Patellofemoral Problems in Total Knee Arthroplasty
Nicholas B. Frisch, MD, MBA
Richard A. Berger, MD
Patellofemoral complications are among the most common postoperative problem associated with the current design of total knee prostheses and are a major cause of revision surgery.1,2,3,4,5,6,7,8,9 Patellofemoral complications affect up to 12% of all total knee replacements done in the United States. Although there is evidence that patellofemoral complications have diminished somewhat over the past decade,10 they are still a common reason for pain and dysfunction after total knee arthroplasty.11,12 In some series, up to 50% of the revisions performed were for a problem related to the patellofemoral articulation,6 and those that were revised for patellofemoral complications have a high risk of reoperation.13
This chapter addresses patellofemoral complications from two perspectives. First, they are considered as they relate to primary total knee arthroplasty, to enable the surgeon to optimize patellofemoral tracking and minimize complications. The various aspects of primary total knee arthroplasty that affect the patellofemoral joint are covered so that these common complications may be avoided. In the second half of this chapter, the diagnosis and management of patellofemoral complications after total knee arthroplasties are addressed.
PATELLOFEMORAL JOINT IN PRIMARY TOTAL KNEE ARTHROPLASTY
Patellofemoral complications are the most common complications in total knee arthroplasty that can be avoided. Patellar tilt is a common radiographic finding and can range from minimal to dramatic. When patellar tilt is associated with a metal-backed patellar component, this can lead to eccentric wear and delamination with the possibility of eventual scoring of the femoral component, which necessitates complete revision surgery (Fig. 59-1). As the problems of the patellofemoral joint progress, subluxation and frank dislocation can occur. Patellar maltracking leads to shear forces over time that can result in either debonding of the patellar prosthesis from the native patella or shearing off of the prosthesis from the fixation lugs (Fig. 59-2). These shear forces on the patella over time can also lead to patella fragmentation or fracture.
The Tight Lateral Retinaculum
It has been the conventional wisdom that patellofemoral maltracking is secondary to a tight lateral retinaculum and, therefore, that the solution to this problem is a lateral retinacular release. Once a lateral retinacular release is performed, however, patellar tracking usually does not improve substantially. A few well-placed towel clips, however, can lead to the false conclusion that the patella is tracking appropriately. In these cases, it is often found postoperatively that the patella is not tracking well, and with time, patellofemoral maltracking and failure occur.
In most total knee arthroplasties with patellofemoral maltracking, a tight lateral retinaculum is not the sole problem; it is really only a symptom indicating that, in some way, the patellofemoral articulation has been altered. Occasionally, such as in a knee with a valgus deformity, real lateral retinacular tightness is identified either preoperatively (Fig. 59-3) or intraoperatively after the arthrotomy has been performed. In these cases, the solution is a lateral retinacular release, which allows the patella to track in a more anatomic position in the trochlear groove. The majority of knees that require total knee arthroplasty are associated with a varus deformity, however, and in the majority of these cases, the patella tracks quite well before arthroplasty. Furthermore, implant designs have evolved to create a more anatomic patellofemoral relationship to improve tracking and minimize patellofemoral complications.14,15 Thus, if the patellar trochlear relationship is recreated, appropriate patellar tracking will result. If that relationship is altered, however poor tracking ensues and manifests as lateral patellar tracking, subluxation, or frank dislocation. Thus, a tight lateral retinaculum is rarely the problem; it is only a symptom that intraoperatively the patellar trochlear relationship has been altered.
Resection of the Patellofemoral Joint
As with the tibiofemoral articulation, the surgeon should take care in resecting the patellofemoral joint. In general, the resected patella should be restored millimeter for millimeter with polyethylene.16,17,18 Thus, the amount of bone resected should precisely equal the thickness of the patellar component to be implanted so that the final thickness
of the patella-prosthesis composite is equivalent to the thickness of the native patella (Fig. 59-4). This is accomplished by first measuring the thickness of the native patella and then cutting it in the coronal plane, removing an amount of patellar bone that is equivalent to the thickness of the component to be inserted. The amount of bone resected is typically between 8 and 10 mm, which should correspond to the thickness of the patella component used. In general, at least 12 mm of patella should be left after resection.
of the patella-prosthesis composite is equivalent to the thickness of the native patella (Fig. 59-4). This is accomplished by first measuring the thickness of the native patella and then cutting it in the coronal plane, removing an amount of patellar bone that is equivalent to the thickness of the component to be inserted. The amount of bone resected is typically between 8 and 10 mm, which should correspond to the thickness of the patella component used. In general, at least 12 mm of patella should be left after resection.
It is important to resect the patella parallel to the coronal plane so that the patellar prosthesis is not tilted on the native patella. This can be accomplished by resecting some of the synovium around the patella so that the patellar tendon is exposed and the cut parallel to it. Booth et al19 have described the patellar nose as also a useful landmark for resecting the patella. A useful rule of thumb is that the thickness of the final patella-prosthesis composite should be on the order of 21 to 23 mm for smaller patients and 24 to 26 mm for larger patients. Thought and care should be taken when the combined thickness of the native patella and the patellar prosthesis is outside this range. There are several different techniques for patellar resurfacing,20 but whether a patellar clamp, a patellar reamer, or a plain saw is used, it is important to adhere to these principles. Achieving a composite thickness that closely approximates the native thickness is associated with greater improvement in quality of life, physical measures, and Western Ontario and McMaster Universities Arthritis Index stiffness scores.21
The consequences of not properly restoring the patellofemoral joint are significant.22,23,24,25 If patellar resection is excessive, so that the final construct of prosthesis plus native patella is significantly less than the native thickness, the quadriceps musculature is put at a substantial mechanical disadvantage, as the patella contributes approximately 30% to the extensor mechanism’s moment arm.26,27 In addition, it is important to note that small differences in the final height of the patella are quite important. For example, if the patella before arthroplasty
measures 25 mm and the final thickness of the resected patella plus the patellar implant is only 20 mm, the overall reduction in patellar thickness is 20%. Therefore, in this particular situation, 5 mm corresponds to a 20% reduction in patellar thickness, which will substantially affect quadriceps function.26
measures 25 mm and the final thickness of the resected patella plus the patellar implant is only 20 mm, the overall reduction in patellar thickness is 20%. Therefore, in this particular situation, 5 mm corresponds to a 20% reduction in patellar thickness, which will substantially affect quadriceps function.26
FIGURE 59-3 Preoperative anteroposterior (A) and patellar view (B) of a knee with a valgus deformity with lateral patellar tracking noted preoperatively. |
Overstuffing the patellofemoral joint, although it improves quadriceps function, increases the resultant patellofemoral force and lateralizes the Q-angle force; this results in a tightened lateral retinaculum and lateral maltracking.28 As the patellofemoral joint is overstuffed with either too thick a patella or an oversized femoral component, the lateral retinaculum is stretched.29 The result is lateral patellar tracking, subluxation, and potentially dislocation.23 In addition to a tight lateral retinaculum secondary to overstuffing of the patellofemoral joint, other patellofemoral problems occur, many of which are associated with an increased Q angle, which further raises the resultant force that tends to pull the patella laterally.24,28,30
The patellofemoral joint can also be overstuffed by failure to resect enough anterior femur.24 Most femoral components are 10 mm thick anteriorly, and in some patients, particularly women, the anterior trochlea that has been resected is not as thick as the component that replaces it (Fig. 59-5). In addition, the fear of notching the anterior femur has led many surgeons not to resect as much trochlea anteriorly as is being replaced by the component. This further leads to overstuffing of the patellofemoral joint. These problems can be avoided by carefully measuring the amount of bone resected from the trochlear groove and by ensuring that cutting guides are positioned appropriately.
Oversizing of the femoral component can also cause overstuffing of the patellofemoral joint.10,24 This is because, in flexion, the anteroposterior dimension of the femur contributes to the patellofemoral joint (Fig. 59-6). This problem is more commonly seen with posterior-stabilized than with cruciate-retaining total knee arthroplasties, secondary to the tendency to upsize the femoral component, which makes it larger than the native anteroposterior diameter of the femur. This also tightens the retinaculum and makes it more likely that patellofemoral tracking problems will occur. Additionally increasing the position of the femoral component anteriorly can result in an increased extensor mechanism arc and therefore result in decreased flexion.31
Many modern femoral components now have a deepened trochlear groove anteriorly, which prevents overstuffing of the joint (Fig. 59-7).15,32,33 It is important to remember, however, that this should be thought of as adjunctive treatment for the patellofemoral joint, and care still must be taken to not overstuff it.
Femoral Component
Femoral component design, size, rotation, and placement all play an important role in patellofemoral tracking.30,32,34,35 The trochlear groove, being part of the femoral component, is related to placement and rotation of the femoral component itself. With the advent of the measured resection technique36 in total knee arthroplasty,
the initial thought was to remove equal amounts of bone from the medial and lateral posterior condyles (Fig. 59-8). However, when equal amounts of posterior condyle are resected in combination with a nonanatomic tibial cut (removing more lateral tibial bone than medial tibial bone), the femoral component is internally rotated on the tibia, which increases the Q angle of the knee and leads to patellar maltracking.32,34,37 To avoid this problem, in the majority of knees, more bone needs to be resected from the posterior medial femoral condyle than from the posterolateral femoral condyle (Fig. 59-9). In many modern total knee systems, the surgical protocols recommend cutting the femoral component with some external rotation relative to the posterior condyles to optimize patellar tracking. Although external femoral rotation relative to the posterior condyles is helpful, a more accurate method of rotating the femoral component is to align the component parallel to the surgical epicondylar axis (Fig. 59-10).
the initial thought was to remove equal amounts of bone from the medial and lateral posterior condyles (Fig. 59-8). However, when equal amounts of posterior condyle are resected in combination with a nonanatomic tibial cut (removing more lateral tibial bone than medial tibial bone), the femoral component is internally rotated on the tibia, which increases the Q angle of the knee and leads to patellar maltracking.32,34,37 To avoid this problem, in the majority of knees, more bone needs to be resected from the posterior medial femoral condyle than from the posterolateral femoral condyle (Fig. 59-9). In many modern total knee systems, the surgical protocols recommend cutting the femoral component with some external rotation relative to the posterior condyles to optimize patellar tracking. Although external femoral rotation relative to the posterior condyles is helpful, a more accurate method of rotating the femoral component is to align the component parallel to the surgical epicondylar axis (Fig. 59-10).
The epicondylar axis can be found by identifying the lateral prominence of the lateral epicondyle and the medial sulcus of the medial epicondyle (Fig. 59-10).38,39 If the medial sulcus is difficult to identify, the entire medial condyle is easy to palpate and is in essence a large prominence. If the center of that large prominence is then identified, that also corresponds to the sulcus of the medial epicondyle.22 The medial sulcus is the insertion point of the deep medial collateral ligament, and overlying this is the fanlike insertion of the superficial medial collateral ligament. These landmarks can be seen and felt intraoperatively in the majority of cases and routinely used to confirm appropriate femoral bone resection and component rotation.
FIGURE 59-9 Appropriate resection of bone from the posterior femoral condyles to avoid internal rotation of the femoral component. The cut is parallel to the epicondylar axis, with more bone being removed form the posteromedial than from the posterolateral condyle. (From Krackow KA. The Technique of Total Knee Arthroplasty. St. Louis, MO: Mosby; 1990:131, with permission.)
Stay updated, free articles. Join our Telegram channelFull access? Get Clinical TreeGet Clinical Tree app for offline access |