All participants provided written informed consent prior to enrolment into the study. All procedures have been approved by the Institutional Review Board of the National Cheng Kung University (NCKUIRB), Taiwan (A-ER-103-139). This prospective study examined the results of the unilateral knee in 28 patients with K-L grade 2 to 3 (11 males and 17 females, aged from 51 to 82-year-old) (Figure 1). All patients consulted for IAHA injection therapy were at least 45 years old and met the American College of Rheumatology (ACR) criteria for knee OA . There were no significant differences in all measured baseline parameters between males and females, detailed description in Table 1. Exclusion criteria were: (1) secondary knee OA following GREES (Group for the Respect of Ethics and Excellence in Science) criteria; (2) a history of knee surgery; and (3) a history of arthrocentesis and an IA steroid or hyaluronic acid injection 3 months before the study. All patients were evaluated by the same observer using the Western Ontario and McMaster University Index (WOMAC) function subscale OA index, and the Visual Analogue Scale (VAS) score before the IAHA, and at week 1, week 2, and week 4 post first IAHA injection. During the study period, no analgesics except acetaminophen were allowed. Moreover, any analgesics were not allowed for 72 h preceding the clinical assessment . Further, the person who did the evaluations was not involved in the IAHA injection to patients.
All patient were given three consecutive weekly injection of 2.5 ml HA (ArtiAid, Maxigen Biotech Inc., Taoyuan City, Taiwan or ARTZDispo, Seikagaku corporation, Takahagi-shi, Japan) into the suprapatellar pouch under ultrasound guidance 3 times weekly (Fig. 2, red arrow). Four times of SF aspiration (Fig. 2, green arrow) was performed during a therapeutic period (Figure 2).
WOMAC was assessed by the investigator before SF aspiration. This questionnaire, first introduced in 1988, is used to assess the health status of patients. It consists of 33 items that evaluate the health and function of the patient from various aspects, including clinical symptoms (5 questions), severity of knee joint stiffness (2 questions), degree of pain (9 questions) and activity of daily living (17 questions). Each question has five subscales where the best situation is scored as never or none and the worst as extreme or always. Here, higher scores are representative of a better health situation and less pain.
Each patient was administrated the VAS by the investigator before SF aspiration. A 100 mm horizontal line served as the VAS and separated the description of “no pain”, graphed as a smiley face, from the other description of “extreme pain”, graphed as a crying face. All patients involved in this study chose scores by free-will without any influence from the investigator.
Ultrasound-guided SF aspiration
SF was aspirated from each study participant by the same investigator under ultrasound guiding  and transported immediately to the laboratory in an icebox at 4°C. The samples were centrifuged for 20 min at 10000 x g, after which the precipitated pellet was discarded and the supernatant pipetted into 2.0 ml Eppendorf tubes and stored at –80°C until analysis.
Determining HA concentration
The HA concentrations of the SF samples were determined by a modified micro-method of the uronic acid assay according to the Morgan-Elsion method . The diluted SF samples were boiled with a sulfuric acid/borax mixture, and after the addition of m-phenylphenol, a pink color developed. The color intensities were then determined in a microplate reader with 560 nm and compared with those of known standard concentrations of HA. All assays of each sample were performed in triplicate to reduce error, the resulting means of which were used for analysis.
Enzyme-Linked Immunosorbent Assay (ELISA)
The concentrations of IL-1β, IL-6, as well as MMP-1, MMP-3, and MMP-13 in the SF were determined by an ELISA detection kit (R&D system, USA) with all procedures following the manual’s instructions. Briefly, a 96-well microplate was coated with a capture antibody overnight then washed by phosphate buffered saline with Tween 20 (PBST) for 3 times. All samples were digested by 4 mg/ml hyaluronidase in a 1: 1 mix, and then incubated at 37°C for one hour . 100 μl of 10-fold diluted SF was loaded into the wells in duplicate and allowed to sit for 2 hours at room temperature. After washing three times, the detection antibody was added and incubated at room temperature for one hour. After washing again, 3,3', 5,5; -tetramethylbenzidine (TMB) chromogen was added and allowed to incubate for 30 minutes. Subsequently, 2% sulfuric acid was added into the wells to stop the chromogenic reaction and measured the absorbance at 405 nm using a plate reader.
Data were expressed as the means ± standard deviation (SD). The paired sample t-test analysis was used to make pairwise comparisons between the groups. Statistical significance was set at p < 0.05.