Background/Objectives: Micro-fragmented adipose tissue is emerging as a promising option for the treatment of various diseases including knee osteoarthritis (OA), though clinical trials are often limited by small sample size and short follow-up periods. Our aim was to evaluate safety and clinical outcomes of an arthroscopic debridement followed by a single injection of micro- fragmented adipose tissue in patients affected by knee OA. Methods: From 2016 to 2020 patients affected by knee OA were enrolled. Micro-fragmented adipose tissue was obtained through the Lipogems kit and intraarticularly injected after an arthroscopic debridement. VAS for pain, Tegner score, Lysholm score, KOOS score subscales variations were assessed from baseline to 3-, 6-, 12-, 24- and 48 months follow-up.

Patients were placed in a supine position and Betadine was used to treat the abdomen.

Anesthesia was performed using disposable blunted cannulas with a modified Klein solution (a mixture of 250 ml of saline solution (4°C), 50 ml of lidocaine (2%) and 1 ml of epinephrine, 1:1000). Two skin incision of approximately 0.5 cm were performed on both sides of the abdomen and the solution was injected in the subcutaneous fat tissue.

Following anesthesia, physicians had to wait at least 15 min before proceeding. Tissue harvesting was performed connecting VacLock vacuum syringe to an aspiration blunt cannula. Approximately the same quantity of tissue was harvested from each side of the patient's abdomen, with a minimum of 100 ml. All procedures were performed by at least two physicians, while the first physician prepared and used the Lipogems ortho kit, the other performed a debridement arthroscopic procedure on the selected knee.

Once the aspiration process was completed, approximately 100 ml of lipoaspirate were processed at each time in the standard 225-ml device. The Lipogems ortho kit is made by a core, a simple disposable device that progressively reduces in size the clusters of adipose tissue (from 1-3.5 mm to clusters of approximately 0.2-0.8 mm) while at the same time eliminating oil and blood residues. The whole procedure takes place in a full immersion system to minimize traumatic actions on cell products. Before starting the procedure, to avoid any cell damage the total absence of air and bubbles in the device was ensured and the core was pre-filled with saline solution from a 1000 ml bag hanging above. It was ensured that the fat that was aspirated was constantly surrounded by a liquid environment, essential to eventually obtain small healthy fat clusters, discarding oil and fat debris. The aspirated fat was pushed from the syringe into the device and through the first reduction filter allowing the corresponding amount of saline to exit from the opposite side reaching the waste bag. Upon each first injection of aspirated fat into this closed system, the floating tissue composed of the lipoaspirate occupied no more than the first half of the core device while keeping it in a vertical position. Once this was observed, the device was shaken and the five stainless steel marbles present inside the system allowed to emulsify oil residues, subsequently removed by gravity along with blood components and the saline solution counterflow. On the other hand, the adipocyte clusters remained or migrated to the top of the device. Once the solution in the device appeared clear and the lipoaspirate yellow, the saline flow was interrupted, and the device rotated by 180°. At this point the process continued by passing the adipose cluster that was obtained through a second size reduction filter and pushing additional fluid from the lower opening of the device. This allowed to obtain a second adipose cluster reduction. At this point the final Lipogems product was ready to be collected in 10 ml syringes from the upper opening and ready for its knee intra-articular injection at the level of the superolateral edge of the patella. This process required approximately 20 min and the final product volume obtained was roughly 25%-30% of the initial tissue that was harvested. Skin portals were sutured and elastic body belt applied and to be maintained for 10 days. Patients were usually discharged the same day of the procedure with a postoperative pain managements protocol (Paracetamol 1000 mg 1 tablet three times a day for 5 days and Ibuprofen 600 mg 1 tablet twice a day for 2 days) and a thromboembolic prophylaxis protocol (low molecular weight heparin for 1 week). Crutches were suggested for the first two days and full weightbearing allowed from the first postoperative day.

Inclusion Criteria:

• age > 18 years, patients affected by radiologically confirmed knee OA according to Kellgren-Lawrence classification, not responsive to at least 3 months of conservative treatment (e.g. physical therapy, exercise).

Exclusion Criteria:

• recent traumatic events involving the index joint, chondromatosis, intra-articular injections of any type in the previous 6 months, malignancy, pregnancy, patients on anticoagulant therapy or suffering from coagulopathies and related disorders, systematic disorders affecting the joints (i.e rheumatoid arthritis). Enrolled patients underwent a knee radiograph (antero-posterior, Rosenberg and lateral projections) as well as MRI before the procedure to stage knee OA and evaluate potential intra-articular loose bodies or diseases different from knee OA (e.g tumors or focal cartilage lesions).

Patients aged > 18 years, affected by radiologically confirmed knee OA according to Kellgren-Lawrence classification, not responsive to at least 3 months of conservative treatment (e.g. physical therapy, exercise).