European recommendations on the use of oral antibiotics for acne

Therapy Brigitte DRÉNO 1 Vincenzo BETTOLI 2 Falk OCHSENDORF 3 Alison LAYTON 4 Håkan MOBACKEN 5 Hugo DEGREEF 6 1 Department of Dermatology, Hotel Dieu, Place Alexis Ricordeau, 44093 Nantes Cedex 01, France 2 Clinica Dermatologica, Azienda Ospedaliera Arispedale S. Anna, Università degli Studi di Ferrara, Corso Giovecca 189, 44100 Ferrara, Italy 3 Klinikum der Johann Wolfgang Goethe-Universitat, Zentrum der Dermatologie und Venerologie (ZDV) Theodor-Stern-Kai 7, D-60590 Frankfurt am Main, Germany 4 Harrogate District Hospital, Lancaster Park Rd, Harrogate, North Yorks, HG2 7SX, UK 5 Department of Dermatology, Sahlgrenska University Hospital, S-413 45 Göteborg, Sweden 6 University Hospital St Rafaël, Department of Dermatology, Kaucijnenvoer, 33, B3000 Leuven, Belgium Reprints: B. Dréno. Fax: (+ 33) 240083117 brigitte.dreno@wanadoo.fr Article accepted on 16/7/2004 Eur J Dermatol 2004; 14: 391-9 European recommendations on the use of oral antibiotics for acne Non-rational prescribing of oral antibiotics in acne is common, and there is currently an unmet need for up-to-date guidelines that specifically address these issues. Presented here is a set of recommendations on the use of oral antibiotics in acne, developed by a group of European acne specialists, designed to be considered by dermatologists and general practitioners in their daily practice throughout Europe. Recommendations cover optimal choice of antibiotic, drug doses, duration of treatment, combination treatment, and maintenance therapy. Key words: Acne vulgaris, oral antibiotics, treatment recommendations. Disclaimer These recommendations have been prepared for dermatologists and general practitioners on behalf of the European expert group on oral antibiotics in acne and reflect the best data available at the time the report was prepared. Caution should be exercised in interpreting the data: the results of future studies may require alteration of the conclusions or recommendations of this report. It may be necessary or even desirable to depart from the recommendations in the interests of specific patients or special circumstances. Just as adherence to these recommendations may not constitute a defence against a claim of negligence, so deviation from them should not necessarily be deemed negligent. T here are many different drugs available for the treatment of acne vulgaris, including oral and topical antibiotics, topical benzoyl peroxide, topical retinoids, oral isotretinoin, oral anti-androgens, and zinc salts. Many guidelines on how these different agents can be optimally used have been published over the years. These include the Ad hoc committee report on the use of systemic antibiotics for treatment of acne vulgaris (1975) [1]; the American Academy of Dermatology Guidelines for care of acne vulgaris (1990) [2]; Treatment of acne vulgaris: guidelines for primary care physicians (1991) [3]; Oral treatment of acne in France (1999) [4]; and most recently the Global alliance to improve outcomes in acne consensus recommendations (2003) [5]. Figure 1 shows the acne treatment algorithm proposed by the Global alliance, in which oral antibiotics, used in combination with topical retinoids, are recommended for the treatment of moderate to severe inflammatory acne. Other than the 1975 ad hoc committee report [1], the other guidelines listed above provide general recommendations on acne management, but fall short of providing detailed guidance on the use of oral antibiotics. This is probably due to the fact that much of the clinical evidence derives from trials that vary greatly in acne definition, end-points, doses used, duration of treatments etc., making direct comparisons difficult. Perhaps as a consequence of this lack of guidance, there is considerable variation in the way in which antibiotics are used to treat acne across Europe, and their inappropriate use is alarmingly common (patients are often treated for excessively long periods of time, and/or there is much variation in antibiotic dosage given). The preferred choice of antibiotic varies from country to country, with first generation cyclines (tetracycline HCl, and oxytetracycline) very popular in the UK, minocycline preferred in Belgium, and lymecycline particularly popular among dermatologists in France, Italy, and Nordic countries. Some of these differences may be accounted for by climate (for example, doxycycline, known to have the potential to cause dose-dependent photosensitivity, is less commonly prescribed in Southern Europe during the summer), and by various pharmacoeconomic considerations. Thus, despite the problems in forming truly evidence-based guidelines on the use of oral antibiotics in acne, there is clearly an unmet need for recommendations on the use of EJD, vol. 14, n 6, November-December 2004 391

Mild Moderate Severe Comedonal Papular/pustular Papular/pustular Nodular 1 Nodular/ conglobate 1 st3 Topical retinoid Topical retinoid + Oral antibiotic + Oral antibiotic + Oral isotretinoin 2 topical antimicrobial topical retinoid +/ BP topical retinoid + BP Alternative 3 Azelaic acid or salicylic acid Alternative topical antimicrobial + alternative topical Alternative oral antibiotic + alternative topical Oral isotretinoin or alternative oral antibiotic + High dose oral antibiotic + topical retinoid + BP retinoid or azelaic acid 4 retinoid +/ BP alternative topical retinoid +/ BP/azelaic acid 4 Alternatives for females 3 See 1 st choice See 1 st choice Oral antiandrogen + topical retinoid/azelaic acid 4 +/ BP Oral antiandrogen + topical retinoid +/ oral antibiotic +/ alternative antimicrobial High dose oral antiandrogen + topical retinoid +/ alternative topical antimicrobial Maintenance: Topical retinoid +/ BP 1 Papulopustular acne with some nodular lesions; 2 Second course in case of relapse; 3 Consider physical removal of comedones. 4 There was no consensus on this alternative recommendation. However, in some countries, azelaic acid prescribing is appropriate practice. BP: benzoyl peroxide. Figure 1. Acne treatment algorithm suggested by Gollnick et al. (2003) [5]. Reprinted from the Journal of the American Academy of Dertatology, Vol. 49 (I suppl.); Gollnick et al. Management of acne: a report from a global alliance to improve outcomes in acne, pages S1-S38, 2003, American Academy of Dermatology, Inc, with permission from Elsevier. oral antibiotics that can be used in the daily practice of physicians treating acne across Europe. This paper therefore presents the available clinical data and expert opinion, followed by a set of detailed and user-friendly recommendations on many aspects of the use of oral antibiotics in acne. Methodology These recommendations were developed over a series of three meetings in 2002 and 2003. During the first two meetings, a core of six independent European acne specialists reviewed current practices around Europe; conducted a systematic literature review (using Medline) covering the years 1992 to 2003; and discussed personal experiences. In a final workshop, the findings of this core group were presented to the wider group of 23 acne specialists, mainly from Europe, but also Brazil and Morocco (Appendix 1) for 392 discussion and review. Recommendations are based on efficacy, practical applicability in daily practice, safety/tolerability, antimicrobial resistance, and pharmacoeconomic considerations. I. Literature review The pathophysiology of acne and rationale for using antibiotics There are a number of pathophysiologic components to acne, including sebaceous gland hyperplasia with seborrhoea; altered follicular growth and differentiation; microbial colonization; and inflammation and other immune responses [5]. The precursor lesion in all acne is the microcomedone, which features altered follicular growth and differentiation, and sebaceous gland hyperplasia with seborrhoea. Microcomedones can then enlarge to form non-inflammatory closed or open comedones, and micro- EJD, vol. 14, n 6, November-December 2004

bial colonization can result in the formation of inflammatory lesions (papules, pustules, or nodules). By using different agents in combination, acne pathophysiology can be targeted from a number of different angles simultaneously, improving therapeutic outcome [5]. For example, topical retinoids (some of which have proven anti-inflammatory properties) target the microcomedone, and are therefore suitable for use in combination with other drugs that target Propionibacterium acnes and inflammation, and may be suitable for maintaining remission following successful treatment [6]. Microorganisms that are commonly present on the skin of patients with inflammatory acne include the yeasts Pityrosporum spp., coagulase-negative staphylococci, and P. acnes [7]. Eradication of Pityrosporum spp., using antimycotic therapy, does not have any effect on acne [8], and use of antibiotics induces resistance in coagulase-negative staphylococci well in advance of observed responses to treatment [9]. Studies have shown that infection with the gram-positive, pleomorphic, anaerobic rod, P. acnes causes inflammation of sterile cysts [10], but that dead P. acnes or living Staphylococcus epidermidis do not cause inflammation [11]. Therefore P. acnes, has been implicated in inflammatory acne lesions. The concentration of P. acnes generally correlates with patient s sebum production [12], but not with the degree of inflammation or the severity of acne [7, 13, 14]. However, the humoral and cellular immune responses to P. acnes correlate with acne severity. Colonization with P. acnes results in secretion of extracellular enzymes, cytokines such as IL-1a, and heat-shock proteins, all of which have mitogenic effects on T-cells [15-19]. Thus P. acnes is associated with inflammatory acne not in a concentrationdependent manner, but in an inflammation-dependent manner. Antibiotics that can both reduce the number of P. acnes and reduce inflammation by different mechanisms are therefore of utility in moderate to severe acne associated with papules, pustules, and nodules, and in acne conglobata (Figure 1). A recent study has shown that inflammation features in the very earliest stages of acne lesion development [20]. Follicles without microcomedonal features had elevated CD3 T-cells, CD4 T-cells, and macrophages in the perifollicular and papillary dermis, as well as changes and activation of vascular intercellular adhesion molecules [20]. Oral antibiotics used in acne Cyclines, macrolides, clindamycin, trimethoprim, cotrimoxazole, and quinolones all have efficacy in acne [21, 22]. Notably, penicillins, cephalosporins, aminoglycosides, and chloramphenicol have very limited effects in inflammatory acne [1, 23, 24]. Cyclines Cyclines (tetracycline HCl, oxytetracycline, lymecycline, doxycycline, and minocycline) have very good efficacy in acne and generally have a good safety profile. Side effects include gastrointestinal disturbance, and some drugspecific effects discussed later in this paper. There is crossresistance within the class, but no cross-resistance to other antibiotic classes. Cyclines are contraindicated in children under 8-12 years (varying according to national licenses) and in pregnancy due to their effect on growing bone tissue (causing inhibition of skeletal growth in the foetus and discoloration of growing teeth). Cyclines form the cornerstone of oral antibiotic therapy in acne, and are discussed in detail later in this paper. Macrolides The utility of oral macrolides (mainly erythromycin) in acne is increasingly limited due to the increasing problem of microbial resistance to these agents [23, 25]. Eady et al. (1989) [26] demonstrated a clear correlation between carriage of erythromycin-resistant P. acnes and poor clinical efficacy. There is frequently cross-resistance between erythromycin and clindamycin. Macrolides are therefore reserved for cases where cyclines are not tolerated, or are contraindicated (e.g. pregnancy and breastfeeding). Clindamycin Although effective, oral clindamycin is rarely used in acne because of potentially serious adverse effects [27]. Disturbance of gastrointestinal flora by this agent can cause overgrowth of Clostridium diffıcile and result in pseudomembranous colitis. Diarrhoea is seen in 5-20% of patients using this agent [28]. Co-trimoxazole and trimethoprim The use of co-trimoxazole (trimethoprim plus sulphamethoxazole) or trimethoprim alone is limited, and these agents are not licensed for use in acne. This is because of the potential for development of serious allergic reactions to the sulphamethoxazole component of co-trimoxazole, which may be seen in up to 3% of patients. Therefore, these agents are limited to situations where there is proven resistance to other agents or as a third-line treatment [22, 29]. Quinolones Although one Japanese study has shown efficacy of levofloxacin in acne [30], oral quinolones are not used in acne because of the small amount of compelling efficacy data, the problems of adverse events (seen in 3-6% of patients), potential for antibiotic resistance (particularly the development of quinolone resistance in commensal bacteria [31]), high price, and unsuitability of these agents for adolescents (due to potential effects on articular cartilage) [32]. Side effects include agitation, headache, hallucination, gastrointestinal disturbance, arthralgia, tendinitis, and photosensitivity. Mechanisms of action of antibiotics in acne Antibacterial actions As already discussed, the density of P. acnes on the skin of acne patients does not correlate well with the degree of inflammation or the severity of acne. Similarly, the magnitude of the reduction in P. acnes counts following antibiotic therapy does not correlate well with clinical efficacy [33, 34]. However, the fact that the presence of P. acnes appears to be associated with and apparently required for the formation of inflammatory lesions; that successful antibiotic treatment of acne is associated with a reduction in the P. acnes population; and that acne associated with erythromycin- or tetracycline-resistant P. acnes does not always respond as well to treatment with those agents [26, 35, 36], may suggest an important role for antibacterial activity in the efficacy of antibiotics in acne. The different classes of agents exert their antibacterial effects in different ways. Cyclines, macrolides and clindamycin inhibit bacterial protein synthesis (by different mechanisms); trimethoprim and sulphamethoxazole interfere with bacterial folate metabolism; whilst quinolones inhibit bacterial DNA gyrase. EJD, vol. 14, n 6, November-December 2004 393

Non-antibacterial actions Non-antibacterial actions include bacterial lipase inhibition [37-39], and anti-inflammatory/immunomodulatory effects [40-43]. P. acnes secretes lipases, which convert diglycerides and triglycerides into free fatty acids. Free fatty acids in turn cause follicular hyperkeratinisation and contribute to the clinical picture of inflammatory acne. It has been shown that cyclines and macrolides inhibit bacterial lipases, independent of their antibacterial effects [37-39]. Macrolides and cyclines also interact with the immune response in a complex way, and many different effects have been observed [40-43]. These include direct, dosedependent inhibition of lymphocyte mitosis; inhibition of phagocytosis; decrease in the secretion of the proinflammatory cytokines TNF-a, IL-1 and IL-6; increase in the secretion of the anti-inflammatory cytokine IL-10; inhibition of leukotaxis; decreased activation of complement protein C3 (only noted with cyclines); modulation of a-msh (demonstrated with minocycline); and inhibition of reactive oxygen species generation [44-47]. One of the main questions today concerning the efficacy of antibiotics in acne is to determine whether the main activity of antibiotics in acne is antibacterial or antiinflammatory. The observed efficacy of antibiotics used at low (sub-mic) doses suggests that the non-antibacterial effects play an important role [34]. Antibiotic resistance in P. acnes Antibiotic resistance in P. acnes was first described in 1979, when erythromycin resistance was found in a single isolate in the USA [48]. Since then, the incidence of resistance has risen, and a recent survey, conducted throughout Europe, showed that at least 50% of acne patients are colonized by erythromycin- and clindamycin-resistant strains of P. acnes, and as many as 20% are colonized with cyclineresistant strains [25]. Resistance emerges through either selection of pre-existing resistant bacterial strains, or through de novo acquisition of a resistant phenotype. Emerging antibiotic resistance in acne has been shown to develop in response to antibiotic prescribing [25]. The duration of treatment required before resistance emerges varies greatly between patients, but the longer the duration of treatment, the more likely antibiotic-resistant P. acnes will emerge, and courses of 6 months are highly likely to result in resistance. There is a recognized correlation between the presence of antibiotic-resistant P. acnes and clinical response to treatment with erythromycin and tetracyclines (although this is less well established) [26, 35, 36]. Antibiotic resistance should therefore be considered as a possible contributory factor to, or possible cause of, therapeutic failure. High dose antibiotics will reduce sensitive strains and allow overgrowth of resistant strains in situ. However, depending on the mechanism of resistance, low dose antibiotics may also encourage overgrowth of resistant strains. In addition, low dose antibiotics may induce de novo antibiotic resistance in other commensal bacteria present (e.g. staphylococci), which often develop resistance much more quickly than resistance in P. acnes. For these reasons, the common practice of using low dose antibiotics for prolonged periods of time should not be recommended. Antibiotic-resistant strains can be transmitted between individuals, and studies have shown that 41-85.7% of untreated close contacts of acne patients under long-term antibiotic treatment harbor erythromycin-resistant strains of P. 394 acnes [25]. Furthermore, 25 out of 39 acne specialists tested were colonized by resistant strains, compared with 0 out of 27 non-dermatologist physicians [25]. Resistant strains can be reduced by using the topical antibacterial agent benzoyl peroxide at the site of application. However no single available agent will fully eradicate antibioticresistant P. acnes. Other ways of preventing the emergence of resistant strains include: 1. Do not use antibiotics where other acne treatments can be expected to bring about the same degree of benefit. 2. Use antibiotics according to clinical need. 3. Do not use antibiotics as a monotherapy [5]. 4. Stop antibiotic therapy when you and the patient agree there is no further improvement or the improvement is only slight (one UK-based study suggested that 6-8 weeks into treatment might be one appropriate time-point at which to assess response to antibiotics [36]). 5. Try to avoid continuing antibiotics beyond six months. 6. Use benzoyl peroxide either concomitantly or pulsed as an anti-resistance agent. 7. Do not switch antibiotics without adequate justification (i.e. re-use the same antibiotic for subsequent courses if patients relapse). Risk factors for developing or acquiring antibiotic-resistant P. acnes include prolonged duration of antibiotic therapy, multiple courses of antibiotics, close contact with acne patients being treated with antibiotics, and poor compliance with treatment [25]. Topical benzoyl peroxide has been shown to be active against fully sensitive and resistant strains of P. acnes. This agent therefore reduces the likelihood of antibioticresistant P. acnes emerging and reduces the number of resistant bacteria in situ [49]. Other anti-resistance agents include topical zinc acetate [50] and oral isotretinoin [51]. The development of antibiotic resistance can also be reduced by ensuring that antibiotics are not used unnecessarily, treatment duration is not excessively long, and patients comply well with treatment [25]. Combining therapy with topical retinoids will also expediate improvement while targeting the microcomedone [5]. In cases where antibiotic resistance is suspected, these drugs are often simply discontinued. Ideally, however, such cases should be managed by first swabbing and culturing to verify the presence of resistant strains, and then using non-antibiotic therapies such as topical benzoyl peroxide, topical or systemic retinoids, hormonal therapies, or systemic zinc salts. Raising antibiotic doses can also be considered. The use of oral cyclines in acne Pharmacokinetics Cyclines can be classified as first generation (tetracycline HCl and oxytetracycline) and second generation (lymecycline, doxycycline, and minocycline). The key difference between first and second generation cyclines is their pharmacokinetic profiles (Table I). First generation cyclines must be taken at least twice daily and their absorption is impaired by food and milk, whereas second generation cyclines can be taken once or twice daily with their absorption unaffected by food. These differences may adversely affect compliance with first generation cyclines (particularly among adolescents), which may lead to therapeutic failure and the development of resistance. There are few pharmacokinetic differences between individual second EJD, vol. 14, n 6, November-December 2004

Table I. Pharmacokinetics of cyclines compared Tetracycline HCl Oxytetracycline Lymecycline Minocycline Doxycycline Absorption 75% 75% 80% 90% 90% Interaction with food Yes Yes No No No Interaction with milk Yes Yes No No Yes (although 30% less than first generation cyclines) Interaction with antacids, iron Yes Yes Yes Yes Yes preparations, and mineral supplements t 1/2 7-8 hours 8-10 hours 10 hours 17-18 hours 16-22 hours T max 3-4 hours 2 hours 3 hours 2-3 hours 2-3 hours Serum protein binding 55-70% 27-35% 50% 80% 90% Clearance Renal Renal Renal Renal Biliary generation cyclines, although it is noteworthy that only doxycycline is cleared by the liver, allowing this agent to be used in patients with renal impairment. Efficacy While the efficacies of first and second generation cyclines have not been compared in large randomized clinical trials, second generation cyclines have been compared in a number of clinical trials (Table II) [52-57]. It can be seen that there is little to choose between these molecules in terms of efficacy, with reductions noted in the number of inflammatory lesions of 51-77% following 3-6 months of treatment. Efficacy and speed of response to treatment can be significantly improved by using topical retinoids concurrently (to target microcomedones, comedones and inflammation). In a study of 242 acne patients, a three month course of lymecycline (300 mg daily) was compared with lymecycline (300 mg daily) plus topical adapalene (0.1%) applied daily [58]. Lymecycline and combined therapy reduced the total number of lesions by a mean of 47.9% and 58.7%, respectively (P = 0.0033); inflammatory lesions were reduced by a mean of 45.6% vs 60.3%, respectively (P = 0.0001); and non-inflammatory lesions were reduced by a mean of 47.6% vs 56.6%, respectively (P = 0.01) [58]. Mobacken (1993) [54] and Campo et al. (2002) [55] examined the effects of treatment for up to 6 months. After 1 month of treatment with lymecycline (600 mg daily), Mobacken (1993) [54] found that facial papules and pustules had reduced in number by 47.5%, compared with 76.5% and 76% following a further 5 months of treatment with lymecycline at 300 mg or 600 mg daily, respectively. Thus it can be seen that additional efficacy can be gained by extending treatment with oral antibiotics beyond 1 month. Campo et al. (2002) [55] compared the efficacy of treatment with lymecycline (300 mg daily for 2 weeks, followed by 150 mg daily) plus topical adapalene, with minocycline (100 mg daily) plus topical adapalene for up to Table II. Effıcacies of second generation cyclines compared Study Design Lymecycline (mean % reduction in number of lesions) Minocycline (mean % reduction in number of lesions) Doxycycline (mean % reduction in number of lesions) 3 months 6 months 3 months 6 months 3 months Schollhammer and Multicentre, randomized, parallel. Alirezai 1994 [52] 1 N=77 Bossuyt et al. Multicentre, randomized, parallel, 2003 [53] 2 investigator-blind. N = 134 Mobacken et al. Open-label, parallel dose-ranging 1994 [54] 3 study. N = 221 Campo et al. Multicentre, randomized, parallel, 2002 [55] 4 investigator-blind. N = 152 Cunliffe et al. Multicentre, randomized, doubleblind, double-dummy. N = 144 1998 [56] 5 Dubertret et al. Multicentre, randomized, doubleblind study. N = 2003 [57] 6 270 72.7 68.4 62.4 58.7 64.5 76 (76.5) 67 77 62 67 50.6 52.2 62 (56.4) 1 This study compared lymecycline (300 mg daily for 2 weeks, then 150 mg daily), minocycline (100 mg daily for 2 weeks, then 50 mg daily), and doxycycline (100 mg daily for 2 weeks, then 100 mg every other day). 2 This study compared lymecycline (300 mg daily) with modified-release minocycline (100 mg daily). 3 In this study, patients received 600 mg lymecycline daily for 1 month, followed by either 600 mg lymecycline daily or 300 mg lymecycline daily. The value in brackets refers to the population randomized to the lower dose. 4 This study compared lymecycline (300 mg daily for 2 weeks then 150 mg daily) plus topical adapalene, with minocycline (100 mg daily) plus adapalene. 5 This study compared lymecycline (300 mg daily for 2 weeks, then 150 mg daily) with minocycline (100 mg daily for 2 weeks, then 100 mg every other day). 6 This study compared lymecycline (300 mg OD), lymecycline (150 mg BD), and placebo. The value in brackets refers to the population randomized to the BD dose. EJD, vol. 14, n 6, November-December 2004 395

Table III. Prices (W) of second generation cyclines per treatment course across Europe at the time of writing Country Lymecycline (300 mg daily for 12 weeks) Minocycline (100 mg daily for 12 weeks) Doxycycline (100 mg daily for 12 weeks) Belgium 64.84 108.71 71.22 France 41.80 73.31 33.60 Germany Not currently available 73.96 32.35 Italy 103.20 1 37.44 34.69 Sweden 64.66 2 Not currently available 82.42 3 UK 47.97 114.68 54.08 1 300 mg daily for 2 weeks, then 150 mg daily for 10 weeks. 2 600 mg daily for 4 weeks, then 300 mg daily for 8 weeks. 3 200 mg daily for 4 weeks, then 100 mg daily for 8 weeks. 6 months, measuring efficacy every 2 months. After 3 months, the number of acne lesions had reduced by 67% and 62% in the lymecycline and minocycline groups, respectively, compared with 77% and 67%, respectively, after 6 months of treatment. These two studies show that patients may benefit from treatment with oral antibiotics for > 1 month, but that there is little advantage in using these agents for > 3 months. In addition, the risk of developing antibiotic resistance is known to increase when treatment is continued beyond 3 months. These issues should be taken into account when considering extending treatment beyond 3 months. 396 Safety and tolerability Tetracycline HCl and oxytetracycline are generally well tolerated, but can produce gastrointestinal disturbances. Like all cyclines, first generation molecules can inhibit skeletal growth in the developing foetus and cause discoloration of growing teeth (particularly with tetracycline HCl). Cyclines are therefore contraindicated during pregnancy and in young children (under 8-12 years, depending on national licenses). Lymecycline has a good safety profile and is generally well tolerated, occasionally causing, like other cyclines, transient mild gastrointestinal disturbances, and rarely allergic reactions. Doxycycline is noted particularly for causing photosensitivity. This effect is dependent on doxycycline dose, UVA intensity, and skin type. Photosensitivity also greatly increases the potential for phototoxicity (burning effects and photo-onycholysis). Bjellerup and Ljunggren (1994) [59] compared the phototoxicity of doxycycline (200 mg daily) and lymecycline (1200 mg daily) using UVA light. In this study, lymecycline induced a slight increase in erythema compared with placebo at 75 J.cm 2. In contrast, doxycycline caused a highly significant increase in erythema compared with placebo at 50, 75, and 100 J.cm 2. Patients and doctors should therefore be conscious of the dangers of UVA solaria when using doxycycline, and this drug should be used with caution in hot climates during the summer. Non-comedogenic sun-screens that protect against both UVB and UVA should be considered when taking doxycycline (depending on dose and climate). Minocycline causes a number of rare but severe side effects, including autoimmune disorders (lupus-like syndrome, autoimmune hepatitis, arthritis, thyroiditis, polyarteritis nodosa) [60-68]; and hypersensitivity reactions (pneumonitis, eosinophilia, serum-sickness-like syndrome, DRESS [Drug Reactions with Eosinophilia and Systemic Symptoms] syndrome, arthritis, vasculitis, and hepatitis) [68]. It has been postulated that minocycline may generate a specific reactive species that could be responsable for such reactions [68]. Other effects include skin hyperpigmentation (particularly in areas exposed to the sun); single organ dysfunction (early onset, dose-related); Sweet syndrome; pseudotumour cerebri [69]; and vestibular disturbances [70]. In a review of minocycline efficacy and safety in acne, Garner et al. (2002) [64] suggested that because of a lack of proven efficacy advantages over other agents, and an uncertain safety profile, there could be no justification in continuing to use minocycline as a first line therapy in acne, and in France, the national regulatory authority (AFSSAPS) now recommends that oral minocycline should be reserved as a second choice oral antibiotic in acne. Pharmacoeconomic considerations One of the problems with attempting to include a cost component in developing pan-european recommendations on antibiotic use is the very large variation in drug prices across Europe (Table III. e.g. minocycline is almost three times more expensive in the UK than in Italy). For this reason, it is difficult to extrapolate data from one country to another. However, pharmacoeconomic analyses can be useful at a national level [71]. One recent study by Bossuyt et al. (2003) [53] compared the cost-effectiveness of minocycline (as Minocin MR ) with lymecycline (as Tetralysal 300 ) for the treatment of acne in the UK, France and Belgium. In this study, treatment success was defined as the percentage of patients in which acne was cleared, much improved, or improved. Treatment success was similar for both agents (98.4 and 91.5% for lymecycline and minocycline, respectively, over all countries). However, cost effectiveness (calculated as treatment cost divided by treatment success) was 65.89, 42.48, and 49.30 Euros per treatment success for lymecycline in Belgium, France, and UK, respectively, compared with 118.81, 80.12, and 125.33 Euros per treatment success for minocycline in Belgium, France, and UK, respectively. This study shows that lymecycline is a more cost-effective acne treatment than minocycline in Belgium, France and the UK, but particularly in the UK. Cyclines and oral contraceptives: an interaction? Acne patients are often concerned about the potential interaction between cyclines and oral contraceptives. Cyclines, and other broad-spectrum antibiotics, could theoretically alter the gut flora, reducing the bacterially-induced hydrolysis of steroid conjugates formed in the liver and gut wall, EJD, vol. 14, n 6, November-December 2004

resulting in a decreased quantity of free steroid for reabsorption, and hence reduced plasma levels of active steroid. There have been occasional anecdotal reports of oral contraceptive failure in patients taking cyclines, and two uncontrolled retrospective studies have claimed failure rates of 1.2 and 1.4 per 100 years of therapy in patients taking cyclines [72, 73]. One controlled retrospective study examined the failure rate of oral contraceptives in 356 patients who had taken oral contraceptives and antibiotics concurrently, compared with control patients [74]. This study showed no statistically significant difference between the two groups. A recent large review concluded that available scientific and pharmacokinetic evidence does not support the hypothesis that antibiotics (with the exception of rifampicin) lower the contraceptive efficacy of oral contraceptives [75]. In addition, the American College of Obstetrics and Gynaecologists states that tetracycline, doxycycline, ampicillin and metronidazole do not affect oral contraceptive steroid levels [76]. The weight of this evidence suggests that oral cyclines do not interfere with oral contraceptives and that alternative forms of contraception are not necessary whilst using these agents. Prescribing cyclines for patients with hyperseborrhoea In a study of 255 acne patients treated for 6 months, Layton et al. (1992) [77] found a correlation between sebum excretion rate and degree of improvement in acne during treatment with erythromycin, minocycline and oxytetracycline, with higher sebum excretion rates being associated with a poorer clinical response to treatment (r = 0.529). The authors suggest that this could be due to a reduced concentration of drug within the follicles as a result of drug dilution with sebum. Layton et al. (1992) [77] suggest that when sebum excretion rates are > 2.5 µg.cm 2.min 1, higher doses of cyclines may be required (e.g. 600 mg lymecycline daily, or up to 200 mg minocycline or doxycycline daily). When using higher doses, patients and physicians should be conscious of the increased risk of side effects (particularly with minocycline). Conclusions Oral antibiotics are a mainstay of treatment for moderate to severe inflammatory acne (including acne on the trunk) and, while a number of questions remain unanswered (e.g. regarding the optimal dose of antibiotic, the optimal duration of treatment and the length of maintenance therapy), there is clearly an unmet need for practical, user-friendly treatment guidelines for these drugs based on our current knowledge. The recommendations set out below provide guidance on some of the key questions regarding oral antibiotic therapy in acne (choice of oral antibiotic, doses, and duration of therapy), as well as combination therapy and maintenance therapy. By providing clear guidance that can be used in daily practice, these recommendations should improve the care of acne patients. II. Recommendations for the use of oral antibiotics Due to wide differences between published acne studies with respect to disease definitions, end-points, doses used, duration of treatments etc., as well as the many unanswered questions that require further investigation, it is not currently possible to derive definitive treatment guidelines backed up by high quality evidence. The recommendations below should therefore be viewed as guidance based on expert opinion of the available data. These recommendations are intended to provide improved patient benefit, and are based on analyses of efficacy, resistance, safety, cost effectiveness, expert opinion, and the needs and limitations of daily practice. First choice of antibiotic 1. Based on advantages in efficacy, safety, and antibiotic resistance, cyclines should be used in preference to other classes of antibiotics in the treatment of acne. 2. Based on pharmacokinetic advantages, second generation cyclines should be used in preference to first generation cyclines. 3. Based on side effect profiles 1 ; lymecycline and doxycycline should be used in preference to minocycline. The choice of agent will depend on the patient characteristics, season, UV exposure and country. Cycline doses 1. Lymecycline should be used at a dose of 300-600 mg daily. 2. Minocycline and doxycycline should be used at a dose of 100-200 mg daily 2. 3. Tetracycline HCl and oxytetracycline should be used at a dose of 1 g daily. Duration of treatment 1. According to the available literature, oral antibiotics should be used for 3 months. In clinical practice, they may be continued longer until clinical improvement is achieved. 2. If oral antibiotics are used for prolonged periods 3, they should only be continued where further clinical benefit is likely, and should always be used in combination with an agent that reduces the likelihood of propionibacterial resistance emerging (e.g. benzoyl peroxide). 3. Compliance should be checked in patients who do not respond well to therapy. Combination therapy 1. Oral antibiotics should not be used alone (as this will target only two pathophysiologic factors of acne). 2. Oral antibiotics should be combined from the start of treatment with a topical retinoid (this will target three pathophysiologic factors of acne, as well as the microcomedone [the precursor of all acne lesions]). 3. Benzoyl peroxide can be used in combination with topical retinoids and oral antibiotics, and should always be considered for patients receiving oral antibiotics for more than 3 months. 4. Oral antibiotics should not normally be combined with topical antibiotics (this may increase the risk of P. acnes resistance and provides no additive benefit). 1 Lymecycline: rare side effects; doxycycline: rare side effects, but dosedependent phototoxicity; Minocycline: rare but severe side effects. 2 200 mg should be reserved for special cases such as hyperseborrhoea, and patients and physicians should be aware of potential side effects at this dose. 3 Studies have shown most improvement occurs within the first 4 months of treatment. EJD, vol. 14, n 6, November-December 2004 397

Maintenance therapy 1. Maintenance therapy should be considered in order to limit relapse. Optimal duration and dosing remains undetermined. 2. Topical retinoids are the treatment of choice for maintenance therapy. 3. Maintenance therapy should be continued for as long as individually needed (this recommendation is not currently evidence-based). 4. Benzoyl peroxide can be added to topical retinoids if necessary (to decrease the number of antibiotic-resistant P. acnes). j Acknowledgements. These recommendations were developed at workshops supported financially by Galderma International. The authors have no financial interest in any of the products related to this work. References 1. Ad Hoc Committee report: systemic antibiotics for treatment of acne vulgaris: efficacy and safety. Arch Dermatol 1975; 111: 1630-6. 2. Drake LA. Guidelines for care of acne vulgaris. J Am Acad Dermatol 1990; 22: 676-80. 3. Taylor MB. Treatment of acne vulgaris. Guidelines for primary care physicians. Postgrad Med 1991; 89: 40-2, 45-7. 4. Oral treatment of acne. Presse Med 1999; 28: 2044-5. 5. Gollnick H, Cunliffe W, Berson D, Dreno B, Finlay A, Leyden JJ, et al. Management of acne: a report from a global alliance to improve outcomes in acne. J Am Acad Dermatol 2003; 49 (1 Suppl): S1-S38. 6. Thielitz A, Helmdach M, Ropke EM, Gollnick H. Lipid analysis of follicular casts from cyanoacrylate strips as a new method for studying therapeutic effects of antiacne agents. Br J Dermatol 2001; 145: 19-27. 7. Leeming JP, Holland KT, Cunliffe WJ. The microbial colonization of inflamed acne vulgaris lesions. Br J Dermatol 1988; 18: 205-8. 8. Jappe U, Ingham E, Henwood J, Holland KT. Propionibacterium acnes and inflammation in acne; P. acnes has T-cell mitogenic activity. Br J Dermatol 2002; 146: 202-9. 9. Marples RR, McGinley KJ. Corynebacterium acnes and other anaerobic diphtheroids from human skin. J Med Microbiol 1974; 7: 349-57. 10. De Young LM, Young JM, Ballaron SJ, Spires DA, Puhvel SM. Intradermal injection of Propionibacterium acnes: a model of inflammation relevant to acne. J Invest Dermatol 1984; 83: 394-8. 11. Kirschbaum JD, Kligman AM. The pathogenic role of Corynebacterium acnes in acne vulgaris. Arch Dermatol 1963; 88: 832-3. 12. McGinley KJ, Webster GF, Ruggieri MR, Leyden JJ. Regional variations in density of cutaneous propionibacteria: correlation of Propionibacterium acnes populations with sebaceous secretion. J Clin Microbiol 1978; 12: 672-5. 13. Cove JH, Cunliffe WJ, Holland KT. Acne vulgaris: is the bacterial population size significant? Br J Dermatol 1980; 102: 277-80. 14. Leyden JJ, McGinley KJ, Mills OH, Kligman AM. Age-related changes in the resident bacterial flora of the human face. J Invest Dermatol 1975; 65: 379-81. 15. Holland KT, Holland DB, Cunliffe WJ, Cutcliffe AG. Detection of Propionibacterium acnes polypeptides which have stimulated an immune response in acne patients but not in normal individuals. Exp Dermatol 1993; 2: 12-16. 16. Holland KT, Aldana O, Bojar RA, Cunliffe WJ, Eady EA, Holland DB, et al. Propionibacterium acnes and acne. Dermatology 1998; 196: 67-8. 17. Jappe U. Akne und Propionibakterien. Deutscher Dermatologe 2002: 583-9. 18. Ashbee HR, Muir SR, Cunliffe WJ, Ingham E. IgG subclasses specific to Staphylococcus epidermidis and Propionibacterium acnes in patients with acne vulgaris. Br J Dermatol 1997; 136: 730-3. 398 19. Eady EA, Cove JH. Is acne an infection of blocked pilosebaceous follicles? Implications for antimicrobial treatment. Am J Clin Dermatol 2000; 1: 201-9. 20. Jeremy AH, Holland DB, Roberts SG, Thomson KF, Cunliffe WJ. Inflammatory events are involved in acne lesion initiation. J Invest Dermatol 2003; 121: 20-7. 21. Leyden JJ. Current issues in antimicrobial therapy for the treatment of acne. J Eur Acad Dermatol Venereol 2001; 15 Suppl 3: 51-5. 22. Meynadier J, Alirezai M. Systemic antibiotics for acne. Dermatology 1998; 196: 135-9. 23. Plewig G, Kligman AM. Acne and rosacea. 3 rd ed. New York: Springer-Verlag. 2000. 24. Technote report 2001. www.ahrg.gov/clinic/evrptfiles.htm#acne. 25. Ross JI, Snelling AM, Carnegie E, Coates P, Cunliffe WJ, Bettoli V, et al. Antibiotic-resistant acne: lessons from Europe. Br J Dermatol 2003; 148: 467-78. 26. Eady EA, Cove JH, Holland KT, Cunliffe WJ. Erythromycin resistant propionibacteria in antibiotic treated acne patients: association with therapeutic failure. Br J Dermatol 1989; 121: 51-7. 27. Ray AJ, Donskey CJ. Clostridium difficile infection and concurrent vancomycin-resistant Enterococcus: colonization in a health care worker: case report and review of the literature. Am J Infect Control 2003; 31: 54-56. 28. Carrasco DA, Vander Straten M, Tyring SK. A review of antibiotics in dermatology. J Cut Med Surg 2002; 6: 128-50. 29. Cunliffe WJ, Aldana OL, Goulden V. Oral trimethoprim: a relatively safe and successful third-line treatment for acne vulgaris. Br J Dermatol 1999; 141: 757-8. 30. Kawada A, Aragane Y, Tezuka T. Levofloxacin is effective for inflammatory acne and achieves high levels in the lesions: an open study. Dermatology 2002; 204: 301-2. 31. Drlica K, Malik M. Fluoroquinolones: action and resistance. Curr Top Med Chem 2003; 3: 249-82. 32. Stahlmann R. Clinical toxicological aspects of fluoroquinolones. Toxicol Lett 2002; 127: 269-77. 33. Leyden JJ, McGinley KJ, Mills OH, Kligman AM. Propionibacterium levels in patients with and without acne vulgaris. J Invest Dermatol 1975; 65: 382-4. 34. Skidmore R, Kovach R, Walker C, Thomas J, Bradshaw M, Leyden J, et al. Effects of subantimicrobial-dose doxycycline in the treatment of moderate acne. Arch Dermatol 2003; 139: 459-64. 35. Leyden JJ, McGinley KJ, Cavalieri S, Webster GF, Mills OH, Kligman AM. Propionibacterium acnes resistance to antibiotics in acne patients. J Am Acad Dermatol 1983; 8: 41-5. 36. Ozolins M, Eady EA, Avery A, Cunliffe WJ, Li Wan Po A, O Neill C, et al. A randomised controlled multiple treatment comparison to provide a cost-effectiveness rationale for the selection of antimicrobial therapy in acne. Report for NHS study 94/48, January 2002. 37. Hassing GS. Inhibition of Corynebacterium acnes lipase by tetracycline. J Invest Dermatol 1971; 56: 189-92. 38. Webster GF, McGinley KJ, Leyden JJ. Inhibition of lipase production in Propionibacterium acnes by sub-minimal-inhibitory concentrations of tetracycline and erythromycin. Br J Dermatol 1981; 104: 453-7. 39. Akamatsu H, Tomita T, Horio T. Effects of roxithromycin on the production of lipase and neutrophil chemotactic factor by Propionibacterium acnes. Dermatology 2002; 204: 277-80. 40. Hamilton-Miller JM. Immunopharmacology of antibiotics: direct and indirect immunomodulation of defence mechanisms. J Chemother 2001; 13: 107-11. 41. Brinkmeier T, Frosch PJ. Oral antibiotics with antiinflammatory/ immunomodulatory effects in the treatment of various dermatoses. Hautarzt 2002; 53: 456-65. 42. Tamaoki J, Kondo M, Kohri K, Aoshiba K, Tagaya E, Nagai A. Macrolide antibiotics protect against immune complex-induced lung injury in rats: role of nitric oxide from alveolar macrophages. J Immunol 1999; 163: 2909-15. 43. Wales D, Woodhead M. The anti-inflammatory effects of macrolides. Thorax 1999; 54 (Suppl 2): S58-S62. 44. Van Vlem B, Vanholder R, De Paepe P, Vogelaers D, Ringoir S. Immunomodulating effects of antibiotics: literature review. Infection 1996; 24: 275-91. 45. Ianaro A, Ialenti A, Maffia P, Sautebin L, Rombola L, Carnuccio R, et al. Anti-inflammatory activity of macrolide antibiotics. J Pharmacol Exp Ther 2000; 292: 156-63. 46. Jain A, Sangal L, Basal E, Kaushal GP, Agarwal SK. Antiinflammatory effects of erythromycin and tetracycline on Propionibacterium acnes induced production of chemotactic factors and reactive oxygen species by human neutrophils. Dermatol Online J 2002; 8: 2. EJD, vol. 14, n 6, November-December 2004

47. Sainte-Marie I, Tenaud I, Jumbou O, Dreno B. Minocycline modulation of alpha-msh production by keratinocytes in vitro. Acta Derm Venereol 1999; 79: 265-7. 48. Crawford WW, Crawford IP, Stoughton RB, Cornell RC. Laboratory induction and clinical occurence of combined clindamycin and erythromycin resistance in corynebacterium acnes. J Invest Dermatol 1979; 72: 187-190. 49. Eady EA, Bojar RA, Jones CE. The effects of acne treatment with combination of benzoyl peroxide and erythromycin on skin carriage of erythromycin resistant propionibacteria. Br J Dermatol 1996; 134: 107-113. 50. Bojar RA, Eady EA, Jones CE, Cunliffe WJ, Holland KT. Inhibition of erythromycin-resistant propionibacteria on the skin of acne patients by topical erythromycin with and without zinc. Br J Dermatol 1994; 130: 329-336. 51. Coates P, Adams CA, Cunliffe WJ, McGinley KT, Eady EA, Leyden JJ, et al. Does oral isotretinoin prevent Propionibacterium acnes resistance? Dermatology 1997; 195 Suppl 1: 4-9; discussion 38-40. 52. Schollhammer M, Alirezai M. Étude comparative de la lymécycline (Tetralysal ), de la minocycline (Mynocine )etdeladoxycycline (Tolexine ) dans le traitement de l acné vulgaire. Réal Thérapeut Dermato-Vénérol 1994; 42: 24-6. 53. Bossuyt L, Bosschaert J, Richert B, Cromphaut P, Mitchell T, Al Abadie M, et al. Lymecycline in the treatment of acne: an efficacious, safe and cost-effective alternative to minocycline. Eur J Dermatol 2003; 13: 130-5. 54. Mobacken H. Oral tetracycline-treatment of acne. Rapid facial improvement, but back lesions are more difficult to treat. Lakartidningen 1993; 90: 2755-7. 55. Campo M, Zuluaga A, Escobar P, Motta A, Argote A, Jaramillo C, Rueda MJ, et al. A comparative study on the effectiveness of lymecicline and adapalene versus minocicline and adapalene in the treatment of acne vulgaris. Proceedings 20th World Congress of Dermatology. Paris, France, 1-5 July 2002: P0005. 56. Cunliffe WJ, Grosshans E, Belaich S, Meynadier J, Alirezai M, Thomas L. A comparison of the efficacy and safety of lymecycline and minocycline in patients with moderately severe acne vulgaris. Eur J Dermatol 1998; 8: 161-6. 57. Dubertret L, Alirezai M, Rostain G, Lahfa M, Forsea D, Niculae BD, et al. The use of lymecycline in the treatment of moderate to severe acne vulgaris: a comparison of the efficacy and safety of two dosing regimens. Eur J Dermatol 2003; 13: 44-48. 58. Cunliffe WJ, Meynadier J, Alirezai M, George SA, Coutts I, Roseeuw DI, et al. Is combined oral and topical therapy better than oral therapy alone in patients with moderate to moderately severe acne vulgaris? A comparison of the efficacy and safety of lymecycline plus adapalene gel 0.1%, versus lymecycline plus gel vehicle. J Am Acad Dermatol 2003; 49: S218-26. 59. Bjellerup M, Ljunggren B. Differences in phototoxic potency should be considered when tetracyclines are prescribed during summer-time. A study on doxycycline and lymecycline in human volunteers, using an objective method for recording erythema. Br J Dermatol 1994; 130: 356-60. 60. Bernier C, Dréno B. Minocycline. Ann Dermatol Venereol 2001; 128: 627-37. 61. Eichenfield AH. Minocycline and autoimmunity. Curr Opin Pediatr 1999; 11: 447-56. 62. Sturkenboom MC, Meier CR, Jick H, Stricker BH. Minocycline and lupus-like syndrome in acne patients. Arch Intern Med 1999; 159: 493-7. 63. Schlienger RG, Bircher AJ, Meier CR. Minocycline-induced lupus. A systematic review. Dermatology 2000; 200: 223-31. 64. Garner SE, Eady EA, Popescu C, Newton J, Li wan po A. Minocycline for acne vulgaris: efficacy and safety. Cochrane Database Syst Rev 2002; 2: CD002086. 65. Goulden V, Glass D, Cunliffe WJ. Safety of long-term high-dose minocycline in the treatment of acne. Br J Dermatol 1996; 134: 693-5. 66. Bachmeyer C, Cadranel JF. Minocycline-induced lupus and autoimmune hepatitis: family autoimmune disorders as possible risk factors. Dermatology 2002; 205: 185-6. 67. Teitelbaum JE, Perez-Atayde AR, Cohen M, Bousvaros A, Jonas MM. Minocycline-related autoimmune hepatitis: case series and literature review. Arch Pediatr Adolesc Med 1998; 152: 1132-6. 68. Knowles SR, Shapiro L, Shear NH. Serious adverse reactions induced by minocycline. Arch Dermatol 1996; 132: 934-9. 69. Chiu AM, Chuenkongkaew WL, Cornblath WT, Trobe JD, Digre KB, Dotan SA, et al. Minocycline treatment and pseudotumor cerebri syndrome. Am J Ophthalmol 1998; 126: 116-21. 70. Corona R. Minocycline in acne is still an issue. Arch Dermatol 2000; 136: 1143-5. 71. Chen SC. Cost-effectiveness analyses: a basic overview for dermatologists. J Cutan Med Surg 2001; 5: 217-22. 72. Hughes BR, Cunliffe WJ. Interactions between the oral contraceptive pill and antibiotics. Br J Dermatol 1990; 122: 717-8. 73. London BM, Lookingbill DP. Frequency of pregnancy in acne patients taking oral antibiotics and oral contraceptives. Arch Dermatol 1994; 130: 392-3. 74. Helms SE, Bredle DL, Zajic J, Jarjoura D, Brodell RT, Krishnarao I. Oral contraceptive failure rates and oral antibiotics. J Am Acad Dermatol 1997; 36: 705-10. 75. Archer JS, Archer DF. Oral contraceptive efficacy and antibiotic interaction: a myth debunked. J Am Acad Dermatol 2002; 46: 917-23. 76. ACOG: The use of hormonal contraception in women with coexisting medical conditions. ACOG Prac Bull 2000; 18: 1-13. 77. Layton AM, Hughes BR, Hull SM, Eady EA, Cunliffe WJ. Seborrhoea an indicator for poor clinical response in acne patients treated with antibiotics. Clin Exp Dermatol 1992; 17: 173-5. Appendix 1. Attendees at meetings of the European expert panel on the use of oral antibiotics in the management of acne Elena Aravijskaya, Russia; Nicole Auffret, France; Rui Bello, Portugal; Vincenzo Bettoli, Italy; Olivier Chosidow, France; Brigitte Dréno, France; Daniele Innocenzi, Italy; Francois Jordaan, South Africa; Martin Kaegi, Switzerland; Jean Marie Lachapelle, Belgium; Hakima Lakhdar, Morocco; Julien Lambert, Belgium; Andrzej Langner, Poland; Alison Layton, UK; Håkan Mobacken, Sweden; Falk Ochsendorf, Germany; Pier Paulo Pedrazetti, Switzerland; Marie Manuela Pronesti, Italy; Franco Rogioletti, Italy; Werner Sinclair, South Africa; Alexandre Jose de Souza Sittart, Brazil; Dente Valentina, Italy; Sarah Wakelin, UK. EJD, vol. 14, n 6, November-December 2004 399