Evaluation of cartilage surface injuries using 3D-double echo steady state (3D-DESS): Effect of changing flip angle from 40° to 90°

Abstract

In magnetic resonance imaging (MRI) with 3D-double-echo steady-state (3D-DESS) sequences, the cartilage-synovial fluid contrast is reported to be better with a flip angle of 90° than with the conventional flip angle of 40°, and the detection rate of knee cartilage injury may be improved. To compare the diagnostic performance and certainty of diagnosis with 3D-DESS images made using two flip angle settings, 40° and 90°, for knee cartilage surface lesions of Grade 2 or above confirmed by arthroscopy. Images were obtained with 3D-DESS flip angles of 40° and 90° at 1.0T in 13 consecutive patients (2 men, 11 women, age range 18-68 years) evaluated for superficial cartilage injury by arthroscopy. Two radiologists classified the presence or absence of cartilage damage of ≥Grade 2 as 'positive (p)' or 'negative (n)', respectively. The rate of agreement with arthroscopic diagnosis was then examined, and the diagnostic performance compared. Diagnostic confidence was assessed scoring the presence or absence of cartilage damage into three categories: 3 = can diagnose with absolute confidence; 2 = can diagnose with a level of certainty as probably present or probably absent; and 1 = cannot make a diagnosis. In a comparison of the rate of agreement between diagnosis by 3D-DESS images and arthroscopy, the rate of agreement was significantly higher and diagnostic performance was better in 90° images for the medial femoral condyle only. Diagnostic confidence was significantly better with 90° flip angle images than with 40° flip angle images for all six cartilage surfaces. In evaluating knee cartilage surface lesions with 3D-DESS sequences, a flip angle setting of 90° is more effective than the conventional setting of 40°.