ORIGINAL CONTRIBUTION
Hidradenitis Suppurativa: MRI Features in Anogenital Disease Nyree Griffin, M.D., F.R.C.R.1 • Andrew B. Williams, M.D., F.R.C.S.2 Simon Anderson, M.D., F.R.C.P.3 • Peter M. Irving, M.A., M.D., F.R.C.P.3 Jeremy Sanderson, M.D., F.R.C.P.3 • Nemesha Desai, B.Sc., F.R.C.P.4 Vicky Goh, M.D., F.R.C.R.1,5 1 Department of Radiology, Guy’s and St Thomas’ National Health Service Foundation Trust, London, United Kingdom 2 Department of General Surgery, Guy’s and St Thomas’ National Health Service Foundation Trust, London, United Kingdom 3 Department of Gastroenterology, Guy’s and St Thomas’ National Health Service Foundation Trust, London, United Kingdom 4 Department of Dermatology, Guy’s and St Thomas’ National Health Service Foundation Trust, London, United Kingdom 5 Division of Imaging Sciences and Biomedical Engineering, King’s College London, London, United Kingdom
BACKGROUND: Hidradenitis suppurativa is a rare chronic inflammatory disorder of apocrine gland-bearing skin, which commonly affects the anogenital region. There has been very little literature to date on the MRI appearances of anogenital hidradenitis suppurativa. OBJECTIVE: The aim of this study was to assess the MRI features of anogenital hidradenitis suppurativa in the largest cohort of patients to be published to date. DESIGN: After an institutional review board waiver, patients
with hidradenitis suppurativa who were undergoing MRI for anogenital disease between 2005 and 2012 were identified from our institutional database. The MRI appearances were recorded by 2 radiologists in consensus, blinded to clinical details. Location of disease, number of tracts, presence of anal fistula, and supralevator involvement were recorded. Patient demographics were also noted. SETTINGS: This study was conducted at the Department
of Radiology, Guy’s and St Thomas’ National Health Service Foundation Trust.
Financial Disclosure: None reported. Poster presentation at the meeting of the European Society of Gastrointestinal and Abdominal Radiology, Edinburgh, United Kingdom, June 12 to 15, 2012. Correspondence: Nyree Griffin, M.D., Department of Radiology, Guy’s and St Thomas’ National Health Service Foundation Trust, Westminster Bridge Road, London SE1 7EH, United Kingdom. E-mail: [email protected] Dis Colon Rectum 2014; 57: 762–771 DOI: 10.1097/DCR.0000000000000131 © The ASCRS 2014
762
PATIENTS: Patients included were those undergoing MRI for anogenital disease in hidradenitis suppurativa between 2005 and 2012. MAIN OUTCOME MEASURES: The distribution of sinus tracts in anogenital hidradenitis suppurativa on MRI was measured. RESULTS: Thirty-one MRIs were performed in 18 patients (15 men; mean age, 46 years). On the baseline MRI, multiple tracts were seen in the natal cleft (16/18; 83%), the perianal (12/18; 61%), the perineal (13/18; 56%), and the gluteal (8/18; 44%) regions. A communication with the anal canal was present in only 4 patients. Three patients had supralevator extension. Seven patients had follow-up MRIs with variable response to interval treatment: 3 of 7 showed responding disease, 3 of 7 showed stable disease, and 1 of 7 showed progressive disease. LIMITATIONS: This study was limited by its relatively small cohort of patients. CONCLUSIONS: In hidradenitis suppurativa, anogenital disease is usually subcutaneous but extensive, with only a minority of patients demonstrating deeper involvement. MRI may help define the extent of anogenital disease and assess response to treatment. KEY WORDS: Anogenital; Hidradenitis suppurativa; MRI; Sepsis; Therapy.
H
idradenitis suppurativa (HS), also known as acne inversa, is a rare chronic inflammatory skin disease that involves apocrine gland-bearing sites. It has a predilection for the axillary and anogenital regions but can also involve the submammary, intermammary, periumbilical, and retroauricular skin. The condition can Diseases of the Colon & Rectum Volume 57: 6 (2014)
763
Diseases of the Colon & Rectum Volume 57: 6 (2014)
TABLE 1. Parameters for MRI protocol Parameter TR/TE, ms TI, ms Echo train length Flip angle, ° Field of view, mm No. of signal averages Section thickness, mm Intersection gap, mm Matrix size No. of sections per stack Bandwidth, Hz
Sagittal STIR
Axial oblique STIR
Coronal oblique STIR
5380/28 130 7 150 300 2 3 1 320 × 100 23 130
5380/28 130 7 150 300 2 4 1 320/100 23 130
5380/28 130 7 150 300 2 3 1 320/100 23 130
TE = echo time; TR = repetition time; TI = inversion time; STIR = short tau inversion recovery.
vary in severity and extent. Smoking is a recognized association. When HS affects the anogenital region, it may present with perianal pain, swelling, pruritus, bleeding, or purulent discharge. It can give rise to inflammatory abscesses and sinus tract formation. The clinical picture may overlap with Crohn’s disease, which may coexist in 17% to 40% of patients with HS.1,2 In the past decade, MRI has become the imaging modality of choice in the characterization of anal fistulas and the delineation of perianal disease because of its superb contrast resolution and multiplanar capability. It has been used mostly in patients with idiopathic anal cryptoglandular sepsis or perianal Crohn’s disease. However, MRI may also have a role in HS. In our institution, a tertiary dermatology referral center, MRI has been used increasingly to assess the extent of anogenital disease to direct and facilitate management in patients with HS. The role of MRI in HS may help to define disease extent, to assess for fistulizing disease, and to exclude Crohn’s disease. Yet, to date, there have been little published data to support its use, with only 1 case report describing the MRI features of perianal HS.3 The purpose of this article is to define the typical MRI appearances at presentation and after therapy in a cohort of patients with anogenital HS.
MATERIALS AND METHODS Patients
Institutional review board waiver was obtained for this retrospective analysis, with no need for informed consent. From the institutional information database, a search was carried out for all patients with the established diagnosis of HS undergoing MRI for the evaluation of anogenital disease between January 2005 and January 2012. Patients before 2005 were excluded, because no imaging studies were archived on the institution picture and archive communication system. The search terms “hidradenitis” and “pelvic MRI” were used. This yielded 18 patients (15 men and 3 women; mean age, 46 years; age range, 24–65 years). Eleven of these patients had 1 MRI study between 2005
and 2012, whereas the remaining 7 patients had between 2 and 4 follow-up MRI studies. Using the hospital electronic patient record database, the patient demographics were recorded for each of these 18 patients. MRI Protocol
MRI was performed on a 1.5-T scanner (Siemens Avanto, Erlangen, Germany). A phased-array surface coil was used with patients imaged in the supine position. No additional patient preparation was required. The protocol included a 3-plane localizer, followed by a sagittal short tau inversion recovery (STIR) sequence through the pelvis. Using the sagittal localizer, axial and coronal oblique (STIR) images were obtained, aligned perpendicular and parallel to the longitudinal axis of the anal canal. Imaging parameters are provided in Table 1. Image Review
Two experienced radiologists with a subspecialty interest in gastrointestinal MRI (>10 years of experience) reviewed all of the baseline and follow-up MRI studies in consensus, using a picture and archive communication system imaging workstation (Centricity, GE Healthcare, Little Chalfont, United Kingdom). Hence, all of the studies were reviewed by both radiologists reading the images together at the same time, with any difference in opinion over the distribution of tracts resolved by consensus. Inflammatory change secondary to HS was defined as high signal intensity seen in the soft tissues on STIR images separate to the blood vessels. The presence of inflammatory involvement was recorded as being absent or present for the following anatomic regions: 1) the perianal region, 2) the natal cleft, 3) the anterior perineal region, 4) the ischioanal fossa, 5) the inguinal region, 6) the gluteal region, 7) the coccygeal/ sacral region, and 8) the supralevator space. Definitions for these anatomic regions were as follows4: 1) the perianal region was defined as the superficial area within 2 cm of the anal orifice; 2) the natal cleft was defined as the midline tissues and adjacent buttock ≥2 cm posterior to the anal orifice; 3) the anterior perineal region was defined as the superficial soft tissues, ≥2 cm, anterior
764
Griffin et al: MRI of Anal Hidradenitis Suppurativa
TABLE 2. Patient demographics Patient 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Age, y
Sex
Ethnicity
Clinical sites of involvement
Smoker
Crohn’s disease
44 48 43 65 51 40 59 31 45 56 35 25 24 47 58 45 65 32
M M M F M M F F M M M M M M M M M M
Black, African White Black, Caribbean White White Black, Caribbean White White White White White Black, African Black, African Black, African White White White Black, Caribbean
Buttocks, perineum, head and neck, axillas Buttocks, perineum, head and neck Buttocks, scrotum, groin Groin, axillas Perineum, axillas Buttocks, scrotum, groin, penis Buttocks, thighs, perianal, anorecto-vaginal fistula Vulva, submammary Groin, buttock Groin, perianal, perineum, axillas Buttock, groin, scrotum, axillas Buttock, groin, scrotum, perineum, perianal Axillas, groin, scrotum, perineum, buttocks Groin, scalp, scrotum, perianal Groin, scrotum, perianal, thigh Perineum, buttocks, scrotum, groin, axillas Perianal, scrotum, buttocks Perineum, buttocks, axillas
Current Current Current Current Current Current Never Never Ex Current Current Never Current Current Current Current Current Never
No No No No No No No Yes No No No No No No Yes No Yes No
M = male; F = female.
to the anal orifice (including scrotum and vulva); 4) the ischioanal fossa was defined as the tissue bounded by the inferior aspect of the levator plates superiorly, the anal sphincter complex medially, the level of the inferior border of the subcutaneous external anal sphincter inferiorly, and obturator internus laterally; 5) the inguinal region was defined as the subcutaneous tissue within both groins; 6) the gluteal region was defined as the subcutaneous tissue overlying the gluteal muscles; 7) the coccygeal/sacral region was defined as the subcutaneous tissue overlying the bony coccyx and/or sacrum; and 8) the supralevator space was defined as the perirectal tissue within the pelvis above the levator plates. When inflammation was present, its morphology was classified as either a tract (defined as a longitudinal tubular high-signal structure on MRI) or as an abscess (defined as a localized spherical collection). When applicable, it was noted whether single or multiple tracts/abscesses were present for each site. In addition, it was also noted whether inflammatory change involved the anal sphincter complex and whether striated muscle in the form of either external anal sphincter or levator plate was breached. The site and level of any enteric openings were recorded if present. The presence of inguinal lymphadenopathy and myositis (defined as high signal on STIR in the gluteus maximus muscle) was also noted. The presence and site of subcutaneous edema within the tissues was also recorded, defined as diffuse subcutaneous high signal on STIR associated with skin thickening.
RESULTS Clinical Findings
Table 2 summarizes the patient demographics, the clinical sites of involvement, the patient smoking status, and the
presence of synchronous Crohn’s disease (established by biopsy, histology, and imaging). As can be seen in Table 2, the majority of patients with anogenital HS were men (n = 15 [83%]). Patients were either white (n = 11/18 [61%]) or Afro-Caribbean (n = 7 [39%]). The diagnosis of synchronous Crohn’s disease was established in only 3 patients. Thirteen patients had had previous perianal surgery, usually involving incision and drainage of abscesses or laying open of anogenital tracts. All of the patients had been treated with one or more courses of antibiotics for their HS. A few patients had received steroids (n = 6/18), infliximab (n = 5/18), or immunomodulators (n = 3/18). Imaging Features Baseline
Table 3 shows the distribution of disease on baseline MRI. At baseline, the most common sites of involvement (in decreasing frequency) were the natal cleft (n = 16/18), the perianal region (n = 12/18), the anterior perineal region (Fig. 1; n = 13/18) and the gluteal region (n = 8/18). Multiple tracts were more characteristic at these sites than solitary tracts, with discrete abscesses being most commonly seen in the gluteal region (n = 3/18). Involvement of the coccygeal/sacral region (n = 5/18), ischioanal fossa (n = 3/18; Fig. 2A), inguinal region (n = 2/18), and supralevator space (n = 3/18) was less frequent. An enteric communication with the anal canal was not common, only seen in 4 patients (Fig. 2B). In these patients, the internal opening was at the dentate line. Three patients had tracts that extended through the levator plate (Figs. 3A and B; patients 7, 10, and 17), with 1 of these patients also having a suspected rectovaginal fistula (patient 7; Fig. 3C). Only 1 of these 7 patients had an established diagnosis of Crohn’s disease. Subcutaneous edema (n = 10/18) was a
765
frequent finding with reactive inguinal lymphadenopathy (Fig. 4) sometimes demonstrated (n = 5/18). The presence of a high signal within the gluteus maximus muscle separate to the tracts was seen in 3 patients, suggestive of myositis (Fig. 5A). Follow-up
T = single tract; TT = multiple tracts; AA = multiple abscesses; LN+ = inguinal lymphadenopathy; IAS = internal anal sphincter; EAS = external anal sphincter.
No No Gluteal edema No No Gluteal edema No No Scrotal edema No No Gluteal/perineal edema Dentate line, 6 o’clock No Dentate line, 3, 5 and 7’o’clock EAS Gluteal, scrotal edema, LN+, gluteal myositis No Levator Rectovaginal fistula, myositis No No Labial/inguinal edema No No Dentate line, 6 o’clock EAS, Levator LN+, gluteal myositis No No Gluteal edema No No LN+ Dentate line, 6 o’clock EAS LN+ No No No No Gluteal edema No No Gluteal edema No Levator No No Gluteal edema, LN+ No No No No Yes Yes No No No No No No No No No No No No No No TT TT No TT, AA No No No TT TT No TT, AA No No TT No TT, AA No No No No No No TT No No T No No No No No No T No No No No No No No No TT No No No No No No No No No TT TT T TT TT No No No TT T T No TT TT TT TT TT No TT, AA No No No No No TT No No No TT TT No TT No No TT No No TT TT TT TT No TT TT TT T TT TT TT TT TT TT TT No TT No No No No No TT TT No No T No No No No No No No No TT TT TT No T TT TT TT No TT TT No TT TT No No TT No 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Natal Coccygeal/ Anterior Patient Perianal Ischioanal cleft sacral Perineal Inguinal Supralevator Gluteal Intersphincteric
TABLE 3. Distribution of tracts on baseline MRI
Enteric opening
Striated muscle
Additional features
Diseases of the Colon & Rectum Volume 57: 6 (2014)
Table 4 shows the change in distribution (if any) on follow-up MRI studies. Follow-up MRI studies were performed in 7 of the 18 patients, with all but 2 undergoing interval medical therapy. A variable response was seen. In 3 patients the number of tracts either decreased in number with reduced signal intensity (Fig. 5) or resolved completely (Fig. 6). The number of tracts was unchanged in 3 patients, although the tracts in 1 of these patients demonstrated lower signal intensity on follow-up, and the gluteal abscesses had resolved. Appearances worsened in 1 patient despite ongoing medical therapy, with greater fluid distension of the tracts and the development of gluteal abscesses (Fig. 7).
DISCUSSION Our results show that anogenital HS typically demonstrates extensive sinus disease on STIR MRI, most commonly affecting the natal cleft, perianal, and perineal regions. Anal sphincter and supralevator involvement only occurs in the minority of patients. Response to treatment as assessed on MRI is variable. The STIR sequence used in our MR protocol is an inversion recovery pulse sequence, which suppresses the signal from fat. It is commonly used in perianal fistula imaging, where tracts appear of high signal intensity against the low signal background. Some centers make use of T2-weighted fat-saturated sequences to provide similar information. The use of a high-resolution T2weighted sequence without fat suppression is not routinely used in fistula imaging but can be helpful in providing high-resolution images of the sphincter complex and as an adjunct to the STIR images if scanning time allows. Gadolinium-enhanced T1 fat-saturated images may also help to identify active disease (where the walls of the tracts and abscesses enhance). However, we very rarely include this in our protocol, because STIR sequences are considered sufficient; intravenous gadolinium is best reserved for complex patients who have been operated on previously. All of our studies were performed on a 1.5-T magnet because of local availability. Higher-strength magnets (eg, 3 T) can be used for pelvic imaging with the greater signal:noise ratio resulting in shorter acquisition times and/or higher spatial resolution, with the latter theoretically leading to greater conspicuity of smaller fistula tracts.5 However, disadvantages include limited availability at present and increased susceptibility artifacts, magnetic field
766
FIGURE 1. Sagittal short tau inversion recovery (STIR) image shows extensive perineal sinus disease (arrows; patient 5); sinus tracts are seen overlying the dorsum of the penis too (arrowheads).
inhomogeneity, and specific absorption rates.6 In practice, the 1.5-T magnet gives sufficiently high-resolution images for diagnostic purposes and is the most commonly used field strength for MRI in clinical centers. HS is a rare inflammatory dermatologic disease. Velpeau7 originally described HS in 1839. In 1922, Schiefferdecker,8 ascribed its etiology to the apocrine glands, which are compound sweat glands that extend through the dermis into the subcutaneous tissue and drain into hair follicles. However, relatively recent histologic studies have demonstrated that HS is an inflammatory disease
Griffin et al: MRI of Anal Hidradenitis Suppurativa
originating from the hair follicle, with inflammation of the apocrine glands occurring as a secondary phenomenon.9–12 Keratin plugging of the follicular unit and subsequent expansion and rupture of the hair follicle lead to spillage of keratin into the surrounding tissue, resulting in an intense inflammatory response. Bacterial superinfection occurs with pathogens including streptococci and staphylococci. In perianal disease, Escherichia coli, Klebsiella, Proteus, and anaerobes may be seen. HS typically develops after puberty, with peak onset in the early 20s, although prepubescent and postmenopausal cases have been described. The prevalence is estimated at 1% of the general European population,13 increasing to 4% in young adult women.14 There is a male predominance of anogenital disease, as seen in our study, although overall the condition is 3 times more common in women. It is associated with smoking and obesity.15 Fourteen of our patients were either active (n = 13) or former (n = 1) cigarette smokers. One third of cases are familial, commonly follow an autosomal dominant mode of inheritance, and have been linked with mutations of the γ-secretase complex.16 Longstanding disease may result in the development of squamous cell carcinoma, particularly at perianal sites.17 Other complications include the formation of anal, urethral, and rectal fistulae and strictures; anogenital lymphoedema; and inflammatory arthropathy. Clinically, initial presentation is with recurrent tender cutaneous nodules, abscesses, and draining sinus tracts. With time, sinus tracts extend to the surrounding tissue, resulting in large areas of chronic inflammation and fibrosis of the skin and associated structures. Severity of disease is variable, but this can be a very debilitating condition.18 The diagnosis is made on clinical grounds,19 when 3 criteria are fulfilled, as follows: 1) typical lesions are present,
FIGURE 2. A, Coronal short tau inversion recovery (STIR) image shows multiple sinus tracts in the ischioanal fossa (arrowheads) and left gluteal region (arrow; patient 6). B, Axial STIR image in the same patient shows transsphincteric fistula with internal opening at the 6 o’clock position at the dentate line (white arrow); there are multiple sinus tracts (arrowheads) in the natal cleft and gluteal region (both subcutaneous and within the muscle) associated with subcutaneous edema.
Diseases of the Colon & Rectum Volume 57: 6 (2014)
767
FIGURE 3. A, Coronal short tau inversion recovery (STIR) image (patient 7) shows a defect in the left ischioanal fossa from previous surgery (*) and tract extending above the left levator plate (arrow). B, Axial STIR in same patient shows supralevator horseshoe around the lower rectum, extending posterior to the vagina (V) with secondary tract seen extending anterolateral to the vagina. Urethra (U) has been labeled. C, Sagittal STIR in same patient shows suspected rectovaginal fistula (arrow) between the anterior wall of the rectum (R) and posterior wall of the vagina (V). Gluteal sinus disease is also seen (arrowhead).
that is, deep-seated painful nodules, abscesses, draining sinus tracts, bridged scars, and tombstone comedos; 2) the patient demonstrates typical topography of involvement, that is, axillas, groin, perineal and perianal region, buttocks, and inframammary and intermammary folds; and 3) disease shows chronicity and recurrence. Histologically, an extensive perifolliculitis is seen.9 Our study had several limitations. There were only a relatively small number of patients (n = 18), but this reflects the relative rarity of this condition and the fact that only the most severe cases were referred for anogenital imaging. It is still the largest cohort to be described in the literature, which only has isolated case reports.3 All patients had been on, or were continuing, with some form
of medical therapy, which again could have influenced anogenital appearances on the MRI study. Thirteen patients had undergone perianal surgery (usually consisting of incision and drainage of abscesses). This could account for the relatively low percentage of patients with abscesses seen on baseline and follow-up MRI studies. However, imaging with anogenital HS while patients are being treated is normal practice. Finally, only 7 patients had a followup MRI study, and because of the variety of individual medical and surgical interventions in this cohort, it was not possible to directly correlate appearances on follow-up MRI with specific therapies. To our knowledge, there has only been 1 case report in the literature describing the MRI features of anogenital
768
FIGURE 4. Coronal short tau inversion recovery (STIR) image (patient 3) shows multiple enlarged reactive inguinal nodes (arrows), perineal sinus tracts (arrowhead), and scrotal edema (*).
HS.3 MRI of the pelvis revealed induration of the subcutaneous tissues over the medial aspects of the thighs, perineum, and mons pubis with multiple subcutaneous tracts and small abscesses. Bilateral enlarged inguinal nodes were seen. There were no fistulous communications with the bladder, urethra, or rectum in their patient. Abscesses were noted to be remote from both the rectum and anus. The distribution of disease in this case report is similar to our cohort, where we demonstrated widespread sinus dis-
Griffin et al: MRI of Anal Hidradenitis Suppurativa
ease characteristically affecting the perianal, anterior perineal, inguinal, and gluteal regions. Subcutaneous edema and inguinal lymphadenopathy were also demonstrated in our cohort. Similarly, in a pictorial review of pathologic conditions affecting the scrotum,20 the authors described the appearances of scrotal HS on MRI, characterized by edema and enhancing microabscesses. There has only been 1 other case report that has described the appearances on barium enema of a patient with perianal HS. Multiple sinus tracks and fistulas were seen arising from the posterior wall of the distal rectum on barium enema.21 This rectal involvement is unusual in HS. We demonstrated anal sphincter involvement in 22% (n = 4/18) of our patients on MRI with the internal opening centered on the dentate line. It has been suggested that fistulas to the anal canal in HS should extend only as far as the lower part of the anal canal, because this is where apocrine glands can be found.22 However, in 3 of our patients, 1 of whom had Crohn’s disease, tracts breached the levator plate; 1 of these patients also had a suspected rectovaginal fistula. Fistulizing disease in HS, although rare, is well recognized in the dermatologic literature.23 Although there may be an overlap in the clinical picture of anogenital HS and Crohn’s disease, the distribution of perianal disease in Crohn’s disease is different from HS. In Crohn’s disease, fistula tracts and/or abscesses are usually seen primarily involving the anal sphincter complex.24 Sinus involvement of the anterior perineum and gluteal and inguinal regions is not typically seen in Crohn’s disease, although the development of anovulvar fistulas is well recognized. Extension into the ischioanal fossa, na-
FIGURE 5. Example of partial response to treatment (patient 10). A, Axial short tau inversion recovery (STIR) image at baseline shows multiple high signal sinus tracts in the natal cleft, perianal region, ischioanal fossa (arrowheads), and gluteal region (arrows). High signal is also seen in the gluteal muscles (*) in keeping with myositis. B, Axial STIR image on follow-up MRI shows a reduction in number and signal intensity of the tracts.
769
Diseases of the Colon & Rectum Volume 57: 6 (2014)
TABLE 4. Distribution of tracts on follow-up MRI scans Interval of second MRI from baseline
Appearances on second MRI
Interval of third MRI from baseline
Appearances on third MRI
2
6 mo
3
9 mo
6
25 mo
Reduced number and signal intensity of tracts in perianal region, no edema Same distribution, but now greater fluid distension, multiple abscesses (left), LN+ No change
31 mo
7 8
18 mo 6 mo
Same distribution, reduced signal intensity, gluteal abscesses and edema resolved No change Labial and inguinal edema only, no tracts
10
10 mo
15
9 mo
Patient
No change Less labial edema, otherwise no change Reduced number and signal intensity of tracts Increased subcutaneous edema, otherwise no change
27 mo 20 mo
18 mo
No change
Interval of fourth MRI from baseline
Appearances on fourth MRI
36 mo
No change
19 mo
No change
LN+ = inguinal lymphadenopathy
tal cleft, and supralevator space are more characteristic in Crohn’s disease.24 There has been no previous literature describing the MRI changes that occur after medical and/or surgical therapy in HS. A variety of medical and surgical treatments may be offered to patients with HS. Prolonged topical and systemic antibiotics are usually first-line medical therapies, with second-line therapies including systemic retinoids, antiandrogens in women, cyclosporin, and more recently tumor necrosis factor-α inhibitors (biologics), such as infliximab (although the use of the latter in HS is
presently unlicensed). Surgical treatments for HS include incision and drainage of abscesses, laying open of tracts, and limited and wide local excision. In the 7 patients with follow-up MRI in our study, a variable response was seen, with some patients showing a reduction in the number and signal intensity of sinus tracts, whereas others remained stable or developed worsening disease. Interval medical therapy (in the form of antibiotics, steroids, infliximab, or immunomodulators) was given to all but 2 of these patients who underwent incision and drainage of abscess collections. These 2 p atients
FIGURE 6. Example of good response to treatment (patient 8). A, Axial short tau inversion recovery (STIR) image at baseline shows several sinus tracts in the natal cleft and perineal region bilaterally (arrows) with labial edema (*). B, On follow-up MRI, there is resolution of the sinus tracts with some residual labial edema noted (*).
770
Griffin et al: MRI of Anal Hidradenitis Suppurativa
FIGURE 7. Example of poor response to treatment (patient 3). A, Axial short tau inversion recovery (STIR) image at baseline shows several small tracts in the perineal region, natal cleft, and left gluteal region (arrows). B, Axial STIR at follow-up shows increase in size and number of sinus tracts (arrows) and development of small left gluteal abscesses (*).
showed little interval change. In the 4 patients who received infliximab, follow-up MRI studies demonstrated an improvement in anogenital disease in 3 of 4 of these patients and worsening disease in the remaining patient. The decision for follow-up imaging was based on a number of factors, including worsening patient symptomatology and a preference by the clinician and patient for repeat imaging to assess response to treatment rather than examination under anaesthetic. As our confidence with using STIR MRI in anogenital HS has developed, clinician acceptance over the years for this imaging technique has also increased. In addition, 5 of the 18 patients in our cohort were on infliximab (patients 3, 6, 8, 10, and 12). Because this is an off-label indication, follow-up MRI to assess response was deemed appropriate in the majority of these patients. It was regarded that a reduction in the number and signal intensity of sinus tracts on follow-up STIR MRI indicated reducing activity of disease, although no direct comparison with clinical outcome was made. This correlation has been shown previously in a study by Savoye-Collet et al25 in patients with perianal Crohn’s disease treated with biologics. They demonstrated that T2 hyperintensity on follow-up MRI decreased in patients demonstrating a clinical response or in remission. They showed that loss of high T2 signal preceded a lack of enhancement on MRI.
CONCLUSION This article describes the MRI distribution of disease in anogenital HS in the largest cohort of patients to be published in the literature to date. This study shows that multiple sinus tracts are usually present characteristically affecting the perianal, anterior perineal, inguinal, and gluteal regions. Subcutaneous edema is a prominent feature. Anal sphincter and levator plate involvement only occurs in a minority of patients. MRI can help define the extent
of disease and associated complications and may potentially determine whether medical therapy needs to be escalated or whether surgical intervention is required. Given that anogenital HS is a chronic and challenging condition to treat, MRI can be a useful tool in the noninvasive management of such patients. REFERENCES 1. van der Zee HH, van der Woude CJ, Florencia EF, Prens EP. Hidradenitis suppurativa and inflammatory bowel disease: are they associated? Results of a pilot study. Br J Dermatol. 2010;162:195–197. 2. Church JM, Fazio VW, Lavery IC, Oakley JR, Milsom JW. The differential diagnosis and comorbidity of hidradenitis suppurativa and perianal Crohn’s disease. Int J Colorectal Dis. 1993;8:117–119. 3. Kelly AM, Cronin P. MRI features of hidradenitis suppurativa and review of the literature. AJR Am J Roentgenol. 2005;185:1201–1204. 4. Taylor SA, Halligan S, Bartram CI. Pilonidal sinus disease: MR imaging distinction from fistula in ano. Radiology. 2003;226:662–667. 5. Kataoka M, Kido A, Koyama T, et al. MRI of the female pelvis at 3T compared to 1.5T: evaluation on high-resolution T2-weighted and HASTE images. J Magn Reson Imaging. 2007;25:527–534. 6. Lee VS, Hecht EM, Taouli B, Chen Q, Prince K, Oesingmann N. Body and cardiovascular MR imaging at 3.0 T. Radiology. 2007;244:692–705. 7. Velpeau A. Dictionnaire de Medicine, un Repertoire General des Sciences Medicales sons le Rapport Theorique et Pratique [in French]. 2nd ed. Paris, France: Z Bechet Jeune; 1839. 8. Schiefferdecker B. In: Die Hautdrusen des Menschen und der Saugetiere, ihre Histologishe und Rassenanatomische Bednutung Sowie die Muscularis Sexualis [in German]. Stuttgart, Germany: Schweizerbart; 1922. 9. Sellheyer K, Krahl D. “Hidradenitis suppurativa” is acne inversa! An appeal to (finally) abandon a misnomer. Int J Dermatol. 2005;44:535–540. 10. Jansen I, Altmeyer P, Piewig G. Acne inversa (alias hidradenitis suppurativa). J Eur Acad Dermatol Venereol. 2001;15:532–540.
Diseases of the Colon & Rectum Volume 57: 6 (2014)
11. Yu CC, Cook MG. Hidradenitis suppurativa: a disease of follicular epithelium, rather than apocrine glands. Br J Dermatol. 1990;122:763–769. 12. Jemec GB, Hansen U. Histology of hidradenitis suppurativa. J Am Acad Dermatol. 1996;34:994–999. 13. Revuz JE, Canoui-Poitrine F, Wolkenstein P, et al. Prevalence and factors associated with hidradenitis suppurativa: results from two case-control studies. J Am Acad Dermatol. 2008;59:596–601. 14. Jemec GB, Heidenheim M, Nielsen NH. The prevalence of hidradenitis suppurativa and its potential precursor lesions. J Am Acad Dermatol. 1996;35:191–194. 15. Sartorius K, Emtestam L, Jemec GB, Lapins J. Objective scoring of hidradenitis suppurativa reflecting the role of tobacco smoking and obesity. Br J Dermatol. 2009;161:831–839. 16. Pink AE, Simpson MA, Brice GW, et al. PSENEN and NCSTN mutations in familial hidradenitis suppurativa (acne inversa). J Invest Dermatol. 2011;131:1568–1570. 17. Maclean GM, Coleman DJ. Three fatal cases of squamous cell carcinoma arising in chronic perineal hidradenitis suppurativa. Ann R Coll Surg Engl. 2007;89:709–712. 18. Matusiak Ł, Bieniek A, Szepietowski JC. Hidradenitis suppurativa markedly decreases quality of life and professional activity. J Am Acad Dermatol. 2010;62:706–708.e1.
771
19. Revuz J. Hidradenitis suppurativa. J Eur Acad Dermatol Venereol. 2009;23:985–998. 20. Kim W, Rosen MA, Langer JE, Banner MP, Siegelman ES, Ramchandani P. US MR imaging correlation in pathologic conditions of the scrotum. Radiographics. 2007;27:1239–1253. 21. Nadgir R, Rubesin SE, Levine MS. Perirectal sinus tracks and fistulas caused by hidradenitis suppurativa. AJR Am J Roentgenol. 2001;177:476–477. 22. Culp CE. Chronic hidradenitis suppurativa of the anal canal: a surgical skin disease. Dis Colon Rectum. 1983;26:669–676. 23. Endo Y, Tamura A, Ishikawa O, Miyachi Y. Perianal hidradenitis suppurativa: early surgical treatment gives good results in chronic or recurrent cases. Br J Dermatol. 1998;139: 906–910. 24. Haggett PJ, Moore NR, Shearman JD, Travis SP, Jewell DP, Mortensen NJ. Pelvic and perineal complications of Crohn’s disease: assessment using magnetic resonance imaging. Gut. 1995;36:407–410. 25. Savoye-Collet C, Savoye G, Koning E, Dacher JN, Lerebours E. Fistulizing perianal Crohn’s disease: contrast-enhanced magnetic resonance imaging assessment at 1 year on maintenance anti-TNF-alpha therapy. Inflamm Bowel Dis. 2011;17:1751–1758.
Hidradenitis Suppurativa: MRI Features in Anogenital Disease Nyree Griffin, M.D., F.R.C.R.1 • Andrew B. Williams, M.D., F.R.C.S.2 Simon Anderson, M.D., F.R.C.P.3 • Peter M. Irving, M.A., M.D., F.R.C.P.3 Jeremy Sanderson, M.D., F.R.C.P.3 • Nemesha Desai, B.Sc., F.R.C.P.4 Vicky Goh, M.D., F.R.C.R.1,5 1 Department of Radiology, Guy’s and St Thomas’ National Health Service Foundation Trust, London, United Kingdom 2 Department of General Surgery, Guy’s and St Thomas’ National Health Service Foundation Trust, London, United Kingdom 3 Department of Gastroenterology, Guy’s and St Thomas’ National Health Service Foundation Trust, London, United Kingdom 4 Department of Dermatology, Guy’s and St Thomas’ National Health Service Foundation Trust, London, United Kingdom 5 Division of Imaging Sciences and Biomedical Engineering, King’s College London, London, United Kingdom
BACKGROUND: Hidradenitis suppurativa is a rare chronic inflammatory disorder of apocrine gland-bearing skin, which commonly affects the anogenital region. There has been very little literature to date on the MRI appearances of anogenital hidradenitis suppurativa. OBJECTIVE: The aim of this study was to assess the MRI features of anogenital hidradenitis suppurativa in the largest cohort of patients to be published to date. DESIGN: After an institutional review board waiver, patients
with hidradenitis suppurativa who were undergoing MRI for anogenital disease between 2005 and 2012 were identified from our institutional database. The MRI appearances were recorded by 2 radiologists in consensus, blinded to clinical details. Location of disease, number of tracts, presence of anal fistula, and supralevator involvement were recorded. Patient demographics were also noted. SETTINGS: This study was conducted at the Department
of Radiology, Guy’s and St Thomas’ National Health Service Foundation Trust.
Financial Disclosure: None reported. Poster presentation at the meeting of the European Society of Gastrointestinal and Abdominal Radiology, Edinburgh, United Kingdom, June 12 to 15, 2012. Correspondence: Nyree Griffin, M.D., Department of Radiology, Guy’s and St Thomas’ National Health Service Foundation Trust, Westminster Bridge Road, London SE1 7EH, United Kingdom. E-mail: [email protected] Dis Colon Rectum 2014; 57: 762–771 DOI: 10.1097/DCR.0000000000000131 © The ASCRS 2014
762
PATIENTS: Patients included were those undergoing MRI for anogenital disease in hidradenitis suppurativa between 2005 and 2012. MAIN OUTCOME MEASURES: The distribution of sinus tracts in anogenital hidradenitis suppurativa on MRI was measured. RESULTS: Thirty-one MRIs were performed in 18 patients (15 men; mean age, 46 years). On the baseline MRI, multiple tracts were seen in the natal cleft (16/18; 83%), the perianal (12/18; 61%), the perineal (13/18; 56%), and the gluteal (8/18; 44%) regions. A communication with the anal canal was present in only 4 patients. Three patients had supralevator extension. Seven patients had follow-up MRIs with variable response to interval treatment: 3 of 7 showed responding disease, 3 of 7 showed stable disease, and 1 of 7 showed progressive disease. LIMITATIONS: This study was limited by its relatively small cohort of patients. CONCLUSIONS: In hidradenitis suppurativa, anogenital disease is usually subcutaneous but extensive, with only a minority of patients demonstrating deeper involvement. MRI may help define the extent of anogenital disease and assess response to treatment. KEY WORDS: Anogenital; Hidradenitis suppurativa; MRI; Sepsis; Therapy.
H
idradenitis suppurativa (HS), also known as acne inversa, is a rare chronic inflammatory skin disease that involves apocrine gland-bearing sites. It has a predilection for the axillary and anogenital regions but can also involve the submammary, intermammary, periumbilical, and retroauricular skin. The condition can Diseases of the Colon & Rectum Volume 57: 6 (2014)
763
Diseases of the Colon & Rectum Volume 57: 6 (2014)
TABLE 1. Parameters for MRI protocol Parameter TR/TE, ms TI, ms Echo train length Flip angle, ° Field of view, mm No. of signal averages Section thickness, mm Intersection gap, mm Matrix size No. of sections per stack Bandwidth, Hz
Sagittal STIR
Axial oblique STIR
Coronal oblique STIR
5380/28 130 7 150 300 2 3 1 320 × 100 23 130
5380/28 130 7 150 300 2 4 1 320/100 23 130
5380/28 130 7 150 300 2 3 1 320/100 23 130
TE = echo time; TR = repetition time; TI = inversion time; STIR = short tau inversion recovery.
vary in severity and extent. Smoking is a recognized association. When HS affects the anogenital region, it may present with perianal pain, swelling, pruritus, bleeding, or purulent discharge. It can give rise to inflammatory abscesses and sinus tract formation. The clinical picture may overlap with Crohn’s disease, which may coexist in 17% to 40% of patients with HS.1,2 In the past decade, MRI has become the imaging modality of choice in the characterization of anal fistulas and the delineation of perianal disease because of its superb contrast resolution and multiplanar capability. It has been used mostly in patients with idiopathic anal cryptoglandular sepsis or perianal Crohn’s disease. However, MRI may also have a role in HS. In our institution, a tertiary dermatology referral center, MRI has been used increasingly to assess the extent of anogenital disease to direct and facilitate management in patients with HS. The role of MRI in HS may help to define disease extent, to assess for fistulizing disease, and to exclude Crohn’s disease. Yet, to date, there have been little published data to support its use, with only 1 case report describing the MRI features of perianal HS.3 The purpose of this article is to define the typical MRI appearances at presentation and after therapy in a cohort of patients with anogenital HS.
MATERIALS AND METHODS Patients
Institutional review board waiver was obtained for this retrospective analysis, with no need for informed consent. From the institutional information database, a search was carried out for all patients with the established diagnosis of HS undergoing MRI for the evaluation of anogenital disease between January 2005 and January 2012. Patients before 2005 were excluded, because no imaging studies were archived on the institution picture and archive communication system. The search terms “hidradenitis” and “pelvic MRI” were used. This yielded 18 patients (15 men and 3 women; mean age, 46 years; age range, 24–65 years). Eleven of these patients had 1 MRI study between 2005
and 2012, whereas the remaining 7 patients had between 2 and 4 follow-up MRI studies. Using the hospital electronic patient record database, the patient demographics were recorded for each of these 18 patients. MRI Protocol
MRI was performed on a 1.5-T scanner (Siemens Avanto, Erlangen, Germany). A phased-array surface coil was used with patients imaged in the supine position. No additional patient preparation was required. The protocol included a 3-plane localizer, followed by a sagittal short tau inversion recovery (STIR) sequence through the pelvis. Using the sagittal localizer, axial and coronal oblique (STIR) images were obtained, aligned perpendicular and parallel to the longitudinal axis of the anal canal. Imaging parameters are provided in Table 1. Image Review
Two experienced radiologists with a subspecialty interest in gastrointestinal MRI (>10 years of experience) reviewed all of the baseline and follow-up MRI studies in consensus, using a picture and archive communication system imaging workstation (Centricity, GE Healthcare, Little Chalfont, United Kingdom). Hence, all of the studies were reviewed by both radiologists reading the images together at the same time, with any difference in opinion over the distribution of tracts resolved by consensus. Inflammatory change secondary to HS was defined as high signal intensity seen in the soft tissues on STIR images separate to the blood vessels. The presence of inflammatory involvement was recorded as being absent or present for the following anatomic regions: 1) the perianal region, 2) the natal cleft, 3) the anterior perineal region, 4) the ischioanal fossa, 5) the inguinal region, 6) the gluteal region, 7) the coccygeal/ sacral region, and 8) the supralevator space. Definitions for these anatomic regions were as follows4: 1) the perianal region was defined as the superficial area within 2 cm of the anal orifice; 2) the natal cleft was defined as the midline tissues and adjacent buttock ≥2 cm posterior to the anal orifice; 3) the anterior perineal region was defined as the superficial soft tissues, ≥2 cm, anterior
764
Griffin et al: MRI of Anal Hidradenitis Suppurativa
TABLE 2. Patient demographics Patient 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Age, y
Sex
Ethnicity
Clinical sites of involvement
Smoker
Crohn’s disease
44 48 43 65 51 40 59 31 45 56 35 25 24 47 58 45 65 32
M M M F M M F F M M M M M M M M M M
Black, African White Black, Caribbean White White Black, Caribbean White White White White White Black, African Black, African Black, African White White White Black, Caribbean
Buttocks, perineum, head and neck, axillas Buttocks, perineum, head and neck Buttocks, scrotum, groin Groin, axillas Perineum, axillas Buttocks, scrotum, groin, penis Buttocks, thighs, perianal, anorecto-vaginal fistula Vulva, submammary Groin, buttock Groin, perianal, perineum, axillas Buttock, groin, scrotum, axillas Buttock, groin, scrotum, perineum, perianal Axillas, groin, scrotum, perineum, buttocks Groin, scalp, scrotum, perianal Groin, scrotum, perianal, thigh Perineum, buttocks, scrotum, groin, axillas Perianal, scrotum, buttocks Perineum, buttocks, axillas
Current Current Current Current Current Current Never Never Ex Current Current Never Current Current Current Current Current Never
No No No No No No No Yes No No No No No No Yes No Yes No
M = male; F = female.
to the anal orifice (including scrotum and vulva); 4) the ischioanal fossa was defined as the tissue bounded by the inferior aspect of the levator plates superiorly, the anal sphincter complex medially, the level of the inferior border of the subcutaneous external anal sphincter inferiorly, and obturator internus laterally; 5) the inguinal region was defined as the subcutaneous tissue within both groins; 6) the gluteal region was defined as the subcutaneous tissue overlying the gluteal muscles; 7) the coccygeal/sacral region was defined as the subcutaneous tissue overlying the bony coccyx and/or sacrum; and 8) the supralevator space was defined as the perirectal tissue within the pelvis above the levator plates. When inflammation was present, its morphology was classified as either a tract (defined as a longitudinal tubular high-signal structure on MRI) or as an abscess (defined as a localized spherical collection). When applicable, it was noted whether single or multiple tracts/abscesses were present for each site. In addition, it was also noted whether inflammatory change involved the anal sphincter complex and whether striated muscle in the form of either external anal sphincter or levator plate was breached. The site and level of any enteric openings were recorded if present. The presence of inguinal lymphadenopathy and myositis (defined as high signal on STIR in the gluteus maximus muscle) was also noted. The presence and site of subcutaneous edema within the tissues was also recorded, defined as diffuse subcutaneous high signal on STIR associated with skin thickening.
RESULTS Clinical Findings
Table 2 summarizes the patient demographics, the clinical sites of involvement, the patient smoking status, and the
presence of synchronous Crohn’s disease (established by biopsy, histology, and imaging). As can be seen in Table 2, the majority of patients with anogenital HS were men (n = 15 [83%]). Patients were either white (n = 11/18 [61%]) or Afro-Caribbean (n = 7 [39%]). The diagnosis of synchronous Crohn’s disease was established in only 3 patients. Thirteen patients had had previous perianal surgery, usually involving incision and drainage of abscesses or laying open of anogenital tracts. All of the patients had been treated with one or more courses of antibiotics for their HS. A few patients had received steroids (n = 6/18), infliximab (n = 5/18), or immunomodulators (n = 3/18). Imaging Features Baseline
Table 3 shows the distribution of disease on baseline MRI. At baseline, the most common sites of involvement (in decreasing frequency) were the natal cleft (n = 16/18), the perianal region (n = 12/18), the anterior perineal region (Fig. 1; n = 13/18) and the gluteal region (n = 8/18). Multiple tracts were more characteristic at these sites than solitary tracts, with discrete abscesses being most commonly seen in the gluteal region (n = 3/18). Involvement of the coccygeal/sacral region (n = 5/18), ischioanal fossa (n = 3/18; Fig. 2A), inguinal region (n = 2/18), and supralevator space (n = 3/18) was less frequent. An enteric communication with the anal canal was not common, only seen in 4 patients (Fig. 2B). In these patients, the internal opening was at the dentate line. Three patients had tracts that extended through the levator plate (Figs. 3A and B; patients 7, 10, and 17), with 1 of these patients also having a suspected rectovaginal fistula (patient 7; Fig. 3C). Only 1 of these 7 patients had an established diagnosis of Crohn’s disease. Subcutaneous edema (n = 10/18) was a
765
frequent finding with reactive inguinal lymphadenopathy (Fig. 4) sometimes demonstrated (n = 5/18). The presence of a high signal within the gluteus maximus muscle separate to the tracts was seen in 3 patients, suggestive of myositis (Fig. 5A). Follow-up
T = single tract; TT = multiple tracts; AA = multiple abscesses; LN+ = inguinal lymphadenopathy; IAS = internal anal sphincter; EAS = external anal sphincter.
No No Gluteal edema No No Gluteal edema No No Scrotal edema No No Gluteal/perineal edema Dentate line, 6 o’clock No Dentate line, 3, 5 and 7’o’clock EAS Gluteal, scrotal edema, LN+, gluteal myositis No Levator Rectovaginal fistula, myositis No No Labial/inguinal edema No No Dentate line, 6 o’clock EAS, Levator LN+, gluteal myositis No No Gluteal edema No No LN+ Dentate line, 6 o’clock EAS LN+ No No No No Gluteal edema No No Gluteal edema No Levator No No Gluteal edema, LN+ No No No No Yes Yes No No No No No No No No No No No No No No TT TT No TT, AA No No No TT TT No TT, AA No No TT No TT, AA No No No No No No TT No No T No No No No No No T No No No No No No No No TT No No No No No No No No No TT TT T TT TT No No No TT T T No TT TT TT TT TT No TT, AA No No No No No TT No No No TT TT No TT No No TT No No TT TT TT TT No TT TT TT T TT TT TT TT TT TT TT No TT No No No No No TT TT No No T No No No No No No No No TT TT TT No T TT TT TT No TT TT No TT TT No No TT No 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Natal Coccygeal/ Anterior Patient Perianal Ischioanal cleft sacral Perineal Inguinal Supralevator Gluteal Intersphincteric
TABLE 3. Distribution of tracts on baseline MRI
Enteric opening
Striated muscle
Additional features
Diseases of the Colon & Rectum Volume 57: 6 (2014)
Table 4 shows the change in distribution (if any) on follow-up MRI studies. Follow-up MRI studies were performed in 7 of the 18 patients, with all but 2 undergoing interval medical therapy. A variable response was seen. In 3 patients the number of tracts either decreased in number with reduced signal intensity (Fig. 5) or resolved completely (Fig. 6). The number of tracts was unchanged in 3 patients, although the tracts in 1 of these patients demonstrated lower signal intensity on follow-up, and the gluteal abscesses had resolved. Appearances worsened in 1 patient despite ongoing medical therapy, with greater fluid distension of the tracts and the development of gluteal abscesses (Fig. 7).
DISCUSSION Our results show that anogenital HS typically demonstrates extensive sinus disease on STIR MRI, most commonly affecting the natal cleft, perianal, and perineal regions. Anal sphincter and supralevator involvement only occurs in the minority of patients. Response to treatment as assessed on MRI is variable. The STIR sequence used in our MR protocol is an inversion recovery pulse sequence, which suppresses the signal from fat. It is commonly used in perianal fistula imaging, where tracts appear of high signal intensity against the low signal background. Some centers make use of T2-weighted fat-saturated sequences to provide similar information. The use of a high-resolution T2weighted sequence without fat suppression is not routinely used in fistula imaging but can be helpful in providing high-resolution images of the sphincter complex and as an adjunct to the STIR images if scanning time allows. Gadolinium-enhanced T1 fat-saturated images may also help to identify active disease (where the walls of the tracts and abscesses enhance). However, we very rarely include this in our protocol, because STIR sequences are considered sufficient; intravenous gadolinium is best reserved for complex patients who have been operated on previously. All of our studies were performed on a 1.5-T magnet because of local availability. Higher-strength magnets (eg, 3 T) can be used for pelvic imaging with the greater signal:noise ratio resulting in shorter acquisition times and/or higher spatial resolution, with the latter theoretically leading to greater conspicuity of smaller fistula tracts.5 However, disadvantages include limited availability at present and increased susceptibility artifacts, magnetic field
766
FIGURE 1. Sagittal short tau inversion recovery (STIR) image shows extensive perineal sinus disease (arrows; patient 5); sinus tracts are seen overlying the dorsum of the penis too (arrowheads).
inhomogeneity, and specific absorption rates.6 In practice, the 1.5-T magnet gives sufficiently high-resolution images for diagnostic purposes and is the most commonly used field strength for MRI in clinical centers. HS is a rare inflammatory dermatologic disease. Velpeau7 originally described HS in 1839. In 1922, Schiefferdecker,8 ascribed its etiology to the apocrine glands, which are compound sweat glands that extend through the dermis into the subcutaneous tissue and drain into hair follicles. However, relatively recent histologic studies have demonstrated that HS is an inflammatory disease
Griffin et al: MRI of Anal Hidradenitis Suppurativa
originating from the hair follicle, with inflammation of the apocrine glands occurring as a secondary phenomenon.9–12 Keratin plugging of the follicular unit and subsequent expansion and rupture of the hair follicle lead to spillage of keratin into the surrounding tissue, resulting in an intense inflammatory response. Bacterial superinfection occurs with pathogens including streptococci and staphylococci. In perianal disease, Escherichia coli, Klebsiella, Proteus, and anaerobes may be seen. HS typically develops after puberty, with peak onset in the early 20s, although prepubescent and postmenopausal cases have been described. The prevalence is estimated at 1% of the general European population,13 increasing to 4% in young adult women.14 There is a male predominance of anogenital disease, as seen in our study, although overall the condition is 3 times more common in women. It is associated with smoking and obesity.15 Fourteen of our patients were either active (n = 13) or former (n = 1) cigarette smokers. One third of cases are familial, commonly follow an autosomal dominant mode of inheritance, and have been linked with mutations of the γ-secretase complex.16 Longstanding disease may result in the development of squamous cell carcinoma, particularly at perianal sites.17 Other complications include the formation of anal, urethral, and rectal fistulae and strictures; anogenital lymphoedema; and inflammatory arthropathy. Clinically, initial presentation is with recurrent tender cutaneous nodules, abscesses, and draining sinus tracts. With time, sinus tracts extend to the surrounding tissue, resulting in large areas of chronic inflammation and fibrosis of the skin and associated structures. Severity of disease is variable, but this can be a very debilitating condition.18 The diagnosis is made on clinical grounds,19 when 3 criteria are fulfilled, as follows: 1) typical lesions are present,
FIGURE 2. A, Coronal short tau inversion recovery (STIR) image shows multiple sinus tracts in the ischioanal fossa (arrowheads) and left gluteal region (arrow; patient 6). B, Axial STIR image in the same patient shows transsphincteric fistula with internal opening at the 6 o’clock position at the dentate line (white arrow); there are multiple sinus tracts (arrowheads) in the natal cleft and gluteal region (both subcutaneous and within the muscle) associated with subcutaneous edema.
Diseases of the Colon & Rectum Volume 57: 6 (2014)
767
FIGURE 3. A, Coronal short tau inversion recovery (STIR) image (patient 7) shows a defect in the left ischioanal fossa from previous surgery (*) and tract extending above the left levator plate (arrow). B, Axial STIR in same patient shows supralevator horseshoe around the lower rectum, extending posterior to the vagina (V) with secondary tract seen extending anterolateral to the vagina. Urethra (U) has been labeled. C, Sagittal STIR in same patient shows suspected rectovaginal fistula (arrow) between the anterior wall of the rectum (R) and posterior wall of the vagina (V). Gluteal sinus disease is also seen (arrowhead).
that is, deep-seated painful nodules, abscesses, draining sinus tracts, bridged scars, and tombstone comedos; 2) the patient demonstrates typical topography of involvement, that is, axillas, groin, perineal and perianal region, buttocks, and inframammary and intermammary folds; and 3) disease shows chronicity and recurrence. Histologically, an extensive perifolliculitis is seen.9 Our study had several limitations. There were only a relatively small number of patients (n = 18), but this reflects the relative rarity of this condition and the fact that only the most severe cases were referred for anogenital imaging. It is still the largest cohort to be described in the literature, which only has isolated case reports.3 All patients had been on, or were continuing, with some form
of medical therapy, which again could have influenced anogenital appearances on the MRI study. Thirteen patients had undergone perianal surgery (usually consisting of incision and drainage of abscesses). This could account for the relatively low percentage of patients with abscesses seen on baseline and follow-up MRI studies. However, imaging with anogenital HS while patients are being treated is normal practice. Finally, only 7 patients had a followup MRI study, and because of the variety of individual medical and surgical interventions in this cohort, it was not possible to directly correlate appearances on follow-up MRI with specific therapies. To our knowledge, there has only been 1 case report in the literature describing the MRI features of anogenital
768
FIGURE 4. Coronal short tau inversion recovery (STIR) image (patient 3) shows multiple enlarged reactive inguinal nodes (arrows), perineal sinus tracts (arrowhead), and scrotal edema (*).
HS.3 MRI of the pelvis revealed induration of the subcutaneous tissues over the medial aspects of the thighs, perineum, and mons pubis with multiple subcutaneous tracts and small abscesses. Bilateral enlarged inguinal nodes were seen. There were no fistulous communications with the bladder, urethra, or rectum in their patient. Abscesses were noted to be remote from both the rectum and anus. The distribution of disease in this case report is similar to our cohort, where we demonstrated widespread sinus dis-
Griffin et al: MRI of Anal Hidradenitis Suppurativa
ease characteristically affecting the perianal, anterior perineal, inguinal, and gluteal regions. Subcutaneous edema and inguinal lymphadenopathy were also demonstrated in our cohort. Similarly, in a pictorial review of pathologic conditions affecting the scrotum,20 the authors described the appearances of scrotal HS on MRI, characterized by edema and enhancing microabscesses. There has only been 1 other case report that has described the appearances on barium enema of a patient with perianal HS. Multiple sinus tracks and fistulas were seen arising from the posterior wall of the distal rectum on barium enema.21 This rectal involvement is unusual in HS. We demonstrated anal sphincter involvement in 22% (n = 4/18) of our patients on MRI with the internal opening centered on the dentate line. It has been suggested that fistulas to the anal canal in HS should extend only as far as the lower part of the anal canal, because this is where apocrine glands can be found.22 However, in 3 of our patients, 1 of whom had Crohn’s disease, tracts breached the levator plate; 1 of these patients also had a suspected rectovaginal fistula. Fistulizing disease in HS, although rare, is well recognized in the dermatologic literature.23 Although there may be an overlap in the clinical picture of anogenital HS and Crohn’s disease, the distribution of perianal disease in Crohn’s disease is different from HS. In Crohn’s disease, fistula tracts and/or abscesses are usually seen primarily involving the anal sphincter complex.24 Sinus involvement of the anterior perineum and gluteal and inguinal regions is not typically seen in Crohn’s disease, although the development of anovulvar fistulas is well recognized. Extension into the ischioanal fossa, na-
FIGURE 5. Example of partial response to treatment (patient 10). A, Axial short tau inversion recovery (STIR) image at baseline shows multiple high signal sinus tracts in the natal cleft, perianal region, ischioanal fossa (arrowheads), and gluteal region (arrows). High signal is also seen in the gluteal muscles (*) in keeping with myositis. B, Axial STIR image on follow-up MRI shows a reduction in number and signal intensity of the tracts.
769
Diseases of the Colon & Rectum Volume 57: 6 (2014)
TABLE 4. Distribution of tracts on follow-up MRI scans Interval of second MRI from baseline
Appearances on second MRI
Interval of third MRI from baseline
Appearances on third MRI
2
6 mo
3
9 mo
6
25 mo
Reduced number and signal intensity of tracts in perianal region, no edema Same distribution, but now greater fluid distension, multiple abscesses (left), LN+ No change
31 mo
7 8
18 mo 6 mo
Same distribution, reduced signal intensity, gluteal abscesses and edema resolved No change Labial and inguinal edema only, no tracts
10
10 mo
15
9 mo
Patient
No change Less labial edema, otherwise no change Reduced number and signal intensity of tracts Increased subcutaneous edema, otherwise no change
27 mo 20 mo
18 mo
No change
Interval of fourth MRI from baseline
Appearances on fourth MRI
36 mo
No change
19 mo
No change
LN+ = inguinal lymphadenopathy
tal cleft, and supralevator space are more characteristic in Crohn’s disease.24 There has been no previous literature describing the MRI changes that occur after medical and/or surgical therapy in HS. A variety of medical and surgical treatments may be offered to patients with HS. Prolonged topical and systemic antibiotics are usually first-line medical therapies, with second-line therapies including systemic retinoids, antiandrogens in women, cyclosporin, and more recently tumor necrosis factor-α inhibitors (biologics), such as infliximab (although the use of the latter in HS is
presently unlicensed). Surgical treatments for HS include incision and drainage of abscesses, laying open of tracts, and limited and wide local excision. In the 7 patients with follow-up MRI in our study, a variable response was seen, with some patients showing a reduction in the number and signal intensity of sinus tracts, whereas others remained stable or developed worsening disease. Interval medical therapy (in the form of antibiotics, steroids, infliximab, or immunomodulators) was given to all but 2 of these patients who underwent incision and drainage of abscess collections. These 2 p atients
FIGURE 6. Example of good response to treatment (patient 8). A, Axial short tau inversion recovery (STIR) image at baseline shows several sinus tracts in the natal cleft and perineal region bilaterally (arrows) with labial edema (*). B, On follow-up MRI, there is resolution of the sinus tracts with some residual labial edema noted (*).
770
Griffin et al: MRI of Anal Hidradenitis Suppurativa
FIGURE 7. Example of poor response to treatment (patient 3). A, Axial short tau inversion recovery (STIR) image at baseline shows several small tracts in the perineal region, natal cleft, and left gluteal region (arrows). B, Axial STIR at follow-up shows increase in size and number of sinus tracts (arrows) and development of small left gluteal abscesses (*).
showed little interval change. In the 4 patients who received infliximab, follow-up MRI studies demonstrated an improvement in anogenital disease in 3 of 4 of these patients and worsening disease in the remaining patient. The decision for follow-up imaging was based on a number of factors, including worsening patient symptomatology and a preference by the clinician and patient for repeat imaging to assess response to treatment rather than examination under anaesthetic. As our confidence with using STIR MRI in anogenital HS has developed, clinician acceptance over the years for this imaging technique has also increased. In addition, 5 of the 18 patients in our cohort were on infliximab (patients 3, 6, 8, 10, and 12). Because this is an off-label indication, follow-up MRI to assess response was deemed appropriate in the majority of these patients. It was regarded that a reduction in the number and signal intensity of sinus tracts on follow-up STIR MRI indicated reducing activity of disease, although no direct comparison with clinical outcome was made. This correlation has been shown previously in a study by Savoye-Collet et al25 in patients with perianal Crohn’s disease treated with biologics. They demonstrated that T2 hyperintensity on follow-up MRI decreased in patients demonstrating a clinical response or in remission. They showed that loss of high T2 signal preceded a lack of enhancement on MRI.
CONCLUSION This article describes the MRI distribution of disease in anogenital HS in the largest cohort of patients to be published in the literature to date. This study shows that multiple sinus tracts are usually present characteristically affecting the perianal, anterior perineal, inguinal, and gluteal regions. Subcutaneous edema is a prominent feature. Anal sphincter and levator plate involvement only occurs in a minority of patients. MRI can help define the extent
of disease and associated complications and may potentially determine whether medical therapy needs to be escalated or whether surgical intervention is required. Given that anogenital HS is a chronic and challenging condition to treat, MRI can be a useful tool in the noninvasive management of such patients. REFERENCES 1. van der Zee HH, van der Woude CJ, Florencia EF, Prens EP. Hidradenitis suppurativa and inflammatory bowel disease: are they associated? Results of a pilot study. Br J Dermatol. 2010;162:195–197. 2. Church JM, Fazio VW, Lavery IC, Oakley JR, Milsom JW. The differential diagnosis and comorbidity of hidradenitis suppurativa and perianal Crohn’s disease. Int J Colorectal Dis. 1993;8:117–119. 3. Kelly AM, Cronin P. MRI features of hidradenitis suppurativa and review of the literature. AJR Am J Roentgenol. 2005;185:1201–1204. 4. Taylor SA, Halligan S, Bartram CI. Pilonidal sinus disease: MR imaging distinction from fistula in ano. Radiology. 2003;226:662–667. 5. Kataoka M, Kido A, Koyama T, et al. MRI of the female pelvis at 3T compared to 1.5T: evaluation on high-resolution T2-weighted and HASTE images. J Magn Reson Imaging. 2007;25:527–534. 6. Lee VS, Hecht EM, Taouli B, Chen Q, Prince K, Oesingmann N. Body and cardiovascular MR imaging at 3.0 T. Radiology. 2007;244:692–705. 7. Velpeau A. Dictionnaire de Medicine, un Repertoire General des Sciences Medicales sons le Rapport Theorique et Pratique [in French]. 2nd ed. Paris, France: Z Bechet Jeune; 1839. 8. Schiefferdecker B. In: Die Hautdrusen des Menschen und der Saugetiere, ihre Histologishe und Rassenanatomische Bednutung Sowie die Muscularis Sexualis [in German]. Stuttgart, Germany: Schweizerbart; 1922. 9. Sellheyer K, Krahl D. “Hidradenitis suppurativa” is acne inversa! An appeal to (finally) abandon a misnomer. Int J Dermatol. 2005;44:535–540. 10. Jansen I, Altmeyer P, Piewig G. Acne inversa (alias hidradenitis suppurativa). J Eur Acad Dermatol Venereol. 2001;15:532–540.
Diseases of the Colon & Rectum Volume 57: 6 (2014)
11. Yu CC, Cook MG. Hidradenitis suppurativa: a disease of follicular epithelium, rather than apocrine glands. Br J Dermatol. 1990;122:763–769. 12. Jemec GB, Hansen U. Histology of hidradenitis suppurativa. J Am Acad Dermatol. 1996;34:994–999. 13. Revuz JE, Canoui-Poitrine F, Wolkenstein P, et al. Prevalence and factors associated with hidradenitis suppurativa: results from two case-control studies. J Am Acad Dermatol. 2008;59:596–601. 14. Jemec GB, Heidenheim M, Nielsen NH. The prevalence of hidradenitis suppurativa and its potential precursor lesions. J Am Acad Dermatol. 1996;35:191–194. 15. Sartorius K, Emtestam L, Jemec GB, Lapins J. Objective scoring of hidradenitis suppurativa reflecting the role of tobacco smoking and obesity. Br J Dermatol. 2009;161:831–839. 16. Pink AE, Simpson MA, Brice GW, et al. PSENEN and NCSTN mutations in familial hidradenitis suppurativa (acne inversa). J Invest Dermatol. 2011;131:1568–1570. 17. Maclean GM, Coleman DJ. Three fatal cases of squamous cell carcinoma arising in chronic perineal hidradenitis suppurativa. Ann R Coll Surg Engl. 2007;89:709–712. 18. Matusiak Ł, Bieniek A, Szepietowski JC. Hidradenitis suppurativa markedly decreases quality of life and professional activity. J Am Acad Dermatol. 2010;62:706–708.e1.
771
19. Revuz J. Hidradenitis suppurativa. J Eur Acad Dermatol Venereol. 2009;23:985–998. 20. Kim W, Rosen MA, Langer JE, Banner MP, Siegelman ES, Ramchandani P. US MR imaging correlation in pathologic conditions of the scrotum. Radiographics. 2007;27:1239–1253. 21. Nadgir R, Rubesin SE, Levine MS. Perirectal sinus tracks and fistulas caused by hidradenitis suppurativa. AJR Am J Roentgenol. 2001;177:476–477. 22. Culp CE. Chronic hidradenitis suppurativa of the anal canal: a surgical skin disease. Dis Colon Rectum. 1983;26:669–676. 23. Endo Y, Tamura A, Ishikawa O, Miyachi Y. Perianal hidradenitis suppurativa: early surgical treatment gives good results in chronic or recurrent cases. Br J Dermatol. 1998;139: 906–910. 24. Haggett PJ, Moore NR, Shearman JD, Travis SP, Jewell DP, Mortensen NJ. Pelvic and perineal complications of Crohn’s disease: assessment using magnetic resonance imaging. Gut. 1995;36:407–410. 25. Savoye-Collet C, Savoye G, Koning E, Dacher JN, Lerebours E. Fistulizing perianal Crohn’s disease: contrast-enhanced magnetic resonance imaging assessment at 1 year on maintenance anti-TNF-alpha therapy. Inflamm Bowel Dis. 2011;17:1751–1758.
