2-1. study design
This study was approved by the Institutional Review Board and was conducted as a prospective cohort study. The surveillance started in April 2016, and based on the collected preliminary data, the data collection was modified from April 2017. Thus, the data used for the analysis in the current study were from April 2017 to February 2020.
Students who were enrolled in the 3rd to 9th grades of elementary and junior high schools affiliated with Chiba University in April 2017 were eligible for enrollment in the study. Written informed consent was obtained from all participants and their guardians before participating in the study. None of the students skipped or repeated the grades during the study period. Participants with lower limb trauma at the time of baseline examination, musculoskeletal or neurological disorders that made it impossible to perform the physical examination alone or to walk independently, history of lower limb surgery, or those without baseline data regarding the presence of AKP, were excluded (Fig. 1).
2-2. Data collection
Data were collected at school under the supervision of teachers when required. Subject demographic data such as sex and age were recorded at the time of participation in this study. Their height and weight data were recorded in January, April, and August of each year.
At the beginning of each fiscal year in April, from 2017 to 2019, the participants underwent a direct musculoskeletal examination by orthopedic surgeons and physical therapists. Each examiner received three to four training sessions prior to the medical assessment for accurate and uniform evaluation. The examination was conducted by at least two examiners per subject to allow one examiner to record the measurement, while the other performed the test.
The presence of tenderness was examined by gentle palpation of the supra- and infra-patellar poles and the tibial tubercle in each knee (14). These three inspection sites were chosen as they are anatomically easy to define and are common sites of AKP in children— often diagnosed as OSD or Sinding-Larsen-Johansson disease (SLJD) (13). AKP was marked as positive if the subject felt pain to any degree by gentle palpation in at least one of the sites.
Three items were measured to assess lower limb tightness, as previously reported: 1) heel-buttock distance (HBD, cm) (11), 2) straight leg raising angle (SLRA, degree) (11), and 3) dorsiflexion angle of the ankle joint at the knee in extended position (DFA, degree) (15). With the subject in the prone position, the examiner measured the HBD by bending the subject’s knees, individually, as far as possible until the examiner felt resistance. The distance from the heel to the buttock was measured using a standard ruler, and the distance was recorded in centimeters to the first decimal point. Thereafter, the subject was placed in a supine position, and each leg was raised with the knees extended to measure SLRA, individually. The angle of the inspection table with the femoral shaft was measured using a large custom-made protractor and recorded in one-degree increments. Finally, the maximum DFA was measured on each side with the knee extended by setting the stationary arm of the goniometer parallel to the fibular shaft and the movement arm parallel to the fifth metatarsal, and was recorded in one degree increments.
In addition to the direct examination, the participants were required to answer a self-reported questionnaire each month to collect data on the presence of AKP and the degree of physical activity. In this questionnaire, each participant palpated the three inspection sites of their own knees and recorded if they felt pain. The participants received instructions with photos of the inspection sites prior to the study. Physical activity was quantified using the Hospital for Special Surgery Pediatric Functional Activity Brief Scale (HSS Pedi-FABS) (16). The HSS Pedi-FABS is a patient-reported outcome measure, with eight validated items designed to quantify movement in children aged 10–18 years (17). The scale ranges from 0 to 30, depending on how physically active the respondents were during the past month.
2-3. Occurrence and protraction of AKP
Subjects without AKP at baseline were assessed for the occurrence and protraction of AKP. The occurrence of AKP was defined as the first time the subject indicated positive pain on either point of palpation. If AKP occurred in one knee after the occurrence of the other knee was recorded, the latter knee was not included in the record as an occurrence of AKP. The left and right sides of the knee were recorded individually for each occurrence. If the patient had knee pain in both knees simultaneously, bilateral AKP was considered.
Protracted AKP was defined as when chronic pain that lasted for more than three months or recurrence after the pain had disappeared for more than three months was observed, according to a previous report (18).
2-4. Statistical Analysis
Summary statistics for baseline variables were created using frequencies and proportions for categorical data and mean and standard deviation (SD) for continuous variables.
To analyze the occurrence of AKP in each knee, we used a multivariable nominal logistic regression analysis to calculate the odds ratio (OR) and investigated the effects of each factor. The analysis excluded the occurrence of bilateral AKP. The model was adjusted for HBD, SLRA, and HSS Pedi-FABS. Covariates for adjustment were selected based on clinical significance and previous studies. The analysis was performed for each fiscal year, and subjects with positive AKP were excluded from the analysis for the next fiscal year.
In the analysis of chronic AKP on each side, the odds ratio (OR) was calculated using the generalized estimating equations for the multivariable logistic regression model, and the effect of each factor was investigated. The model was adjusted for age, HBD, SLRA, DFA, and HSS Pedi-FABS. Covariates for adjustment were selected based on clinical significance and previous studies (6). The influence of age was analyzed by comparing each age group from 9 to 14 years, against 8 years as the reference. HBD was analyzed by classifying the participants based on whether their heel could touch their buttock, creating two groups: the heel contact group (HBD = 0mm) and the non-contact group (HBD > 0mm).
All p-values were two-sided. Statistical significance was set at p < 0.05. All statistical analyses were performed using SAS software (version 9.4; SAS Institute, Cary, NC, USA).