Antibiotic therapy in acne is a time-honored practice whose mechanism is only recently thoroughly understood. Initially, it was assumed that the reduction in Pro- pionibacterium acnes was the sole mechanism of antibiotic efficacy in acne, but it is now understood that certain antibiotic drugs are also potent anti-inflammatory agents via nonantibiotic mechanisms. In addition, the induction of resistance has made antibiotic therapy problematic in many patients.
Benzoyl peroxide (BP) is a topical disinfectant that was originally used as a peeling agent for acne. Its mechanism of action is through lowering P. acnes populations by oxidative killing, and the drug is extremely effective as a topical agent. When applied to the skin, BP breaks down into benzoic acid and hydrogen peroxide (1,2). It assumed that the peroxide accounts for the majority of bactericidal activity, but no studies have been performed to assess the activity of benzoic acid in acne.
The major side effect of BP is irritation, which usually is easily managed with moisturizers. However, BP has been reported as a contact sensitizer in as many as
4% of patients and can reach nearly 75% when applied to leg ulcers (3), but in clini- cal acne practice actual contact allergy is rarely noted. As a heavy prescriber of the drug, I see, at most, a case every few years.
Various concentrations of BP are available, but there is no convincing data to prove that high concentrations are more effective than lower ones. P. acnes reduction is as effective by 2.5% as 10% BP (4), and one small study shows therapeutic equiv- alence between 2.5%, 5%, and 10% BP gels and a lower rate of irritation with 2.5% than the higher concentrations (5). BP washes are useful in particular for trunk acne since they can cover a large area easily, but in the past have been of fairly low potency. Newer formulations have been designed to have greater substantivity and are capable of P. acnes reductions near that of traditional gels and creams. As a single agent, BP is superior to clindamycin (6). Combination products of BP plus erythromycin or clindamycin have been developed and are more effective clinically than either product alone (6,7).
Topical and oral erythromycin and topical clindamycin have been well-established acne treatments for decades, but have become much less effective in the past
15 years or so due to the acquisition of resistance by P. acnes. Resistant bacteria are now induced quickly by macrolide therapy because most patients have a portion of
their normal skin flora that is genetically resistant, and that subgroup expands under the selective pressure of therapy (8 – 11). Resistant bacteria make for acne that resists therapy and erythromycin resistant strains are typically resistant to clin- damycin and vice versa.
Resistance can be combated by the addition of BP to topical macrolide regi- mens. It has been clearly shown that such combination products are not only more effective than monotherapy with macrolides, but also do not permit the sur- vival of resistant populations of P. acnes (6).
Other macrolides for example, azithromycin have been reported in small studies to be of value in acne (12), but no data is available on the effect of resistance on the utility of these drugs.
The tetracycline family of antibiotics are extremely useful in acne because they have multiple modes of action, functioning as antibiotics that reduce bacterial populations, and as anti-inflammatory drugs that attack acne from a second front.
Tetracyclines, especially doxycycline and minocycline are highly anti- inflammatory in many cell systems (Table 1). Neutrophil and monocyte chemotaxis is inhibited through calcium chelation, blunting the migration of cells to the follicle (13). Granuloma formation in vitro (14) and in vivo (15) is inhibited; with mino- cycline and doxycycline roughly 10-fold more active than tetracycline. In this model, macrolides and cephalosporines were inactive. Protein kinase C is inhibited (15), perhaps interfering with signal transduction. Generation of reactive oxygen species and the oxidative burst in neutrophils is decreased (16). Nitric oxide pro- duction is modulated (17). Matrix metalloprotease and collagenase activity is inhib- ited (18 – 20). In vivo, tetracyclines have been demonstrated to be highly active in treating purely inflammatory diseases including rheumatoid arthritis, bullous pem- phigoid, and sarcoidosis (21). Nonantibiotic derivatives of doxycycline have been recently developed that are highly anti-inflammatory and even antineoplastic through inhibition of angiogenesis and may be of use in acne and other inflamma- tory diseases (22 – 24).
Concentrations of tetracyclines that are below the antibiotic threshold still have anti-inflammatory activity. Low doses of doxycycline and minocycine that do not affect bacterial growth decrease the production of neutrophil chemo- attractants by P. acnes (25,26). Subminimal inhibitory doses also retain the ability to inhibit inflammation in vivo and improve diseases such as acne, rosacea, and periodontitis (27 – 29).
Antibiotic resistance is less a problem with the tetracyclines than the macrolides, but resistance in P. acnes has been documented. In general, tetracycline resistant strains are cross-resistant to doxycycline but sensitive to minocycline (30).
Choice of oral antibiotic by dermatologists for treating acne has shifted over the past few decades. Currently, the once frequently prescribed tetracycline is used relatively infrequently, with the majority of patients treated with doxycycline or minocycline. Tetracycline has multiple disadvantages, including greatest effect of diet on absorption, lower anti-inflammatory and antibacterial activity, and lower effect on acne lesions (30 – 32). There are few studies that address the relative potency of these two drugs in treating acne, and the few that do are fairly small and do not involve the more severe patients and manage to show only equivalence (33). However, there is good reason to believe that minocycline is the stronger drug. In my experience, there have been many patients with significant acne who fail to respond to doxycycline yet have an excellent response when switched to minocy- cline. The reason for this may lie in the greater lipophilicity of minocycline and the greater activity in a lipid milieu. This is reflected in a 10-fold greater reduction of P. acnes by minocycline when compared with doxycycline (32).
The side effect profile (Table 2) of doxycycline and minocycline also differs, most notably in the incidence of photosensitivity with doxycycline and the occurrence of hypersensitivity reactions with minocycline. Photosensitivity is very common at higher doses of doxycycline. The minocycline hypersensitivity reactions are uncom- mon and include urticaria, serum sickness-like reactions, and what has been termed a lupus-like reaction that in reality is probably not an activation of systemic lupus erythematosus but a generalized drug-induced reaction that resembles lupus (34).