It has been reported that injuries involving the posterior cruciate ligament (isolated and combined) account for 5 to 20 per cent of all injuries of the ligaments of the knee3,7,10,11,20,28. The objective and subjective outcomes of non-operative treatment of an isolated rupture of the posterior cruciate ligament have been reported to be good or excellent8,13,27,30,31. Unfortunately, combined injuries involving the posterior cruciate ligament are not uncommon5,17,25.
Excessive posterior laxity at 90 degrees of flexion (3+ or more) often is found on physical examination of a knee with a ruptured posterior cruciate ligament, and the clinical course for such knees is not always benign6,9,18,27. Replacement of a ruptured posterior cruciate ligament with a graft has become increasingly popular for patients who have excessive posterior laxity2,6,12,23,24,28.
The primary biomechanical rationale for insertion of a graft is to restore normal posterior laxity and, when necessary, to repair torn secondary stabilizing structures. However, reports of residual excessive posterior laxity after reconstruction of the posterior cruciate ligament are becoming more frequent2,19,24. Gradual stretching of the graft is one explanation for increased laxity; however improper placement of the graft and failure to pre-tension the graft correctly are other possible explanations. The biomechanics of posterior cruciate reconstructions has received relatively little attention1,4,14,16,29.
There were several objectives of this study. The first was to compare the relative displacements of a trial isometer wire at four designated points on the femur and then to compare the displacement at one of these points with the displacement of the distal end of a graft that had its proximal end fixed at that point. Our second objective was to measure the force generated in the femoral end of the graft (that is, within the intra-articular portion of the graft) as a function of the amount of pre-tension applied directly to the bone block of the free end of the graft in the tibial tunnel. Third, we sought to determine the amount of pre-tensioning of the graft that is needed to restore normal anterior-posterior laxity at 90 degrees of flexion (the laxity-matched pre-tension). Finally, we compared the anterior-posterior laxity values at selected angles of knee flexion before and after section of the posterior cruciate ligament and after insertion of the graft at the laxity-matched pre-tension.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Funds were received in total or partial support of the research or clinical study presented in this article. The funding sources were National Institutes of Health Grant RO1 AR40330 and the Dorothy and Leonard Straus Fund.
†Biomechanics Research Section, Department of Orthopaedic Surgery, Rehabilitation Center, University of California at Los Angeles, 1000 Veteran Avenue, Los Angeles, California 90024-1795. E-mail address for Dr. Markolf: kmarkolf@ortho.medsch.ucla.edu.
‡Department of Orthopaedic Surgery, Texas Tech University Health Sciences Center, Lubbock, Texas 79301.
Twelve fresh-frozen normal, stable knees were obtained from cadavera of individuals who had been forty-nine to seventy-nine years old at the time of death. The ends of the femur and tibia were potted in cylindrical molds of polymethylmethacrylate so that they could be gripped in the fixtures of the testing apparatus. The skin and soft tissues around the knee were left intact during testing.
Testing Apparatus
An anterior-posterior force (±200 newtons) was applied manually to an undercarriage bar attached to the tibial fixture of the apparatus, and tibial displacement was recorded with use of a spring-loaded transducer22. The tibia was not allowed to rotate. Neutral rotation at each angle of flexion tested was defined as midway between the tibial rotations produced by five newton-meters of internal and external tibial torque. The force applied to the tibia, the resultant force in the posterior cruciate ligament or the graft, and the tibial displacement were recorded. Knee laxity was defined as the resultant anterior-posterior displacement of the tibia, relative to the femur, when a 200-newton anterior-posterior force was applied to the tibia.
In order to define 0 degrees of flexion, the potted end of the femur was clamped such that the femoral shaft was horizontal, with the patella facing down, and a 2.5-newton-meter extension moment was applied to the tibia. Isometer testing also was performed with the specimen in this inverted orientation to eliminate the posterior subluxation of the tibia that would have occurred if, after section of the posterior cruciate ligament, the tibia was extending upward against gravity during testing. A specially designed isometer containing a displacement transducer was used to record the displacement, relative to the tibia, of a trial wire connected to a femoral point (or to the tibial end of the graft). To begin the test, the electrical output of the instrumented isometer was set to zero when the knee was in 0 degrees of flexion. The knee was flexed slowly to 120 degrees while the displacement of the isometer wire and the angle of flexion were recorded. Isometer readings were recorded as positive when the wire retracted into the joint, which corresponded to lengthening of the graft.
Testing Procedures
First, the intact knee was subjected to anterior-posterior testing, with the tibia in neutral rotation, at 0, 10, 30, 45, 70, and 90 degrees of flexion. A cylindrical block of subchondral femoral bone containing the femoral origin of the posterior cruciate ligament was mechanically isolated and was attached to a miniature load-cell mounted on the femur (Fig. 1), which recorded the resultant force in the ligament32. Anterior-posterior testing was repeated, with the tibia locked in neutral rotation, after installation of the load-cell. The posterior cruciate ligament was removed, and anterior-posterior testing was performed again.
The knee was mounted for isometer testing with the trial wire attached to the femur at one of four designated points. The femoral origin of the posterior cruciate ligament could be considered as approximately elliptical in shape, with the major axis of the ellipse oriented approximately in the anterior-posterior direction and the minor axis, in the proximal-distal direction (Fig. 1). The location that we designated as the central point was at the center of the ellipse. The axis of the cylindrical cap of bone and of the load-cell also passed through this point. The proximal and distal points were on the minor axis of the ellipse at the proximal and distal margins, respectively, of the femoral origin of the ligament. The anterior point lay on the major axis of the ellipse at the anterior margin of the femoral origin of the ligament, near the femoral trochlea. A point on the posterior margin of the femoral origin of the ligament was not tested because this location is never considered for attachment of a graft in the clinical setting. A small screw with an attachment hook was inserted at each point for connection of the isometer trial wire. Slight tension of the trial wire was maintained by a spring within the plunger of the isometer.
After the measurements made with the isometer trial wire had been recorded, an eleven-millimeter hole for a tibial tunnel was created with use of a drill and a drill-guide. The posterior (target) point of the drill-guide was placed at the center of the tibial insertion of the posterior cruciate ligament, approximately one centimeter distal to the tibial plateau. The included angle between the drill and the tibial plateau was approximately 50 degrees.
The grafts were prepared from whole bone-patellar ligament-bone specimens obtained from a tissue bank; each specimen was split longitudinally into two halves, yielding two grafts. A ten-millimeter-wide section of tissue was prepared from the central portion of each half. The bone at the patellar end was reinforced with wire loops and was potted in a small cylinder of polymethylmethacrylate to make a cap on the graft identical in size to the isolated cap of bone that contained the posterior cruciate ligament; the femoral load-cell was attached to this cap. The bone block at the tibial end of the graft was interwoven with four strands of high-strength Dacron line and was passed over and around the posterior edge of the tibial plateau and into the tibial tunnel. The bone block was free in the tunnel because of a 0.5-millimeter difference between its diameter and that of the hole to the tunnel. The Dacron lines exited the tunnel distally and passed through a split clamp mounted to the tibia. The lines were clamped when the desired level of pre-tension of the graft had been achieved. The graft was oriented such that its wide dimension was aligned with the wide dimension of the intact posterior cruciate ligament. The acrylic surface (at which the graft fibers exited the acrylic cap on the graft) was flush with the surface of the intercondylar notch. This placement simulated as closely as possible the geometric configuration of a graft located at the center of the femoral attachment of the intact posterior cruciate ligament.
With the tibia free to rotate, a calibrated spring-scale was used to apply tension to the Dacron lines with the knee flexed 90 degrees while force was recorded simultaneously by the femoral load-cell. Four levels of tension were applied to the graft: forty-five, ninety, 134, and 189 newtons. Three trials were performed with each level of tension, and the average of the load-cell readings was determined.
Next, the spring-scale was used to apply a trial level of pre-tension to the Dacron lines with the knee in 90 degrees of flexion while an anterior force (approximately twenty-two newtons) was applied to the tibia (as is done commonly at an operation). The Dacron lines were clamped to maintain the level of pre-tension of the graft during anterior-posterior testing with the tibia in neutral rotation. Through trial and error, we found a level of pre-tension that, with a 200-newton force applied to the tibia, produced the same anterior-posterior laxity (within one millimeter) at 90 degrees of flexion as that recorded after installation of the load-cell on the posterior cruciate ligament. This pre-tension was designated the laxity-matched pre-tension. Anterior-posterior testing with the tibia in neutral rotation was repeated after the insertion of the graft at the laxity-matched pre-tension.
Statistical Analysis
A two-way analysis-of-variance model with repeated measures was used to determine the significance of differences among the mean anterior-posterior laxities for the various conditions: the intact knee, after installation of the load-cell, after section of the posterior cruciate ligament, and after insertion of the graft at the laxity-matched pre-tension. Multiple pairwise comparisons between the means at specific angles of flexion of the knee were made with use of the Student-Newman-Keuls procedure. Similar analysis-of-variance models (with pairwise comparisons) were used to compare the isometer measurements associated with the four different points on the femur and to compare the pre-tensioning force ratios (ratio of the force recorded by the load-cell to the force applied to the distal end of the graft) at 30 and 90 degrees of flexion. The level of significance was p = 0.05.
Isometry
The isometry measurements that we recorded with use of a trial wire are in general agreement with those reported by Ogata and McCarthy26, who also found that use of an anterior point of attachment for a trial wire increased the relative displacement of that wire (that is, the wire was drawn into the joint) during flexion of the knee. We believe that the proximal, distal, and central points of attachment were associated with acceptable changes in the relative displacement of the wire over the range of motion that was tested.
Our results contrast with those of Grood et al.15, who concluded that errors in the placement of the femoral tunnel in the anterior-posterior direction would be more acceptable than those in the proximal-distal direction. However, those authors performed tests on knees with an intact posterior cruciate ligament and applied a 100-newton posterior force to the tibia throughout the range of motion, whereas we performed testing on posterior-cruciate-deficient knees (as would be done at an operation) and made a special effort to eliminate posterior loading of the tibia during isometer testing.
We think our findings are encouraging for a clinician contemplating the intraoperative use of an isometer, as the measurements that were recorded with the trial wire were an accurate indicator of the behavior of the graft in situ.
Pre-Tensioning of the Graft
We pre-tensioned the graft at 90 degrees of flexion because this is the position used clinically for the posterior drawer test and it is the position of maximum laxity for a knee with a ruptured posterior cruciate ligament.
We found that the resultant force measured at the femoral load-cell (which represented force in the intra-articular portion of the graft) was always less than the tension that was applied to the tibial end of the graft. This finding is probably due to frictional loss as the graft passes up and over the posterior tibial plateau and into the joint. We believe that the tibial end of the graft should be fixed and the femoral end should be pre-tensioned before fixation. This ensures that the pre-tension force is transmitted directly to the intra-articular portion of the graft.
The laxity-matched pre-tension varied a great deal. Although the mean value for the twelve knees was 43.0 newtons, pre-tension of eighty-nine newtons or more was needed in three knees. We believe that the surgeon should apply more than forty-three newtons to the femoral end of the graft with the knee flexed 90 degrees in order to ensure that laxity will be restored in all patients. The potential consequences of over-tensioning with respect to the forces in the graft is discussed in Part II of this study21.
Anterior-Posterior Laxity
The two to three-millimeter increase in laxity at 70 and 90 degrees after installation of the load-cell is best explained by considering the angle of pull of the intact ligament on the cap of bone during an applied posterior-drawer force. At angles of flexion near 90 degrees, the fibers of the ligament are pulling almost perpendicular to the axis of the load-cell. This subjects the cap of bone to the maximum amount of cantilever deflection and produces additional posterior laxity. At smaller angles of flexion, the collateral ligaments and the capsule help to resist posterior force, and the component of ligament-pull acting at right angles to the load-cell is diminished.
The coring cutter used to isolate the cap of bone produced a gap of about three millimeters, which is approximately equal to the mean increase in anterior-posterior laxity at 70 to 90 degrees after installation of the load-cell. When a posterior tibial force is applied at an angle of flexion near 90 degrees, theoretically the cap of bone could come into contact with the wall of the tunnel and reduce the recorded load. Although we could not be certain that a gap remained during all of the anterior-posterior tests, periodic visual checks of the gap demonstrated that the cap of bone clearly was deflected but did not touch the wall. Even if contact had occurred, it would not have explained the scatter in the pre-tension values, as they were relatively low.
After section of the posterior cruciate ligament, the mean increase in anterior-posterior laxity at 90 degrees of flexion was approximately 58 per cent greater than the increase at 30 degrees of flexion (the angle at which the Lachman test is normally performed to test for rupture of the anterior cruciate ligament) (Fig. 3). This finding reinforces the accepted clinical practice of performing the posterior drawer test at 90 degrees of flexion to establish the diagnosis of rupture of the posterior cruciate ligament.
NOTE: The authors thank Steve Jackson, for his assistance in the testing and the data analysis, and Stephen Liu, M.D., for his guidance in the initial phases of the work. The tissues used for the patellar ligament grafts were kindly provided by the Musculoskeletal Transplant Foundation.