Objective. To investigate the frequency of ultrasound (US)-detectable involvement of the subtalar joint (STJ), to compare clinical versus US assessment of the STJ, and to compare different scanning approaches to the STJ in juvenile idiopathic arthritis (JIA). Methods. Clinical and US assessments were performed independently in 50 ankles with clinically active JIA. US abnormalities of the STJ were investigated using a lateral, medial, and posterior scanning approach and scored semiquantitatively. Agreement was tested using kappa statistics. A control group of 10 healthy subjects was examined.Results. Clinical and US evaluations detected synovitis in 24 of 50 (48.0%) and 27 of 50 (54.0%) of STJs, respectively. US detected synovitis in 10 of 26 STJs (38.5%) recorded as normal on clinical evaluation, but was negative in 7 of 24 STJs (29.2%) diagnosed as having involvement on clinical examination. Agreement between clinical and US assessments was fair (k 5 0.32). US abnormalities were more frequently detectable using the lateral scanning approach. All patients with US abnormalities in the medial and/or posterior side of the STJ had also US abnormalities on the lateral scanning approach, but the reverse was not true. Intra-and interobserver agreements for the lateral scanning approach were satisfactory for both detecting involvement and scoring US abnormalities. None of the 17 STJs of healthy controls showed US abnormalities. Conclusion. US may increase the precision of the evaluation of the STJ in JIA. The observed high frequency of STJ involvement on US suggests to include this joint in US scanning protocols devised for children with JIA. Synovitis is more frequently detected using the lateral scanning approach.