Recurrent dislocation of the patella (RDP) is a relatively common clinical knee injury that severely affects the child’s athletic ability and daily life. Most scholars recommend surgical treatment. There are many surgical methods used to treat RDP, all of which aim to control patellar instability and restore the normal motion track of the patella. In recent years, as research related to the medial patellofemoral ligament (MPFL) has intensified, reconstruction of the MPFL has achieved good clinical results for the treatment of RDP in adults, with most literature reporting a high postoperative success rate and a re-dislocation rate of only 0G-10G. The goal of MPFL reconstruction is to restore the stable structure of the medial patella and limit its lateral dislocation. limiting its dislocation to the lateral side. A variety of surgical approaches to reconstruct the MPFL have been reported in the literature. The main difference between the different surgical techniques is the method of grafting and fixation used. For MPFL reconstructive surgery in adults, most authors opt for a medullary tract + squeeze screw fixation method on the femoral side because there is no need to consider damage to the epiphysis. The medial tuberosity is anatomically positioned next to the MPFL stop, which maximally mimics the femoral side of the MPFL stop and avoids damage to the distal femoral epiphysis due to the preparation of the femoral side of the medullary tract. In addition, the autologous tendon is more suitable than the allograft tendon in terms of type and body habitus. From January 2008 to July 2010, a total of 6 cases (7 knees) of RDP were treated with arthroscopic-assisted major retractor tendon transposition and reconstruction of MPFL, with good clinical results, which are reported below. Clinical data I. General information In this group, there were 6 cases of RDP (including 1 case of both knees), 2 males and 4 females; age ranged from 10 to 14 years old, with an average of 12 years old. The average time from injury to surgery was 13 months (1-25 months). Inclusion criteria: All patients complained of a history of two or more lateral dislocations of the patella, at least one of which was seen in the hospital. All reported preoperative instability, tenderness or stepping sensation in the knee joint, which was obvious when going up and down stairs, and were unable to engage in strenuous physical activities such as fast running and jumping. Clinical examination or examination under anesthesia could reproduce patellar dislocation, and three of them had positive patellar extrapolation and extrapolation fear test. Exclusion conditions: habitual patellar dislocation and fixed patellar dislocation were excluded, patellar dislocation due to other surgical treatment was excluded, patellar dislocation due to cerebral palsy or other neuromuscular pathologies was excluded, and other ligament injuries combined with surgical treatment were excluded. Pre-operative evaluation was performed with detailed history of patellar dislocation and patellar instability and Lysholm score, clinical examination with attention to recording knee mobility, measurement of lower limb force line and Q angle, special examination including patellar tilt test, patellar extrapolation and extrapolation fear test. All patients underwent preoperative lateral and axial patellar radiographs, CT and MRI of the knee joint to determine the lateral patellofemoral angle (LPFA), patellar tilt angle (PTA), and tibial tuberosity-femoral glide distance (TG-TMG). -(TT-TG) to understand the development of patellofemoral joint and femoral condyle. Third, the surgical method MPFL reconstructive surgery technique after the onset of anesthesia, are to first assess the mobility of the patella. Fithian et al. suggested that the diagnosis of patellar instability should be based on the perception of a “soft” or absent endpoint when the patella is dislocated laterally in the fully extended or flexed 30ºposition of the knee; and, in the flexed 30ºposition, the patella is “soft” or absent at the endpoint. In the 30º position, the patella can be pushed more than 10 mm laterally (from the central position). 1, Arthroscopy A standard arthroscopic approach was established for diagnostic arthroscopy. The patellar trajectory and patellar mobility are examined. The accompanying articular cartilage damage should be examined in detail and documented, and if unstable cartilage fragments are present, they need to be removed and other injuries treated symptomatically. 2.Lateral support band release For cases where the lateral patellar support band is too tight, the patella is obviously tilted outward, and internal pushing cannot reach the center of the femoral slide, arthroscopic electroknife is used to release the patella, and the release process pays attention to protecting the patellar stop of the lateral femoral muscle to prevent the patella from dislocating medially. The skin incision was made at the medial epicondyle of the femur, with a length of about 3 cm, and the stop of the greater trochanter was found at the node of the retractor muscle, which was separated proximally to the tendon ventral junction and cut off, paying attention to the protection of the femoral/N artery crossing the fissure of the retractor tendon. The free end of the graft was closed with a No. 2 Ethibond non-absorbable suture, and the end of the suture was used as a traction line. 4, MPFL patellar stop was selected by preparing the patellar tunnel by making a longitudinal incision at the medial edge of the patella with a length of approximately 3 cm, bluntly separating the subcutaneous tissue with the fingers through the incision and probing the superior pole of the patella and the medial edge of the patella. The mid-superior 1/3 of the medial edge of the patella was selected as the center of the patellar tunnel, and a 4.5-mm drill was used to create the bone tunnel. The tunnel exit can be located at the outer edge of the patella (tunnel across the patella) or on the anterior surface of the patella (“L” shaped tunnel, Figure 7). When creating the tunnel, an ACL guide can be used to place the guide pin and then drill the hole along the guide pin, which prevents the drill from accidentally entering the joint and damaging the articular cartilage of the patella. If the graft is large, the tunnel can be enlarged appropriately. It is important to note that when using the “L” shaped tunnel, the tunnel length should be greater than 2 cm to prevent the bone bridge of the tunnel from breaking. 5. Assess isometricity by passing the greater retractor tendon through the medial patellar soft tissue channel, attaching it to the patellar stop and tensioning it. Full range of knee flexion and extension was performed to check the length change of the graft during knee flexion and extension. The graft was required to be essentially isometric (5 mm change in length) during knee flexion and extension. The reconstructed MPFL should be able to play the role of “reins”, so that the patella enters the femoral carriage smoothly during early knee flexion, and ensure that the graft does not impact or rub against the femoral condyle. When fixing the patellar side of the graft, the knee should be kept in flexion at 60º~90ºand the patella should be placed in the center of the femoral glide tunnel, using isometric adjusters (Synthes, Paoli, Penn) to maintain an initial tension of approximately 5 Newtons on the graft. After the graft is passed through the patellar tunnel, one stitch is closed for temporary fixation. The patella is then checked again for mobility in the flexed 30º or extended position, at which point a very firm end point should be felt; the knee is then flexed and extended, and the range of knee flexion and extension should be unaffected; and, pushing the patella laterally in the flexed 30º position, the patella is displaced outward in a range of 7 to 9 mm. After confirming proper tension of the graft, the graft is reflexed and sutured to itself, or to soft tissue structures such as the anterior patellar periosteum and the lateral patellar support band. After fixation, the excess tendon is excised and the medial patellar support band and incision are closed. Arthroscopy is again used to examine whether the patellar trajectory has returned to normal during full range of knee flexion and extension. In particular, at the onset of knee flexion, it is important to ensure that the patella can enter the femoral glide without any impingement or obstruction. 7, Lateral semi-internal displacement of patellar tendon stop For cases with CT measurement of TT-TG greater than 20 mm and Q angle greater than 20 degrees, we used the method of distal rearrangement by placing the lateral semi-internal displacement of patellar tendon stop under the proximal tibia periosteum. 8. After MPFL reconstruction surgery postoperative management and rehabilitation planning MPFL reconstruction surgery, patients need to be fixed in a long-leg cast for 4 weeks (especially for patients who underwent lateral semi-internal displacement of the patellar tendon stop). Isometric contraction of the quadriceps muscle was started on postoperative day 2, and flexion exercises were started 4 weeks postoperatively. After 3 months postoperatively, patients were allowed to start jogging and light physical activity. When resuming physical activity, patients can wear a simple knee brace and patella stabilization brace for protection. If the patient’s knee flexion and extension range of motion and quadriceps muscle strength return to normal, full physical activity can begin, which generally takes 6 months. IV. Postoperative evaluation Postoperative evaluation is performed to understand the patient’s symptoms, the function of the affected knee, to determine the presence of patellar re-dislocation and the feeling of patellar instability, and to perform the Lysholm score again. Objective assessment included knee mobility, abnormal patellar activity, muscle strength, patellar extrapolation and extrapolation fear test. Postoperative axial plain films of the patella were routinely taken. The preoperative and postoperative Lysholm scores were statistically analyzed using the t-test, and a statistically significant difference was defined as P < 0.01. Results All 6 cases were followed up after surgery for 12 to 36 months, with a mean of 18 months. All patients returned to normal knee mobility at the 12-month postoperative follow-up, none had patellar re-dislocation and no sensation of patellar misalignment or subluxation, and the Lysholm score improved from 76.7±8.7 preoperatively to 95.6±5.7 postoperatively (P=0.000). external thrust test and external thrust fear test were negative at 0 degrees and 30 degrees of knee flexion. All patients were able to palpate cord-like structures on the medial aspect of the patella. No infection, neurovascular injury or rupture of the grafted tendon occurred. The patients felt satisfied and were willing to undergo the same type of surgical treatment. Postoperative radiographs: the patella was well repositioned and there was no recurrent dislocation or subluxation of the patella. Discussion RDP is most common in females aged 10 to 17 years, mainly in patients with 2 or more episodes of lateral patellar dislocation, which are mostly self-replacing.RDP can be caused by poor healing of the joint support tissues after one or more traumatic dislocations. 1/3 of patients have no clear history of trauma, but it more often occurs in knees with one or more underlying anatomical abnormalities that predispose the patella to dislocation or subluxation. The causes of patellar instability include: (1) relaxation of the medial patellar support band; (2) contracture of the lateral patellar support band; (3) hypoplastic femoral epicondyle; (4) small femoral intercondylar depression; (5) knee valgus deformity; (6) knee retroflexion deformity; (7) increased femoral anteversion angle or femoral internal rotation; (8) tibial external rotation; (9) patellar tendon stop deviation; and (10) high patella. The passive stabilizing structures of the knee that prevent the patella from dislocating laterally consist of two parts: the bony stabilizing structures of the femoral talus and the soft tissue stabilizing structures of the medial knee joint. Therefore, surgical methods for treating RDP can be divided into 2 major categories: soft tissue surgery and bony surgery. In 1915, Albee introduced the concept of femoral glide dysplasia and proposed the method of glide osteotomy to correct glide dysplasia and thus treat patellar dislocation. Subsequently, gluloplasty was introduced, which can also be used to correct glulopalatal dysplasia. However, both suffer from high surgical difficulty, the reconstructed patellofemoral joint still does not fit and the possibility of damaging the articular cartilage, and the surgical success rate is low. Therefore, most authors have shifted the focus of treatment to the medial soft tissues of the knee regardless of the presence of femoral condylar dysplasia. The soft tissue stabilizing structures, in turn, are most important in the MPFL, which accounts for 53% to 80% of the total medial block force. Sallay reported that 94% of patients with acute patellar dislocation have MPFL tears, and Cofield reported that approximately 44% of patients with traumatic patellar dislocation eventually develop RDP; therefore, the majority of patients with RDP have MPFL fractures and functional loss, and MPFL reconstruction should be performed first in these patients. Nomura et al. reported intermediate results of artificial ligament reconstruction of the MPFL for RDP, with an excellent rate of 96%. We used a transposition of the greater collecting tendon to reconstruct the MPFL, and the node of the greater collecting tendon as the stop of the greater collecting tendon was anatomically positioned adjacent to the stop of the MPFL, which could maximize the simulation of the femoral side stop of the MPFL without damaging the distal femoral epiphysis by preparing the femoral side marrow tract. In addition, the autologous tendon is more suitable than the allograft tendon in terms of type and body habitus. Intraoperative arthroscopic direct dynamic observation of the patellofemoral joint alignment relationship and motion trajectory can be combined with other surgical modalities to deal with intra-articular lesions while clarifying the lesion, which helps to relieve symptoms; the surgery is less traumatic without affecting the aesthetics and avoids the adverse effects of large tissue trauma from incisional repositioning, and patients can be discharged earlier. However, not all RDPs require lateral patellar support band release, as unnecessary lateral support band release may cause increased patellar laxity and excessive release may also cause medial subluxation of the patella. We performed arthroscopic release of the lateral support band for those with positive preoperative patellar sliding test and negative patellar tilt test, and paid attention to protecting the patellar stop of the lateral femoral muscle during release to prevent medial dislocation of the patella. In this group, four patients underwent lateral support band release. The normal value of Q-angle in knee extension was 10° to 20°, with an average of 14°. Increase in Q-angle may increase patellofemoral instability, and four cases (5 knees) in this group had Q-angle >20°, and all of them underwent distal patellar rearrangement (lateral semi-internal displacement of the patellar tendon stop). In this study, the average follow-up time was only 13 months, and further follow-up is needed to understand the long-term results of this procedure, whether the transplanted ligament can adapt to the growth of the child and whether it will increase the pressure on the patellofemoral joint causing degeneration in the long term. Conclusion Arthroscopic-assisted reconstruction of the MPFL with a large retractor tendon graft is an effective surgical method for the treatment of RDP in children, and the initial follow-up results were satisfactory. Because of the short follow-up period, further studies are needed to understand the long-term effects of this surgical approach.