Lerman J E, Gray D S, Schweitzer M E, Bartolozzi A
Department of Radiology, Nassau County Medical Center, East Meadow, NY, USA.
J Comput Assist Tomogr. 1995 Jul-Aug;19(4):604-7. doi: 10.1097/00004728-199507000-00019.
To investigate the accuracy of diagnosing anterior cruciate ligament (ACL) tears with axial MR imaging.
Two blinded independent observers retrospectively reviewed axial T2 weighted or FSE fat suppressed imaging of the knee from 47 patients. Arthroscopy had demonstrated a complete tear of the ACL in 25 patients and a normal ACL in 22 patients. The two criteria used to diagnose ACL tears on sequential axial imaging were (a) lack of visualization of the ACL in its expected course or (b) focal increased signal in the expected course of the ACL.
Observer 1 demonstrated a sensitivity of 92% (23 of 25) and a specificity of 100% (22 of 22) in diagnosing ACL tears (chi 2 = 39.6, p < 0.001). Observer 2 demonstrated a sensitivity of 92% (23 of 25) and a specificity of 82% (18 of 22) in diagnosing ACL tears (chi 2 = 26.1, p < 0.001). The Kappa value was 0.75 (agreement in 41 of 47 cases), indicating a high degree of intraobserver agreement.
Axial T2 weighted or FSE fat suppressed imaging provides an accurate means of confirming the diagnosis of ACL tears.
