Choosing appropriate portals is a key step in the planning of every arthroscopic surgery. When attempting to execute an anatomic ACL reconstruction, it is imperative to attain optimal visualization of the native anatomy with unrestrained scope movement. Below are the descriptions of three different techniques to reach anatomic positioning of the femoral tunnel using independent drilling through different accessory anteromedial portals.

The three-portal technique, which is described below, employs the high anterolateral portal (LP), the central anteromedial portal (CP), and the accessory anteromedial portal (AMP) (Fig. 23.2). It is best suited for anatomic SB or DB ACL reconstruction allowing an excellent balance between adequate three-dimensional visualization of the knee structures and optimal angle of attack for the instruments used for placing anatomic femoral tunnels [4].

An accessory medial portal gives the arthroscopist a better viewing angle of the resident’s ridge, the bifurcate ridge, and the posterior wall of the lateral femoral condyle. However, both the lateral and medial portals can have limited viewing when one has to flex the knee past 90° vertical.

There is a third viewing option that overcomes the limitations as presented for lateral and medial viewing portals. This is a superior accessory anteromedial portal (SAM) first described by Dinesh Patel in the 1980s. The knee is placed in a figure four position, placing the operative leg over the contralateral leg. The borders of the medial aspect of the inferior patella and the superior medial femoral condyle are palpated. Just below, one can readily palpate a soft spot in the capsule. The scalpel blade is then gently directed inferolaterally, and the scope is introduced into the knee joint. This superior placement allows for a second lower medial portal that can be made in a horizontal fashion, for drilling of the ACL femoral tunnel (Fig. 23.6).

This portal essentially gives one an axial view of the entire lateral femoral condyle, particularly the relationship of the bifurcate ridge, the resident’s ridge, and the posterior wall of the lateral femoral condyle (Fig. 23.7a–c). This allows for greater accuracy of k-wire placement than any of the two other portals. Further with a little varus pressure, one can flex the knee more than 90° without compromising the view during k-wire placement and drilling of the tunnel.

When the knee is placed in 70–90° of knee flexion in a vertical position, the SAM provides an axial view of the tibial plateau where the relationship between the base of the tibial spine and the anterior aspect of the tibia and the menisci can be far more accurately assessed.

If one desires to be as objective as possible, I believe that the SAM provides better opportunity to achieve this.

This inside-out technique is not applicable to the knee without passive flexion over 140°, as less flexion results in blowout of the tunnel. In this deep flexion, reduced joint cavity volume may disturb view to the attachment area. Care must be taken to avoid damage to the articular cartilage of the medial femoral condyle.

Good fixation may be achieved with an interference screw, but not with a button around the tunnel opening on the lateral cortex because of softer bone quality and shorter tunnels in the physis. Thus, it is our opinion that this technique is applied for the anatomical rectangular tunnel ACL reconstruction with a bone-patellar tendon-bone graft or bone-quadriceps tendon graft, not for a round tunnel reconstruction with soft tissue graft [47, 48].

The distal thigh is kept horizontal using a leg holder with the calf hanging. In addition to the routine anterolateral (AL) and anteromedial (AM) portals, the far anteromedial (FAM) portal is created 2–2.5 cm posterior to the AM portal and just above the medial meniscus (Fig. 23.8) [46]. This portal makes it possible for instruments to get more perpendicular access to the ACL femoral attachment area on the lateral wall of the notch.

The fibrous tissue, including the ACL stump, on the superior-posterior half of the lateral wall of the intercondylar notch is thoroughly removed using a radiofrequency device through the FAM portal, while the posterior third of the lateral wall of the notch is simultaneously viewed via the AM portal with a 45° oblique arthroscope. Mechanical shavers may not be utilized in order to preserve subtle undulation of the bony surface around the attachment area. After cleaning up, the attachment area is clearly delineated by the resident’s ridge, anteriorly; upper cartilage margin, superiorly; and posterior cartilage margin, posteriorly (Fig. 23.9) [23, 42, 49].

Two points are marked with a 5-mm distance in the center of the attachment area along its long axis to the resident’s ridge using radiofrequency device and a microfracture awl.

With the knee deeply flexed over 140° while viewing with the arthroscope via AM portal, two guide pins are drilled from the marked points to the lateral femoral cortex via the FAM portal and then overdrilled with a 5.0-mm cannulated acorn drill bit (Fig. 23.10). The two round holes are dilated into one parallelepiped socket with the 5 × 10-mm cannulated dilator (Figs. 23.11a, b and 23.12).

Many authors experimentally studied the use of the AMP for the execution of the femoral socket in ACL reconstruction. Cadaveric specimens, synthetic knee models, and computational knee models were evaluated with regard to the optimal knee flexion angle to achieve anatomic positioning of the guide pin for preparing the femoral socket [3, 5, 8, 15, 16, 36, 37, 39, 44, 55, 64]. Zantop et al. used 60 bone models to recreate the drilling of the PL bundle tunnel using three different simulated knee flexion angles at 70°, 90°, and 110°, in combination with either a low or a high AMP, to define which was the safest choice to avoid lateral femoral condyle cartilage damage [64]. The findings suggest that flexing the knee 110° and using the low AMP minimize the risk of cartilage damage. Nakamura et al. tested the same knee flexion angles in ten cadaveric specimens in combination with a FAM portal, also finding that higher knee flexion angles better avoid cartilage damage [37]. Farrow et al. utilized seven fresh frozen cadaveric knees to drill guide pins in the anatomical AM and PL bundle positions using an accessory anteromedial approach with 90°, 110°, and 130° of knee flexion [8]. The exit of the guide pins in the lateral femoral cortical bone was identified, and the distance between each pin and lateral gastrocnemius, articular cartilage of the lateral femoral condyle, and lateral collateral ligament were measured. Again, safer distances were achieved when the knee was flexed more than 110°. To better understand the influence of the knee flexion angle and the resulting tunnel length and inclination, Badeski et al. used nine cadaveric specimens with the knee flexed at 90°, 110°, and 130° to find increasingly horizontal PL bundle femoral tunnels and decreasing risk of femoral tunnel blowout when increasing knee flexion angles [5]. There is no clear consensus, but the majority of studies conclude that within a range of 100–130°, the resulting tunnels will have sufficient length to allow graft bone interface for proper fixation and graft healing. Lower flexion angles increase the odds of the guide pin hitting lateral structures like the common peroneal nerve, iliotibial tract, biceps tendon, popliteal tendon, lateral collateral ligament, and the lateral gastrocnemius. The risk of tunnel blowout and subchondral damage to the posterior aspect of the lateral femoral condyle was also associated with lower flexion angles. The optimal knee flexion angles vary with the shape and size of the ACL footprint that is being reconstructed. Higher flexion angles result in longer tunnels in comparison to decreased flexion angles. Controlling all of these variables to provide the best individualized treatment for every patient is challenging and requires meticulous planning and accomplishment.

Comparison of the AMP techniques with the TT technique was also studied experimentally. Bedi et al. used 18 cadavers to compare the femoral tunnel length and obliquity at 100°, 110° and 120° of knee flexion, concluding that the higher the flexion angle, the more oblique the tunnels. This could lead to shorter tunnels and posterior cortical wall blowout [6]. The same author showed that at time zero, the AMP technique better restored the Lachman and pivot-shift test, while TT approach leads to enlargement of the tibial tunnel aperture due to the eccentric position and over-reaming in the posterolateral direction when trying to recreate the anatomic position [7]. Tudisco et al., on the other hand, found no significant difference on anteroposterior (AP) translation at time zero, but again found lower pivot shift with the AMP group than the TT group [2]. Many other studies found experimentally that the femoral tunnels were drilled closer to the anatomic position with adequate coverage of the footprint area by the use of the AMP technique as compared to the TT technique [12, 13, 54, 56].

Koutras et al. and Franceschi et al. showed that although the AMP techniques lead to more anatomical femoral sockets than the TT technique, no significantly better clinical outcomes were found [10, 27]. Noh et al. in a randomized controlled trial found that the Lysholm score and the AP translation were significantly improved in the AMP group in comparison to the TT group, while International Knee Documentation Committee (IKDC) and Tegner activity scale had no significant difference [38]. The Danish ACL Reconstruction Registry, a large prospective cohort, identified a higher rate of ACL re-tear with the use of the AMP technique explaining that higher in situ forces are experienced by the anatomically placed graft in comparison to a more vertical graft and that surgeons likely encounter a steep learning curve when shifting to an anatomical technique after years of using a TT technique [43]. To better understand the learning curve associated with transition from TT to AMP technique, Inderhaug et al. demonstrated significant improvement with regard to anatomical positioning of the femoral tunnel in a surgeon transitioning from TT to anatomical technique by providing post-op feedback by 3D CT scan after a first series of surgeries [22]. The assessment of the tunnel position by 3D CT scan regularly demonstrated that AMP and outside-in had similar results, both superior to TT groups [45, 51] (Takeda and Shin).

Although not always feasible, it is possible to achieve an anatomic femoral tunnel aperture position by the use of TT technique. Kopf et al. used the tibial tunnels created for a DB ACL reconstruction and the AMP in 113 patients to check if they could get guide pins to the center of the native femoral footprint and found that by the use of the AM tibial tunnel only, 4 % of the center of femoral footprints could be reached by the guide pin, followed by 64 % when using the PL tibial tunnel and 100 % using the AMP [25].