Diagnostic accuracy of ultrasonography of the distal sesamoidean ligaments (DSLs) has been reported as poor compared to MRI. Higher ultrasound frequency equals higher spatial resolution and may therefore result in improved diagnostic accuracy. To assess and compare diagnostic accuracy for two different ultrasound frequencies compared to 0.31T MRI for injuries of the DSLs and suspensory ligament branches (SLBs). Horses with metacarpo-/metatarsopahalangeal region lameness were included. The metacarpo-/metatarsopahalangeal region was examined ultrasonographically using a 9MHz (US9) and a 16MHz (US16) transducer and subsequently by 0.31T MRI. Cross-sectional area-measurements (CSAs) at specific zones were compared using two tailed paired T-tests and Wilcoxon signed rank test. Sensitivity (Se), specificity (Sp), negative- (NPV) and positive predictive values (PPV) were determined with 0.31T MRI as the gold standard. 13 horses were included. A total of 17 SLB and 16 DSL injuries were identified with MRI. US9 identified 8/17 SLB-injuries (Se=0.47, Sp=0.88, NPV=0.44, PPV=0.89) and 4/16 DSL-injuries (Se=0.25, Sp=0.93, NPV=0.85, PPV=0.44). US16 identified 10/17 SLB-injuries (Se=0.59, Sp=1, NPV=0.56, PPV=1) and 8/16 DSL-injuries (Se=0.50, Sp=0.96, NPV=0.9, PPV=0.72). CSA measurements did not differ between US transducers. CSA measurements differed significantly between US (9+16MHz pooled) and MRI (p<0.05, median difference=0.298cm2). US16 had higher diagnostic accuracy than US9 and a high frequency transducer should be preferred for examinations of the SLBs and DSLs. CSA differs significantly between US and MRI, hence comparisons between modalities should be made with caution. Low number of horses included. Low-field MRI was used as gold standard. Higher diagnostic accuracy obtained with higher ultrasounds frequency.
