➢ Regardless of the drilling technique utilized, nonanatomic grafts (for example, vertical femoral tunnels) do not restore normal kinematics and may be associated with a higher failure rate.
➢ Anteromedial drilling allows placement of both the femoral and tibial tunnels within their native insertion sites. On the other hand, transtibial drilling compromises the ability of the surgeon to place either the tibial or femoral tunnel anatomically.
➢ Although nonanatomic anterior cruciate ligament (ACL) reconstructions result in measureable kinematic changes (e.g., rotational or translational), clinical outcome studies have not shown a difference between the two techniques.
Anterior cruciate ligament (ACL) surgery continues to evolve as we gain a better understanding of the anatomy and biomechanics of the ACL. In the late 1980s, there was a shift from a two-incision (rear-entry guide) to a single-incision (transtibial) technique. The impetus driving this was the potential to achieve equivalent tunnel placement more quickly and in a more cosmetic fashion1. However, when anatomic studies clearly defined the ACL femoral insertion site on the lateral intercondylar wall and not the roof, it was discovered that even when performed as described in the Surgical Techniques supplement of The Journal of Bone & Joint Surgery in 20052, the traditional transtibial technique placed the femoral tunnel higher (toward the apex of the notch as opposed to the wall) on the intercondylar notch, outside the native insertion site (Fig. 1)3-6. Some surgeons postulated that, even with adjustments to the traditional transtibial technique, transtibial drilling would always place the femoral tunnel higher on the intercondylar notch than the location of the native insertion site (Fig. 2, A). In order to avoid placing the femoral tunnel on the intercondylar roof as is commonly associated with the transtibial technique (also known as the vertical tunnel), the anteromedial portal technique was developed to allow more accurate femoral (and tibial) tunnel placement (Fig. 2, B)7.
The most recent development in the evolution of ACL surgery is the concept of anatomic reconstruction, which strives to place both the femoral and tibial tunnels within their native insertion sites. Studies have shown that it is much more difficult to place the femoral tunnel accurately with use of a transtibial technique6,8-13. Also, attempting to place the femoral tunnel in its insertion site with the transtibial technique has been shown to compromise the tibial tunnel placement14-17. Drilling the femoral tunnel via an anteromedial portal allows independent and more accurate placement of the tibial and femoral tunnels. Nonanatomic or vertical ACL grafts are not positioned to control rotational stability and are associated with higher failure rates18-22. Regardless of the drilling technique utilized, ACL reconstructions that utilize the native femoral and tibial footprints achieve better biomechanical and clinical outcomes than those with grafts outside the native footprints, including vertical graft placement18-20.
Table I summarizes the advantages and disadvantages of the transtibial and anteromedial drilling techniques. The purpose of this article is to cover in greater detail the history of this evolution and the clinical and biomechanical studies comparing these two techniques.
Why Question ACL Outcomes and Techniques?
Much of the recent emphasis on cadaveric and in vivo studies to determine the ability of the anteromedial and transtibial techniques to reproduce an anatomic ACL reconstruction has stemmed from a common observation that nonanatomic grafts have a higher failure rate. A series of 122 failed ACLs was evaluated19. A nonanatomic graft was found in 88%, with 61% entirely on the intercondylar femoral roof. The investigators also noted that 44% of the failed ACLs underwent revision without correction of the tunnel location, and all of these grafts also failed. Eighty-three percent of the failed ACLs had undergone a transtibial technique, which was the more common technique during the collection period.
The Multicenter ACL Revision Study (MARS) group cited a technical cause of failure of the primary reconstruction in 276 (60%) of 460 revision ACL reconstructions23. Femoral tunnel malposition was reported in 219 (47.6%) of the primary reconstruction failures, and it was the only cause of failure in 117 (25.4%) of the failures. The failed femoral tunnels were either too vertical, too anterior, or both (35.9%, 29.9%, and 26.5%, respectively). In addition to ACL revisions, the Multicenter Orthopaedic Outcomes Network (MOON) found that transtibial as opposed to anteromedial portal drilled ACL reconstructions had a higher rate of all repeat ipsilateral knee surgeries, which that group proposed may have been due to tunnel positioning24.
Nonanatomic grafts have been reported as a cause of ACL technical failure in multiple smaller series as well18,25,26. Garofalo et al. reported femoral tunnel malposition in 79% and tibial tunnel malposition in 21% of thirty-one ACL failures25. Taggart et al. reported femoral tunnel malposition in 60% of twenty ACL failures26. In a separate series of twenty-six ACL failures, the failed ACL grafts were more vertical (69.6° as opposed to the native anteromedial bundle angle of 56.2° and the native posterolateral bundle angle of 55.5°)18. They noted that, compared with the anatomic location, the failed ACL grafts were too posterior and medial on the tibial side, and too anterior on the femoral side.
Additional recent outcome studies on ACL reconstruction have also revealed the need for an improvement in return to athletic play. The Multicenter Orthopaedic Outcomes Network (MOON) study group reported that only seventy-two (72%) of 100 soccer athletes and sixty-one (65%) of ninety-four football players returned to play, with only thirty-three (43%) of seventy-seven football players reporting return to play at their pre-injury level27,28. The decreased rate of return to play may reflect a persistent knee instability, which may be secondary to nonanatomic graft placement.
Effect of Drilling Techniques on Tunnel Characteristics in Cadavers
Awareness of the importance of tunnel placement has led to many cadaveric studies evaluating the ability of the transtibial and anteromedial portals to consistently place the femoral tunnel in an anatomic position (Table II). Furthermore, recent studies have also evaluated the benefits and drawbacks associated with the different drilling techniques in their attempts to place the femoral tunnel anatomically; specifically, these studies evaluated the effects of tunnel length, aperture size, graft obliquity, and risk of damage to the medial tibial spine and posterolateral structures.
Tunnel Location and Orientation
The ability of the transtibial technique to place the femoral tunnel in an anatomic position was first studied by Arnold et al., who claimed that reaching the femoral attachment site “through a correctly placed tibial tunnel . . . proved to be impossible.”6 The success associated with use of the transtibial technique to anatomically place the femoral tunnels in double-bundle ACL reconstruction was studied by Giron et al. These investigators noted that, via transtibial drilling, only twelve of fourteen femoral guidewires could reach the anteromedial bundle attachment and only nine of fourteen wires could reach the posterolateral bundle8.
In what was, to our knowledge, the first cadaveric study to directly compare the two techniques, Gavriilidis et al. demonstrated that the center of the transtibial femoral tunnel was 6.2 mm from the center of the native ACL footprint, whereas the center of the anteromedial tunnel was a distance of 2.8 mm from the center of the native ACL footprint9. Similar results suggesting more anatomic placement with anteromedial drilling were found in three recent trials (Table I), including one by Bedi et al., in which the anteromedial drilled femoral tunnel was in the center of the femoral footprint 100% of the time whereas the transtibial tunnel was, on the average, 1.9 mm anterior and 3.3 mm superior10,11,20.
With more advanced imaging techniques, such as three-dimensional digitized images and fused computed tomography (CT) scans, cadaveric studies can determine the percentage match between the native femoral footprint and the tunnel footprint. Tompkins et al. demonstrated that 97.7% of anteromedial drilled tunnels were in the native femoral footprint as compared with only 61.2% of transtibial drilled tunnels10. Gadikota et al. reported similar results in eight cadaveric specimens, with 27.1% versus 13.6% of the transtibial and anteromedial femoral tunnels outside the femoral footprint11.
In addition to the ability to place the tunnels anatomically, Bedi et al. evaluated the overall obliquity of the reconstructed cadaveric ACLs created by the two drilling techniques. They found that appropriate coronal obliquity was better achieved with use of anteromedial drilled femoral tunnels than it was with transtibial drilled tunnels29.
The ability of the transtibial technique to allow for an anatomic femoral tunnel with different graft types has also been investigated. A cadaveric study found that, with meticulous tibial tunnel placement with use of computerized navigation, a bone-patellar tendon-bone ACL reconstruction can be performed within 0.3 mm of the femoral insertion with <10 mm of graft-tunnel mismatch30. In a similar cadaveric study in which the technique described above was used, the authors evaluated the smaller tibial tunnels necessitated by hamstring autografts and found that the femoral tunnel was 7.6 mm from the native ACL femoral insertion and only occupied 30% of the footprint31.
Because of increasing awareness of the association between the transtibial technique and vertical grafts, the proponents of that technique have advocated for the use of a more medial starting point for the tibial tunnel in an attempt to place the femoral tunnel in an anatomic location. Miller et al. evaluated the effects of utilizing the more medial tibial tunnel to reach the anatomic femoral insertion14. Compared with the standard central tibial starting point, a more medial starting point brought the femoral tunnel closer to horizontal (10:40 versus 10:14 o’clock). After femoral drilling, however, the more medially drilled tibial aperture was larger (118.6 versus 106.3 mm2) and had a larger tibial tunnel volume (3016.0 versus 2632 mm3), potentially compromising fixation. The enlarged transtibial tibial tunnels were even larger than the tibial tunnels observed when the anteromedial portal technique was utilized for femoral drilling. Bedi et al. had similar findings and reported that tibial apertures were 38% larger in transtibial drilled femoral tunnels as compared with anteromedial drilled tunnels20.
Tunnel Length and the Risk of Posterior Wall Blowout and Damage to the Medial Femoral Condyle
In a comparison of cadaveric anteromedial and transtibial ACL reconstructions, Bedi et al. raised the concern that the anteromedial drilled femoral tunnels were more likely to violate the posterior wall (19.4% versus 0%) and were more likely to have an intraosseous length of <25 mm (41.7% versus 16.7%) than the transtibial drilled femoral tunnels were29. They also noted that increasing knee flexion allowed for more coronal obliquity but created a higher risk of critically short tunnels and posterior wall blowout when a standard 6-mm femoral offset guide was used. These findings were in contrast to those from a series of additional studies that demonstrated longer femoral tunnels with less risk of posterior wall blowout with increasing knee flexion32,33.
In a study comparing tunnel length, the femoral tunnels were <30 mm long in 26% of the knees in the anteromedial group but in only 2% of the knees in the transtibial group32. Three cadaveric studies found that anteromedial femoral tunnel drilling allowed for a minimum mean tunnel distance of 30 mm (30.5, 34, and 34 to 39 mm, respectively), depending on the flexion angle and guidewire type32-34. Dave et al. demonstrated that, with increasing knee flexion from 90° to 120°, the mean length of the femoral tunnel (measured with a straight guidewire) increased from 34.4 to 39.9 mm33. They also demonstrated that four of the eight guidewires were within 3 mm of the posterior wall when the tunnel had been drilled with the knee in 90° of flexion but that none of the eight guidewires were within 3 mm of the posterior wall when the tunnel had been drilled with the knee in 120° of flexion. These studies suggest that hyperflexion allows for a safe tunnel length without risk of posterior wall blowout, a finding that was verified in a study of 106 consecutive patients who had no posterior wall blowout fractures and in whom all anteromedial femoral tunnels were drilled to a length of >30 mm35.
Although short femoral tunnels are not reported to be a problem with the transtibial technique, Heming et al. found that the process of creating an anatomic femoral tunnel with use of that technique might lead to the creation of a short tibial tunnel36.
Drilling the femoral tunnel from the anteromedial portal also introduces the risk of damage to the medial femoral condyle, whereas this risk does not exist with use of the transtibial technique. Careful arthroscopic needle localization of the anteromedial portal, combined with knee hyperflexion while drilling, can allow access to the femoral insertion site without damaging the medial femoral condyle. The use of a flexible guidewire and reamer or a half-round reamer are other strategies that can be used to safely bypass the medial femoral condyle while accessing the femoral insertion site37,38.
Risk of Neurovascular Injury
When comparing the two reconstruction techniques, Gadikota et al. noted that the anteromedial femoral tunnel was more likely to exit posterior relative to the lateral epicondyle (three specimens versus zero), potentially risking damage to posterolateral structures11. Proponents of the anteromedial technique claim that hyperflexion decreases this risk. A separate study performed with use of ten cadaveric knees found that, with the knee flexed to 90° or more, the biceps tendon was never pierced and the mean distance to the peroneal nerve was 28.6 mm (range, 25 to 32 mm), with a mean distance of 44 mm at 120° of flexion39.
Effect of Drilling Techniques on Tunnel Characteristics in Vivo
Cadaveric studies do not always translate into in vivo results, as was discovered with GORE-TEX ACL grafts (W.L. Gore and Associates, Flagstaff, Arizona)40,41. Fortunately, there are now multiple in vivo studies available to evaluate the effect of anteromedial and transtibial drilling on both femoral and tibial tunnel characteristics (Table III).
In an in vivo study in which postoperative radiographs were used to evaluate the two drilling techniques in seventy patients, it was determined that 86% of the anteromedial drilled femoral tunnels and 57% of transtibial drilled femoral tunnels were within 5% of anatomic reference values12. Chang et al. utilized radiographs to determine that the anteromedial ACL reconstructed grafts had less sagittal obliquity (i.e., were less vertical) than transtibial reconstructions (a mean of 55.9° versus 61.7°, respectively), but intraoperative measurements revealed that the femoral tunnels were shorter in the anteromedial group than they were in the transtibial group (a mean of 34.2 versus 43.3 mm, respectively)32.
Abebe et al. utilized newly developed three-dimensional magnetic resonance imaging (MRI) analyses of ACL reconstructions to directly compare the femoral attachment sites of eight patients who had a transtibial reconstructed ACL and eight patients who had an anteromedial reconstructed ACL13. They found anteromedial drilled tunnels were, on the average, 3 mm from the center of the ACL attachment site as compared with 9 mm in the transtibial drilled tunnels.
With the development of new transtibial drilling techniques purporting to alter the tibial tunnel starting point to allow the surgeon to reach the anatomic femoral site via the tibial tunnel, research began to focus on the articular location of the tibial tunnel. In a comparison of the anteromedial and the more medial transtibial techniques, with twenty patients in each group, Silva et al. noted that the center of the tibial tunnel was more posterior with use of the transtibial technique than it was with the anteromedial technique (55.4% versus 44.4%, respectively, of the anteroposterior distance of the tibia), yet the center of the femoral tunnels remained lower in the anteromedial tunnels than in the transtibial tunnels (34.7% versus 24.0%, respectively, of the height of the femoral condyle)15.
In a prospective study of thirty patients who underwent primary autogenous bone-patellar tendon-bone ACL reconstruction by high-volume sports-medicine surgeons with either transtibial or anteromedial portal drilling, MRI analysis showed no significant difference in the position of the reconstructed femoral footprint. However, on the tibial side, the transtibial tunnel centers were an average of 5.23 mm posterior to the center of the anatomic footprint as compared with the anteromedial tunnel centers, which were 0.8 mm anterior, and the sagittal obliquity of the transtibial drilled tunnels were 66.9° as compared with 52.2° for ACLs with anteromedial tunnels (native ACL obliquity was 53.5°)16. These values were consistent with those of drilled tunnels in a similar study in which the transtibial sagittal obliquity was found to be 72° as opposed to 53° for the anteromedial group (native ACL obliquity was 52°)17. The former study demonstrates that, with careful tibial tunnel placement, the femoral tunnel can be placed anatomically at the cost of a more posterior tibial tunnel causing a vertical graft secondary to a nonanatomic tibial tunnel16. Proponents of the transtibial approach may argue that the anterior portion of the tibial footprint only represents the anterior flare of the ACL insertion and that the functional portion of the ACL inserts only into the posterior portion of the tibial footprint. Placing an ACL graft more posterior in the tibia can also aid in preventing the impingement that can occur with vertical transtibial grafts; however, a more posterior tibial tunnel leads to a more vertical graft, which is associated with higher failure rates18-22.
Beginning in 2011, anteromedial and transtibial portal drilling was evaluated for double-bundle ACL reconstruction in two studies by Kopf et al.42,43. The two transtibial portals were compared with the anteromedial portal drilling for their ability to achieve an anatomic placement of the anteromedial bundle. Via the anteromedial tibial tunnel and the posterolateral tibial tunnel, transtibial drilling could place the femoral anteromedial bundle within its anatomic origin 4.4% and 60.2% of the time as compared with 100% of the time with anteromedial portal drilling42. The results of transtibial double-bundle tunnels were evaluated with use of three-dimensional CT scans; the tibial tunnels were medial to the anatomic posterolateral position, and the femoral tunnels were anterior to both the anatomic anteromedial and posterolateral positions43.
The ability to more consistently place the ACL graft in its native insertion site with the anteromedial technique than with the transtibial technique can have implications for graft passage. With the transtibial technique, a graft can always be passed in a straight line. However, the independent drilling of the anteromedial technique may require the graft to make an intra-articular turn while being passed. This potential turn is easier to navigate with a soft-tissue graft, whereas a bone block can be more challenging and may require manipulation with an arthroscopic probe during graft passage.
Effect of Drilling Techniques on Biomechanics
The above evidence suggests differences between the ability of the anteromedial and transtibial techniques to place the tibial and femoral tunnels in an anatomic position. In addition to the evaluation of these differences, multiple studies have evaluated the biomechanical differences of anteromedial and transtibial drilling techniques (Table IV).
A series of laboratory studies has raised biomechanical concerns for vertically placed grafts, regardless of drilling technique. A cadaveric study first demonstrated that vertical grafts were able to control anterior tibial translation but were not able to control rotational loads or anterior translation with combined valgus and internal rotational torque44. It was then demonstrated that a more anatomic reconstruction was better able to control translation under these combined loads45. Anatomic and vertical ACL grafts in fresh-frozen ACL deficient knees were compared with use of virtual reconstructions. The anatomic ACLs demonstrated greater strain and fiber elongation in anterior tibial translation and internal rotation5. The investigators theorized that the greater strain and elongation suggested an improved ability to resist translation and rotation.
The first comparative biomechanical analysis of the anteromedial and transtibial techniques only evaluated anterior tibial translation, and the investigators found that neither technique completely reproduced the anterior tibial translation of an intact knee and that there was no difference between the two techniques at any flexion angle (0°, 30°, 60°, or 90°). They found that the anteromedial but not the transtibial technique was able to restore the normal anterior translation with a combined valgus and internal rotation torque in full extension. Neither technique was able to restore normal anterior translation at 30° of knee flexion, although the anteromedial technique was significantly closer46.
The second biomechanical analysis of the two techniques compared anterior tibial translation as well as rotational control. Both Lachman and anterior drawer testing showed that the anteromedial technique was able to reproduce the constraint of the native ACL, whereas the ability of the transtibial technique to reproduce the constraint of the native ACL was shown by the results of anterior drawer testing but not Lachman testing. The transtibial technique was unable to reproduce native rotational control as measured by a manual or instrumented pivot shift; the anteromedial technique may have overconstrained the knee, as it provided more rotational control than the native ACL20. Biomechanical analysis of twelve cadaveric knees that underwent double-bundle ACL reconstruction showed that tensioning the posterolateral bundle in >30° of flexion overconstrained the knee compared with the intact state, whereas tensioning in 0° to 15° of flexion did not lead to overconstraint when the anteromedial bundle was tensioned in <75° of knee flexion47. The long-term importance of potentially overconstraining rotation is unknown. However, with regard to translational constraint, a randomized clinical trial on single-bundle ACL reconstruction found no difference in outcome scores or cartilage damage at three years in knees that restored normal anteroposterior laxity and those that overconstrained anteroposterior laxity by 2 mm48.
A recent biomechanical comparison of two different femoral tunnel positions (the center of the femoral origin and the center of the anteromedial bundle), both drilled through the anteromedial portal, found that both femoral tunnel locations were able to restore anterior translation. The study also demonstrated that, overall, centrally drilled femoral tunnels as opposed to femoral tunnels in the center of the anteromedial bundle may better approximate rotational knee stability without loss of anterior translational control49.
Schairer et al. compared the in vivo kinematics of anteromedial and transtibial portal drilled ACL reconstructed knees under translational and rotational loads by making use of the intact contralateral knee as a control50. In the anteromedial group, the kinematics of every measure was similar to the controls. However, in the transtibial group, there was more medial tibial translation and external rotation than there was in the control group (8.4° versus 2.9°). They also found increased lateral compartment contact area in the anteromedial group as compared with the control group, but not in the transtibial group. The importance of this finding is unknown.
Comparative Studies of Anteromedial and Transtibial ACL Reconstructions
Despite good-quality studies on tunnel placement, complications, and biomechanical outcomes, well-designed prospective randomized controlled trials with long-term follow-up have not yet demonstrated clear differences in clinical outcomes associated with the two drilling techniques. A few comparative studies exist that compare the clinical outcomes of the two techniques51-53.
A prospective randomized controlled trial comparing anteromedial and transtibial portal drilling in ACL reconstruction with hamstring autograft with 86% follow-up (sixty-five of seventy-six patients) demonstrated no clinical difference between the Lysholm knee score and KT-1000 anterior laxity score at twelve months after surgery51. Unfortunately, this study did not clarify the method of randomization or indicate whether it was blinded, and it did not evaluate rotational laxity. Furthermore, many patients are only just beginning to return to their preoperative activities by twelve months after surgery; thus, a longer follow-up is likely needed to determine any clinically significant difference.
A blinded cross-sectional study of autograft bone-patellar tendon-bone ACL reconstructions in soccer players with two to five-year follow-up revealed that patients who underwent anteromedial as compared with transtibial drilling techniques had a significantly shorter time to walking without crutches, return to daily activities, jogging, training, and play. Anteromedial drilling also led to improved stability, as determined with use of the KT-1000 anterior laxity score, the Lachman test, the pivot-shift sign, and objective International Knee Documentation Committee (IKDC) scores. The IKDC score was higher for the anteromedial group; however, no significant differences were found in other clinical outcomes measures, including the visual analog score for satisfaction with surgery and the Lysholm, Tegner, or Short Form-12 scores52.
A review of the Danish Knee Ligament Reconstruction Register of over 9000 primary ACL reconstructions from 2007 through 2010 indicated a revision rate of 5.16% with use of the anteromedial technique as compared with 3.20% with use of the transtibial technique. Secondary end points of the study included an increased relative risk of 2.86 for a positive pivot shift and 3.70 for sagittal instability with use of the anteromedial rather than the transtibial technique53. The findings of decreased stability with the anteromedial technique conflict with the majority of prior cadaveric and clinical studies, and the investigators noted that the majority of the surgeons in the anteromedial group were learning the technique. With regard to their observation that the anteromedial group had a higher revision rate than the transtibial group did, the investigators theorized that, because the ACLs in the anteromedial group may experience higher forces due to their more anatomic position, they may be more likely to fail.
Unlike the results reported in the aforementioned Danish study, the MOON group in the U.S. recently reported the opposite result24. In a comparative study of 380 patients with ACL reconstructions from a prospective cohort from the years 2002 and 2003, transtibial as opposed to anteromedial portal drilling was associated with an increased risk of repeat ipsilateral knee surgery at six years postoperatively, with an odds ratio of 2.49. Although the reasons for this association were not studied, the authors hypothesized that the increased risk of repeat surgery in the transtibial group may have been due to decreased rotational control in a nonanatomic graft, leading to “higher forces on the meniscus and altered pressure distribution on the articular cartilage.” They also suggested that the transtibial group may have had more vertical tunnels, which can lead to a symptomatic cyclops lesion. Despite the increased risk of ipsilateral knee surgery, the Knee injury and Osteoarthritis Outcome Scores Quality of Life (KOOS-QOL) subscale score was unchanged between the groups.
ACL reconstruction techniques have begun a transition toward drilling the femoral tunnel through the anteromedial portal rather than through transtibial drilling in an attempt to improve clinical outcomes by improved graft placement. The abundance of cadaveric and in vivo models has demonstrated that anteromedial portal drilling more consistently and reproducibly places the graft in a more anatomic position, which has been shown to lead to fewer failures than nonanatomic or vertical grafts. While both drilling methods improve anterior translation, anteromedial portal drilling leads to better rotational control. Despite the improvement in tunnel position and biomechanics, the comparative studies to date have failed to show any clinically significant differences in outcome scores or revision rate between transtibial and anteromedial portal drilling in ACL reconstruction (Table V).
Source of Funding: No outside funding was used for this study.
Investigation performed at the UPMC Center for Sports Medicine, Pittsburgh, Pennsylvania
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. One or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. No author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
- Copyright © 2014 by The Journal of Bone and Joint Surgery, Incorporated