Radiographic Evaluation

Specific radiological assessments have to be performed to address tunnel placement, tunnel enlargement, type and location of hardware, degenerative changes, and posterior cruciate ligament (PCL) insufficiency.

The performed radiographs should routinely include the following:

Fig. 57-1 Lateral radiograph in maximal extension (orange line represents Blumensaat line) demonstrating an excessively anterior placement of the tibial tunnel. Sagittal hardware localization is easy to identify.

Fig. 57-2 Clinical and radiological view of extensive hyperextension of a right knee. The orange line is the Blumensaat line demonstrating anterior impingement of the graft although the tibial tunnel (white line) is placed correctly, due to the hyperextension of the knee. Thus, in case of hyperextension, the need for a more posterior placement of the tibial tunnel as normally recommended has to be evaluated carefully.

Fig. 57-3 A 45-degree posteroanterior weight-bearing radiograph according to Rosenberg, indicating degenerative medial joint narrowing compared with the contralateral knee (arrow).

Fig. 57-4 A 45-degree posteroanterior weight-bearing radiograph according to Rosenberg, showing narrow notch configuration (so-called “gothic notch”).

Fig. 57-5 Posterior stress radiographs of both knees after anterior cruciate ligament (ACL) reconstruction of the right knee demonstrating unrecognized posterior cruciate ligament insufficiency of the right knee.

In cases with clinically apparent varus or valgus malalignment or posterolateral insufficiency, long standing radiographs should be made to assess the need for additional axis correction. In case of tunnel enlargement, a computed tomography (CT) scan is indicated to clearly analyze its dimensions (Fig. 57-6).

Fig. 57-6 A and B, Conventional radiographs demonstrating tibial tunnel enlargement. C, Computed tomography (CT) scan performed to exactly assess the dimension of the tunnel enlargement.

Concomitant Pathology

The stabilizing functions of the single knee ligaments are not strictly separated. Secondary graft failure can often be found in combination with unrecognized concomitant ligament pathology, even in perfect ACL reconstructed knees. In fact, all the knee restraints work as a unit and normal knee stability can only be achieved if all ligaments are intact. In cases of ACL insufficiency, secondary restraints such as the posterior oblique ligament (POL) and the posterior horn of the medial meniscus might compensate for the instability. However, if there is medial collateral ligament (MCL) insufficiency (especially POL) and/or a subtotal medial meniscus loss, the freshly reconstructed ACL might not be able to absorb the particular forces and might stretch out. In principle this can be applied to any other combination of ligamentous knee injuries. Thus all directions of knee instability, especially rotatory instabilities, have to be assessed carefully during clinical (and radiological) examination. In detail these instabilities include the PCL (see Fig. 57-5), lateral or posterolateral rotatory instability (PLRI; Fig. 57-7), anteromedial rotatory instability (AMRI), anterolateral rotatory instabilities (ALRI), and MCL or posteromedial rotatory instabilities (PMRI; Fig. 57-8).

Fig. 57-7 Posterolateral rotatory instability. A, Clinical view demonstrating increased external rotation of the right knee. B, Radiograph after posterolateral reconstruction, which in this case was not successful. C, Intraoperative view demonstrating severe posterolateral opening.

Fig. 57-8 Posteromedial rotatory instability. A, Schematic illustration demonstrating increased stress on the anterior cruciate ligament (ACL) in a case of medial instability. B, Intraoperative view demonstrating severe medial opening.

Now we have to question which of these concomitant pathologies needs to be addressed surgically in revision procedures. If an AMRI or PMRI is present, a certain grade of instability might be left untreated in a revision case, as the influence of these instabilities is still not clearly defined and the outcome of surgery (MCL, POL, PMRI) often is unsatisfactory. If there is severe valgus opening, a posteromedial reefing (e.g., that according to Hughston) or a ligament grafting could be done.^17,18 Whether a microperforation of the medial structures according to Rosenberg is successful in these conditions needs further clinical research. In cases of lateral instabilities including PLRI and ALRI, the situation is different. In former years a grossly positive pivot shift was often treated by an additional anterolateral stabilization such as according to Lemaire^19–21 (Fig. 57-9). At present, an additional extraarticular procedure is not routinely recommended; however, in revision reconstructions one might make an exception. In cases with only a slight anterior translation combined with a gross pivot shift, one might consider an additional anterolateral stabilization in selected cases if the femoral ACL graft position is lateral (10-o’clock position). If, for example, a revision reconstruction with lateral femoral tunnel placement does not restore the pivot shift during revision surgery, one might consider an additional Lemaire procedure in these rare cases. However, only one peripheral concomitant pathology should always be addressed during ACL revision reconstruction, the PLRI (see Fig. 57-7), because the high failure rate of ACL reconstruction in combination with a PLRI is well known.^22–24

Fig. 57-9 Lateral extraarticular tenodesis using a strip of the iliotibial band in a case of anterolateral rotatory instability (Lemaire procedure).

In principle, single-staged repair of any of these combinations is possible if boundary conditions allow for the planned surgical procedure. One major exception has to be assessed carefully, which is clinically apparent as the “fixed posterior subluxation.” Normally this phenomenon is discussed in the context of PCL insufficiency and repair. However, it has been demonstrated that approximately 15% of all patients with PCL insufficiency underwent previous isolated ACL reconstruction.²⁵ The reason for this failure often is a wrongly interpreted positive Lachman test, which in the case of PCL insufficiency and an intact ACL might demonstrate an increased way with a firm endpoint. Thus it is of great importance to always evaluate the PCL in ACL deficient knees. Additionally, the PCL deficient knee easily subluxates to posterior during ACL graft fixation, which might lead to a fixed posterior subluxation.²⁶ This results in changed femorotibial biomechanics and high patellofemoral reaction forces. Patients who demonstrate a fixed posterior subluxation often suffer from very early and severe degenerative lesions. If a fixed posterior subluxation is diagnosed, the initial treatment includes the wear of a special brace (PTS Brace, Medi GmbH, Bayreuth, Germany), which pushes the tibia to anterior.^25,26 Depending on boundary conditions and ligamentous instability, revision ACL reconstruction as well as PCL reconstruction might be indicated.²⁶

Hardware Removal

Hardware removal might result in distinct postoperative morbidity or large bone defects (e.g., transfixation devices or deeply countersunk metal interference screws) (Fig. 57-10). Thus older fixation devices need to be removed only if they compromise new tunnel placement or graft fixation. If metal fixation devices do not completely affect new tunnel placement, they often can be pushed into the cancellous bone by serial tunnel dilation (Fig. 57-11). Biodegradable interference screws, even if they are not degraded during revision surgery, do not need to be removed because they can easily be overdrilled. Only a careful washout of particles from the joint cavity is needed to prevent later chondral lesions and/or inflammatory response. Hardware removal also should be performed if older fixation devices clearly are responsible for local pain or discomfort (Fig. 57-12).

Fig. 57-10 Completely countersunk femoral transfixation device. Its removal necessitated an additional lateral approach and opening of the lateral femoral cortex.

Fig. 57-11 Tibial metal screw that is pushed into the cancellous bone by serial tunnel dilation.

Fig. 57-12 Hardware that needs to be removed due to local irritation or pain.

Due to the variability of fixation devices available on the market, hardware removal can be extremely frustrating. Attempts to remove hardware using instruments that do not fit exactly to the fixation devices might result in large bone defects and iatrogenic lesions. Thus in our experience a set of specially designed revision screwdrivers can be very helpful. These instruments fit to most of the fixation devices available (Fig. 57-13).

Fig. 57-13 A set of specially designed revision screwdrivers.

(By permission of KarlStorz, Tuttlingen, Germany.)

Most of the standard metallic interference screws can be removed nicely in a single-staged procedure. However, if metal screws have been deeply countersunk (tibial site, femoral outside-in) and need to be removed for revision reconstruction, one might consider a two-staged procedure because the removal in these circumstances could be very time consuming. Moreover, if the removal of metal implants and ACL revision reconstruction are planned in the same procedure, the patient should be informed that it might be possible to end in a two-staged procedure if problems occur during implant removal. Sometimes problems occur during surgery; for example, if there is an incompletely incorrect tunnel (see later discussion) and the metal screw is large (9 × 25 mm), the additional removal of that screw might lead to a large bone defect, communicating with the old and desired new tunnel. In these cases, bone grafting in a two-staged procedure might be necessary. Sometimes the removal of metallic cross-pin devices, especially the Arthrex TransFix might be tricky, because the manufacturer recommends the countersinking of the implant (see Fig. 57-10). Based on our own experience, we always schedule patients with a metallic TransFix for a two-staged procedure.

Tunnel Malplacement

One of the major keys to successful ACL reconstruction is a correct tunnel placement. Tunnel malplacement can lead to graft impingement, elongation, and failure; loss of range of motion; and high tibiofemoral contact forces.

According to the current literature the femoral tunnel should be drilled in the 10-o’clock position for right knees or in the 2-o’clock position for left knees.^27–29 This means that a lateral femoral tunnel position should be achieved in order to control rotation as well as anterior displacement of the knee. A femoral tunnel drilled at the 12-o’clock position (“high-noon” or central cruciate) might be appropriate to control anterior translation but does not allow for rotational stabilization.^30,31 Often in these cases a normal Lachman test with firm endpoint combined with a positive pivot-shift test can be found in the clinical examination.

Radiographically the tibial tunnel should be placed directly posterior to the intersection point of the Blumensaat line and the tibial joint line (lateral view in full knee extension) in order to avoid anterior impingement of the graft. Placement of the tibial tunnel too anteriorly might lead to an extension deficit due to notch impingement, which possibly results in long-term graft failure.¹⁴

Excessively anterior placed tibial tunnels can cause high-tension forces to the graft in knee flexion (“nutcracker knee”). In addition to graft elongation and secondary failure, this may lead to a flexion deficit and increased tibiofemoral pressure in flexion. These forces might result in early and massive chondral defects, even in young patients.

In addition to early chondral changes or graft failure, tunnel malplacement also leads to painful knees (femoral anterior placement) in many of these cases. If loss of range of motion and an intact (sometimes very tight) ACL are found in combination with tunnel malplacement, arthroscopic graft resection and arthrolysis have to be evaluated. If performed, these should be followed by intensive physiotherapy to regain full range of motion prior to ACL revision reconstruction.

Classification of Existing Tunnel Positions

At our institution we established a classification of the position of formerly placed tunnels to aid in the specific planning of the revision procedure. Tunnel positions are diagnosed using conventional radiographs (lateral view in maximal extension, 45-degree posteroanterior weight-bearing radiograph) and are graded as follows:

Fig. 57-14 A, Completely correct femoral and tibial tunnels. B, Completely incorrect femoral and incompletely incorrect tibial tunnel. (White line represents the Blumensaat line.)

Surgical Management

In general, the type of graft used during the index surgery should be known. If a bone–patellar tendon–bone (BPTB) graft was used, an incompletely incorrect tunnel at the femoral site might not create problems if the bone plug was not countersunk because a certain osseous fill-up of the defects is present. If a soft tissue graft was used during index surgery, an incompletely incorrect tunnel might offer the most challenging problem.

Correctly positioned tunnels measuring less than approximately 8 mm in diameter can be reused if primary reconstruction was performed with a soft tissue graft, depending on the desired type of fixation (isolated versus hybrid). In these cases, tunnel preparation includes removal of intratunnel soft tissue using a drill or a shaver followed by drilling or dilation of the re-created tunnel to the desired diameter. The important effect of this procedure is a débridement of the tunnel wall by removing sclerotic bone so that graft incorporation can proceed (Fig. 57-15).

Fig. 57-15 Reuse of a completely correct femoral tunnel. Tunnel is cleaned by shaver and drill bit (A, B) followed by serial dilation (C, D).

In cases of complete incorrect tunnel placement or complete bony replacement of the previous graft (e.g., BPTB), new tunnel preparation can be done as in primary ACL reconstruction. If an old and completely incorrect tunnel shows excessive enlargement, it should be filled with a cancellous bone plug or, as an alternative, with a biodegradable interference screw prior to graft fixation to prevent collapse between the old and new tunnels (Fig. 57-16).

Fig. 57-16 Completely incorrect femoral tunnel (shown with tunnel enlargement) is filled with a biodegradable interference screw prior to graft fixation to prevent collapse between the old (o) and new (n) tunnels.

Surgically the most demanding cases are those with incompletely incorrectly placed tunnels. Drilling a correctly positioned tunnel in these cases might lead to huge bone defects. To achieve stable tunnel conditions, we suggest initially drilling a correctly placed tunnel with a diameter of only 4 to 5 mm and then using the serial dilation technique to the desired diameter. This procedure leads to a compaction of cancellous bone from the newly created tunnel into the old tunnel. In critical cases with very large bone defects or low bone density, a biodegradable interference screw or a cancellous bone plug can be placed into the old tunnel prior to dilation (Fig. 57-17). However, if there is any uncertainty, a two-staged procedure should be performed.

Fig. 57-17 Filling of an incompletely incorrect tunnel with a biodegradable interference screw prior to drilling and serial dilation of the new tunnel.

Tunnel direction must be verified in the coronal plane in addition to the tunnel entry position. Because many surgeons use a transtibial technique for femoral tunnel creation, the direction of a new tunnel drilled via the anteromedial portal diverges from the old tunnel, which finally improves graft fixation even if the tunnel entry site is enlarged (Fig. 57-18). Thus the anteromedial portal technique should be recommended routinely in all ACL revision procedures.

Fig. 57-18 Intraoperative and radiological views of femoral tunnel divergence. o, Old tunnel; n, new tunnel.

The importance of an intraoperative impingement test to evaluate tibial tunnel position must be emphasized. This test can be performed by placing the dilatator into the tibial tunnel. In full knee extension, the likelihood that anterior graft impingement will occur can be easily assessed. If excessive hyperextension was found in the clinical and radiological examination, the tibial tunnel should be placed more posterior than normally recommended to avoid notch impingement. In this situation the intraoperative impingement test should be performed very attentively.

Tunnel Enlargement

Bone tunnel enlargement commonly occurs following an extracortical (nonanatomical) fixation technique combined with soft tissue grafts due to intratunnel graft motions or with BPTB grafts on the tibial site^16,32 (windshield-wiper and bungee-cord effect) (Fig. 57-19). An additional biological factor that might lead to tunnel enlargement is inflow of synovial fluid between the tunnel wall and the graft during cyclical loading, which leads to high intratunnel pressures and the activation of osteoclasts by synovial cytokines. In these cases, radiographically the tunnel enlargement often appears pear–shaped below the joint line (see Fig. 57-19). The inflow of synovial fluid into the tunnel depends on the type of graft and its fixation.

Fig. 57-19 A, Lateral radiograph demonstrating femoral tunnel enlargement with semi-anatomical fixation device. B, Lateral radiograph demonstrating tibial pear-shaped tunnel enlargement, which might be due to inflow of synovial fluid.

Factors that inhibit synovial inflow are as follows:

In contrast, the inflow of synovial fluid is advanced by the following:

If excessive tunnel enlargement can be seen on regular radiographs, a computed tomography (CT) scan is indicated to identify its exact dimensions. Based on CT findings of tunnel enlargement, the question of whether a two-staged procedure is necessary directly depends on the desired type of graft and fixation technique.

If a two-staged procedure is indicated, the first step includes autologous or allogenic cancellous bone or bone substitute filling followed by revision ACL reconstruction. On the femoral site, the impaction of a cylindrical iliac crest graft or a cylindrical synthetic filler (β-tricalcium phosphate [TCP]) can be nicely performed arthroscopically (Fig. 57-20). The use of bone chips on the femur might require special instruments if performed arthroscopically.¹⁶ On the tibial site one might use a single or two cylindrical plugs, which should be impacted. However, in most cases a tight filling with this technique is difficult. Thus we recommend using autologous bone chips, which are eventually mixed with synthetic materials (β-TCP) to tightly fill the old tunnel. In this case one should be careful not to compact bone chips into the joint cavity. We therefore recommend not to débride soft tissue from the tibial tunnel aperture site, which can later prevent bone chip access to the joint.

Fig. 57-20 Arthroscopic filling of an enlarged femoral tunnel with a cylindrical synthetic bone substitute (β-TCP).

A new CT scan before revision ACL reconstruction should be done to evaluate the exact bony recovery. If filling of the defects was successful, revision ACL reconstruction can be performed using the same methods as for primary reconstruction. The minimum time between bone grafting and revision reconstruction should be 3 months; 6 months is optimal.

Graft Selection

Similarly to primary ACL reconstruction, we have to differentiate grafts with bone blocks from soft tissue grafts and autografts from allografts. At present, the use of synthetic ligament substitutes is obsolete because of high failure rates and a high incidence of chronic knee inflammation.

Due to a rising number of revision ACL reconstructions performed, the use of allografts is becoming more popular, especially in the United States. Advantages of allografts include reduced operative trauma (no donor site morbidity), decreased operation time, smaller incisions, and the available choices of graft sizes in multiple ligament reconstructions. One disadvantage is longer graft incorporation time compared with autografts. Thus, the post-treatment needs to be adapted. The use of allografts with huge bone blocks might be helpful if huge tunnel enlargement is present. However, the risk of disease transmission must be considered and discussed with the patient.

Hamstring tendons have become increasingly popular in ACL surgery due to decreased donor site morbidity, improved cosmesis, and at least identical clinical outcome compared with autologous BPTB grafts if modern fixation techniques are used.³³

At our institution we routinely use hamstring tendon autografts (four-stranded semitendinosus tendon) for primary and revision ACL reconstruction if local boundary conditions (tunnel enlargement) allow for their use (see Fig. 39-3). In revision ACL reconstruction it might happen that the favored graft already has been harvested for primary ACL reconstruction. Thus if the use of an autograft is desired, it has to be decided whether another graft from the ipsilateral knee will be used or whether the graft will be harvested from the uninjured contralateral knee. If ipsilateral hamstrings were already taken for previous surgery, we routinely harvest the contralateral semitendinosus tendon. In cases when no hamstring tendons are left, we prefer autologous quadriceps tendon grafts rather than BPTB autografts or different allografts (e.g., hamstring tendon, BPTB, tibialis anterior tendon, Achilles tendon).

Graft Fixation

The standard procedure in our institution for ACL revision reconstruction is a direct fixation technique by the use of biodegradable interference screws and hybrid fixation (Table 57-1) at both sites. General surgical principles of this technique, as well as possible intraoperative problems and their solution strategies, are described elsewhere in this text (see Chapter 46). The use of interference screws allows for an anatomical fixation at the level of the joint line, which has been shown to increase graft isometry³⁴ and knee stability by reducing the graft length to its intraarticular portion, thus increasing construct stiffness.^35,36 The use of interference fit fixation additionally reduces intratunnel graft motion, which is a common side effect of nonanatomical (extracortical) graft fixation.

Table 57-1 Graft Fixation Technique

At present it is generally recommended to perform hybrid fixation at the tibial site, not only in revision but also in primary ACL reconstruction, for the following reasons:

An easy and secure method for tibial backup fixation is a suture of the linkage material over a bony bridge. To do so, a monocortical drill hole is created 2 cm distally of the tibial tunnel exit site. Then one strand of each attached suture is passed through the hole and tied over the created bony bridge³⁷ (see Chapter 46).

For femoral hybrid fixation with interference screw fixation, we prefer the use of a biodegradable spherical device (EndoPearl, Linvatec, Largo, FL). The EndoPearl is sutured to the femoral end of the graft and achieves an internal locking between the graft and the tip of the interference screw. Thus it increases fixation strength, especially in cases of tunnel enlargement and low bone density (see Chapter 39).

Technical problems of graft fixation in revision ACL reconstruction often appear due to graft–tunnel mismatch, especially in cases with completely correct or incompletely incorrect tunnel positions. The principles of tunnel placement depending on previously drilled tunnel positions were previously explained. If, after appropriate tunnel creation, the diameter of the femoral tunnel still is 1 to 3 mm larger than the graft diameter, femoral hybrid fixation with a biodegradable interference screw plus the EndoPearl device prevents graft slippage and allows for secure graft fixation (e.g., for a 9-mm tunnel and 7-mm graft, use a 9-mm EndoPearl and 8-mm interference screw).

In cases with rather insecure tibial interference screw fixation, we favor the use of a suture button instead of the suture over a bony bridge. Manual rotation of the button tightens the linkage material, thus preventing graft slippage (see Chapter 46). Furthermore, in cases with impaired tibial bone quality, we suggest the use of oversized screws (diameter of 11–12 mm; Fig. 57-21) or the use of two screws (sandwich technique; Fig. 57-22) for tibial graft fixation.

Fig. 57-21 Metallic and biodegradable interference screws are available in different sizes and diameters. Thus a precise matching of graft and tunnel diameter and screw size can be performed.

Fig. 57-22 The use of two screws for tibial fixation in a case of tibial tunnel enlargement.