Although hidradenitis suppurativa (HS) is not primarily an infectious disease, antibiotics are widely used to treat HS. Recent microbiological data show that HS suppurating lesions are associated with a polymorphous anaerobic flora, including actinomycetes and milleri group streptococci, and can therefore be considered as polymicrobial soft tissue and skin infections. Analysis of the literature provides little information on the efficacy of antibiotics in HS but suggests a beneficial effect of certain antimicrobial treatments, depending on the clinical severity of the disease. Patients must be informed and should agree with the treatment strategy before starting antibiotic treatments.
Key points
- •
Targeted systemic antibiotherapy can dramatically improve hidradenitis suppurativa (HS) but randomized controlled trials are lacking to recommend these treatments.
- •
Emergence of antibiotic resistance is a major concern.
- •
Empiric antibiotherapy for HS should target a broad spectrum, including anaerobes; carefully avoid underdosing; and rapidly stop treatment in case of lack of efficacy.
- •
Clinical remission can be obtained in patients with Hurley stage 1 HS using oral antibiotics (clindamycin-rifampin or rifampin-moxifloxacin-metronidazole) combinations and short-lasting remission or improvement in stage 2 and 3 with intravenous antibiotics. However, in the absence of a validated maintenance treatment, relapses are the rule, indicating the need for a combined antibiotic-surgical treatment.
- •
A multidisciplinary team including dermatologists, infectious diseases physicians, microbiologists, and surgeons is required to optimize the care of these patients.
Although the pathogenesis of inflammation in hidradenitis suppurativa (HS) is not fully understood, different types of antibiotics (ABs; tetracyclines, rifampin, clindamycin) are widely and increasingly used to treat HS in the absence of validated efficacy data. Considering the prevalence of HS, this issue is critical because potential antimicrobial resistance may hamper the efficacy of ABs and induce comorbidities associated with altered commensal microflora.
Rationale/background
- •
The constant and typical clinical features that define an active HS disease are abscesses, which are also a classic sign of an infectious process.
- •
Specific ABs are effective in HS.
- •
In a so-called inflammatory disease, the lack of reports on the efficacy of nonsteroidal antiinflammatory drugs (NSAIDs) and steroids, which are typical antiinflammatory drugs, is surprising and suggests that this disease may be not purely inflammatory.
- •
Some ABs definitely known to lack an antiinflammatory action (ertapenem, ceftriaxone) are effective in HS. Ertapenem is a β-lactam antibiotic inducing bacterial lysis and thus promoting inflammation. In addition, imipenem, a carbapenem, is considered an immunoenhancing antibiotic that increases leukocyte chemotaxis and has no antiinflammatory activity.
- •
In a microbiological study, prolonged cultures saved for 2 weeks were obtained by deep HS lesion biopsies, guided with ultrasonography in order to avoid surface contamination and by swab specimens, and in 4 to 7 days isolated a commensal flora, mostly anaerobic, varying with Hurley severity staging (detailed information is provided elsewhere in this issue). This study may explain the previous results in the literature: sterile because of a too-short length of culture, variable because of a rich commensal flora, and delay in growing depending on the species. These results also question the relevance and pathogenicity of this commensal flora in HS. However, if this flora is considered opportunistic in a genetically predisposed individual, this may explain the failures of ABs when they are not targeted against the isolated flora and also when the length of treatment is too short to eliminate bacteria in a host who has an innate immune anomaly, not allowing the clearance of these germs in the usual length of time.
Rationale/background
- •
The constant and typical clinical features that define an active HS disease are abscesses, which are also a classic sign of an infectious process.
- •
Specific ABs are effective in HS.
- •
In a so-called inflammatory disease, the lack of reports on the efficacy of nonsteroidal antiinflammatory drugs (NSAIDs) and steroids, which are typical antiinflammatory drugs, is surprising and suggests that this disease may be not purely inflammatory.
- •
Some ABs definitely known to lack an antiinflammatory action (ertapenem, ceftriaxone) are effective in HS. Ertapenem is a β-lactam antibiotic inducing bacterial lysis and thus promoting inflammation. In addition, imipenem, a carbapenem, is considered an immunoenhancing antibiotic that increases leukocyte chemotaxis and has no antiinflammatory activity.
- •
In a microbiological study, prolonged cultures saved for 2 weeks were obtained by deep HS lesion biopsies, guided with ultrasonography in order to avoid surface contamination and by swab specimens, and in 4 to 7 days isolated a commensal flora, mostly anaerobic, varying with Hurley severity staging (detailed information is provided elsewhere in this issue). This study may explain the previous results in the literature: sterile because of a too-short length of culture, variable because of a rich commensal flora, and delay in growing depending on the species. These results also question the relevance and pathogenicity of this commensal flora in HS. However, if this flora is considered opportunistic in a genetically predisposed individual, this may explain the failures of ABs when they are not targeted against the isolated flora and also when the length of treatment is too short to eliminate bacteria in a host who has an innate immune anomaly, not allowing the clearance of these germs in the usual length of time.
Clinical studies: review
Methods
A search was performed on PubMed using the terms “hidradenitis suppurativa” or “acne inversa” and “antibiotics,” from 1950 to 2015. Excluding randomized controlled trials (RCTs), which are described elsewhere in this issue, studies including at least 10 patients were retained. Minor studies with fewer than 10 patients were also reviewed but are briefly mentioned here. Studies using concomitant oral or topical non-AB medications were excluded.
Parameters studied were design and objectives (improvement vs remission), number of patients, outcome with mode of clinical assessment (clinical score of severity used), length of follow-up, initial severity, and side effects. Results are summarized in Table 1 .
Study | Aim of the Study | Type of Study | Patients/Treatment | Outcome Measures | Results | Side Effects | Limitations | Follow-up |
---|---|---|---|---|---|---|---|---|
Systemic AB | ||||||||
Bettoli et al, 2014 |
|
| pts: 23 (16 F, 7 M) do: clindamycin 600 mg; rifampin 600 mg du: 10 wk | Sartorius: r = improvement >25% reduction of flare-up | sc: 20 of 23 re: 17 of 20 (85%) dro b : 3 of 23 Sartorius mean reduction: 45% Flare-up: mean reduction: 60% | se: 3 of 23 (13%), mostly nausea and vomiting st: 1 of 3 | No randomization No control group | No |
Nassif et al, 2012 |
|
| pts: 30 Hurley: II, 18 pts; III, 12 pts do: ertapenem 1 g IV daily du: 6 wk | Hurley: r = reduction of at least 1 stage | Hurley: re = 17 of 30 8 pts (III>II) 8 pts (II>I) 1 pt (II>0) 13 of 30 pts (unchanged) |
| No randomization No control group Retrospective | No |
Join-Lambert et al, 2011 |
|
| pts: 28 (20 F, 8 M) Hurley l (6) Hurley ll (10) Hurley lll (12) do: metronidazole 1.5 g/d Rifampin 10 mg/kg/d Moxifloxacin 400 mg/d du: 4–6 wk; possible repeated courses | cr: clearance of all inflammatory lesions including hypertrophic scars | cr: 16 of 28 Hurley I: 6 of 6 (2.4 a ) Hurley II: 8 of 10 (3.8 m a ) Hurley III: 2 of 12 (6.2 m a first pt) (12 m a second pt) | se: GI disorders (64%) Vaginal candidiasis (35% of F) | No randomization No control group Different length of treatment Pretreatment with IV ceftriaxone and oral metronidazole in 14 patients Different severities Retrospective | Follow-up on: 14 of 16 cr Duration: 2–12 mo Secondary prophylaxis: 13 of 14 (TMP-SMZ 12 pts) or doxycycline (1 pt) Relapse: 7 of 14 |
Van der Zee et al, 2009 | Expand and validate the basis of the combination clindamycin + rifampin |
| pts: treated, 47; assessed, 34 5 different dosages, details not specified 23 of 34 do: rifampin 600 mg, clindamycin 600 mg du: 10 wk | Hurley Physician GA pi: <75% ti: >75% | re: 28 of 34 (82%) pi: 12 of 28 (35%) ti: 16 of 28 (47%) nr: 6 of 34 (18%) Length of treatment: No difference among: 10 wk, >10 wk, <10 wk High percentage of nonresponders in group 10 wk or longer No predominant response in severe cases | se: 13 of 34 (38%) 9 of 13 diarrhea 9 of 34 (26%) stopped for side effects 6 of 9 (diarrhea) No cases positive for Clostridium difficile colitis | No randomization No control group Different severities Heterogeneity of the included groups and few patients each group Retrospective | 13 of 34 Relapse 8 of 13 (61%) After 5.0 mo (mean) |
Gener et al, 2009 | Evaluate efficacy Clindamycin + rifampin Different severity (Hurley I, II, III) |
| pts treated: 116 pts analyzed: 70 do: clindamycin 600 mg + rifampin 600 mg du: 10 wk | Sartorius: improvement statistically significant Hurley Pain and Suppuration score (numerical scale 0–10) QoL score Patient GA | pts: 70 cr: 8 of 70 (11%) si: 59 of 70 (82%) nr: 1 of 70 w: 2 of 70 Pain score (7>3) Suppuration score (6>2) | se: 10 of 70 nausea, diarrhea, abdominal pain 1 of 10 skin eruption st: 8 of 10 | No randomization No control group Different severities Decision to treat on clinical ground | No |
Mendonca et al, 2006 | Assess efficacy: clindamycin and rifampin |
| pts: 14 (9 F, 5 M) do: clindamycin 600 mg Rifampin 600 mg du: 10 wk | Complete remission (parameter not detailed) | cr: 10 of 14 (71%) (8 rifampin/clindamycin 2 rifampin/minocycline) | se: 6 of 14 diarrhea 2 of 6 clindamycin > minocycline 4 of 6 dro | No randomization No control group Retrospective No data on disease severity |