Value of low-field magnetic resonance imaging in diagnosing meniscal tears in the canine stifle: a prospective study evaluating sensitivity and specificity in naturally occurring cranial cruciate ligament deficiency with arthroscopy as the gold standard.

Abstract

OBJECTIVE: To evaluate the sensitivity and specificity of low-field magnetic resonance imaging (lfMRI) for detection of meniscal tears in the canine stifle. STUDY DESIGN: Double-blinded prospective clinical study. ANIMALS: Forty-two consecutive stifles of dogs (>or=20 kg; n=34) with clinical and radiologic signs suspicious for cranial cruciate ligament (CCL) insufficiency. METHODS: Each stifle had 7 predefined lfMRI sequences using a 0.5 T magnet with a human knee coil. After lfMRI, diagnostic arthroscopy was performed by 1 surgeon unaware of the MRI findings. After completion of the study MRI images were read by 1 investigator, unaware of the intraoperative findings. RESULTS: At arthroscopy, 22 stifles had meniscal tears requiring subtotal meniscectomy. Of these only 14 were identified by lfMRI. Overall sensitivity and specificity of lfMRI for detection of meniscal tears were 0.64 (95% confidence interval [95% CI]=0.43, 0.80) and 0.90 (95% CI=0.70, 0.97), respectively. Positive and negative predictive values were 0.88 (95% CI=0.64, 0.97) and 0.69 (95% CI=0.50, 0.83), respectively. Neither the state of dislocation of vertical longitudinal tears nor the amount of CCL rupture had an influence on lfMRI accuracy (P=1.00). CONCLUSION: lfMRI was of low diagnostic accuracy in detecting meniscal tears, when compared with arthroscopy. Especially negative lfMRI findings should be interpreted with caution. CLINICAL RELEVANCE: When using lfMRI as a noninvasive preoperative screening tool for the diagnosis of meniscal tears, a high percentage of missed meniscal tears has to be expected.