Biomechanical comparison of single- and double-bundle medial patellofemoral ligament reconstruction

Recurrent patellar dislocation is common clinically, primarily in adolescents [1, 2]. This instability seriously affects patients’ functional movement in daily life. It is more common in patients whose medial support structure poorly heals in the knee joint after traumatic dislocation or in patients who have abnormal anatomical structures that can easily be dislocated [3].

From the anatomical and biomechanical studies of MPFL, it is found that MPFL is the primary passive restraint maintaining static stability that resists the dislocation of the patella and controls patellar tracking, providing approximately 53–60% of restraining force [2, 4]. Patients with patellar dislocation often have torn MPFL, as revealed by MRI [5]. Thus, MPFL reconstruction has great clinical implications for the treatment of patellar dislocation [6].

Conservative treatment for the recurrent patellar dislocation is less effective, and many surgical treatments have been advocated at present [7]. There have been some disputes about graft tension, tunnel location, graft fixation angles, and so on in MPFL reconstruction [8]. For example, over-tight MPFL grafts can lead to increased graft tension, limited range of motion in the knee joint, graft rupture, and iatrogenic patellar dislocation [9]. The key to successful reconstruction is to restore patellar original anatomical morphology. Since the locations of structures attaching to the medial femoral condyle vary, the relationship between MPFL near the patellar side and vastus medialis obliquus (VMO) is complex. Therefore, it causes certain controversies and confusions in MPFL anatomical origins, insertions, and the relationship between MPFL and the oblique beam tendon of VMO. These confusions will cause difficulties in MPFL reconstruction, leading to uncertain postoperative effects. The double-bundle MPFL reconstruction was developed recently to simulate anatomical structures by using two patellar tunnels [10]. However, the biomechanical properties of double-bundle MPFL reconstruction remains poorly understood. Consequently, in this study we attempted to evaluate the biomechanics of isometric single-bundle MPFL reconstruction and anatomic double-bundle MPFL reconstruction.

Six fresh frozen adult cadaveric knee specimens were obtained for this study. These specimens were comprised of 15-cm distal femurs, 15-cm proximal tibias, and their surrounding soft tissue structures. The average ages of the subjects were 57 ± 8 years old (range 48–72 years old). Cadaveric specimens had no prior surgery to the knee joint area. Deformities and apparent kinematic alterations when extended and flexed were not observed in these specimens. All operations were performed by one experienced orthopedist. A midline incision of the knee joint was performed, and the skin and subcutaneous fat were resected, retaining the ligament structures, joint capsules, and distal quadriceps. In order to remove the tibial and femoral bone marrow tissue, electric drill reaming was performed. Thus, a 9-mm diameter iron screw (the length of both the distal femur and proximal tibia was 20 cm) was inserted into the pulp cavity after drying. Then, polymethyl methacrylate (PMMA) was used for the fixation of the screw (Fig. 1).

A No. 2 Ethibond suture was used to suture the nylon string on the distal end of the distal quadriceps (simulating the contraction of the distal quadriceps through the suspension of the pulley). Two parallel holes were drilled from the lateral margin of the patella to the geometric center, and a steel wire was lead through the parallel hole. In our study, a biomechanical material dynamic mechanical testing system (Bose 520, Australia) was used for data collection. A nylon rope through the pulley was used to suspend 170 N of the axial load on the quadriceps (simulating quadriceps contraction). The steel wire through the patella was linked with the hook of the machine arm equipped with the system (providing the lateral shift of the patella).

The test protocol is based on previous studies [11, 12]. First, each specimen was fixed at 0° to test the force needed when the patella was laterally shifted 10 mm at the speed of 0.5 mm/s, and the test was repeated three times. This test was repeated when knee flexion was at 0°, 15°, 30°, 45°, 60°, and 90°. All six specimens were tested in the four states, including MPFL intact, MPFL torn, single-bundle MPFL reconstruction, and double-bundle MPFL reconstruction.

In normal knees, the force required to shift the patella 10 mm laterally at the 0° of the flexion is 74.3 ± 10.7 N (Table 1). Most of the time, this force increased as the flexion angle increased (P < 0.05), except for the 45° flexion angle.

When comparing the force required in different groups at the same flexion angle (Fig. 3), a similar trend was found in the first four angles. Furthermore, it was found that the MPFL torn group requires a significantly smaller force to permit a 10-mm patella lateral shift at flexions of 0°, 15°, 30°, and 45°, compared to the other groups. The force required in the MPFL torn group reduced from 12.8 to 38.8% at these four angles, compared to the other groups. At a flexion of 15°, the double-bundle reconstruction group required a greater force, compared to the single-bundle reconstruction group (P < 0.05). Interestingly, no statistical difference was found at flexions of 60° and 90°. During the entire flexion, no significant difference was found between the intact group and deficient groups.

In our study, we found that both single-bundle and double-bundle MPFL reconstruction can restore the stability of the patella. In addition, these results imply that double-bundle reconstruction greatly enhances the capacity to resist the patella’s dislocation before the patella entering the femoral trochlea. This indicates that double-bundle reconstruction has angular synergy effects, simulating the MPFL wide footprint in the patella at a smaller flexion angle [13]. Regarding the precision, in our study in an intact knee joint when knee flexion is at 0° and 15°, a 10-mm patellar shift requires 74.3 ± 10.7 N and 74.7 ± 10.6 N of lateral stress, respectively; and it requires 101.5 ± 13.8 N when knee flexion is at 90°. This is close to the results of other cadaveric patellar biomechanical models, which support the validity of the current experimental setup [2, 14, 15].

Results in current study show that both single- and double-bundle MPFL reconstruction can help restore the biomechanics of the patellofemoral joint. The forces required for 10-mm lateral shift is greater in the double-bundle MPFL reconstruction than that in the single-bundle MPFL reconstruction when knee flexion was at 15°. This is possibly due to the synergism of the bundles at that degree of knee flexion, as they form an angle of 12° to 15° in the double-bundle MPFL reconstruction. Another possible explanation would be that the extensive footprint of MPFL is imitated in the double-bundle reconstruction, which requires larger resistance to the patellar shift before the patella gliding into the femoral trochlea.

Previous study has shown that the patella’s resistance to the outer edge stress is at the lowest level during knee flexion at 20° to 30° [16]. In fact, from the literature such angles are clinically common angles for patellar dislocation [17‐19]. Based on these biomechanical results, we can speculate that double-bundle MPFL reconstruction has advantages in preventing patellar lateral recurrent dislocation after operation. In some clinical studies, it was shown that the double-bundle reconstruction achieved far better clinical outcomes in the long-term follow-up compared to single-bundle reconstruction [20].

No statistical differences were found in force required among groups at flexion of 60° and 90°. It is reasonable since the medial facet of lateral femur condyle resisted the lateral shift after entering femoral trochlea at higher flexion angle, despite the torn MPFL. Meanwhile, the force required was increasing in all groups because of the resistance from femur condyle.

Kang et al. researched the anatomy of MPFL and put forward the concept of bi-functional bundles [10]. It has been reported that double-bundle anatomical reconstruction is consistent with the patellofemoral ligament in terms of anatomical characteristics [21]. The application of the double-bundle anatomical reconstruction and reconstruction of the straight bundle below the patellofemoral ligament can help restore a low level of stability. In the case of knee flexion that occurs earlier before the patella gliding into the femoral trochlea, the contraction of the upper tilt bundle and medial femoral tilt muscle bundle can stabilize the patella [13]. The coupling of the upper tilt bundle reconstruction and medial femoral oblique muscle can strengthen the dynamic and static stability of the knee joint.

The limitations of this study include the small number of samples, samples from older subjects who had different geometric structures, biomechanical properties in the patella compared to those from youth. This may have effects on the experiment. Even though no pathological characteristics or abnormal patella rotation were observed in the included samples, VMO and femoral trochlear dysplasia may still disrupt the imitation of patellar dislocation.

We found that both single-bundle and double-bundle MPFL reconstruction can restore the stability of patella and that the force required to shift the patella 10 mm laterally in the double-bundle MPFL reconstruction is higher than that in the single-bundle reconstruction at 15°. Furthermore, double-bundle reconstruction has angular synergy effects that simulate the MPFL wide footprint in the patella. This enables it to have greater capacity to resist patellar lateral dislocation before the patella entering the femoral trochlea at a smaller flexion angle. Further studies should be done to verify its clinical efficacy.