Clinical Diagnostic Value of 5 T MRI for Knee Injuries: A Comparison Study With 1.5 or 3 T

ABSTRACT Background Preliminary studies have shown the potential of 5 T MRI in cardiac, neurovascular, and abdominal imaging. However, the clinical diagnostic value of 5 T for assessing knee injuries remains unclear. Purpose To compare the image quality, anatomic visibility, and diagnostic performance of 1.5, 3, and 5 T in paired MRI knee studies. Study Type Prospective. Population Ninety patients underwent 1.5–5 or 3–5 T paired MRI scans. Field Strength/Sequence 2D T1‐weighted and 2D and 3D fat‐suppressed proton density‐weighted at 1.5, 3, and 5 T. Assessment Three radiologists independently performed a quantitative assessment of signal‐to‐noise ratio (SNR) and a qualitative evaluation of overall image quality, edge sharpness, artifacts, fat‐suppression homogeneity, and anatomic visibility by using a 5‐point Likert scale. The ability to detect knee injuries was determined using arthroscopy results as the reference standard. Statistical Tests Differences between 5 and 1.5/3 T MRI were analyzed using the Wilcoxon signed–rank test or Student's t test. The sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) were used to assess diagnostic performance. DeLong's test was used to determine the significance of AUC differences. Results Median SNRs at 5 T were higher than those at 3 T (56.8–287.1 vs. 47.5–229.1; p < 0.046) and 1.5 T (69.5–258.4 vs. 35.1–188.9; p < 0.001) across all acquisitions. Moreover, 5 T MRI demonstrated significantly better overall image quality and edge sharpness than those demonstrated by 3 T (median range: 3–5 vs. 3–4; p = 0.048 and 3–5 vs. 2–4, p = 0.033) or 1.5 T (3–5 vs. 3–4; p = 0.035 and 3–5 vs. 2–3; p = 0.017). Finally, 5 T MRI showed a higher AUC value than 3 T (94%–100% vs. 81%–90%) and 1.5 T (97%–100% vs. 81%–90%), particularly for trochlear, tibial plateau, and femoral condyle cartilage defects. Data Conclusion Compared to 1.5 and 3 T MRI, it significantly improved image quality, anatomical visibility, and diagnostic performance, especially for subtle cartilage defects. Evidence Level 1. Technical Efficacy Stage 2.