
`cycline, or lymecycline) are the preferred options;
`minocycline has signifi cant adverse eff ects.98 Co-
`trimoxazole should be avoided because the sulfa-
`methoxazole component has signifi cant side-eff ects.
`Quinolones are not recommended in adolescents due to
`arthropathy risks and because oral ciprofl oxacin shows
`rapid selectivity that promotes resistance.99 Amino-
`glycosides and chloramphenicol have very limited
`eff ects79 and oral clindamycin, although eff ective, has
`the potential for signifi cant adverse eff ects such as
`pseudomembranous
`colitis. There
`is
`increasing
`resistance
`to
`the macrolides
`(erythromycin and
`azithromycin) and
`trimethoprim
`that
`is causing
`worldwide concern.
`The use of antibiotics for acne has been questioned
`owing to resistance concerns, especially since they are
`used for long periods at low doses.100 Concomitant
`benzoyl peroxide can reduce problems with bacterial
`resistance,101 whereas concomitant
`treatment with
`diff erent oral and topical antibiotics should be avoided.
`Data from a large well-reported RCT indicated that
`6–8 weeks is an appropriate time to assess response.64 If
`an individual does not respond to antibiotics or stops
`responding, there is no evidence that increasing the
`frequency or dose is helpful. Such strategies increase
`selective pressure without
`increasing
`effi cacy.82
`Antibiotics should be stopped if no further improvement
`is evident. Antibiotics should not be routinely used for
`maintenance because alternatives exist with similar
`effi cacy and preventative action.79,82 Benzoyl peroxide
`protects against resistance by eliminating resistant
`bacteria: the Global Alliance to Improve Outcomes in
`acne (2003) recommends that if antibiotics must be
`used for longer than 2 months, benzoyl peroxide should
`be used for a minimum of 5–7 days between antibiotic
`courses to reduce resistant organisms from the skin.77
`
`Oral contraceptives
`(COCs) contain an
`Combined oral contraceptives
`oestrogen (ethinylestradiol) and a progestogen. COCs are
`frequently prescribed for women with acne because
`oestrogen suppresses sebaceous gland activity and
`decreases
`the
`formation of ovarian and adrenal
`andro gens. Progestogen-only
`contraceptives often
`worsen acne and should be avoided in women who
`have no contraindications
`to oestrogen-containing
`
`366
`
`www.thelancet.com Vol 379 January 28, 2012
`
`6
`
`

`

`Seminar
`
`A
`
`B
`
`Figure 3: Before (A) and after (B) view of a woman with severe acne treated with a course of isotretinoin
`Reproduced with permission from Amy Derick; full patient consent was received.
`
`studies were generally of poor quality and inconclusive.
`Another systematic review found some benefi t for
`acupuncture with moxibustion, but the quality of
`included studies was limited.109 A systematic review of
`four RCTs of tea-tree oil in 2000 did not fi nd conclusive
`evidence of benefi t,110 although a recent well-reported
`study of 60 people in Iran with mild-to-moderate acne
`showed a modest reduction in lesion count and few
`local adverse eff ects when compared with placebo,
`suggesting that larger trials might be worthwhile.111
`CAM cannot be recommended for acne treatment
`because it is not supported by good evidence—CAMs
`might work, but the key studies have not been done, or
`when done, they have been inconclusive or reported
`poorly. CAM therapy for acne is a research gap that
`needs to be addressed given the high degree public
`interest and spending on CAM approaches.
`
`Special clinical problems
`The depth and extent of acne scarring varies and can be
`improved by multiple procedures including subcision,
`punch excision, laser resurfacing, dermabrasion, and
`chemical peels.27,112 Increasingly acne scarring is being
`treated with fractionated laser treatments—a technique
`that produces thousands of microthermal areas of
`dermal ablation separated by areas of untreated skin,
`with fewer side-eff ects and a quicker healing period than
`ablative lasers.113
`Whereas open comedones can often be extracted with
`minimal skin trauma, cysts and closed comedones
`provide more challenging targets for acne surgery. Closed
`comedones can be nicked with a bevelled needle before
`expression wi