Use of Autologous Adipose-Derived Stromal Vascular Fraction to Treat Osteoarthritis of the Knee

Methods:

Adipose Harvest. Using no oral or parenteral sedation, standard wetting solution (1 liter Lactated Ringer's, 50 milligrams of 1% lidocaine, and 1 cc of 1:1000 epinephrine) was infused through small incisions created with the tip of a #11 scalpel blade in the abdomen or flank using a standard multi-hole infusion cannula. A super-wet plus technique (2 volumes of wetting solution to 1 volume of proposed fat aspirate) was used to infuse the solution into the deep and superficial fat compartments. Twenty minutes was allowed for maximum vasoconstrictive effect of the epinephrine. Fat was harvested using standard Suctioned-Assisted Liposuction (SAL) method with a 3mm cannula at approximately .5 - .7 atmosphere of vacuum, aspirating approximately 200-300 cc of lipoaspirate into a sterile tissue processing container (GID SVF-1, Lousiville, CO) for each knee to be treated.

Adipose processing method. Lipoaspirate was harvested and processed all the way to the generation of the SVF within a single GID SVF-1 sterile disposable device. The GID SVF-1 unit contains the washing mechanism, the mesh filter, and the centrifuge capability in the same device. After harvest the lipoaspirate was washed three times with 37C Lactated Ringer's Solution with 20 mg Cipro and 5,000 units heparin per liter and the fluid portion removed using the mesh filter system, leaving dry adipose inside the canister. The canister was weighed to determine the amount of dry adipose available for further processing. The washed adipose was disaggregated using Type I collagenase (Worthington, Lakewood, NJ) at a concentration of 200 CDU/ml of total catalytic volume where total catalytic volume is the volume of the adipose tissue plus an equal volume of 37C Lactated Ringer's Solution. The collagenase was injected into the canister through a sterile .22 micron filter (Millex-MP, Millipore, Cork, Ireland). The device with adipose, buffer, and collagenase was then placed into an incubated shaker for 40 minutes at 38C at 150 RPM, see Figure 1. After disaggregation human albumin solution was added to achieve a concentration of 2.5% and stop any further collagenase activity. The device was then centrifuged at 800 g for 10 minutes. The supernatant, including all floating cells and debris and the aqueous phase, was removed using a port on the top of the device and discarded. The SVF pellet at the bottom was resuspended in 10 ml of sterile LR accessed via the central port on the device using a 14G 5.5 inch spinal needle (Abbocath-T, Hospira, Sligo, Ireland). A sample of 0.5 ml of the resuspension was collected in a 1.5 cc Eppendorf tube to be used for cell counting and assay. The resuspended SVF cells were concentrated into a 3 ml dose using a sterile 15 ml Falcon tube at 400 g for 4 minutes.

SVF counting method. The SVF cell count and viability was assessed using an ADAM MC image cytometry system (Bulldog Bio, Portsmouth, NH). The ADAM MC uses propidium iodide staining to count viable and non-viable nucleated cells. A 100 µL aliquot of the assay sample was diluted with a 1:5 ratio (1 part sample to 5 parts sterile LR) to adjust the sample within the operating limits of the ADAM MC. The differential stains were applied and aliquots loaded into the disposable cassette that is utilized with the ADAM device. The results from the ADAM counting device provide the concentration of SVF cells in the resuspension syringe and the percentage viability. The total volume of the resuspension in the syringe was multiplied by the concentration to give the total number of resuspended mononucleated cells (no adipose cells, no RBCs, and no fragments included in the counting process).

Injection and image guidance method. The patient was placed in the supine position and the area over the lateral suprapatellar region of the knee was prepared in the usual sterile fashion using an iodine prep solution (1% Iodophor) followed by a 70% Isopropyl Alcohol solution. This location was used due to ease of access into the intra-articular space and to avoid injecting into the fat pad of the knee. No sedation or pain medication was administered to the patient. An ethylchloride topical anesthetic was sprayed on the lateral knee until the skin color changed to white. Lidocaine local anesthetic was injected using a 25 gauge needle to numb the skin and subcutaneous tissues. Using an M-Turbo Sonosite ultrasound system, the joint space was identified and under live ultrasound guidance the knee was aspirated using an 18 gauge/1.5 inch needle, if fluids were available for aspiration. Then all of the 3 cc of buffered solution of SVF was slowly injected into the intra-articular space through the same 18 gauge/1.5 inch needle. The needle was then be removed and direct pressure over the injection site will placed for approximately ten (10) seconds. Hemostasis after injection was confirmed, and then the injection site was cleaned with an alcohol wipe and covered with a sterile band-aid. The patient was given crutches and asked to be non-weight bearing on the injected knee for two (2) days. The patient was allowed to bend and flex the knee as long as non-weight bearing condition is maintained.

Pain and mobility assessment methods. Assessment of knee pain was done using the PROMIS pain instruments, a validated pain scale system developed under funding by the NIH. PROMIS (Patient Reported Outcomes Measurement Information System) is a system for measures of patient reported health status for physical, mental, and social well-being, including three measurements of pain (www.nihpromis.org). The PROMIS Pain Intensity instrument (3a) assesses how much a person hurts. The PROMIS Pain Behavior instrument (Bank 1.0) measures behaviors that indicate to others that an individual is experiencing pain. The PROMIS Pain Interference instrument (Bank 1.0) measures the consequences of pain on relevant aspects of one's life. Patient responses are converted to numeric values in a validated scale. Responses to each question in a pain instrument range from 1 to 5 (1=not at all, 2=a little bit, 3=somewhat, 4=quite a bit, 5=very much), and are summed over the number of questions in the bank to create a raw score. The raw score is converted to a standardized score using an iterative response method with a mean of 50 and SD of 10, based on a large sample of the general population of the United States. The validated PROMIS pain scales provide interval data and allows calculation of the mean and SD, and use of t-tests of significance. Data was recorded pre-operatively (time 0), and at follow-up points at 2 weeks, 4 weeks, 6 weeks, and 12 weeks.

Additionally patients were asked to respond to a pain and mobility questionnaire with one question on pain and one question on mobility. The questionnaire asked the patient to compare their increase/decrease in pain and mobility relative to before surgery, using a ± 1-5 scale, see Figure 2. On this scale negative numbers correspond to decreasing pain/mobility, and positive numbers correspond to increasing pain/mobility. This questionnaire provides ordinal data and allows calculation of the median and significance using Wilcoxon Ranked Sums nonparametric testing.