PURPOSE: To attempt to visualize the ligament with standard 1.5-tesla magnetic resonance imaging (MRI) in the acute anterior cruciate ligament (ACL)-torn knee, and if it is visible, attempt to characterize it as torn or intact at its femoral, meniscal, and tibial attachment sites. METHODS: This was a retrospective MRI study based on arthroscopic findings of a known ACL tear in 72 patients between the years 2006 and 2010. Patients all had hamstring ACL reconstructions, no concomitant lateral collateral ligament, or posterolateral corner injury based on imaging and physical examination, and had a preoperative 1.5-tesla MRI scan with standard sequences performed within 3 weeks of the injury. Two fellowship-trained musculoskeletal radiologists retrospectively reviewed the preoperative MRI for visualization of the anterolateral ligament (ALL) for concomitant tears. Inter- and intraobserver reliability was calculated. Learning effect was analyzed to determine if radiologists' agreement improved as reads progressed. RESULTS: Both radiologists were able to visualize the ALL in 100% of the scans. Overall, ALL tears were noted in 26% by radiologist 1 and in 62% by radiologist 2. The agreement between the ligament being torn or not had a kappa of 0.54 between radiologists. The agreements in torn or not torn between radiologists in the femoral, meniscal, and tibial sites were 0.14, 0.15, and 0.31. The intraobserver reliability by radiologist 1 for femoral, meniscal, and tibial tears was 0.04, 0.57, and 0.54 respectively. For radiologist 2, they were 0.75, 0.61, and 0.55. There was no learning effect noted. CONCLUSIONS: ALL tears are currently unable to be reliably identified as torn or intact on standard 1.5-tesla MRI sequences. Proper imaging sequences are of crucial importance to reliably follow these tears to determine their clinical significance. LEVEL OF EVIDENCE: Level IV, therapeutic case series study.
PURPOSE: To attempt to visualize the ligament with standard 1.5-tesla magnetic resonance imaging (MRI) in the acute anterior cruciate ligament (ACL)-torn knee, and if it is visible, attempt to characterize it as torn or intact at its femoral, meniscal, and tibial attachment sites. METHODS: This was a retrospective MRI study based on arthroscopic findings of a known ACL tear in 72 patients between the years 2006 and 2010. Patients all had hamstring ACL reconstructions, no concomitant lateral collateral ligament, or posterolateral corner injury based on imaging and physical examination, and had a preoperative 1.5-tesla MRI scan with standard sequences performed within 3 weeks of the injury. Two fellowship-trained musculoskeletal radiologists retrospectively reviewed the preoperative MRI for visualization of the anterolateral ligament (ALL) for concomitant tears. Inter- and intraobserver reliability was calculated. Learning effect was analyzed to determine if radiologists' agreement improved as reads progressed. RESULTS: Both radiologists were able to visualize the ALL in 100% of the scans. Overall, ALL tears were noted in 26% by radiologist 1 and in 62% by radiologist 2. The agreement between the ligament being torn or not had a kappa of 0.54 between radiologists. The agreements in torn or not torn between radiologists in the femoral, meniscal, and tibial sites were 0.14, 0.15, and 0.31. The intraobserver reliability by radiologist 1 for femoral, meniscal, and tibial tears was 0.04, 0.57, and 0.54 respectively. For radiologist 2, they were 0.75, 0.61, and 0.55. There was no learning effect noted. CONCLUSIONS: ALL tears are currently unable to be reliably identified as torn or intact on standard 1.5-tesla MRI sequences. Proper imaging sequences are of crucial importance to reliably follow these tears to determine their clinical significance. LEVEL OF EVIDENCE: Level IV, therapeutic case series study.
Authors: Volkan Kızılgöz; Ali Kemal Sivrioğlu; Hasan Aydın; Türkhun Çetin; Gökhan Ragıp Ulusoy Journal: J Int Med Res Date: 2018-01-19 Impact factor: 1.671