In patients with chronic posterolateral patellar instability who have had more than two patellar dislocations, the medial patellofemoral ligament can be reconstructed using small incisions with in situ dissection. The strength of the reconstruction should be greater than the strength of the original ligament to counteract the factors that predispose to patellar instability. Although this technique has evolved considerably over the past two decades, the surgical technique still needs to be improved and the related issues still need further research.
Maintaining proper ligament tension after reconstruction and accurate graft placement are key issues in medial patellofemoral ligament reconstruction. In some cases, medial patellofemoral ligament reconstruction needs to be combined with other surgical techniques to address problems such as talus dysplasia, malalignment, and high patella.
The surgical treatment of chronic lateral patellar instability is divided between bony and soft tissue procedures, the former including anterior tibial tuberosity movement proximal or distal and glide reconstruction, and the latter mainly medial patellofemoral ligament reconstruction and medial support band shortening. Several anatomic biomechanical studies have demonstrated that the main role of the medial patellofemoral ligament is to limit the lateral dislocation of the patella during 0-30 degrees of knee flexion.
It has also been shown that medial patellofemoral ligament deficiency is the most important risk factor for chronic posterolateral patellar instability. Therefore, theoretically, the main way to treat chronic lateral patellar instability is to reconstruct the medial patellofemoral ligament. Since the rise of medial patellofemoral ligament reconstruction in the 1990s, it has now become the treatment of choice for cases with at least two previous patellar dislocations.
Although the treatment of patellar instability has progressed considerably in the last two decades, further improvements in surgical technique are still needed. In addition, the high complication rate (26%) after medial patellofemoral ligament reconstruction needs to be of great concern to physicians. Following the guidelines in this article will help improve the outcome after medial patellofemoral ligament reconstruction.
Medial patellofemoral ligament anatomic reconstruction vs. non-anatomic reconstruction
Placement of the insertion is critical to the recovery of function after ligament reconstruction, and this is also important for medial patellofemoral ligament reconstruction. However, there is a lack of research on the optimal placement of inserts, and the need for anatomic reconstruction of the medial patellofemoral ligament is controversial. The clinical significance of non-anatomic femoral tunnels in medial patellofemoral ligament reconstruction is also controversial.
Reconstruction of normal femoral ligament attachment
The need for reconstruction of the normal femoral ligament attachment has been confirmed by several studies, and Elias et al. performed a biomechanical experiment using a computerized knee model to investigate the effect of reconstruction on patellofemoral interaction forces and pressure distribution. Bollier et al. found that the anterior position of the femoral tunnel may result in overloading of the medial patellar cartilage, leading to osteoarthritis and pain. Camp et al. found that non-anatomic femoral attachment abnormalities identified by imaging were a risk factor for surgical failure. The investigators found an 80% risk of dislocation within four years of surgery in patients with abnormal attachment of the graft to the femoral end.
Thaunat and Erasmus concluded that when the femoral tunnel is too close to the proximal end of the limb, limb extension tends to trigger graft laxity, and limb flexion causes graft tension, which is characterized by anterior knee pain and limited limb flexion. Although the graft used for medial patellofemoral ligament reconstruction is stronger than the normal ligament, excessive graft tension due to knee flexion can lead to fixation failure and even recurrent patellar dislocation.
Conversely, if the femoral tunnel is too close to the distal end of the limb, the graft is tense in extension and relaxed in flexion. Smirk recommends avoiding the trochanteric node as a starting point for graft fixation because fixation there can result in too much graft tension in flexion and too much graft laxity in extension. Based on the above clinical and experimental studies, researchers concluded that the alignment of the femoral tunnel should be as similar as possible to the original anatomy.
Mrlegari et al. conducted a biomechanical laboratory study using cadaveric knee joints and found that the use of abnormal anatomic attachment points allowed the graft to conform to the normal anatomic trajectory and maintained normal pressure on the patellofemoral joint compared to normal anatomic attachment points. correlation. This may be due to the fact that the difference between the actual fixation point and the anatomical fixation point was small and therefore did not cause a significant clinical difference, or it may be due to the short follow-up period and the subsequent development of osteoarthritis in the patellofemoral joint in some patients.
Ostermeier et al. compared static medial patellofemoral ligament reconstruction with dynamic non-anatomic reconstruction (using the medial collateral ligament as a pulley) and found that dynamic reconstruction had significantly less impact on the patella and less graft tension than static reconstruction. deie et al. found that dynamic medial patellofemoral ligament non-anatomic reconstruction was clinically more effective and prevented the recurrence of dislocation.
Clinical significance of the patellar attachment site
There are relatively few studies on the anatomical reconstruction of the medial patellofemoral ligament at the patellar attachment site, and Kang et al. concluded that the patellar attachment should be divided into two bundles: the inferior and superior bundles. Farr et al. suggested using the semitendinosus as a graft, thus mimicking the anatomic attachment of the medial patellofemoral ligament in the larger patellar area.
MOchizuki found that the semitendinosus was not a perfect replacement for the medial patellofemoral ligament because its proximal fiber bundle was linked to the medial femoral muscle and its distal fibers were attached to the middle patellar ligament rather than directly to the patella. The contraction of the medial femoral muscle increases the tension of the medial patellofemoral collateral ligament to maintain patellar stability during knee flexion. The reconstruction of the medial patellofemoral ligament at the patella in this case is a non-anatomic reconstruction. It would be inappropriate to drill a hole in the patella to create a ligamentous attachment point that does not exist in order to restore an anatomic reconstruction.
Choice of attachment point
Servien et al. concluded that anatomic reconstruction of the medial patellofemoral ligament is more difficult. The researchers analyzed femoral tunnels in 29 patients and found that only 20 (69%) patients were well positioned by plain film analysis.
In a 2007 laboratory study, Schottle et al. first identified the anatomical attachment points of the femur on lateral radiographs: 1 mm anterior to the tangent to the posterior femoral cortex (reference line), 2.5 mm distal to the perpendicular line traced through the initial part of the medial femoral condyle, and proximal to the perpendicular line traced through the most posterior part of the Blumensaat line. In a study by Redfern et al.
A lateral plain radiograph of the knee. The blue dot is the anatomic attachment point of the graft at the femur as indicated by Schüttle et al. and the red line is used as a reference to locate the femoral attachment point; the red dot is the attachment point of the graft at the patella as indicated by Barnett et al. and the yellow line is used as a reference to locate the patellar attachment point. The blue dot is the point of anatomic attachment of the graft to the femur as indicated by Barnett et al. If the anterior-posterior distance between the medial femoral condyles is considered to be 100% (shown by the yellow arrow), then the attachment point of the graft is located at a distance of 40% posteriorly, 50% distally, and 60% anteriorly. C Abnormal femoral tunnel position Lateral view of the knee after medial patellofemoral ligament reconstruction. This patient developed severe postoperative anterior knee pain and medial patellar instability.
Although these imaging landmarks are highly reproducible, the curve of the posterior lateral femoral cortex varies depending on the weight-bearing situation. Stephen therefore concluded that the posterior femoral cortical curve does not consistently predict the exact site of graft attachment to the femur. To avoid these limitations, Stephen et al. correlated the graft attachment point with the dimensions of the medial femoral condyle using the normalized dimension of joint morphology: if the anterior-posterior distance between the medial femoral condyle is considered to be 100%, then the graft attachment point is located 40% posteriorly, 50% distally, and 60% anteriorly.
These imaging landmarks can be used as an aid to accurately locate the attachment point intraoperatively and can be used as a tool to assess the patient’s persistent pain and dysfunction postoperatively. However, intraoperative localization of the graft attachment point in the C-arm is a rough estimate and cannot be used as the sole criterion for the location of the graft attachment point. Its final location is based on an accurate understanding of the relevant anatomy. Again a larger surgical incision is made in order to prepare to grasp its anatomy. Only then can the exact anatomical attachment site of the graft be accurately grasped and such surgery performed with complete certainty.
Bbarnett et al. proposed the use of anatomic imaging landmarks for the anatomic localization of the graft in the patella. The graft attachment site occupies 33% of the total length of the patella and is located at the intersection of the proximal 1/3 and distal 2/3 of the longitudinal axis of the patella.
It is also worth noting the individual variability of the medial patellofemoral collateral ligament at the beginning of the femur and the attachment point of the patella. Siebold suggests that the graft can be positioned at the femoral-patellar attachment point using an arthroscopic extra-articular approach from the knee joint. This removes the influence of individual variables on the graft attachment site and theoretically prevents postoperative complications.
Ideal non-isometric medial patellofemoral collateral ligament
The concept of isometric is derived from the literature related to anterior cruciate ligament surgery in the 1960s. It was based on the idea of no significant stretching of the ligament during normal range of motion of the knee, i.e., the length remained essentially constant, thus avoiding fixation failure due to overstretching, but clinical practice proved this concept to be incorrect. At present, ACL reconstruction no longer seeks to be isometric but to restore its anatomical structure and normal function. If the experience gained from ACL reconstruction is applied to medial patellofemoral ligament reconstruction, the goal should be to restore the anatomical structure and normal function rather than absolute isometric. Therefore, knowledge of the relevant anatomy and normal function of the medial patellofemoral ligament is essential.
Most authors believe that the medial patellofemoral ligament is not isometric within the range of motion of the knee, and Smirk et al. performed an anatomical study of 25 embalmed cadaveric specimens to assess isotonicity of the knee joint from 0° to 120° of flexion. The researchers defined ligament changes of less than 5 mm as isometric and showed that the medial patellofemoral ligament remained isometric only from 0° to 70°. In another study evaluating isometricity, Steensen et al. found a range of patellofemoral ligament length variation of 5.4 mm for knee motion from 0° to 90° of flexion and 7.2 mm for variation from 0° to 120°.
Victor found that the medial patellofemoral ligament is non-isometric: the proximal bundle is tense when the knee is in extension and the distal bundle is tense when the knee is flexed at 30°. In contrast, Stephen found that the normal medial patellofemoral ligament was isometric from 0° to 110° of the knee. The variability in the results of these studies may be due to their experimental approach, as all experiments were performed on normal knee cadaver specimens.
Earlier studies of the anterior cruciate ligament of the knee found that a small change in its origin and termination point could constitute a large change in the ligament length during knee motion. This also applies to the medial patellofemoral ligament; Steensen et al. found that altering the patellofemoral ligament at the femoral origin affected its length during knee motion. Changing the patellofemoral ligament at the point of attachment had less effect on the length change.
In a clinical study, Tateishi et al. demonstrated that changing the position of the femoral origin during patellofemoral ligament reconstruction affected the change in graft length. The researchers further demonstrated that the location of the center of the femoral tunnel determines the pattern of graft length change. If the location of the femoral attachment point has a strong influence on the pattern of graft length change and if the pattern of graft length change is closely related to prognosis, then the location of the femoral tunnel is critical to achieving a good postoperative outcome.
Erasmus believes that patellar height plays a very important role in the isometricity of the medial patellofemoral ligament, and the higher the patellar position the more pronounced the nonisotropy. In patients with significant high patella, the tibial tuberosity should be relocated distally to the limb. In this way, the degree of nonisometric patellofemoral medial ligament can be reduced and the tension of the graft can be precisely controlled. tateishi demonstrated in a clinical study that the isometricity of the graft correlates with the degree of high patella.
Triantafillopoulos et al. used the semitendinosus muscle as the graft and performed dynamic femoral fixation using two different glides – the medial interval, and the posterior third of the medial collateral ligament. The average change in graft length from 0° flexion to 90° was 4 mm with the medial interosseous muscle as the glide and 1 mm with the medial patellofemoral ligament as the glide. although the medial interosseous muscle was less isometric as the glide, it was relatively more stable and helped maintain patellar stability.
Parker et al. conducted a comparative study of patellofemoral dynamics after isometric reconstruction versus anatomical reconstruction on cadaveric specimens and found that the specimens in the isometric reconstruction group did not return to normal patellofemoral dynamics performance at any flexion angle. In contrast, the anatomically reconstructed knee showed the same patellofemoral trajectory as normal during motion from 0° to 28° of flexion. Neither approach maintained normal patellofemoral trajectory in greater knee flexion. However, non-isometric medial patellofemoral ligament reconstruction is more effective than isometric reconstruction in maintaining patellofemoral dynamics.
The non-isometric nature of the graft should be similar to the normal medial patellofemoral ligament. To achieve this, Thaunat and Rrasmus suggested that the graft should be isometric from 0° to 30° of flexion of the knee. This is both the so-called ideal non-isometric. Since the tendency to dislocate the patella is greatest from 0° to 30° of flexion, the probability of patellar dislocation is reduced. The effect of graft relaxation decreases with increasing knee flexion.
CT 3D reconstructions of the knee at 0°, 30°, 60°90°, and 120° of flexion are shown from A-E, respectively. The red dot is the anatomical attachment point of the medial patellofemoral ligament at the femur as indicated by Stephen. The red line represents the normal medial patellofemoral ligament and the blue line represents the position of the medial patellofemoral ligament after graft reconstruction.B The blue arrow on the image indicates the abnormal anterior femoral tunnel position in a patient with severe knee pain and medial patellar instability after reconstruction. The length of the graft is the distance between the femoral attachment point and the patellar attachment point. According to Smirk and Morris’ definition of isometric (less than 5 mm difference in length), the normal medial patellofemoral ligament is isometric from 0° to 30° of knee flexion, as shown in the lower right inset. However, as the knee flexion increases, the graft gradually relaxes. The longest graft was found at 30° of knee flexion. Therefore, the knee should be fixed in 30° of flexion using the anatomical attachment point as the starting point.
Fixation of the graft according to the blue line trajectory maintains isotonicity throughout the range of motion of the knee, but is less clinically effective. Fixation of the graft in this position may counteract the instability by providing greater local strength, but excessive local pressure may also aggravate medial patellar cartilage dysplasia and ultimately worsen the condition. The above scenario can be used to explain patients with postoperative anterior knee pain. Therefore, in cases with lateral patellar stability, the graft should be kept isotonic only in 0° to 30° of knee flexion.
Importance of graft tension
In addition to the femoral tunnel, which plays an important role in the outcome of patellofemoral ligament reconstruction, graft tension is another important factor. Thaunat et al. reported two cases of limited knee motion due to excessive graft tension, one patient was unable to straighten the knee and the other was unable to flex it. If the graft tension is excessive it can cause medial patellar subluxation during knee fixation.
Considering the high chance of concomitant medial articular surface injury, excessive weight bearing on the medial patellofemoral articular surface is avoided during reconstruction. Moreover, over-tightening of the graft leads to excessive tension, which in turn leads to fixation failure. Especially in patients with concurrent lateral ligament release, excessive graft tension can lead to medial patellar subluxation of medical origin. However, too little graft tension may result in too little medial restriction leading to lateral recurrent instability.
In the absence of lateral forces acting on the patella, the normal medial patellofemoral ligament is free of tension, so a conceptual analysis of the so-called graft tension is inappropriate. When the graft is tense, it restricts knee motion. The purpose of medial patellofemoral ligament reconstruction is to replace the torn ligament with a stronger graft while maintaining tension similar to that of a normal ligament. In cases where the anatomy of the glide is normal, it is simpler to apply appropriate tension to the graft. In cases with more severe glide dysplasia, the lack of anatomic landmarks to locate the center of the patella intraoperatively makes it more difficult to grasp the tension of the graft, and there is a tendency to over-tension the graft.
In a controlled biomechanical study using knee specimens, Beck et al. found that low graft tension (2 N) stabilized the patella without increasing pressure on the medial patellofemoral unit. Excessive tension limited lateral patellar movement and increased medial patellofemoral pressure. One way to obtain the optimal tension of the graft from a practical point of view is to use the contralateral patella as a reference. Intraoperatively, the degree of patellar displacement is compared between the two sides. In patients with symptoms on both sides, the lateral transverse displacement of the patella is considered normal at less than half the width of the patella.
Another controversial and important issue is the degree of knee flexion during tensioning of the graft. Thaunat recommends tightening the graft in full knee flexion, and to achieve this they stretch the patella proximally with a bone hook, which puts more tension on the patellar ligament than the graft during quadriceps contraction.
Farr tightened the graft at 30° of knee flexion, resulting in ligamentous laxity in knee flexion and ligamentous tension in full knee extension; Yoo believes that the optimal angle of graft fixation is 30°, while LeGrand recommends fixation at 45° to 60° of knee flexion to ensure that the patellar glide atella would engage in the trochle. The focus is on ensuring that there is no restriction of knee motion after graft fixation and that there is no adverse effect of lateral displacement of the patella from 0° to 30° of knee flexion. The graft should only be tightened during lateral displacement of the patella.
Influence of patellar instability factors on clinical outcome
Habitual lateral patellar dislocation is a major complication after medial patellofemoral ligament reconstruction. So far it has not been possible to confirm that graft fixation failure is due to graft tear, laxity or other instability factors. Lateral knee instability is multifactorial, with failure of medial patellofemoral ligament fixation in addition to poorly developed and misaligned glides (e.g., tibial tuberosity to glide depression distance greater than 20 mm, patellar tilt angle greater than 20°) and high patella. In the presence of these risk factors, medial patellofemoral ligament reconstruction may not be sufficient to achieve a good clinical outcome, so these risk factors need to be eliminated when performing medial patellofemoral ligament reconstruction.
Wagner et al. found that severe dysplasia of the talus was strongly associated with poor clinical outcomes. Wagner suggested that glide plastic reconstruction should be considered in patients with severe glide dysplasia. However, their conclusion was based on one case report (Level of Evidence IV), while in another case report Steinner et al. concluded that there was no significant correlation between glide dysplasia and medial patellofemoral ligament reconstruction.
Other risk factors for patellar instability (medial soft tissue defects, high patella, etc.) are more important in medial patellofemoral ligament reconstruction than patellar talus dysplasia. Correction of these risk factors can reduce the danger of glider dysplasia and increase stability. Because of the high complication rate of glide reconstruction, it is only indicated for patients with severe glide dysplasia and other surgical procedures that do not ensure patellofemoral stability. Therefore, glide reconstruction should only be used as a remedial surgical measure.
Wagner found that approximately 58% of patients had a coexisting high patella, with 70% of these patients having a patellar index between 1.2 and 1.3, which may be the main reason for their failure to affect clinical outcomes. At present, the patellar index of the tibial tuberosity that requires distal displacement of the high patella remains unknown.
Wagner found that an abnormal tibial tuberosity to glide depression distance could lead to poor clinical outcomes and therefore recommended changing the tibial tuberosity position to restore a normal tibial tuberosity-glide depression distance (approximately 12 mm). The ultimate goal is to reduce graft loading.
Medial patellofemoral ligament reconstruction alone is indicated by recurrent lateral patellar dislocation with a tibial tuberosity to glide depression distance of 20 mm, a positive apprehension test after 30° of knee flexion, a patellar Caton-Deschamps index of less than 1.2, and Grade A glide dysplasia.
Conclusion
The current approach to recurrent lateral patellar dislocation with at least one recurrence is to reconstruct the medial patellofemoral ligament ligament through a small incision with a stronger graft to resist the uncorrected risk factors that predispose to patellar instability. Medial patellofemoral ligament reconstruction is complex and requires extensive experience to avoid complications arising from abnormal femoral tunnel position or inadequate graft tension. Proper attachment of the graft to the femur and proper tension are determinants of the postoperative outcome.
Furthermore, in some cases, medial patellofemoral ligament reconstruction needs to be combined with other surgical steps to eliminate other risk factors for patellar instability, such as glide dysplasia, abnormal alignment, and high patella. Understanding the anatomy and function of the medial patellofemoral ligament is essential to ensure good long-term postoperative results.