Approval from the hospital ethics committee was obtained before this study started (IRB Number: 2021-038). 71 patients with PCL injury, identified in the Medical Record Registration System of our hospital from January 2015 to December 2020, were eligible for inclusion of this study. The inclusion criteria were subjects with an MRI scan in the picture archiving and communication system (PACS); the PCL rupture was identified by arthroscopy. Subjects with previous osteotomy or fracture, meniscal repair, or partial meniscectomy, which would influence the measurement of PTS and/or meniscal slope (MS); subjects with PCL avulsion fractures or ACL injuries; subjects with discoid meniscus were excluded from this study. Among the 71 patients, 16 patients with ACL injury, five patients with PCL avulsion fractures, three patients with severe menisci injury (Bucket Handle Tear) that can affect the accuracy of the MS, one patient with previous ACL reconstruction, one patient with lateral discoid meniscus were excluded from this study.
As a result, 45 patients with PCL rupture were included in the study. Among them, some concomitant injuries were presented and included partial meniscus rupture, cartilage injury, bone marrow edema, but those could not influence the accuracy of the measurements. The PCL-intact control group, matched by age and sex was built to include 45 individuals who came to the radiology department of our hospital for knee-MRI-scanning only with anterior knee pain and without a history of knee injury. And the MRI of each individual was retrospectively reviewed by two experienced orthopedists to ensure the participants’ eligibility for the study. The informed consent requirement was waived by the Institutional Review Board of the First Affiliated Hospital of Chongqing Medical University.
The MRI examinations were performed within one week prior to surgical planning in patients with PCL injuries. All examinations were performed with the same 1.5T MRI scanner (Siemens Magnetom Essenza, Germany). The coronal and sagittal planes were scanned with the T1-weighted turbo spin-echo (TSE) sequence and proton density (PD) TSE with the fat-suppressed (FS) sequence, and the axial plane was scanned with the PD-TSE-FS. The layer thickness was set to 3 mm, the slice gap was 0.5 mm, the field of view (FOV) was 160 mm, and the matrix size was 512×512.
Anatomic parameters were obtained on MRI by two observers in a blinded and randomized fashion through PACS workstation to determine inter-observer reliability. Both observers were trained in consensus in the measuring methods. The values measured by the 2 observers were averaged for the statistical analysis.
The literature suggested that separate assessment of the PTS was not reliably possible on lateral radiographs, the reliable and reproducible methods conducted on MRI or computed tomography (CT) images were recommended to measure the PTS 6. It varies in the literature regarding the definition of the lateral tibial axis on MRI 15-17, we conducted the method described by Hudek et al. 17, which was reported to be the most repeatable and reliable method to measure sagittal tibial slopes on MRI 15：a T1-weighted mid-sagittal cut was selected with the appearance of PCL-tibial insertion point, the tibial axis was defined as the line through both center of circles drawing on the sagittal image, and tibial axis was used to measure the medial and lateral PTS (Fig. 1) and MS (Fig. 2) on the other two sagittal images (the mid-medial tibial plateau cut and the mid-lateral tibial plateau cut), respectively.
MTD is the depth of concavity of the medial plateau in the middle of the articular region. The perpendicular distance between the line connecting the uppermost superior-anterior and posterior cortex edges of the medial tibial plateau, and the lowest point of the medial plateau concavity 18. (Fig. 3)
The average values of the variables measured by both observers were used for all analyses. The Independent t-test was conducted to compare the parameters between the two subgroups, and binary logistic regression analysis was used to determine the probability of PCL injury in an individual based on the measured covariates. All analysis was conducted by a coauthor independently via the SPSS software (Version 22.0; IBM Corp), and the P value of < 0.05 was considered statistic significant. The interclass correlation coefficient (ICC) was calculated to assess the reliability of each parameter, with a value of more than 0.75 indicating excellent agreement.
G*Power 3.1 (Heinrich-Heine-Universitat Dusseldorf, Dusseldorf, Germany) was used to perform power analysis. Regarding the difference in the lateral PTS between the two groups, for a large effect size (0.56), the results indicated that at least 40 patients were needed in this study (alpha, 0.05; power, 0.8).