Impact of patellar height on unicompartment knee arthroplasty: does patella baja lead to an inferior outcome?

Though a number of series have been published documenting good results after UKA, there has been a paucity of literature on the role of UKA on the length of the patella tendon or the incidence of patella baja after UKA [1‐4, 18]. In spite of indications for this procedure in unicompartmental arthritis having been established, there has been some controversy regarding the status of the patellofemoral joint [1, 2, 4, 5]. A recent study [5] had concluded that patellar height had no consistent major effects on early clinical outcome after UKA, and hence the patellar height might not be considered as a strict separate patient-selection criterion. The purpose of this study was to evaluate the patients for a change in patellar height by ISR and BPI after UKA and to see if there is a difference in clinical and radiological outcomes for these patients. Before the study we had assumed that at mid-term follow-up a progression in patellofemoral arthritis may lead to progressive impingement, anterior knee pain and adversely affect the outcome in patients who develop patella baja post-operatively. However, our results showed a significant decrease in only the stair-climbing component of the functional aspect of KSS for patients with patella baja assessed with ISR while this decrease was not significant in patients with patella baja assessed with BPI, and hence we had to reject our previous hypothesis.

Patella baja refers to an inferior position of the patella in the sagittal plane and can be determined with numerous radiographic measurements [13, 14, 19]. Most involve a ratio between a measure of patellar length and a measure of the distance between an aspect of the patella and a landmark on the tibia. The use of a ratio rather than an absolute length compensates for variations in patients’ height. We used both the ISR and BPI to measure the patellar height after UKA. The BPI takes the joint-line into consideration. In contrast, the ISR ignores the joint-line and relates more to the patellar-tendon length. Grelsamer [7] described a decrease in BPI values and non-decreased ISR values to indicate an elevation of the joint-line, defined as pseudo-patella baja. However, an elevation of the joint-line is hardly possible after UKA, since the contralateral compartment is not affected by the surgery [5]. It may nevertheless be that UKA created a “step” in the joint-line from the medial to the lateral compartment, and referring to the upper polyethylene border for the measurements, BP values might be decreased by such a step [5]. Finally, BPI values might have been decreased by a true decrease of the patellar height. The decrease in patellar height after TKR has been reported in more than 50–64 % [8, 9] of cases and patella baja develops in 9–25 % of patients without lateral release, and all patients with patellar release [9, 10]. Weale et al. [10] reported that there was no significant change in patellar height after UKA at 8 months or 5 years after surgery, while Naal et al. [5] did observe a significant decrease in patellar height as per BPI and no significant decrease as per ISR at short-term follow-up. Our results at mid-term follow-up are also in agreement with these results of Naal et al. [5] which they observed at short-term follow-up. Also, in contrast to a study by Weale et al. [10] we observed an overall decrease in patellar tendon length and that this change was not progressive over the follow-up period. In contrast to TKA, fewer patients develop patella baja after UKA and this may be due to the minimally invasive incision, avoiding patellar eversion and minimal trauma to Hoffas’ fat pad and extensor mechanism. Much of the literature concerning UKA has advocated relative under-correction of the alignment of the knee with the presumption that overcorrection increases the risk of failure by progression of arthritic change of the lateral compartment [20‐22]. Scott et al. [23] reported that overcorrection of alignment increases the risk of degenerative change in the contralateral compartment. However, progression of arthritis in the lateral compartment could be simply caused by the natural progression of the underlying arthritic disease [24]. Barrett and Scott [25] reported that under-correction of alignment increases the wear of polyethylene implants and the recurrence of the deformity. Kennedy and White [16] reported that the best result can be obtained when the mechanical axis runs through the center of the knee joint or slightly medial from the center of knee joint, and we aimed for the same in our cases. The position of the implant did not affect the development of patella baja. This may be because limited exposure and careful soft tissue resection prevented damage to the patellar tendon. In spite of a small incision, it was observed that a large proportion of implants were placed in an acceptable position. Moreover, as UKA involves only the medial compartment, it does not effectively change the position of the joint line [17].

Meneghini et al. [9] studied over 1,000 patients with TKA and concluded that a decrease in patella tendon length occurs; stair and function scores after TKA are adversely affected compared with patients in whom the ISR is not decreased. Other studies have associated patella infera after TKA with decreased post-operative range of motion, extensor lag, anterior knee pain, anterior polyethylene impingement and wear, and diminished outcomes [5‐9]. In contrast, after UKA no consistent major effects on early clinical outcome were seen and a weak negative correlation between lower pre-operative BP values and the post-operative knee extension, and a weak negative correlation between lower pre-operative IS values and post-operative knee scores were seen [5]. Our results did show a significant difference only in stair-climbing scores at final follow-up for patients with post-operative patellar baja. However, when the scores were compared between those patients who had pre-operative patella baja according to both ISR and BPI, who became normal after surgery, and those who remained patella baja after surgery, there was no significant difference between them. Thus, it shows that a corrective effect of surgery on pre-operative patellar baja is not present.

The current study shows a decline in functional scores over time in patients undergoing UKA. However, this decrease did not correlate with patellar height. The decrease in functional status may be related to the increase in age of the patients, considering that the mean age at which the patients underwent UKA in our study was 64.2 years.

Progression of patellofemoral arthrosis after UKA is a possibility and Berger et al. [1] observed patellofemoral symptoms were present in 1.6 % of patients at 10 years which increased markedly to 10 % of patients at 15 years. Hernigou and Deschamps [26] reported that after UKA the patellofemoral joint was affected by degenerative changes and patellar impingement against the femoral component and that both factors negatively influenced the functional outcome. Patellar impingement with the Oxford UKA prosthesis, as in our study, may not be a problem as during the surgical procedure care is taken to remove the bone anteriorly until there is at least 4 mm clearance for the front of the bearing in full extension. Price et al. [4] ignored the state of the patellofemoral joint in the absence of anterior knee pain at surgery and at 10 years follow-up; 97 % of knees reviewed had no patella-femoral pain and also the main cause for revision in their series was progression of osteoarthritis in the lateral compartment. At an average follow-up of 6.4 years we had 5.22 % of patients who complained of anterior knee pain. However, in only one patient with associated anterior knee pain was a TKA performed.

We do have limitations in this study, in that it is a retrospective review of cases. However, there are many strengths like use of a single implant, a single surgeon being the main operating surgeon, medium duration of follow-up and a sufficiently large number of cases: all of which may offset the study limitation.

In summary, this study documents a decrease in patellar height after UKA by both ISR and BPI. In this study, 7.3 % developed post-operative patella baja according to ISR while 7.8 % developed patella baja as per BPI. Patients with a decrease in patellar height as per ISR have a decrease in stair-climbing score at mid-term follow-up. There was no effect on the correction of pre-operative patella baja to normal post-operatively. The overall KSS and pain score are not affected by a change in patellar height and neither is there a significant progress in patellofemoral osteoarthritis among patients with patella baja compared to normal patella.