CORR Insights®: What Is the Maximum Tibial Tunnel Angle for Transtibial PCL Reconstruction? A Comparison Based on Virtual Radiographs, CT Images, and 3D Knee Models

Where Are We Now?

Although a relatively recent article suggested that double-bundle posterior cruciate ligament (PCL) reconstructions can achieve similar clinical outcome scores as primary ACL reconstruction [4], most surgeons feel this procedure remains more technically demanding and yields less predictable biomechanical stability and patient-reported outcomes than its more-common ACL counterpart.

PCL reconstruction is challenging for many reasons: a high proportion of associated collateral ligament injuries, chronic posterior force because of gravity during postoperative recovery, larger graft diameters, coronal plane alignment, surgeons’ relative unfamiliarity because the injury is less common, and increased technical complexity because of anatomy. The killer turn phenomenon, whereby a graft is subject to friction against the anterior edge of the tibial tunnel, has been validated to attenuate the graft and enlarge the tibial inlet, effectively leading to anterior displacement of the graft, which is mechanically unfavorable [6]. Although the anatomic factors that create the killer turn are effectively fixed because of the necessary tunnel placement and graft trajectory in anatomic reconstruction techniques, potentially modifiable variables to mitigate this effect exist and include changing the method of fixation (suspensory versus aperture), retaining the native remnant (to provide padding against frictional forces), smoothing the transition zone at the exit of the tibial tunnel (to limit point loading at the turn), and changing the angle of the tunnel (to reduce the angle acuity at the aperture) [7].

Of these, changing the tunnel angle is the simplest to address intraoperatively, and thus identifying the best angle for PCL reconstruction to improve circumferential graft contact with bone while minimizing the killer turn angle is an important direction of study.

The current study [15] adds to the body of evidence helping surgeons make better preoperative and intraoperative decisions. It simplifies some parameters for tunnel placement. An anteromedial approach can use a guide angle of 60° and a starting point 60 mm below the joint line to create a 60-mm tunnel length. Similarly, an anterolateral approach can use a 50° guide angle and a starting point 50 mm below the joint line to create an approximately 50-mm tunnel length, albeit at a slightly less favorable angle. The results are important because nonanatomic tibial tunnel placement leads to increased posterior tibial translation in biomechanical analyses. Sadly, tibial PCL tunnel positions are inconsistently described, and this variability limits clinical application [10].

In practice, I use an anteromedial approach with a guide angle set to 60°. I corroborate my tunnel placement in three ways before drilling the tibial tunnel: with direct arthroscopic visualization with a 70° arthroscope from the anterior lateral portal to confirm medial to lateral placement between the tibial spines, with direct arthroscopic visualization from the posteromedial portal with a 30° arthroscope from the accessory posteromedial portal to confirm superior to inferior guide placement, and with lateral fluoroscopy with a guide pin in the planned tunnel to confirm that the trajectory follows the hourglass contour of the posterior tibial metaphysis without violating the back wall. Based on the authors’ findings, I can now also assess the tunnel length and feel confident that a length of approximately 60 mm provides additional confirmation of the optimal tunnel placement.

Although I expect these measurements might vary based on the patient’s size, the authors of the current study [15] found that no patient anthropomorphic characteristics (such as age, gender, height, and BMI) affected the tunnel length or angle. Undiagnosed collateral injury, varus alignment, and inappropriate tunnel position are the three most commonly cited reasons for recurrent instability after PCL reconstruction [11, 12]. I think an anteromedial approach is the most clinically useful because the starting position and suggested guide angle of 60° places the tunnel well inferior to associated ligament reconstructions commonly encountered in the multiply injured knee.

Where Do We Need To Go?

Although the authors’ findings are useful and easily memorizable as the “three sixties,” the authors assume that the starting point is always about 1.5 cm medial to the tubercle, and that the guide matches the radiographic measurements discussed. However, the guide very likely does not match because of the interposed soft tissue of the portals and fat pad, as well as other soft tissue anatomy (an intact ACL, for example). How do we identify which of the modifiable variables in PCL reconstruction matter? How do we improve these variables to better standardize techniques?

Currently, studies have stated that allograft reconstruction results appear to be equivalent to those of autograft reconstructions [3], and inlay techniques appear equivalent to transtibial techniques [14]. Single-bundle techniques have been considered to be equivalent to double-bundle techniques, but more-recent studies suggest improved stability with double-bundle reconstructions and have supported the concept of codominance between the anterolateral and posteromedial bundles [9, 13]. Because single-bundle and double-bundle techniques use the same single tibial tunnel, any difference in stability is not likely because of the killer turn effect at the entry into the tibial tunnel. Interestingly, the codominance theory has been applied to the posterolateral ligaments as well as the PCL bundles [8], with the conclusion that the two bundles protect each other by functioning in a load-sharing, codominant fashion, with no component dominating at any flexion angle.

The anterolateral bundle is larger and tensioned at 90°, the flexion angle at which most mechanical testing is performed. Because double-bundle techniques have never been reliably shown to be superior for ACL reconstruction, we must ask, with the addition of the posteromedial bundle, why is it favorable, given its smaller size and the fact that it is routinely tensioned in extension? Although anatomic studies have evaluated the codominance theory [1, 8], a detailed biomechanical analysis is incomplete and clinical studies are lacking. A recent meta-analysis was inconclusive, largely because of the inclusion of only four randomized controlled trials [5]. Furthermore, what other factors may be at play in the posterior cruciate-deficient knee and its subsequent reconstruction that may compromise outcomes?

How Do We Get There?

Multicenter trials are one solution but require substantial resources, time, and energy. Because of the relative rarity of PCL injuries treated with surgery, the completion of additional randomized controlled trials is challenging, particularly in isolation. In the meantime, we may have to resign ourselves to alternative directions of inquiry.

Regarding the morphology of the ideal tibial tunnel, biomechanical cadaveric models using pressure sensors placed anterior to the tibial graft at the aperture of the tibial tunnel in the posterior knee can quantify the forces applied to the graft at the killer turn, providing biomechanical validation to the theoretical advantages proposed in this model.

Another consideration, although it may be outside of the scope of the current study [15], is the effect of posterior tibial slope. Recently, there has been increased interest in the effect of increased tibial slope on the fate of ACL reconstruction, and osteotomies have been developed to intentionally alter slope in the sagittal plane. In turn, there is concern that patients prone to PCL injury may have a reduced tibial slope [16]. Changes in posterior tibial slope could affect the trajectory of tibial tunnels, and future studies might review a series of knees with PCL injuries. CT and 3D modeling studies could be applied to knees that are at greater risk of PCL injury because of sagittal malalignment. If varus coronal plane alignment can be associated with PCL failure, sagittal alignment might play an equal or larger role. Currently, the relationships between coronal and sagittal plane malalignment are not well described but could be synergistic.

As noted, an evolving body of biomechanical data suggests there are benefits to double-bundle reconstruction. This may be simply because our biomechanical testing apparatus is becoming more sophisticated and better attuned to subtle differences than in the past. The most-sensitive apparatus remains patient and clinical studies reporting 2-year clinical follow-up with biometric testing for stability, and patient-reported outcome measures will be the ultimate test of the double-bundle technique. We would be wise to remember that in double-bundle ACL reconstructions, the clinical benefits never delivered on the promise of the biomechanical results.

We are learning that PCL injuries, similar to ACL injuries, are rarely isolated and probably involve injury to the posteromedial and posterolateral structures, to some extent. A further understanding of the codominance theory would benefit from not only an assessment of the individual PCL bundles, but also the relationship between the injured PCL and the adjacent posterolateral and posteromedial structures. Sequential sectioning studies whereby the PCL knee is subjected to additional capsular insult will help us better understand what additional stabilization is needed to prevent the mechanical failure of PCL reconstructions [2].

In general, a better understanding of the contributions of osseous and ligamentous contributions to knee stability will be more difficult than simply reducing the killer turn, and these avenues of research can lead us well into the coming decades as we seek to improve outcomes for patients with posterior knee instability.