Arthroscopic Medial Plication for Patellar Instability

Published on 10/03/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 1629 times

CHAPTER 10 Arthroscopic Medial Plication for Patellar Instability

The proper treatment of patellar instability requires a comprehensive understanding of patellofemoral anatomy and biomechanics, accurate physical examination skills, and appropriate imaging studies. Surgical options abound for this problem, but the patella is a sensitive structure, and inappropriate or overly aggressive procedures can have disastrous consequences. For patellar instability, I have found that less is more, and that arthroscopic realignment is the procedure of choice for the patient with routine instability that requires surgical intervention.

An acute dislocation of the patella may be described as a primary disruption of the patellofemoral relationship, where the patella is displaced out of the femoral sulcus. The direction of dislocation is most commonly lateral, although superior,¹ medial,² intra-articular,¹ and vertical intercondylar³ have also been described. This chapter will deal only with the most common lateral patellar dislocation.

The incidence of acute dislocation of the patella is difficult to assess accurately, and has only been addressed in a few studies. McManus and colleagues ⁴ have reviewed the records of 94,875 pediatric visits to their emergency room over a 4-year period and found some evidence of patellar dislocation in 55 (0.05%), although only 33 (0.03%) could be proven to be acute dislocations. Cash and Hughston⁵ reported treating 399 patients with this disorder over a 30-year period (13.3 patients/year). Castelyn and Handelberg⁶ have reported an incidence of acute patellar dislocation of 2.44% in their series of knee injuries. More recently, Nietosvaara and associates⁷ have reported an incidence of acute patellar dislocation in 43/100,000 adolescents/year (0.04%).

Most reports in the literature agree that acute patellar dislocation occurs in a young population, with an average age of approximately 20 years most commonly reported.⁴^,⁵ However, within this group, the incidence of redislocation appears to be significantly higher in patients who sustain their first dislocation at an earlier age.⁴^,⁸ Although earlier studies suggested that females appeared to be at a higher risk for both acute and recurrent dislocation.⁸^–¹⁰ more recent data suggest that the gender incidence may be equal.¹¹

The natural history of acute dislocation has been addressed in a number of studies. Hawkins and coworkers ¹² treated 20 patients conservatively (3 weeks of immobilization). At 40 months of average follow-up, 3 had redislocated (15 %), 4 had apprehension or complaints of instability (20%), and 15 had pain associated with patellofemoral (PF) crepitus. Remarkably, 100% were able to return to work and recreational sports. Cofield and Bryan⁹ studied 48 patients with acute dislocations treated conservatively (closed reductions and immobilization for 1 to 6 weeks). Of these, 44% redislocated and 27% went on to require subsequent surgery. If subjective criteria are include, 52% were considered failures.

McManus and colleagues ⁴ reviewed 28 patients with acute dislocations; 21 were treated without surgery. Five patients redislocated and 11 were considered symptomatic. Cash and Hughston⁵ have reported a redislocation rate of 20% to 43% among first-time dislocators treated with immobilization alone, with the rate depending on the presence of congenital predisposition (patellofemoral dysplasia). In a more recent study of 74 acute patellar dislocations, Atkin and assoociates¹¹ reported that 58% of patients were still symptomatic during strenuous activities at an early follow-up of 6 months.

Review of these studies suggests that the natural history of acute patellofemoral dislocation is that of a high percentage of redislocation (20% to 40%) and continued symptoms.

ANATOMY

The patella is the largest sesamoid bone in the human body ¹³ and, like other sesamoid bones, most likely evolved as a mechanism to protect the adjacent tendon (quadriceps in this case) from abrasion and to improve distribution of forces across the tendon.¹⁴ In addition, the patella acts to improve the biomechanical leverage of the quadriceps during extension, acting as a fulcrum,¹⁵ with its greatest effect seen at 20 degrees of flexion.¹⁶

Most animal species load the knee in a flexed position and have a well-stabilized patella maintained within the confines of the trochlea throughout the range of motion. Humans, on the other hand often load the knee near full extension, with the patella out of the confines of the trochlea and susceptible to instability. Interestingly, in great apes, the femoral diaphysis is straight and the trochlea is flat, with no elevation of the lateral trochlea ridge; however, the patella remains stable because the flexed angle of their knee. In humans, the distal femur evolved into an obliquity angle of 8 to 10 degrees, necessitating the development of a deeper trochlea sulcus, with an elevated lateral trochlear lip, to stabilize the patella.¹⁷

Patellar tracking is dependent on many anatomic factors. Static factors include bony anatomy and ligamentous structures. Dynamic factors include muscular interaction and change in anatomic relationships related to the range of motion of the knee. Static bony anatomy that contributes to PF stability includes the depth and shape of the trochlea, shape of the patella, femoral anteversion, and tibial rotation, along with the relative valgus alignment of the femoral-tibial articulation.

The Q angle is defined as the angle between the quadriceps mechanism and the patellar tendon, and is a helpful measure of patellar tracking. The greater the anatomic valgus, or the greater the external rotation present in the tibia, the larger the Q angle will be, resulting in a laterally directed force vector (Fig. 10-1). A normal Q angle is typically considered to be less than 15 degrees. Increased femoral anteversion will also result in a high Q angle by causing internal rotation of the femur relative to the tibia.

FIGURE 10-1 The Q angle is measured from the center of the patella to the anterior superior iliac spine (ASIS) superiorly and the tibial tubercle inferiorly. This measurement is a useful clinical method for evaluating patella alignment and tracking. An angle of more than 15 degrees is often considered abnormal.

(Modified from Halbrecht JL. Patella problems in athletes. In Johnson DJ, Pedowitz, RA, eds. Practical Orthopaedic Sports Medicine and Arthroscopy. Philadelphia: Lippincott, Williams & Wilkins, 2006:677-707.)

From full knee extension to approximately 20 degrees of flexion, the patella rests superior to the trochlea groove and is stabilized by a combination of ligamentous and muscular forces. Patellar stability at this point is determined by overall quadriceps tension, with medial-sided stability contributed by vastus medialis obliquus (VMO) contracture and medial patellofemoral ligament tension, whereas lateral stability is contributed by the lateral retinaculum. After approximately 25 degrees of flexion, the patella becomes engaged in the trochlear groove and stability is maintained by bony congruity with the trochlea.

Causative Factors

Disruption or distortion of any of these anatomic factors can result in patellofemoral instability.

Medial Patellofemoral Ligament Insufficiency

Numerous studies have shown that the medial patellofemoral ligament (MPFL) is the most important stabilizer of the patella to lateral translation. Hautamaa and associates ¹⁸ have shown in a cadaver biomechanical study that with serial sectioning the MPFL contributes 50% of the patellofemoral stability (Fig. 10-2).¹⁹ In a similar study, Desio and coworkers¹⁹ have shown that the MPFL contributes 60% of the stability, with the medial patellomeniscal ligament contributing an additional 13%. Clinical studies on patients with patellar dislocation also seem to confirm this finding, with both surgical and MRI evidence of disruption of the MPFL after patella dislocation.²⁰^–²²

FIGURE 10-2 The medial patellofemoral ligament contributes more than 50% of the stability of the patella (P) to lateral dislocation, shown here in a cadaver dissection. The arrow next to P is on the patellomeniscal ligament. The curved arrow is on the patellotibial ligament, which has been reflected from the tibia. The large arrow points to the medial patellofemoral ligament, which is supported by a probe. The arrowhead is on the superficial medial collateral ligament.

(From Desio SM, Burks RT, Bachus KN. Soft tissue restraints to lateral patellar translation in the human knee. Am J Sports Med. 1998;26:59-65.)

Trochlear Dysplasia

Trochlear dysplasia is thought to be a contributor to patellar instability, but is often difficult to measure accurately radiographically. Magnetic resonance imaging (MRI) is the most accurate method of measurement, with an MRI study showing a 100% sensitivity and a 96% specificity when a criteria of less than a 3-mm trochlear depth was used, measured 3 cm above the femorotibial articulation.²³ However, trochlear dysplasia is rare and, even when present, does not usually require direct treatment. In addition, patients with unilateral instability have trochlear dysplasia bilaterally, but only instability on the side where they have disrupted their MPFL. For rare cases of patella instability thought to be caused by severe trochlear dysplasia, several procedures have been described to deepen the trochlea (trochleoplasty), including osteotomy and elevation of the lateral trochlea with bone grafting ²⁴ or undermining and deepening of the trochlea groove.²⁵ It is my experience that even in the presence of trochlear dysplasia, trochleoplasty is usually not necessary. The vast majority of cases of patellar instability can be managed with proximal soft tissue realignment, and occasionally with the addition of a tibial tubercle osteotomy.

Excessive Femoral Anteversion.

An increase in femoral anteversion will result in relative external rotation of the tibia. This will cause an increase in the Q angle and a tendency toward lateral tracking of the patella. Correction of patellar malalignment in this situation can theoretically be accomplished through a femoral osteotomy. However, because of the obvious invasiveness and risks, this procedure is very rarely recommended. As noted, treatment is directed at stabilization of the proximal soft tissues. If absolutely necessary, the relative external rotation of the tibia can be addressed by osteotomizing the tibial tubercle and moving it medially to decrease the Q angle. In my experience, this is necessary in less than 10% of cases.

Excessive Tibial External Rotation.

Excessive external rotation of the tibia can be a primary anatomic finding, rather than a result of femoral anteversion, and will also result in lateral displacement of the tibial tubercle and an increased Q angle. When this rotation is extreme, and the Q angle exceeds 20 degrees, consideration may be given to correcting this angle by osteotomy and medialization of the tubercle. However, in our experience, a proximal soft tissue alignment is still often quite successful. We reserve the addition of an osteotomy to cases with chondral damage to the lateral facet of the patella or lateral trochlea, where elevation of the tubercle will unload the articular surface, or to those rare cases documented by CT or MRI as having a large offset between the tibial tuberosity and trochlear groove (tibial tuberosity–trochlear groove [TTTG] distance). This is more accurate than the clinical measurement of the Q angle and, in the rare case when the offset is more than 20 mm, I will consider an osteotomy of the tibial tubercle to correct the TTTG by moving the tubercle medially by 1 cm or more to obtain a TTTG distance closer to 10 to 15 mm.

Weak Vastus Medialis Obliquus.

The VMO is a distinct muscle grouping of the vastus medialis that has a distinct nerve supply ²⁶ and whose fibers insert into the superomedial aspect of the patella at approximately 65 degrees to the longitudinal axis.²⁷ There can be anatomic variation on the insertion site to the patella. A more medial insertion results in a more medial stabilizing effect. A more superior insertion changes the force vector, providing less medial stabilizing force, and may be a contributor to lateral instability. Weakness of the VMO removes a significant dynamic stabilizing influence on the patella. Strengthening of this muscle is an important component of any rehabilitation program. The VMO contracts maximally in the terminal range of extension, and can be best strengthened with terminal, short-arc extension exercises.

PATIENT EVALUATION

History

The diagnosis of acute dislocation of the patella is not difficult if the patella is obviously dislocated at the time of presentation. However, if the patella spontaneously reduces, the diagnosis may be difficult. The patient will often report that the knee gave way or “popped out of place” during a twisting activity. The patient may give a history of previous subluxation episodes or a history of dislocation of the opposite knee. Patients with generalized hyperlaxity may also give a history of shoulder or ankle instability.

Understanding the mechanism of injury may be helpful in determining the diagnosis. Two mechanisms for acute lateral patellar dislocation have been proposed, direct and indirect. The indirect mechanism of injury results from a powerful quadriceps contracture against an internally rotated femur (externally rotated tibia) and usually involves a sudden twisting motion on a firmly planted foot, such as a sudden change of direction while playing soccer with cleats. The direct mechanism of dislocation involves a direct blow to the medial aspect of the patella, resulting from a fall or contact with another player during athletic activity. A combination of these two mechanisms may contribute to a single injury (Fig. 10-3).

FIGURE 10-3 Two mechanisms for acute patella dislocation. A, Indirect mechanism. A powerful quadriceps contracture is imposed on an internally rotated femur (externally rotated tibia). B, Direct mechanism. Illustrated here is a direct blow to the medial aspect of patella, usually by a direct fall onto the knee.

Physical Examination

The first step in evaluating a patient for instability is to inspect the knee with the patient standing. Valgus alignment, foot pronation, and the presence of squinting patellae should be noted. The Q angle is an essential parameter for evaluating instability. It is measured from the anterosuperior iliac spine (ASIS) to the center of the patella, and from the center of the patella to the tibial tubercle. This angle should be evaluated in full extension as well as at 30 degrees of knee flexion and seated at 90 degrees. There is controversy regarding which evaluation is most useful, and there has been significant intraobserver error reported.²⁸ In my practice, the 0- and 30-degree Q angles are most helpful.

The lateral retinaculum is then assessed for tightness, with the patient supine and the leg fully extended. If the examiner cannot lift the lateral edge of the patella 15 degrees, or to a neutral plane parallel to the examination table, the patellar tilt test is positive and indicates a tight lateral retinaculum. The examiner should also attempt to translate the patella medially. Movement of less than 15 mm, or one quadrant of the patellar width, indicates a tight lateral retinaculum (Fig. 10-4).

FIGURE 10-4 Positive patellar tilt test. A, This is defined as the inability to lift the lateral facet of the patella more than 15 degrees (or to neutral; B) and indicates a tight lateral retinaculum.

Laxity of the medial retinaculum (medial patellofemoral ligament) is then assessed by lateral translation of the patella and is compared with the opposite knee. The ability of the examiner manually to translate the patella laterally, more than 50% of the width of the patella, is a sign of instability, especially if this excursion is greater than the opposite knee. Subjective apprehension on the part of the patient with this maneuver indicates a positive apprehension test and also suggests instability.

The J sign is a dynamic test that demonstrates severe lateral translation of the patella with active terminal extension of the knee, and often is seen with more severe cases of instability, especially with a high-riding patella. The patient is asked to extend the knee actively against gravity. As the patella disengages from the trochlea, it jumps laterally, making an inverted J sign (Fig. 10-5).

FIGURE 10-5 The J sign is an indication of significant lateral patellar tracking. It is demonstrated here as the patella jumps out laterally as the knee is actively extended, creating a an inverted letter J.

Acute Patellar Dislocation

When examining a first-time dislocator, confusion with an anterior cruciate ligament (ACL) tear is common. The patient will usually present with a large hemarthrosis that will be tender to palpation along the medial retinaculum. There may also be tenderness along the lateral trochlea and medial patella from direct contact and shearing forces during the relocation phase of the injury. Often, however, the pain may be diffuse, and the diagnosis will remain in doubt. Demonstration of a hemarthrosis by needle aspiration, associated with tenderness along the medial retinaculum and subjective apprehension with lateral translation of the patella, will help suggest the diagnosis. In the case of an osteochondral fracture, marrow fat globules may be evident in the knee aspirate.

Diagnostic Imaging

The standard evaluation of patients with PF symptoms will include an anteroposterior (AP), lateral and 45-degree tangential view of the patella (Merchant view; Fig. 10-6). The AP view is rarely helpful, but will reveal the presence of a bipartite patella or a patellar fracture. The lateral view is useful to identify loose bodies and patella alta or baja, and will reveal the presence of early degenerative changes with inferior or superior patella osteophytes. The lateral view will also reveal calcification or traction spurs associated with chronic patellar-quadriceps tendinitis, and will demonstrate irregularity of the tibial tubercle associated with Osgood-Schlatter disease. Trochlear depth and dysplasia may also be determined on the lateral view (Fig. 10-7),²⁹ although there is poor reproducibility and interobserver reliability.³⁰

FIGURE 10-6 A, 45-degree patellofemoral view (Merchant view). This is extremely useful x-ray for evaluating the patellofemoral joint. B, Merchant view radiograph demonstrating lateral patellar tilt and translation.

(A modified from Halbrecht JL. Patella problems in athletes. In Johnson DJ, Pedowitz, RA, eds. Practical Orthopaedic Sports Medicine and Arthroscopy. Philadelphia: Lippincott, Williams & Wilkins, 2006:677-707.)

FIGURE 10-7 A true lateral x-ray can be helpful for determining trochlear depth (arrow).

The Merchant view is useful to evaluate patellar alignment. Various measurements can be performed on this x-ray that are useful in assessing patellar tracking, including the congruence angle, tilt angle, and lateral translation (Fig. 10-8).

FIGURE 10-8 The Merchant view can be used to measure the congruence angle and patellar translation relative to the trochlear groove. In this patient with patellar instability, the values for both measurements were significantly abnormal. Red lines demonstrate abnormal congruence angle.

For acute dislocations that have spontaneously reduced, radiographs are often normal. However, the radiographs will be helpful if they reveal an osteochondral fracture fragment from the medial patellar facet or lateral trochlea. An avulsion fracture of the retinaculum off the medial border of the patella is even more diagnostic. Teitge ³¹ has suggested the use of an axial oblique view to visualize this avulsion better. The incidence of chondral and osteochondral fracture is up to 46% in patients with acute patellar dislocation,³² although most of these are only visible by arthroscopic and MRI evaluation. Occasionally, the diagnosis of patellar dislocation can only be made by the demonstration of a torn medial retinaculum as seen by MRI (Fig. 10-9). In several studies, MRI was thought to be diagnostic in 81% to 87% of cases, whereas surgical evaluation was diagnostic in confirming a torn medial retinaculum in 94% to 100% of cases.²⁰^–²²

FIGURE 10-9 MRI can be useful to diagnose patellar instability, especially in acute cases in which disruption of the MPFL can be demonstrated (arrow).

MRI may also be useful to find subtle changes of chondromalacia in the patella and trochlea associated with the instability, or to observe chondral fractures and loose bodies. Computed tomography (CT) scanning can be useful to evaluate the patellofemoral joint in lesser degrees of flexion, when the standard 45-degree Merchant view does not demonstrate anticipated malalignment,³¹^,³³ although I rarely order this study. Both MRI and CT scanning can be used to measure the TTTG distance; I prefer to use MRI, which provides similar information without radiation exposure, and also provides additional information about the status of the MPFL and cartilaginous structures (Fig. 10-10).

FIGURE 10-10 A, CT scanning is useful to evaluate patellar tracking in varying degrees of extension. B, It may be used as an objective way to measure the relationship of the tibial tuberosity (TT) to the trochlear groove (TG) to determine a true Q angle (using the TTTG distance).

TREATMENT

Conservative Management

The traditional treatment for a primary acute patellar dislocation has been nonoperative, although the exact method of treatment is controversial. Most literature studies recommend some type of immobilization followed by an aggressive rehabilitation program.⁵^,⁹^,¹² However the benefit of immobilization following patellar dislocation has not been proven, and results after immobilization and early range of motion appear to be the same.⁵^,⁹ Furthermore, there does not seem to be a correlation between the length of immobilization and results.⁵^,⁹^,¹²

Long-term results suggest that the natural history of nonoperative treatment of patellar instability is not as favorable as commonly thought. In a group of 20 patients with acute dislocations treated conservatively, Hawkins and coworkers ¹² have reported a 20% incidence of ongoing instability and a 15% incidence of pain and crepitus. In their series of 48 patients with acute dislocations, Cofield and Bryan⁹ reported a 44% incidence of redislocation; 27% of these patients went on to subsequent surgery and, taking into account subjective criteria, 52% were considered failures. In 21 patients treated nonoperatively, McManus and colleagues⁴ have reported 5 redislocations and 11 patients who remained symptomatic. Cash and Hughston⁵ have reported a redislocation rate of 20% to 43%, depending on anatomic evidence of dysplasia predisposing to instability. In a more recent study, Fithian and associates³⁴ have shown that of 189 patients followed over a period of 2 to 5 years after an acute patellar dislocation, 17% of first-time dislocators and 49% of patients with a previous history of dislocation had recurrences. The risk for recurrence was higher in women.

Surgical Management

With the advent of less invasive methods of treatment, and with more critical reviews of the results of nonoperative treatment, more authors are recommending surgical treatment, even for acute patellar dislocation.¹⁰^,²¹^,³⁵^,³⁶ Relative indications for surgical treatment of a patellar dislocation include failure of conservative treatment, the presence of an osteochondral fracture (loose body), recurring instability, and significant residual subluxation on a postreduction Merchant view radiograph. Contraindications to the treatment of patients with patellas instability are rare, but would include patients with pain syndromes or reflex sympathetic dystrophy (RSD) that could be made worse with surgery. Care must also be taken with patients with unusual underlying pathology, such as Ehlers-Danlos syndrome, although I have successfully treated patients with such rare conditions as Charcot-Marie-Tooth disorder and nail patella syndrome.

Arthroscopic Technique

When surgery is selected as the treatment option, initial arthroscopic evaluation is routinely recommended to confirm the diagnosis and address any loose chondral or osteochondral fractures. The exact method of stabilization is controversial, and numerous procedures have been advocated, both open and arthroscopic.

Role of Lateral Release.

Various surgical techniques have been proposed to correct patellar instability. Lateral release alone has been described, although results have been mixed, with a high incidence of recurring instability.³⁷^–³⁹ Lateral release alone does not address the disrupted anatomy of the medial retinaculum and is not considered by most to be an effective treatment for true patellar instability or malalignment (subluxation).⁴⁰ I do not recommend isolated lateral release for patellar instability.

Role of Medial Retinacular Repair and Reefing.

When surgery is indicated for patellar instability, most recommend treatment for the disrupted medial retinacular structures and medial PF ligament, often called proximal soft tissue realignment. Complex surgical reconstruction of the medial structures, such as the extensive open reconstruction as described by Insall and colleagues,⁴¹ have fallen out of favor, and minimally invasive open procedures for direct anatomic repair of the medial retinaculum and medial patellofemoral ligament have become the preferred method of repair.³⁶ Several arthroscopically assisted procedures have also been reported,⁴²^–⁴⁴ as well as an all- inside arthroscopic method,⁴⁵ which is my preferred technique at this time.

Whether to add a lateral release at the time of medial repair is controversial. Several authors routinely perform a lateral release at the time of medial repair,³⁶ whereas others have shown no advantage to adding a lateral release²² or individualize the decision based on tightness of the lateral retinaculum.³² My current recommendation is to base the decision on the tightness of the lateral retinaculum at the time of surgery, with a tendency toward recommending the lateral release if there is any doubt. An overly tight lateral retinaculum tethers the patella laterally, and will inhibit proper realignment despite medial reefing, especially in more chronic cases of recurring instability.

Distal bony realignment procedures are reserved for patients with severe cases of malalignment recurring instability and a high Q angle, or for patients who have failed previous proximal soft tissue realignment.

Arthroscopically Assisted Proximal Realignment

Initial recommendations for arthroscopic patellar realignment consisted primarily of arthroscopically assisted techniques using a medial incision. Yamamoto ⁴⁴ has treated 30 acute patellar dislocations with arthroscopic lateral release, along with an arthroscopically assisted repair of the medial retinaculum. The transcutaneous passage of sutures through the retinaculum using a large curved needle was recommended, although the sutures were still tied through a medial skin incision. Only acute dislocations were treated. Reported results were excellent, with only one case of redislocation.

Small ⁴³ reported a modified version of the Yamamoto technique, also using an arthroscopically assisted method and a small medial incision (Fig. 10-11). Patients with acute and recurrent dislocations were included, as well as those with malalignment and subluxation. Results were good to excellent in 92.5% of their 24 patients (27 knees), according to a subjective questionnaire. There were two recurrent subluxations, one reoperation for arthrofibrosis, and one superficial infection.

FIGURE 10-11 Arthroscopically assisted patellar realignment is performed by passing sutures percutaneously and then making an incision to tie the sutures.

Henry and Pflum ⁴² have described an arthroscopically assisted technique using cannulated needles, but tied the sutures through a medial incision as well. No follow-up series or results were reported.

All-Arthroscopic Proximal Realignment:

A number of authors have described entirely arthroscopic proximal realignment procedures, with good results.³⁶^,⁴⁵^–⁴⁷ All these procedures have the advantage of eliminating any type of medial incision. The technique for arthroscopic realignment presented here is my preferred method of treatment for most patients with uncomplicated patellar instability.

Surgery is performed under general anesthesia with a thigh holder in place. A tourniquet is applied, but rarely inflated. Before plication, a healing response is created along the medial retinaculum by gently abrading the tissue with a rasp or shaving with a whisker blade (Fig. 10-12). Medial retinacular sutures are introduced percutaneously using an epidural needle (Tuohy needle; Teleflex Medical (Rusch), Deluth, Ga). An epidural needle is essential because the noncutting edge on the inner bevel of the tip prevents cutting or damaging the suture. The needle is placed adjacent to the patella and a no. 1 PDS suture is passed manually through the needle and retrieved arthroscopically through an accessory superolateral portal (Fig. 10-13). Although nonabsorbable braided suture could be used, I prefer an absorbable suture in this highly sensitive area to avoid postoperative irritation from permanent knots.

FIGURE 10-12 Arthroscopic view of medial retinacular abrasion with a shaver to stimulate a healing response prior to tying sutures.

FIGURE 10-13 A, Schematic illustration of the initial step of arthroscopic patellar realignment by percutaneous passage of a suture through an epidural needle inserted adjacent to the patella. B, Arthroscopic view of needle insertion.

The needle is gently withdrawn from the retinaculum but not out of the skin. The needle is then redirected subcutaneously approximately 2 to 3 cm posteriorly and reinserted through the retinaculum (Fig. 10-14). This creates a loop of suture that is again retrieved through the same accessory portal. The needle is withdrawn completely and the process is repeated until four or five sutures are in place. The sutures are retrieved through an accessory proximal lateral portal and clamped for later imbrication of the medial retinaculum (Fig. 10-15). An arthroscopic lateral release is then performed with a standard electrocautery device. After the lateral release, the medial sutures are tied inside the joint from the proximal lateral or anteromedial portal using standard arthroscopic knot-tying techniques (Fig. 10-16).

FIGURE 10-14 Second step of arthroscopic realignment. The needle is withdrawn to a subcutaneous depth, leaving the initial limb of suture within the needle. The needle is then reinserted 2 to 3 cm posteriorly, creating an internal loop of suture.

FIGURE 10-15 Both limbs of suture are retrieved and clamped for later tying. An average of four to six sutures is placed in a similar fashion for a typical case. Tying is performed after lateral retinacular release, if indicated.

FIGURE 10-16 A, Preoperative arthroscopic view of lateral patellar instability. B, Postoperative arthroscopic view of centralized and stable patella after arthroscopic realignment.

Alignment and stability are checked and additional sutures added as necessary. I generally use four to six sutures. Occasionally, some dimpling of the skin will occur in the area of the sutures as the intra-articular knots are tied. The dimpling resolves with resorption of the sutures and postoperative rehabilitation, and has not been a problem. However, if desired, the dimpling can be addressed intraoperatively by spreading gently in the subcutaneous tissues with a small scissor through the medial portal to release the subcutaneous tissue.

PEARLS

1. Use a beveled needle such as a Tuohy epidural needle to avoid damage to the sutures.

2. Take large bites of medial retinacular tissue ( 2 to 3 cm minimum).

3. Use an average of four to six sutures for a typical case.

4. Stimulate a healing response along the medial retinaculum before reefing using a whisker blade or a rasp.

5. Insert sutures before lateral release is performed to avoid extravasation.

6. Tie knots from the superolateral portal.

7. Initial puckering of the medial subcutaneous tissue may occur, but this resolves with rehabilitation.

8. Avoid stiffness postoperatively by allowing range of motion to begin at 1 week.

9. For rare patients with severely damaged native tissue as a result of previous surgery or unusual collagen disorders, consider reconstruction of the MPFL with graft (less than 5% of cases).

10. Distal realignment with a tibial tubercle osteotomy may rarely be added when the TTTG distance is more than 20 mm (less than 10% of cases)

Postoperative Management

Postoperative treatment involves a brace locked in full extension for 1 week, followed by range of motion and physical therapy for 2 to 3 months. Patients are instructed to begin quadriceps contractions and ankle pumps immediately postsurgery in the recovery room. After the first week, the brace is unlocked to enable patients to begin range of motion exercises, but bracing is continued for 3 more weeks until quadriceps strength returns. Patients are not allowed to flex past 90 degrees for 4 weeks, but may begin weight bearing immediately in the brace. Running is allowed at 3 months, followed by return to full twisting and impact sports between 4 and 6 months, depending on return of quadriceps strength.

Avoiding Complications

Complications are rare with this procedure, which is one of the attractive features of an entirely arthroscopic approach. Several technical pearls may be useful to mention. Some dimpling or bunching up of the subcutaneous tissue and skin may occur with this technique as tension is applied to the sutures, and the sutures are tied inside the joint. This will work itself free during rehabilitation, and has not been a problem in my experience. However, if the surgeon wishes to address this intraoperatively, a small dissecting scissor may be inserted through the medial portal and the skin freed from the sutures with gentle spreading of the scissor blades. Overtightening would be difficult to achieve with this technique, and I have yet to encounter this complication. This is also one of the advantages of an arthroscopic approach as opposed to an open reefing. Under arthroscopic guidance, one can titrate the medial reefing under direct vision by inserting as many sutures as necessary and taking as much of a bite of medial tissue as required to obtain a stable and neutral tracking patella. I typically take a 2- to 3-cm bite of tissue and use from four to six sutures.

Postoperative stiffness is possible, even with an arthroscopic approach to patellar realignment, and is a potentially devastating complication. This can be avoided by following the aggressive postoperative rehabilitation protocol outlined earlier. I have successfully avoided this complication using this technique and an accelerated rehabilitation program. It is essential to begin knee motion within 1 week of surgery.

Although I prefer an all-arthroscopic method of proximal realignment, open and arthroscopically assisted proximal realignment procedures can also be effective in preventing recurring instability. However, these procedures are more frequently ac companied by complications, including the risk of joint stiffness, scar tissue, and overcorrection. Medial PF reconstruction using a graft is reserved for rare cases of failed arthroscopic realignment, or patients with inadequate native tissue. I reserve distal realignment for patients with instability associated with significant arthrosis, or those rare patients with a TTTG distance of more than 20 mm on overlapped CT or MR imaging. In these patients, biomechanical unloading and bony realignment of the patellofemoral joint are indicated, in conjunction with a soft tissue arthroscopic realignment.

OUTCOMES

In a review of my 5-year results,⁴⁵ 93% of patients reported significant subjective improvement. The average Lysholm score improved from 41.5 to 79.3 (P < .05). Preoperative and postoperative radiographs were measured for congruence angle, lateral patellofemoral angle, and lateral patellar displacement, and all showed significant improvement postoperatively (P < .05; Fig. 10-17). There were no complications and no redislocations. Patients reported a significant improvement in pain, swelling, stair climbing, crepitus, and ability to return to sports (P < .05). Although the average Q angle in the study was 11 degrees, this procedure was successfully performed on patients with Q angles up to 20 degrees. None of the patients required a second operation for débridement of scar tissue or manipulation. All patients regained full range of motion as compared with the opposite side (Table 10-1).