Secondary Logo

Journal Logo


Whole Patellar Allograft for Total Knee Arthroplasty after Previous Patellectomy

Busfield, Benjamin, T; Ries, Michael, D

Author Information
Clinical Orthopaedics and Related Research: September 2006 - Volume 450 - Issue - p 145-149
doi: 10.1097/01.blo.0000223980.44455.21


Results of total knee arthroplasties (TKA) in patients who have patellectomies have been less favorable than in patients with intact patellae.8,9,12,13,15,17 Patellectomy compromises the extensor mechanism by decreasing the quadriceps excursion and lever arm with a resultant decrease in strength.1,10,21 Posterior cruciate-substituting rather than posterior cruciate-retaining designs have been advocated to compensate for quadriceps inefficiency, but the degree of constraint necessary as it relates to outcome is controversial.9,17

Cybex data for quadriceps strength greater than 40 ft/lb at 30°/second testing speed have been shown to be a prognostic factor for patients who have TKAs with patellectomies.13 Several reconstructive procedures have been proposed to reconstitute bone stock in the patella and restore maximum function of the extensor mechanism. Surgical options include autogenous bicortical and unicortical structural iliac crest bone grafts, and resected femoral condyles or resected tibial plateaus.2,11,19,20 Because the grafts do not contain soft tissue for suture fixation to the extensor mechanism, they are placed in a synovial pouch.2 However, adequate autogenous graft material may not restore patellar thickness, and late resorption or migration of the graft can occur. The average thickness of the unresected patella measured during TKA is 24 mm.6 Some authors have reported short-term outcomes using autogenous grafts for patellar augmentation after patellectomies.2,11,20 However, the long-term function of these grafts in restoring patellar height and quadriceps function is not clear. Extensor mechanism allografts have been used successfully to augment extensor mechanism deficiency after TKA when there is discontinuity in the extensor mechanism.3,16 Usually a whole patellar allograft (patellar ligament, whole patella, quadriceps tendon) is used. Although attenuation of the tendinous soft tissue component of the allograft has been reported as a complication, bony resorption or patellar migration does not seem to compromise results of this technique.3,16 To restore the full thickness of the patella, we have used whole patellar allografts in patients who had patellectomies with TKAs with intact extensor mechanisms. The allograft was placed underneath the soft tissue sleeve of the extensor mechanism so attenuation of the allograft tendon would not be expected to compromise extensor mechanism strength while still retaining patellar thickness with a large solid patellar bone graft.

We compared the functional outcome and complications in a consecutive series of patients who had TKAs after patellectomies and who had whole patellar allografting with reported outcomes and complications of patients who had TKAs after patellectomies, including those with other patellar bone-grafting techniques.


We retrospectively reviewed a consecutive series of eight patients (10 knees) with TKAs and previous patellectomies treated with whole patellar allografting. We then compared their results with those of historical controls. Only patients who had a previous surgical patellectomy were included. Patients who had osteolysis, patellar fracture, or any remaining patellar bone stock were excluded (Table 1). One patient was lost to followup. The patient had moved out of the area and could not be contacted. We based our results on the remaining seven patients (nine knees). Patients were an average age of 54 years (range, 47-59 years). There were five men and two women. One patient had bilateral whole patellar allografting and TKAs. One whole patellar allograft revision procedure was performed for allograft resorption and flexion instability. The allograft was revised to a new whole patellar allograft with TKA revision to a thicker tibial insert. The patients who had primary and revision TKAs of the ipsilateral knee were included in the analysis (Patients 5 and 9). The followup averaged 44.5 months (range, 39-48 months).

Patient Characteristics and Secondary Surgeries

The initial indication for patellectomy was chondromalacia in seven knees and patella fracture in two knees. The average time from patellectomy to TKA was 17.8 years (range, 5 months-37 years). The indication for primary TKA was posttraumatic arthritis in one knee, osteoarthritis (OA) in four knees, and preexisting tibiofemoral OA with an acute comminuted distal femur fracture in one knee. The indication for revision TKA was infection in one knee, periprosthetic fracture in one knee, and patellar allograft resorption with flexion instability in one knee. One patient had a history of HIV, hepatitis B, and diabetes mellitus (DM), two patients had a history of hepatitis B(one of whom also had hepatitis C, cirrhosis, and malnutrition), one patient was taking steroids for temporal arthritis, and one patient was taking steroids after a cardiac transplant.

The knee was exposed though a vertical medial or midline arthrotomy through the extensor mechanism for performing primary or revision TKA. A posterior cruciate-retaining TKA with an oxidized zirconium femoral component and dished tibial insert was used for primary TKAs (Genesis, Smith and Nephew, Memphis, TN). A fresh-frozen extensor mechanism allograft (consisting of the quadriceps tendon, patella, patellar ligament, and tibial tubercle in continuity) was obtained and thawed in warm saline before implantation. The patellar cartilage was removed using a scalpel to decrease the immunogenicity of the allograft. No patellar prosthesis was implanted in the allograft to avoid problems of late patellar component loosening. The tibial tubercle bone block was excised, and the remaining quadriceps and patellar tendons were sutured into the underside of the patient's intact extensor mechanism soft tissue sleeve using #1 Ticron sutures (Fig. 1). The allograft patellar ligament was positioned deep to the host patellar ligament, and the allograft quadriceps tendon was positioned deep to the host quadriceps tendon. Several sutures were placed to secure the allograft. The host patellar ligament, quadriceps tendon, and the knee were flexed through a range of motion (ROM) from 0° to 90°. If the allograft tilted or impinged on the tibial insert, it was repositioned before final suture repair. The allograft extensor mechanism was positioned so the distal pole of the patella was 1 cm proximal to the tibiofemoral joint line as recommended by Figgie et al.4 A posterior cruciate-retaining femoral component with dished tibial (Genesis®) implant was used for all primary TKAs without sacrificing the posterior cruciate ligament, except in Patient 6. Patients 6 and 8 had distal femoral condyle fractures that compromised the collateral ligaments and necessitated the use of constrained implants.

Fig 1A
Fig 1A:
D. Intraoperative photographs show the (A) soft tissue sleeve of the extensor mechanism retracted laterally after TKA; (B) a whole patellar allograft with sutures positioned anteriorly; (C) sutures placed through the allograft tendon and tied anteriorly over the host quadriceps tendon and patellar tendon; and (D) the host extensor mechanism soft tissue sleeve closed side-to-side covering the anterior surface of the allograft.

Patients were treated with the same postoperative physical therapy protocol used for patients who have primary TKAs. Unrestricted active assisted and passive ROM exercises and ambulation weightbearing as tolerated were started 1 day postoperatively and advanced as tolerated.

Data collected included Knee Society knee and function scores, ROM, quadriceps strength, and extensor lag. Separate knee and function scores were assigned based on the criteria described by Insall et al.7 Quadriceps strength was assessed subjectively during a physical examination by the senior author (MDR) as excellent, good, fair, or poor. This was equivalent to muscle strength grading against force on a five-point scale: 5 = excellent with full strength, 4 = good with mild weakness, 3 = fair with moderate weakness, and 2 = poor with significant weakness. Radiographs of the patella were taken with the knee flexed at 30°. Radiographic data were collected by one of the authors (BTB) who was blinded to the outcomes. Patellar thickness and tilt were measured on radiographs taken immediately and at the most recent followup (Fig 2).

Fig 2A
Fig 2A:
B. (A) A postoperative radiograph of the patella shows restoration of anatomic patellar thickness. (B) A radiograph taken 4 years postoperatively shows partial patellar resorption, resulting in loss of patellar thickness. The quadriceps maintains excellent function despite partial loss of patellar thickness.


Two of the six patients who had primary TKAs had subsequent reoperations, one because of infection and another because of aseptic allograft resorption and fragmentation. One of the three patients who had revision TKAs had a subsequent reoperation for infection. The remaining six patients (four who had primary TKAs and two who had revision TKAs) had improved quadriceps strength and Knee Society scores, and two patients had resolution of significant extensor lags. The six patients with intact reconstructions had average knee and function scores of 85 and 67, respectively. One patient (Patient 1) had complete resolution of a 35° extensor lag and another patient (Patient 5) had resolution of a 45° extensor lag despite the development of partial allograft resorption. The mean quadriceps strength increased an entire grade to 4.5 points.

The average patellar thickness 6 weeks postoperatively was 24 mm. This decreased to 15 mm at the most recent followup. The average patellar tilt was 10° 6 weeks postoperatively and 1° at the most recent followup.

There was a high failure rate. Two patients required amputations and one patient required allograft patella revision. Two of the failures occurred as a result of infections. Patient 6 had a history of hepatitis, malnutrition, cirrhosis, and ipsilateral foot infection with a fixed equinus contracture. The patient had symptomatic OA develop and was scheduled for TKA. The patient fell and sustained a comminuted intraarticular distal femur fracture that was treated with a constrained TKA. A postoperative hematoma developed requiring evacuation. Because of subsequent infection, we removed the implants and inserted an antibiotic cement spacer before performing delayed revision TKA. After a repeat, the patient elected to have an amputation rather than continue with surgical treatment. Patient 7 had a history of septic arthritis, which was treated with two-stage débridement, insertion of an antibiotic cement spacer, and delayed primary TKA. The patient has recurrent infection but refused arthrodesis. He was treated with a repeat two-stage débridement, insertion of antibiotic cement spacer, and delayed revision TKA. The patellar allografting was performed at the second revision TKA. However, recurrent infection developed and the patient elected amputation 5 months after revision TKA. The third failure in this group occurred in a patient who had diabetes and was HIV-positive and hepatitis B-positive who had bilateral whole patellar allograft reconstructions with TKAs. The initial results were extremely gratifying for the patient, who was nonambulatory before surgery. After bilateral TKAs with patellar allografting, the patient was able to return to unlimited walking activities. He could ascend and descend stairs foot-over-foot without support. However, fragmentation of the left patellar allograft occurred 1 year postoperatively. He also had symptoms of flexion instability develop. Revision to a thicker tibial insert and patellectomy were recommended, but the patient refused the patellectomy. He had repeat whole patellar allograft reconstruction and polyethylene insert exchange. At more than 3 years followup, the patient has an excellent outcome with no extensor lag, 135° ROM, and quadriceps strength of 5 with knee and function scores of 100 and 70, respectively.


Knee kinematics and biomechanics are altered after patellectomy.1,18 Some authors recommend using a posterior-stabilized or a more constrained TKA in this patient population to restore AP stability, whereas some studies showed no difference in outcome with a cruciate-retaining TKA.1,9,14,17 Restoration of patellar height with a whole patellar allograft may improve quadriceps strength, AP stability, and knee function.

The limitations of our study include a small study size, use of historical control groups, and only a 3-year followup. We placed allografts in patients who had medical comorbidities or history of infection, which may have increased the risk of postoperative infection. We also included primary and revision surgeries. It is difficult to discern whether the TKA or the patellar allograft was primarily responsible for the improvement in knee scores, quadriceps strength, and instability.

Reconstruction of the patella to restore quadriceps function may improve outcomes, therefore bone grafting procedures have been proposed to reconstruct the patella in patients with patellectomies who have TKAs. Techniques using autografts from the iliac crest, bone from the femoral TKA osteotomies, or femoral head and femoral condyle allografts to replace the patella have been described.2 The technique was used in five revision TKAs and one bilateral primary TKA. At an average followup of 75 months, there were six excellent results and one complication of reflex sympathetic dystrophy. 2 Autograft fixation did not deteriorate with time.2 A similar technique has been proposed using the resected tibial plateau.11,20 In eight patients, four of the autografts had to be resected because of loss of fixation with resultant extensor lag.11,20 Knee scores ranged from 55 to 95 points postoperatively. Another technique involves internal fixation of a flat piece of structural unicortical iliac crest bone graft to the patellar remnant.19

Nonvascularized autograft or allograft used to restore patellar height can resorb with time. The decreased thickness of a patellar bone graft may lead to decreased quadriceps strength and TKA function, although this was not the case during the 3-year followup of our patients. The early postoperative patellar thickness in our patients after allograft patella reconstruction was 24 mm, and results with this technique show an increase in quadriceps strength of one grade and resolution of two large extensor lags. Morselized nonstructural bone grafting can lead to alteration in the shape of the graft with time and decreased thickness, as was seen in a study on revision TKAs with a patellar shell.5 A series of nine patients with more than 2 years followup had significant improvements in mean knee and function scores, with postoperative scores of 91 and 84, respectively. However, four of the patients had a laterally molded patella after consolidation of the bone mass and a 10% loss of patellar height.5 In an iliac crest bone graft series, there was loss of 25% of patellar height over the average 75 months followup.2 Slow patellar resorption did occur in our series, although the average patellar thickness of 15 mm was not associated with a decrease in quadriceps strength with time. In a study of whole patellar allograft reconstruction used for salvage of a disrupted extensor mechanism, the change in thickness of patellar bone stock with time was not observed, presumably because the focus of the reconstructive technique is restoration of continuity of the extensor mechanism.3 Changes in thickness with time have not been reported in studies of autografts used for restoration of patellar thickness, but this may be a valuable predictor of long-term function and provide a quantitative method to compare results of different bone grafting techniques.2,5

Autograft techniques to restore patellar thickness after patellectomy require suture fixation of the bone graft into a synovial pouch to maintain its position. We used suture fixation between the allograft quadriceps tendon to host tendon, and allograft patellar ligament to host patellar ligament. This fixation provided reliable tendon to tendon healing to maintain the allograft position. There were no cases of allograft patellar migration.

Whole patellar allograft reconstruction can be an effective technique to increase patellar thickness and quadriceps function for patients who have TKAs after previous patellectomies. Patients who benefited the most were those who had large preoperative extensor lags or residual quadriceps weakness after TKAs with previous patellectomies. However, there was a high complication rate. The outcome does not seem favorable enough to warrant routine whole patellar allografting during TKA for a previous patellectomy because of the risk of allograft complications. The potential benefits of improved quadriceps mechanics may outweigh the risks of complications for patients with significant quadriceps weakness requiring TKAs after previous patellectomies.


1. Bayne O, Cameron HU. Total knee arthroplasty following patellectomy. Clin Orthop Relat Res. 1984;186:112-114.
2. Buechel FF. Patellar tendon bone grafting for patellectomized patients having total knee arthroplasty. Clin Orthop Relat Res. 1991;271:72-78.
3. Emerson RH Jr, Head WC, Malinin TI. Extensor mechanism reconstruction with an allograft after total knee arthroplasty. Clin Orthop Relat Res. 1994;303:79-85.
4. Figgie HE3rd, Goldberg VM, Heiple KG, Moller HS 3rd, Gordon NH. The influence of tibial-patellofemoral location on function of the knee in patients with the posterior stabilized condylar knee prosthesis. J Bone Joint Surg Am. 1986;68:1035-1040.
5. Hanssen AD. Bone-grafting for severe patellar bone loss during revision knee arthroplasty. J Bone Joint Surg Am. 2001;83:171-176.
6. Hitt K, Shurman JR2nd, Greene K, McCarthy J, Moskal J, Hoeman T, Mont MA. Anthropometric measurements of the human knee: correlation to the sizing of current knee arthroplasty systems. J Bone Joint Surg Am. 2003;85:115-122.
7. Insall JN, Dorr LD, Scott RD, Scott WN. Rationale for The Knee Society clinical rating system. Clin Orthop Relat Res. 1989;248: 13-14.
8. Joshi AB, Lee CM, Markovic L, Murphy JC, Hardinge K. Total knee arthroplasty after patellectomy. J Bone Joint Surg Br. 1994;76:926-929.
9. Kang JD, Papas SN, Rubash HE, McClain EJ. Total knee arthroplasty in patellectomized patients. J Arthroplasty. 1993;8:489-501.
10. Kelly MA,Insall JN. Patellectomy. OrthopClin North Am. 1986;17:289-294.
11. Lakshmanan P, Wilson C. Total knee arthroplasty in a patellectomised posterior cruciate ligament-deficient knee: a new technique of patellar tendon bone grafting. Knee. 2004;11:481-484.
12. Larson KR, Cracchiolo A, Dorey FJ, Finerman GA. Total knee arthroplasty in patients after patellectomy. Clin Orthop Relat Res. 1991;264:243-254.
13. Lennox DW, Hungerford DS, Krackow KA. Total knee arthroplasty following patellectomy. Clin Orthop Relat Res. 1987;223:220-224.
14. Marmor L. Unicompartmental knee arthroplasty following patellectomy. Clin Orthop Relat Res. 1987;218:164-166.
15. Martin SD, Haas SB, Insall JN. Primary total knee arthroplasty after patellectomy. J Bone Joint Surg Am. 1995;77:1323-1330.
16. Nazarian DG, Booth RE Jr. Extensor mechanism allografts in total knee arthroplasty. Clin Orthop Relat Res. 1999;367:123-129.
17. Paletta GA, Laskin RS. Total knee arthroplasty after a previous patellectomy. J Bone Joint Surg Am. 1995;77:1708-1712.
18. Pinskerova V, Maquet P, Freeman MA. The anatomic literature relating to the knee from 1836 to 1917: an historic note. Clin Orthop Relat Res. 2003;410:13-18.
19. Tabutin J. Osseous reconstruction of the patella with screwed autologous graft in the course of repeat prosthesis of the knee. Rev Chir Orthop Reparatrice Appar Mot. 1998;84:363-367.
20. Tirveilliot F, Migaud H, Tillie B, Vielpeau C, Flautre B, Gougeon F. Patellar reconstruction during total knee arthroplasty after previous patellectomy. Rev Chir Orthop Reparatrice Appar Mot. 2003; 89:613-620.
21. Wendt PP, Johnson RP. A study of quadriceps excursion, torque, and the effect of patellectomy on cadaver knees. J Bone Joint Surg Am. 1985;67:726-732.
© 2006 Lippincott Williams & Wilkins, Inc.