This study was a retrospective case control study. The MRI scans were obtained from patients who had received a reconstruction operation for ACL injuries, between January 2004 and January 2011, while, the control group had an MRI for other knee problems, rather than ACL injury. They were either found to have isolated meniscal injuries, or no pathology. The exclusion criteria were performed following the results from an MRI scan with bad quality (other hospitals), greater than grade II Outerbridge osteoarthritic changes, patient's age older than 55-year-old, multiple ligamentous injury and foreigner. Following the rules above, total of 226 patients' MRI images were obtained. Among them, the ACL injured group was composed of 120 patients; 76 male with the average age of 37.9 years (range, 19 to 50 years) and 44 female with the average age of 42.5 years old (range, 20 to 55 years old). The causes of ACL injuries were traffic accidents for 38 patients, playing soccer for 23, playing basketball for 15, and some other reasons for 44. Non-ACL injured group consisted of 106 patients, and among them, there were 71 male patients with an average age of 40.5 years (range, 17 to 53 years) and 35 female patients with an average age of 47.2 years (range, 21 to 54 years).
For each included patient, the 1.5 Tesla Knee MRI (Magnetom Avanto-Siemens, Image J program ver. 1.44, JAVA, California, CA, USA) was collected and the slice spacing was 3 mm. The NW, bicondylar width, medial condylar width, and lateral condylar width of each patient were measured at an intermediate imaging, among the obtained ones, where popliteal grooves are seen under the T1 coronal view of MRI. The medial-to-lateral condyle size ratio (M:L ratio) was calculated by dividing the medial condyle width by the lateral condyle width, and the NWI was calculated by dividing the NW by bicondylar width (Fig. 1).
The axial T2 sequence was used to measure the width of the notch entrance (NE). This width was measured at each slice containing the notch, from the first slice with cartilage on both condyles to the last slice with bony connection between the medial and the lateral condyle (Fig. 2).
The widths of the subjects in each group were measured twice with three observers. The descriptive statistics were used for the calculation of notch entrance width, medial condyle width, lateral condyle width, M:L ratio, bicondylar width, and the NWI. The first study was performed with the differences of bony morphology between men and women, and the differences between the ACL injured and the non ACL-injured were analyzed. All comparisons were performed with a student t-test and the alpha level was set at 0.01. All MRI measurements were performed with three observers. Each observer measured the values twice. The intra and inter observer reliability were calculated using the interclass correlation coefficient (ICC), and the statistical analysis was made with SPSS ver. 18 (SPSS Inc., Chicago, IL, USA).