Lateral Radiographs Of The Knee

Recently, several studies have described the limitations of the axial view for patellar alignment evaluation. Walker and colleagues [44] questioned the value of the axial view and compared the axial view with CT of the patellofemoral joint. They found that these two modalities often give conflicting results and concluded the axial view to be of limited value due to lack of sensitivity, as ''even florid examples [of maltracking] must be missed.'' Many researchers now advocate using the lateral view of the knee, in various degrees of flexion, to evaluate for alignment at the patellofemoral joint. The basis of this argument rests on the fact that in most knees, the patella is fully engaged in the trochlea by 30 degrees of flexion. Many patients with mild subluxation or tilt are not diagnosed on axial images as their abnormally aligned patellae have corrected at the angle the axial images are obtained. It is not possible to obtain an axial image at less than 20 degrees of flexion; special equipment and elaborate technique is required to obtain these images at angles less than 30 degrees.

Maldague and Malghem [45] have shown how the lateral radiograph can be used to evaluate for malposition of the patella. Dupont and Guier [46] added to their grading scheme (Fig. 14). There is a technical advantage to this technique. A true lateral view can be obtained in full extension or in varying degrees of

Maldague Knee

Fig. 14. Schematic representation of patellar position on lateral views, according to the degree of external rotation (tilt), modified Maldague-Malghem's classification in 4 stages. (A) Stage 1 (normal position): both lines (central ridge, CR; lateral edge, LE) are concave posteriorly and separated by 5 to 10 mm. The anterior line is the lateral edge, the posterior line the central ridge. (B) Stage 2 (false lateral profile as described by Maldague and Malghem, or minor lateral patellar subluxation): both lines are superimposed. Only one straight line is visible. (C) Stage 3 (overwhelming lateral profile, or pronounced lateral subluxation): the anterior line is the central ridge, the posterior line is the lateral aspect of the patella and is convex. (D) Stage 4. The central ridge is no longer visible. The patella often covers the anterior cortex of the femoral metaphysis and appears ovoid in shape [46]. (From Dupont JY, Guier CA. Comparison of three standard radiologic techniques for screening of patellar subluxations. Clin Sports Med 2002;21(3):389-401; with permission.)

Fig. 14. Schematic representation of patellar position on lateral views, according to the degree of external rotation (tilt), modified Maldague-Malghem's classification in 4 stages. (A) Stage 1 (normal position): both lines (central ridge, CR; lateral edge, LE) are concave posteriorly and separated by 5 to 10 mm. The anterior line is the lateral edge, the posterior line the central ridge. (B) Stage 2 (false lateral profile as described by Maldague and Malghem, or minor lateral patellar subluxation): both lines are superimposed. Only one straight line is visible. (C) Stage 3 (overwhelming lateral profile, or pronounced lateral subluxation): the anterior line is the central ridge, the posterior line is the lateral aspect of the patella and is convex. (D) Stage 4. The central ridge is no longer visible. The patella often covers the anterior cortex of the femoral metaphysis and appears ovoid in shape [46]. (From Dupont JY, Guier CA. Comparison of three standard radiologic techniques for screening of patellar subluxations. Clin Sports Med 2002;21(3):389-401; with permission.)

flexion to the point where the patella engages the trochlea. Murray and colleagues [47] compared lateral radiographs (taken in 15 to 30 degrees of flexion) of the knee with axial views and found the lateral films had increased sensitivity and specificity for correlation to patellofemoral pain, previous dislocation, and malalignment. Similar results were obtained by Dupont and Guier [46].

The lateral radiograph can also allow assessment of the condition of the trochlea and patellar height. The assessment of patella height is important as patella alta is associated with an increased risk for dislocation. A simple method is the modified Insall-Salvati method, which takes into account the wide variety of patellar shapes that are encountered [48]. This method is highly reproducible and does not rely on exact positioning. The patellar articular surface is compared with the length of the patellar tendon from the tibial tubercle to the most inferior aspect of the patellar articular surface. This ratio of patellar tendon/patellar length is measured and a normal value is less than 2 (Fig. 15). Trochlear depth, accurately measured by the lateral knee radiograph, is an important factor in patellar mala-lignment. A shallow trochlea places one at risk for dislocation and subluxation. Malghem and Maldague [49] demonstrated that a depth in the proximal trochlea of less than 5 mm increases the risk of instability.

Fig. 15. Modified Insall-Salvati method to determine patellar height. The ratio of B/A should normally be less than 2.

CT and MR

MR and CT can both be used to study the patellofemoral articulation and each has intrinsic advantages and disadvantages. Both modalities have the advantage of being able to study the knee directly in degrees of flexion less than 30 degrees, without overlapping structures adding confusion. A static study involves acquiring axial images at a fixed degree of flexion, usually less than 20 degrees. Kinematic studies acquire images while the knee moves from flexion to extension, and the images acquired can be viewed in a cine mode. These modalities may be useful in patients with symptoms that suggest malalignment but without evidence of the diagnosis by radiographs or physical exam [50,51].

Schutzer and colleagues [37] in 1986 evaluated the patellofemoral joint with CT. They described three malalignment patterns. These include subluxation without tilt, subluxation with tilt, and tilt without subluxation. Their study demonstrated that in asymptomatic normal controls, the patella is either centered or slightly medially displaced by 10 degrees of flexion. Therefore, a patella is considered subluxed if the congruence angle is greater than 0 with the knee in 10 degrees or more of flexion. The patellar tilt angle in asymptomatic controls always was greater than 8 degrees and usually was more than 15 degrees. The authors conclude that 8 degrees is the lower limit of normal tilt on CT [37].

Static MR techniques, such as the one described by Koskinen and colleagues [52], can be used to obtain accurate traditional patellofemoral indices such as lateral patellar tilt, lateral patellofemoral angle, lateral patellar displacement, sulcus angle, and congruence angle. The advantages of static MR over CT include its ability to characterize the status of the soft tissues of the knee. Patellar height can also be assessed reliably with MR. Miller and colleagues [53] compared sagittal MR images and lateral radiographs, and found good correlation between the two modalities. They compared patellar tendon length to patellar length ratios of lateral knee radiographs and midline sagittal MR images. CT sagittal reconstructions theoretically could provide similar information. Pfirr-mann and colleagues [54] compared MR and lateral knee radiographs to demonstrate that MR is a reliable test for trochlear dysplasia.

Kinematic MR and CT

The goal of kinematic studies is to better evaluate patellar tracking and to image the PF joint in a manner that more closely mimics physiologic conditions. Advances in fast imaging technology have allowed the development of practical kinematic techniques [50]. Earlier kinematic studies were performed with passive movement of the knee while images were obtained at different degrees of flexion. More recently, active movement kinematic studies let the patient move the knee through quadriceps contraction while being imaged [55]. Active contraction kinematic studies can identify malalignment and maltracking in patients who otherwise would appear normal. Shellock and colleagues [56] demonstrated the advantages of stressing the patellofemoral joint with ''loaded'' kinematic MR studies. This was accomplished by the patient performing quadriceps contraction against resistance supplied by weights. Their study showed improved ability to identify alignment abnormalities compared with unloaded active kinematic exams.

McNally and colleagues [51] described a useful loaded kinematic MR technique in their study comparing static and kinematic MR. With this technique, the patient was placed supine in the magnet with the knees strapped loosely together at about 30 degrees of flexion. Images were obtained while the patient extended the knee against a balloon, which provided resistance (loading). Multiple fast gradient echo sequences were acquired while extension took place. The imaging took about 2 minutes, with the balloon controlling the rate of extension. The axial slices closest to the center of the patella were selected and compiled for a cine loop. This loop was viewed on a PACS station and subjectively quantified. In the study, the authors rated the observed subluxation (as defined by the lateral movement of the patella in relation to the trochlea) as grade 1 (mild), grade 2 (moderate), or grade 3 (severe). They then compared the given grade of subluxation to multiple patellofemoral measurements, including femoral sulcus angle, sulcus depth, lateral patellar angle, patellar lateralization, and patella-patellar tendon ratio near extension. Their study showed that with increasing grade of mal-tracking there was a progressive worsening of the above patellofemoral indices, suggesting their simple subjective grading method's utility.

In a separate study, O'Donnell and colleagues [57] compared tracking patterns in 50 patients with anterior knee pain to 50 asymptomatic controls using the protocol described by McNally and coworkers. They demonstrated that increasing degrees of patellar lateralization relate to increasing severity of symptoms in patients. They also showed that many normal controls show mild lateralization near full extension, and thus conclude that this phenomenon is likely a normal variant rather than pathologic.

Fulkerson and colleagues [8] described a technique for unloaded (non-weighted) kinematic CT of the patellofemoral joint obtained at various degrees of flexion. This technique allowed for the evaluation of tilt and subluxation at various degrees of flexion. Tilt was measured by comparing the posterior con-dylar line to the lateral patellar facet. Congruence angle was used to measure subluxation. The examination was performed in about 20 minutes and at a cost similar to standard knee radiography. The values obtained were compared with the guidelines set forth by Schutzer and colleagues [37].

Competing Viewpoints

Dye [58] has challenged the idea that malalignment (without subluxation) by itself causes patellofemoral pain, and has questioned some of the measurements that have been traditionally used to evaluate patients with anterior knee pain. In his study there was no statistical difference in Qangle and congruence angle in patients with patellofemoral pain and asymptomatic controls. He also concluded that osseous landmarks on radiography often do not match the contour of the underlying cartilage, which challenges the concept of patellar tilt. Ninety percent of patients with a diagnosis of malalignment improve with conservative treatment even though their malalignment is not surgically addressed. Many patients with bilateral patellar tilt have symptoms in only one knee. Therefore, Dye asserts that loss of tissue homeastasis, which is the root cause of patellofe-moral pain, is likely due to ''supraphysiologic'' overload [58]. Other authors, such as Thomee and colleagues, have also pointed out problems with the theory of malalignment as the cause of anterior knee pain [43,59].

Was this article helpful?

0 0

Post a comment