PATIENT EVALUATION

ASIS, anterosuperior iliac spine.

Anatomy

The anatomical landmarks of the patellofemoral joint are the seven facets of the posterior patella (Fig. 9-1). The medial and lateral facets are each divided into roughly equal thirds and the seventh facet is the odd facet located on the extreme medial border of the patella. On the trochlear side, the patella articulates with the femoral joint surface on the medial and lateral sides of the trochlear groove. The contact area pressures vary in location and load depending on the level of flexion. The patellofemoral contact, load, and tracking are also affected by anatomical variants, as described by Wiberg and Baumgartl (Fig. 9-2). Also, anatomic abnormalities such as patella baja, patella alta, trochlear dysplasia, rotational alignment of the femur and tibia, foot alignment, tight lateral retinaculum, and vastus medialis strength also play a role in the tracking and articulating pressures and contact locations (Fig. 9-3).

FIGURE 9-1 Seven facets described on the articular side of the patella. Illustrated are the superior medial facet (SMF), middle medial facet (MMF), inferior medial facet (IMF), superior lateral facet (SLF), middle lateral facet (MLF), inferior lateral facet (ILF), and odd facet.

FIGURE 9-2 Anatomic variants of the patella as described by Wiberg and Baumgartl.²³ I. Equal medial and lateral facets which are both slightly concave (normal anatomy), II. Smaller medial than lateral facet, medial facet is flat or slightly convex, III. Very small medial facet which is convex, IV. Without a medial ridge or medial facet.

FIGURE 9-3 Anatomy of the medial aspect of the knee. The MPFL provides 53% of the restraint to lateral displacement of the patella. The patellomeniscal ligament and medial retinacular fibers are responsible for 22% of patellar restraint.

Alignment can also be assessed with the Q angle. As the knee angle changes the Q angle changes. It is largest with the knee in full extension. This has been proposed to be the reason for patellar subluxation/dislocation with the deforming force being greatest in the most unstable position. The foot-up model of the dynamic Q angle is based on the premise that increased foot pronation results in tibial internal rotation producing an in-turned knee posture and “functional” valgus.

Anatomic variations responsible for patellofemoral problems vary, depending on the pathology. Excessive femoral anteversion, tibial torsion, patella alta, and shallow trochlea are all causes of patellar instability, and pain caused by instability (Fig. 9-4).

FIGURE 9-4 Trochlear dysplasia as described by Dejour. Type A, Shallow trochlea. Look for the crossing sign on the lateral x-ray. Type B, Flat trochlea. Look for trochlear spur on lateral x-ray. Type C, Lateral convexity and medial hypoplasia. Look for double contour on lateral. Type D, Cliff type; supratrochlear spur and double contour on lateral x-ray.

Diagnostic Imaging

Proper diagnostic imaging is very important in the diagnosis of the root cause of the problem. Patellofemoral pain versus instability can have similar causes. Limb alignment may be assessed on clinical examination. The first images that should be obtained because of availability and ease of use are x-rays. Obtaining standing anteroposterior (AP) lateral views, skyline views of the patellofemoral joint, and standing notch views help in assessing the bony anatomy of the knee. Standing long leg views are important to obtain to assess alignment. Long leg views can assess varus and valgus malalignment, and skyline views can assess patellofemoral cartilage thickness, gross patellar tracking, and whether there is increased patellar tilt. However, routine radiographs have been shown to identify fewer than 50% of osteochondral loose bodies.

When the patella does not engage in the trochlea by 15 to 20 degrees of knee flexion, patella alta may be present. Patella alta may be radiographically assessed in many ways. Laurin and Merchant axial radiographs are obtained with the knee flexed 20 and 40 degrees, respectively (Fig. 9-5). These assess patellar tilt, but one tangential radiograph obtained at 30 degrees of flexion is sufficient in most cases. The x-ray beam is projected caudad at an angle of 30 degrees from the plane of the femur. A line is drawn along the lateral facet of the patella, and a second line is drawn between the condyles of the trochlea anteriorly. Normally, the angle between these two lines will be open laterally. However, if the lines are parallel or the angle opens medially, the patella is probably tilted.

FIGURE 9-5 A, Diagram of the Merchant skyline radiographic view. The sulcus angle ABC is bisected by BD and line BE passes from the deepest point of the sulcus through the most posterior point on the ridge of the patella. The positive congruence angle DBE indicates lateral translation. B, The Laurin angle is negative, indicating patellar lateral tilt.

Teitge and colleagues¹ have described a radiographic technique that can be helpful for diagnosing patellar instability. They obtained bilateral axial radiographs of the patellofemoral joints in anatomic position, with constant medial and lateral force applied to the patellae with an instrumented device, and then axial radiography was repeated. It was found that a 4-mm increase in medial or lateral patellar excursion compared with the patellar excursion on the asymptomatic knee correlated with patellar instability. Stress radiographs are helpful in identifying patients with congruity of the articular surfaces whose knees may subluxate or dislocate because of deficient ligamentous structures. Patients who are unable to relax the extensor mechanism because of pain or who have bilateral symptoms are not candidates for stress radiography.

Once x-rays have been evaluated, limb alignment may be further evaluated for rotational malalignment with an overlay alignment CT scan to measure femoral anteversion and tibial torsion. CT has been shown to be more sensitive than axial radiography in delineating patellar malalignment²; it allows axial cuts of the patellofemoral articulation at angles less than 20 degrees of knee flexion, which enhances the detection of subluxation as the patella loses the stabilizing function of the lateral femoral condyle. Another role for CT is in identifying lateralization of the tibial tubercle, as measured by the distance between the tibial tubercle and the trochlear sulcus (Fig. 9-6). An axial CT image demonstrating the femoral trochlear groove is superimposed on an axial image of the tibial tubercle. A line is drawn on this superimposed image between the posterior margins of the femoral condyles. Two lines are drawn perpendicular to this line, one bisecting the femoral trochlear groove and the other bisecting the anterior tibial tuberosity. The distance between these two lines determines the extent of lateralization of the tibial tubercle. Values greater than 9 mm have been shown to identify patients with patellofemoral malalignment, with a specificity of 95% and a sensitivity of 85%.³

FIGURE 9-6 Top left, (A) Frontal scans of hips and (B) first transverse section. This should image the greater trochanter, femoral neck, femoral head, and acetabulum of both sides. Right, (C) Frontal scan of the knees and second and third transverse sections as seen on the frontal scan of the knees. (D) The second transverse section is made through the middle of the lateral condyle of the femur, below the lower pole of the patella. (E) The third section should image the tibial condyles and fibular head of both sides. Lower left, (F) Frontal scan of the ankles with the fourth reference line for the transverse section through the lower end of the tibia, which bisects the most distal area of the tibia. (G) The fourth tranverse section should image the tibia and fibula on both sides.

The mean tibial tuberosity-trochlear groove (TTTG) distance is 12 mm in the normal knee. However, in people with a history of patellar dislocation, it has been shown to be more than 20 mm (as measured by CT, with the knee extended) in 56% of individuals (Fig. 9-7).⁴ There has been an association between trochlear and patellar dysplasia and patellofemoral arthritis. Of individuals with isolated patellofemoral osteoarthritis, 42% had a Wiberg type II patella.⁵ The sulcus angle and TTTG distance have been shown to have the strongest correlations with patellar instability.⁶

FIGURE 9-7 A, Diagram of TTTG distance. This also shows the sulcus angle (SA). B, The TTTG distance as measured by superimposing axial CT images of the tibial tubercle and the trochlear groove with the knee in extension.

A rotational CT scan is used to look at femoral anteversion and tibial torsion. Normal measurements for femoral anteversion vary widely and there is significant controversy regarding acceptable normal levels. The femoral anteversion range is 14 to 28 degrees, whereas the mean tibiofibular torsion range is 16 to 50 degrees.⁷

MRI is more effective at identifying articular-cartilage damage directly for large lesions or indirectly by changes in the underlying bone for smaller lesions. Sallay and associates⁸ have used MRI to visualize the pathoanatomic features of patellar dislocations. They identified the essential lesion of patellar dislocations as a tear of the medial patellofemoral ligament (MPFL) off the femoral insertion. The location of the injury was confirmed by surgical exploration. Although other authors have identified avulsions of the MPFL from the patella with the use of MRI alone, this is an uncommon location for injury and may be overinterpreted on MRI studies of patients.

Joint Geometry

Stability of the patellofemoral joint is dependent on the shape of the trochlear groove and the steepness of the slope of the patellar facets. The trochlear groove was first analyzed radiographically by Merchant with the skyline radiographic view performed at 45 degrees of knee flexion. This assesses the patellar position within the groove. The congruence angle of Merchant⁹ is used to identify lateral translation and the lateral patellar angle of Laurin is used to identify lateral tilt.¹⁰

The lateral facet of the patella is responsible for 60% of the load across the normal patellofemoral joint. As the knee flexes, the contact force on the patellofemoral joint increases. Conversely, as the knee moves toward full extension, the contact force decreases and the patella leaves the trochlear groove as it moves proximally. At this point, the patella is dependent on the soft tissues for lateral stability. Contact points on the femoral trochlea and patella change as the knee flexes more (Fig. 9-8).¹¹ The starting point of the patella is lateral, with a catching mechanism that moves the patella medially into the trochlear groove at initial flexion. The patella then tracks in the trochlear groove moving distally, laterally, and posteriorly in line with the groove orientation.

FIGURE 9-8 Knee flexion causes the patellofemoral joint contact area to move distally and posteriorly across the surface of the femur and proximally across the patella. However, in deep flexion, the patellar contact area changes to two separate areas as it bridges the femoral intercondylar notch.

The amount of load across the patella increases with the degree of flexion, so that at 20 degrees the load across the patellofemoral joint is more proximal in the trochlear groove and more distal on the patella. The load across the patellofemoral joint increases with larger vector forces associated with increased knee flexion. The loads are larger more distally on the trochlea and, as noted, the lateral facet translates 60% of this load. The patellar surface that articulates with the femur changes from distal to proximal on the patella. Therefore, the larger loads seen on the patella are more proximal.¹² The forces have been estimated to increase to 20 times body weight with jumping, 7.6 times with deep squats, and 3.3 times with stairs.¹³

Therefore, abnormal loads resulting in cartilaginous damage are usually seen in the areas represented by higher flexion with chronic patellofemoral osteoarthritis (OA). Loads are also seen to be higher in knees with patella alta (Fig. 9-9). However, in cases of patellar instability and maltracking, the areas of pathologic findings represent the cause of the maltracking. A subluxating patella will be damaged with the dislocation and, more specifically, relocation of the patella, whereas a lateral tracking patella will have increased wear of the lateral facet and lateral trochlea.

FIGURE 9-9 A, Measurement of the Insall-Salvati ratio, which is the ratio of the patellar tendon length (PT) to the length of the patella (PL), as measured on a lateral knee xray. The PL/PT is normally 0.8 to 1.2. B, Measurement of the patellotrochlear index, which is a reflection of the functional height of the patella TO/PL It is used to examine for patella alta if the PL/PT ratio is negative.

Classification of Patellofemoral Disorders

A clinical classification of patellofemoral disorders was developed by Merchant¹⁴ based on cause. The five major categories are trauma, patellofemoral dysplasia, idiopathic chondromalacia patella, osteochondritis dissecans, and synovial plicae. Also, pigmented villonodular synovitis (PVNS) can present as an asymptomatic finding, but can cause patellofemoral pain.¹⁵ In this chapter, we will present examples of these different causes of patellofemoral problems and their arthroscopic presentations.

Holmes and Clancy¹⁶ have proposed a clinical classification of anterior knee pathology that has three major classification classes: patellofemoral instability, patellofemoral pain with malalignment, and patellofemoral pain without malalignment (Box 9-1).

First, it is first important to characterize the primary complaint, instability versus pain. Second, what is the primary cause of the complaint? Once the differential diagnosis is narrowed down, diagnostic imaging can be used to work up the diagnosis. Finally, arthroscopy can be used to evaluate the cause and degree of the pathology further.

TREATMENT

Arthroscopic Assessment of the Patellofemoral Joint

Arthroscopy can provide direct visualization of the patellofemoral joint and improve on the diagnostic accuracy of the cause of patellofemoral pain or instability. The patellofemoral chondral surfaces and medial and lateral restraints of the patella can be directly examined.

Position and Setup

The patient is placed supine on the operative table after anesthesia. A wide variety of anesthesia can be used—epidural, general, spinal, or local. Once the patient is under anesthesia, a physical examination should be performed, looking at range of motion, patella tracking, patella crepitus, and tightness of the lateral retinaculum, and a ligamentous examination of the knee should be carried out. Always compare the results with the contralateral knee. A leg holder should not be used, becaue it can interfere with use of the superior patellar portal in the examination of the patellofemoral joint. A tourniquet may or may not be placed.

Once the patient is prepped and draped, marks for the portals should be made. The superomedial portal is placed 2 cm above the superomedial pole of the patella, just medial to the central quadriceps tendon. This is a portal favored by Pidoriano and Fulkerson.¹⁷

However, Grana and coworkers¹⁸ have found that viewing patellar tracking is equivalent from a superior or inferior portal. Nissen and colleagues¹⁹ have favored the use of standard arthroscopic portals. The other three standard portals include the inferior lateral, inferior medial, and superior lateral portals. All portals can be used to maximize the visualization of the patellofemoral joint, as well as the rest of the knee joint. The use of the superomedial portal has the benefit of being less impaired if a post is used during arthroscopy compared with the superolateral portal. However, the superolateral portal may provide a better view of the lateral facet of the patella and the lateral side of the trochlea.

The evaluation of the patellofemoral joint should be divided into the assessment of joint surfaces and joint tracking.

Technique

As with any arthroscopic procedure of the knee, this is initiated with a thorough history and physical and imaging protocol. After this is carried out and determining that surgery is appropriate for an individual patient, many decisions are required. We find that the arthroscopic procedure performed for a patient with an isolated patellofemoral problem is often one of confirmation and/or exclusion of other diagnoses. By the time the patient and surgeon have decided an arthroscopic procedure should be performed, all nonoperative solutions have been exhausted. A decision to perform a specific procedure for the treatment of a problem, such as a medial patellofemoral ligament reconstruction for patellofemoral instability, has been made. Most surgeons prefer to perform a diagnostic arthroscopy prior to performing the definitive procedure that has been chosen.

Having confirmed with the patient in the preoperative holding area the procedure and knee that is being operated, on the patient is put on the operating room table in the supine position. After anesthesia has been induced, an examination of the uninvolved and involved knee is performed. Side to side differences are noted. Differences in the examination in the sleeping or anesthetized patient, compared with the awake patient, are also noted.

After examination, a tourniquet is applied to the leg prophylactically. We generally do not inflate the tourniquet unless necessary during the case for visualization purposes. Use of a leg holder versus a side post is left to the surgeon’s preference. We use a side post that can be flipped down to allow complete access to the superior aspect of the knee for accessory portals (Fig. 9-10).

FIGURE 9-10 A, Leg with side post in down position. B, Leg with side post in up position.

Initial evaluation after prepping and draping is performed through an inferior lateral patella portal (Fig. 9-11); care must be made to place this portal laterally enough away from the patella tendon and inferior enough from the patella so that easy manipulation up into the patellofemoral joint can be performed without injuring the articular cartilage. We prefer to perform the routine evaluation of the patellofemoral joint, medial gutter, medial compartment, lateral compartment, and intercondylar notch, perform the posterior examination, and then concentrate on the involved pathology in the patellofemoral region. It is important to understand that fluid placed under pressure with a pump or gravity will affect normal mechanics within the knee. Also, the portals used can affect the patella and put it in a less than physiologic position when evaluating the knee. The superomedial portal can be useful in visualizing patellar tracking as well as looking at the angle of tilt of the patella (Fig. 9-12). The lateral facet should align with the trochlear groove by 20 to 25 degrees of knee flexion (Fig. 9-13) and the apex of the patella by 35 to 40 degrees of knee flexion (Fig. 9-14).

FIGURE 9-11 View of patellofemoral joint from the infralateral portal of the left knee.

FIGURE 9-12 View of the patellofemoral joint from the superomedial portal of the left knee; patellar tilt with lateral subluxation at full extension.

FIGURE 9-13 View of the patellofemoral joint from the superomedial portal of the left knee; patellar tilt with lateral subluxation at 20 degrees of flexion.

FIGURE 9-14 View from the superior medial portal of the left knee, with correction of patellar tilt; lateral subluxation at 40 to 50 degrees of flexion.

Grading of all articular cartilage lesions should be performed at this time, as well as sizing them and locating them geographically on the patella and trochlear groove. This should be documented by videography or drawings at the conclusion of the procedure, which can then be used by the surgeon and patient to plan definitive treatment if none has been done during the procedure.

Arthroscopy performed under local anesthesia can be beneficial in determining the cause of the patient’s pain, specifically if it is not intra-articular. Most intra-articular pain generators will be anesthetized with the injection of the local anesthetic into the joint. This is not true, however, with extra-articular pain generators, and therefore potential differentiation can be performed at the time of arthroscopic evaluation.

Evaluation of Joint Surfaces to Help Determine Treatment Options

Cartilage Lesions

The patellofemoral joint must first be assessed by the much less invasive techniques of x-ray evaluation using the sunrise view and the 30-degree bent knee lateral view and AP view of the knee. The most relevant x-ray is the sunrise view taken at 20 degrees of flexion. Kijowski and colleagues²⁰ have found that the most sensitive radiographic findings for degeneration of articular cartilage in the patellofemoral joint are marginal osteophytes. Joint space narrowing, subchondral sclerosis, and subchondral cysts were noted to be insensitive for osteoarthritis of the patellofemoral joint. MRI has also been used to assess the patellofemoral joint. Advantages include direct visualization; it is the most sensitive and specific diagnostic test available.

The patellofemoral joint surface may be examined directly with arthroscopy. Gomes and associates²¹ have proposed a technique to assess the articular surface of the patella for cartilage breakdown more effectively. They used a Kirschner wire to maintain a 45-degree lateral tilt after a lateral retinacular release. However, a lateral retinacular release can be counterproductive and even devastating in some patients, specifically those with an already hypermobile patella. This technique, however, could be used to assist in some operative procedures to address chondral deficits.

If a patient has patellofemoral pain that presents as an articular type of pain, it may be treated conservatively, and often improves. However, if the pain persists for a prolonged follow-up period, it may then be necessary to assess the joint surfaces and treat the patient surgically. There has traditionally been a poor correlation between patellofemoral pain and patellofemoral cartilage lesions. Kettunen and coworkers,²² however, have found that patients who had patellofemoral pain syndrome were more likely to have severe lesions of the patella or trochlear surfaces. These patients also had more functional limitations than patients with mild or no lesions. Assessment of the articular surface is done using the modified Outerbridge classification (Table 9-3).

TABLE 9-3 Modified Outerbridge Classification of the Articular Surface

Grade	Macroscopic Description
0	Normal
I	Softening or swelling
II	Fissuring or small erosions
III	Fragmentation, fissuring, deeper erosions, but not to subchondral bone
IV	Erosions down to subchondral bone

Findings with patellofemoral arthroscopy may direct our understanding of the cause of patellofemoral pain. The location of pathologic lesions on the patella may help us determine whether the pain and lesions are secondary to malalignment, maltracking, or instability.

If the cause of pain is secondary to instability, a significantly different pattern of wear is seen; this is often caused by the acute dislocation. The lesion is usually described as being located on the medial patella and lateral trochlea because of damage from the relocation of the patella within the trochlear groove.

Although the cartilaginous lesions are usually the cause of the pain, other causes of the pain and pathologic lesions are plicae, tight lateral retinaculum, synovitis, a released lateral retinaculum, or a torn or incompetent MPFL. The lateral retinaculum is an extension of the fibrous aponeurosis of the vastus lateralis muscle to the lateral patella. It is best seen with the arthroscope from the lateral infrapatellar portal.

The plicae are bands of remnant synovial tissue irritated by overuse or injury. They resemble a tendon and are usually found in the medial side of the knee at the level of the patella. The mechanism of the pain is thought to be that the plica gets caught between the patella and femur and causes a localized inflammation that results in patellofemoral pain.

Synovitis is usually seen as a diffuse inflammatory response, showing a red inflamed synovium. However, a patient can also have anterior knee pain from PVNS, which, although rare, has been reported.¹⁵ In the one reported case, the PVNS was located in the patellofemoral joint (Fig. 9-15).

FIGURE 9-15 Example of PVNS in Patellafemoral joint.

(From Edwards MR, Tibrewal S. Patello-femoral joint pain due to unusual location of localized pigmented villonodular synovitis—a case report. Knee. 2004;11:327-329.)

The MPFL is a thin structure found in the second layer on the medial side of the knee. It is responsible for 50% to 60% of the restraint of the patella from translating laterally out of the groove. An incompetent MPFL is usually caused by an acute injury. The MPFL normally appears to attach anteriorly at the superomedial border of the patella, where it blends with the tendon of the vastus medialis. Posteromedially, the MPFL attaches to the anterior aspect of the medial epicondyle of the femur. A tear is usually seen on the femoral side.

Medial instability is almost always caused by iatrogenic overrelease of the lateral retinaculum. In the past, this was overused as a method of treatment of lateral tracking because of the ease and speed of the procedure. The assumption was that the release of a tight lateral retinaculum would solve the patellar tracking problems. It is now believed that an isolated lateral release is almost never the answer in the treatment of patellofemoral pathology.

PEARLS& PITFALLS

PEARLS

• Arthroscopy is very sensitive and useful for evaluating the patellofemoral joint.

• the normal TT-TG distance is less than 12 mm.

• Use of a side post allows access to the leg from the superior medial portal.

• Superior portals (medial and lateral) allow unique visualization of the patella articulation with the femoral trochlea.

• An isolated lateral retinacular release is rarely indicated.

PITFALLS

• Many findings seen at arthroscopy may be asymptomatic.

• Obtaining the TT-TG distance requires specific CT software and is dependent on the radiologist’s understanding of the measurement.

• Use of a leg holder inhibits the surgeon’s ability to access the superior aspect of the knee.

• Superior lateral portals are difficult to use with a post and are at risk of developing fistulas.

• Medial subluxation can occur as a complication with a lateral release.

CONCLUSIONS

The diagnosis of patellofemoral problems requires a proper history and physical examination, as well as the use of radiography, CT, and MRI. However, arthroscopy of the patellofemoral joint is an important tool in the diagnosis of causes of patellofemoral pain and/or instability. Cartilaginous lesions, patellofemoral instability, and articular dysplasia can be assessed using this technique and, in some cases, surgical treatment can be directed and assisted by arthroscopy. It is important to remember that treatment should be based on arthroscopic findings in conjunction with clinical signs and symptoms.