Robert de LaPrade et al.
The goal of this study was to determine if untreated grade III injuries of posterolateral knee structures (PLS) contribute to increased force on an ACL reconstructed graft. This was accomplished through measuring the force in an ACL graft during joint loading for an ACL reconstructed (ACLR) knee with otherwise intact structures, and in the same knees with the ACLR after selected cutting of specific posterolateral knee structures.
Eight fresh frozen cadaveric knees were utilized. An ACL reconstruction using a central third patellar tendon graft was performed. The tibial bone plug was fixed to a tensioning jig (on the tibia) through a ligament augmentation device (LAD), which had a buckle transducer applied to it to measure force in the graft. Loads applied were 67 N of anterior and posterior load, 12 Nm of varus and valgus moment, and 5 Nm of internal and external rotation moment. Tests were first performed on the knee with PLS intact, and then after sequential individual sectioning of the fibular collateral ligament (FCL), popliteofibular ligament (PFL), and popliteus tendon (PLT).
The ACL graft force was significantly higher, with the FCL transected during varus loading at both 0º (p < .01) and 30º (p < .01) of knee flexion than it was for the same loading of the joint with intact PLS. In addition, coupled loading of varus and internal moment at 0º (p < .02) further increased graft force beyond that with varus alone. These findings remained significant with additional sequential cutting of the PFL and popliteus tendon.
In conclusion, we found that during joint loading a significant increase in force on an ACL graft occurred during varus and coupled varus-internal rotation moments in knees with grade III PLS deficiency compared to the same reconstruction in the PLS intact knee. We believe this study supports the clinical observation that untreated grade III PLS injury contributes to ACL graft failure, by developing higher forces on the ACL graft and suggests that the varus-thrust gait pattern seen in some of these patients places an ACL graft at increased risk for failure. It is recommended that grade III PLS injuries with evidence of varus instability be repaired/reconstructed at the time of ACL reconstruction to decrease the chance of postreconstruction ACL graft failure.
Injuries to the posterolateral structures of the knee can be easily missed by a less than complete knee exam, especially when there is a concomitant anterior cruciate ligament tear. While recognized as a cause of frank instability, grade III posterolateral knee structure (PLS) injuries (1) have also been implicated as a leading cause of ACL graft failure (21).
Over the past few years, the complexity of the posterolateral corner of the knee has become better understood. The unique biomechanics of the convex lateral tibial plateau, and its articulation with the convex lateral femoral condyle, create a setting for marked instability with even normal gait in the face of PLS injury. Seebacher et al. (24), described the posterolateral anatomy as a three layered complex which helped in its understanding clinically. More recently, articles further describing the anatomy of individual structures of the posterolateral knee and the diagnosis of posterolateral rotatory instability (PLRI) have provided a clearer understanding, and in turn, have sparked further study, of these debilitating injuries (11,12,13,14,25,27,28).
Anterior cruciate ligament reconstruction is one of the most popular surgical procedures in orthopaedics. Failure rates for primary reconstructions are reported to be between 0.7% and 8% (8). Results of graft revisions are much less predictable (7), with failure rates of 5 to 52% (8,19). Many authors have suggested that the main cause of graft failures is unrecognized, and therefore untreated, PLRI (9,11,21).
Markoff (17) measured the force in a simulated intact ACL with and without the PLS and found increased ACL force in varus loading after cutting the PLS. This study may not be applicable to reconstruction of the ACL, since the reconstruction procedure alters the joint kinematics and load sharing between ligamentous structures (F. Wentorf, unpublished data, 1998).
The goal of this study was to determine if untreated grade III injuries of the PLS contribute to increased force on an ACL reconstructed graft to support or not support the clinical suspicion that untreated grade III PLS injury contributes to ACL graft failure. Results would be used as a guide to determine the advisability for reconstructing grade III injured PLS of the knee at the time of ACL reconstruction.
MATERIALS AND METHODS:
We studied 8 fresh frozen cadaveric knees which demonstrated no evidence of previous surgeries, instability by clinical exam, or evidence of more than superficial chondromalacia at the time of dissection. The specimens were frozen at -20 degrees Celsius and allowed to thaw overnight, prior to testing. The specimens were cut on each side to within 15 cm of the joint line. Skin and subcutaneous tissues were then removed from the knees and the following structures were identified: iliotibial band and its components, fibular collateral ligament, fabellofibular ligament, components of the long and short head of the biceps femoris, mid-third lateral capsular ligament, popliteofibular ligament, popliteal aponeurotic attachments to the lateral meniscus, and the ligament of Wrisberg. Polymethylmethacrylate was used to pot the femur and tibia/fibula, which were all separately drilled with metal screws at their ends prior to potting to prevent any rotational slippage in the knee test machine.
The central third of the patellar tendon (10 mm), along with 10 millimeter wide by 25 millimeter long bone blocks off the patella and tibial tubercle, was harvested for the ACL graft. A small medial arthrotomy provided visualization of the tibial and femoral attachment sites of the ACL after resection of the fat pad. The anterior intermeniscal ligament was carefully preserved. The native ACL was then resected and a reconstruction was performed with the femoral and tibial tunnels placed at the attachments of the native ACL. Arthex (Naples, Florida) ACL guides were utilized to drill the tibial and femoral tunnels.
The femoral bone block was secured in the femoral tunnel with a 9 millimeter by 25 millimeter cannulated Arthrex interference screw. The tibial bone plug was secured to a 5 cm length of ligament augmentation device (LAD) (3M, St. Paul, MN), looped at both ends, with a No. 5 nonabsorbable, double looped suture. The looped sutures were attached to a tensioning jig, with a screw mechanism to apply tension, which was rigidly secured to the proximal tibia with K-wires. The jig was centered over the tunnel such that the force applied to the graft complex was parallel to the tunnel. A buckle transducer was placed on the LAD, using similar techniques to those reported previously by this lab (10,15), to measure the forces seen in the ACL graft (Figure 1). The knee was then mounted in the testing machine and the ACL graft was secured to the tensioning jig after the reconstructed ligament had been exercised through a series of flexion/extension and anterior/posterior forces on the knee.
The baseline tension on the graft was determined by applying a 67 N (15 lb) distal traction load with a calibrated spring scale to the ACL graft-LAD complex with the knee in 0º of flexion and then the graft fixed to the tibial jig. This duplicated the force seen on an ACL graft when fixation of the tibial bone block would be performed during an in vivo ACL reconstruction. Once this load was set, it was allowed to counterbalance with the proximal tibia, as would occur clinically during the fixation of the tibial bone block. The baseline calibration force on the graft was established as the force present on the graft when an anterior force of 67 N (15 lb) at 30º of knee flexion (Lachman test) was applied to the tibia. The force on the graft, as measured by the buckle transducer, was recalibrated via the screw tensioning device to this baseline force using this same technique (as necessary) after each testing cycle for each individual knee flexion angle to account for any graft or other tissue creep.
A pilot study on 4 knees was performed prior to the initiation of this study to determine the specific joint loads for this study. Loads which were tested in the pilot study, but not included as part of this study because no graft loading was seen, were posterior drawer at 0º, 30º, and 90º, and valgus with coupled internal or external rotation at 0º and 30º. Although external rotation moments and coupled posterior-external rotation moments did not increase the force on the ACL graft, they were included in this study because of their use clinically in the assessment of abnormal motion testing for the posterolateral knee (3,4,5,29). In addition, the force and moment values chosen for loading these knees were based upon the pilot data. It was found that applied forces or moments higher than those performed resulted in graft failure (bone block fracture (three) or bone-tendon interface failure (one)) for those motions which increased graft force because of the large increase in motion which occurred during the sequential cutting studies.
The following loads were applied to the knee testing machine (15): at full extension: 67 N anterior tibial force, 5 Nm internal and external tibial rotation, 12 Nm of varus and valgus moment, and 5 Nm internal rotation coupled with 12 Nm of varus; at 30 degrees: 67 N anterior drawer, 5 Nm internal and external tibial rotation, 12 Nm of varus and valgus, 12 Nm varus coupled with 5 Nm internal rotation, and 67 N posterior drawer coupled with 5 Nm external rotation; and at 90 degrees: 67 N anterior drawer and 67 N posterior drawer coupled with 5 Nm external rotation. The force on the ACL was recalibrated by the screw mechanism on the tibial tensioning jig back to the baseline graft force seen with a 67 N simulated Lachman test at 30º of knee flexion after each testing sequence for each angle of flexion. The entire testing sequence was performed a total of 3 times. An instrumented spatial linkage (ISL) (10) was used to measure six-degrees-of-freedom joint motion.
Joint loading and motion testing was performed on the knees after simulated ACL reconstruction in the normal state and after sequential cutting, and repeated simulated ACL reconstruction, of what we feel are the most important posterolateral structures based on our pilot and prior studies (11,12,13,14,27,28): (1) the fibular collateral ligament (near its attachment on the femur), (2) the popliteofibular ligament (near its attachment on the posteromedial fibular styloid), and (3) the popliteus tendon (at its origin on the femur). We did not investigate the effect of varying the cutting sequence.
Statistical analysis for the forces on the anterior cruciate ligament graft was performed for differences between the normal and the sequential cutting states of the posterolateral structures during joint loading using Student’s t-test. Analysis was also performed for the differences in anterior/posterior translation, varus/valgus opening, and internal/external rotation between the normal and sequential cutting states, for each motion analyzed. Statistical significance was assumed for P values less than .05.
Force on the ACL Graft
The ACL graft force was noted to change with joint loading when the reconstruction was performed with the PLS absent compared to the same reconstruction performed with the PLS present (Figure 2). Figure 3 summarizes the increase (or decrease) in force seen on the ACL graft, relative to the intact PLS state, during joint loading after ACL reconstruction at each sequential sectioning of the FCL, PFL, and popliteus tendon compared to the baseline state of ACL reconstruction with the PLS intact. Significant changes in force seen on the graft during testing are summarized in the following paragraphs.
There was a significant increase in force on the ACL graft after the FCL was sectioned during varus loading at both 0º (p<.01) and 30º (p<.01) of knee flexion. In addition, coupled loading of varus and internal rotation at 0º (p<.02) and 30º (p<.001) further increased graft force relative to varus alone for this testing state. A significant decrease in graft force was seen with application of a valgus force at 30º of knee flexion after the FCL was cut (p<.04).
Additional sectioning of the PFL and popliteus tendon (after the FCL was cut) revealed similar increases in force on the ACL graft relative to the knee with PLS intact with no significant changes occurring between each additional cutting state. The force on the ACL graft was significantly increased compared to the PLS intact state during varus loading at 0º for the FCL and PFL sectioned, (p<.05); 0º for the FCL, PFL and popliteus tendon sectioned, (p<.02); 30º for the FCL and PFL sectioned, (p<.04); and 30º for the FCL, PFL and popliteus tendon sectioned (p<.01). In addition, a significant increase in force, compared to the PLS intact state, on the ACL graft was seen for coupled loading of varus and internal rotation at both 0º (FCL and PFL sectioned, p<.04; FCL, PFL, and popliteus tendon sectioned, p<.01) and 30º (FCL and PFL sectioned, p<.001; FCL, PFL, and popliteus tendon sectioned, p<.001) of knee flexion.
Changes in Joint Motion
A summary of the changes in varus/valgus opening (Figure 4A), external rotation/internal rotation (Figure 4B), and anterior/posterior displacement (Figure 4C) after sequential sectioning of the PLS and ACL reconstruction compared to ACL reconstruction in the PLS intact state, are presented in Figure 4. Significant changes in joint motion seen with sequential sectioning of the PLS are noted in the graphs. We reemphasize that the graft was reset to the original setting force, simulating an ACL reconstruction for the particular state, after each new cut state in the cutting sequence.
The joint loads which resulted in a significant increase in force on the ACL graft after cutting a PLS structure were varus moments and coupled varus - internal rotation moments. An internal rotation moment by itself did not result in a significant increase in graft force and it can be extrapolated that the main load which increased the force on the ACL graft was a varus moment. The loads applied in our study were chosen for convenience of the test set up. It is instructive to relate these forces to those occurring in vivo. An estimate of these forces can be obtained from Prodromos et al., who measured joint moments in patients before and after high tibial osteotomy (22). These authors presented adduction joint moment (varus) in terms of Body Weight (BW) X Height (HT). They arbitrarily assigned a value of 4% for normal. The highest adduction moment was 6% BW X HT. To compare with our values, these values must be converted to Nm. A typical body weight is chosen as 155 pounds, or 690 N. A typical height is chosen as 5 ft 10 in, or 1.8 m. Four percent BW X HT then becomes 50 Nm and 6% BW X HT becomes 75 Nm. In a worst case, all of this moment would be taken by the passive structures. For the highest graft force (71 N) measured with an applied coupled varus and internal rotation moment of 12 Nm at 0º, if the graft force increased proportionately in relation to joint varus moment, 50 Nm would cause 50 Nm/12 Nm X 71 N = 296 N of force on the graft. Seventy-five Nm (6% BW X HT adduction moment) would cause 444 N of force on the graft. Rowden et al., (23) reported that the failure strength of a patellar tendon ACL graft was 416 N immediately post op. Since animal studies have shown ACL grafts to weaken with time before gaining strength with remodeling (2,18), the forces that may occur for the patient with a high adduction moment gait (6% BW X HT) would be suspected to be at risk of graft stretching or rupture, especially for a patient with a varus thrust gait pattern.
Noyes et al. have argued that the high adduction moment associated with a varus thrust gait increases lateral soft tissue forces and theorized that this high adduction moment could place an ACL reconstruction at risk (20). Other studies have demonstrated that the ACL acts as a secondary restraint to varus angulation (16,17,26). Our study demonstrates that in varus loading absence of the PLS does in fact significantly increase the load in the ACL graft, which places it at an increased risk for failure. It is recommended that the PLS be repaired or reconstructed at the time of ACL reconstruction in these patients. It is also recommended that a complete assessment of the source of the varus moment be performed prior to ACL reconstruction. It is important to determine if the increase is varus is solely due to a grade III PLS injury or is combined with a varus aligned knee. It is recognized that in some knees a proximal tibial osteotomy may also need to be performed to address increased varus joint line opening for those patients with a mechanical axis which passes medial to the central axis of the knee due to medial compartment arthritis or a congenital tendency to genu varus in patients with chronic injuries.
While PLRI has been demonstrated to cause both an increase in varus and external rotation in the knee (3,4,5,6,12,13,29), external rotation associated with PLS sectioning resulted in an unloading of the ACL grafts in this study. While an increase in external rotation has been demonstrated to cause clinical disability and loss of function in patients (5,6,11,12,21), it does not appear to be a main cause of an increase in abnormal loading of the ACL graft in patients with grade III PLRI. It is recognized that our study is a static study and that coupled muscle forces and increases in external rotation could contribute to ACL graft failure clinically through loading forces which we could not measure in the in vitro state.
Analysis of our motion differences for intact knees with ACL reconstructions and sequential sectioning of the PLS and reset reconstructions in these knees revealed significant increases in varus opening for an applied varus moment after each sectioned structure, indicating that all three of these PLS structures (FCL, PFL, and popliteus tendon) contribute to prevention of increased varus joint line opening with the ACL graft present (Figure 4A). Our data also show that abnormal joint laxity can result even with an appropriately performed ACL reconstruction if other structures, in this case the PLS, are not repaired or reconstructed as well. It is recognized that only one sequential cutting sequence was utilized for this study (based upon our pilot study data) and variation in cutting sequences may yield further information on the importance of each individual anatomic structure on preventing increases in varus opening and the resultant increase in force seen in the ACL reconstructed knee. It appears from our data that the main structure which prevents an increase in force on the ACL graft when injured is the FCL. Further studies are planned with variations in the sequential sectioning to determine if significance increases in force on the ACL graft are seen with isolated popliteus complex injuries.
The force in an ACL graft increased during varus and coupled varus - internal rotation moments in knees with grade III PLS deficiency compared to the force in the same knee with the ACL reconstruction and PLS intact. It is recommended that strong consideration be given to repairing or reconstructing grade III PLS injuries, especially with evidence of varus instability, at the time of ACL reconstruction to decrease the chance of postreconstruction ACL graft failure.
This paper was supported by a grant from the Minnesota Medical Foundation, grant
#MMF/SMF-1034-97. Appreciation is expressed to Arthex Inc., Naples, Florida, for donation of the ACL reconstruction equipment and interference screws for this study.
Address correspondence and reprint requests to:
Robert F. LaPrade, M.D., Department of Orthopaedic Surgery, University of Minnesota, Box 492 UMHC, 420 Delaware St. S.E., Minneapolis, MN 55455.