PATELLAR AND FEMORAL BONE MORPHOLOGY AND ITS ASSOCIATION WITH LOADING IN YOUNG ADOLESCENT BOYS AND GIRLS

INTRODUCTION

High levels of physical activity or high body mass index (BMI) during growth may negatively influence bone and cartilage, but little is known about how loading relates to the shape of the patella and femur. It is well established that bone shape is primarily determined during growth, and specific variations in bone shape are associated with a higher risk of osteoarthritis (OA). Therefore, we aim to identify the association between the 3D shape of the patella and femur bones and loading factors (i.e., body mass index (BMI) and sports participation) in young adolescents.

OBJECTIVE

Our objective is twofold: 1) to determine the differences in bone shape between boys and girls, and 2) to determine which bone shape variations are associated with loading parameters (i.e., BMI and sports participation).

METHODS

Data from 1912 participants, aged 14.1 (SD: 0.67), standardized BMI for age and sex (BMI-SDS) of 0.42 (1.20), were retrieved from the Generation R study. The Generation R study is a large population cohort study that follows children from fetal life until adulthood. A subset of participants who underwent knee MRI (3.0T, Discovery MR750w, GE Healthcare, Milwaukee, WI, USA) at the 13-year follow-up time point were included in the current study. Imaging was performed with two knees in full extension using a water excitation Gradient Recalled Acquisition in Steady State sequence. Patellae and distal femora were automatically segmented using a method that combines multi-atlas and appearance models. Two statistical shape models (SSM) were built based on the automatically segmented left and right patellae and femora. Shape modes explaining at least 1% of the total population variation were included in the analyses. Differences between boys and girls were determined using a 2-sample T-test. Generalized estimating equation models, separate for boys and girls, were used to analyze the association between BMI-SDS, sports participation (yes or no), and shape variation. Bonferroni correction was used to correct for multiple testing.

RESULTS

A total of 3638 patellae and 3355 femora were included in the shape models. Eleven patellar and fourteen femoral shape modes explained at least 1% of the total variation and were retained for analysis. Eight out of the eleven (modes 1-4, 6, 8, 10, and 11) patellar and twelve out of the fourteen (modes 1-10, 12, and 14) femoral shape modes showed significant differences between boys and girls. Four patella and two femur modes were significantly associated with BMI in both boys and girls, while four patella and seven femur modes were significantly associated in either boys or girls only (Table 1). Patella shape mode 1 was significantly associated with sports participation in both boys and girls, as well as BMI in boys only. Femur shape mode 1 was associated with sports participation in girls and BMI in both boys and girls (Table 1; Figure 1). Femur mode 3, explaining variation in medial anterior condyle and epicondylar width, was only associated with BMI in boys. A higher BMI was associated with a slightly thicker medial anterior condyle and a narrower epicondylar width. Furthermore, patella mode 4, explaining variation in thickness of the dorsal side of the medial apex, was only associated with BMI in girls, with higher BMI being associated with a thicker dorsal side of the medial apex.

CONCLUSION

Multiple differences in shape variations between boys and girls were identified. Sports participation was associated only with the shape mode, explaining size, whereas BMI demonstrated more associations, indicating that BMI is a more important factor. However, sex-specific differences exist. For instance, a higher BMI was associated with a thicker dorsal medial apex in girls, but not in boys. In contrast, a slightly thicker medial anterior condyle was associated with BMI in boys, but not in girls. These varying relationships between BMI and shape in boys and girls highlight the necessity of conducting shape analyses separately during growth.