Use of microfragmented autologous adipose tissue (LIPOGEMS®) as a sphincter-saving therapy for complex anal fistulas: a single-center retrospective study Authors: Carlos Bustamante-Recuenco, Ramón Sanz-Ongil, Soledad Oliart, Aritz Equisoain-Azcona, Javier García-Quijada, Jorge González-Cavero, Daniel Arenas-Escribano, Francisco Javier Angulo Morales Published: 2026 ABSTRACT BackgroundComplex anal fistulas remain a therapeutic challenge due to high recurrence rates and the risk of incontinence associated with conventional treatments. Microfragmented autologous adipose tissue has emerged as a sphincter-sparing option with potential regenerative and anti-inflammatory properties. This study aimed to evaluate the safety and efficacy of its injection into the fistula tract combined with closure of the internal opening.MethodsA single-center, retrospective, observational study was conducted between June 2019 and May 2025. All patients with complex anal fistulas treated with microfragmented adipose tissue processed with LIPOGEMS® system were included. Demographic data, fistula characteristics, surgical history, perioperative variables, postoperative morbidity, healing rate, and hospital stay were analyzed. Follow-up visits were performed weekly in the first month, and at 3, 6 and 12 months thereafter.ResultsA total of 28 patients with complex fistula were included. Mean patient age was 51.2 years (SD: 15.3), with predominance of males (69.7%). Most fistulas were medium or high transsphincteric (85.7%) and the main location was posterior (57.1%). 75% of patients had undergone previous curative-intent procedures. Median operative time was 71 minutes, and median processed adipose volume was 14 ml. Two postoperative complications were recorded (one abdominal hematoma-CD I and one perianal abscess-CD IIIa). After a median follow-up of 11.2 months, the overall healing rate was 46.4%, higher in primary (57.1%) and posterior fistulas (56.3%). Median hospital stay was one day, and median clinical closure occurred at 2.7 months.ConclusionsLIPOGEMS® therapy appeared safe, feasible, and associated with encouraging healing outcomes, offering advantages in functional preservation and recovery. Larger prospective studies are needed to confirm these results and refine its role in fistula management. Background The incidence of anal fistula in the European population is estimated at approximately 1.69 to 1.83 cases per 10.000 individuals, with an associated incidence for cryptoglandular infection and Crohn’s disease (CD) of 0.86 and 0.76 respectively [1]. Complex anal fistulas account for approximately 33% to 42% of all cases, highlighting the clinical significance of this subgroup [2, 3]. Multiple treatments have been described, but despite the surgical advances, contemporary series report a moderate-high recurrence rate, ranging from approximately 22% to 32% in local case-series [2, 4] and showing a greater range of 4.9% to 60.7% in larger systematic reviews and meta-analysis [3, 5, 6]. Postoperative anal incontinence remains one of the most serious complications in this field, making sphincter-preserving procedures essential for this patient population. Among these less invasive procedures, the use of mesenchymal stem cells (MSCs) derived from autologous adipose tissue has gained attention over the years. By secreting mitogenic and immunomodulatory mediators, MSCs can actively shape and regulate the surrounding microenvironment, acquiring a regenerative and immunomodulatory potential that has been confirmed in both in vivo and in vitro studies [7, 8]. In the field of perianal fistula treatment, this biological effect may promote the healing of complex or refractory fistulas without compromising continence, an outcome that has been evaluated in several studies with promising results [9–12]. However, traditional MSC-based treatment involves a three-step procedure: obtention of the autologous adipose tissue, lab-processing through expansion or enzymatic digestion, and finally the injection of the final product in the fistula tract [13]. This involves a high cost, both economically and in terms of patient burden, and the completion of complex regulatory frameworks due to the adipose tissue processing [13–15]. Also, evidence about its safety and effectiveness is still scarce in anal fistulas secondary to cryptoglandular infection. In this regard, the LIPOGEMS® system enables the procedure to be completed in a single surgical intervention, as adipose tissue is processed through a rapid mechanical method [16]. It also preserves the native stromal microenvironment, potentially enhancing cell viability and therapeutic efficacy [9, 10]. In addition, it offers a cost advantage by avoiding cell expansion, with its overall cost positioned between that of simple autologous fat grafting and commercial stem cell–based products [9, 17]. Furthermore, it eliminates the need to comply with Good Manufacturing Practice (GMP) requirements [15]. In order to assess the clinical efficacy of this specific treatment we decided to perform a retrospective review since its adoption in our center. Material and methods Study design and setting This is a retrospective case series of consecutive patients treated at a single tertiary referral center between June 2019 and May 2025. The study protocol was approved by the local Institutional Review Board and conducted in accordance with the Declaration of Helsinki [18]. Outcomes The primary outcome was to determine the fistula recurrence rate in our center. The assessment of the safety of the procedure, reflected by the analysis of the postoperative complication rate, was determined as the secondary objective. Patient selection and inclusion criteria Inclusion criteria:Adults older than 18 years with complex anal fistula based on the American Society of Colon and Rectal Surgeons (ASCRS) definition: Involvement of >30% of the external anal sphincter or presence of transsphincteric tracts affecting a significant sphincter portion, suprasphincteric, extrasphincteric or horseshoe configurations; fistulas related to inflammatory bowel disease, prior radiation, malignancy, preexisting fecal incontinence, or chronic diarrhea; recurrent, branching, multiple-tract fistulas and anterior fistulas in women [17].6 months of follow-up visits completed at the time of the data collection. Adults older than 18 years with complex anal fistula based on the American Society of Colon and Rectal Surgeons (ASCRS) definition: Involvement of >30% of the external anal sphincter or presence of transsphincteric tracts affecting a significant sphincter portion, suprasphincteric, extrasphincteric or horseshoe configurations; fistulas related to inflammatory bowel disease, prior radiation, malignancy, preexisting fecal incontinence, or chronic diarrhea; recurrent, branching, multiple-tract fistulas and anterior fistulas in women [17]. Adults older than 18 years with complex anal fistula based on the American Society of Colon and Rectal Surgeons (ASCRS) definition: Involvement of >30% of the external anal sphincter or presence of transsphincteric tracts affecting a significant sphincter portion, suprasphincteric, extrasphincteric or horseshoe configurations; fistulas related to inflammatory bowel disease, prior radiation, malignancy, preexisting fecal incontinence, or chronic diarrhea; recurrent, branching, multiple-tract fistulas and anterior fistulas in women [17]. 6 months of follow-up visits completed at the time of the data collection. 6 months of follow-up visits completed at the time of the data collection. Exclusion criteria:Poorly controlled inflammatory bowel disease at time of surgery.Untreated or undrained anorectal abscess.Rectovaginal fistula.Pregnancy.Uncontrolled diabetes mellitus.Known coagulopathy or connective tissue diseases.Loss of follow-up. Poorly controlled inflammatory bowel disease at time of surgery. Poorly controlled inflammatory bowel disease at time of surgery. Untreated or undrained anorectal abscess. Untreated or undrained anorectal abscess. Rectovaginal fistula. Rectovaginal fistula. Pregnancy. Pregnancy. Uncontrolled diabetes mellitus. Uncontrolled diabetes mellitus. Known coagulopathy or connective tissue diseases. Known coagulopathy or connective tissue diseases. Loss of follow-up. Loss of follow-up. Perianal fistulas were classified as primary if no previous curative-intent surgical treatment (CIST) had been received, and as recurrent otherwise. In this last case, patients were included either in the SOG (Single Operation group) or in the ROG (Reoperated group) group according to the number of previous surgeries (one or more than one). Preoperative assessment and classification All patients underwent a standardized preoperative evaluation including clinical examination and radiological complementary test (endoanal ultrasound-EUS and/or Magnetic Resonance Imaging-MRI), which were performed in every case to define anatomy and tract complexity. Fistulas were classified according to the Park´s, St James University Hospital (SJUH) and Garg classifications [19–22]. A draining seton was placed in all patients for six weeks prior to definitive treatment. Data collection, variables and database A retrospective database was created from electronic clinical records, endoanal ultrasound reports, operative notes, and outpatient follow-up visits. Collected variables included: demographics, comorbidities, number and type of prior surgeries, pre- and postoperative Wexner incontinence score, location of external and internal fistula openings (EFO, IFO), number of tracts, intraoperative findings and perioperative complications. Follow-up and definitions Patients were followed weekly during the first month, and at 3, 6 and 12 months postoperatively. At each visit, a thorough medical history was obtained and a physical examination was performed. Clinical healing was defined as complete closure of internal and external openings with absence of discharge or drainage from the tract. Additional radiological evaluations were performed when recurrence or postoperative adverse events were suspected, including incomplete closure of the internal or external fistula openings, persistent drainage, ongoing perianal pain or discomfort, fever, perianal swelling or mass, or new-onset fecal incontinence. Fistula persistence was defined as continued drainage, failure of fistula tract closure, or the need for reintervention within 6 months. Recurrence was defined as the reappearance of any of the aforementioned clinical signs after initial healing. STROBE guidelines were followed during the project development and manuscript writing and edition [23]. Statistical analysis Statistical analysis was performed using SPSS 27.0® software (IBM, SPSS Statistics for Windows, Version 27.0. Armonk, NY: IBM Corp.). Categorical variables were presented as counts and percentages. Due to the small sample size (n = 28), Shapiro–Wilk test was used to assess the distribution of quantitative variables. These were expressed as mean and standard deviation (SD) if normally distributed, or as median, interquartile range (IQR) or range (min–max), as appropriate. For bivariate analysis, normally distributed quantitative variables were compared using Student’s t-test, whereas non-normally distributed variables were analyzed using the Mann–Whitney U test. Pre and postoperative Wexner score were compared with the non-parametrical Wilcoxon Test. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. Statistical significance was set at p < 0.05. Surgical procedure All operations were performed by experienced proctologists in collaboration with two plastic surgeons. We followed the technique reported by Naldini et al.[16], that can be resumed in five steps:1-Obtaining autologous adipose tissue. Adipose tissue was obtained from the lower or lateral abdominal region, depending on the subcutaneous fat distribution (Fig. 1A). Prior to fat extraction, the subcutaneous layer was infiltrated with a total of 360 mL of modified Klein solution (500 ml saline solution, 20 ml lidocaine and 1 ml adrenaline). It was injected using a single-use 17G cannula through a subcentimeter, suprapubic skin incision (Fig. 1B). Fat harvesting was performed using a 13G blunt-tip cannula attached to a 20-mL Vaclock® syringe (Fig. 1C-E). The abdominal incision was closed with a single, monofilament 4/0 stitch. Collaboration with Plastic Surgery department was needed for this step.Fig. 1A Liposuction area delimitation in the infraumbilical lower abdominal region. B Klein solution infiltration through a subcentimeter incision 2 cm above the pubic symphysis. C and D Liposuction performance on both sides of the marked area with a 13G blunt-tip cannula attached to a 20-mL Vaclock® syringe. E Harvested fat tissue appearance in the syringe after liposuction2-Mesenchymal stem cell isolation using the LIPOGEMS® system. The aspirated content was transferred into 10-mL syringes and left to decant (Fig. 2A and B). Afterwards, it was subsequently processed with the LIPOGEMS® device, designed to progressively microfragment adipose clusters while simultaneously removing oil residues and blood pro-inflammatory components. This step is detailed in Fig. 2(C to E). Processing was carried out entirely under continuous saline immersion to reduce mechanical stress. The resulting microfragmented adipose product was collected in a 10-mL syringe and allowed to settle so that the excess saline could decant (Fig. 2F). Finally, the prepared graft was distributed into 1-mL syringes fitted with a 22G, 30-mm needle, which would be used for the final injection in the internal opening and fistula tract.Fig. 2 A and B Transference of harvested fat tissue to 10 ml syringes, followed by decantation. C Filling of the cylinder with saline solution. Initial D and final E appearance of the cylinder content after its shaking, maneuver that allows the spheres to micro-fracture the collected fat tissue for its final use. F Extraction of the final product from the cylinder in 10 ml syringes. Afterwards, the prepared graft will be distributed into 1-mL syringes, to be used in the fistula tract treatment3-Patient positioning for fistula approach. In case of anterior fistula the patient was put in prone jackknife position. Lithotomy position with padded leg holders was adopted otherwise.4-Management of internal opening and fistula tract. First, a mucosal flap is fashioned over the internal opening using electrocautery, ensuring that it extends at least 1 cm beyond the fistula orifice (Fig. 3A, B). After, the internal orifice is closed with 2/0 polydioxanone stitches, making sure that the muscular layer is included in them (Fig. 3C and D). Finally, the flap is closed using an absorbable, 3/0, multifilament suture (Fig. 3E, F). Core-out of the extrasphinteric fistula tract is performed when possible. If not, external opening is lightly enlarged to facilitate drainage.Fig. 3A Initial view of the fistula tract canalized with a seton. B Internal orifice after performance of the mucosal flap. C and D Internal orifice closure with 2/0 polydioxanone stitches, making sure that the muscular layer is included in them. Closure of the mucosal flap can be completed in an horizontal E or vertical F direction depending on tissue tension and internal orifice characteristics5-Micro-fragmented adipose tissue infiltration. After completion of the core-out (Fig. 4A), the MSC preparation is injected along the course of the fistula tract, in the mucosal flap and in the internal opening in the muscular layer (Fig. 4B to D). The amount injected vary depending on the amount obtained and the length of the fistula tract. A volume less than 6 mL is considered insufficient for adequate treatment.Fig. 4A Final view after performance of the core-out. B Injection of the MSC preparation around the fistula tract. C and D Injection of the MSC preparation in the mucosal flap and in the internal opening and the mucosal flap approaching it from both sides of the fistula tract 1-Obtaining autologous adipose tissue. Adipose tissue was obtained from the lower or lateral abdominal region, depending on the subcutaneous fat distribution (Fig. 1A). Prior to fat extraction, the subcutaneous layer was infiltrated with a total of 360 mL of modified Klein solution (500 ml saline solution, 20 ml lidocaine and 1 ml adrenaline). It was injected using a single-use 17G cannula through a subcentimeter, suprapubic skin incision (Fig. 1B). Fat harvesting was performed using a 13G blunt-tip cannula attached to a 20-mL Vaclock® syringe (Fig. 1C-E). The abdominal incision was closed with a single, monofilament 4/0 stitch. Collaboration with Plastic Surgery department was needed for this step.Fig. 1A Liposuction area delimitation in the infraumbilical lower abdominal region. B Klein solution infiltration through a subcentimeter incision 2 cm above the pubic symphysis. C and D Liposuction performance on both sides of the marked area with a 13G blunt-tip cannula attached to a 20-mL Vaclock® syringe. E Harvested fat tissue appearance in the syringe after liposuction 1-Obtaining autologous adipose tissue. Adipose tissue was obtained from the lower or lateral abdominal region, depending on the subcutaneous fat distribution (Fig. 1A). Prior to fat extraction, the subcutaneous layer was infiltrated with a total of 360 mL of modified Klein solution (500 ml saline solution, 20 ml lidocaine and 1 ml adrenaline). It was injected using a single-use 17G cannula through a subcentimeter, suprapubic skin incision (Fig. 1B). Fat harvesting was performed using a 13G blunt-tip cannula attached to a 20-mL Vaclock® syringe (Fig. 1C-E). The abdominal incision was closed with a single, monofilament 4/0 stitch. Collaboration with Plastic Surgery department was needed for this step. Fig. 1A Liposuction area delimitation in the infraumbilical lower abdominal region. B Klein solution infiltration through a subcentimeter incision 2 cm above the pubic symphysis. C and D Liposuction performance on both sides of the marked area with a 13G blunt-tip cannula attached to a 20-mL Vaclock® syringe. E Harvested fat tissue appearance in the syringe after liposuction A Liposuction area delimitation in the infraumbilical lower abdominal region. B Klein solution infiltration through a subcentimeter incision 2 cm above the pubic symphysis. C and D Liposuction performance on both sides of the marked area with a 13G blunt-tip cannula attached to a 20-mL Vaclock® syringe. E Harvested fat tissue appearance in the syringe after liposuction 2-Mesenchymal stem cell isolation using the LIPOGEMS® system. The aspirated content was transferred into 10-mL syringes and left to decant (Fig. 2A and B). Afterwards, it was subsequently processed with the LIPOGEMS® device, designed to progressively microfragment adipose clusters while simultaneously removing oil residues and blood pro-inflammatory components. This step is detailed in Fig. 2(C to E). Processing was carried out entirely under continuous saline immersion to reduce mechanical stress. The resulting microfragmented adipose product was collected in a 10-mL syringe and allowed to settle so that the excess saline could decant (Fig. 2F). Finally, the prepared graft was distributed into 1-mL syringes fitted with a 22G, 30-mm needle, which would be used for the final injection in the internal opening and fistula tract.Fig. 2 A and B Transference of harvested fat tissue to 10 ml syringes, followed by decantation. C Filling of the cylinder with saline solution. Initial D and final E appearance of the cylinder content after its shaking, maneuver that allows the spheres to micro-fracture the collected fat tissue for its final use. F Extraction of the final product from the cylinder in 10 ml syringes. Afterwards, the prepared graft will be distributed into 1-mL syringes, to be used in the fistula tract treatment 2-Mesenchymal stem cell isolation using the LIPOGEMS® system. The aspirated content was transferred into 10-mL syringes and left to decant (Fig. 2A and B). Afterwards, it was subsequently processed with the LIPOGEMS® device, designed to progressively microfragment adipose clusters while simultaneously removing oil residues and blood pro-inflammatory components. This step is detailed in Fig. 2(C to E). Processing was carried out entirely under continuous saline immersion to reduce mechanical stress. The resulting microfragmented adipose product was collected in a 10-mL syringe and allowed to settle so that the excess saline could decant (Fig. 2F). Finally, the prepared graft was distributed into 1-mL syringes fitted with a 22G, 30-mm needle, which would be used for the final injection in the internal opening and fistula tract. Fig. 2 A and B Transference of harvested fat tissue to 10 ml syringes, followed by decantation. C Filling of the cylinder with saline solution. Initial D and final E appearance of the cylinder content after its shaking, maneuver that allows the spheres to micro-fracture the collected fat tissue for its final use. F Extraction of the final product from the cylinder in 10 ml syringes. Afterwards, the prepared graft will be distributed into 1-mL syringes, to be used in the fistula tract treatment A and B Transference of harvested fat tissue to 10 ml syringes, followed by decantation. C Filling of the cylinder with saline solution. Initial D and final E appearance of the cylinder content after its shaking, maneuver that allows the spheres to micro-fracture the collected fat tissue for its final use. F Extraction of the final product from the cylinder in 10 ml syringes. Afterwards, the prepared graft will be distributed into 1-mL syringes, to be used in the fistula tract treatment 3-Patient positioning for fistula approach. In case of anterior fistula the patient was put in prone jackknife position. Lithotomy position with padded leg holders was adopted otherwise. 3-Patient positioning for fistula approach. In case of anterior fistula the patient was put in prone jackknife position. Lithotomy position with padded leg holders was adopted otherwise. 4-Management of internal opening and fistula tract. First, a mucosal flap is fashioned over the internal opening using electrocautery, ensuring that it extends at least 1 cm beyond the fistula orifice (Fig. 3A, B). After, the internal orifice is closed with 2/0 polydioxanone stitches, making sure that the muscular layer is included in them (Fig. 3C and D). Finally, the flap is closed using an absorbable, 3/0, multifilament suture (Fig. 3E, F). Core-out of the extrasphinteric fistula tract is performed when possible. If not, external opening is lightly enlarged to facilitate drainage.Fig. 3A Initial view of the fistula tract canalized with a seton. B Internal orifice after performance of the mucosal flap. C and D Internal orifice closure with 2/0 polydioxanone stitches, making sure that the muscular layer is included in them. Closure of the mucosal flap can be completed in an horizontal E or vertical F direction depending on tissue tension and internal orifice characteristics 4-Management of internal opening and fistula tract. First, a mucosal flap is fashioned over the internal opening using electrocautery, ensuring that it extends at least 1 cm beyond the fistula orifice (Fig. 3A, B). After, the internal orifice is closed with 2/0 polydioxanone stitches, making sure that the muscular layer is included in them (Fig. 3C and D). Finally, the flap is closed using an absorbable, 3/0, multifilament suture (Fig. 3E, F). Core-out of the extrasphinteric fistula tract is performed when possible. If not, external opening is lightly enlarged to facilitate drainage. Fig. 3A Initial view of the fistula tract canalized with a seton. B Internal orifice after performance of the mucosal flap. C and D Internal orifice closure with 2/0 polydioxanone stitches, making sure that the muscular layer is included in them. Closure of the mucosal flap can be completed in an horizontal E or vertical F direction depending on tissue tension and internal orifice characteristics A Initial view of the fistula tract canalized with a seton. B Internal orifice after performance of the mucosal flap. C and D Internal orifice closure with 2/0 polydioxanone stitches, making sure that the muscular layer is included in them. Closure of the mucosal flap can be completed in an horizontal E or vertical F direction depending on tissue tension and internal orifice characteristics 5-Micro-fragmented adipose tissue infiltration. After completion of the core-out (Fig. 4A), the MSC preparation is injected along the course of the fistula tract, in the mucosal flap and in the internal opening in the muscular layer (Fig. 4B to D). The amount injected vary depending on the amount obtained and the length of the fistula tract. A volume less than 6 mL is considered insufficient for adequate treatment.Fig. 4A Final view after performance of the core-out. B Injection of the MSC preparation around the fistula tract. C and D Injection of the MSC preparation in the mucosal flap and in the internal opening and the mucosal flap approaching it from both sides of the fistula tract 5-Micro-fragmented adipose tissue infiltration. After completion of the core-out (Fig. 4A), the MSC preparation is injected along the course of the fistula tract, in the mucosal flap and in the internal opening in the muscular layer (Fig. 4B to D). The amount injected vary depending on the amount obtained and the length of the fistula tract. A volume less than 6 mL is considered insufficient for adequate treatment. Fig. 4A Final view after performance of the core-out. B Injection of the MSC preparation around the fistula tract. C and D Injection of the MSC preparation in the mucosal flap and in the internal opening and the mucosal flap approaching it from both sides of the fistula tract A Final view after performance of the core-out. B Injection of the MSC preparation around the fistula tract. C and D Injection of the MSC preparation in the mucosal flap and in the internal opening and the mucosal flap approaching it from both sides of the fistula tract Results Demographics A total of 28 patients (table 1) with complex anal fistula treated with LIPOGEMS® between June 2019 and May 2025 were included in the study. Mean age was 51.12 years (SD: 15.32), with 68% (19/28) of males and 32% (9/28) of female subjects. American Society of Anesthesiologists (ASA) most frequent classification was II (22/28; 79%) and III (6/28; 21%). Arterial hypertension (13/28, 46%) was the main comorbidity, followed by diabetes mellitus and hyperlipidemia, being each one detected in 8 patients (29%). Crohn´s disease (CD) was observed in 4 patients (14%).Table 1Demographic dataN=28Age, mean years (SD*)51.21 (15.32)Gender, n (%) Male19 (67.9) Female9 (32.1)BMI, mean Kg/m2 (SD)28.69 (5.17)ASA score, n (%) I0 (0) II22 (78.6) III6 (21.4) IV0 (0)HTN, n(%)13 (46.4)DM, n (%)8 (28.6)HL, n (%)8 (28.6)CAD, n (%)0 (0)COPD, n (%)1 (3.6)BA, n (%)1 (3.6)SAHS, n (%)2 (7.1)CKD, n (%)1 (3.6)Anticoagulant/antiplatelet therapy, n (%)1 (3.6)IBD, n (%)4 (14.3) Biologic treatment4 (100) Montreal Classification A2L1B1p1 (25) A2L2B1p1 (25) A2L3B1p2 (50)SD Standard Deviation, BMI Body Mass Index, ASA American Society of Anesthesiologists, HTN Hypertension, DM Diabetes Mellitus, HL Hyperlipidemia, CAD Coronary Artery Disease, COPD Chronic Obstructive Pulmonary Disease, BA Bronchial Asthma, SAHS Sleep Apnea–Hypopnea Syndrome, CKD Chronic Kidney Disease, IBD Inflammatory Bowel Disease Table 1Demographic dataN=28Age, mean years (SD*)51.21 (15.32)Gender, n (%) Male19 (67.9) Female9 (32.1)BMI, mean Kg/m2 (SD)28.69 (5.17)ASA score, n (%) I0 (0) II22 (78.6) III6 (21.4) IV0 (0)HTN, n(%)13 (46.4)DM, n (%)8 (28.6)HL, n (%)8 (28.6)CAD, n (%)0 (0)COPD, n (%)1 (3.6)BA, n (%)1 (3.6)SAHS, n (%)2 (7.1)CKD, n (%)1 (3.6)Anticoagulant/antiplatelet therapy, n (%)1 (3.6)IBD, n (%)4 (14.3) Biologic treatment4 (100) Montreal Classification A2L1B1p1 (25) A2L2B1p1 (25) A2L3B1p2 (50)SD Standard Deviation, BMI Body Mass Index, ASA American Society of Anesthesiologists, HTN Hypertension, DM Diabetes Mellitus, HL Hyperlipidemia, CAD Coronary Artery Disease, COPD Chronic Obstructive Pulmonary Disease, BA Bronchial Asthma, SAHS Sleep Apnea–Hypopnea Syndrome, CKD Chronic Kidney Disease, IBD Inflammatory Bowel Disease Demographic data SD Standard Deviation, BMI Body Mass Index, ASA American Society of Anesthesiologists, HTN Hypertension, DM Diabetes Mellitus, HL Hyperlipidemia, CAD Coronary Artery Disease, COPD Chronic Obstructive Pulmonary Disease, BA Bronchial Asthma, SAHS Sleep Apnea–Hypopnea Syndrome, CKD Chronic Kidney Disease, IBD Inflammatory Bowel Disease Perioperative data In relation to fistula characteristics, (tables 2 and 3). 24 of the 28 patients had a mid or high transsphincteric fistula. Posterior location was the most frequently observed (16/28; 57%). The 3 low fistulas detected had an anterior location and were observed in women. According to SJUH classification, grade III (22/28) and IV (5/28) supposed the 96% of the sample. No fistula was amenable to fistulotomy when Garg classification was applied and referral to a fistula expert was recommended in 8 cases (29%). Grade III-A (15/28; 54%) and III-B (5/28; 18%) were the most frequent ones.Table 2Perioperative dataN=28Etiology, n (%) Primary7 (25) Recurrent21 (75)Fistula location, n (%) Anterior12 (42.9) Posterior16 (57.1)Number of fistula tracts, n (%) 125 (89,3) 23 (10.7)Preoperatory imaging test, n (%) Endoanal ultrasound26 (92.9) MRI*6 (21.4)Fistula type according to radiological sphincter involvement, n (%) Low transsphincteric fistula3 (10.7) Mid transsphincteric fistula12 (42.9) High transsphincteric fistula12 (42.9) Suprasphincteric fistula1 (3.6)Previous surgical interventions, median (IQR)3 (4)Previous Curative Intent Surgical Treatment (CIST), n (%) Single operation group (SOG)11 (39.28) Laser therapy5 (18) Partial fistulectomy4 (14.4) LIFT*2 (7.1) Reoperated group (ROG)10 (36) Laser therapy7 (25) Partial fistulectomy5 (18) LIFT3 (10.7) Endorectal advancement flap (EAF)2 (7.1) Fistula plug insertion4 (14.4) Operative time, median min (range)71 (47–121) Core-out extrasphincteric tract, n (%)13 (46.5) Amount of processed micro-fragmented adipose tissue, median ml (range)14 (6–20)MRI Magnetic resonance imaging, LIFT Ligation of the Intersphincteric Fistula Tract, IQR Interquartile rangeTable 3Preoperative perianal fistula classificationN=28Park´s classification, n (%) Submucosal0 (0) Intersphinctheric0 (0) Transsphincteric27 (96.4) Suprasphincteric1 (3.6) Extrasphincteric0 (0)St. James University Hospitalclassification, n (%) Grade I: Simple linear intersphincteric fistula0 (0) Grade II: Intersphincteric fistula with intersphincteric abscess or secondary tract0 (0) Grade III: Transsphincteric fistula22 (78.6) Grade IV: Transsphincteric fistula with abscess or secondary tract5 (17.9) Grade V: Suprasphincteric and extrasphincteric fistula1 (3.6)Garg classification, n (%) Grade I: Low linear intersphincteric or transsphincteric fistula0 (0) Grade II: Low intersphincteric or transsphincteric fistula0 (0) Grade III-A: High linear Transsphincteric fistula15 (53.6) Grade III-B: Fistula with associated Crohn's disease, sphincter injury, post radiation exposure or anterior fistula in a female5 (17.9) Grade IV-A: High Transsphincteric fistula with abscess2 (7.1) Grade IV-B: High Transsphincteric fistula with multiple tracts3 (10.7) Grade IV-C: High Transsphincteric fistula with Horseshoe abscess2 (7.1) Grade V-A: Transsphincteric with intersphincteric supralevator extension0 (0) Grade V-B: Suprasphincteric fistula1 (3.6) Grade V-C: Extrasphincteric fistula0 (0) Table 2Perioperative dataN=28Etiology, n (%) Primary7 (25) Recurrent21 (75)Fistula location, n (%) Anterior12 (42.9) Posterior16 (57.1)Number of fistula tracts, n (%) 125 (89,3) 23 (10.7)Preoperatory imaging test, n (%) Endoanal ultrasound26 (92.9) MRI*6 (21.4)Fistula type according to radiological sphincter involvement, n (%) Low transsphincteric fistula3 (10.7) Mid transsphincteric fistula12 (42.9) High transsphincteric fistula12 (42.9) Suprasphincteric fistula1 (3.6)Previous surgical interventions, median (IQR)3 (4)Previous Curative Intent Surgical Treatment (CIST), n (%) Single operation group (SOG)11 (39.28) Laser therapy5 (18) Partial fistulectomy4 (14.4) LIFT*2 (7.1) Reoperated group (ROG)10 (36) Laser therapy7 (25) Partial fistulectomy5 (18) LIFT3 (10.7) Endorectal advancement flap (EAF)2 (7.1) Fistula plug insertion4 (14.4) Operative time, median min (range)71 (47–121) Core-out extrasphincteric tract, n (%)13 (46.5) Amount of processed micro-fragmented adipose tissue, median ml (range)14 (6–20)MRI Magnetic resonance imaging, LIFT Ligation of the Intersphincteric Fistula Tract, IQR Interquartile range Perioperative data MRI Magnetic resonance imaging, LIFT Ligation of the Intersphincteric Fistula Tract, IQR Interquartile range Table 3Preoperative perianal fistula classificationN=28Park´s classification, n (%) Submucosal0 (0) Intersphinctheric0 (0) Transsphincteric27 (96.4) Suprasphincteric1 (3.6) Extrasphincteric0 (0)St. James University Hospitalclassification, n (%) Grade I: Simple linear intersphincteric fistula0 (0) Grade II: Intersphincteric fistula with intersphincteric abscess or secondary tract0 (0) Grade III: Transsphincteric fistula22 (78.6) Grade IV: Transsphincteric fistula with abscess or secondary tract5 (17.9) Grade V: Suprasphincteric and extrasphincteric fistula1 (3.6)Garg classification, n (%) Grade I: Low linear intersphincteric or transsphincteric fistula0 (0) Grade II: Low intersphincteric or transsphincteric fistula0 (0) Grade III-A: High linear Transsphincteric fistula15 (53.6) Grade III-B: Fistula with associated Crohn's disease, sphincter injury, post radiation exposure or anterior fistula in a female5 (17.9) Grade IV-A: High Transsphincteric fistula with abscess2 (7.1) Grade IV-B: High Transsphincteric fistula with multiple tracts3 (10.7) Grade IV-C: High Transsphincteric fistula with Horseshoe abscess2 (7.1) Grade V-A: Transsphincteric with intersphincteric supralevator extension0 (0) Grade V-B: Suprasphincteric fistula1 (3.6) Grade V-C: Extrasphincteric fistula0 (0) Preoperative perianal fistula classification St. James University Hospital classification, n (%) Including both CIST and non-CIST previous surgeries, 18 patients (64%) had undergone ≥3 surgical interventions prior to LIPOGEMS® therapy (Table 3). When considering CIST frequency, laser fistula closure had been attempted in 12 cases (5 in SOG and 7 in ROG), partial fistulectomy in 9 cases (4 in SOG and 5 in ROG), and LIFT in 5 cases (2 in SOG and 3 in ROG). In the ROG group, the most frequent combinations were laser closure with LIFT or fistula plug insertion, with three patients in each group. Median operative time was 71 minutes (Range: 47–121), and the median of amount of injected processed fat was 14 ml (6–20). Removal of the extrasphincteric tract was only possible in 13 cases. No intraoperative adverse events were recorded. Postoperative course Only two postoperative complications were observed: a minor abdominal wall hematoma (secondary to liposuction) that did not require specific treatment, and a perianal abscess that was drained under local anesthesia in the outpatient clinic. This last case presented as a late event and was due to a superficial incomplete wound healing, as no fistula recurrence was observed. As shown in Table 4, an initial healing rate of 53% was obtained at 6 months, which decreased to 46% after two recurrences were confirmed (at 7 and 8 months respectively) following an initial complete closure. The overall healing rate was higher in case of primary (4/7; 57%) and posterior fistula (9/16; 56%). Patients with abscess or multiple fistula tracts presented a lower healing rate (40% vs 56%), whereas no statistical significance was obtained (p=0,241). No statistical significance was observed in the healing rate-amount of injected processed tissue relation in our series (55% in <12 ml cases, 40% in >12 ml cases, p=0.368). In cases that healed after LIPOGEMS®, nearly 3 months passed until the discharge stopped. In terms of functional results, a non-significant difference of 0.12 points between pre and postoperative Wexner score was registered (p=0.343). A median LOS of 1 day was recorded. The overall follow-up was 11.2 months (3.43–69.13), being slightly longer in the healed patient’s subgroup (12.5; 6.53–27.57).Table 4Postoperative courseN=28Postoperative complications, n (%)2 (7.2) ≤30 days1 (3.6) >30 days1 (3.6)Clavien-Dindo classification, n (%) I1 (3.6) II0 (0) IIIa1 (3.6) IIIb0 (0) IV0 (0)Reoperation, n (%)0 (0)Readmission, n (%)1 (3.6)Abdominal SSI, n (%)0 (0)Abdominal hematoma, n (%)1 (3.6)Perianal abscess, n (%)1 (3.6)Systemic events, n (%)0 (0)Healing rate, n (%) Overall healing rate13 (46.4) Initial healing rate15 (53.5)Healing rate according to previous treatment Primary fistula4 (57.1) Recurrent fistula9 (42.8) Combined treatment fistula4 (40)Healing rate according to fistula location Anterior fistula4 (33.3) Posterior fistula9 (56.3)Healing rate according to radiological sphincter involvement Low transsphincteric fistula1 (33.3) Mid transsphincteric fistula5 (41.7) High transsphincteric fistula6 (50) Suprasphincteric fistula1 (100) Fistula persistence, n (%)13 (46.4) Fistula recurrence, n (%)2 (7.1) Time until discharge stop, median months (range)2.25 (1–5)Wexner Incontinence Score, median (range) Preoperative0.81 (0–14) Postoperative0.93 (0–15) LOS*, median days (range)1 (1-1) Follow-up of healed patients, median months (range)12.50 (6.53–27.57) General follow-up time, median months (range)11.20 (3.43–69.13)SSI Surgical Site Infection, LOS Length of stay Table 4Postoperative courseN=28Postoperative complications, n (%)2 (7.2) ≤30 days1 (3.6) >30 days1 (3.6)Clavien-Dindo classification, n (%) I1 (3.6) II0 (0) IIIa1 (3.6) IIIb0 (0) IV0 (0)Reoperation, n (%)0 (0)Readmission, n (%)1 (3.6)Abdominal SSI, n (%)0 (0)Abdominal hematoma, n (%)1 (3.6)Perianal abscess, n (%)1 (3.6)Systemic events, n (%)0 (0)Healing rate, n (%) Overall healing rate13 (46.4) Initial healing rate15 (53.5)Healing rate according to previous treatment Primary fistula4 (57.1) Recurrent fistula9 (42.8) Combined treatment fistula4 (40)Healing rate according to fistula location Anterior fistula4 (33.3) Posterior fistula9 (56.3)Healing rate according to radiological sphincter involvement Low transsphincteric fistula1 (33.3) Mid transsphincteric fistula5 (41.7) High transsphincteric fistula6 (50) Suprasphincteric fistula1 (100) Fistula persistence, n (%)13 (46.4) Fistula recurrence, n (%)2 (7.1) Time until discharge stop, median months (range)2.25 (1–5)Wexner Incontinence Score, median (range) Preoperative0.81 (0–14) Postoperative0.93 (0–15) LOS*, median days (range)1 (1-1) Follow-up of healed patients, median months (range)12.50 (6.53–27.57) General follow-up time, median months (range)11.20 (3.43–69.13)SSI Surgical Site Infection, LOS Length of stay Postoperative course SSI Surgical Site Infection, LOS Length of stay Discussion In this retrospective case series, the use of microfragmented adipose tissue processed with the LIPOGEMS® system for the treatment of complex anal fistulas achieved a moderate healing rate of 46%, with a low complication profile and no significant impact on continence. These findings should be interpreted in the context of a highly selected and challenging population, as most patients had mid-to high transsphincteric fistulas and a complex surgical history. In recent years, sphincter-preserving techniques have gained popularity due to their potential to achieve better functional outcomes while effectively treating the fistula. In this setting, the most established is the placement of a draining or loose seton, which maintains tract patency, allows ongoing drainage, and reduces the risk of abscess. This strategy is particularly appropriate for patients unwilling to accept the risk of incontinence or in those with complex fistulas or fistulizing Crohn´s Disease [17]. Nevertheless, this therapy serves only as a bridge to surgery or palliative treatment, since it does not permit fistula healing. Many other conservative techniques have been developed. The use of plugs and fibrin glue emerged in the early 2000´s, but their poor long-term results had led to their lack of use [17, 24–26]. Although platelet-rich plasma (PRP) injection has demonstrated promising outcomes, its ability to consistently reduce long-term recurrence remains uncertain [27, 28]. Moreover, the ESCP guidelines for cryptoglandular anal fistula only recommends its use as an add-on treatment to LIFT [29]. Laser-based therapies may constitute the main sphincter-saving technique at present due to its reproducibility and safety, even though results are modest in case of complex anal fistula [30]. Traditional sphincter-preserving procedures, such as LIFT and endorectal advancement flap (EAF), must also be discussed. Fistula tract ligation offers a very low incontinence risk (0.6%−1.6%) together with an acceptable healing rate (69%), even though factors as prior surgery, horseshoe extensions, or Crohn´s Disease negatively influence its efficacy [29, 31, 32]. For its part, EAF is typically reserved for more complex or high fistulas, achieving higher healing rates of around 74–75%, but at the cost of increased incontinence rates, ranging from 7 to 20% depending on the thickness of the flap used [29, 31, 33]. Even though It is not mentioned in the ESCP guidelines [29], the use of autologous-derived mesenchymal stem cells has been previously investigated. Meta-analysis and long-term studies showed a clinical remission rate between 50% and 70%, being slightly lower in case of cryptoglandular origin [34, 35]. In this sense, the LIPOGEMS® processing system was developed as a cheaper and simpler MSC-based treatment, as no cell expansion is needed and treatment is completed in a single-step surgery [16]. The available literature about the use of LIPOGEMS® system is scarce. Thus, a preliminary results analysis is mandatory at this point to determine whether further research is worthwhile. In this sense, the present study constitutes the second largest case LIPOGEMS® series to date. The population included showed a great heterogeneity in etiology, fistula characteristics and previous surgical history. When contrasted with Naldini et al. original article [16], our sample showed a greater recurrent fistula rate (75% vs 58%) and a broader inclusion criteria, as patients with IBD, multiple fistula tracts and anticoagulant/antiplatelet therapy were included. In this sense, the case series published by Topal and Testa et al. were more similar to the present one, as the above mentioned characteristics did not suppose the patient exclusion [36, 37]. In this sense, we believe these differences justify the lower healing rate (46%) reported in this study compared with those reported in the literature (74% by Naldini et al., 70% by Topal et al. and 68% by Testa et al.) [16, 36, 37]. In fact, in the same Naldini and Testa studies a relationship between previous CIST and a lower healing rate (present in 75% of our sample) is found. Moreover, a 52% increased risk of fistula recurrence in patients with prior anal surgery has been previously reported by García-Aguilar and Mei [6, 38]. This higher fistula complexity did not cause a higher morbidity rate in our study. In fact, our morbidity rate was slightly lower than the one communicated by Testa et al. (7% vs 10%) [37]. Thus, this surgical procedure keeps a low complication rate as one of its advantages in spite of the increased patient complexity observed in the recent studies. We consider laser-based fistula therapy to be one of the main sphincter-preserving techniques due to its wider adoption compared with other modalities. A lower cost than PRP and MSC therapy as well as a positive security profile constitutes its main benefits. Also, collaboration with plastic surgery is not needed and therefore surgical scheduling process is simpler. Moreover, the healing rates reported in recent meta-analysis are non-inferior to other therapies [30, 39], even though a recent Spanish study by De la Portilla et al. [40] reported a significantly lower percentage (44.3%). In this regard, adipose-tissue MSC treatment may offer a higher and more durable success rate [13, 34] even though comparative studies have not been completed yet. The main advantage over laser ablation and LIFT is, in our opinion, the broader indications of MSCs therapy (with or without LIPOGEMS®) considering fistula local characteristics. Parameters as fistula tract length >4–5 cm, multiple tract existence, rectovaginal fistula, interposed collections and IBD etiology are considered contraindications for laser and LIFT but not for MSCs treatment [40, 41]. It also should be noticed that even though these are well-known recurrence factors, healing rates of MSC seem to be similar in the short-term and even higher in the long-term when compared to laser therapy [13, 34]. This difference is maintained in the short-term when comparing laser with LIPOGEMS®-specific case series [16, 36, 37]. As a result, we believe this treatment must be considered in complex fistula management. EAF, for its part, is applicable to more complex scenarios but, as previously mentioned, carries a higher risk of incontinence [33]. It is also a technically demanding procedure with a significant learning curve, whereas MSCs and laser treatment are easier to perform [42]. This study has several limitations. Selection bias is present due to its retrospective nature. Also, the patient sample included was small in statistical terms and showed a great heterogeneity. These two facts could have limited the utility of the statistical results and the external validity. Also, for the absence of control group any comparison with other surgical techniques was impossible to perform. In the clinical setting, the lack of a standardized radiological follow-up protocol may have led to underdetection of fistula recurrence [43], although 24 of 28 patients underwent follow-up EUS due to suspected recurrence. Furthermore, the relatively short follow-up period of the study may have resulted in an underestimation of the late recurrence rate. Nevertheless, considering our results regarding fistula healing and morbidity rates, we believe that the preliminary evidence obtained from this case series is encouraging and useful in this difficult field, and that further research is justified. Conclusions Complex fistula treatment with microfragmented autologous adipose tissue is a valid therapeutic option in perianal fistula management, as it offers a similar healing rate to other sphincter-saving techniques with a low postoperative complication rate. As the fourth study in this field, the present work adds evidence about MSCs LIPOGEMS® therapy indications, efficacy and safety in perianal fistula. However, further prospective, comparative studies are needed to better define its indications and to draw robust conclusions regarding its effectiveness.