This was a retrospective study comparing HAi+rESWT with HAi alone for the management of sPFOLP. All subjects were treated at the outpatient department of the Center for Joint Surgery, Southwest Hospital, Third Military Medical University (Army Medical University), Chongqing, China between January 1st, 2014 and January 31st, 2018.
This retrospective study was approved by the Ethics Committee of the Southwest Hospital, Third Military Medical University (Army Medical University), Chongqing, China (No. KY201915). All procedures that involved human participants were performed in accordance with the ethical standards of the institutional and national research committee and with the 1975 Helsinki declaration and its later amendments or comparable ethical standards. Written informed consent was obtained from all individual participants included in this study.
All female and male adults aged 18-65 years with unilateral sPFOLP who were treated with respectively HAi+rESWT or HAi alone at the outpatient department of the Center for Joint Surgery, Southwest Hospital, Third Military Medical University (Army Medical University), Chongqing, China between January 1st, 2014 and January 31st, 2018 were eligible for being enrolled in this study. There were no subjects in the hospital's database who had suffered from sPFOLP and were treated with rESWT alone.
Diagnosis of sPFOLP was based on the following criteria: unilateral or bilateral pain of the knee joint for more than three months, particularly when going up and down the stairs and relieved after rest; aggravation of symptoms over time; no history of perceived local trauma; presence of a focal patellar bone marrow edema and/or localized osteonecrosis evidenced by MRI; and exclusion of rheumatoid arthritis, gout, tuberculosis and other inflammatory diseases.
Exclusion criteria for being enrolled in this study were: bilateral sPFOLP (because in case of bilateral sPFOLP pain perception could be different from pain perception in case of unilateral sPFOLP), joint degeneration with radiographs showing narrow joint space / Kellgren-Lawrence Grades II to IV, valgus or varus deformity >5 degrees of the knee, previous treatments with intra-articular injections and/or ESWT (both radial and focused), contraindications of rESWT (signs of local infection and/or tumor, serious blood dyscrasia, blood-clotting disorders, treatment with oral anticoagulants) and incomplete availability of radiographs and MRI images of the affected knee joint in the hospital’s database for confirmation of diagnosis and evaluation of treatment outcome.
All subjects diagnosed with sPFOLP were provided with a detailed description of the benefits and potential risks of treatment selection prior to treatment. All subjects were offered HAi+rESWT or HAi alone; no subject was offered rESWT alone. In select cases subjects were offered surgery (not considered in this study). In addition, all subjects received education with special attention to reduction of going up and down the stairs, squatting and specific activities, as well as non-opioid analgesics (Celecoxib 200 mg, oral, once per day) if necessary.
We identified n=36 subjects in the hospital's database with unilateral sPFOLP who had made an informed decision for being treated with HAi+rESWT. These subjects received five intra-articular injections of 2.5 mL HA (ARTZ Dispo, Seikagaku Corporation, Tokyo, Japan) each into the affected knee joint, with one injection per week. In addition, subjects received rESWT immediately before each HA injection, which was performed with the Swiss DolorClast device (EMS Electro Medical Systems, Nyon, Switzerland) using the EvoBlue handpiece and the 15-mm applicator. Each rESWT session consisted of 2000 radial extracorporeal shock waves (rESWs) applied to the area of the patellar osteochondral lesion as determined by MRI in sagittal and axial positions. Tender points around the patella served as auxiliary treatment areas. The knee flexion was 45 degrees during rESWT. The positive energy flux density (ESW+) of the rESWs was between 0.04 and 0.08 mJ/mm2, corresponding to an air pressure of 1.8-2.5 bar. The rESWs were applied at a frequency of 6-8 rESWs/second. No anesthesia or analgesic drugs were applied during the rESWT sessions. All rESWT sessions were performed by the same therapist.
Furthermore, we identified n=45 subjects in the hospital's database with unilateral sPFOLP who had made an informed decision for being treated with HAi alone. Injection of HA was performed as described for the subjects who were treated with HAi+rESWT.
Baseline characteristics of the subjects are summarized in Table 1.
Outcome measurements and assessments
The primary clinical outcome was pain during normal daily life activities, measured on a visual analogue scale (VAS) ranging from zero (no pain at all) to ten (maximum, unbearable pain). The VAS pain score was assessed by the same examiner at baseline (BL) and six weeks (W6), three months (M3) and six months (M6) after BL, as well as during a last follow-up examination (LE) that took place at 37.6 ± 1.7 (mean ± standard error of the mean; SEM) months after BL (range, 12-59 months).
Treatment success was defined as individual improvement of the VAS pain score by at least 60% at M3.
Secondary clinical outcomes were the Western Ontario and McMaster University Osteoarthritis Index (WOMAC) score and measurements of the area of the patellar bone marrow edema (PBME) on magnetic resonance imaging (MRI) scans of the knee. The WOMAC score was assessed by the same examiner at BL, W6, M3, M6 and LE. Magnetic resonance imaging was performed with an 0.2 Tesla Artoscan-C device (Esaote, Genoa, Italy) at BL, M3, M6 and LE. The Materialise Mimics software (Version 19.0; Materialise, Leuven, Belgium) was used to determine the maximum PBME area on sagittal, fat suppression sequence images by two independent, experienced radiologists, whose results were averaged.
For all investigated variables, mean and SEM were calculated at BL, W6, M3, M6 and LE. This was separately performed for the subjects who were treated with HAi+rESWT and the subjects who were treated with HAi alone. Because this was a retrospective two-cohort study, tests of baseline differences between the two cohorts were performed (according to  tests of baseline differences in randomized controlled trials (RCTs) are illogical). After determining using the D’Agostino and Pearson omnibus normality test whether the distribution of the variables age, body weight, body height and body mass index (BMI) of the subjects in both cohorts at BL were consistent with a Gaussian distribution, baseline comparisons were performed with unpaired Student's t test (mean body weight and mean body height), Mann Whitney test (mean age and mean BMI) and Fisher's exact test (gender). These tests showed statistically significant differences between the two cohorts at BL (Table 1).
Accordingly, changes in mean VAS pain score, mean WOMAC score and mean PBME area over time after treatment were assessed using the general linear model repeated measures (GLM-RM) procedure, with treatment as fixed factor, the investigated variable (VAS pain score, WOMAC score, PBME area) as dependent variable, and the subjects' gender, age, body weight, body height and BMI as covariates. Then, the post hoc Mann Whitney test was applied for pairwise comparisons. Because each data set was used in two statistical analyses (GLM-RM procedure and post-hoc test), in all analyses an effect was considered statistically significant if its associated P value was smaller than 0.025. Calculations were performed using SPSS (Version 26; IBM, Armonk, NY, USA) and GraphPad Prism (Version 8.2.1; GraphPad software, San Diego, CA, USA).
Treatment success (i.e., number of subjects with individual improvement in VAS pain score by at least 60%) was tested with Chi-square test.
No subject was lost during the follow-up period, i.e., all subjects passed the examinations at BL, W6, M3, M6 and LE. As a result, complete capture of all data from all subjects was achieved.