PURPOSE: To compare the effect of hyaluronic acid (HA) or platelet-rich plasma (PRP) on pain, physical function, quality of life and knee joint morphology in patients with knee osteoarthritis severity II-III.

BACKGROUND: Knee osteoarthritis affects quality of life significantly because it is the most common joint disease and causes considerable disability. Pathogenesis is multifactorial, nevertheless reduced cartilage production, increased destruction, and synovial inflammation are important factors in the osteoarthritis process. Today, symptomatic drugs are commonly used in the treatment of osteoarthritis, but these treatments have limited effects on cartilage degeneration. Intraarticularly, hyaluronic acid (HA) and platelet-rich plasma (PRP) treatments have been used for osteoarthritis due to pain and functional effects. HA has been shown to reduce the levels of collagen degradation products and maintain normal cartilage metabolism. PRP is thought to have positive effects on clinical and tissue healing due to the numerous growth factors involved. However there is no research to prove definitively that one of the two applications in knee osteoarthritis is superior to the other.

METHOD: 120 patients between the ages of 50-70, OA severity II-III will be included in the study. Patients will be stratified according to the severity and age of OA, and two groups will be randomly assigned as HA and PRP. HA and PRP injections will be performed two times and one month apart. Outcome measures are pain, physical function, quality of life, muscle strength, WORMS, and patient satisfaction. Each patient will be examined at baseline, first, third, sixth, ninth and twelfth months.

PROJECT TEAM N.Fusun Toraman (NFT) cell phone: +90 505 355 9045 e.mail: fusuntoraman@gmail.com Meral Bilgilisoy Filiz (MBF) cell phone: +90 505 647 5840 e.mail: mbilgilisoy@gmail.com Aslı Karadag Ozdemir (AKO) cell phone: +90 555 276 6334 e.mail:aslikaradagoz@gmail.com Yasemin Karaman (YK) cell phone: +90 533 375 0070 e.mail: dryaseminkaraman@hotmail.com Sevim Yıldız (SY) cell phone: +90 535 967 3870 e.mail: syildiz07@yahoo.com

PROJECT PRESENTATION AND ETHICAL APPROVAL The project team and the physicians of the Physical Therapy and Rehabilitation Clinic were informed about the research topic, content and applications and their suggestions were recorded (August 2017). The research was submitted to the Antalya Training and Research Hospital Ethical Committee and Ethic approval was obtained (September 21, 2017 / Decision number 2017-177). The research was presented as a 3001-TUBITAK project on December 2017, and Contract is signed (September 07, 2018 / Project number 118S014). Project start date was determined as September 21, 2018 by TUBITAK.

The Contract of Project was submitted to the Research and Development Office of Antalya Training and Research Hospital and Purchasing Officers were determined by them (September 13, 2018 / 00076415138). Technical specifications were prepared for the purchase of Regenflex Starter and EasyPRP, bidding process was made, contract was written and products purchased (October 30, 2018 / 00079677083). Technical specification was prepared for the purchase of the computer, bidding process was made, contract was prepared and computer purchased (October 02, 2018 / 00077859953). Technical specification was prepared for the laboratory testing, bidding process was made, contract was prepared (November 22, 2018).It was learned that the research project should be sent to Turkey Pharmaceuticals and Medical Devices Agency Department of Clinical Research at the meeting entitled as Good Clinical Researches which was held on November 30, 2018, at Ministry of Health, Ankara. The project was sent to Turkey Pharmaceuticals and Medical Devices Agency Department of Clinical Research on December 17, 2018 and start time of the project was changed to January 07, 2019. The mandatory amendment on the commencement date of the project was reported to TUBITAK by writing an official letter with attached information related with the Basic Principles of Clinical Trials. Additionally, a similar official letter sent to TUBITAK was written to the Ethical Committee of Antalya Training and Research Hospital and Clinical Research and Development Unit on December 19, 2018.The project is approved by Turkey Pharmaceuticals and Medical Devices Agency Department of Clinical Research (January 03, 2019/No:71146310-511.06-E.1116/subject:2018-157) PATIENT RECRUITMENT: Hospital records of patients admitted to Outpatient Clinics of Physical Therapy and Rehabilitation with the diagnostic code M17.0-M17.9, M24.1, M25.5, between 50-79 ages, and between March and August 2018 were controlled in order to determine how many months were necessary for patient selection and radiological results of the patients were evaluated. At the end of the evaluation, it was decided that two months of the patient recruitment time were sufficient. It is decided to recruit the patients diagnosed as knee osteoarthritis after clinical, laboratory and radiological examination and referred to the project team between January and March 2019. Diagnosis of the knee osteoarthritis is made according to American College of Rheumatology diagnostic criterias. The patients are informed about the content of the study by the principal investigator and the research assistant (NFT, and AKO, respectively]. The criterias of inclusion and exclusion, follow-up times and importance of follow-up are explained to the patients and consent is obtained from the patients who meet the criteria. NFT ve AKO examine the patients and select the patients using inclusion and exclusion criterias. If the sufficient number of patients do not meet the criterias for inclusion between the dates specified, the patients are referred from Physical Therapy and Rehabilitation Outpatient Clinics of Ataturk State Hospital. The patients are selected using the inclusion and exclusion criterias.

ASSESSMENT PROCEDURE: NFT and AKO get information related with age, sex, medical and family history, comorbidities, menopausal characteristics of the female patients, radiologic and laboratory diagnostic tests, education, occupation, socioeconomic status, habits, medication use, cognitive level, physical activity status, dominant extremity. Then activity and rest pain is asked, body composition, physical performance tests, and isokinetic test are applied, questionnaires of disease-specific and overall quality of life are filled. Under AKO supervision physical performance tests are performed by a single occupational therapist, and isokinetic test and body composition measures are performed by a single physiotherapist. The weight-bearing knee radiologies are evaluated independently by SY and NFT according to Kellgren Lawrence criterias. Randomization is performed by MBF after all examinations are completed. The groups are stratified as PRP and HA using covariate adaptive randomization as age and osteoarthritis severity covariate.MRI is performed after appointment of radiology department. Two of MRI in the morning and two of MRI in the afternoon are performed daily. The patients who can not perform MRI examinations are recommended to undergo sedation. The patients who do not want to undergo MRI with sedation are excluded from the study by establishing the definition of non-volunteers. MRI is performed at the beginning and at the end of the study and the evaluation will be made by the two radiologist independently (SY, and YK). Total of 16 patients- eight patients of the HA group in the morning time, eight of the patients of the PRP group in the afternoon as daily- are injected by MBF. Injections are performed by using suprapatellar intraarticular route, ultrasound guided and 20G injector and are repeated after one month. Three passive flexion and extension movement is performed after injection, and after fifteen minutes of the rest the patient is sent to home. On the day of the injection, it is recommended that the patient does not perform intensive activities for 24 hours and should not remain standing for a long time. If there is pain and swelling on the knee, it is recommended to use 500 mg of paracetamol which does not contain codeine, cold application three times a day and not to use nonsteroid antiinflammatory medicines. If pain persists, it is recommended to communicate with the physiatrist who performs injection (MBF).Exercise brochure is prepared for both of the groups to be performed one day after injection and is recommended to continue for one year. Cold application, analgesic and NSAID use, exercise compliance/adherence, and adverse events related with the injections are asked to the patients on each of the evaluation time and the results are recorded.Patient satisfaction with a 6-point Likert scale is evaluated on each time of the evaluation. Each of the patient is examined at baseline, and also at the first, at the third, at the sixth, at the ninth and at the twelfth months for all of the outcome measures. MRI is assessed at baseline and at the twelfth month. Follow-up is provided by phone calling and inviting the patient.

DATA COLLECTION PROCEDURE: The groups are encoded and randomized by MBF. The questionnaires are filled by AKO. All of the results of examination, laboratory, radiography and outcome measures will be saved in a separate file for each of the patient by AKO. The data in the files is recorded to the computer by AKO in every weekend. The written files and computer data are checked by MBF.

DATA ANALYSIS PROCEDURE: Dependent variables are pain, physical performance measures, muscle strength, quality of life, patient satisfaction, Whole-Organ Magnetic Resonance Imaging Score (WORMS). Independent variables are PRP and HA. Sample size is determined as a minimum clinical important difference (MCID) of 1.33 on a scale of 0-10 on the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC). It is calculated that at least 46 patients should be in each group as a total sample size of 92 with 0.80 power and a standard deviation of 2.25 by taking into account the MCID in WOMAC. Considering the drop outs, it is decided to take 60 patients for each group and 120 patients in total. Statistical analysis is performed using SPSS version 18. The frequency analysis is used for the demographic data, and the variance homogenity and distribution of the variables are analysed with Levene test and Kolmogorov Smirnov test respectively. The mean and standard deviation are calculated for the baseline outcome measures which are normally distributed and homogen and independent t-test is used to compare the groups. Median, minimum and maximum values are calculated for the baseline outcome measures which are not normally distributed and non-homogen and Mann-Whitney U test is used for the group comparison. The significance is set as p<0.05. The effect of time and group is analysed by using repeated two-way analysis of variance (RM-ANOVA) or two-way analysis of covariance (RM ANCOVA) taking into account the difference in baseline measurements. In RM-ANOVA or RM-ANCOVA tests, the selection of univariate or multivariate approach is selected according to the data distribution, the sphericity and epsilon value. A multivariate approach is chosen for variables which are not normally distributed. The univariate approach is used when the sphericity test is not significant in variables that fulfill the normal distribution assumption. When the sphericity test is significant, the cut-point of epsilon value which is 0.750 is checked. If the epsilon is ≥ 0.750, the highest value of univariate test approach and if the epsilon <0.750 the multivariate approach is recorded. The effect sizes calculated are recorded for group and time effects for RM-ANOVA and RM-ANCOVA tests. The effect sizes are interpreted as follows: negligible effect of between 0.00-0.10, the small effect of between 0.10-0.30, the medium effect of between 0.30-0.50; high effect of between 0.50-0.70; very high effect of between 0.70-0.90, and perfect effect of between 0.90-1.00. The post hoc Bonferroni test is used to compare the group effect and time effect. Significance value for the pain, the quality of life, the muscle strength, the physical function measures, and the patient satisfaction is p <0.01 (0.05/5), and for MRI measure is p <0.025 (0.05/2). MRI results by the two independent researcher (YK and SY) and baseline knee radiology are evaluated by the two independent researcher (SY and NFT). Interobserver reliability for MRI and baseline knee radiology results is calculated by using Pearson or Spearman correlation according to the distribution properties of the variables. The correlation between pain, physical function, muscle strength and magnetic resonance change will be calculated by Pearson or Spearman correlation according to the variables distribution properties. Mean imputation technique is used for missing values on a variable. All of the statistical analysis is performed by NFT and is checked by Full Software Engineering Advertising Promotion Tourism Organization Industry Trade Limited Company.

Inclusion Criteria:

• Between 50-70 years of age
• Kellgren Lawrence (KL) radiological grade II-III knee OA
• Sedanter
• Unilateral knee pain
• Volunteer

Exclusion Criteria:

• KL radiological grade ≥2 of contrlateral knee
• Thrombocytopenia (<150.000/mm3)
• Coagulopathy
• Osteoarthritis of multiple joints
• Acute exacerbation of any joint
• Intraarticular steroid, PRP or HA injection within 6 months
• Non-steroidal anti inflammatory drugs or steroid use within 6 weeks
• Diabetes and cancer
• Severe anemia (hemoglobin <10)
• Presence of heart and lung disease that can not perform physical function tests
• Secondary inflammatory arthritis
• Knee surgery or knee injury
• Planning for joint replacement within the next 3 months
• Difficulty in answering the questionnaires
• Limitation in joint range of motion
• Presence of genu varum / valgum >50.

There are five investigators in this study. N.Fusun Toraman is project manager and blind for the study groups. Yasemin Karaman and Sevim Yıldız are radiologists and blind for the study groups. Meral Bilgilisoy Filiz and Aslı Karadag Ozdemir know the groups, they will select the participants, and Meral Bilgilisoy Filiz will apply intraarticular injections via ultrasound. The participants of the groups do not know each other.