Measurements of equine foot parameters show limited agreement between radiographs and low-field magnetic resonance imaging.

Background: Equine foot radiographs are commonly obtained to measure anatomical conformation parameters. Comparison of measurements between radiographs and low-field magnetic resonance imaging (MRI) has not been extensively explored.

Objectives: To compare foot parameter measurements between radiographs and low-field MRI, and assess the effect of hoof wall markers on visualising the hoof capsule (during MRI) and facilitating measurements.

Study design: Comparative cadaveric analytical study.

Methods: Radiography and MRI of nine equine cadaver front feet were performed with and without hoof wall markers, which were lead strips for radiography and a water-soaked hoof bandage for MRI. Intra-observer reliability and inter-modality agreement were calculated using intra-class correlation coefficients (ICC) with 95% confidence intervals (CI) and p < 0.05.

Results: Intra-observer repeatability was generally good, apart from distal dermal frontal measurements. There was limited agreement between radiographic and MRI measurements. Results are presented as RAD indicating those obtained with radiography and T1, T2* or STIR indicating those obtained with the relevant MRI sequence; m is added if a marker was used. Founder distance only showed good agreement for radiographic and T1 measurements with markers; ICC 0.78 (CI 0.33-0.95 p = 0.004). Inter-modality comparisons for distal phalanx rotation were limited by intraobserver repeatability. Good agreement was noted for sole thickness and epidermal sole thickness measurements with markers; sole thickness (RADm vs. T1m ICC 0.81 [CI -0.04-0.96], p < 0.001; RADm vs. T2*m ICC 0.86 [CI 0.51-0.97], p < 0.001; RADm vs. STIRm ICC 0.91 [CI 0.66-0.98], p < 0.001) and epidermal sole thickness (RADm vs. T1m ICC 0.88 [CI 0.55-0.97], p < 0.001; RADm vs. T2*m ICC 0.83 [CI 0.41-0.96], p = 0.002; RADm vs. STIRm ICC 0.80 [CI 0.31-0.95], p = 0.004). Radiographic measurements with and without markers often had good to excellent agreement; for some parameters, hoof wall markers were associated with reduced intra-observer repeatability. The water-soaked hoof bandage aided MRI hoof capsule visualisation; limitations included reduced repeatability and unattainable distal measurements.

Main limitations: Small sample size.

Conclusions: The limited agreement between radiographic and MRI measurements suggests these modalities are not interchangeable in equine foot assessment. Hoof wall markers do not benefit foot measurements.