Magnetic Resonance Imaging for Distinguishing Perianal Hidradenitis Suppurativa from Fistulizing Crohn Disease

Mika Yamanaka-Takaichi; Mariana Yalon; Edward V Loftus; Anna B Grimaldo; Austin Todd; David Anthony Wetter; Hamid R Tizhoosh; Eric J Dozois; Afsaneh Alavi; Eric C Ehman

doi:10.1159/000547570

Magnetic Resonance Imaging for Distinguishing Perianal Hidradenitis Suppurativa from Fistulizing Crohn Disease

Dermatology. 2025;241(5-6):437-446. doi: 10.1159/000547570. Epub 2025 Jul 25.

Authors

Affiliations

¹ Department of Dermatology, Mayo Clinic, Rochester, Minnesota, USA.
² Department of Radiology, Mayo Clinic, Rochester, Minnesota, USA.
³ Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, Minnesota, USA.
⁴ Mayo Clinic Alix School of Medicine, Mayo Clinic College of Medicine and Science, Rochester, Minnesota, USA.
⁵ Division of Clinical Trials and Biostatistics, Mayo Clinic, Rochester, Minnesota, USA.
⁶ Artificial Intelligence and Informatics, Mayo Clinic, Rochester, Minnesota, USA.
⁷ Kimia Lab, University of Waterloo, Waterloo, Ontario, Canada.
⁸ Division of Colon and Rectal Surgery, Mayo Clinic, Rochester, Minnesota, USA.
⁹ Department of Dermatology, Mayo Clinic, Rochester, Minnesota, USA, alavi.afsaneh@mayo.edu.

PMID: 40716427
DOI: 10.1159/000547570

Abstract

Introduction: Both hidradenitis suppurativa (HS) and Crohn disease (CD) are characterized by suppuration and granulomatous inflammation, which may result in formation of tunnels, sinus tracts, or fistulas. We sought to determine the differences in characteristics of fistulizing perianal CD vs. perianal HS on magnetic resonance imaging (MRI).

Methods: This retrospective cohort study included patients diagnosed with CD, HS, or both diseases who underwent medical examination at the Mayo Clinic between January 1, 1998, and July 31, 2021. We further selected patients who had documented perianal lesions and underwent pelvic MRI. Two abdominal radiologists blinded to clinical diagnosis reviewed the MR images for the presence or absence of 10 characteristics. For characteristics that differed significantly between patients with CD and HS, odds ratios were calculated. A radiomics-based MRI scoring system was constructed according to the odds ratios and was validated.

Results: On MRI, significantly more patients in the HS group (n = 49) than the CD group (n = 74) had subcutaneous tunnels (35% vs. 4%; p < 0.001), soft-tissue inflammation of subcutaneous tissue and skin (35% vs. 5%; p < 0.001), and inguinal lymphadenopathy (65% vs. 45%; p = 0.02). Significantly more patients in the CD group than the HS group had transsphincteric fistula (64% vs. 35%; p = 0.002), intersphincteric fistula (58% vs. 29%; p = 0.001), mesorectal lymphadenopathy (26% vs. 4%; p = 0.002), and rectal inflammation (32% vs. 12%; p = 0.01). The proposed MRI scoring system for distinguishing HS from CD had a sensitivity of 0.77, specificity of 0.80, and area under the curve of 0.84.

Conclusion: MRI may be a valuable tool in distinguishing perianal draining tunnels in HS from fistulizing perianal CD. The MRI scoring system we created could help in the clinical decision-making process. Combining MRI with other investigational and clinical findings may improve diagnostic accuracy for these challenging perianal diseases.

Keywords: Crohn disease; Fistula; Hidradenitis suppurativa; Inflammatory bowel disease; Perianal.

MeSH terms

Adult
Crohn Disease* / complications
Crohn Disease* / diagnostic imaging
Diagnosis, Differential
Female
Hidradenitis Suppurativa* / complications
Hidradenitis Suppurativa* / diagnostic imaging
Humans
Magnetic Resonance Imaging*
Male
Middle Aged
Rectal Fistula* / diagnostic imaging
Rectal Fistula* / etiology
Retrospective Studies
Young Adult