Reproducibility and agreement of radiographic assessment of carpal deformities in foals

The Pivot Point (PP) method is commonly used in the radiographic assessment of carpal deformities in young foals, as the range of deviation may influence treatment choice. The aims of this study were to assess the intra- and interobserver reproducibility of the PP method and subjective radiographic evaluation without line drawing to establish the anatomical site responsible for carpal deviation in foals and to evaluate the agreement between these two techniques.

Anonymized radiographs of foals presented for investigation of carpal deformity or prematurity were retrospectively and independently reviewed by six readers. Readers were first asked to subjectively identify the origin of the deviation and then apply the PP method and calculate the angle of deviation (PP_Angle). A second reading in a different randomized order was performed at least two weeks after the first reading. The carpi with the highest variability in PP_Angle measurements were reviewed in consensus by two other radiologists who did not perform the measurements.

A total of 52 radiographs from 25 foals were selected. Good intraobserver reproducibility was observed for all variables, with no significant differences between the first and second readings by the same reader. Measurement of the angle of deviation using the PP method had a high intraobserver reproducibility (correlation coefficient of 0.93, p < 0.05). PP and subjective evaluations revealed strong intraobserver reproducibility for the origin of deviation (Cramer coefficients of 0.4 and 0.5, respectively; p < 0.05). There was strong agreement between PP and subjective evaluation for establishing the origin of deviation for all readers (Cramer coefficient 0.41; p < 0.05). Conversely, interobserver reproducibility for PP and subjective evaluations was low (Kappa values of 0.26 and 0.20, respectively; p < 0.05). Higher variability of PP_Angle was found in limbs with lateral bowing of the distal radius.

The results of this study suggest that the PP method can reliably be used by the same reader for follow-up of carpal deformities and that there is no need to draw lines if the only required information is the origin of deviation, while measurements by different readers on the same patient may be misleading.