Structural ultrasound of joints and tendons in healthy children: development of normative data

Musculoskeletal ultrasound is a well accessible technique to assess disease activity in children with juvenile idiopathic arthritis. Knowledge of reference values of joint structures is indispensable to differentiate between physiological and pathological finding. The aim of this study was to assess the structural sonographic features of joints and tendons in healthy children from several age groups (0.2-18 year), and develop a set of normative data.

Greyscale ultrasound was performed in 500 healthy children (age 0.2-18 years) according to a predefined scanning protocol (Additional file 1) including the shoulder, elbow, wrist, second metacarpophalangeal joint, hip, knee, ankle, and first metatarsophalangeal joint). Demographic data and values of cartilage thickness, tendon diameters, and the degree of capsular distention measured by bone-capsular distance (BCD) were collected. Differences according to the sex were assessed by unpaired t-test. Single and multiple regression analyses were performed between the ultrasound outcomes and covariates such as age, height, weight and body mass index. Growth charts and tables were developed with respect to age. Nonparametric quantile regression was applied using the R-packages quantreg and quantregGrowth.

A total of 195 male and 305 female volunteers were included between the age of 0 and 18 years (mean age 8.9; range: 0.2-17.9 years). Cartilage diminished markedly as children aged, and cartilage of the boys was significantly thicker compared to the girls in all joints (p < 0.001). In addition, cartilage became thinner as children's height and weight increased (beta regression coefficients between - 0.27 and - 0.01, p < 0.0001). Capsular distention (i.e., BCD > 0 mm) was uncommon in the ankle, wrist and MCP2 (resp. in 3, 6, and 3% of cases). It was more common in the suprapatellar and parapatellar knee, MTP1 and posterior recess of the elbow (resp. in 34, 32, 46, and 39% of cases). In the hip, some capsular distention was always present. Age was found to be the best predictor for BCD (beta regression coefficients between 0.05 and 0.13, p < 0.0001). Height was, in addition to age, a good predictor of tendon diameter (beta regression coefficients between 0.03 and 0.14, p < 0.0001). Growth curves and tables for each variable were developed.

Reference values of sonographic cartilage thickness, BCD and diameters of tendons at several joints were established from 500 healthy children, aged between 0.2 and 18 years. Growth charts and tables were developed to distinguish normal findings from pathology in children with complaints suspicious of arthritis.