Abstract

Background

Hidradenitis suppurativa (HS) is a chronic inflammatory skin condition characterised by recurrent painful boils in flexural sites, such as the axillae and groin, that affects about 1% of the population, with onset in early adulthood.

Objectives

To assess the effects of interventions for HS in people of all ages.

Search methods

We searched the following databases up to 13 August 2015: the Cochrane Skin Group Specialised Register, CENTRAL in the Cochrane Library (Issue 7, 2015), MEDLINE (from 1946), EMBASE (from 1974), and LILACS (from 1982). We also searched five trials registers and handsearched the conference proceedings of eight dermatology meetings. We checked the reference lists of included and excluded studies for further references to relevant trials.

Selection criteria

Randomised controlled trials (RCTs) of all interventions for hidradenitis suppurativa.

Data collection and analysis

Two review authors independently assessed study eligibility and methodological quality and performed data extraction. Our primary outcomes were quality of life, measured by a validated dermatology‐specific scale, and adverse effects of the interventions.

Main results

Twelve trials, with 615 participants, met our inclusion criteria. The median number of participants in each trial was 27, and median trial duration was 16 weeks. The included studies were conducted over a 32‐year time period, from 1983 to 2015. A single RCT that was underpowered to detect clinically meaningful differences investigated most interventions.

There were four trials of anti‐TNF‐α (tumour necrosis factor‐alpha) therapies, which included etanercept, infliximab, and adalimumab. Adalimumab 40 mg weekly improved the Dermatology Life Quality Index (DLQI) score in participants with moderate to severe HS by 4.0 points relative to placebo (95% confidence interval (CI) ‐6.5 to ‐1.5 points), an effect size approximately equal to the DLQI minimal clinically important difference. We reduced the evidence quality to 'moderate' because the effect size was based on the results of only one study. In a meta‐analysis of two studies with 124 participants, standard dose adalimumab 40 mg every other week was ineffective compared with placebo (moderate quality evidence). In a smaller study of 38 participants, of whom only 33 provided efficacy data, infliximab 5 mg/kg treatment improved DLQI by 8.4 DLQI points after eight weeks. Etanercept 50 mg twice weekly was well tolerated but ineffective.

In a RCT of 200 participants, no difference was found in surgical complications (week one: risk ratio (RR) 0.78, 95% CI 0.58 to 1.05, moderate quality evidence) or risk of recurrence (after three months: RR 0.96, 95% CI 0.68 to 1.34, moderate quality evidence) in those randomised to receive a gentamicin‐collagen sponge prior to primary closure compared with primary closure alone.

RCTs of other interventions, including topical clindamycin 1% solution; oral tetracycline; oral ethinylestradiol 50 mcg with either cyproterone acetate 50 mg or norgestrel 500 mcg; intense pulsed light; neodymium‐doped yttrium aluminium garnet (Nd:YAG) laser; methylene blue gel photodynamic therapy; and staphage lysate, were relatively small studies, preventing firm conclusions due to imprecision.

Authors' conclusions

Many knowledge gaps exist in RCT evidence for HS. Moderate quality evidence exists for adalimumab, which improves DLQI score when 40 mg is given weekly, twice the standard psoriasis dose. However, the 95% confidence interval includes an effect size of only 1.5 DLQI points, which may not be clinically relevant, and the safety profile of weekly dosing has not been fully established. Infliximab also improves quality of life, based on moderate quality evidence.

More RCTs are needed in most areas of HS care, particularly oral treatments and the type and timing of surgical procedures. Outcomes should be validated, ideally, including a minimal clinically important difference for HS.

Author(s)

John R Ingram, Pick‐Ngor Woo, Ser Ling Chua, Anthony D Ormerod, Nemesha Desai, Anneke C Kai, Kerry Hood, Tara Burton, Francisco Kerdel, Sarah E Garner, Vincent Piguet

Abstract

Plain language summary

Treatments for hidradenitis suppurativa

Background

Hidradenitis suppurativa (HS) is a long‐term, distressing skin condition involving multiple painful boils in skin creases, such as the armpits, groin, and genital region, estimated to affect about 1 in 100 people. It typically begins in early adulthood and has a large impact on quality of life because of pain, scarring, and low self‐esteem. Doctors and the general public have largely ignored the condition, in part because people with HS do not wish to draw attention to their condition, so there is a relative lack of evidence to guide treatment.

Review question

What are the beneficial and harmful effects of treatments for hidradenitis suppurativa in terms of changes in quality of life and side effects?

Study characteristics

Our review included only randomised controlled trials (RCTs); we included 12 trials, containing a total of 615 people. In most cases, only a single trial that was too small to provide meaningful results investigated the treatments. There was no RCT evidence to support several quite commonly used treatments. The average duration of the trials was four months, long enough to check whether a treatment works initially but not long enough to show the duration of disease control or to detect delayed side effects.

Key results

The evidence from two trials for clindamycin lotion applied to the skin and oral tetracyclines was relatively weak, despite these antibiotics being standard treatments for mild to moderate HS. There were four pharmaceutical industry‐sponsored trials of anti‐TNF‐α (tumour necrosis factor‐alpha) therapies, which included etanercept, infliximab, and adalimumab. Of these, a trial of etanercept did not find benefit, whereas a small trial of infliximab reported an improvement in quality of life after eight weeks. A larger trial, including 154 participants, investigated adalimumab. There was no benefit for moderate to severe HS at standard psoriasis doses of 40 mg every other week, but 40 mg weekly did improve quality of life. The estimate of quality of life improvement ranged from a level that probably would help people with HS to a level that might not be enough to justify use of adalimumab. The trial found no increase in serious side effects, including infections, but it was not large enough to detect rare effects. There were no trials investigating when to perform surgery or what surgical procedure to consider. One trial looked at inserting an antibiotic sponge into wounds after removal of HS lesions, but found no benefit compared with surgery without the antibiotic sponge. There were three trials of laser‐type treatments, but the trial quality was too low to recommend these therapies.

Quality of the evidence

Our review has highlighted a need for more clinical trials to give better evidence to guide treatment choices in HS. More trials of oral treatments are required as well as surgical studies. Future trials should include patient‐reported outcomes, such as quality of life and pain.

Author(s)

John R Ingram, Pick‐Ngor Woo, Ser Ling Chua, Anthony D Ormerod, Nemesha Desai, Anneke C Kai, Kerry Hood, Tara Burton, Francisco Kerdel, Sarah E Garner, Vincent Piguet

Reviewer's Conclusions

Authors' conclusions 

Implications for practice 

Overall, our review found that knowledge gaps predominate over robust evidence for the treatment of hidradenitis suppurativa (HS). Only 12 RCTs met our inclusion criteria, with a total of 615 participants. Imprecision due to small numbers of participants led us to downgrade the quality of evidence for several of our comparisons. In the context that most interventions were investigated by a single randomised controlled trial (RCT), for which a median of 27 participants were included for 16 weeks, it is difficult to draw meaningful conclusions about adverse effects, particularly rare or delayed effects.

Jemec 2012 and the systematic reviews discussed above provide a framework for HS treatment based on disease severity and lack of response to previous treatments. However, it is difficult to provide strong recommendations from our review because the RCT data are often limited in terms of study size and quality. For example, we have not found sufficient evidence to determine the effects of topical clindamycin or oral tetracycline, despite these being standard treatments for mild to moderate HS. Only one small trial of 30 participants has investigated topical clindamycin 1% solution compared with vehicle placebo (Clemmensen 1983), and the results of the two outcome measures, an unvalidated participant diary and a physician‐reported severity score, were not consistent. We found no placebo‐controlled RCTs of oral tetracyclines for inclusion in our review. Jemec 1998 provided low quality evidence regarding the effect of oral tetracycline compared with topical clindamycin in 46 participants. The study reported no difference in efficacy between the two groups in terms of pain, HS Severity Score, and Physician Global Assessment, but there was a difference in the participant global self‐assessment outcome of 28 mm on a visual analogue scale (VAS) from 0 to 100 mm in favour of oral tetracycline. Overall, the evidence is too limited to provide a recommendation. Similarly, we cannot use the data from Mortimer 1986 to provide a recommendation regarding the endocrine interventions under investigation because of its small size.

Moderate quality evidence does exist for adalimumab therapy. Kimball 2012 demonstrates that adalimumab weekly improves quality of life, although the 95% confidence interval includes an effect size of only 1.5 Dermatology Life Quality Index (DLQI) points, which may not be clinically relevant (Basra 2008; Basra 2015). Meta‐analysis of Kimball 2012 and Miller 2011 shows that adalimumab every other week (EOW) is ineffective. This evidence is likely to affect the treatment of only a relatively small subset of HS patients because anti‐TNF‐α therapy is usually reserved for severe HS refractory to other treatments because of high cost and the potential for serious adverse effects. In particular, the safety profile of weekly treatment has not been established because Kimball 2012 was not powered to detect rare or delayed adverse effects (AEs), and ongoing psoriasis biologic safety registers do not include recipients of adalimumab weekly therapy. In addition, adalimumab weekly therapy is likely to cost twice as much as the standard dosing used for psoriasis, which may further restrict its availability. Results from the PIONEER studies of adalimumab therapy for HS, which should be reported in full in the near future, may help to improve confidence in the estimates of effect size for adalimumab weekly therapy. The evidence for infliximab is less robust, being based on a smaller study; however, the effect size of 8.4 DLQI points is likely to be clinically relevant. The evidence regarding etanercept is only moderate quality in a study of 20 participants with results suggesting that, even at a dose double that licensed for psoriasis, etanercept may not provide benefit in HS.

We downgraded the within‐participant studies of neodymium‐doped yttrium aluminium garnet (Nd:YAG) laser, intense pulsed light (IPL), and topical photodynamic therapy (PDT) to low quality evidence because of imprecision and a lack of blinding in the absence of sham treatments. Implications for practice arising from Buimer 2008 are limited by a high risk of performance bias and unclear risk of bias in most of the other domains due to incomplete study reporting. Furthermore, there was no difference found in the rate of surgical complications or risk of recurrence in the group randomised to receive a gentamicin‐collagen sponge prior to primary closure compared with primary closure alone, so a change in practice in this case cannot be recommended. The trial of staphage lysate was a small study, and there is insufficient evidence to warrant a change of practice based on the results of Angel 1987 alone.

The three studies in 'Studies awaiting classification' may alter the conclusions of the review once assessed.

Implications for research 

Our review has highlighted a need for further RCTs to improve the evidence base for most interventions in HS. One exception perhaps is in the field of biologic therapies, where the evidence is already of higher quality than for other interventions, and there are ongoing studies of adalimumab and anakinra therapy. The HS Priority Setting Partnership (PSP) (Ingram 2014) recently gave highest priority to the question "What is the most effective and safe group of oral treatments in treating HS? (e.g. antibiotics, hormonal treatments, retinoids, immunosuppressants, metformin, steroids)", and our review has highlighted important gaps in the evidence base for these commonly used treatment options.

In terms of RCT design, trials should include a power calculation and recruit sufficient participants to avoid problems with imprecision due to being underpowered. We also found that outcome measure heterogeneity is likely to be an issue in the absence of agreement about the key outcomes that should be included. Selected outcomes should be validated, and ideally, the minimal clinically important difference for the primary outcome should be determined in HS, to ensure that treatment success or failure is clearly defined. The outcomes of a trial should be prospectively declared in a clinical trial database, including the nature and timing of the primary outcome. An intention‐to‐treat analysis, with a predetermined method for dealing with missing data, should be incorporated to minimise the potential for attrition bias.

As well as trials of medical therapy, our review has demonstrated a need for more surgical RCTs. In particular, the RCT evidence base remains weak for timing of surgery, type of surgical procedure, and optimal postoperative wound care in HS. The HS PSP rated both "What is the best surgical procedure to perform in treating HS e.g. incision & drainage, local excision, wide excision?" and "What is the best method of wound care after surgery or for active disease? (e.g. skin grafts, secondary intention, dressings)" in the top 10 priorities for HS research. Although we included three laser or light RCTs in our review, we were not able to make treatment recommendations due to low quality evidence. Future trials in this area should incorporate a sham treatment control to minimise performance and detection bias, and if a within‐participant design is chosen, we recommend that the unit of randomisation should be the left and right sides of the same anatomical site, selecting participants with bilateral HS.

Comparison with a skin disease, such as vitiligo, with a similar prevalence demonstrates the need for more RCTs in HS to guide treatment. The updated Cochrane review for vitiligo (Whitton 2015) included 96 trials containing 4512 participants, which represents eight times as many RCTs compared with our HS review.

Get full text at The Cochrane Library