Isolated patellofemoral joint (PFJ) arthritis affects ∼10% of the population over the age of 40, with a mean age of onset reported between 40 and 55 years old. There is a relative preponderance in women compared with men, which may be a result of the increased frequency of subtle PF dysplasia and malalignment commonly observed in women. Despite its incidence, PFJ arthritis is rarely debilitating or severely limiting, and isolated patellofemoral arthroplasty (PFA) represents only 1% of primary knee arthroplasty procedures. Additionally, this lower rate of PFA procedures, compared with what would be predicted based on disease prevalence, could be partially attributable to the mixed results observed with early-generation implants. Instead, many surgeons have likely utilized other surgical options, including arthroscopic debridement, patellar unloading procedures (e.g., tibial tubercle osteotomy), cartilage grafting and restoration techniques, and even patellectomy. However, these have had satisfactory results on the order of 60% to 70%. Total knee arthroplasty (TKA) has also been successfully used for the treatment of isolated PFJ arthritis. However, it represents a more extensive treatment option that may not be ideal or necessary for the young, active patient typically seeking treatment of PFJ arthritis.
Patellofemoral arthroplasty, when performed well, positioned well, and in the appropriately selected patient, ameliorates the symptoms of arthritis while retaining bone stock and the cruciate ligaments, thereby maintaining normal knee kinematics. In successful PFAs, functional outcomes have been remarkable, particularly compared with TKA. , Unfortunately, registry data have historically shown relatively high failure rates, with a cumulative percent revision over 5 years for any reason ranging from 8% to 18%, the majority (42%) related to progressive arthritis, suggesting poor patient selection. A study conducted by van der List et al. evaluating over 9000 PFAs yielded 5-, 10-, 15-, and 20-year PFA survivorship rates of 91.7%, 83.3%, 74.9%, and 66.6%, respectively, with an annual revision rate of 2.18%. Earlier studies and even contemporary analyses have found high rates of early failures related to patellar maltracking, often secondary to trochlear component malposition. While these problems have been mostly eliminated with onlay designs, which can be predictably positioned perpendicular to the femoral AP axis, many surgeons have shied away from PFA in favor of ongoing, often ineffective, nonoperative treatments or TKA. This chapter will discuss the common complications and pitfalls in PFA, and review strategies to minimize the risk of failures and their treatment options when they occur.
Appropriate patient selection is paramount to success in PFA. Failure to do so increases the risk of persistent pain, patient dissatisfaction, or an early “failure” requiring reoperation. There are several crucial aspects of a patient’s history that the surgeon should elicit when considering PFA. First, the patient’s pain should be triggered and exacerbated by activities that increase pressure and strain in the PFJ, including hyperflexion, kneeling, squatting, lunging, ambulating on stairs or hills, prolonged sitting with knee flexed and standing from a seated position. In contrast, patients should experience little if any pain when walking on level ground. If a patient has as much pain while walking on level ground as when climbing stairs, this should alert the surgeon that the individual would not benefit from PFA and may have more extensive intraarticular disease or pain originating from outside of the knee. Second, patients should describe their pain as predominantly anterior or anterolateral. Patients with isolated PFJ arthritis typically note retropatellar and/or peripatellar pain; a more diffuse pain pattern is not consistent with isolated PFJ arthritis. There are several risk factors for PFJ arthritis that should also be considered, such as a history of patellar dislocation/subluxation and direct anterior trauma to the knee. Physical examination should corroborate these historical elements, with reproducible pain with patellar compression and inhibition, anterior crepitus with moving between flexion and extension, and the absence of medial or lateral tibiofemoral joint line tenderness. Patellar tracking (or maltracking) should be assessed as well as the Q angle, as both may predispose the patient to PFJ arthritis and will need to be considered and addressed at the time of eventual PFA.
Careful analysis of imaging studies is also imperative, as the pattern of patellofemoral disease and absence of substantial tibiofemoral cartilage wear are critical determinants of success in PFA. Radiographs include weight-bearing anteroposterior (AP), weight-bearing mid-flexion posteroanterior (PA), lateral, and 30-degree patella skyline views to evaluate the extent of osteoarthritis (OA) in each compartment. , Radiographs should also be scrutinized for the presence of patella alta or baja and the presence of tilt or subluxation. Routine preoperative MRI is also recommended to ensure appropriate patient selection. The critical findings should include advanced patellofemoral chondromalacia involving the lateral patellar facet and/or lateral trochlea or diffuse patellofemoral arthritis. Isolated medial patellar facet and/or medial trochlear degeneration, or no more than Grade II to III patellofemoral chondromalacia, should be considered contraindications to PFA given subpar outcomes in those cohorts. Equally important, MRI should also confirm a lack of Grade III or IV chondral thinning in the tibiofemoral compartments or subchondral tibiofemoral condylar edema. However, focal, full-thickness femoral condylar defects can be treated with chondral grafting techniques and do not necessarily preclude PFA.
Patients must also be screened for conditions that predispose to failure. A history including inflammatory arthritis, chondrocalcinosis, neuropathic joint, joint instability, and/or morbid obesity makes rapid progression to tricompartmental disease likely; thus, these factors should be considered contraindications to PFA. Though not an absolute contraindication, the presence of depression should give the surgeon pause, as coexisting mental health issues can diminish patient satisfaction despite a technically successful surgery. Kazarian et al. found that patients with lower preoperative SF 36 mental health section scores were less satisfied and less likely to have their expectations met after PFA. Efforts should be made in these patients to address preexisting mental health comorbidities before surgery or to make sure that the patients understand the impact of preexisting characteristics on their perception of outcomes after PFA. Additionally, patients who require narcotic medications for PF pain or who display apparent symptom magnification should not be considered candidates for PFA unless these risk factors are modifiable.
Patellar Maltracking and Instability
Patellar and trochlear component malpositioning and soft-tissue imbalance may result in patellar maltracking, which is the most common technical error resulting in early failure after PFA. At present, patellar maltracking can be reduced with appropriate implant selection, thorough preoperative planning, meticulous soft-tissue management, and postoperative rehabilitation programs focused on quadriceps strengthening. However, despite optimal implant selection and positioning, patellar maltracking can still occur, particularly in cases of preoperative severe extensor mechanism malalignment or in the presence of a high Q angle. While tracking is often optimized with appropriate trochlear component external rotation and medialization of the patellar component, additional measures may be needed. After confirming component positioning, perform a limited lateral facetectomy of the portion of the patella that is not covered by the polyethylene button. This will often add some degree of slack in the lateral retinaculum. The next step would be to perform a lateral retinacular recession or lateral release. The need for more extensive interventions to achieve appropriate patellar tracking—including medial patellofemoral ligament reconstruction, anteromedial soft-tissue realignment, or tibial tubercle anteromedialization—would ideally be identified during the preoperative planning stages or intraoperatively and can be performed as necessary.
There is a growing body of literature elucidating the disparity in patellar maltracking and overall failures between inlay and onlay implant designs, with most modern implants falling into the latter category. Although inlay- and onlay-style implants have not been directly compared, data from the Australian registry report significantly higher 5-year cumulative revision rates for inlay- versus onlay-style designs (>20% vs. <10%, respectively). This is consistent with findings of one single-surgeon series that found that patients undergoing PFA with a first-generation inlay PFA had a 17% incidence of patellar maltracking, resulting in a relatively high need for secondary surgery or revision. The same series found that those who had a second-generation implant, using an onlay design, had an incidence of patellar maltracking of less than 1%. It is also consistent with a meta-analysis of nonrandomized trials that showed an eightfold higher likelihood of reoperation or revision after inlay-style PFA—most commonly due to patellar maltracking—compared with TKA, whereas these rates were equivalent when onlay-style PFAs were used. Consistently, the differences between the inlay and onlay designs have been most striking in terms of patellar maltracking, with inlay-style designs reporting an incidence of maltracking in 17% to 36% of patients versus <1% of patients who received onlay-style prostheses. , The impact of implant selection in avoiding this common pitfall merits an in-depth discussion of the nuances of their respective designs.
Inlay designs have a narrower anterior femoral flange width and often suboptimal trochlear implant geometry, which makes them more prone to patellar instability and subluxation. , Additionally, the sagittal radius of curvature is often relatively obtuse, resulting in a tendency to flex the implant, which leads the proximal portion of the flange to be prominent and offset from the anterior femoral surface, further increasing the risk for maltracking, snapping, and clunking as the knee transitions through the initial 30 degrees of flexion. , The limited proximal extension of many inlay trochlear components on the anterior femur puts the patient at further risk of maltracking and subluxation as the patella initially articulates on the native anterior femoral cortex in full extension and then transitions onto the trochlear implant surface at approximately 10 to 20 degrees of flexion. This transition point can lead to patellar catching and subluxation. Finally—and perhaps most impactful on patellar tracking—inlay designs are positioned based on normal trochlear morphology. The requisite trochlear inlay implant positioning, such that its articulating surfaces are flush with the surrounding articular cartilage, makes inlay trochlear components prone to placement in excessive internal rotation ( Fig. 13.1 ), which inadvertently puts the patella at greatest risk of subluxation and maltracking. , , ,
In contrast, newer-generation onlay-style implants are positioned perpendicular to the AP axis of the distal femur based on the same anterior femoral resection as performed in TKA. , , This controlled, standardized amount of external rotation eliminates the influence and variability of native trochlear inclination and, thus, the tendency to internally rotate the trochlear implant so common with inlay-style designs. , In the same vein, Valoroso et al. found that trochlear resection in neutral rotation, using an onlay-style trochlear component, improved patellofemoral congruence and reduced patellar tilt. Newer onlay-style implants have improved other elements of trochlear implant geometry as well, including the sagittal radius of curvature, medial-lateral implant breadth, optimized tracking angle, and proximal extension of the anterior flange, all of which have further enhanced patellar tracking. , Onlay-style prostheses have a sagittal radius of curvature compatible with most distal femora, allowing flush placement proximally and in the intercondylar region, thereby avoiding the tendency to flex the implant. They also have an increased medial to lateral width that covers the entire anterior surface of the articulating distal femur. This, in combination with the increased proximal extent of the trochlear component, enables the patellar component to engage with the trochlear component throughout the critical range of motion, avoiding the patellar snapping and catching commonly observed with inlay-style designs. Converting an inlay-style trochlear prosthesis to an onlay-style implant can correct patellar maltracking when it is due to component malposition (internal rotation, in particular). , Hendrix et al. reported that at a mean of 5-year follow-up, these patients experienced improved Bristol knee pain and function scores, with only a small proportion of patients requiring revision to TKA due to progression of arthritis.
Tibiofemoral Disease Progression
In the absence of maltracking, the primary mode of failure is progression of tibiofemoral arthritis progression, accounting for 42% of revisions performed within 5 years of surgery and most late failures. With regard to the earlier discussion on patient selection, it is worth repeating that a critical preoperative assessment including patient history, imaging studies, and physical examination is imperative to avoid pursuing PFA in patients with early signs of diffuse degenerative disease or with systemic risk factors for progression. Interestingly, multiple series have reported that the risk of revision surgery due to tibiofemoral disease progression is increased when the indication for PFA was primary OA compared with when it is done for PF arthritis secondary to trochlear dysplasia or posttrauma. This makes intuitive sense given that patients with trochlear dysplasia have anatomic abnormalities isolated to the PF compartment that selectively puts it at risk for degenerative arthritis, whereas patients with primary osteoarthritis have a greater tendency for developing more diffuse disease in the knee. Progression to advanced tibiofemoral arthritis was also associated with obese patients at 4-year follow up. Surprisingly, tibiofemoral disease progression has not been statistically associated with patient age, gender, history of prior knee surgery, patellar height, patellofemoral OA severity, patellar and femoral component sizes, or concomitant performance of lateral release.
When tibiofemoral arthritis disease progression does occur, the common resolution is to convert the PFA to a TKA. Clinical outcomes of conversion TKA after prior PFA are typically more comparable to primary TKA than a revision TKA, particularly since the patella often does not require revision. With regard to surgical technique, the conversion generally requires removal of only the trochlear component, as the patellar button is typically retained unless grossly loose, oxidized, worn, or malpositioned. Removing even a well-fixed trochlear implant is typically easy and can be done with a curved ¼-inch osteotome without compromising bone stock. In fact, generally primary TKA components are used, without a need for stems, augments, or bone graft ( Fig. 13.2 ). In a series from van Jonbergen et al., there was a trend toward increased stiffness following conversion from PFA to TKA, with 3 patients requiring manipulation under anesthesia. However, this was not significant and this cohort similarly showed Knee Society scores, Western Ontario and McMaster Universities Arthritis Index (WOMAC) scores, and complication rates equivalent to primary TKA. Others have found that while duration of surgery, postoperative pain relief, and range of motion after revision PFA to TKA were similar to primary TKA, patients undergoing conversion TKA were at increased risk for infection and for requiring further surgery.