Extensor mechanism complications are the most common reason for reoperation after the use of modern condylar total knee arthroplasty designs. 1,9–12,18,21,22 These complications include patellofemoral instability, patellar fracture, patellar component wear and loosening, extensor mechanism disruption, and soft tissue impingement syndromes. Several of these complications may lead to reoperation requiring removal of the patellar implant. In many instances, a new patellar component can be implanted with good clinical results. 3,19
In some cases, however, the magnitude of patellar bone loss or concern about the ability to obtain adequate fixation have prompted the practice of not attempting to replace the patellar component. Reasons for patellar bone loss include prior bone resection, fracture, osteonecrosis, or osteolysis. The treatment options in this situation include patellectomy, patellar resection arthroplasty, or patellar bone grafting. 2,10,12,13,19 In patients with a patellar resection arthroplasty, the functional outcome is less optimal when compared with results of implanting another patellar prosthesis. 2 Some instances of postoperative patellar fracture and persistent retropatellar pain have been reported after patellar resection arthroplasty. 19 Studies of patellar resection arthroplasty to date have described patients with relatively short followup. 2,10,19
The goals of the current study were to evaluate the clinical and functional results of patellar resection arthroplasty for the severely compromised patella for which insertion of another patellar component was not possible. An additional goal of the study was to elucidate factors associated with a successful or poor outcome of a patellar resection arthroplasty.
MATERIALS AND METHODS
Between 1978 and 1996, 33 patients (37 knees) with a total knee arthroplasty had a reoperation in which the patella was retained but the patellar component was removed and not replaced because of severe compromise of patellar bone stock that precluded implantation of another patellar prosthesis. Patients treated with revision operation of an infected prosthesis and patients with a patellar fracture at the time of revision were excluded. Two of the 33 patients died before the 2-year followup, and the remaining 31 patients (35 knees) constitute the basis of this study. There were 17 men (two bilateral) and 14 women with an average age of 66.6 years (range, 32–85 years) at the time of their patellar resection arthroplasty. The underlying diagnoses leading to the original knee arthroplasty was osteoarthrosis in 27 patients, rheumatoid arthritis in six patients, osteonecrosis in one patient, and posttraumatic arthritis in one patient. The patellar resection arthroplasty followed a primary knee arthroplasty in 18 knees (16 patients) and a revision arthroplasty in 17 knees (15 patients).
At the time of patellar resection arthroplasty, the procedure was done as an isolated procedure consisting of patellar component removal only in 11 knees (nine patients) or patellar component removal and tibial polyethylene exchange in eight knees (six patients). In 15 knees (15 patients), the femoral and tibial components were revised in conjunction with the patellar resection arthroplasty and one additional knee (one patient) had femoral component revision, patellar resection arthroplasty, and retention of the existing tibial component. Data subanalysis was done comparing patients who had isolated patellar component resection (Group I) and patients who had patellar component resection and revision of the tibial or femoral component or both (Group II).
The patellar components removed were metal-backed in seven and all-polyethylene patellar components in 28 knees. The reasons for removal of the existing patellar component included aseptic loosening in 29 knees (27 patients) and severe patellar polyethylene wear and associated patellar osteolysis in six knees (four patients). After removal of the patellar component and any remaining bone cement, the existing patellar shell was allowed to articulate against the femoral component in 23 knees (21 patients). A patelloplasty, consisting of shaping the remaining bony contours of the patellar shell to promote tracking and provide an extensor mechanism fulcrum was done in 12 knees (nine patients). Adjunctive procedures to facilitate and improve patellar tracking included lateral retinacular release in 16 knees (14 patients) and an associated vastus medialis advancement in nine knees (eight patients) (Table 1).
Clinical and radiographic data on all patients were collected prospectively. Patients were contacted at 1 year, 2 years, 5 years, and every 5 years subsequently on a routine basis. Some patients were evaluated more frequently on the basis of young age or symptoms, if deemed necessary. Knee scores were calculated using the Knee Society knee scoring system consisting of a score for pain and a score for function, each with a maximum of 100 points. 15 There were five deaths of causes unrelated to the knee operation: one at 17 years, one at 9 years, two at 6 years, and one at 4 years. All patients who died had been evaluated postoperatively. Among the 26 patients (30 knees) who were alive at the end of the study period, 17 patients (19 knees) were evaluated most recently with physical and radiologic examinations including an interview, and nine patients (11 knees) returned detailed questionnaires. Final followup averaged 7.9 years (range, 2–18 years). No patients were lost to followup.
The changes in knee pain and function scores were evaluated using analysis of variance (ANOVA). Statistical significance was attributed to observations with less than 0.05 likelihood occurring by chance.
The mean preoperative Knee Society pain and function scores were 50.9 points (range, 0–95 points) and 51.4 points (range, 0–90 points) respectively. These improved significantly to a mean pain score of 58.4 points (range, 20–100 points) and function score of 65.6 points (range, 10–100 points) at 2 years postoperatively (p < 0.05). At final followup these scores had declined to a mean of 57.4 points (range, 0–100) for pain and 52.1 points (range, 0–100) for function (p < 0.05). Comparing the Knee Society scores for patients who had an additional procedure with patients who did not, the scores were significantly higher for patients who had vastus medialis advancement (p < 0.05) or lateral retinacular release (p < 0.04). The Knee Society scores did not differ significantly for patients who had patelloplasty (p = 0.6).
Range of Motion
The mean arc of knee motion was 84° (range, 0°–110°) preoperatively and improved to 105° (range, 0°–135°) at 2 years postoperatively (p < 0.2) and 102° (range, 0°–120°) (p < 0.3) at latest followup. Before the index surgery, 11 patients had an extensor lag of 10° or greater. One patient had an extensor lag of 45° and another had an extensor lag of 50°. At final followup only one patient had an extensor lag greater than 10° (15°).
Before resection arthroplasty two patients (two knees) had no pain, two patients (two knees) had mild pain, three patients (three knees) had moderate pain, and 24 patients (28 knees) had severe pain. At 2 years followup the pain was rated as none in 23 patients (24 knees), mild in three patients (five knees), moderate in two patients (three knees), and severe in three patients (three knees). At final followup the incidence of pain was none in 14 patients (16 knees), mild in one patient (one knee), moderate in six patients (eight knees), and severe in 10 patients (10 knees).
Lateral radiographs were available in 34 of 35 knees (31 patients) at the latest followup and Merchant views were available in 31 of 35 knees (28 patients) at the latest followup. Of the 31 knees (28 patients) with Merchant views, patellae were located centrally (Fig 1) in 15 knees (14 patients), laterally subluxated (Fig 2) in nine knees (eight patients), fractured (Fig 3) in six knees (five patients), and dislocated in one knee (one patient). Patients with fractured or subluxated patellas had poorer functional scores when compared with patients in whom the patella remained centrally located (p < 0.05).
Six knees (four patients) required reoperation: three at 2 years, one at 5 years, and two at 6 years. All were treated with patellectomy. The indication for reoperation was pain with patellar fracture in four and pain without fracture in two knees. Revision of femoral and tibial components at the time of patellectomy also was done in four knees (four patients): two had tibial and femoral loosening, one had axial tibiofemoral malalignment, and in one without clear tibial or femoral loosening to exclude the old tibial and femoral implants as a possible source of pain.
One patient had a deep venous thrombosis develop and one had a postoperative myocardial infarction. There was one superficial wound infection that was treated successfully with local wound care and oral antibiotics.
Data Subanalysis (Group I Versus Group II)
Patients having patellar resection arthroplasty in isolation (Group I) and patients receiving revision of tibial or femoral components or both in addition to patellar resection arthroplasty (Group II) were subanalyzed with respect to outcome of the procedure (Table 1). With the numbers available there was no statistically significant difference between the two groups with respect to function (p < 0.06). However, at final followup pain scores were better in patients in Group II than in patients in Group I (p < 0.03). Furthermore, there were significantly more reoperations in patients in Group I compared with patients in Group II (p = 0.01).
Patellofemoral complications represent a substantial proportion of the problems reported after total knee arthroplasty. 1,7–9,11,12,16,18,21,22 When patellar component failure necessitates implant removal, treatment options include insertion of another patellar prosthesis, removal of the patellar component without insertion of another prosthesis, patellectomy, or patellar bone graft. Usually, reimplantation of another patellar implant is possible and preferable as this provides the best pain relief and functional outcome. 2–4,14,20 Some authors have reported a high complication rate associated with this procedure, which may be attributable to poor remaining bone stock at the time of revision surgery. 3 Patellectomy is associated with markedly inferior functional results, difficulties with weakness or delayed disruption of extensor mechanism perhaps attributable to abnormal knee biomechanics, diminished quadriceps torque and strength, and ligament instability. 8,17 Patellar impaction bone grafting or insertion of the patellar component into the extensor mechanism have been proposed as alternatives. 6,13 The long-term result of these procedures is unknown at this point.
There are some reports in the literature on the results of patellar resection arthroplasty. 2,10,19 An early report, (nine knees), concluded that patellar resection arthroplasty was the preferable alternative to patellectomy as none of these patients had poor results or reoperation at an average followup of 55 months. 10 Barrack et al 2 observed that the clinical outcome of patellar resection arthroplasty at 30 months followup was inferior to patellar component reimplantation. Pagnano et al 19 reported on the results of patellar resection arthroplasty in 34 knees, including knees with patellar fractures and deep infection, and reported an improvement in Knee Society scores after patellar resection arthroplasty. In their experience, 10 patients had persistent mild or moderate anterior knee pain at followup of 3.5 years and no patients required reoperation.
Although not strictly comparable, the results of the current study are consistent with results in published reports. 2,10,19 Patients with preoperative patellar fracture and deep infection were excluded because these diagnoses present confounding variables that complicate the analysis of results of patellar resection arthroplasty. The findings of the current study showed that patellar resection arthroplasty provided relief of pain for most but not all patients and to a lesser degree improvement of function for some patients. One observation of this study was the marked improvement in the preoperative extensor lag.
Patients treated with isolated patellar resection arthroplasty had significantly poorer pain scores at final followup, when compared with patients who had revision of other knee components at the time of the patellar resection arthroplasty (p < 0.03). They also had more reoperations (p = 0.01). Functional scores also were poorer but the difference did not reach statistical significance. The reason for the difference in outcome between the two groups probably is multifactorial. The majority of the patients having isolated patellar implant resection had extensor mechanism tracking problems as evidenced by the high rate of lateral retinacular release or vastus medialis advancement or both in this group. The authors theorize, but cannot prove, that patients treated with isolated patellar component resection may have had unrecognized or untreated underlying tibial or femoral component malrotation or both that contributed to the initial failure and also may have played a part in the ongoing extensor mechanism problems, even after resection arthroplasty. These findings imply that tracking of the extensor mechanism should be evaluated carefully at the time of resection arthroplasty and appropriate steps should be taken to rebalance the extensor mechanism or solve the problems leading to extensor mechanism failure.
At final followup patellar subluxation or dislocation was evident radiographically in 10 knees. The importance of this radiographic finding is not known as there is a report indicating that radiographic patellar subluxation without clinical symptoms is common. 5 Nevertheless, the functional outcomes were poorer in patients with radiographic evidence of patellar subluxation, dislocation, or fracture (p < 0.05).
The current study suggests that patellar component resection arthroplasty, consisting of removal of the existing patellar component and associated patelloplasty, provides a reasonable treatment option for a failed patellar component associated with deficient patellar bone.
The patella should be resurfaced during revision knee arthroplasty if the thickness generally is greater than 8 to 10 mm. In cases of severe bone deficiency, patellar resection arthroplasty is a reasonable option to consider and is preferred to patellectomy. The extensor mechanism also should be examined carefully and realignment procedures done if necessary, especially in patients having isolated patellar resection. At the time of patellar resection arthroplasty tibial and femoral component stability and axial and rotational alignment should be scrutinized, and if major problems are identified, measures should be taken to solve them.
1. Aglietti P, Buzzi R, Gandenzi A: Patellofemoral functional results and complications with the posterior stabilized total condylar knee prosthesis. J Arthroplasty 3:17–25, 1988.
2. Barrack RL, Ingraham R, Matzkin E, Rorabeck CH, Engh GA: Results of revision total knee with patellar replacement versus patellar bony shell. Clin Orthop 356:139–143, 1998.
3. Berry DJ, Rand JA: Isolated patellar component revision of total knee arthroplasty. Clin Orthop 286:110–115, 1993.
4. Boyd Jr AD, Ewald FC, Thomas WH, Poss R, Sledge CB: Long-term complications after total knee arthroplasty with or without resurfacing of the patella. J Bone Joint Surg 75A:674–681, 1993.
5. Brick GW, Scott RD: The patellofemoral component of total knee arthroplasty. Clin Orthop 231:163–178, 1988.
6. Buechel FF: Patellar tendon bone grafting for patellectomized patients having total knee arthroplasty. Clin Orthop 271:72–78, 1991.
7. Campbell DG, Mintz AD, Stevenson TM: Early patellofemoral revision following total knee arthroplasty. J Arthroplasty 10:287–291, 1995.
8. Dennis DA: Extensor mechanism problems in total knee arthroplasty. Instr Course Lect 46:171–180, 1997.
9. Doolittle KHD, Turner RH: Patellofemoral problems following total knee arthroplasty. Orthop Rev 17:696–702, 1988.
10. Drakeford MK, Tsao AK, Lavernia C, Krackow KA, Hungerford DS: Resection arthroplasty for failed patellar component. Orthop Trans 17:992, 1994.
11. Grace JN, Rand JA: Patellar instability after total knee arthroplasty. Clin Orthop 237:184–189, 1988.
12. Grace JN, Sim FH: Fracture of the patella after total knee arthroplasty. Clin Orthop 230:168–175, 1988.
13. Hanssen AD: Bone-grafting for severe patellar bone loss during revision knee arthroplasty. J Bone Joint Surg 83A:171–176, 2001.
14. Hungerford DS: Management of extensor mechanism complications in total knee arthroplasty. Orthopedics 17:843–844, 1994.
15. Insall JN, Dorr LD, Scott RD, Scott RN: Rationale of the Knee Society clinical rating system. Clin Orthop 248:13–14, 1989.
16. Johnson DP, Eastwood DM: Patellar complications after knee arthroplasty: A prospective study of 56 cases using the kinematic prosthesis. Acta Orthop Scand 63:74–79, 1992.
17. Laskin RS: Management of the patella during revision total knee arthroplasty. Orthop Clin North Am 29:355–360, 1998.
18. Lynch AF, Rorabeck CH, Bourne RB: Extensor mechanism complications following total knee arthroplasty. J Arthroplasty 2:135–140, 1987.
19. Pagnano MW, Scuderi GR, Insall JN: Resection of the patellar component for inadequate bone stock in revision and reimplant total knee arthroplasty. Clin Orthop 356:134–138, 1998.
20. Rand JA, Chao EY, Stauffer RN: Kinematic rotating-hinge total knee arthroplasty. J Bone Joint Surg 69A:489–497, 1987.
21. Stuart MJ, Larson JE, Morrey BF: Reoperation after condylar revision total knee arthroplasty. Clin Orthop 286:168–173, 1993.
22. Wright J, Ewald FC, Walker PS, et al: Total knee arthroplasty with the kinematic prosthesis: Results after five to nine years: A follow-up note. J Bone Joint Surg 72A:1003–1009, 1990.