Observational Study of Growth-Related Knee Pain in Japanese Footballers Aged 12–15 Years: A Subsequent Series Following Children Aged 8–12 Years

This is a subsequent series of studies that examined the factors for growth-related injuries in children aged 8–12 years. We aimed to explore factors associated with growth-related knee pain (GRNP) in Japanese footballers aged 12–15 years. A total of 600 data (200 football players in 7–9th grades) were collected from their parents. Clinical characteristics (age, height, and weight), sports-specific variables (dominant leg, position, and surface), weekly training volume (total days and hours of football), and experience of GRKP (e.g., Osgood–Schlatter disease) within the past 1 year were assessed using a web-based questionnaire. Independent t -tests and chi-square tests were used to investigate the difference between groups with and without GRKP. Logistic regression analyses were also performed to explore risk factors for youth athletes. Additionally, a receiver operating characteristic curve was constructed for the variables used in the logistic regression model. Among junior high school footballers in Japan, 21.0% had experience with GRNP in the past 1 year. Weekly training hours were significantly and positively associated with GRNP (odds ratio, 1.047; 95% confidence interval, 1.016–1.079). The optimal cut-off value for training hours per week was 12.5 h, with low-to-moderate sensitivity (44.4%) and specificity (73.8%). However, clinical and sport-specific characteristics in Japanese footballers aged 12–15 years were not identified as risk factors for recent GRNP. To decrease the risk of GRNP in young Japanese footballers, controlling training volume, which combines adequate recovery or introducing multi-sports events, may be a key strategy for injury prevention.


Introduction
To proceed with injury prevention strategically, understanding the risk factors and mechanisms of injuries in each sport and playing category is essential.Risk factors are divided into intrinsic (i.e., age, sex, body composition, and leg dominance) and extrinsic (i.e., environment and surface) components, which render athletes susceptible to injury.Furthermore, the mechanism of sports injury is used to describe the inciting event (i.e., playing/sports situation) directly associated with the onset of injury.Exposure to matches and training (total load) have been introduced as practical examples of inciting events for lower leg stress fractures in football (Bahr & Krosshaug, 2005).This scenario is what is called "overuse injury." Football is popular among many generations worldwide; however, players of different ages and circumstances are not at equal risk of injuries.Children and adolescents are not the same as adults, and young football players often suffer from growth-related injuries (Faude et al., 2013;Juch et al., 2023;Rössler et al., 2016).Apophyseal injuries are common in young footballers; however, the incidence of different apophyseal injuries varies substantially, depending on the location and age group.For Observational Study of Growth-Related Knee Pain in Japanese Footballers Aged 12-15 Years Sasaki and Nagano example, the peak incidence of foot-ankle time-loss apophyseal injuries (e.g., severe disease) occurs in patients aged <11 years, whereas those of the knee (e.g., Osgood-Schlatter disease) occur in patients aged <14 years (Materne et al., 2022).Another study also showed similar results that Osgood-Schlatter and Sever's diseases had their peak prevalence in football players <13-14 years and <11 years, respectively (Le Gall et al., 2006;Price et al., 2004).Furthermore, knee overuse injuries, including the Osgood-Schlatter disease, for children aged 9-14 years represented the greatest burden on better performance, and complications were more likely reported by older players (12-14 years) than younger players (9-11 years) (Leppänen et al., 2019).Therefore, young footballers aged 13-14 years are the main targets for the prevention of growth-related knee injuries.
Recently, factors associated with growth-related knee pain (GRNP) in Japanese footballers <12 years have been explored (Iwame et al., 2019;Sasaki et al., 2023).Data from East Asia are valuable because almost all epidemiological reports on growth-related or overuse conditions in children are from Western countries (Gaulrapp & Nührenbörger, 2022;Materne et al., 2022;Post et al., 2020;Rössler et al., 2018).Additionally, previous studies in Japan could not target the main age at which GRNP occurs most frequently (approximately 13-14 years old), although they could identify the target age for growth-related heel pain (around 11 years old).Therefore, this study aimed to explore the factors associated with GRNP in Japanese footballers aged 12-15 years as a study series following a previously published study in children aged 8-12 years old (Sasaki et al., 2023).In the cultural context of Japan, 3 rd to 6 th grade children aged between 8-12 years are categorised as elementary school students, whereas 7 th to 9 th grade adolescents aged 12-15 years are classified as junior high school students.We hypothesised that (1) experience of GRNP within the past one year would be higher in 7-9th grade youth footballers compared with those previously reported in 3 rd to 6 th grade children and that (2) body size (e.g., height and weight) and training volume (e.g., playing hours/days per week) in footballers aged 12-15 years would be linked to the prevalence of GRNP, a trend that is common to footballers aged 8-12 years in Japan.

Data Collection and Setting
The present study was a cross-sectional study based on data from children in 3 rd to 6 th grade; therefore, detailed survey items (screening and main questionnaires) have been described previously (Sasaki et al., 2023).Anonymous online questionnaires in a web-based survey (Rakuten Research, Inc., Japan) were used for data collection.Each parent of footballers in 7 th to 9 th grade answered information about their child, such as clinical characteristics (age, grade, weight, and height), sportsspecific variables (dominant leg, playing position, and surface), weekly training volume (total days and hours football), and experience of GRNP (e.g., Osgood-Schlatter disease) within the past 1 year.
Detailed data setting for recruiting target participants is shown in Fig. 1.In the present study, the desired sample size (a total of 600 data, 200 footballers in each 7 th to 9 th grade) was achieved over three days (August 1-3, 2022).We confirmed the required sample size for t tests (d = 0.5, α level = 0.05, 1-β level = 0.95, allocation ratio = 5) and chi-square tests (w = 0.3, α level = 0.05, 1 − β level = 0.95, larger df = 7) by using prior power analyses.Finally, participants with GRNP and without injury were set at 126 and 389, respectively.This study was approved by the Ethics Committee of Tokyo Ariake University of Medical and Health Sciences (approval no.0382).

Statistical Analysis
Descriptive statistics were expressed by continuous and categorical variables.Definitions of each category were based on a previous series of 3 rd to 6 th grade football players (Sasaki et al., 2023).Independent t-tests and chi-square tests were used to compare the difference between the two groups (with GRNP vs. without knee injury).Effect sizes for univariate analyses were calculated using Choen 's d (<0.2,trivial;small;medium;and >0.8,large) and Cramer's V (<0.1,trivial;small;medium;and >0.5,large).Logistic regression analysis was also performed, and odds ratios (OR) and 95% confidence intervals (CI) were shown as the results.Variables with p < 0.05 in the univariate analysis (independent t-tests and chi-square tests) were entered into a multivariate model.Additionally, a receiver operating characteristic (ROC) curve was constructed to investigate the sensitivity and specificity of the variables used in the logistic regression model.Areas under the ROC curve and 95% CI were determined, and the optimal cut-off values were calculated.All statistical analyses were performed using SPSS Statistics version 22.0 for Windows.

Results
Experience of GRNP was observed in 21.0% (126 of 600 players aged 12-15 years; see Fig. 1).Table I shows the characteristics of youth footballers with and without GRNP within the recent 1 year.Players with GRNP had more training days (p = 0.011, d = 0.259) and hours (p = 0.002, d = 0.323) per week than those without knee pain.In contrast, clinical (age and body size) and sports-specific (dominant leg, position, surface) variables did not differ among the two groups.Observing the logistic regression model, more weekly training hours (OR = 1.047; 95% CI, 1.016-1.079;p = 0.003) were significantly associated with the experience of GRNP (Table II).Weekly training days were excluded from the final logistic regression model due to the multicollinearity (ρ = 0.645, p < 0.001 for Spearman's rank correlation coefficient).The area under the ROC curve for training hours per week was 0.612 (95% CI, 0.555-0.669),and the optimal cut-off value was 12.5 h per week (sensitivity, 44.4%; specificity, 73.8%) for Japanese football players aged 12-15 years (see Fig. 2).

Discussion
The present study is a subsequent series of studies that examined the factors associated with GRNP in 3 rd to 6 th grade footballers (aged 8-12 years) (Sasaki et al., 2023).Among the 600 young Japanese

Sasaki and Nagano
Observational Study of Growth-Related Knee Pain in Japanese Footballers Aged 12-15 Years footballers in the 7-9 th grade (aged 12-15 years), 21.0% (126 players) had experienced GRNP in the past 1 year.The prevalence of GRNP in the 7 th to 9 th grades was approximately two-fold higher than that in the 3 rd to 6 th grades (11.2%, as shown by Sasaki et al., 2023), which could prove our first hypothesis.Another study (Iwatsu et al., 2023) also investigated the point prevalence of knee pain among young Japanese players (7,234 participants from several sports) aged 6-15 years and showed that 13-year-old athletes had the highest rate of knee pain at 19.1%, which is similar to the present results.In an epidemiological study conducted in a Middle Eastern country (Qatar), the prevalence and occurrence rates of apophyseal knee injuries in young elite footballers aged 9-18 years were a median of 33.2% and 13.1 years old, respectively (Materne et al., 2022).In Qatar, the prevalence of apophyseal injuries in the knee joint was >30%, which was a surprisingly high proportion when compared with that in the present results; this may be due to the difference in study design, type of participants, and cultural context.Knee overuse problems, including apophysitis, lead to a great burden on both participation and performance in youth competitive football (Leppänen et al., 2019); therefore, coaches in the 13-14 years category should be aware of growth-related knee injury in order to prevent it.
Observing the clinical characteristics of young Japanese footballers aged 12-15 years, participants with knee pain did not differ in age and body size from those without painful episodes.Furthermore, sport-specific factors (dominant leg, position, and surface) were not associated with any episodes of GRNP among young Japanese football players.These results partially differ from those of our second hypothesis.According to a previous study series targeting 3 rd to 6 th grade children (Sasaki et al., 2023), age, height and weight were significantly associated with GRNP; however, this trend is not consistent with the present results for 7 th to 9 th grade children.A recent study by Murayama et al. (2023) showed that the actual age of peak height velocity (PHV) was 12.78 ± 0.93 years according to data of 207 Japanese boys observed over 12 years (2,484 observations), which is similar to the age of PHV at 12.61 years in Greek children (Kleanthous et al., 2022).Kleanthous et al. (2022) also observed that growth acceleration (take-off) began at the age of 9.86 years.Thus, 3 rd to 6 th grade Japanese children reported in a previous series (Sasaki et al., 2023) were just fit in the growth spurt for body size, while present youths in 7 th to 9 th grade are beginning to experience a decrease in body size velocity.Therefore, for youth footballers aged 12-15 years, chronological age or body size, such as height and weight, would not be principal risk factors for GRNP in Japan.Meanwhile, dominant leg, position, and surface did not affect the occurrence of GRNP throughout the 3 rd to 9 th grades (i.e., all footballers aged 9-15 years).Common views between previously reported children (3 rd to 6 th grade) and present youth (7 th to 9 th grade) footballers in Japan, the proportion of right-in-leg dominance (87.1% and 87.3%, respectively), playing as fielders (92.6% and 91.4%, respectively), and usually on the clay ground surface (77.0% and 77.1%, respectively) did not differ among school categories.In other words, circumstantial context may not affect the experience of GRNP in Japan.
Focusing on the training volume for youths in Japan, players with recent GRNP had significantly more weekly days and hours than those without knee pain.The training volume of 7 th to 9 th grade youth with knee pain was 1.5 days and 4.1 h on average, larger than those of 3 rd to 6 th grade children with knee pain shown in the previous report (Sasaki et al., 2023), which partly agreed with our second hypothesis.Additionally, when using the logistic regression model, more weekly training hours were identified as a significant risk factor for GRNP (OR = 1.047; 95% CI, 1.016-1.079).This means that for an increase of 1 h in training volume per week in Japanese footballers aged 12-15 years, the corresponding increase in the probability of GRNP within 12th month was 4.7%.Previous reports on footballers aged 8-12 years also showed a similar trend, in which more training hours per week (Sasaki et al., 2023) or training for >10 h per week (Iwame et al., 2019) were associated with a painful episode in the knee using logistic regression analysis.In youth sports, several recommendations regarding sports participation volumes have recently been made to prevent overuse injuries.One of these approaches is that young athletes should participate for fewer hours per week than their age or <16 h per week (LaPrade et al., 2016).In the present study, the cut-off value of training volume for footballers aged 12-15 years was 12.5 h per week using the present ROC analysis, which agrees with the recent guidelines shown by LaPrade et al. (2016).Jayanthi et al. (2015) also reported that injured young athletes (14.1 ± 2.1 years old, mean ± SD) significantly participated in more total hours of organised sports activities compared with uninjured athletes (11.2 vs. 9.1 h per week, on average).Additionally, current recommendations suggest that children who specialise in single sports activities should be closely monitored for indicators of burnout, overuse injury, or potential decrease in performance (LaPrade et al., 2016).A reason why increased sports participation volume in childhood is the sports specialisation, which is defined as year-round intensive training in a single sport with the exclusion of other sports (Jayanthi et al., 2013).In Japan, >75% of junior high school-aged footballers perform a single sport (football only); this proportion is much higher than that of elementary school-aged footballers at 43% (Nagano & Oyama, 2023a).Nagano and Oyama (2023a) also revealed that the appearance of overused knee injuries in junior high school was significantly associated with weekly sports participation, which was common in the present Observational Study of Growth-Related Knee Pain in Japanese Footballers Aged 12-15 Years Sasaki and Nagano study.Especially in Japan, specialised groups who continued a single specific sport in team sports had a greater prevalence of overuse injuries than a non-specialised group who previously participated in several sports, a trend that differed from that of individual sports (i.e., track and field and swimming; Nagano & Oyama, 2023b).Therefore, youth footballers categorised in team sports should control their training volume (combining adequate recovery) or introduce multisport events (other than football) to decrease the risk of overuse injuries, including growth-related pain.
Our study has several limitations.First, this was a cross-sectional study using a self-report questionnaire completed by parents, which might have included some bias.Second, a specific physician did not diagnose the knee symptom; therefore, GRNP in the present study may contain various conditions.However, we believe that a study series following 3 rd to 6 th grade football players focusing on GRNP is crucial.This challenge resulted in 1,800 Japanese children aged 8-15 years.Third, the outcome of the ROC curve analysis was not an excellent model (area under the curve, 0.612), with 44.4% sensitivity and 73.8% specificity.The present cut-off value of training volume (12.5 h per week) may be relatively suitable for detecting players without knee pain but not for detecting painful occasions.

Conclusions
The present study aimed to explore the factors associated with GRNP in Japanese footballers aged 12-15 years.Among junior high school footballers, 21.0% had experiences of GRNP in the past 1 year.More weekly training hours were significantly and positively associated with GRNP (OR, 1.047; 95% CI, 1.016-1.079).Additionally, the optimal cut-off value for training hours per week was 12.5 h, with low-to-moderate sensitivity (44.4%) and specificity (73.8%) when explored using ROC curve analysis (area under the curve, 0.612; 95% CI, 0.555-0.669).However, clinical (age and body size) and sportspecific (dominant leg, position, and surface) characteristics were not associated with any episode of GRNP.To decrease the risk of GRNP in young Japanese footballers, controlling the training volume, which combines adequate recovery or introducing multi-sports events, may be a key strategy for injury prevention.

Fig. 2 .
Fig. 2. Receiver operating characteristic curve for training hours per week.AUC: area under the curve; CI: confidence interval.

TABLE I :
Characteristics of Japanese Footballers (Aged 12-15 Years) with and without GRKP Note.SD: standard deviation; Cat: category; ES: effect size; GRNP: growth-related knee pain.