The study by Breeman et al. is an excellent multi-center, randomized, controlled trial initiated in 1999 that enrolled 1715 patients who were randomly allocated to receive or not receive patellar resurfacing during total knee arthroplasty. The authors examined functional knee scores, cost-effectiveness, and the need for subsequent surgery. It is considered to be a Level-I study and is the largest randomized controlled trial of patellar resurfacing to date. Previous studies were either nonrandomized trials or smaller randomized trials1-4, and the existing systematic reviews were based on only a small number of studies with a high level of evidence5,6. Despite everything that has already been stated in the literature regarding patellar resurfacing, the debate continues with no definitive answer. This topic continues to appear on the program at many of the knee arthroplasty meetings, in the form of either a debate or the most recent opinions—which are often contradictory. Thus, although the article by Breeman et al. is an excellent one, the big question is…does it finally settle this controversy?
Sixteen (2.1%) of the 759 patients who did not receive resurfacing eventually underwent a resurfacing less than five years after the primary arthroplasty. Only two patients who received a resurfacing sustained a patellar fracture requiring further surgery. A patella-related reoperation was necessary in 2% of the nonresurfacing group and only 1% of the resurfacing group. The proportion of patients requiring a reoperation of any type in the affected knee was higher in the nonresurfacing group than in the resurfacing group: 5.8% compared with 4.4% for minor to intermediate reoperations and 2.9% compared with 1.6% for major reoperations. The authors, however, noted that the differences were not significant and were probably due to random variations. Despite that, my impression is that there is a trend toward more frequent problems in the nonresurfacing group. Another extremely important finding is that patients who need to undergo a late resurfacing typically do not do as well as patients who had their patella resurfaced during the original arthroplasty. This provides yet another reason for resurfacing the patella during knee arthroplasty.
The cost of having the patella resurfaced added significantly to the total cost of the arthroplasty, but this was offset by the smaller number of reoperations required in the resurfacing group than in the nonresurfacing group. The economic analysis therefore yielded a tie between the two groups.
This study fails to answer some very important questions. Several parameters were not examined: posterior cruciate-substituting versus posterior cruciate-retaining designs, the effect of knee alignment, and the effect of the degree of osteoarthritis of the patellofemoral joint, which is perhaps the most important question of all. It is possible, for example, that a posterior cruciate-substituting design yields better results if the patella is resurfaced, that a valgus knee does better with patellar resurfacing, and that a knee with severe osteoarthritis of the patellofemoral compartment does better with patellar resurfacing. I am postulating this only because that is my own personal impression, and these questions were not answered by this study. I believe that the sample size in the study was large enough that these parameters could be examined if the necessary data were available. It is hard for me to imagine that a knee with a severe valgus deformity and severe osteoarthritis of the patellofemoral compartment does as well without patellar resurfacing as with patellar resurfacing. I could be wrong—but this article does not provide me with that answer. I would encourage the authors, if it is possible, to look at subgroups of patients and analyze the effects of knee alignment, the degree of osteoarthritis, and cruciate-substituting versus cruciate-retaining designs. Also, the present study should be extended to ten years.
In summary, this article actually reassures me, as someone who resurfaces all patellae—but at the same time it probably gives a degree of reassurance to those surgeons who never resurface. I therefore do not believe that opinions will be changed on the basis of this article, and thus the controversy will continue. It is an excellent article analyzing a huge series of patients, but I would encourage the authors to examine the subgroups further.