In the treatment of complex anal fistulas transplant of freshly collected autologous adipose tissue mechanically fragmented or centrifuged adipose tissue (CAT) might be an alternative to in vitro expanded autologous or allogeneic adipose-derived stem cells, showing remarkable efficacy in diverse therapeutic indications.

The aim of our study is to evaluate randomly the efficacy and safety of the use of CAT in the healing process of complex anal fistulas, except for Crohn's disease (CD) related fistulas.

This monocentric randomized controlled study was designed to test the efficacy and safety of injection of CAT for treatment of complex fistula-in ano, not associated with CD . The study was conducted at the Second Surgical Unit of the University Hospital of Ferrara, located in the North East of Italy. All consecutive patients referred to our clinic with complex anal fistula were screened. Patients were enrolled from May 2016 through May 2019.

Eligible patients were randomly assigned, according to a 1:1 scheme, to receive treatment with or without CAT (120 patients totally). The primary endpoint was defined as the proportion of patients whose fistula was completely healed within 4 weeks after surgery. Fistula healing was defined as the absence of any anal symptom, with no discharge from the fistula and a closed external opening confirmed on clinical evaluation. Fistula healing was eventually confirmed by a pelvic MRI at 3 months after surgery.Moreover, we analyzed postoperative pain measured by Visual Analogue Scale for 2 weeks after surgery to evaluate whether CAT injection could modulate pain, and recurrence at 6 months to evaluate the long-term efficacy of CAT injection.

Faecal continence was assessed before surgery and at 6-months follow-up visit using the Cleveland Clinical Florida Fecal Incontinence (CCF-FI) questionnaire. Adipose tissue from periombelical or lateral abdomen was harvested from these patients by liposuction to prepare CAT using Coleman's technique.Complex anal fistulas were treated by conventional surgery. Once the seton (present in the most of patients) was removed, after identification of internal opening, necrotic and inflamed tissues were excised using a "cone-like" fistulectomy or debrided with a wire brush to remove granulation tissue by the help of fistuloscope (VAAFT modified, without destroying the fistulous track with monopolar electrode). Before CAT injection, the internal opening was treated in two ways: by simply closing it using 2-0 Vicryl stitches on the muscular and mucosal layers or by performing a mucosal advancement rectal flap.

CAT was injected into the submucosal layer surrounding the internal opening, and around the fistula tract wall (in case of VAAFT modified) or inside the perianal wound after fistulectomy, using a 1-mL syringe equipped with a 22-gauge and 30-mm length needle, dividing injections at all quadrants of fistula wall, or distributing the CAT inside the perianal wound, at different levels by using multiple passes from the internal to the external openings until there was firm swelling surrounded the fistula tract. The external opening was approximated loosely with a suture to allow drainage of the fistula, but not extrusion of the centrifuged adipose tissue. The amount of product injected may vary depending on the total amount harvested but it should be not less than 10 ml. Patients were assessed for fistula healing, pain (VAS) and AEs at 1 week, 2 weeks, 4 weeks, 8 weeks, three months, and 6 months after treatment. Assessment consisted of clinical examination and AEs and VAS registration. Pelvic MRI was performed at 3 months after surgery. The Cleveland Clinical Florida Fecal Incontinence (CCF-FI) questionnaire was administered before surgery and at 6-months follow-up visit. At the final visit patients were also asked about their overall satisfaction with the procedure, with ratings from very satisfied to very dissatisfied.

Inclusion Criteria:

Patients aged 18 years or older with complex perianal fistulas

Exclusion Criteria:

Inflammatory bowel diseases, Crohn's Disease (CD) or Ulcerative Colitis (UC) Hepatitis B or C virus Infection with Human immunodeficiency virus Surgery for a malignant tumor during the previous 5 years (excluding in situ carcinoma) Previous pelvic radiotherapy Autoimmune disease Active tuberculosis Symptoms of septicemia Breastfeeding or pregnant women Women unwilling to use contraception during the study Psychiatric disorders Alcoholism Drug dependency Allergy to local anesthetics

De-identified individual participant data for all primary and secondary outcome measures will be made available.

Data will be available within 12 months of study completion

Data access requests will be reviewed by an external indipendent review Panel. Requestores will be required to sign a Data Access Agreement.