For the operative procedure, in 18 knees with patellar maltracking (17 patients), the lateral parapatellar approach17 was used to integrate a lateral release and to move the tibial tubercle fragment proximally and medially. In two knees (two patients) the medial parapatellar approach was used. In all knees an osteotomy of the tibial tubercle was done. After removal of the hinged prosthesis and lavage, vital bone was exposed by curettes and a high-speed burr (Midas Rex, Fort Worth, TX). At the tibial side, the defects were reconstructed with impaction grafting (16 knees) (Fig 2). In four knees with uncontained defects of the tibial wall, a sandwich technique was used (Fig 1D). The intramedullary canal was filled with solid cancellous slices of 10–15 mm height. After contouring, these allograft slices were hammered into the intramedullary canal to achieve tight press-fit fixation. Then, the central hole for the conical stem was created by a high-speed burr. Impaction of the cancellous and sandwich bone grafts was achieved using trial conical component stems with increasing diameters. The grafts were made from a mean of 2.6 femoral heads approximately equivalent to 250 cc of bone graft. At the femoral side, large structural allograft reconstructions were required in 16 knees: 10 femoral heads (Fig 2B), two proximal femoral allografts, and four distal femoral allografts (Fig 1C-D). All allografts used in this series were fresh-frozen and treated without radiation. The remaining four patients received impacted homologous cancellous bone from a mean of 2.3 femoral heads. Additional autologous cancellous bone and tricortical iliac crest grafts were used in three knees (three patients).
The LCS Classic Revision System (DePuy Johnson & Johnson, Warsaw, IN) is nonmodular, whereas the LCS-Universal (DePuy Johnson & Johnson) provides modular stem length and diameter. The LCS Classic Revision System has a fixed short conical stem that is proportional between sizes (Figs 1D, 2D). The length of the stem varies from 8.5–10.5 cm with a base diameter of 2–2.5 cm. Therefore, the stems could be inserted in variable positions in the host intramedullary canal and no offset components were necessary. The LCS Classic revision implants are porous-coated on the undersurfaces and on the proximal portion of the stem. In the current series, in 18 knees (17 patients) the LCS Classic prosthesis was used. In two patients receiving large bulk allografts, the LCS Universal prosthesis was used.
There were no posterior stabilizers or cams used in this series. Intrinsic stability at the articulating surfaces was provided by the congruency of the PE rotating platform and the femoral curvature. As a result of the two concavities of the liners, the jump factor for the standard version is 6.1 mm and 9.1 mm for the deep-dish version. A residual laxity of the collateral ligaments or the posterior capsule must be less than this critical limit to avoid spin-out of the platform or dislocation. Sixteen deep-dish and four standard PE platforms were used. The total height of the PE and metal tibial components was a mean 17.5 mm (range, 10–25 mm). For the thickness of the tibial metal component there are two choices, 5 and 15 mm, and the thickest PE platform is 20 mm. For 22.5 and 25 mm the thicker metal platform was used. Component fixation was done without cement in eight knees (seven patients) at the tibial and femoral sides, if sufficient primary stability was provided by the host bone and the bone grafts. In five knees (five patients), there were femoral cemented fixation and tibial cementless fixation (Fig 2B). Both implants were cemented in seven knees (seven patients). Cementation mostly was restricted to the cut bone surface (Fig 2B). For the more recent revisions in our series cemented anchorage or press-fit diaphyseal stem fixation with the modular revision prosthesis was preferred for very large defects.
At the time of insertion of the hinged prosthesis, the ligaments had been excised or cut at the epicondyles. However, at the revision operation, a scarred soft tissue envelope was seen in all knees. Consequently, stability could be reestablished with extensive subperiosteal release techniques that were done primarily at the tibial condyles. In all knees, a substantial posterior tibial and femoral subperiosteal release was necessary to cope with the discrepancy of the flexion and extension gap. Additionally the posterior femoral condyles were reconstructed meticulously with bulk allografts and adequately sized femoral components to provide stability in flexion (Figs 1C, 2B). During surgery, ligamentous stability and correct mobility of the rotating platform were tested using the trial components.
In four knees (four patients) with deep infection of the hinged prostheses, insertion of the LCS revision components and the bone grafts was done as a two-stage procedure after implantation of a spacer and eradication of the infection.
One to 3 days after surgery, the patients started walking with two crutches or a rolling walker. Range of motion exercises also were initiated at that time. Weightbearing was reduced to 15 kg for 10–12 weeks. Patients were advised to use a cane until they were confident with their stability.
The clinical assessment was done by two orthopaedic surgeons on the basis of the Knee Society score.14 Neither surgeon was involved in the revision surgery. Radiographic analysis included AP, lateral, and Merchant views. For location of radiolucent lines, the zones defined by the Knee Society were used.9 Additional AP views with varus and valgus stress were obtained within 3 months after surgery. Radiographic analysis was done by two of the authors (JF and BB). Clinical and radiographic followups were done 3 months after surgery and then annually. The mean followup was 5.3 years (range, 2.3–10.5 years). Three patients died by the time of the most recent evaluation. Two of these patients were included in the study, because the followup at the last examination was more than 2 years.
The study was approved according to the law and regulations of our institution and country. Written informed consent was obtained.
The clinical outcome essentially was improved. The Knee Society score increased from a preoperative mean of 21 points (range, 7–49 points) to a postoperative mean of 81.2 points (range, 59–94 points) at the last followup (Pearson correlation coefficients for the two observers, preoperative, 0.97393; postoperative, 0.96731). The preoperative functional Knee Society score was a mean 13.7 points (range, 0–45 points), and the last postoperative result was 81.2 points (range, 30–100 points) (Pearson correlation coefficients for the two observers, preoperative, 0.81815; postoperative, 0.95014). For the functional score, deductions mainly resulted from general infirmity. The two patients who died had Knee Society scores of 93 and 85 points and functional scores of 80 points each at 5 years followup.
Nineteen of the 20 LCS prostheses showed adequate clinical and radiographic fixation (Figs 1D, 2B). Two years after revision surgery, one painful, aseptically loosened cementless tibial component had to be revised. Circumferential radiolucent lines more than 2 mm thick were seen only in this patient. For the remaining patients, there was no implant migration. Radiolucent lines with a width less than 1 mm were seen in a few knees in Zones 1 and 4 at the femoral and tibial sides. The lines never occurred in combination. In five knees (five patients) a dense sclerotic halo was seen at the peripheral stem. There was no progression of this halo.
All massive bone grafts reinforced bone stock. There was no resorption or collapse (Figs 1D, 2B). The central cancellous bone grafts showed remodeling with the trabeculae positioned parallel to the direction of loading (Fig 2B-C). The tibial cancellous bone graft in the patient with tibial loosening had been integrated entirely at the rerevision. As a result, surgical technique clearly was facilitated.
In all 20 knees (19 patients) of the current series, adequate clinical and radiographic ligamentous stability was restored by soft tissue sleeve releases and bone reconstruction (Fig 2C). In one patient, posterior dislocation of the revision arthroplasty occurred after adequate trauma. Ligamentous stability had been seen at the last examination before the trauma and no giving-way phenomena had been reported. She was significantly overweight and had a contralateral amputation of the midfemur.
Problems of the extensor mechanism did not occur in this series. Although there had been tilting, subluxation, or luxation of the patella in most cases before the revision operation, there was normal patella alignment on all postoperative radiographs. In six knees (six patients) with patella replacement, bone ingrowth was visible. The femorotibial angle was a mean 5.8° (range, 1° – 11°) (Pearson correlation coefficients for the two observers, 0.89281).
There was one specific complication associated with our surgical technique and our choice of implant; the failure of the tibial component mentioned previously resulted from cementless fixation in an extensive cancellous bone graft. There were two unspecific major complications: one traumatic dislocation mentioned previously and one hematogenous infection of the revision prosthesis from a tooth infection 1.5 years after the revision surgery. The patient was treated with two-stage revision to an S-ROM mobile hinged-bearing implant (DePuy Johnson & Johnson). Rerevision with bone grafting and the LCS revision system was not attempted because of the patient’s general infirmity at the time of reimplantation. All three patients with major complications (tibial loosening, traumatic dislocation, and deep infection) were not included in the mean postoperative score. Minor complications in nine patients (nine knees) did not influence the outcome. Six of these patients (six knees) sustained a longitudinal fissure at the remnants of the tibial condyles at the time of prosthetic insertion. All fissures healed primarily after intraoperative screw fixation and bone grafting. Two postoperative hematomas were treated with irrigation. One patient sustained a traumatic tibial tubercle pull-off, requiring fixation with cerclage wires.
The new surgical approach was chosen to improve implant fixation, to reinforce bone stock, to restore ligament tension, and to avoid problems of the extensor mechanism.
As in the alternative studies, our series is not controlled, which makes comparisons with those studies difficult.1,13,15,18 Our study also is limited by the number of patients and the length of followup, because treatment of failed hinged prostheses is rare. Statistical analysis was not possible.
Revision from a hinge to a rotating platform prosthesis is logistically and technically demanding with a long operating time. For patients with low physical demands or patients in a reduced general medical condition hinged megaprostheses are a treatment option, but they rarely have been used in nonneoplastic knee revisions.23 In tumor treatment, Kawai et al16 reported 32 mechanical failures in 82 segmental replacements at a minimum of 2 years followup. If arthrodeses are used for failed hinges, the fusion rate is reported to reach only 56% as a result of poor remaining bone stock.5
In the literature there are only four systematic reports (68 cases) on revisions of hinges and the followups are shorter than in the current series (Table 1). In contrast to our series, hinges or highly constrained cemented prostheses were used. Inglis and Walker13 reported poor results for fixed–axis hinges in revisions of 40 failed hinged arthroplasties, because 16 rerevisions were done. The remaining three reports1,15,18 include only 28 patients; the clinical and radiographic outcomes and the complications of these patients are shown in Table 1.
To provide adequate fixation of revision arthroplasties, the least constraint possible should be used to avoid increased transmission forces at the bone and cement interfaces.22,23 As a result of the mobility of the LCS rotating platform, high conformity of the femoral shape is possible without adverse constraint. Therefore, in the current series, small prosthesis stems could be used. The clinical midterm and long-term benefits of the LCS rotating platform have been reported for primary and revision arthroplasties.4,6,24 If constrained knee prostheses need extensive intramedullary fixed stems to enlarge interface stability, complications also are caused by significant bone loss resulting from stress shielding.12 The distance of the shielding has been shown to be equivalent to the length of the stems.3 Barrack et al2 reported localized pain at the end of the stems in 11–14% after revision arthroplasty. With the use of modular press-fit stems in revision knee arthroplasty, Vince and Long25 presented a 23% failure rate of a constrained condylar-type prosthesis with 2–6 years followup. If a cemented, stemmed prosthesis is chosen for revision knee arthroplasty, the sclerotic surface of the intramedullary canal makes reliable interdigitation of the cement questionable.13,18
Massive bone grafts constantly reinforced bone stock in our series. The failed hinged prostheses caused extensive ice cream cone-like bone defects that needed higher volumes of cancellous bone grafting than have been reported to date.4 Impaction grafting and the sandwich technique have produced high primary stability and reliable secondary stability of the implants with evident bone remodeling. For revisions of bicondylar knee arthroplasties, successful results with impaction bone grafting in combination with cementless or cemented prostheses have been published.21,27 Bradley4 reported on 19 knees with impaction grafting of at least 90 cc morselized bone using the same revision implant as in the current series in their series with an average followup of 33 months. There was one mechanical failure. The Knee Society score, including the functional score, was a mean 147 points (range, 102–198 points). In the current series, structural allografts have shown convincing midterm results which are comparable with those published in a multicenter study.7 In contrast to impaction grafting, late infection and resorption may represent a higher risk in large bulk allografts in the long-term.7 Downgrading of the bone defects by the surgical technique used in the current series has an essential benefit. In case of a rerevision, the bone defects are less extensive compared with the situation at the last revision.
Why could ligamentous stability be restored in our series? After removal of hinged prostheses, Kim18 reported that there is a scar tissue sleeve present to provide soft tissue stability, which supports intrinsic stability of the revision implant. In the current series, this tissue sleeve was equilibrated by meticulous release techniques and joint line reconstruction with structural bone grafts. Ligament stability was provided in extension and flexion by structural allograft reconstruction of the posterior femur and by posterior femoral and tibial release. Minor residual soft tissue instability could be tolerated because of the conformity provided by the concavity of the LCS rotating platform PE insert. In the current prospective study, all failed hinges could be changed to a stable rotating platform prosthesis. Constrained or hinged prostheses were not used in this series. However, when persistent severe instability is seen during revision knee arthroplasty of hinges, a revision implant with high intrinsic stability should be available.
In contrast to revisions from hinges to hinges reported in the literature, our technique did not cause any problems of the extensor mechanism.1,15 This might result from the anatomic femoral groove of the component and the less constrained type of implant.
For treatment of failed hinged knee arthroplasties less constrained smaller components provided adequate fixation in our series. Massive bone grafts reliably reinforced bone stock. Extensive soft tissue releases restored ligament tension in all knees. Problems of the extensor mechanism did not occur. One cementless tibial component with extensive cancellous bone grafting became loose. In the long term the special advantage of the new procedure presented in our series may emerge from the reconstruction of ligament tension and bone stock.
1. Barrack RL, Lyons TR, Ingraham RQ, Johnson JC: The use of a modular rotating hinge component in salvage revision total knee arthroplasty. J Arthroplasty 15:858–866, 2000.
2. Barrack RL, Rorabeck C, Burt M, Sawhney J: Pain at the end of the stem after revision total knee arthroplasty. Clin Orthop 367:216–225, 1999.
3. Bourne RB, Finlay JB: The influence of tibial component intramedullary stems and implant-cortex contact on the strain distribution of the proximal tibia following total knee arthroplasty: An in vitro study. Clin Orthop 208:95–99, 1986.
4. Bradley GW: Revision total knee arthroplasty by impaction bone grafting. Clin Orthop 371:113–118, 2000.
5. Brodersen MP, Fitzgerald Jr RH, Peterson LF, Coventry MB, Bryan RS: Arthrodesis of the knee following failed total knee arthroplasty. J Bone Joint Surg 61A:181–185, 1979.
6. Buechel Sr FF, Buechel Jr FF, Pappas MJ, Dalessio J: Twenty-year evaluation of the New Jersey LCS rotating platform knee replacement. J Knee Surg 15:84–89, 2002.
7. Clatworthy MG, Ballance J, Brick GW, Chandler HP, Gross AE: The use of structural allograft for uncontained defects in revision total knee arthroplasty: A minimum five-year review. J Bone Joint Surg 83A:404–411, 2001.
8. Engh GA: Bone Defect Classification. In Engh GA, Rorabeck CH (eds). Revision Total Knee Arthroplasty. Baltimore, Williams & Wilkins 63–120, 1997.
9. Ewald F: The Knee Society total knee arthroplasty roentgenographic evaluation and scoring system. Clin Orthop 248:9–12, 1989.
10. Fitzek JG, Barden B: Rekonstruktionsmöglichkeiten nach Fehlschlägen in der Knieendoprothetik. In Imhoff AB (ed). Fortbildung Orthopädie. Darmstadt, Steinkopff 157–172, 2000.
11. Hanssen AD, Trousdale RT, Osmon DR: Patient outcome with reinfection following reimplantation for the infected total knee arthroplasty. Clin Orthop 321:55–67, 1995.
12. Hoeffel DP, Rubash HE: Revision total knee arthroplasty: Current rationale and techniques for femoral component revision. Clin Orthop 380:116–132, 2000.
13. Inglis AE, Walker PS: Revision of failed knee replacements using fixed-axis hinges. J Bone Joint Surg 73B:757–761, 1991.
14. Insall JN, Dorr LD, Scott RD, Scott WN: Rationale of the Knee Society clinical rating system. Clin Orthop 248:13–14, 1989.
15. Jones RE, Skedros JG, Chan AJ, Beauchamp DH, Harkins PC: Total knee arthroplasty using the S-ROM mobile-bearing hinge prosthesis. J Arthroplasty 16:279–287, 2001.
16. Kawai A, Lin PP, Boland PJ, Athanasian EA, Healey JH: Relationship between magnitude of resection, complication, and prosthetic survival after prosthetic knee reconstructions for distal femoral tumors. J Surg Oncol 70:109–115, 1999.
17. Keblish PA: The lateral approach to the valgus knee: Surgical technique and analysis of 53 cases with over two-year follow-up evaluation. Clin Orthop 271:52–62, 1991.
18. Kim YH: Salvage of failed hinge knee arthroplasty with a Total Condylar III type prosthesis. Clin Orthop 221:272–277, 1987.
19. Kirk PG: Selecting an Implant: A Comparison of Revision Implant Systems. In Engh GA, Rorabeck CH (eds). Revision Total Knee Arthroplasty. Baltimore, Williams & Wilkins 137–165, 1997.
20. Rinta-Kiikka I, Alberty A, Savilahti S, et al: The clinical and radiological outcome of the rotating hinged knee prostheses in the long-term. Ann Chir Gynaecol 86:349–356, 1997.
21. Samuelson KM: Bone grafting and noncemented revision arthroplasty of the knee. Clin Orthop 226:93–101, 1988.
22. Scuderi GR: Revision total knee arthroplasty: How much constraint is enough? Clin Orthop 392:300–305, 2001.
23. Springer BD, Hanssen AD, Sim FH, Lewallen DG: The kinematic rotating hinge prosthesis for complex knee arthroplasty. Clin Orthop 392:283–291, 2001.
24. Stiehl JB: World experience with Low Contact Stress mobile-bearing total knee arthroplasty: A literature review. Orthopedics 25(2 Suppl):S213–S217, 2002.
25. Vince KG, Long W: Revision knee arthroplasty: The limits of press fit medullary fixation. Clin Orthop 317:172–177, 1995.
26. Westrich GH, Mollano AV, Sculco TP, et al: Rotating hinge total knee arthroplasty in severely affected knees. Clin Orthop 379:195–208, 2000.
© 2004 Lippincott Williams & Wilkins, Inc.
27. Whiteside LA, Bicalho PS: Radiologic and histologic analysis of morselized allograft in revision total knee replacement. Clin Orthop 357:149–156, 1998.