Copyright © 2004 by The Journal of Bone and Joint Surgery, Inc.

Commentary & Prospective

Commentary & Perspective on
"Long-Term Changes of the Nonresurfaced Patella After Total Knee Arthroplasty"
by Hsin-Nung Shih, MD, et al.

Commentary & Perspective by
Robert L. Barrack, MD*,
Tulane University Health Sciences Center, New Orleans, Louisiana

Despite numerous studies in recent years, the issue of whether or not to resurface the patella during primary total knee arthroplasty remains unresolved. The advocates of routine resurfacing claim that the prevalence of anterior knee pain is higher in the nonresurfaced knee, functional capacity is higher with resurfacing, and that a substantial percentage of nonresurfaced knees will require subsequent resurfacing. Proponents of nonresurfacing point to the complications associated with resurfacing, such as suboptimal resection of the native patella, which can lead to symptoms and the need for reoperation. Occasionally, severe complications such as patellar tendon rupture or patellar fracture are related to patellar resurfacing. Clearly, this is not a totally benign procedure. The additional time and expense that is associated with resurfacing the patella is also an issue. According to recent reports in the literature, the prevalence of anterior knee pain or symptoms relating to the patellofemoral joint following total knee arthroplasty is approximately 10% to 20% on the average1,2. Whether or not the prevalence is higher without patellar resurfacing remains hotly debated. One study reported significantly higher rates of anterior knee pain (p < 0.0001) in the nonresurfaced knee3, whereas another reported an equal incidence4.

A middle-ground approach is so-called selective resurfacing, a term that describes the resurfacing of the patella for specific indications only. Since only 10% to 20% of patients with a resurfaced or unresurfaced patella will experience anterior knee pain following total knee arthroplasty, it seems excessive to resurface every patella and expose 100% of patients to the additional operative time, expense, and risk associated with this aspect of total knee arthroplasty when only a small percentage will potentially benefit. Unfortunately, no study has been able to identify the patient factors that consistently predict who will experience anterior knee pain following total knee arthroplasty with an unresurfaced patella. The report by Shih et al. adds valuable new information to the debate and may prove to be of value in the selection of patients for patellar resurfacing.

One aspect of this issue that has not been explored is the fate of the unresurfaced patella over time. The results indicate that a specific preoperative radiographic parameter may prove to be of value in predicting which patients with an unresurfaced patella are likely to experience anterior knee pain and functional limitations following total knee arthroplasty. Of the forty-one patients with abnormalities on the Merchant view preoperatively, a striking 98% (which represented only 18% of the study group) sustained degenerative changes of the patellofemoral joint over the ensuing years. More importantly, a significantly higher percentage of patients with this finding experienced anterior knee pain as well as difficulty with climbing stairs or arising from a chair (p < 0.001). Among the patients who did not have an abnormal Merchant view preoperatively, the incidence of symptoms was very low, in the range frequently reported for patients with a resurfaced patella. This would seem to provide a crucial missing piece of the puzzle—the predictive factor for anterior knee pain following total knee arthroplasty. There are a number of confounding factors in this study, however, that are typical of the difficulties encountered in interpreting the results of other studies of patellar resurfacing. These include the variability in surgical technique, component design, and patient factors.

The authors utilized a standard medial parapatellar approach. They suggested that changing to a different approach, such as the subvastus, might limit laxity of the capsular structures, an occurrence that they thought led to the early lateral tilt that was observed in some patients. This is an interesting hypothesis that bears further exploration. They attempted to place the femoral component in 3° of external rotation and center the tibial component on the tibial tubercle. There is no mention of whether the femoral and tibial components were placed centrally or toward the lateral side, which might improve patellar tracking. Component factors are probably more relevant than technique factors in judging the results of this study. The PCA component they used was an early-generation cementless design that may not have been well suited for articulation with an unresurfaced patella. The design of the femoral component has evolved since that time to incorporate deeper, more conforming patellofemoral grooves that extend farther distally.

The parameter most unique to this study, however, was the patient population. The study group was almost 90% female and included individuals of small stature, by most standards, with an average height of only 152 cm (5 ft) and weight of 66 kg (145 lb). All of these factors affect the anatomy of the patellofemoral groove and the stresses on the patellofemoral joint. Both height and weight have been described as having an impact on the likelihood of anterior knee pain following total knee arthroplasty1,2. Because of these nuances of surgical technique, component design, and patient population, the results of the current study may not be generalizable to the practices of all surgeons.

Despite these limitations, the authors have made a valuable contribution to the debate. It seems clear from their data that the vast majority of patients who undergo total knee arthroplasty do not require patellar resurfacing, since <20% of their patients exhibited the radiographic risk factor they described. The fact that patients with an abnormal preoperative Merchant view were at high risk for anterior knee pain and diminished functioning following total knee arthroplasty without patellar resurfacing does not mean that these patients would fare better with patellar resurfacing. They might well do better, but they could also do worse or be at high risk for patellar fracture, subluxation, or component loosening. Whether the preoperative Merchant view proves to be of importance in predicting anterior knee pain following total knee arthroplasty with an unresurfaced patella, and thus a selection criterion for patellar resurfacing, is therefore not established by this study. Further study of this parameter in future investigations of this topic is certainly warranted, however.

*The authors did not receive grants or outside funding in support of their research or preparation of this manuscript. One or more of the authors received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity (DePuy). No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated.


1. Barrack RL, Bertot AJ, Wolfe MW, Waldman DA, Milicic M, Myers L. Patellar resurfacing in total knee arthroplasty: a prospective, randomized, double-blind study with five to seven years of follow-up. J Bone Joint Surg Am. 2001;83:1376-81.
2. Barrack RL, Wolfe MW, Waldman DA, Milicic M, Bertot AJ, Myers L. Resurfacing of the patella in total knee arthroplasty. A prospective, randomized, double-blind study. J Bone Joint Surg Am. 1997;79:1121-31.
3. Waters TS, Bentley G. Patellar resurfacing in total knee arthroplasty: A prospective, randomized study. J Bone Joint Surg Am. 2003;85:212-7.
4. Barrack RL, Bertot AJ, Wolfe MW, Waldman DA, Milicic M, Myers L. Patellar resurfacing in total knee arthroplasty. J Bone Joint Surg Am. 2001;83:1376-81.