Work

Optimizing Knee MRI: Diagnostic Performance of a 3D PDW SPAIR-Based Short Protocol

Year: 2025
Type: article
Abstract: Objectives: This study aimed to evaluate the usefulness of a short magnetic resonance (MR) protocol for knee evaluation, using 3D PDW SPAIR sequences compared with traditional 2D ones. Methods: A prospective analysis included 76 patients with knee pain. MR was performed using a 1.5 T scanner. The standard protocol consisted of multiplanar 2D proton density-weighted (PDW) SPectral Attenuated Inversion Recovery (SPAIR) and additional T1-weighted (T1W) and T2-weighted (T2W) sequences, with a total acquisition time of 17 min. The simulated short protocol included a 3D PDW SPAIR sequence with isotropic voxels and a slice thickness of 0.6 mm, coronal T1W, and gradient echo (GRE) axial sequences, with a total acquisition time of 9 min. Two radiologists independently reviewed images and collected pathological processes. Results: The 3D PDW SPAIR sequence demonstrated a significantly higher subjective image quality compared to standard 2D sequences [κ = 0.712 (p < 0.001) vs. κ = 0.144 (p = 0.63); p < 0.001]. Artifacts were not significantly different between the two protocols (p = 0.201). Qualitative assessments showed superior ratings for 3D images (excellent quality: 72.4% vs. 26.3% for 3D and 2D, respectively; p < 0.001). Diagnostic performance was comparable between the two protocols for ACL injuries, medial and lateral collateral ligament injuries, and chondropathies. Three-dimensional sequences were more effective in detecting medial meniscal injuries (p < 0.001), particularly radial and complex tears, likely due to higher spatial resolution and multiplanar reconstruction capability. Conclusions: The 3D PDW SPAIR sequence offers advantages in knee MRI study, including improved image quality, reduced acquisition time, and superior detection of meniscal injuries.
Language: English
Subfield: Surgery
Field: Medicine
Open Access status: gold
Estimated APC paid: $2,896
OpenAlex ID: W4413199393
FWCI: 1.785
Cites: 47
Cited by: 2
Related to: 10