History: Signs and Symptoms

ACL injuries occur frequently in sports that involve running, jumping, and cutting movements. They can occur without contact when the foot is anchored to the playing surface—usually by way of cleats or a rubber sole—and the body rotates beyond the tolerance of the ligament as the knee buckles. Thus, it is important to ask the patient how the injury occurred and the position of the knee during the injury, which may also allude to the ACL bundle injury pattern.¹⁵ This may be associated with an audible “pop.” Asking whether the athlete was able to continue to play will give you an idea of the severity of the injury. Knee pain and a hemarthrosis are usually present acutely. A complaint of instability is also common, especially with walking downhill or down stairs.

Physical Examination

Inspect and palpate for an effusion. If a large effusion is present, consider aspiration for pain relief, and inspect the aspirate for any fat globules, which would be suggestive of a fracture. Check the range of motion; if it is limited, magnetic resonance imaging (MRI) should be ordered to ensure that no displaced meniscal tear is present. The physical examination of an isolated ACL tear is usually significant for a side-to-side difference with regard to Lachman and pivot-shift maneuvers. If a discrepancy between the Lachman and pivot-shift maneuvers exists, this may signify a partial tear involving either the AM or PL bundles. The PL bundle is mainly responsible for rotational stability, and a large pivot shift will be evident if it is torn. Similarly, the AM bundle is mainly responsible for translational stability when the knee is flexed, and a large Lachman maneuver will be present if the AM bundle is torn. A KT-1000 test can also be used to confirm a side-to-side difference in anterior translation. A more prominent anterior drawer compared with a Lachman test should alert the examiner to consider a concomitant posteromedial or posterior horn medial meniscal injury. Varus/valgus instability testing should be performed to ensure no collateral injury is present. Dial testing and posterolateral drawer testing at 30 degrees should be performed to assess for a posterolateral knee injury. Gait analysis should be performed to inspect for any underlying varus laxity. Tests for possible meniscal pathology should also be performed (i.e., joint line tenderness, McMurray maneuver), but it may be difficult to distinguish between a lateral meniscal tear and a bone contusion acutely. Thus, appropriate imaging is important.

Imaging

A complete knee series consisting of weight-bearing anteroposterior and notch, lateral, and patellofemoral sunrise views should be obtained. The soft tissues can be inspected for an effusion. The bony anatomy should be inspected for any fractures or subtle signs of a rotational injury, such as segond or reverse segond capsular lesion. In addition, the status of the physes and any arthritic changes should be noted. Long cassette films should be obtained for any patient with varus alignment on examination or if any arthritic changes are noted on the knee series obtained. This will help determine whether an osteotomy should be performed. An MRI is essential not only to confirm an ACL injury but, more importantly, to assess for any concomitant ligamentous or cartilage injuries that will affect the operative plan. The PL bundle is more easily visualized on coronal sectioning. Specifically, it may be seen at the level of the first cut, which includes the PCL.

Indications

The absolute indications for double-bundle ACL reconstruction are evolving. Even though single-bundle ACL reconstruction is considered the “gold standard,” the technique can be improved. Gait analysis after single-bundle reconstruction has demonstrated that rotatory instability persists.⁵ Furthermore, biomechanical cadaveric studies have shown that even lowering the femoral insertion site to the 3- or 9-o’clock position does not fully prevent rotatory instability.¹⁶ Clinically, as many as one-fifth of the patients do not resume preinjury activities and usually complain of vague instability symptoms that objectively correspond to a mild persistent pivot shift.¹⁷ In comparison, double-bundle ACL reconstruction does restore the rotational component in a cadaveric model.¹⁴ It has been suggested that a positive pivot shift after ACL reconstruction is correlated with the development of later osteoarthrosis.¹⁸ Perhaps with reconstruction of both the AM and PL bundles, the decreased rotational instability will provide improved overall knee kinematics and may prevent or slow the degenerative changes seen after single-bundle ACL reconstruction.⁴ A contraindication to performing the double-bundle technique is in the young athlete with open physes. Two tunnels would risk physeal arrest with subsequent malalignment and possible leg length discrepancy.

Anesthesia and Positioning

The operative extremity is identified by the patient and initialed by a member of the surgical team. All patients undergo a preoperative femoral nerve block in the holding area by our anesthesia colleagues. The patient is then placed in a supine position and given intravenous conscious sedation. A careful exam under anesthesia is performed and recorded to document the Lachman and pivot-shift maneuvers. Again, the senior author is interested in correlating the exam with the tear pattern of the individual bundles of the ACL. A tourniquet is applied to the proximal thigh. The extremity is then secured within a circumferential leg holder placed at the level of the tourniquet. The foot of the operating table is completely retracted to permit hyperflexion of the knee, which is crucial for later placement of the PL femoral tunnel. The contralateral extremity is placed within a well-leg holder with the hip flexed approximately 90 degrees and abducted and externally rotated away from the surgical field to allow unobstructed access to the operative knee (Fig. 25-1). The leg is elevated for 5 minutes, and the tourniquet is then inflated. The knee is prepped and sterilely draped.

Fig. 25-1 Leg positioned to allow for range of motion between full extension and 120 degrees of flexion.

Anterior Cruciate Ligament Graft Preparation

Two tibialis anterior allografts are individually fashioned as a double loop. The folded length of each graft should be approximately 12 cm for sufficient graft tissue. The grafts are trimmed to a folded diameter of 7 mm for the PL bundle and 8 mm for the AM bundle. A #2 braided suture is whipstitched up and down both ends of the graft for 3 cm. The stitch depth is alternated, and care is taken to avoid penetrating the suture and risking weakening or breaking. The graft is then passed through the closed-looped Endobutton (Smith & Nephew, Andover, MA). Two Fiberwire sutures (Arthrex, Naples, FL) (one stripped and one nonstripped for later identification) are placed within the button holes. A 2–0 absorbable suture is tied through both strands of the folded graft to secure them once the graft is passed within the closed-looped Endobutton. Each graft is marked to alert the surgeon when to engage or “flip” the Endobutton (Fig. 25-2).

Fig. 25-2 Doubled-over tibialis anterior allograft with whipstitch. Anteromedial diameter, 8 mm; posterolateral diameter, 7 mm. Endobutton-CL pictured at right.

Surgical Landmarks

With the knee flexed approximately 45 degrees, the inferior pole of the patella is marked. The inferior extent of the lateral parapatellar portal begins at the level of the inferior pole of the patella and extends proximally for approximately 2 cm. The medial parapatellar portal begins at the level of the inferior pole of the patella and extends distally along the medial aspect of the patellar tendon for approximately 2 cm. The high placement of the portals allows the arthroscopic instruments to enter the knee above the level of the fat pad. The 11 scalpel blade is angled approximately 45 degrees to the skin and distally toward the notch to safely enter the knee without harming the articular cartilage. A low AM accessory portal will be placed with the assistance of a spinal needle to ensure proper trajectory for the PL femoral tunnel (Fig. 25-3). The knee is flexed 90 degrees, and the 11 scalpel blade is angled upward as it enters the skin at the level of the spinal needle marking. The portal is extended proximally approximately 1 cm, avoiding injury to both the underlying meniscus and the articular cartilage.

Fig. 25-3 Arthroscopic portal placement and skin incisions. AMP, Accessory medial portal; MP, medial portal; LP, lateral portal.

Diagnostic Examination

The knee is flexed approximately 25 degrees, and the arthroscopic trochar is placed in the lateral parapatellar portal angled toward the notch and then redirected beneath the patella. The patellofemoral joint is visualized. Any cartilage defects are addressed as needed. The arthroscope is then dropped down over the trochlea and into the notch to grossly view the ACL. The knee is then extended, and a valgus stress is applied to open the medial compartment. The entire meniscus is visualized and probed for stability. Medial meniscal tears are usually seen with chronic ACL tears. Any meniscal pathology is addressed with either repair or débridement depending on the location and character of the tear. The knee is brought through a range of motion to inspect the femoral condylar articular surface for any defects. The tibial plateau articular cartilage is also inspected for any defects. The knee is then placed in a figure-four position to view the lateral compartment. The PL bundle is best visualized in this position. The entire meniscus is once again observed and probed for stability. Lateral meniscal longitudinal tears are commonly seen with acute ACL tears at the junction of the middle and posterior thirds. The posterior third of the lateral meniscus is more easily viewed with the knee flexed to 90 degrees; the middle third with the knee flexed to 60 degrees; and the anterior third with the knee flexed to 30 degrees. Any meniscal pathology is addressed with either repair or débridement. The lateral femoral condylar articular surface is inspected for any defects as the knee is brought through an entire range of motion.

Specific Steps

Attention is then redirected to the notch. Any obstructing fat pad or ligamentum mucosum is removed with the shaver or ArthroCare Coblation device. The bundles of the ACL are carefully dissected with a small ArthroCare Coblation device to fully appreciate the injury tear pattern. Because the two bundles are differentially tensioned depending on the position of the knee at the time of injury, the tear pattern can be quite different for each bundle.¹⁸ We are currently performing a prospective study to correlate the preoperative examination (i.e. KT-1000, examination under anesthesia [pivot-shift and Lachman maneuvers]) with the individual tear pattern of each bundle (Fig. 25-4). The remnant of the ACL bundles is then débrided from both their femoral and tibial insertions. The ArthroCare Coblation device is used to mark the AM and PL femoral and tibial footprints (Fig. 25-5). No notchplasty is performed.

Fig. 25-4 Anterior cruciate ligament (ACL) tear of anteromedial (AM) and posterolateral (PL) bundles, each from femoral insertion. Preoperative exam under anesthesia demonstrated a 3+ Lachman and a 3+ pivot shift. LFC, Lateral femoral condyle.

Fig. 25-5 Anteromedial (AM) and posterolateral (PL) tibial (A) and femoral (B and C) footprints marked. LFC, Lateral femoral condyle.

An 18-gauge spinal needle is directly visualized as it passes from the location of the low AM portal onto the medial face of the lateral femoral condylar notch in the region of the previously marked PL bundle (Fig. 25-6). Once satisfactory trajectory for the PL bundle with the spinal needle has been determined, remove the spinal needle and make the low AM portal with an 11 scalpel blade angled upward to avoid cutting the anterior horn of the medial meniscus. Pass a -mm Steinman pin via the low AM portal onto the medial face of the lateral femoral condylar notch. Place the pin approximately 8 mm posterior to the anterior articular margin and approximately 5 mm superior to the inferior articular margin of the lateral femoral condyle. Once correct placement of the pin has been obtained, bring the knee into approximately 120 degrees of hyperflexion and tap it into place with a mallet (Fig. 25-7). Ensure that the pin did not penetrate the medial meniscus, and then slide the 7-mm acorn reamer over the Steinman pin and have it rest against the medial face of the lateral femoral condylar notch. Ream to a depth of 25 mm. Then place the Endobutton drill within the previously reamed canal, and drop your hand before starting the drill, which will maximize the PL tunnel length. With the knee still flexed at 120 degrees, have your assistant place his or her hand on the lateral aspect of the knee and push the biceps tendon inferiorly, which will deflect the common peroneal nerve away from the drill trajectory. Barely perforate the lateral femoral cortex with the Endobutton drill, and quickly retract it backward to minimize the risk of injury to the common peroneal nerve. Measure the transcondylar length, and choose the appropriately sized continuously looped Endobutton. If the length is greater than 35 mm, replace the 7-mm acorn reamer within the PL tunnel and dilate the tunnel depth to 30 mm by hand. There should be at least 15 mm of graft tissue within the tunnel (Fig. 25-8).

Fig. 25-6 A, B, Accessory medial portal visualization with an 18-gauge needle.

Fig. 25-7 A, Guide pin placement for posterolateral femoral tunnel. B, 7-mm acorn reamer as inserted from accessory medial portal. LFC, Lateral femoral condyle.

Fig. 25-8 Posterolateral (PL) femoral tunnel. LFC, Lateral femoral condyle.

Make a 3- to 4-cm incision over the AM surface of the tibia for creation of the tibial tunnels and passage of the grafts. This is in a location midway between the tibial tubercle and the posteromedial border of the tibia. Dissect the soft tissues both medially and laterally for easy access to the future tibial AM and PL tunnels. Identify the tibial PL footprint, which is located just medial to the posterior horn of the lateral meniscus and anterior to the PCL. The footprint should have been previously marked with the ArthroCare Coblation device. Place the ACL director guide (Smith & Nephew, Andover, MA) at 55 degrees with the tip centered within the PL tibial footprint (Fig. 25-9). Drill a Steinman pin through the guide and within the joint so that a few millimeters are visible. Identify the AM tibial footprint with the ArthroCare Coblation device, which is a couple of millimeters anterior to the ideal single-bundle reconstruction location: the posterior aspect of the anterior horn of the lateral meniscus, downward sloping side of the medial tibial eminence, and approximately 7 mm anterior to the PCL. Reset the ACL director guide to 45 degrees and place it within the previously marked AM tibial footprint (Fig. 25-10). Drill a second Steinman pin through the guide and within the joint so that a few millimeters are visible. Note that the PL tibial guide pin is quite vertical in comparison with the AM tibial guide pin, which is quite horizontal. This pin attitude also prevents impingement within the notch because the PL tunnel is centered within the knee and the AM tunnel is horizontal. Once again, no notchplasty is necessary. The PL tibial guide pin also enters the AM surface of the tibia more posteromedially than the AM tibial guide pin, which is more lateral and centered (Fig. 25-11). Retract the skin on the tibia to allow placement of a 7-mm reamer over the PL tibial guide pin, and place a curette within the joint overlying the PL pin to protect the articular surfaces. Ream the PL tunnel, and remove debris with a shaver. Then place the 8-mm reamer over the AM tibial guide pin, and once again place a curette overlying the pin to protect the articular surfaces. Ream the AM tunnel, and remove debris with a shaver. There should be an approximately 1-cm bony bridge between the two tibial tunnels.

Fig. 25-9 A, Posterolateral (PL) tibial insertion landmarks. Anterior cruciate ligament (ACL) director guide set at 55 degrees. B, Guide pin drilling. AM, Anteromedial; Lat men, lateral meniscus; PCL, posterior cruciate ligament.

Fig. 25-10 A, Anteromedial (AM) tibial insertion landmarks, guide pin drilling. B, Anterior cruciate ligament (ACL) director guide set at 45 degrees. PL, Posterolateral.

Fig. 25-11 A, Guide pins for posterolateral (PL) and anteromedial (AM) tibial tunnels. B, External view of guide pins.

Then direct your attention to the AM femoral footprint. Guide the AM femoral tunnel off the previously made femoral PL tunnel and not off the back wall of the notch, which is traditionally done with over-the-top femoral tunnel drill guides. Place the Steinman pin 3 mm posterior and slightly superior to the previously drilled femoral PL tunnel. The Steinman pin can be placed transtibially via the previously drilled AM tunnel or via the low AM portal, whichever will allow the proper trajectory to be low enough within the notch (Fig. 25-12). Hyperflex the knee as the Steinman pin is then tapped into place. Then place an 8-mm acorn reamer over the pin and pass it within the joint to rest against the notch wall. Ream the AM femoral tunnel to a depth of 40 mm. Place the Endobutton drill within the AM tunnel. Your hand should be dropped as the knee is maintained in the hyperflexed position to maximize tunnel length and ensure that the anterolateral femoral cortex will be perforated to allow passage of the Beath pin distal to the tourniquet. Measure the transfemoral diameter, and choose the appropriately sized closed-looped Endobutton. Ideally, there should be at least 15 to 20 mm of graft tissue within the canal.

Fig. 25-12 Anteromedial femoral guide pin landmarks. TT, Transtibial; MP, medial portal; PL, posterolateral.

With the knee hyperflexed, place a Beath pin via the low AM portal, through the PL femoral tunnel, and out of the skin on the lateral aspect of the knee. Once again, you want to drop your hand and push the biceps tendon inferiorly as you pass the pin to protect the common peroneal nerve from injury. Tie a long looped suture through the eyelet of the Beath pin, which is pulled intraarticularly and grasped out the tibial PL tunnel with arthroscopic suture retrievers. Place the 7-mm prepared PL allograft within the long looped suture attached to the Beath pin, and pull it through the respective tibial and femoral tunnels. Flip the Endobutton, and pull the graft to ensure proper engagement. Pass the Beath pin through the tibial and femoral AM tunnels with the knee hyperflexed to ensure the pin exits the thigh distal to the tourniquet and remains sterile. Depending on the trajectory of the tunnels, the Beath pin may first need to be passed via the low AM portal through the AM femoral tunnel and the long looped suture pulled out through the AM tibial tunnel with arthroscopic suture retrievers. Place the 8-mm prepared AM allograft within the long looped suture attached to the Beath pin, and pull it through the respective tibial and femoral tunnels. Flip the Endobutton, and pull the graft to ensure proper engagement (Fig. 25-13).

Fig. 25-13 Posterolateral (PL) graft in place; anteromedial (AM) graft position represented by Fiberwire sutures in tunnel. Note that the position of the two bundles is parallel with the knee in full extension.

Cycle the knee through a full range of motion from approximately 0 to 120 degrees, approximately 20 to 30 times, while maintaining tension on both graft ends to remove any slack, and check the isometry. Tension the PL bundle first between 0 and 10 degrees of flexion. Then tension the AM bundle with the knee in approximately 60 degrees of flexion (Fig. 25-14). Tibial fixation is achieved with bioabsorbable interference screws, which are the same diameter as the corresponding tunnel. One staple is used as adjunctive fixation for each graft on the tibial side. Note that the reconstructed state recreates the crossing pattern of the PL and AM bundles (Fig. 25-15).

Fig. 25-14 Anteromedial (AM) and posterolateral (PL) grafts in situ. The PL bundle is partially obscured by the AM bundle.

Fig. 25-15 Crossing pattern of anteromedial (AM) and posterolateral (PL) bundles. A, Parallel in extension. B, Crossed in flexion. LFC, Lateral femoral condyle.

Rehabilitation

Postoperatively, the patient is placed in a hinged knee brace. Full weight-bearing is allowed with the knee locked in extension. Continuous passive motion (CPM) is started immediately from 0 to 45 degrees of flexion and is increased by 10 degrees per day until the maximal obtainable flexion permitted by the CPM is achieved for 2 consecutive days. This is usually after 1 to 2 weeks. The brace is unlocked at 1 week, and crutches are maintained until quadriceps control is reestablished, typically in 4 to 6 weeks. The accelerated rehabilitation protocol described by Irrgang is implemented with return to contact sports at 6 months with a brace after successful function testing.¹⁹

Complications

We have had three graft failures, all occurring after returning to sport. Two were sustained during contact injuries while playing collegiate football. The third occurred in a noncompliant patient 3 months after reconstruction when she returned to playing high-school basketball without a brace. Four patients have undergone staple removal for symptomatic hardware. To date, after 192 double-bundle ACL reconstructions, we have had no fractures and no radiographic signs of femoral condylar avascular necrosis or tunnel widening.

Results

Several in vivo functional biomechanical studies demonstrate that the kinematics are not completely restored with single-bundle reconstruction.^5,6 Amis¹⁵ has demonstrated that even if the femoral tunnel is placed in a lower position than the traditionally described location, the kinematics are still not normal with regard to rotational stability. In contrast, Yagi et al¹⁴ published their results of double-bundle ACL reconstruction in a cadaveric model in which rotational stability was restored.

Clinical results of double-bundle ACL reconstruction surgery are still evolving (Table 25-1).^12,20–25 In 1987, Zaricznyj¹² published the first clinical results for double-bundle ACL reconstruction. Twelve of the fourteen patients had excellent results. In 1999, Muneta et al²⁴ published preliminary results suggesting that the double-bundle procedure showed a better trend with respect to anterior stability. In 2001, Hamada et al²² published a 2-year follow-up on 160 consecutive patients who underwent single or bisocket ACL reconstructions, demonstrating no statistical significant difference between the two techniques for IKDC, KT measurements, or thigh muscle strength. A trend for better anterior stability was observed in the double-bundle group. No assessment of rotatory stability was mentioned. Furthermore, in 2004, Adachi et al²⁰ performed a randomized prospective study of 108 patients (55 single bundle; 53 double bundle) with an average of 32 months of follow-up (24 to 36 months). No statistically significant difference was noted with regard to knee joint stability (KT-2000) or to proprioception. There was a statistically significant difference with regard to a decreased incidence of notchplasty for the double-bundle group compared with the single-bundle group. The authors did not obtain IKDC results or assess for rotatory stability. Aglietti et al²¹ have an unpublished series of 75 patients (25 single bundle, 50 double bundle) that demonstrates a lower side-to-side KT difference, a lower number of patients with a postoperative pivot shift, and better IKDC functional results for patients who underwent double-bundle ACL reconstruction. In contrast, Yagi et al²³ also have an unpublished series of 60 prospectively randomized patients (20 double bundle, 20 AM reconstruction, 20 PL reconstruction). At 1 year, there were no statistically significant differences regarding side-to-side KT measurements, IKDC functional results, or patients with a postoperative pivot shift.

The senior author has performed a total of 186 primary double-bundle ACL reconstructions to date. The average follow-up is 12 months. The average postoperative side-to-side KT measurement difference is 1.2 mm. Six patients had a 1+ pivot shift and the remainder demonstrated no pivot. In addition, we have recently compared our double-bundle ACL reconstruction, early range-of-motion data with a cohort of single-bundle reconstructions performed by the senior author. The comparison demonstrated with statistical significance that the double-bundle ACL reconstruction patients have consistently achieved earlier better range of motion at 1 week, 4 weeks, and 12 weeks postoperative follow-up (Table 25-2).