AmJClinDermatol DOI10.1007/s40257-017-0267-z EVIDENCE-BASED REVIEW Oral Antibacterial Therapy for Acne Vulgaris: An Evidence- Based Review Amanda Bienenfeld1 • Arielle R. Nagler2 • Seth J. Orlow2 (cid:2)SpringerInternationalPublishingSwitzerland2017 Abstract Conclusion Thisarticleprovidesasystematicevaluationof Background To some degree, acne vulgaris affects nearly thescientificevidenceoftheefficacyoforalantibioticsfor every individual worldwide. Oral antibiotic therapy is acne.Duetoheterogeneityinthedesignofthetrials,thereis routinelyprescribedforthetreatmentofmoderatetosevere insufficientevidencetosupportonetype,dose,ordurationof inflammatory acne; however, long-term use of oral oral antibiotic over another in terms of efficacy; however, antibiotics for acne may have unintended consequences. due to increasing resistance to antibiotics, dermatologists Objective Theaimofthisstudywastoprovideasystematic shouldheedconsensusguidelinesfortheirappropriateuse. evaluation of the scientific evidence on the efficacy and appropriateuseoforalantibioticsinthetreatmentofacne. Methods A systematic search of MEDLINE was con- Key Points ducted to identify randomized controlled clinical trials, systematic reviews, and meta-analyses evaluating the effi- The use of oral antibiotics is reserved for patients cacy of oral antibiotics for acne. Overall, 41 articles that with moderate to severe inflammatory acne. examined oral antibiotics compared with placebo, another oral therapy, topical therapy, alternate dose, or duration Tetracyclines areconsideredfirst-linetherapy,while were included in this study. macrolides and trimethoprim/sulfamethoxazole are Results Tetracyclines, macrolides, and trimethoprim/sul- acceptable alternative agents. famethoxazole are effective and safe in the treatment of Itisrecommendedthatoral antibioticsbeprescribed moderate to severe inflammatory acne. Superior efficacy of with concurrent topical therapy for improved one type or class of antibiotic could not be determined, efficacy and to combat antibiotic resistance. therefore the choice of antibiotic is generally based on the side-effectprofile.Althoughdifferentdosingregimenshave Oral antibiotics used in the treatment of acne may beenstudied,thereisalackofstandardizedcomparatortrials have unintended effects on non-target bacteria, and to determine optimal dosing and duration of each oral the clinical implications of this warrant further antibiotic used in acne. The combination of oral antibiotics exploration. withatopicaltherapyissuperiortooralantibioticsalone. & SethJ.Orlow 1 Introduction [email protected] Acne vulgaris is the eighth most prevalent disease world- 1 NewYorkUniversitySchoolofMedicine,NewYork,NY, wide. Almosteveryindividualbetween15and 17yearsof USA age is affected [1, 2]. Acnemay be associated with lasting 2 TheRonaldO.PerelmanDepartmentofDermatology,New side effects, including facial scars, feelings of low self- YorkUniversitySchoolofMedicine,240East38thStreet, 11thFloor,NewYork,NY10016,USA esteem, and depression [3]. A.Bienenfeldetal. Several independent, interacting factors contribute to TheidentifiedstudieswereevaluatedusingtheStrengthof acne pathogenesis, including increased sebum production, RecommendationTaxonomy(SORT)3-pointgradingscale increased keratinization of the follicular epithelium, [8]. Finally, the most up-to-date consensus guidelines for inflammation, and overgrowth of normal skin microflora, acne treatment from expert committees were reviewed for particularlyGram-positivePropionibacteriumacnes[4–6]. their recommendations. These factorsareinter-related andsynergistic; therefore, it is difficult to tease out the primary or initiating process. Sebum, the nutrient source of P. acnes, is necessary for 3 Results acnedevelopment[5].Inturn,P.acnes,relyingonsebum, forms biofilms in sebaceous sites, which promotes follic- 3.1 Guidelines ular hyperkeratinization and stimulates the host inflam- matoryresponse[6,7].Theresultinginflammationleadsto Several consensus acne treatment guidelines have been follicular wall rupture, and the downstream inflammatory published; summaries of eight consensus guidelines from response stimulates further tissue destruction and scar separate expert committees are included in Table 1. Five formation [6]. In addition, increased follicular keratiniza- guidelines report that oral antibiotics are appropriate in tionpromotesevenmoreproliferationofP.acnes,creating moderate–severe inflammatory acne [3, 9–12]; one rec- a complex cycle [4]. ommends use in widespread mild–severe inflammatory Acne therapies include topical and systemic treatments, acne or moderate nodular/conglobate acne[13]; and one targeting different aspects of acne pathogenesis. Oral recommendstheuseoforalantibioticsonlyinnodularacne antibiotics are routinely prescribed for the treatment of [14]. moderate–severeinflammatoryacneandarebelievedtobe Alloftheguidelinesrecommendthecombinationoforal successful because they target both the bacteria and asso- antibiotics with a topical retinoid, and possibly benzoyl ciated inflammation. The antibiotics used in the treatment peroxide, and also recommend combination therapy ofacneincludetetracyclines,macrolides,clindamycin,and becauseittargetsmultiplepathophysiological mechanisms trimethoprim/sulfamethoxazole. However, despite their [15], achieves faster, more complete resolution of come- efficacy, prolonged use of oral antibiotics may potentially donal and inflammatory lesions [13], and combats the alter the skin and gut microbiome and contribute to development of antibiotic resistance [11, 15, 16]. Tetra- antibioticresistance,agrowingglobalconcern.Theaimof cyclines are considered first-line in oral antibacterial ther- this review was to critically evaluate the published litera- apy,butiftetracyclinesarecontraindicatedornottolerated, tureontheroleoforalantibioticsinthetreatmentofacne, macrolides may be used as second-line therapy as well as the specific classes, doses, and duration of [3, 11, 14, 16]. Additionally, several statements recom- treatment. mend limiting the use of erythromycin due to increased resistance[3,10].Oneguidelineaddressedthetreatmentof acne in children and recommends that in children under 8 2 Methods years of age, for whom tetracyclines are contraindicated, macrolides and trimethoprim/sulfamethoxazole may be AMEDLINEsearchwasconductedtoidentifyrandomized used with careful surveillance [10]. controlled trials, systematic reviews, and meta-analyses that evaluated oral antibiotic therapy for acne. Only peer- 3.2 Antibiotic Efficacy reviewed articles published in English prior to 30 September2016were considered forinclusion. Thesearch Oral antibiotic efficacy in acne is likely to be multifacto- terms ‘acne (vulgaris)’, ‘oral antibiotics’, ‘tetracycline’, rial. Antibiotics may target P. acnes and the underlying ‘doxycycline’, ‘minocycline’, ‘macrolides’, ‘ery- inflammation contributing to acne development. Tetracy- thromycin’, ‘azithromycin’, ‘trimethoprim’, ‘trimetho- clines, macrolides, and clindamycin inhibit bacterial pro- prim/sulfamethoxazole’, and ‘clinical trial’ were used, and tein synthesis, while trimethoprim/sulfamethoxazole 543 articles were identified. After titles and abstracts were interferes with bacterial folate metabolism. Additionally, screened for relevance, a total of 50 full-text publications tetracyclines and, to a lesser extent, macrolides, inhibit were reviewed for content. Only articles evaluating oral neutrophil chemotaxis, cytokine production, and macro- antibiotics compared with another oral therapy, topical phage function [15, 17]. therapy, placebo, or alternate dose or duration were con- The impact of antibiotics on P. acnes in the skin is sidered. Twenty-eight full-text articles were identified that controversial. In vitro testing has shown P. acnes are fulfilled these criteria, with a further cross-reference of sensitive to tetracycline, doxycycline, minocycline, oxyte- bibliographiesidentifyinganadditional13articles(Fig. 1). tracycline,erythromycin,andpenicillin,aswellastoother OralAntibacterialTherapyforAcneVulgaris Fig.1 Identificationand selectionofstudies n o Records iden(cid:2)fied through Addi(cid:2)onal records iden(cid:2)fied it a database searching through other sources c ifit (n = 543) (n = 13) n e d I Titles and abstracts screened for relevance ng (n = 556) ni e e r c S Records excluded (not relevant) (n = 506) y Full-text ar(cid:2)cles assessed bilit for eligibility gi (n = 50) Eli Full-text ar(cid:2)cles excluded (trials not randomized or controlled) (n = 9) d Studies included in e qualita(cid:2)ve synthesis d u (n =41) cl n I antibacterials active against Gram-positive organisms reducetheproductionofproinflammatorycytokines,suchas [18,19].However,invivo,dataareinconsistent.Onestudy interleukin-8 and tumor necrosis factor-a [25, 26]. Tetra- foundareductioninP.acnesafter3 weeksoftetracycline, cyclineanderythromycinsignificantlyinhibitthereleaseof minocycline, and clindamycin treatment, but not after reactiveoxygenspeciesfromhumanneutrophils[27].These treatment with penicillin or ampicillin [20]. Meanwhile, anti-inflammatory effects have been utilized in acne treat- other in vivo studies have found no significant changes in ment,aswellasintreatingotherinflammatorydiseasessuch skin bacterial flora in response to tetracycline or doxycy- asperiodontitisandrheumatoidarthritis[28,29]. cline treatment despite clinical improvement [20–22]. Improvements in acne with systemic antibiotics may be 3.2.1 Efficacy of Individual Antibiotics seen without a simultaneous reduction inP. acnes. In fact, while the concentration of P. acnes correlates with sebum Only a few randomized controlled trials examined the production, P. acnes concentration does not correlate with efficacy of antibiotics in the treatment of acne compared the amount of inflammation or acne severity. Instead, the with placebo. We identified nine randomized, placebo- inflammatory response in reaction to P. acnes has been controlled trials that investigated the efficacy of tetracy- correlated with acne severity [16]. Thus, it is believed that cline [30], doxycycline [22, 31], minocycline [32, 33], antibioticsthatarebothantibacterialandanti-inflammatory roxithromycin [34], trimethoprim/sulfamethoxazole are most effective in the treatment of acne. [35,36],andclindamycin[37](Table 2).Otherantibiotics, Variousanti-inflammatoryactionscontributetotheeffi- suchaslymecycline,erythromycin,andazithromycinhave cacy of antibiotics in the treatment of acne. Tetracycline, demonstrated efficacy in acne but have not been evaluated erythromycin, and clindamycin inhibit neutrophil chemo- in randomized, placebo-controlled trials and are therefore taxis[23,24],andtetracycline,doxycycline,andmacrolides not included in this review. A.Bienenfeldetal. Duration Simultaneoususewithtopicalretinoidorretinoid/BPfor3–4months.Ifalternativetherapiesareinappropriate,alongerdurationisacceptablewithconsistentfollow-upandre-evaluations Clinicaleffecttakesapproximately4–8weeks.Theidealdurationis3months;minimumduration6–8weeks,maximumduration12–18weeks.Antibioticsshouldbetaperedordiscontinuedwheninflammatorylesionsadequatelyresolve.Ifitisnotpossibletodiscontinueantibiotics,switchtoacombinationproductwithBPat2months Dosage MinocyclineERat1mg/kgissafest,butnoevidencethatitismoreefficacious.Doxycyclineeffectivebetweenthe1.7and2.4mg/kgrange.SDDof20mgbidor40mgdaily Tetracycline:500mgbid;doxycycline:50–100mgbid;lymecycline:150–300mgdaily;minocycline:50–100mgbid;erythromycin:500mgbid;TMP/SMX:800SMX/160TMP;trimethoprim:300mgbid nevulgaris Treatmentrecommendation Evidencesupportstheuseoftetracyclines,macrolides,TMP/SMX,trimethoprim,amoxicillin,andcephalexinforacne.Thetetracyclineclassisrecommendedfirst-lineunlesscontraindicated.Restricttheuseoferythromycinbecauseofitsincreasedriskofbacterialresistance.Oralantibioticmonotherapyisdiscouraged;instead,concomitantuseoforalantibioticswithtopicalretinoidsorretinoid/BPisrecommended.Maintenancetherapywithtopicaltherapiesafteroralantibioticsisrecommended Atopicalretinoidandantimicrobialtherapyshouldbetriedbeforeinitiatingoralantibiotics.Tetracyclines,macrolides,andtrimethoprimareeffectiveforacne.Cephalosporins,fluoroquinolones,aminoglycosides,chloramphenicol,sulfonamides,andgyraseinhibitorsshouldNOTroutinelybeusedforacnebecauseoflackofefficacyandsafetyconsiderations.Oralantibioticsshouldnotbeusedasmonotherapy;combinationtherapyoforalantibioticsandtopicalretinoidsisrecommendedtoclearinflammatorylesionsandcomedonesfasterthanantibiotictherapyalone.Therefore,topicalretinoidsshouldbestartedattheinitiationofantibiotictherapy.Useatopicalretinoidformaintenancetherapyforcontinuedsuccess.Ifretreatmentisnecessary,thesameantibioticshouldbeusedifitwereeffective,otherwiseuseanalternativeantibiotic c a seinthetreatmentof Typeofacne(fororalantibiotics) Moderatetosevereinflammatoryacneresistanttotopicaltherapy Moderatetosevereinflammatoryacne u c oti ve antibi yof Impro009 oral dem etoe(23) usguidelinesfor ommittee heAmericanAcaDermatology heGlobalAlliancOutcomesinAcnupdatesfrom200 ns C T T e s n Table1Co Study(year) Zaengleinetal.(2016)[3] Thiboutotetal.(2009)[9][updatedfromGollnicketal.(2003)[15]] OralAntibacterialTherapyforAcneVulgaris d, Duration 3months’durationisrecommendebutlongerdurationmaybeuseduntilclinicalimprovementisachieved.Maximumduration:6months Discontinueortaperwithin1–2months,oncenewinflammatorylesionshavestoppedemerging osage ymecycline:300–600mgdaily;minocycline/doxycycline:100–200mgdaily;tetracycline:1mgdaily oxycycline:50–100mgdailyorbid;tetracycline:500mgbid;minocycline:50–100mgdailyorbid;minocyclineER:1mg/kgdaily;erythromycin:250–500mgdailyorbid;TMP/SMX:160–800mgbid D L D Treatmentrecommendation Tetracyclines,macrolides,clindamycin,trimethoprim,cotrimoxazole,andquinolonesarealleffectiveinacne.Tetracyclinesshouldbeusedasfirst-lineoverotherclassesofantibiotics;second-generationtetracyclineshouldbeusedbeforefirst-generationtetracycline;lymecyclineanddoxycyclineshouldbeusedbeforeminocyclinebasedonsafetyprofile.Oralantibioticmonotherapyisdiscouraged;atopicalretinoidshouldalwaysbeprescribedwithoralantibiotics.Whenoralantibioticsareusedforprolongedperiodsoftime(over3months),BPshouldbeaddedtotheregimen.Oralantibioticsshouldnotbeadministeredwithtopicalantibiotics.Maintenancetherapyshouldbewithatopicalretinoid±BP.Ifretreatmentisnecessary,usethesameantibioticasthepriorcourse Inchildrenolderthan8yearsofage,tetracycline,doxycycline,andminocyclinearecommonlyusedinthetreatmentofacne.Inchildrenunder8yearsofage,erythromycin,azithromycin,andTMP/SMXmaybeusedwithcaution.Erythromycinuseislimitedsecondarytoincreasedresistance.Avoidantibioticmonotherapy;combinewithtopicalretinoid±BP.Ifretreatmentisnecessary,usethesameantibioticasthepriorcourse ororal vereacne Typeofacne(fantibiotics) Notaddressed Moderatetoseinflammatory a d ommittee uropeanExpertGrouponOralAntibioticsinAcne heAmericanAcneandRosaceSocietyconvenedapanelofpediatricdermatologists,pediatricians,anddermatologists(EndorsedbyTheAmericanAcademyofPediatrics) ue C E T n Table1conti Study(year) Drenoetal.(2004)[16] Eichenfieldetal.(2013)[10] A.Bienenfeldetal. e bl si s o p n e h w s s k k e e e e w w d d 2 6 se se 1 Duration Minimum: Notaddres Notaddres Limitedto y Dosage Doxycycline:100–200mg/day;tetracycline:500–1000mg/day;minocycline100–200mg/day;lymecycline:300–600mg/day;erythromycin:500–1000mg/da Notaddressed Notaddressed Notaddressed ethoprim/sulfamethoxazole m Treatmentrecommendation Moderateacne:oralantibiotics,suchasdoxycycline,minocycline,lymecycline,orerythromycincombinedwithatopicalretinoidandtopicalBP.Severeacne:6–8weeksofrecommendedregimenformoderateacne;ifnoresponseisseen,thenoralisotretinoinisrecommended.Oralantibioticsshouldneverbeadministeredasmonotherapyorwithtopicalantibiotics Oralantibioticsshouldbereservedformorewidespreaddisease.Systemictreatmentshouldalwaysbecombinedwithatopicalretinoid±BP.Doxycyclineandlymecyclineshouldbechosenbeforeminocyclineortetracycline Oralantibiotictreatmentshouldbeinitiatedwithfixedcombinationadapalene0.1%/BP2.5%;ifaresponseisseen,continuetreatmentforatleast12weeks.Whencleared/almostcleared,usetopicalretinoidformaintenance.Ifnoresponse,checkcomplianceandconsiderchangingtheoralagent Thepreferredoralantibioticsaretetracyclinesandmacrolides.Oralantibioticsshouldalwaysbeusedwithatopicalretinoidtherapy.BPshouldbeaddedtotheregimenifthedurationoforalantibioticsexceeds8–12weeks.Second-linetreatmentfornon-respondersorpatientswhocannottoleratetherapyisoralisotretinoin TMP/SMXbialdosedoxycycline,tri o Typeofacne(fororalantibiotics) Moderatetosevereacne Widespreadmildtomoderatepapulopustularacne,severepapulopustular,ormoderatenodular/conglobateacne Moderatetoseverepredominantlypapulopustularacne Nodularacne SDDrelease,submicr d e ued Committee South-EastAsiaStudyAlliance ExpertsofficiallynominatedbytheEuropeanDermatologyForumortheEuropeanAcademyofDermatologyandVenereology 17acneexpertsreviewingtheguidelinesfromtheGlobalAlliance,EuropeanDermatologyForum,AmericanAcademyofDermatology,andgroupsfromAsia AsianWorkingGroup BPERbenzoylperoxide,extend Table1contin Study(year) Gohetal.(2015)[11] Nastetal.(2012)[13] Gollnicketal.(2016)[12] Abad-Casintahanetal.(2011)[14] bidtwicedaily, OralAntibacterialTherapyforAcneVulgaris aevidence nestudies;consensus of no= vel or-III Le I I I II I II I II II all-dy; als,stu bo bo triol place place olledcontr vs. vs. ontrase- e e cc Treatment Tetracyclinevs.placebo Doxycyclinevs.placebo Doxycyclinevs.placebo Minocyclinevs.placebo Minocyclinevs.placebo Roxithromycinvs.placebo Trimethoprim/sulfamethoxazol Trimethoprim/sulfamethoxazol Clindamycinvs.placebo qualityindividualrandomizedyclinicaltrial,cohortstudy, antibioticscomparedwithplacebo DesignTypeofacne Randomized,double-blind,placebo-controlledDiagnosisofacnevulgarisn=clinicaltrial.Age12–29years,51 Randomized,double-blind,placebo-controlledInflammatoryacnevulgarisn=crossovertrial.Age14–27years,62 Randomized,multicenter,double-blind,placebo-Moderateacnevulgarisn=controlledtrial.Age18yearsandolder,51 Mild–moderateacnevulgarisRandomized,multicenter,double-blind,placebo-controlledcrossovertrial.Age14–34years,n=43 Randomized,multicenter,double-blind,placebo-Moderate–severeacnevulgarisn=controlledtrial.Age12–30years,451 Randomized,double-blind,placebo-controlledInflammatoryacnevulgarisn=crossovertrial.Age14–30years,46 Randomized,double-blind,placebo-controlledDiagnosisofacnevulgarisn=crossovertrial.Age7–35years,43 Randomized,double-blind,placebo-controlledDiagnosisofacnevulgarisn=crossovertrial.Age11–28years,33 Randomized,double-blind,placebo-controlledtrial.Moderate–severeacnevulgarisn=Agerangenotprovided,83 systematicreview/meta-analysisofrandomizedcontrolledtrialswithconsistentfindings,high-eta-analysisoflower-qualityclinicaltrialsorstudieswithinconsistentfindings,lower-qualitfrombenchresearch,usualpractice,opinion,disease-orientedevidence,orcaseseries Table2Efficacyoforal Study(year) Laneetal.(1969)[30] Plewigetal.(1970)[31] Skidmoreetal.(2003)[22] HersleandGisslen(1976)[32] Fleischeretal.(2006)[33] Ferahbasetal.(2004)[34] Hersle(1972)[35] MacDonaldetal.(1972)[36] ChristianandKrueger(1975)[37] aLevelofevidence:I=II=systematicreview/mguidelines,extrapolations A.Bienenfeldetal. 3.2.1.1 Tetracycline The efficacy of tetracycline was success, graded by a dichotomized Evaluator’s Global assessedinarandomized,double-blind,placebo-controlled SeverityAssessmentscale,comparedwith8.7%ofpatients trial in 51 patients with acne vulgaris [30]. Improvement in the placebo group (pooled analysis: p\0.001). was evaluated based on clinical and photographic assess- ment. After 6 weeks of tetracycline therapy, 54% of 3.2.1.4 Roxithromycin Roxithromycin, a semi-synthetic patients showed improvement compared with 7% of macrolide, was evaluated in a crossover study [34]. Sub- patientsintheplacebogroup(p\0.01).Atthecompletion jects were treated with roxithromycin 150 mg twice daily of the 3-month trial, clinical improvements were noted in or placebo for 4 weeks and then underwent a 2-week 75%ofthetetracyclinegroupandonly33%oftheplacebo washout period without any treatment, followed by the group (p\0.01). crossover of treatment for 4 weeks. Subjects treated with roxithromycininthefirstphaseofthestudyhadsignificant 3.2.1.2 Doxycycline A placebo-controlled, crossover reductions in median acne scores after 4 weeks compared trial was performed in 62 patients with inflammatory acne with baseline (p\0.001), whereas patients treated with to evaluate the efficacy of doxycycline 100 mg daily placebodidnot(p[0.05).Attheendofthesecondphase, compared with placebo [31]. Subjects were treated with and after both groups had been treated with a 4-week doxycycline or placebo for 4 weeks during the first phase, course of roxithromycin, all subjects had significantly followed by a 4-week rest period and a second 4-week decreased median acne scores compared with baseline phase in which the medications were switched. In each (p\0.001), and no difference was seen between groups phase, patients treated with doxycycline had a significant (p[0.05) [34]. reduction in inflammatory lesion count compared with the initiation of that phase (average lesion count decreased by 3.2.1.5 Trimethoprim/Sulfamethoxazole Trimetho- 26% in the first phase [p\0.001] and 24% in the second prim/sulfamethoxazole is typically used when other phase [p\0.05]); however, patients treated with placebo antibiotics are contraindicated or not tolerated by the had no significant reductions in inflammatory lesions in patient.Someconsideritonlyasathird-lineagent[10,16]. either phase of the study [31]. The study did not statisti- Two studies examined the efficacy of trimethoprim/sul- callycomparetheresultsofthedoxycyclinegroupwiththe famethoxazole in randomized, double-blind, placebo-con- placebo group. Another study investigating doxycycline trolled, crossover trials [35, 36]. Hersle [35] demonstrated 20 mg taken twice daily showed a reduction in the total a significant reduction in acne scores after 5 weeks of number of lesions by 52%, compared with 18% with pla- therapy with trimethoprim/sulfamethoxazole compared cebo after 6 months of treatment (p\0.01) [22]. with baseline (p\0.001) and with the placebo group (p\0.001). In the second crossover trial, the authors did 3.2.1.3 Minocycline Despite the use of many different not present the raw data but reported no significant dif- oral antibiotics for acne, extended-release minocycline ference between treatment and placebo groups in the administered at 1 mg/kg daily is the only US FDA-ap- reduction in number of inflammatory and non-inflamma- proved antibiotic specifically for moderate–severe inflam- tory lesions at the endof each phase (p[0.05). However, matoryacne[10].Acrossoverstudyevaluatingtheefficacy when the authors analyzed inflammatory and non-inflam- of minocycline showed a significant reduction of acne matory lesions separately, they concluded that there was a lesion score in the first phase of the minocycline group significantreductioninthenumberofinflammatorylesions (p\0.05), while the placebo-treated group acne score did compared with placebo, but there was no significant notdecreasesignificantly(nopvalueprovided)[32].There reduction in the number of non-inflammatory lesions was no comparison between the minocycline and placebo compared with placebo [36]. groups, and there was no washout period between the two phases of the crossover study to evaluate the results in the 3.2.1.6 Clindamycin When compared with placebo, oral second phase of the trial. clindamycin is effective in the treatment ofacne [37].In a Two 12-week, phase III studies examined safety and double-blind, placebo-controlled trial, clinical improve- efficacy in 615 subjects treated with extended-release ment in acne was defined as at least 50% reduction in the minocycline 1 mg/kg daily compared with 309 patients numberofpapulesandpustulescomparedwiththenumber treated with placebo [33]. In both studies, patients treated at initiation. After 13 weeks of therapy, 86% of clin- with extended-release minocycline had a significant mean damycin-treated patients and 38% of placebo-treated percentage reduction in inflammatory lesion count (43.1 patients had clinical improvements (p\0.005). The clin- and 45.8%) compared with placebo (31.7 and 30.8%) damycinandplacebogroupshada50and21%reductionin [p\0.001inbothstudies].Furthermore,16.6%ofpatients the number of comedones (p\0.05) and a 78 and 26% in the extended-release minocycline group had treatment reductionininflammatorylesions(p\0.01), respectively; OralAntibacterialTherapyforAcneVulgaris however, the study did not report differences between the Studies have also compared different antibiotic classes clindamycinandplacebogroups.Additionally,participants in the treatment of acne. Three studies found similar were allowed to continue any therapies that they were efficacy of doxycycline and azithromycin [43–45]. Azi- already receiving at the start of the trial, including thromycin is generally prescribed in pulsed doses, but antibacterial soaps and cleansers [16]. with heterogeneity in scheduling and dosing of pulses. A randomized, double-blind study of patients with moderate 3.2.2 Comparison by Class of Oral Antibiotics acne confirmed the non-inferiority of pulsed azithromycin (500 mg daily for 3 days in week 1, then 500 mg weekly Despitetheroutineuseofantibioticsforacne,thereislittle for 9 weeks) compared with daily doxycycline, and evidence that supports the use of one class of antibiotic reported no difference in the incidence of adverse events over another. Additionally, there is considerable hetero- [45]. Studies that compared daily doxycycline with pulsed geneityintheliteraturewithregardtodesignandoutcome azithromycin, by Kus et al. [44] (500 mg/day for 3 days/ measures in studies examining efficacy of various antibi- week in month 1, then 2 days/week in month 2, and then otics [38], making it difficult to establish evidence-based 1 day/week in month 3) and Babaeinejad et al. [43] guidelines.Consequently,thechoiceofantibioticisusually (500 mg daily for 4 consecutive days per month for 3 determined by the side-effect profile. months), also showed comparable efficacy between the Different tetracycline antibiotics have been studied in two antibiotics for moderate inflammatory acne. However, the treatment of acne, but there is little evidence of supe- Babaeinejad et al. reported a significantly greater riority of one type over another (Table 3). A large, ran- improvement in patients older than 18 years of age domized, observer-blinded trial in patients with mild– receiving daily doxycycline compared with pulsed azi- moderate acne found comparable efficacy in reducing the thromycin [43]. number of inflammatory lesions compared with baseline A randomized, multicenter, open-label, controlled trial betweenminocyclineandoxytetracyclineat18weeks[39]. compared minocycline 100 mg daily with roxithromycin Two studies examined the efficacy of lymecycline com- 150 mg daily and faropenem 200 mg three times daily in pared with minocycline [40, 41]. Grosshans et al. [40] patients with moderate to severe inflammatory acne. A performed a randomized, multicenter, double-blind, dou- significant decrease in lesion count was noted, compared ble-dummytrial.Attheendof12weeks,nodifferencewas with baseline, in all three treatment groups, but no signif- noted between the lymecycline- and minocycline- treated icantdifferencebetweengroupsatweek4or4 weekspost- groups in the percentage reduction of inflammatory lesion treatment.Additionally,therewasnosignificantdifference (50.6 and 52.2%, respectively) or total lesion count (44.5 in adverse events between groups [46]. However, the and 42.9%, respectively). Bossuyt et al. [41] compared inabilitytodetectsignificantdifferencesbetweenantibiotic extended-release minocycline 100 mg daily with lymecy- classes is not surprising given the small sample size cline 300 mg daily for 12 weeks. At the end of the trial, (n = 150) and short treatment duration [42]. again, no significant difference was noted between the Oral erythromycin has been effective in the treatment lymecycline and minocycline groups for mean percentage of acne. A randomized, double-blind, clinical trial that reductions in inflammatory lesion count (63 and 65%, studied the efficacy of erythromycin and tetracycline respectively) and total lesion count (58 and 56%, respec- showed that, at week 12, erythromycin significantly tively), or median percentage reduction in non-inflamma- reduced inflammatory lesion count by 67% and non-in- tory lesion count (54 and 49%, respectively) [41]. flammatory lesion count by 22%, and tetracycline sig- As a result, a systematic review of clinical trials on the nificantly reduced inflammatory lesion count by 64% and efficacy of tetracycline, minocycline, doxycycline, and non-inflammatory lesion count by 34%. No differences in lymecycline for the treatment of mild–moderate inflam- efficacy were seen between the erythromycin and tetra- matoryacnefoundnosignificantdifferenceinreductionof cycline groups [47]. Despite the equivalent efficacies, inflammatory or non-inflammatory lesion count among erythromycin is not recommended for acne due to various tetracyclines [38]. Furthermore, there is no evi- increased microbial resistance [15]. dence that second-generation tetracyclines, such as Finally, trimethoprim has been studied in the treatment minocycline, are more effective than first-generation of acne. In a small, randomized, single-blinded trial, the tetracyclines, despite superior pharmacokinetics and the use of trimethoprim 100 mg (three times per day for higher cost of second-generation tetracyclines [38, 42]. 4 weeks, then twice daily for 4 weeks) was as effective in A Cochrane review concluded that the more expensive, reducing lesion count as oxytetracycline 250 mg (three extended-release minocycline formulation is neither more times per day for 4 weeks, then twice daily for 4 weeks). effective nor safer than other oral antibiotics [42]. However,twopatientsinthetrimethoprimtreatmentgroup A.Bienenfeldetal. oface n Levelevide I I I I II I II II II I II and cline ne eroxide, minocy nem c p g e Treatment Oraltetracyclinesinthetreatmentofa Minocycline,oxytetracycline,benzoylerythromycin/benzoylperoxide Randomizedcontrolledtrialscomparinatanydosewithothertherapies Minocyclineandlymecycline Minocyclineandlymecycline Doxycyclineandazithromycin Doxycyclineandazithromycin Doxycyclineandazithromycin Minocycline,roxithromycin,andfarop Tetracyclineanderythromycin Oxytetracyclineandtrimethoprim o n 5 d orclasses Typeofacne Mild–moderateinflammatoryacnevulgaris Mild–moderateacnevulgaris Inflammatoryacnevulgaris Moderate–severeacne Acneseveritygradebetween1and(Leedsscale) Moderateacnevulgaris Atleast10inflammatorylesions,anmorethanthreenodules Moderateacnevulgaris Moderate–severeinflammatoryacnevulgaris Moderate–severeacnevulgaris Inflammatoryacnevulgaris oforalantibioticscomparedwithotherantibiotictypes Design Systematicreview Randomized,observer-blinded,clinicaltrial.Agen=12–39years,649 Systematicreview Randomized,multicenter,double-dummycontrolledn=trial.Age12–32years,144 Randomized,investigator-blindedcontrolledtrial.Agen=12–30years,134 Randomized,double-blindclinicaltrial.Agerangenotn=provided,100 Randomized,investigator-blindedclinicaltrial.Agen=18–30years,51 Randomized,double-blind,non-inferioritystudy.Agen=14yearsandolder,240 Randomized,multicenter,open-labelcontrolledtrial.n=Age16yearsandolder,150 Randomized,multicenter,double-blindclinicaltrial.n=Age14–30years,200 Randomized,single-blindedclinicaltrial.Age14yearsn=andolder,53 dencecategories,seeTable2 Table3Efficacy Study(year) Simonartetal.(2008)[38] Ozolinsetal.(2005)[39] Garneretal.(2012)[42] Grosshansetal.(1998)[40] Bossuytetal.(2003)[41] Babaeinejadetal.(2011)[43] Kusetal.(2005)[44] Maleszkaetal.(2011)[45] HayashiandKawashima(2011)[46] Gammonetal.(1986)[47] Gibsonetal.(1982)[48] aForlevelofevi