The normal ACL anatomy can be readily delineated by orthogonal sagittal and coronal MRI images [8]. However, due to its oblique course, 5–10% of the normally imaged anterior cruciate ligaments are difficult to visualize by orthogonal sagittal MRI [9]. Additionally, artifacts from the popliteal artery, partial volume, osteoarthritis, and post traumatic fibrous scarring increase the false-positive diagnosis of an ACL tear [10].
Nowadays, identifying ACL partial tear is of utmost importance in the management of ACL tears as it improves the outcome of ACL reconstruction. However, the diagnosis of partial ACL tears is diagnostically challenging owing to the striated appearance of a normal ACL which is aggravated with the use of fluid-sensitive sequences mimicking the appearance of partial ACL [11, 12].
To overcome these anatomical and technical limitations, various techniques were utilized to allow delineation of the entire course of the ACL in a single slice through parallel imaging along the ACL longitudinal axis [13].
Several methods to obtain oblique sagittal images were described in literature; Buckwalter and Pennes acquired an imaging plane angled 15° to the standard sagittal plane [9], and Do-Dai et al sliced parallel to the medial border of the lateral femoral condyle to develop oblique sagittal images for the ACL [14]. Nakanishi et al. sliced on axial image at an angle of 10° [15]. In this study, we acquired the oblique sagittal image plane parallel to the inter-condylar surface of the lateral femoral condyle.
Oblique coronal images were also utilized in previous studies to evaluate for ACL injuries. Staeubli et al. stated that the best sequence to visualize the ACL diagonal course is achieved by oblique coronal images as it clearly delineates the relation of the ACL attachment to the inter-condylar notch as well as its orientation towards the posterior cruciate ligament [16]. Hong et al. also concluded that oblique coronal MRI sequences increased the diagnostic efficacy in differentiating between partial and complete ACL tears [17]. In this study, we acquired the oblique coronal image plane parallel to femoral inter-condylar roof along the course of the ACL.
In the present study, we investigated the accuracy, sensitivity, and specificity of utilizing additional oblique sagittal and oblique coronal images with standard sagittal images both separately and in combination with the evaluation of complete and partial ACL tears.
This study showed that the addition of oblique sagittal and oblique coronal images to the standard sagittal images whether separately or in combination (approaches B, C, and D) improved the specificity and accuracy of the diagnosis of complete ACL tears in comparison to the standard sagittal images only (approach A). However, there was no statistical significance in diagnosis of complete ACL tear obtained from a combined use of oblique sagittal and coronal images (approach D) in comparison to the use of only one of the oblique images (approaches B and C. This was in agreement with the study performed by Kwon et al., which demonstrated that additional oblique imaging improved the specificity for diagnosing complete ACL tear. It also showed that either method of oblique imaging alone is sufficient with no statistical significance detected in combining both oblique methods [18].
As for diagnosis of partial ACL tears, this study revealed that the accuracy, sensitivity, and specificity for approach A in diagnosis of partial ACL tear was significantly lower than that for approaches B, C, and D (p value < 0.05). This was also in agreement with the findings described by Kwon et al. [18].
However, on the contrary to Kwon et al. [18] which stated that there was no statistical significance between the use of a single oblique method and their concurrent use in the diagnosis of ACL partial tears, this study revealed that the accuracy, sensitivity, and specificity of concurrent utilization of oblique sagittal and coronal images (approach D) for detection of partial ACL tear were significantly higher than that for approaches B and C (p value < 0.05). This is likely attributed to the advantage of the concurrent use of both obliques in detection of the entire course and femoral attachment of the ACL on two different planes enhancing the accuracy of diagnosis of changes in ACL thickness, signal intensity, continuity, and contour.
In 2018, Ghasem et al. [19] showed that the addition of oblique sagittal to standard images was better than the addition of oblique coronal images in the detection of complete and partial ACL tears with PPV and NPV of 78% and 88% in partial and 83% and 100% in complete ACL tears. Our results showed that there was no statistical significance in using either sagittal oblique or coronal oblique separately in diagnosis of partial and complete ACL tears; however, our study revealed that there is a statistical significant difference in diagnosing partial ACL tears when both oblique methods were performed together instead of performing it separately. The PPV and NPV for (approach D) in partial ACL tears were 94% and 95% in comparison to 82% and 85% in (approach B) and 83% and 87% in (approach C).
Our study showed that there is a higher interobserver agreement in diagnosis of complete and partial ACL tears using approaches B, C, and D in comparison to approach A, being highest with approaches C and D. This is in agreement with Kosaka et al., which showed higher diagnostic performance and interobserver agreement when using additional oblique coronal images [20]. The main reason for this is that oblique coronal images clearly delineate the course and femoral attachment of the ACL.
The present study had some limitations: First, the inconsistency as regards the duration between the onset of symptoms or time of injury and the MRI examination. However, the time interval between MRI examination and arthroscopy was limited to 1 month. Secondly, the difference in slice thickness between the standard sagittal images (4-mm slice thickness) and oblique sagittal and coronal images (2-mm slice thickness) was not taken into consideration in our study when comparing the different imaging techniques.