Researchers make their case for patellar resurfacing

Findings suggest the procedure may lead to less postop anterior knee pain; some critics disagree.

Issue: Issue 4 2003

British flag A prospective, randomized study of post-arthroplasty anterior knee pain showed a significantly lower prevalence of pain in resurfaced patellae than in nonresurfaced patellae.

The study by Tim S. Waters, MRCS, and George Bentley, ChM, FRCS, at the Royal National Orthopaedic Hospital in Middlesex, England, began with 514 consecutive primary press-fit condylar total knee replacements. After exclusions for frailty of the patella and losses during follow-up, the study included 474 knees in 390 patients.

The results showed overall prevalence of anterior knee pain at 25.1% (58 of 231 knees) in the nonresurfaced group and 5.3% (13 of 243 knees) in the resurfaced group (P<0.0001).

“We changed our practice at the Royal National Orthopaedic Hospital because of this paper. We now resurface the patella whenever we can,” Waters told Orthopaedics Today.

Survey planned

Water said he and Bentley are planning a survey of the British Orthopaedic Association about orthopaedic surgeons’ attitudes toward resurfacing the patella. The last such survey was performed in 1996.

While a survey will take time to gauge reaction to the study results, Waters said the paper has generated much interest among British orthopaedists, whose anecdotal evidence suggests a shift in attitude toward resurfacing.

“Surgeons realize that postoperative anterior knee pain is difficult to predict. We found no correlation with the radiographic findings, and others such as John H. Newman et al from the Avon Orthopaedic Center, Bristol, England, have shown no correlation with the appearance of the patella at operation,” Waters said.

“Moreover, the studies show that there is nothing to be lost by resurfacing, and there seems to be no reason not to resurface.”

Surgeons differ

The debate over resurfacing and a lack of indisputable evidence helped generate the study by Waters and Bentley, whose stated purpose in the study was to clarify the role of patellar resurfacing during total knee arthroplasty, particularly with regard to anterior knee pain.

While the study by Waters and Bentley found less anterior knee pain in resurfaced patellae, other studies have shown that pain linked to resurfacing and the issue of patellar resurfacing is a matter of ongoing debate.

René Verdonk, MD, MS, of Gent University Hospital, Gent, Belgium, said he respected the quality of the study but doubted its results. “No other study indeed has shown such discrepancies,” he said.

Verdonk, who is a member of the Orthopaedics Today Editorial Advisory Board, also noted that decreasing anterior knee pain could be done with more extensive Hoffa fat pad removal or by desensitizing the patella.

He cited the results of a 2001 total knee arthroplasty study by Henkel et al in the Journal of Arthroplasty (performed at the Schulthess Clinic in Zurich, Switzerland) as well as the recommendation of American orthopaedic surgeon Peter A. Keblish Jr., MD, supporting that point.

Hans J. Refior, MD, chairman of the department of orthopaedics at the University of Munich, Germany, and an editor of Orthopaedics Today, also said the U.K. study would not change his general practice of not resurfacing.

Refior, who described the study and its follow-up as “excellent,” speculated that the lack of anterior knee pain may be due to the prosthesis design rather than the resurfacing.

Randomized study

The patients were randomized to receive either resurfacing or retention of the patella. They were also randomized to either a cruciate-substituting or a cruciate-retaining prosthesis as part of a separate trial.

The mean duration of follow-up was 5.3 years (2 to 8.5 years). Investigators assessed patients by the Knee Society rating system, a clinical anterior knee pain score and the British Orthopaedic Association patient satisfaction score.

Patients were divided into three categories: those who had one knee replaced, those who had both knees replaced and were contralateral knee symptomatic, and those with multiple joint arthritis or considered medically infirm.

The anterior knee pain rating is as follows: 0 — no pain; I — mild pain that does not intrude on daily activities; II — moderate pain that is a nuisance, patient not considering further surgery; III — severe pain, patient considering further surgery.

Thirty-five patients who received bilateral knee replacement underwent resurfacing on one side only.

Parapatellar approach

The operation was performed through a standard medial parapatellar approach. Assessment was performed with the examiner blind to resurfacing.

A lateral release was done on 76 knees (31.3%) in the resurfaced group and 61 knees (26.4%) in the nonresurfaced group. The procedure was performed if lateral patellar tracking was impaired.

All patients were managed with antibiotic prophylaxis from anesthesia until the wound healed. Other methods of prophylaxis were used for deep-vein thrombosis until discharge.

Each knee was splinted in extension for 48 hours: static quadriceps exercises were started without delay. Weight-bearing began after 24 hours, and flexion was initiated at 48 hours once the splint was removed.

Of the 58 with anterior knee pain in the nonresurfaced group, 34 were Grade I under the clinical anterior knee pain score, eight were Grade II and 16 were Grade III.

There was one case of component loosening, and 10 of 11 patients who underwent secondary resurfacing had complete relief of anterior knee pain.

Overall postoperative knee scores were lower in the nonresurfaced group, and there was a significant difference among patients with osteoarthritis (P<0.01).

There was no significant difference between the groups with regard to postoperative function scores, and patients with bilateral knee replacement were more likely to prefer the resurfaced side.

In further results presented earlier this year at the American Academy of Orthopaedic Surgeons meeting, which compared 67 PFC Sigma knees (with randomized patellar resurfacing) with knees receiving only the original PFC implant, patients with the nonresurfaced patella had significantly more anterior pain than the resurfaced group.

For your information:

Waters TS, Bentley G. Patellar resurfacing in total knee arthroplasty: a prospective, randomized study. JBJS. 85A:212-217; 2003.

Waters TS, Bentley G. Anterior knee pain following knee replacement: comparison of patellar resurfacing and non-resurfacing in PFC and PFC Sigma knee replacements. #189. Presented at the American Academy of Orthopaedic Surgeons 70th Annual Meeting. Feb. 5-9, 2003. New Orleans.