Chapter 61 Disorders of the Patellofemoral Joint

Basic Anatomy and Biomechanics

The patella is the biggest sesamoid bone in the human body. It is the link from the powerful quadriceps muscle to the patellar tendon. The patella is the connection element in the extensor mechanism, receiving the convergent quadriceps’ fibers in its superior pole and the patellar tendon in its inferior pole.

Only the superior two thirds of the patella have an articular surface. The distal pole is extra-articular and serves as the patellar tendon insertion. Its anterior surface is convex, and on its articular side, the patella is divided by a longitudinal median ridge. This ridge divides the patella into medial and lateral facets, but overall, seven facets are described. A transverse ridge may also exist. In the most medial zone there is a secondary ridge, which delineates the odd facet. The cartilage in the articular surface of the patella is the thickest in the human body, reaching up to 7 mm.

Patellar shape, however, is not constant. Three different patellar types have been described by Wiberg,¹¹⁰ and a type 4 was later described by Baumgartl, the “Jaegerhut” patella, with no medial facet and consequently no median ridge (Fig. 61-1).

Figure 61-1 The four types of patella according to the Wiberg classification as seen on axial views are illustrated here.

A rich arterial plexus supply is present. A complex vascular anastomotic ring lying in the thin layer of loose connective tissue that covers the dense fibrous rectus expansions surrounds the patella. Six main arteries compound this ring. Two intraosseous systems were described by Scapinelli ⁹³: the mid patellar vessels, which enter the vascular foramina situated in the anterior surface; and a second system, which arises from the polar vessels, from the anastomosis behind the patellar ligament. Later, Bjorkstrom⁸ described vessels entering from the quadriceps and from the medial and lateral retinacula.

The trochlea is situated in the distal part of the femur. The normal trochlea is formed by medial and lateral facets, divided by the trochlear groove (TG). The lateral facet is bigger and more prominent, and extends more proximally than the medial facet. Distally, in the transition from the trochlea to the femoral condyles, a groove can be observed in each side. The medial condylotrochlear groove is discrete and at times almost imperceptible; the lateral groove is more marked. The trochlear groove ends in the notch, and this can roughly serve as a reference for establishing surgical trochlear limits—distally from the notch, in a V-shaped form, going proximally through the grooves. Proximally, before the cartilage starts, cortical bone is covered by adipose tissue and synovium, which avoids patellar contact with the cortical femoral bone.

The TG deepens distally, and some controversy still exists regarding its orientation. If its deepest points are taken into account, the natural TG is most often aligned so that it deviates distally and laterally in relation to the femoral shaft axis.⁹⁷

The quadriceps muscle is formed by four parts: rectus femoris, vastus medialis, vastus lateralis, and vastus intermedius. These muscles converge to a tendon 5 to 8 cm superior to the patella. Fibrous expansions arise from the vastus lateralis and medialis, blending with the lateral and medial retinacula, respectively. The most inferior part of the vastus medialis is known as the vastus medialis obliquus (VMO) and inserts in the patella at a mean of 47 ± 5 degrees from the femoral axis in the coronal plane.³⁴ Similarly, a vastus lateralis obliquus can be described, with a more vertical orientation (35 ± 4 degrees).

The patellar tendon arises from the inferior pole of the patella. Its average length is 4.6 cm (3.5 to 5.5 cm), and its width is between 24 and 33 mm.⁸⁷ It inserts in the tibial tubercle, which usually is a little lateralized in relation to the long axis of the tibia; thus the distal orientation of the patellar tendon is also lateralized (valgus). The posterior part of the patellar tendon is separated from the synovial membrane of the joint by the infra-patellar fat pad, and from the tibia, more distally, by a bursa.

Medially, the retinacular expansion of the vastus medialis merges with layers 1 and 2. Other portions of the middle layer persist as a separate layer, forming the medial patellofemoral ligament (MPFL), below the deep fascia and superficial to the joint capsule. The MPFL runs transversely from the patella to the femur. Some controversy exists about its femoral insertion, but it seems reasonable to define it as near the medial epicondyle, just proximal and posterior to it, distal to the adductor tubercle. The patellar attachment is wider than the femoral one, extending from the proximal and medial corner of the patella over approximately half of its length.⁵ The mean length of the MPFL is 53 to 55 mm, and its width varies between studies, ranging from 3 to 30 mm and widening at its attachments.^5,^19,^87,¹⁰⁶ As it approximates the patella, it is overlaid by the distal part of the VMO, and some of its fibers merge into the deep aspect of the muscle. Other bands with secondary functions have been described that link the patella to the tibia and the medial meniscus.

On the lateral side, the superficial oblique retinaculum runs from the iliotibial band (ITB) to the patella. The epicondylopatellar band, the transverse retinaculum (from the ITB to the patella), and the patellotibial band form the deep transverse retinaculum. This connection from the ITB may play a role in patellar lateral displacement and explains the contribution of a tight ITB to patellofemoral pathology.

Patellofemoral biomechanics can be understood as a complex interplay of factors that allow the quadriceps to exert its primary functions—knee extension and deceleration—especially during gait. To allow proper function, one must assume that the patella is located in the trochlear groove and that no instability is present. Continuity of the extensor mechanism should be mandatory to allow proper force transmission, and no pain should be present that would otherwise inhibit the quadriceps function.

The patella increases the moment arm of the extensor mechanism. It concentrates the tension of the converging quadriceps fibers and transmits it to the patellar tendon. In complete extension, a coronal result is produced, and no sagittal forces are expected (this is not completely true, as the retinacula exert some posterior displacement forces in the patella when the knee is near extension). As the knee flexes, however, a posteriorly directed force vector becomes clear, and this raises the patellofemoral joint reaction force. The greater the degree of flexion, the greater is the resultant force vector ⁵³ (Fig. 61-2).

Figure 61-2 The patellofemoral joint reaction force (PFJRF) becomes higher as the knee flexion angle increases. In complete extension, M1 and M2 are in opposite directions, but in the same plane; the resultant PFJRF is almost zero. As flexion increases, M1 and M2 converge, and the vector PFJRF increases.

This posteriorly directed resultant vector is also important for patellar stability in the coronal plane. The trochlear lateral facet is deeper near the TG and becomes more prominent (higher) as it extends laterally. The patellar shape follows this principle: the crest is posterior, and the lateral facet more anterior. Thus, the articulation of the patellofemoral (PF) lateral facets is not in the coronal plane but is oblique in relation to it, with its more medial part posterior to its more lateral part. As a result, when the quadriceps contracts, the resultant posteriorly directed force vector tends to bring the mobile part of the articulation (the patella) medially.

Also in the coronal plane, multiple quadriceps insertions should be noted, along with their different angles of action. The VMO and the vastus lateralis obliquus (VLO) mainly act in an oblique manner in relation to the longitudinal direction. Based on this, malfunctioning of one (VMO) or hyperfunctioning of the other (VLO) can cause coronal displacement and, to a greater extent, instability. If the force-producing capacity of each muscle head is proportionate to its cross-sectional area, the VMO could contribute 10% to total quadriceps tension, and if completely relaxed, it can cause tension to swing laterally to approximately 6 degrees ³⁴ (Fig. 61-3).

Figure 61-3 Anatomic view showing the valgus alignment of the extensor mechanism and the oblique contributions of medial retinaculum, vastus medialis, lateral retinaculum, vastus lateralis, and iliotibial tract.

The angle of the quadriceps insertion and the angle difference of the patellar tendon insertion are other causes of a laterally directed force vector (valgus orientation of the extensor mechanism). This difference can be measured during the physical examination by tracing two lines that intersect each other in the center of the patella: one is traced from the patella to the anterior iliac spine, representing the quadriceps tension line; the other is traced from the patella to the tibial tubercle and represents the patellar tendon reaction force line. This is called the Q angle, and in normal subjects, it is expected to not exceed 15 or 20 degrees. Women have the greatest values.

Soft tissue restraints also play a fundamental role in coronal plane force balance. In complete extension, normal patellae are not engaged in the trochlear groove. This engagement starts at approximately 20 degrees of flexion, when the distal and lateral part of the patella touches the upper and proximal part of the trochlea, which comprises the lateral facet. Because the patella is not engaged before this point, only soft tissue stabilizers act to ensure its coronal location. On the lateral side, the retinaculum is directly linked to the ITB, and tension in the ITB causes the patella to track in a more lateral direction.⁶³ On the medial side, the MPFL contributes 50% to 60% of the restraint to patellar lateral displacement at 0 to 20 degrees of flexion, with a mean failure load of 208 N.^5,⁷⁹ Although MPFL insufficiency is not the cause of lateral dislocation, one cannot assume lateral dislocation without its insufficiency or rupture.

With initial patellofemoral contact, as a result of the articular surface orientation, a medial patellar shift is produced when the patella engages and follows the trochlear groove. As the flexion angle of the knee is increased, the contact area of the patella progresses proximally, while the trochlear contact area progresses distally. From extension to 90 degrees of flexion, the patella holds the quadriceps tendon away from the femur, but with additional degrees of flexion, an extensive area of contact is formed between the tendon and the trochlea. At between 90 and 135 degrees of flexion, the patella rotates and the ridge that separates the medial and odd facets engages the femoral condyle. At 135 degrees, separate lateral and medial (limited to the odd facet) contact areas are formed ^40,⁴¹ (Fig. 61-4).

Figure 61-4 Illustration demonstrating the patellofemoral joint contact areas according to the knee flexion angle.

Patellofemoral Disorders: Analysis

Clinical Symptoms

Pain: In patellofemoral disorders, pain is usually anterior, not well localized, and diffuse. It may be referred to the medial retinaculum or, more rarely, to the lateral retinaculum. Sometimes the pain is located below the patella as a bar. Posterior knee pain, although rare, may also be possible, especially if an articular effusion is present. Usually, the pain is worse during or after activities that overload this joint, such as those that demand jumping, running, or vigorous quadriceps contraction. It can also be felt during stair climbing or descending. Patellofemoral joint reaction force reaches 3.3 times body weight in stair climbing or descending, and 7.6 times during squatting.^62,⁸⁸ Prolonged sitting with the knees flexed is a classical situation that produces or increases patellofemoral pain; this is commonly known as the “movie sign” (or “theater sign”). Although classic, this sign is not specific.

Instability or feeling of instability: Instability is not the best definition for patellar dislocation. The word “instability” describes better the subjective feeling of an unstable knee.

Instability may be objective, as found in patellar dislocation. It is a mechanical phenomenon whereby the patella loses contact with the trochlea. It first occurs during high-energy activity and is always followed by an effusion, hemarthrosis, or even sometimes hematoma, caused by capsule rupture and subcutaneous blood diffusion.

Instability may also be subjective, caused by quadriceps inhibition (reflex) as the result of pain, without loss of contact of the articulating surfaces. It occurs with low-energy activities, such as walking or ascending or descending stairs.

Careful analysis of the clinical history and the patient’s description of the episode is useful in differentiating them. Objective instability is easy to diagnose when the patella is still dislocated, when a medical report indicates that a doctor relocated it, or when x-rays have been taken and show the dislocation or objective signs of it, such as a bone fracture (avulsion) on the medial part of the patella or on the lateral femoral condyle.

Locking or catching: This is the third great complaint and must be differentiated from those produced by meniscal tears. The main difference from meniscal locking is the fact that the knee is not able to flex or to extend, but in a bucket-handle tear the knee is able to flex. Sometimes, locking in extension may be produced by a reflex mechanism of quadriceps and hamstrings contracture due to pain to avoid contact of the opposed cartilage surfaces. Locking episodes are usually momentary, but in some cases can last for longer periods and can be really painful. Catching may be produced by cartilage lesions or by irregularities in the patella or trochlea as they glide over each other.

When collecting the clinical history, the physician should ask about associated conditions, which are common in patellofemoral disorders, such as global pathology—low back pain, hip catching, ankle pain or sprains—and, of course, symptoms in the opposite knee. Time from the onset of symptoms must be determined.

Physical Examination

Physical examination starts by looking at how the patient rises from their chair and walks into the physician’s office. This information is interesting because the patient is acting naturally. In the office, evaluation starts with the patient standing. Asymmetry at the level of the shoulders or pelvis is noted when the patient is standing and sitting. The patellae should face forward. If femoral anteversion is present, they will face inward. Coronal and rotational alignments are observed. Genu valgum, tibial torsion, and limb discrepancy are also noted at this moment. The overall extensor mechanism alignment is checked, and the Q angle can be already estimated. From the back of the patient, subtalar eversion, which would produce compensatory internal tibial rotation, should be looked for.

The same observations done when the patient was standing should be done when he is walking. Rotational deformities are especially exacerbated during gait. Muscle hypotrophy can be noted. Any limp will become evident.

With the patient supine and the hips and knees extended, the Q angle measurement can be done effectively. Care should be taken because a laterally displaced patella will cause underestimation of its value. Knee flexion can correct this by bringing the patella into the center of the trochlear groove, but no agreement has been reached on the best Q angle measurement method (flexion or extension), or even on its applicability.⁹⁹ Normal individuals, in general, will not present with values greater than 20 degrees.

Asking the patient to contract both quadricepses will allow comparison of the contraction pattern and the muscular mass. The VMO bulk should also be noted at this moment. Active and passive movements of flexion and extension of the knee will allow patellar tracking assessment. The J sign, seen when the patella shifts abruptly medially, and then down in the trochlear groove as flexion progresses, similar to an inverted J, is sometimes found, meaning patellar lateral displacement in extension. Palpation and specific tests are then performed:

Palpation: Palpation should start in the less painful areas. The patella, patellar tendon, quadriceps tendon, and lateral and medial retinacula should be investigated for tenderness. The patellar facets may even be palpated and inspected for tenderness, but it should be noted that some structures (retinacula) are interposed, and this can confuse interpretation of the pain source. The distal-to-proximal position of the patella can be assessed by observation and palpation.

Glide test: Patellar medial and lateral glide should be tested at full extension and at 30 degrees of flexion. For this purpose, one can divide the patella into four vertical quadrants. Displacement of less than 1 quadrant or more than 3 is considered abnormal (Fig. 61-5).

Tilt test: The lateral and medial margins of the patella should be in the same horizontal plane, and its transverse axis should be elevated beyond the horizontal plane. Significant tilt in the physical examination correlates with magnetic resonance imaging (MRI)⁴⁶ or computed tomography (CT) scan tilt measures greater than 10 degrees. Anterior palpation should be done during the arc of movement in the search for pain and crepitus. The degree of flexion at the time the signals or symptoms are produced should be noted.

Smillie test⁹⁸ or apprehension test: This is performed with the knee extended. The examiner grasps the patella with his fingers and applies a laterally directed force, trying to dislocate it while holding the tibia with the other hand. This laterally directed force applied over the patella causes apprehension in the patient as he feels that the patella is about to dislocate. It is also called the apprehension test because it is the patient’s positive reaction that will determine test positivity. For adequate examination, the quadriceps should be relaxed. The test should be performed bilaterally, and comparison with the opposite side may help. It is not useful in acute dislocations because pain and fear will be present even before the physical examination. In chronic cases, it reflects well the insufficiency of the patellar restraints (notably the MPFL) (Fig. 61-6).

Muscle stiffness: Hamstrings are tested with the patient supine and the opposite leg extended and flat over the table. The hip is flexed 90 degrees, and the knee is then extended as far as possible. The popliteal angle is observed and compared with the opposite side. The quadriceps is tested with the patient prone, and in most patients the heels should touch the buttocks. The ITB is tested with the Ober test: with the patient in lateral decubitus position (the side to be tested is up), the hip is extended and abducted, and the knee is extended. From this position, the thigh is released and is allowed to adduct. Most patients will be able to touch the examination table with the medial aspect of the knee.

Figure 61-5 Clinical evaluation of mediolateral displacement of the patella. It should be tested at 0 degrees and 30 degrees of knee flexion and recorded in millimeters or quadrants.

Figure 61-6 The Smillie or apprehension test. The physician displaces the patella laterally, and the patient shows apprehension and fear of dislocation. Test positivity is determined by the patient’s reaction, not by the physician’s ability to dislocate or sublux the patella.

Imaging

X-ray Analysis

X-ray analysis is the first step before any other investigation of the knee is undertaken. Combined with the history and physical examination findings, it will guide and allow subsequent imaging procedures, and at times will even make them unnecessary. The basic protocol is almost uniform to the various situations, chronic or urgent. The only necessary condition is that the patient is able to stand up on the affected limb. Basic standard x-rays are described in the following paragraphs.

AP View

The anteroposterior (AP) view must be done in monopodal stance as long as the patient is able to do so. In younger patients (before 50 years old), it should be performed in 15 to 20 degrees of flexion; in older patients and in those who have antecedents of knee trauma or surgery, it should be done in 30 or 45 degrees of knee flexion (Schuss or Rosenberg). The AP analysis is not really helpful for patellofemoral problems. It will allow bone quality analysis, alignment assessment, and evaluation of femorotibial-associated pathology or arthritis. Gross patellar instability or displacement (Fig. 61-7) can also be observed in this view, along with malformations such as bipartite patella or fracture. A loose body in the lateral gutter may be found, representing a lateral condyle fracture that occurred during patellar dislocation.

Figure 61-7 In rare cases, patellar tilt and subluxation are so pronounced that they can be evidenced even on anteroposterior (AP) views. In this case, note the high and tilted patella.

In the AP view, we can observe a bipartite or a multipartite patella. It is the result of incomplete fusion of an ossification center and is described to have a frequency between 0.005% and 1.66%.^10,¹⁰² The accessory fragments commonly are located close to the superolateral border. The edges of the fragment are smooth, and this allows differentiation from fracture. Bipartite or multipartite patellae are often bilateral.

Lateral View

This is the most interesting view of the knee. The reliability of its interpretation depends on the technical quality of the image. It is essential to have a perfect superimposition of the two posterior condyles. The image is done in monopodal weight bearing with an angle of flexion between 15 and 20 degrees. Some authors propose to take the lateral view in full extension, but its accuracy in determining patellar height is controversial because different degrees of quadriceps contraction could modify the patellar height; also, if the patient has knee hyperextension, this could yield a false-positive patella alta and false information about patellar engagement. Nevertheless, the location of the patella in relation to the trochlea provides interesting data.

X-ray analysis has to be systematic and should follow the guidelines provided in the following sections.

Trochlea

In a normal knee, Blumensaat’s line is continued anteriorly by the trochlear groove line, which should stay posterior to the projection of the femoral condyles (facets). In 1987, Henri Dejour described the crossing sign, which characterizes trochlear dysplasia on the sagittal view. The crossing point represents the exact location where the deepest point of the trochlear sulcus reaches the same height as the femoral condyles, meaning that the trochlea becomes flat in this location (Fig. 61-8).

Figure 61-8 In a normal trochlea, the trochlear groove is posterior to the facets (A). In a flat trochlea, the facets and the trochlear groove are in the same plane. In the lateral view, this is demonstrated by the crossing sign (B). When the trochlea is convex, the “groove” does not exist, and the lateral part of the trochlea rests in front of the medial facet, which can be demonstrated in lateral view as the double-contour sign (C).

The position of the trochlear sulcus line is abnormal in relation to the anterior femoral cortex. In a study performed by Dejour and associates,³⁰ in normal knees the trochlear sulcus line was at a mean distance of 0.8 mm posterior to a line projected from the anterior femoral cortex, and in those knees with dysplastic trochleae its mean position was 3.2 mm forward to the same line. This increases the contact force between the patella and the trochlea (anti-Maquet effect) (Fig. 61-9).

Figure 61-9 Trochlear groove position and depth in relation to the anterior femoral cortex. A line (X) is drawn tangential to the anterior femoral cortex along its most distal 10 cm, extending below the articular surface. The trochlear sulcus is anterior to this line (left figure). The trochlear depth is the distance A/B (measured in millimeters) along a line subtended 15 degrees from the perpendicular to the tangent of the posterior femoral cortex (right figure).

The crossing sign has been found in 96% of the population with antecedents of true patellar dislocation, and in only 3% of healthy controls.²⁹ The first published classification (Henri Dejour) divided dysplasia into three grades, according to the level of the crossing sign. Other noted signs included the following: (1) the deepness of the trochlea as measured by a line that traced 15 degrees from another one perpendicular to the femoral shaft and tangential to the posterior condyles (see Fig. 61-9), and (2) the “bump,” which was defined as the distance between a line drawn tangential to the anterior femoral cortex and the highest point of the trochlea.

The classification in three grades has some limitations, as was corroborated by the work of Remy and colleagues.^89,⁹⁰ They showed that interobserver reproducibility of trochlear analysis was low, especially for type II dysplasia. This led to a new study performed in 1996 by Dejour and Le Coultre, which analyzed 177 cases of patellar instability and included radiographs along with preoperative and postoperative CT scans. Based on this analysis, a new and more precise classification with four grades of trochlear dysplasia was defined.^26,²⁸ Two new signs were added to the crossing sign. The first is the supratrochlear spur, which represents a global prominence of the trochlea and plays a role similar to a ski jump when the patella engages the trochlea. The second sign is the double contour, which is the radiographic line that ends below the crossing sign and represents the subchondral condensation of the hypoplastic medial facet on the lateral view. In 2002, the Lillois group conducted a new interobserver study⁹¹ and concluded that “this new classification system is more reproducible than the former 3-type system proposed. The crossing sign and the supratrochlear spur are the most reproducible signs” (Fig. 61-10).

Figure 61-10 For analysis of trochlear dysplasia, a true profile is needed with a perfect superimposition of the posterior femoral condyles. The three trochlear dysplasia signs are the following: the crossing sign; the supratrochlear spur; and the double-contour sign, which goes below the crossing sign.

This classification system (Figs. 61-11 and 61-12) is based mainly in the lateral view, although CT may assist in differentiation between types. Four types, based on the three dysplastic signs described, are included:

• Type A: presence of crossing sign in the true lateral view. The trochlea is shallower than normal but is still symmetrical and concave.

• Type B: crossing sign and trochlear spur. The trochlea is flat in axial images. All of the trochlea is prominent.

• Type C: presence of crossing sign and the double-contour sign on the lateral view. No prominence is seen, and in axial views, the lateral facet is convex and the medial facet hypoplastic.

• Type D: combines all previously mentioned signs, which are the crossing sign, supratrochlear spur, and double-contour sign. In the axial view, clear asymmetry of the height of the facets, also referred to as a cliff pattern, is evident.

Figure 61-11 Trochlear dysplasia classification according to Dejour. Type A: crossing sign in lateral true view is present. The trochlea is shallower than normal but is still symmetrical and concave. Type B: crossing sign and trochlear spur. The trochlea is flat or convex on axial images. Type C: the crossing sign and the double-contour sign (the second representing the densification of the subchondral bone of the medial hypoplastic facet). On axial computed tomography (CT) scan views, the lateral facet is convex. Type D: all mentioned signs are combined, that is, crossing sign, supratrochlear spur, and double-contour sign that goes below the crossing sign. On axial CT scan views, a cliff pattern is seen.

Figure 61-12 Trochlear dysplasia assessed in lateral views. The crossing sign, the supratrochlear spur, and the double-contour sign should be assessed to classify it. Type A: only crossing sign is present. Type B: crossing sign and trochlear spur can be seen. Type C: crossing sign and double-contour sign representing the medial hypoplastic facet are present. Type D: all the mentioned signs are combined: crossing sign, supratrochlear spur, and double-contour sign.

Patella

Grelsamer et al ⁴⁴ did a study describing three types of patella, based on the ratio between the length of the patella and the length of the articular surface. Most patellae exhibit a ratio between 1.2 and 1.5 and are classified as type I. Those with a ratio greater than 1.5 give the appearance of having a long nose; this is type II. Those with a ratio less than 1.2 (short nose) are type III (Fig. 61-13).

Figure 61-13 Patellar shape on lateral views according to Grelsamer. The ratio (A/B) between the length of the patella (A) and the length of its articular surface (B) is calculated. Most patellae exhibit ratios between 1.2 and 1.5.

The shape of the patella on the lateral view is correlated with the tilt and with the global morphology of the patella. In a normal patella, with no tilt, the most posterior part visible in the lateral view should be the median longitudinal ridge. The lateral facet projection is located slightly anterior. In tilted patellae, these relations are lost, and the overall anteroposterior size of the patella appears increased (Fig. 61-14).

Figure 61-14 Lateral view of severe patellar tilt and subluxation. Note that the projection is a perfect lateral view (perfect posterior condyle superimposition). The patella, however, is not in front of the femur but it is lateral to it, and the patellar anteroposterior diameter is increased. Additionally, severe trochlear dysplasia can be noted.

The tilt evaluation has been described by Maldague and Malghem.⁶⁹ Three positions are described: normal position, in which the lateral facet is in front of the crest; mild tilt, in which the two lines (lateral facet and crest) are on the same level; and severe tilt, which shows the lateral facet behind the crest (Fig. 61-15).

Figure 61-15 Tilt evaluation in lateral views as described by Maldague and Malghem. In the normal position, the lateral facet is in front of the crest. When mild tilt is present, the lateral facet and the crest are on the same level. When severe tilt occurs, the lateral facet is behind the crest.

Patellar Height

Patella alta or infera is essentially diagnosed on the lateral view. Patellar height must be measured using an identified index. The main indexes used in the literature are listed here:

• The Caton-Deschamps index ^14,¹⁵: ratio between the distance from the lower edge of the patellar articular surface to the anterosuperior angle of the tibia outline (AT) and the length of the articular surface of the patella (AP). A ratio (AT/AP) of 0.6 or smaller reveals patella infera, and a ratio greater than 1.2 indicates patella alta (Fig. 61-16).

• The Insall-Salvati index ⁵⁶: ratio between the length of the patellar tendon (LT) and the longest sagittal diameter of the patella (LP). Insall determined that this ratio (LT/LP) is normally 1. A ratio smaller than 0.8 indicates patella infera, and greater than 1.2 patella alta (Fig. 61-17).

• The Blackburne-Peel index ⁹: ratio between the length of the perpendicular line drawn from the tangent to the tibial plateau to the inferior pole of the articular surface of the patella (A) and the length of the articular surface of the patella (B). The normal ratio (A/B) was defined as 0.8. In patella infera it is smaller than 0.5, and in patella alta it is greater than 1.0 (Fig. 61-18).

Figure 61-16 The Caton-Deschamps index (AT/AP) is the ratio between the distance from the lower edge of the patella’s articular surface to the anterosuperior angle of the tibia outline (AT) and the length of the articular surface of the patella (AP).

Figure 61-17 The Insall-Salvati index (LT/LP) is the ratio between the length of the patellar tendon (LT) and the longest sagittal diameter of the patella (LP).

Figure 61-18 The Blackburne-Peel index (A/B) is the ratio between the length of the perpendicular line drawn from the tangent to the tibial plateau until the inferior pole of the articular surface of the patella (A) and the length of the articular surface of the patella (B).

Several factors must be considered when a decision is made regarding which index to use. Blackburne-Peel needs good superimposition of medial and lateral tibial plateaus. Insall-Salvati is not a good choice in the presence of Osgood-Schlatter disease or sequelae of it. Finally, the Caton-Deschamps method seems the easiest to use, especially for surgical planning.

Axial View

The axial view has been described at different angles of knee flexion and different positions of the x-ray cassette. Our common approach is to perform 30-degree axial views as described by Ficat, who also described axial views at 60 degrees and 90 degrees (Fig. 61-19). Radiographs are obtained with the knee flexed over the edge of the table, the beam directed proximally, and a perpendicular cassette in place. Images beyond 45 degrees of flexion, however, are less informative as they show the lower part of the trochlea and the patella as fully engaged, many times correcting tracking abnormalities (Fig. 61-20). These high–flexion angle images are not necessary.²³ Lower flexion angles, although capable of showing better the maltracking signs, are technically demanding and at times impossible. With a well done image, one can assess the relation between the femoral trochlea (at 30 degrees, the lateral facet should appear with two-thirds total trochlear width) (Fig. 61-21) and the patella (with the lateral facet also composing two thirds). Tilt, congruence, and cartilage thickness can also be appreciated.

Figure 61-19 Axial views performed at 30, 60, and 90 degrees of knee flexion. Note how the trochlear shape changes as flexion increases and the medial facet seems bigger. It is not necessary to routinely perform 60-degree and 90-degree flexion x-rays in common patellofemoral (PF) disorders.

Figure 61-20 Sagittal model demonstrating how, in the same patient, different images can fail to show the abnormal trochlear shape. Proximally (A), the trochlea is flat, but it deepens distally (B and C). For this reason, it is important to have a full computed tomography (CT) scan acquisition of the trochlea. Axial views done with angles superior to 30 degrees could miss the proximal trochlear dysplasia.

Figure 61-21 Axial view of the patellofemoral articulation at 30 degrees of knee flexion. The medial facet appears smaller and corresponds to one third of the trochlear width, while the lateral facet corresponds to two thirds of it. The patella is centered, and its long axis is horizontal, meaning that no tilt is present.

Alternative methods have been proposed. The main choices are discussed in the following sections.

Merchant View ⁷⁶

The Merchant view is obtained with the patient in the supine position and the knees flexed at 45 degrees over the edge of the table. The lower limbs rest on an angled platform. The x-ray beam is angled toward the feet, 30 degrees from horizontal, and the film cassette is positioned 30 cm below the knees. The x-ray beam strikes the cassette at a 90-degree angle, imaging both knees simultaneously. Two angles are measured on this view: the sulcus angle and the congruence angle.

• The sulcus angle (Fig. 61-22) (defined by Brattström) is the angle formed by two lines drawn from the deepest point of the trochlear groove to the highest point on the medial and lateral femoral condyles. This measurement reveals the shape of the groove; the greater the sulcus angle, the flatter is the trochlea. The average sulcus angle on the merchant view measures 138 degrees (standard deviation [SD] ± 6) and is equal in males and females. Values superior to 150 degrees are considered abnormal.

• The congruence angle (Fig. 61-23) is measured by bisecting the sulcus angle to construct a reference line, and then projecting a second line from the apex of the sulcus angle to the lower point of the subchondral articular surface of the patella (apex). If the line drawn from the patellar apex is lateral to the reference line, then the angle is positive; if the patellar apex is medial to this line, a negative value is assigned; measurement of the normal congruence angle averages −6 degrees (SD ± 11 degrees; abnormal if greater than +16 degrees).

Figure 61-22 The sulcus angle defined by Brattström and popularized by Merchant is the angle formed by the two lines drawn from the deepest point of the trochlear groove to the highest point on the medial and lateral femoral condyles.

Figure 61-23 The congruence angle is measured by bisecting the sulcus angle to construct a reference line and then projecting a second line from the apex of the sulcus angle to the lower point of the subchondral articular surface of the patella (apex). If the line drawn from the patellar apex is lateral to the reference line, then the angle is positive; if it is medial to this line, a negative value is assigned.

Laurin View ^64,⁶⁵

This image is obtained with the patient sitting and the knee flexed 20 degrees. The x-ray cassette is held approximately 12 cm proximal to the patellae and is pushed against the anterior thighs. The x-ray beam is directed cephalic and 20 degrees superior from the horizontal point. Two measurements are made on this view: the lateral patellofemoral angle and the patellofemoral index.

• The lateral patellofemoral angle (LPFA) (Fig. 61-24) is formed when one line connects the superior points of the medial and lateral trochlear facets and a second line is drawn tangent to the lateral facet of the patella. It measures tilt and subluxation and should open laterally in normal knees (97% open laterally and 3% are parallel; no LPFA opening medially was found in normal knees in Laurin’s study).

• The patellofemoral index (Fig. 61-25) is the ratio (M/L) between the thickness of the medial joint space (M) and that of the lateral joint space (L). It should measure 1.6 or less (Fig. 61-26).

Figure 61-24 The lateral patellofemoral angle (LPFA) is formed by one line connecting the superior points of medial and lateral trochlear facets and a second line tangent to the lateral facet of the patella.

Figure 61-25 The patellofemoral index is the ratio (M/L) between the thickness of the medial joint space (M) and the lateral joint space (L).

Figure 61-26 A-C, Radiologic differentiation between tilt and subluxation. A, Subluxation can be detected using the congruence angle (the normal range is demonstrated in the figure). B, The congruence angle is within the normal range, but tilt is noted as the patellar transverse axis is lifted off from the horizontal. Another possible method of assessment could be the lateral patellofemoral angle (on computed tomography [CT] scans, the posterior femoral condyles could be used as reference). C, Both tilt and subluxation are present.

Malghem and Maldague Lateral Rotation View (30 Degrees LR)⁷⁰

This view is obtained in 30 degrees of knee flexion while one examiner pulls the forefoot laterally. The cassette is held over the patient’s thighs, and the x-ray beam is directed cranially. Patellar position (centered or subluxated) is defined according to Merchant’s congruence angle. In the authors’ series, the 30-degree lateral rotation (LR) view was superior to standard 45-degree axial views in detecting patellar subluxation. In 27 knees operated on for patellar instability, 45-degree routine views depicted subluxation in only seven cases, and 30-degree LR views demonstrated it in all cases. Additionally, when both views showed signs of instability, the degree of subluxation was greater in the 30-degree LR view.

In acute or chronic patellofemoral instability, medial patellar avulsions (Figs. 61-27 and 61-28) can be demonstrated and should not be confused with bipartite patella. Other important data provided by axial views include patellar shape and joint line thickness. Patellar shape is evaluated according to Wiberg’s classification (Fig. 61-29). Joint space thickness is diminished in arthritis. Axial views at 30 degrees allow assessment of which side of the articulation is affected (usually the lateral side). Information on the size of osteophytes and on joint line narrowing is provided. Iwano and coworkers⁵⁷ used the following simple staging system of lateral patellofemoral osteoarthritis (OA) (Fig. 61-30):

• Stage I, mild OA: joint space measures at least 3 mm.

• Stage II, moderate OA: joint space measures less than 3 mm, with no bony contact.

• Stage III, severe OA: bony contact in less than one quarter of the joint surface.

• Stage IV, very severe OA: joint surfaces entirely touch each other.