The main achievement of this study was that we proposed a new femoral tunnel classification system based on three-dimensional CT imaging to guide ACL revision surgery on the basis of comprehensive consideration of the femoral tunnel shape, position, diameter, and posterior wall of the OT, and it was proved that this classification system had high reliability and could be used in clinical practice.
The classification aimed to describe the position of the femoral tunnel with surrounding bony landmarks for the selection of different operation method for ACL revision. Zauleck et al. found LIR was a reliable bony landmark helping the surgeon analyze and judge the anterior and posterior location of the tunnel [23]. However, it was difficult to identify the specific location of LIR on CT two-dimensional imaging due to the influence of the OT [17, 18], and the LIR was almost entirely dependent on relative anatomical knowledge and clinical experience in the Robert femoral tunnel classification, which substantially decreased the objectivity of the results [12]. Therefore, in this study, we successfully located the precise position of LIR in the femoral model through the three-dimensional CT reconstruction technology and the anatomical characteristics of LIR, and proposed our femoral tunnel classification system according to the position between the LIR and OT (intersection, tangency, and separation). Moreover, the consistency of the four types in the high senior group was significantly higher than that in the low senior group, while the consistency of intra-observer reliability all exceeded 0.73, so the new classification system also had high reliability and stability.
Due to the shortage of clinical experience, the Type I off-target type tunnel was often misplaced in an extreme forward position by lower seniority doctors, but we found this type had a minimal impact on ACL revision, because we could directly placed the New Tunnel (NT) in the ACL femoral footprint to complete the first-phase revision (Fig. 4a). In addition, this study did not further divide Type II straddled type into forward, vertical, and forward vertical subtypes like the Robert type II (Fig. 4b) [12], because the research results of Robert et al. have shown that this classification was easily confused in the narrow ACL femoral footprint, and the three subtypes have little influence on the choice of revision strategies. Regardless of the subtype of OT, directly placing the NT in the narrow ACL femoral footprint would intersect with OT to form a huge tunnel, and expanding the OT will make NT more anterior, which would make OT unable to be used in first-phase revision. Therefore, the Type II straddled type corresponding to Robert type II in this study was not further divided into subtypes, due to the tunnel expansion was an important factor contributing to the failure of ACL revision and the enlarged tunnel frequently led to tunnel burst [19, 20].
In order to make the choice of revision strategy more instructive, the Type III anatomical type was further divided into three subtypes combined with OT diameter and posterior wall: IIIa in-suit type, IIIb enlarged type and IIIc posterior type. Firstly, the OT could be directly expanded by 0.5 ~ 1.0 mm to complete the first-phase revision as NT in the IIIa in-suit type (Fig. 4c). Secondly, we should further consider the diameter and the volume of intercondylar fossa in whether to expand OT, to circumvent the possibility of graft failure or intercondylar fossa impact in IIIb enlarged type (Fig. 4d) [21]. Finally, in the IIIc posterior type, we should also focus on the femoral graft fixation method to avoid the fixation failure caused by the tunnel burst after extrusion fixation, which requires detailed preoperative planning (Fig. 4e). It's noteworthy that the division of these subtypes was based on the diameter and posterior wall of the OT rather than the position. Although the position of OT in the anatomical region would have differences in the in-situ tension and mechanical properties of the graft, the difference would not directly determine the success or failure of the revision process just as the diameter and posterior wall of femoral tunnel in complex revision cases, and whether the tunnel should be located in high or low position in ACL anatomical regions were difficult to reach a consensus [22]. Nevertheless, the first-phase revision should be considered for enlarging the tunnel regardless of high or low position in the lateral wall of the intercondylar fossa in the Type IIIa in-suit type.
According to the tunnel shapes in clinical practice, we further proposed the Type IV irregular type based on improving the accuracy of the Robert classification system, which could more systematically cover clinical surgery. For example, the femoral double tunnel was difficult to make a selection of revision strategy, and even if the position of the double tunnel was correct, the expanded double tunnel could still probably run through a huge tunnel (Fig. 4f) [23]. Therefore, only the second-phase revision or OTT reconstruction could be chosen to deal the difficult problem.
There are some limitations in this study. Firstly, only 150 cases CT imaging were included in the retrospective study, but the patients included were from multi-center hospitals, all of whom need ACL revision surgery and the location of femoral tunnels were complex. Secondly, only the femoral tunnel was chosen to make a classification in this study, but the clinical revision strategy still needed to combine the tibial tunnel in clinical practice, even though it was less important than the femoral tunnel. Finally, the LIR was determined by the positional relationship of the anatomical femoral axis and the Blumensaat line, so it may differ from the actual situation due to individual differences. However, making the classification system by the exact anatomical parameters was more objective than the Robert type.