FormalPara Key Summary Points

Early accessible office-based procedures that require minimal resources such as local excision and unroofing may facilitate timely surgical management of hidradenitis suppurativa

This review compared local excision and unroofing across 55 included articles based on a number of variables, including recurrence and complication rates

Recurrence rates for unroofing (14.5%) were found to be half those of excision (30%) across patients

Complication rates for unroofed lesions (12.5%) were found to be lower than for excised lesions (26%)

However, existing data are limited by low-quality uncontrolled studies with small sample sizes, variable reporting of outcomes, and lack of uniform definitions for recurrence and remission

Introduction

Hidradenitis suppurativa (HS) is a chronic inflammatory skin disorder centered around the hair follicle that appears to be the end result of a complex interplay of genetics, microbes, immune dysregulation, and environmental factors [3]. The hallmark of progressive disease is the creation of suppurative and connected dermal tunnels, tracts, bridging nodules, sinuses, and fistulas. This structural alteration of the skin has historically contributed to the belief that HS is a surgical disease, with wide or radical excisions performed to remove abnormal areas. However, such an approach is associated with significant morbidity, including nerve damage, thrombosis, bleeding, infection, dehiscence, and scar contracture [30]. With the introduction of improved medical therapies including broad-spectrum antibiotics, hormonal therapies, and biologic therapies, the role for surgery is shifting though it is still essential to current management.

HS is now considered both a medical and surgical disease, recognizing that the fundamental physical alteration of the skin in later stages of disease needs to be addressed alongside the other known mechanisms of the inflammatory component of the disease in all stages. There are numerous procedural modalities for HS, all with the common goal of altering or removing areas of abnormal skin that are typically not responsive to existing medical therapies alone. Simple procedures that can be performed in an outpatient office-based setting include incision and drainage (an incision made over suppurative nodules to facilitate drainage), unroofing (insertion of a probe into abscesses and sinuses of HS followed by a systematic removal of all involved tissue down to the base of the tracts), and limited or wide excision (removal of diseased skin with a margin of normal skin). Other procedures that require additional equipment include CO2 laser (laser excision, marsupialization, and destruction of tracts, usually with second intention healing), neodynium-yttrium aluminum garnet laser (Nd:YAG) (laser destruction of hair follicles), and electrosurgery (loop electrode removal of diseased skin down to the subcutaneous layer). Surgical approaches that typically require an operating room setting and surgical specialist include radical excision (removal of skin even beyond the clinically evident areas in an effort to capture hidden disease). These treatments vary markedly in their technique, level of invasiveness, and morbidity.

This review focuses on excision and unroofing for several reasons. First, excision in its various forms (limited/local, wide, and radical) is traditionally the most common procedure for HS and is considered a surgical mainstay, especially for Hurley III patients. Of note, it is rare for authors to uniformly define the diameter or exact width of an excision in the literature, making it difficult to discriminate among limited (removal of diseased area with < 1 cm margin), wide (removal of diseased area with several centimeters of normal skin), and radical (removal of entire anatomic subunits) excision. The size of excision determines whether this can be a realistic in-office procedure for dermatologists, with limited excision typically performed under local anesthesia and radical excision reserved for an operating room setting with surgical specialists. In contrast, unroofing has a much smaller literature base and is almost always performed in the outpatient setting. Unroofing generally involves insertion of a probe into tunnels of HS followed by a systematic removal of all involved tissue down to the base of the tracts [58]. Although these two procedures are very different, they can be used on overlapping types of HS lesions, including inflammatory nodules, tunneling sinus tracts, abscesses, and scarring. They are perhaps the more accessible of the many surgical options, requiring no additional office equipment or specialists in many cases. Incision and drainage is also a simple in office procedure, but was not reviewed because it carries a 100% recurrence rate and is best reserved for relief of pain secondary to acute inflammatory nodules or abscesses [16, 36].

Dermatologists can play a crucial role in the timing and type of procedures utilized for HS because they often serve as the medical home for patients with HS, optimizing medical management while coordinating referrals for surgical and other specialists. This review intentionally focuses on two office-based procedures that can be performed by dermatologists: excision and unroofing. This review summarizes data on recurrence and complication rates for excisions and unroofing to better assist physicians attempting to navigate these surgical treatment decisions.

This article is based on previously conducted studies and does not contain any studies with human participants or animals performed by any of the authors.

Methods

A literature search for articles was conducted using the PubMed database from 1 January 1985 to 31 December 2019 (Fig. 1) using the search terms (“hidradenitis suppurativa OR acne inversa”) AND (“surgery”), AND (“procedure”), AND (“excision”), AND (“unroofing” OR “deroofing”). A filter was used for this search to include only English language articles and full-text articles. The titles and abstracts were reviewed, and articles were included if they discussed surgical management of HS and excluded if they did not discuss surgery, focused mainly on reconstructive technique or wound healing, focused only on other procedural modalities (e.g., CO2 laser), were short communications (such as letters to the editor or viewpoints), were reviews or meta-analyses, did not have HS as the primary focus or only focused on management of HS sequelae (e.g., lymphedema), or focused only on description of surgical technique. Duplicate articles were then removed. Patient demographics, clinical characteristics, HS characteristics, HS medical and surgical intervention, and postoperative complications were collected for each study using a pre-determined standardized data form. Descriptive statistics were calculated, and recurrence and complication data were first totaled across the included studies at both the patient and the lesion level. These data were stratified by procedure (excision versus medical therapy) and summarized using counts and frequencies. Rates of recurrence and complication were compared between the two procedural groups using chi-square tests or Fischer’s exact tests where appropriate.

Fig. 1
figure 1

Diagram of literature review

Full size image

Results

A total of 1621 potentially relevant unique articles were identified from the initial literature search. Of these, 594 articles were excluded because they did not discuss surgery, 207 articles were excluded because of having the sole focus on reconstruction and/or wound healing, 76 articles were excluded because of having a focus only on other procedural treatments (e.g., CO2 laser), 48 articles were excluded because they were short communications (e.g., letters to the editor), 187 articles were excluded because they were reviews or meta-analyses, 192 articles were excluded because HS was not the primary focus or only discussed management of sequelae of HS (e.g., lymphedema), and 92 articles were excluded because they focused only on descriptions of surgical technique (Fig. 1). Fifty-five unique articles were ultimately selected based on inclusion/exclusion criteria, including 32 retrospective studies (58%), 12 case series/case reports (22%), and 11 prospective studies (20%). Across all studies, 3914 patients were included for analysis. A summary of all articles is included in Table 1.

Table 1 Summary of articles
Full size table

Excision

Forty-five articles (81%) comprising 2816 patients discussed excision in the treatment of HS: 60% of these articles were retrospective studies (n = 27), 22% were case series/case reports (n = 10), and 18% were prospective studies (n = 8). See Table 2 for a summary of excision data. Hurley staging was provided in 34.7% (n = 976) of these patients (Hurley stage 1: 5.2%, n = 51; Hurley stage 2: 31.3%, n = 305; Hurley stage 3: 63.1%, n = 616). Excision studies that reported an average duration of disease (n = 15) had a mean duration of 19.7 years. Seven excision studies reported an average time from disease onset to excision, with a mean of 7.6 years. Follow-up periods were reported differently across all the excision studies (Table 1). Of the studies that provided a mean follow-up, the mean follow-up period was an average of 40 months (median 21.2, IQR = 45.9, n = 26).

Table 2 Summary of recurrence and complication data across excision and unroofing studies
Full size table

Excision: Recurrence

Eighty-two percent of studies reported the recurrence rate (n = 37). Seventeen percent of studies had defined criteria for recurrence of HS lesions (n = 8), such as “recurrence of a lesion within 5 mm of the operated area” or “an inflammatory boil immediately within the scar or within less than 0.5 cm from the scar.” None of the studies had identical recurrence criteria. Of the 37 articles that reported recurrence data after excision (representing 1840 patients), the recurrence rate was 30% (n = 554 patients). Three studies reported recurrence by lesion number, with 98 of 371 lesions recurring across these studies (26.4%) [21, 42, 13]. One study reported recurrence by Hurley stage as follows: 3 lesions recurred in Hurley 1 patients (8.1%), 12 lesions recurred in Hurley 2 patients (32.4%), and 22 lesions recurred in Hurley 3 patients (59.5%) [21]. Hurley stage was not found to be a significant risk for recurrence based on this data. Seven excision studies reported on recurrence by specific site on the body. Of the 62 patients who had recurrence reported by site, 25 (40%) were in the inguinal area, 17 (27%) axillary, 15 (24%) genital, 4 (6%) perineal, and 1 (1%) in the submammary area. Of the 311 lesions that had recurrence, 139 (45%) recurred in the axillary area, 62 (20%) inguinal, 36 (12%) perineal, 10 (3%) submammary, and 3 (1%) genital.

Excision: Complications

Sixty percent of excision studies had specific data on the complication rate and type (n = 27). The complication rate calculated across studies was 31.2% (409 complications/1311 patients). The top three reported complications were infection (n = 13), postoperative bleeding (n = 10), and dehiscence (n = 7), and other reported complications included pain and contracture/stricture of scar, delayed wound healing, hypergranulation, decreased mobility, and hypertrophic scar. Three studies reported the complication rate by lesion number, with a 26.1% complication rate across lesions (96 of 367 lesions had a complication; 4 lesions were excluded from the complication analysis by one author [21]). The top three complications among these three studies included wound dehiscence (n = 58 lesions), infection (n = 8 lesions), and contracture (n = 7 lesions), and other reported complications included lymphedema and scar retraction. No excision studies reported on complication rate by anatomic site.

Excision Combined with Medical Therapy

Medical therapy alongside excision was reported in 12 studies (804 patients), with variable timing of perioperative administration of medical therapy. Thirty-three percent (n = 4) of these articles were prospective studies, 42% (n = 5) were retrospective studies, and 25% (n = 3) were case series/case reports; 66% (n = 8) of these articles also reported data on postoperative complications, with a combined complication rate of 32% (150/457 patients). Of the studies that reported the postoperative recurrence rate in the setting of adjuvant medical therapy, the mean recurrence rate was 13% (114/863 patients). The majority (n = 11) of these studies reported on the use of antibiotics and biologic medications, with only two mentioning retinoids [47, 62]. Of these, one study discussed post-surgical use of biologics [18], four studies detailed use of biologics and antibiotics in the weeks or months preceding surgery [37, 39, 56, 63], three detailed antibiotic and/or biologic use both pre- and post-surgical intervention [41, 46, 64], one detailed intraoperative use of a gentamicin sponge [11], and two studies did not specify exact timing [47, 54]. Prens et al. [46] reported significantly longer healing times with the use of adalimumab (five patients) and infliximab (one patient) started several months before surgery, held for 1–2 weeks perioperatively, and then re-instated postoperatively within 1–2 weeks. Worden et al. [64] performed a multivariate logistic regression analysis to analyze factors that negatively impacted healing and found a significant negative impact on healing if immune modulating therapy (infliximab, adalimumab, prednisone) was not held at least 2 weeks before surgery. Only one study detailed the use of adjuvant biologic therapies, including infliximab (eight patients) and ustekinumab (three patients) initiated 2–3 weeks after closure and continued for an average of 10.5 months [18]. This did result in a significantly lower rate of local recurrence, time to recurrence, disease progression, and time to progression in the combined therapy group compared with the surgery only group [18]. There was no significant difference in wound healing between the combined therapy and surgery only group.

Unroofing

Six articles (11%) comprising 222 patients discussed unroofing in the treatment of HS: one was a retrospective study, three were prospective studies, and two were case series/case reports. See Table 2 for a summary of unroofing data. Hurley staging was provided in four studies (n = 174) (Hurley stage 1: 7.5%, n = 13; Hurley stage 2: 77.6%, n = 135; Hurley stage 3: 14.9%, n = 26). Only one unroofing study reported an average duration of disease, which was 14.9 ± 9.4 years (range 4–30) [26]. Two unroofing studies reported on the timing of unroofing following disease onset, reporting a mean of 18.3 years [26] and median of 12 years [7]. The mean follow-up period for the unroofing studies was 240 days (SD = 207, n = 6).

Unroofing: Recurrence

Five of the six unroofing studies reported recurrence rates [7, 15, 26, 38, 58]. Only two studies reported recurrence criteria, such as “an inflammatory nodule or abscess located within 2 cm of the scar” or self-reported “newly described disease adjacent to or within the previously operated area.” Recurrence across patients was reported as 14.5% in 3 studies (8 recurrences/54 patients) [15, 26, 38] and recurrence across lesions was reported in 3 studies at a rate of 20% (133 recurrences/657 lesions) [7, 15, 58]. Time to recurrence was reported by only one study and ranged from 14 months to 87.5 months from the operative date [7]. No unroofing studies reported on recurrence by anatomic site.

Unroofing: Complications

Four unroofing studies reported complication data; one was a case report [38], and three reported complication rate across lesions [7, 15, 58]. A case report of one patient reported no complications [38]. There was a 12.5% complication rate across unroofed lesions (83 complications/666 lesions). The top three reported complications included postoperative bleeding (n = 3), hypergranulation tissue (n = 1), and wound infection (n = 1). No unroofing studies reported on complication by anatomic site.

Unroofing Combined with Medical Therapy

Data on the use of medical therapy alongside unroofing were reported in only one study including 8 patients with 28 lesions [26]. Intravenous azithromycin and oral metronidazole were used for 5–7 days prior to unroofing, sinuses were washed daily with saline that contained gentamicin, and intravenous azithromycin, oral metronidazole, and levofloxacin were used for 5–7 days postoperatively; 82.1% (24/28 lesions) showed improvement, and there were no recurrences within the 3-month follow-up period; 28.5% of lesions had complications (8 complications/28 lesions), and the three most common were pain (n = 4), oozing (n = 3), and fever (n = 2).

Excision and Unroofing (Combined Data)

Four studies reported data on excision and unroofing collectively in retrospective studies, representing 876 patients [25, 35, 36, 59]. Two studies reported grouped recurrence rates across multiple procedures (e.g., excision, unroofing) and did not specify intervention [35, 36]. Kohorst et al. [36] reported a grouped recurrence rate of 24.4% for a group of 590 patients who underwent excision, unroofing, and incision and drainage, whereas Kohorst et al. [35] had a patient-reported recurrence rate of 40.2% for a group of 111 patients who underwent excision, unroofing, or curettage. For the group of 111 patients, unroofing and excision had similar recurrence risks, and neither Hurley II nor III patients had an increased risk of postoperative recurrence compared with Hurley I patients.

Three of these studies specified postoperative complications, with an overall rate of 4% (81 complications/765 patients) [25, 36, 59]. The top complication was infection (n = 2). Each additional complication was reported once and included wound dehiscence, bleeding, contracture, skin graft loss, delayed wound healing, anemia, stricture, tightness of skin, burning/numbness of skin, and hypoalbuminemia.

Overall Recurrence and Complication Rate Results

Recurrence rates at both the patient and lesion level were both significantly associated with type of procedure. Thirty percent of patients treated with excision had a recurrence compared with 15% of those treated with unroofing (p = 0.015). Similarly, 26% of lesions recurred after excision, compared with 20% after unroofing (p = 0.02). Complication rates at the lesion level were also significantly associated with procedure, with rates after excision more than double those after roofing (26% vs. 12%, p < 0.001). The complication rate after combined medical and surgical therapy did not differ between procedures. Because no patients were either reported to have complications after unroofing or recurrences after unroofing combined with medical therapy, comparisons were not possible in these categories.

Discussion

This review demonstrates that excision and unroofing significantly differ in recurrence and complication rates when examined by both patients and lesions. The recurrence rate of unroofing was found to be half that of excision across patients and slightly lower across unroofed lesions. Unroofing also has a significantly lower complication rate than excision. Unroofed lesions have half the complication rate of excised lesions, though complications such as bleeding and pain are similar across the two modalities.

The existing data also suggest that recurrence rates may be lower in surgical intervention when used alongside medical therapy. The recurrence rate when excision was combined with medical treatments such as antibiotics and biologics was less than half that of surgery alone and was the lowest recurrence rate found in this review. Because only one unroofing study described a combined medical and surgical approach with antibiotics, a comparison of excision and unroofing was not possible. Despite the lower recurrence rate for excision combined with medical therapy, studies reported concerns about poor wound healing with immunomodulating interventions such as biologics or systemic steroids. The complication rate for the excision with the medical therapy group was equivalent to the rate of excision alone, but significantly longer healing time was noted with the use of biologic therapy including adalimumab and infliximab. The exception was one study [18] that used only adjuvant biologic therapy initiated several weeks after closure, where no difference in wound healing was found.

This review highlights that both excision and unroofing may play important roles in the current management of HS. The existing data suggest that either approach can be successfully employed based on factors such as disease severity, extent of involvement, clinical setting, provider comfort, and available resources. The lower recurrence and complication rates for unroofing may be particularly relevant to any practitioner looking for an effective and fast technique, as unroofing has been noted for its tissue-saving approach, suitability as an office procedure that rarely requires general anesthesia, and high patient satisfaction rate [58]. These data also begin to suggest that combining medical therapies with surgical intervention may improve surgical outcomes, although the ideal timing of these medical therapies and balancing them with wound healing concerns remain unclear.

The North American HS guidelines [3] and HS ALLIANCE Zouboulis et al. [66] make recommendations for surgical intervention based on lesion type and Hurley stage. The Hurley staging system is a widely used clinical tool that was developed to help guide surgical management by classifying patients according to physical signs of HS. Hurley I has traditionally been characterized as a non-surgical state that is medically managed, with single or multiple isolated nodules or abscesses without dermal tunnels or scarring. Hurley II denotes recurrent abscesses and single or multiple widely separated lesions with tunnel formation and scarring. Hurley III has diffuse involvement across a regional area and multiple extensive interconnected tunnels, abscesses, and scarring. A traditional treatment model dictates medical therapy only for Hurley I, with surgical intervention reserved for Hurley II and III for disease that is uncontrolled by medical therapy. This review found that Hurley staging was inconsistently reported in existing studies both pre- and postoperatively, making it impossible to compare outcomes across Hurley stage for each type of procedure. Moving forward, specifying the stage of HS in studies of surgical intervention for HS both pre- and postoperatively may provide important context for results and will be instrumental in determining the efficacy of specific interventions for defined types of HS lesions and optimizing surgical intervention for HS.

There is a growing body of literature on combination therapy with medical treatments and surgical intervention. In our review of the limited existing data, we found a lower recurrence rate for excision combined with medical therapy compared with the recurrence rate for excision alone, and the complication rate was comparable. Comparison of recurrence for unroofing with and without medical therapy was limited by the limited number of combination therapy unroofing studies (only one study of eight patients). A higher complication rate was seen in this study compared with unroofing without medical therapy, but this comparison is highly limited given this is a small number of patients in a single study. Antibiotics and biologic therapy were the two most common adjuvant therapies in this review. Our findings align with other reports in the literature of using medical therapy alongside surgery, which the North American HS guidelines deemed generally beneficial with minimal risk of increased postoperative complications [3]. The Safety and Efficacy of Humira (Adalimumab) for HS Peri-Surgically study (SHARPS study), which assessed the role of adalimumab in combination with HS surgery by randomizing participants to receive placebo or adalimumab before and after surgery, will be a helpful addition to understanding the extent to which biologic therapy might affect surgical outcomes in the setting of severe HS [51]. [“Safety and Efficacy of Humira (Adalimumab) for Hidradenitis Suppurativa (HS) Peri-Surgically (SHARPS Study)—Full Text View—ClinicalTrials.gov,” n.d.]. Similar trials and prospective controlled studies are needed to definitively compare recurrence and complication rates between surgery alone versus combination therapy and to determine optimal medical intervention alongside surgery.

Several reviewed studies and other published reviews also comment on the promising nature of early intervention based on small prospective and larger retrospective studies, with a potentially higher cure rate for surgical intervention at an early Hurley stage compared with later Hurley stages [1, 32, 50, 53]. Based on these studies, excision and unroofing may be particularly useful for focal areas of disease. Meaningful comparison of disease duration prior to excision versus unroofing was limited by the very small number of unroofing studies that reported this information. The potential impact of early surgical intervention with medical therapy is not definitively known, but our increasing comprehension of HS pathophysiology reinforces the idea that structural alteration of the skin occurs quickly and often aggressively. Thus, this review provides data on two surgical interventions that can play an early role for practitioners interested in this approach.

This review is limited by low-quality studies with small sample size, uncontrolled retrospective and prospective studies, and case reports that did not control for confounding variables. Data on duration of follow-up, medical comorbidities, specific anatomic region, size of excision, and type of reconstruction were not uniformly collected across studies, and outcome rates are variably reported, with some studies reporting data based on number of patients and others reporting data based on number of lesions. Recurrence and complication data by specific site on the body were scarce, reported in only a small number of excision studies and no unroofing studies. The number of studies included in this review is limited, particularly for the topic of unroofing. As a result, it is difficult to provide a true definitive comparison between unroofing and excision to guide treatment choices. Finally, although recurrence is used as a primary outcome, uniform definitions for recurrence and remission in HS do not exist, making it difficult to compare efficacy across studies [57]. Defining fundamental outcomes terminology and outcome measures for HS is currently ongoing through the International Dermatology Outcome Measures (IDEOM) initiative [19].

Conclusions

HS is a disease that has marked variability in its presentation and progression, requiring both medical and surgical management that is tailored to the type of HS lesion. Definitive conclusions about recurrence and complication rates for excision and unroofing are limited by the quality of the current literature. However, based on available data, recurrence and complication rates for unroofing appear to be lower than for excision. Though preliminary, it also appears that continuing medical therapy in the perioperative period may be helpful in terms of limiting recurrence, though concerns about delayed wound healing with biologic therapy have been reported. Further systematic prospective studies are needed to better compare complication and recurrence rates across these procedures in HS, especially in the context of concomitant medical therapy.