Patients
We retrospectively reviewed the records of 425 patients who underwent ACL reconstruction between February 2001 and December 2005. All patients were managed by the senior author (S.-J.K.). The inclusion criteria for the present study were (1) a single-bundle ACL reconstruction with use of bone-patellar tendon-bone autograft, (2) a unilateral ACL injury without a contralateral knee injury, (3) no axial or rotational malalignment of the lower extremity (<5 mm of mechanical axis deviation as measured on standing hip-knee-ankle radiographs12), (4) no history of surgery involving the lower extremity, (5) no chondral lesion greater than grade 2 (fissuring and fragmentation measuring <0.5 in [<13 mm] in diameter) according to the Outerbridge system13 at the time of surgery, (6) an intact meniscus or prior partial meniscectomy with maintained hoop tension (that is, a meniscectomy involving less than one-third of the entire meniscus), and (7) a duration of follow-up of at least twenty-four months. Forty-six patients who were managed with isolated ACL reconstruction (Group I) and twenty-three patients who were managed with anatomical posterolateral corner reconstruction of the lateral collateral ligament (LCL) and popliteal tendon (Group II) met these criteria and were included in the present study.
Patients who had an isolated ACL injury with less than grade-1 varus laxity on physical examination (<3 mm of laxity as compared with the normal, contralateral side) did not require surgery. Posterolateral rotatory instability was diagnosed on the basis of the reverse-pivot-shift test and >10° of external rotatory laxity (as compared with the normal, contralateral knee) on the dial test with the knee in 30° and 90° of flexion and with the patient in the prone position and under general or spinal anesthesia. Informed consent to take part in the present study was given by all patients, and institutional review board approval was obtained.
Surgical Procedure
Single-Bundle Technique for ACL Reconstruction with a Bone-Patellar Tendon-Bone Autograft
A bone-patellar tendon-bone graft with a width of 10 mm was harvested. The patellar and tibial bone blocks were of a trapezoidal shape and measured 20 mm long and 8 mm deep. A tibial tunnel was drilled with use of a 10-mm-diameter cannulated reamer. A femoral guide pin was positioned at 10:30 o'clock on the right knee or 1:30 o'clock on the left knee, with the knee flexed 70° to 90°, and the femoral socket was reamed to a depth of 25 mm through the tibial tunnel. The previously prepared bone-patellar tendon-bone graft tendon was passed through the tibial tunnel, across the joint, and into the femoral socket. The graft was secured within the femoral socket with a bioabsorbable interference screw. The graft was pretensioned by pulling it tightly and moving the knee through a full range of motion ten times. The graft then was fixed within the tibial tunnel with a bioabsorbable interference screw at 10° to 15° of knee flexion.
Posterolateral Corner Reconstruction of the LCL and Popliteal Tendon with a Tibialis-Posterior-Tendon Allograft
On the lateral aspect of the knee, the plane between the iliotibial band and the biceps femoris was developed to expose the lateral epicondyle of the femur, the Gerdy tubercle, and the fibular head. An ACL guide was placed with one tip on the point 10 mm inferior to the knee joint line and 5 mm medial to the proximal tibiofibular joint, and the other tip was placed on the Gerdy tubercle. A guide pin was inserted under fluoroscopic guidance, and then a 7-mm tibial tunnel was created. A 7-mm fibular tunnel also was created at an angle of 70° to the floor in an anteroinferior-to-posterosuperior direction with use of an ACL guide placed with one tip at the point just posteromedial to the LCL and the other tip at the anteroinferior aspect of the fibular head. With use of a looped wire, each end of the posterior tibialis allograft, which was sutured in a whipstitch fashion, was passed through the tibial and fibular tunnels in an anterior-to-posterior direction and was secured with bioabsorbable interference screws. The EndoPearl fixation device (CONMED Linvatec, Largo, Florida) was attached at each end of the graft. The end of the graft for the popliteal tendon, exiting from the tibial tunnel, was routed to a Kirschner wire that was provisionally inserted at a tentative isometric point on the anterior one-fifth of the popliteal sulcus, about 15 mm distal and 5 mm anterior to the femoral epicondyle, and an isometric test was performed. The isometric point was confirmed to have <2 mm of migration of the rerouted graft. After the isometric point was determined, a 7-mm femoral tunnel was created over a Kirschner wire. A leading suture was connected to the EndoPearl fixation device and was used to pull the graft for the popliteal tendon into the femoral tunnel. The graft was pretensioned ten times through a knee range of motion. A bioabsorbable screw was inserted into the femoral tunnel, and the graft was secured. The other end of the graft for the LCL, exiting from the fibular head, was routed to the anterior margin of the lateral epicondyle, and an isometric test was performed in the same manner. A 7-mm femoral tunnel was made, and the graft was preconditioned, followed by femoral fixation with a bioabsorbable interference screw in the femoral tunnel. The femoral tunnel was created in the anterosuperior direction at an angle of 20° to the transverse line of the femoral shaft to minimize interference with the ACL tunnel14,15. A more detailed description of the surgical technique in the present study is provided in the study by one of us (S.-J.K.) and colleagues16 (Fig. 1).
Postoperative Rehabilitation
Isolated ACL Reconstruction
All patients were permitted immediate weight-bearing as tolerated and knee motion. By the twelfth week after surgery, swimming, cycling, and jogging were permitted. Return to sports involving jumping, pivoting, or sidestepping was allowed six months after surgery.
Posterolateral Corner Reconstruction Combined with ACL Reconstruction
The reconstructed grafts were protected by immobilization for the first four weeks. Isometric quadriceps-strengthening exercises and mobilization of the patella were initiated immediately after surgery. Weight-bearing as tolerated was allowed in this period. At four weeks after surgery, flexion of the knee as tolerated was permitted. At six to eight weeks after surgery, closed-chain kinetic exercises were started. At ten to twelve weeks after surgery, stationary cycling, stair-stepping, and single-leg stance were allowed. Full squatting was prohibited until three months after surgery. At four to five months after surgery, swimming and fast walking in a pool were allowed. At six months after surgery, return to full activity was allowed.
Clinical Assessments
Clinical outcomes were determined on the basis of the data that were obtained before surgery and at the time of the twenty-four-month follow-up. Ligament laxity was examined with the Lachman, varus, pivot-shift, reverse-pivot-shift, and dial tests. The Lachman and varus tests were graded according to the amount of side-to-side difference (as compared with the normal, contralateral side) as 0 (<3 mm), 1+ (3 to 5 mm), 2+ (6 to 10 mm), or 3+ (>10 mm). The pivot-shift test was graded as 0 (absent), 1+ (subluxation), 2+ (jump), or 3+ (transient lock). For determination of the amount of external rotation on the dial test, the angle between the axis of the medial border of the foot and the femur was measured.
Varus laxity was quantified on the basis of varus stress radiographs made with 30° of knee flexion using a Telos Stress Device (Telos, Marburg, Germany) with a 150-N varus load17. Varus laxity was measured according to the method described by Jacobsen18. One horizontal line was drawn tangential to the most proximal subchondral bone portion of both tibial plateaus. Another line was drawn tangential to the most distal portion of both femoral condyles in the same manner. One line perpendicular to the tibial-side horizontal line was drawn tangential to the most lateral border of the tibia. The distance between the two horizontal lines on this vertical line was regarded as the varus laxity. Anterior translation was examined with a KT2000 arthrometer (MEDmetric, San Diego, California) at 30° of knee flexion at the standard force of 134 N. To assess reliability, each evaluation was performed twice by two different orthopaedic surgeons. The average of the two individual mean values was used to determine the final side-to-side difference. Functional outcomes were assessed with use of Lysholm scores and the International Knee Documentation Committee (IKDC) form. Radiographs were made at the time of the twenty-four-month follow-up and were compared with those that were made preoperatively. These radiographs included posteroanterior weight-bearing views with the knee in 15° and 30° of flexion as well as lateral and skyline views. The radiographs were classified according to the system of Kellgren and Lawrence19.
Statistical Analysis
SPSS software (version 18.0; Statistical Package for the Social Sciences, SPSS, Chicago, Illinois) was used for statistical analyses. The Mann-Whitney U test was used to compare the outcomes for both groups, including Lysholm knee scores (ranked continuous data). For the side-to-side differences on the dial test, KT2000 arthrometer testing, and varus stress radiographs (continuous data), two-sample t tests were used to compare the two groups. For comparison of the IKDC scores (categorical ordinal data), the chi-square test was employed. The interobserver reliability of the measurements of laxity on stress radiographs, KT2000 arthrometer testing, and the dial test was evaluated with use of the intraclass correlation set at a 95% confidence interval. The level of significance was set at p < 0.05.
Source of Funding
There were no external funding sources for this study.
The mean age at the time of surgery was 36.0 years (range, twenty to fifty-four years) in Group I and 36.4 years (range, twenty-one to fifty-three years) in Group II (see Appendix). Group I included thirty male patients and sixteen female patients, and Group II included sixteen male patients and seven female patients. The injuries in Group I included thirty-seven sports injuries (80.4%), six motor-vehicle injuries (13.0%), and three falls (6.5%). The injuries in Group II included sixteen sports injuries (69.6%), four motor-vehicle injuries (17.4%), and three falls (13.0%). The average interval between the injury and the operation was 7.2 months (range, 0.7 to twenty-eight months) in Group I and 7.8 months (range, one to thirty months) in Group II.
Clinical Assessments
In Group II, the reverse pivot-shift test was negative for twenty-two patients (95.7%) after posterolateral corner reconstruction. The mean side-to-side difference on the dial test at 30° and 90° of knee flexion improved significantly (p < 0.01) from 15.3° to 4.8° (intraclass correlation coefficient for intertester reliability, 0.661) and from 11.3° to 3.2° (intraclass correlation coefficient for intertester reliability, 0.650), respectively, when the preoperative values were compared with the values at the time of the twenty-four-month follow-up. Rotational stability analysis revealed that nineteen patients had similar or more stability on the injured side than on the normal side. Of the four patients with laxity, three had a side-to-side difference of <10° in terms of external rotation and one had a difference of >10°. With regard to varus laxity, twenty-one patients had a grade of 0 and two had a grade of 1. Varus stress radiographs revealed that the mean side-to-side difference in displacement decreased significantly from 3.7 mm preoperatively to 0.5 mm at the time of the twenty-four-month follow-up examination (intraclass correlation coefficient for intertester reliability, 0.797) (p < 0.05) (Table I).
With regard to anterior laxity as measured with the KT2000 arthrometer, the mean side-to-side difference in displacement decreased substantially from 6.2 mm preoperatively to 2.2 mm postoperatively (intraclass correlation coefficient for intertester reliability, 0.854) in Group I and from 6.5 mm preoperatively to 1.6 mm postoperatively (intraclass correlation coefficient for intertester reliability, 0.867) in Group II. Postoperatively, the mean side-to-side difference was significantly greater for Group I (2.2 ± 1.0 mm) than for Group II (1.6 ± 0.8 mm) (p = 0.031) (Table II). At the time of the twenty-four-month follow-up, seven knees (15.2%) in Group I and two knees (8.7%) in Group II had a positive Lachman test (grade 1) and three knees (6.5%) in Group I and one knee (4.3%) in Group II had a positive pivot-shift test (grade 1).
Substantial improvements in terms of both Lysholm and IKDC scores were observed in both groups between the preoperative and follow-up examinations. The mean Lysholm score improved from 73.2 to 93.2 in Group I and from 64.4 to 90.1 in Group II (Table III). At the time of the latest evaluation, ten knees (21.7%) in Group I were classified as IKDC grade A (normal), twenty-eight (60.9%) were classified as grade B (nearly normal), four (8.7%) were classified as grade C (abnormal), and four (8.7%) were classified as grade D (severely abnormal). Therefore, thirty-eight (82.6%) of forty-six knees had a satisfactory result. In Group II, six knees (26.1%) were classified as grade A (normal), fourteen (60.9%) were classified as grade B (nearly normal), two (8.7%) were classified as grade C (abnormal), and one (4.3%) was classified as grade D (severely abnormal). Therefore, twenty (87.0%) of the twenty-three knees had a satisfactory result. There was no significant difference between the groups with regard to the IKDC and Lysholm scores at the time of the twenty-four-month follow-up (p = 0.) (Table IV).
Complications
In Group I, two knees (4.3%) had a flexion deficit of >5° and three knees (6.5%) had an extension deficit of >5° in comparison with normal, contralateral knee. In Group II, one knee (4.3%) had a flexion deficit of >5° and two knees (8.7%) had an extension deficit >5°. A correction loss of posterolateral rotational instability of >10° was found in one knee (4.3%) in Group II. Progressive radiographic osteoarthritic changes as compared with the preoperative findings were identified in one knee (2.2%) in Group I and in one knee (4.3%) in Group II at the time of the twenty-four-month follow-up examination. There was no morbidity associated with posterolateral corner reconstruction. Revision surgery was performed, in the same manner as the primary operation, for one patient with recurrent posterolateral rotatory instability.
The frequency of ACL instability combined with posterolateral corner injuries has not been investigated thoroughly. Several studies have suggested that as many as 11% of ACL tears are associated with posterolateral corner injuries9,20. The present series showed that thirty-two (7.5%) of 425 patients who were managed with ACL reconstruction during the same period also had a posterolateral corner injury. The results of anatomical posterolateral corner reconstruction combined with ACL reconstruction were satisfactory in terms of rotational, varus, and anteroposterior stability restoration.
The present study was conducted to investigate the influence of anatomical posterolateral corner reconstruction on anteroposterior laxity and clinical outcomes in cases of posterolateral rotatory instability associated with ACL injury. We compared the results for knees with combined ACL and posterolateral corner injuries with those for knees with isolated ACL injuries. Our hypothesis was that a combined injury of the ACL and posterolateral corner would lead to worse results than an isolated ACL injury because a combined injury results from higher velocity trauma and is associated with more tissue damage as compared with an isolated injury. In addition, longer operative and tourniquet times may increase tissue damage and may be associated with poor clinical outcomes. However, in the present study, the clinical results for knees with combined ACL and posterolateral corner injuries were as good as those for knees with isolated ACL injuries, with no significant differences between the groups in terms of Lysholm and IKDC scores and with less anteroposterior laxity after combined posterolateral and ACL reconstruction than after isolated ACL reconstruction.
One reason for the finding that there was less laxity after combined ACL and posterolateral corner reconstruction may be related to the fact that patients managed with isolated ACL reconstruction have minimal posterolateral instability that does not need surgical intervention but that does cause more laxity in the anterior-posterior direction. Lorbach et al.21 investigated knee kinematics in a study of cadaver knees and found increased tibial external rotation ranging from 4.8° to 5.0°, but not exceeding 5.0°, after cutting of the ACL. Therefore, external rotation of <10° on the dial test might indicate a partial injury that is pathologic but does not need any surgical treatment. However, in such cases, the rotational laxity would increase stress on the reconstructed ACL. Chhabra et al.22 reported that the anatomical two-bundle approach to ACL reconstruction yielded increased rotational stability and was associated with good functional recovery. Cadaveric studies also demonstrated that anatomical femoral tunnel placement (at the 10 o'clock position) led to better control of tibial rotation than did placement of a tunnel higher in the notch23,24. Achieving a more anatomical femoral tunnel placement during ACL reconstruction might restore better rotational, especially posterolateral, stability.
The present study showed that only one patient (4%) had overcorrection of posterolateral rotational instability, which is lower than the rates reported in other studies. Fanelli et al.25 reported that 71% of their patients with posterolateral corner insufficiency experienced overcorrection with use of the Clancy technique. Kim et al.26 showed that the modified biceps rerouting technique yielded 33% overcorrection. We believe that our better results may have been due to the more anatomical and isometric location of the graft in our technique. Anatomical reconstruction was effective for restoring stability27,28 and reducing the limitation of knee motion after surgery29. In addition, osteoarthritic changes were not remarkable compared with those reported in other short-term studies involving patients with normal menisci and articular cartilage at the time of surgery30,31.
The anatomy of the posterolateral corner is complex, and there is considerable variation among individuals. Consequently, a number of reconstructive techniques have been described, which can be a source of confusion32,33. It is technically challenging to reconstruct all of the principal structures (LCL, popliteal tendon, and popliteofibular ligament) of the posterolateral corner in an isometric manner. In our technique, we reconstructed the LCL and the popliteal tendon but not the popliteofibular ligament. Biomechanical studies have revealed that LCL-popliteal tendon reconstruction has demonstrated results that have been comparable with those of popliteal tendon-popliteofibular ligament reconstruction and popliteofibular ligament-LCL reconstruction34. Höher et al.35 reported that among posterolateral corner structures the popliteal tendon appeared to be a major stabilizer under the posterior tibial loading condition.
In recent years, biomechanical studies have demonstrated that posterolateral corner structures interact functionally with the cruciate ligaments. Markolf et al.36 measured the force in a simulated intact ACL with and without posterolateral knee structures and found increased force on the ACL with varus loading after cutting the posterolateral corner structures. LaPrade et al.8 reported that an increase in external rotation force caused clinical disability and loss of function, but that the main cause of increased abnormal loading of the ACL in patients with posterolateral rotatory instability was the LCL. Zantop et al., in a dissection study, demonstrated that the fibular insertion of the LCL is moved anteriorly and is thereby tightened in internal tibial rotation, which aligns the LCL with the ACL to be a primary restraint to withstand anterior tibial translation under the combined rotatory load37.
The present study had some inherent limitations. First, the comparison between the two groups was not completely justified. We could not compare ACL and posterolateral corner reconstruction with isolated ACL reconstruction among patients who had ACL and posterolateral corner injuries because of the rarity of isolated ACL reconstruction in such patients. Second, we could not clarify the degrees of tibial external rotation in Group I, which made it difficult to evaluate the correlation between the degrees of external rotation and the gray zone (the status of posterolateral structures that lie between normal and pathological posterolateral rotatory instability). Third, in Group I, varus instability was determined on the basis of physical examination. Although all of the patients in Group I had less than grade-1 varus instability, varus stress radiographs would have shown a better quantified measurement. Fourth, the measurement of external rotatory instability with the dial test was not very accurate and was not quantified with use of instruments such as stress radiographs, although two orthopaedic surgeons measured the laxity twice, and the mean of the individual means was determined.
However, until now, there has been no single definitive tool for the diagnosis of posterolateral rotatory instability. Among physical examinations for posterolateral rotatory instability, the dial test is the most sensitive38,39 and is not overlooked. Experimental sectioning of posterolateral corner structures resulted in a substantial increase in external rotation at all angles of flexion as compared with the intact knees. When the posterior cruciate ligament (PCL) was sectioned after the posterolateral corner structures had been cut, an exaggeration of external rotation was produced, but when the ACL was sectioned after the posterolateral corner structures had been cut, there was no added external rotation40. Therefore, ACL injuries combined with posterolateral rotatory instability are more difficult to diagnosis with the dial test than PCL and posterolateral corner injuries are.
In conclusion, the results of the present short-term study indicate that combined ACL and posterolateral corner injuries can be successfully treated with arthroscopic ACL reconstruction and anatomical posterolateral reconstruction for the popliteal tendon and LCL. The results of combined ACL and posterolateral corner reconstruction were similar to the results of isolated ACL reconstruction in terms of clinical scores and even better in terms of anteroposterior laxity improvement.
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. None of the authors, or their institution(s), have had any financial relationship, in the thirty-six months prior to submission of this work, with any entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. Also, 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.