The Journal of Arthroplasty xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis Young-Hoo Kim, MD , Jang-Won Park, MD, Jun-Shik Kim, MD, Hyun-Keun Oh, MD Joint Replacement Center, Ewha Woman’s University School of Medicine, Seoul, Republic of Korea
a r t i c l e
i n f o
Article history: Received 19 January 2015 Accepted 18 April 2015 Available online xxxx Keywords: clinical outcomes radiographic outcomes revision total knee arthroplasty constrained condylar knee prosthesis
a b s t r a c t The purpose of this study was to determine long-term clinical and radiographic results. One hundred and ninetyfour patients (228 knees) underwent revision TKA with use of a constrained condylar knee prosthesis. The mean duration of follow-up was 14.6 years (range, 11 to 16 years). The mean pre-revision Knee Society knee scores (43.5 points) and function scores (47.0 points), and Western Ontario and McMaster Universities Osteoarthritis index scores (88 points) were improved significantly (P = 0.002) to 85.6, 68.5, and 25 points, respectively, at 14.6 years follow-up. Eighteen knees (8%) had re-revision. Four knees were re-revised for infection. Kaplan– Meier survivorship analysis revealed that the 16-year rate of survival of the components was 94.7% as the end point of loosening and 92% as the end point of revision. © 2015 Elsevier Inc. All rights reserved.
Revision total knee arthroplasty (TKA) presents numerous challenges including stabilization of ligamentous laxity and marked bone loss. To manage some of these challenges, the legacy constrained condylar knee system (modification of the original Total Condylar III design) [1,2] was developed to resist coronal plane moments due to deficient soft-tissue constraints. The issue with constrained condylar total knee prosthesis is the advantage of added stability that comes at the price of potentially loosening due to transfer of forces to the other interfaces. In recent years, authors of several studies have investigated the clinical and radiographic results of revision TKAs with the use of legacy constrained condylar knee prosthesis (LCCK; Zimmer, Warsaw, Indiana) [3–6]. However, most of these studies had less than 10 years of follow-up and included a small number of patients [7,5,8]. The purpose of the current retrospective study was to determine the long-term clinical results of the patients after revision with these constrained prostheses, including the Knee Society knee and function scores [9], the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores [10] and radiographic results including fixation of the components and osteolysis. Materials and Methods From January 1995 to February 2003, 424 revision TKAs in 355 patients were carried out by a senior surgeon (YHK). Of the 355 patients, No author associated with this paper has disclosed any potential or pertinent conflicts which may be perceived to have impending conflict with this work. For full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2015.04.019. Reprint requests: Young-Hoo Kim, MD, The Joint Replacement Center, Ewha Woman’s University, MokDong Hospital, 911-1, MokDong, YangChun-Ku, Seoul, Republic of Korea 158-710.
107 patients were excluded because the constrained condylar knee prosthesis was not used for the revision (107 patients had a revision using a posterior cruciate ligament substitute prosthesis), 28 patients were excluded because they declined to participate, and 15 patients were excluded because they had infected knees. Two hundred and five patients were enrolled and all of these 205 patients required condylar constrained total knee prosthesis to substitute the distal femoral and/or proximal tibial bone defects. Data on 205 patients were entered into the database. However, eleven of the 205 patients were lost to follow-up or died. Therefore, only 194 patients (228 knees) comprised the study (Fig. 1). The study was approved by the institutional review board, and all patients provided informed consent before undergoing revision TKA. We retrospectively reviewed the data for prospectively followed patients. Ninety-seven of the 194 patients (50%) were reported previously, at a mean of 7.2 years of follow-up [3]. The current study group included 168 women and 26 men who had a mean age (and standard deviation) of 65.0 ± 10.4 years (range, 26 to 86 years) at the time of revision surgery. The mean body mass index (and standard deviation) was 26.9 ± 4.1 kg/m 2 (range, 19.1 to 40.5 kg/m 2). The mean follow-up was 14.6 years (range, 11 to 16 years). Previous surgeries were performed once in 110 knees (48%), two times in 88 knees (39%), and three times in 30 knees (13%). Evaluation of bone defect was performed using the Anderson Orthopaedic Research Institute (AORI) bone defect classification [11]. The reason for revision was aseptic loosening with AORI type F2B and/or T2B bone loss in 98 knees (43%), aseptic loosening with AORI type F2A bone defect and/or T2A in 68 knees (30%) and wear of tibial polyethylene with instability in 62 knees (27%). The demographic data on the patients are summarized in Table 1. Epidural normotensive anesthesia was used for 146 patients (75%), general anesthesia in 41 (21%), and spinal anesthesia in seven (4%). All
http://dx.doi.org/10.1016/j.arth.2015.04.019 0883-5403/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Kim Y-H, et al, Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.04.019
2
Y-H. Kim et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx Table 1 Demographic Data on 194 Patients (228 Knees).
Fig. 1. CONSORT flow diagram.
procedures were performed with tourniquet inflation to 250 mmHg. One hundred and 41 knees (62%) were approached through a previous anterior midline skin incisional scar, and the remaining 87 knees (38%) were approached through a previous medial parapatellar skin incisional scar. All patients received medial parapatellar capsulotomy. A quadriceps snip [12] was required in 164 knees (72%) for adequate exposure. Two knees (0.9%) had a tibial tubercle osteotomy. All knees were revised with the use of LCCK prosthesis. The indication for using this prosthesis was a bone stock deficiency with insufficient medial or lateral collateral ligaments but an intact, quadriceps mechanism. All of 42 posterior cruciate retaining TKA before revision had an insufficient posterior cruciate ligament and medial or lateral collateral ligaments because of bone defects. The medullary canal of femur was reamed to the point of mild resistance but not to so-called cortical chatter. A modular stem with distal and/or posterior femoral augments was used to fill the flexion space and to achieve balance with the extension gap. Femoral (lateral and/or medial condyles) augmentation was required in 145 knees (64%) and femoral cone was required in one knee (0.4%). Twenty centimeters long femoral medullary stem was used in 198 knees (87%) and 14.5 cm long femoral medullary stem was used in 30 knees (13%). The osseous surfaces were meticulously prepared with a saw or a highspeed burr to increase surface contact area. Simplex P cement (Howmedica, Rutherford, New Jersey) was digitally pressurized into the metaphyseal bone while the modular femoral component was press-fit in 198 knees (87%) (in these knees, metaphysis and the joint surfaces were cemented and medullary stem was press fitted). In the remaining 30 knees, with severe osteoporosis were fixed using cement in both stem and the femoral cut surface (in these knees, joint surfaces, metaphysis and medullary stem were cemented). The medullary canal of tibia was reamed to the point of mild resistance but not so-called cortical chatter. Proximal tibial augmentation (medial or lateral plateau) was required in 135 knees (59%). Twenty
Gender (M/F)
26/168
Agea (years) Heighta (cm) Weighta (kg) Body mass indexa (kg/m2) Primary diagnosis (no. [%] of patients/no. of knees) Osteoarthritis Osteonecrosis of the medial femoral condyle Childhood tuberculous arthritis Previous surgeries per knee Previous surgeries 1 time 2 times 3 times Preoperative knee laxity Medial collateral ligament laxity Lateral collateral ligament laxity Reasons for revision (no. [%] of knees) Aseptic loosening with AORI type F2B and/or T2Bb bone loss Aseptic loosening with AORI type F2A and/or T2A bone defects Wear of tibial polyethylene with instability Mean duration between primary and revision total knee arthroplasties (range) Prosthesis before revisionc (no. [%] of knees) Cementless Miller-Galante Cementless porous-coated anatomic Cemented low contact stress Cemented press-fit condylar Cemented anatomic modular Cemented NexGen cruciate-retaining flex Cemented Scorpio Cemented Advance Cemented Omnifit Duration of follow-upa (years)
65.0 ± 10.4 (26–86) 154.8 ± 8.4 (139–178) 64 ± 10 (43–92) 26.9 ± 4.1 (19.1–40.5)
186 (96)/216 4 (2)/8 4 (2)/4 1.5 times (1 to 3 times) 110 knees (48%) 88 knees (39%) 29 knees (13%) 228 knees 201 knees (88%) 27 knees (12%) 98 (43)
68 (30)
62 (27) 12.8 years (8 years to 28 years)
50 (22) 48 (21) 28 (12) 24 (11) 24 (11) 18 (8) 14 (6) 12 (6) 10 (4) 14.6 (11–16)
a
The values are given as the mean, with the range in parentheses. AORI: Anderson Orthopaedic Research Institute. c The NexGen and Miller-Galante Prostheses are manufactured by Zimmer (Warsaw, Indiana); the Scorpio and Omnifit prostheses, by Stryker (Mahwah, New Jersey); the Advance prosthesis is manufactured by Wright Medical (Arlington, Tennessee); cemented porous-coated anatomic prosthesis was manufactured by Howmedica (Mahwah, New Jersey); and the cemented low contact stress, cemented press-fit condylar and anatomic modular prostheses are manufactured by DePuy, Warsaw, Indiana. b
centimeters long tibial medullary stem was used in 210 knees (92%) and 14.5 cm long tibial stem was used in 18 knees (8%). Two hundred and ten knees (92%) were fixed with so-called hybrid fixation and the remaining 18 knees (8%) with severe osteoporosis were fixed using Simplex P cement in both tibial stem and the proximal cut surface. Average polyethylene thickness was 14.5 mm (range, 10 to 25 mm). The patellae were resurfaced in 203 knees (89%) during the time of revisions, which were not resurfaced during the primary TKA. Twenty knees (9%) were not able to be resurfaced and left alone because of insufficient bone stock. In the remaining five knees (2%), a trabecular patellar prosthesis was used to resurface the patella. All of 203 unresurfaced patellae during the time of primary surgery were found to have denuded cartilage at the time of revision surgery. Twenty knees that were resurfaced during the time of primary TKA were found to have worn of polyethylene patellar prosthesis. Mean tourniquet time was 71 minutes (range, 45 to 101 minutes) and mean surgical time from skin to skin was 99 minutes (range, 81 to 124 minutes). A splint was applied with the knee in 15° of flexion and was maintained for the first 24 hours after the operation to relieve pain. Subsequently, the knee then was placed on a continuous passive motion machine twice daily for 30 minutes each time for 10 to 15 days. No
Please cite this article as: Kim Y-H, et al, Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.04.019
Y-H. Kim et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
patient had continuous-passive-motion therapy at home. All of the patients began standing at the bedside or walking with crutches or a walker twice daily for 30 minutes each time under the supervision of a therapist on the second postoperative day. The patients used crutches or a walker with full weight-bearing for six weeks and used a cane when needed thereafter. The average length of stay in hospital was 16 days (range, 14 to 21 days). The readmission rate was 8%. The records of the 194 patients were entered into an ongoing computerized registry that was updated continuously. All of the knees were reviewed by a research associate not connected with the original surgery, and the data were entered into a computerized record. Follow-up evaluation was scheduled at postoperative intervals of three months, one year, and yearly thereafter. At these intervals, the patients were evaluated and radiographs were made. Preoperative and postoperative review data were recorded according to the scoring systems of the Knee Society [9], the WOMAC [10], and a 10-point visual analog scale for patient satisfaction. We grouped the visual analog scale responses into four categories [13]: ≤ 2 indicated fully dissatisfied; 3 to 5, somewhat dissatisfied; 6 to 8, satisfied, and 9 to 10, fully satisfied. The grouping is somewhat arbitrary and has not been validated, but we believed that it provides a subjective impression of the degree of patient satisfaction. Standing anteroposterior radiographs, including the femoral head and ankle, as well as supine, lateral, and skyline patellar radiographs of the knee were obtained under fluoroscopic control at each followup. Two observers, who were not involved with the surgery, assessed the radiographs for the alignment of the limb, the position of the components, and the presence and location of all radiolucent lines at the cement-bone or cement-implant interface, according to the recommendations of the Knee Society [9]. The skyline patellar radiographs were examined for patellar tilt, subluxation, or dislocation. If there was no alignment or position change of the components and a radiolucent line was incomplete and b 1 mm in width, the radiographic results were defined as good. If there was no circumferential radiolucent line or component migration, the component was defined as well fixed. Good interdigitation between cement and bone was defined when there was no radiolucent line between cement and bone. The joint line was determined on anteroposterior radiographs, made with the patient supine before and after revision by measuring the distance between the tip of the fibular head and the distal margin of the lateral femoral condyle preoperatively and the tip of the fibular head and the distal margin of the lateral femoral component postoperatively [14]. Statistical Analysis The normality of the distributions of the variables was tested with use of the Shapiro–Wilk test. Clinical variables measured preoperatively and at the final follow-up were compared by using a paired t test. A P value of b0.05 was considered to be significant. Kaplan–Meier survivorship analyses [15] were used in the analysis, with a re-revision of TKA as end point and the confidence interval at certain time points was calculated with the formula of Greenwood [16]. Results The median Knee Society knee and function scores were 43.5 points (range, 0 to 59 points) and 47.9 points (range 0 to 60 points) points, respectively, before the revision and improved to 85.6 points (range 16 to 91 points) and 58.5 points (range, 15 to 78 points), respectively, at final follow-up. These improvements were significant (P b 0.001). The Knee society knee and function scores were decreased gradually at each follow-up because of patients’ ages. The median preoperative pain score derived from KS knee score was 10.4 points (and standard deviation of 18.2 points) and 45.1 points (and standard deviation of 8.1 points) at the final follow-up. The median total preoperative and postoperative WOMAC scores were 88 points (and standard deviation of
3
15 points) and 28 points (and standard deviation of 15 points), respectively (P b 0.001). The difference in mean range of motion between preoperatively and postoperatively was not significant (P = 0.145). The mean range of motion was 94° preoperatively and 108° at the time of the final follow-up (Table 2). Three patients (2%) had femoral stem tip pain and four patients (2%) had tibial stem tip pain during weight bearing. Stem tip pain was mild in all patients, and did not require revision for stem tip pain. Mean joint line elevation was 2.0 mm in the current series. It did not affect on the revision rate. The position of the femoral and tibial components improved significantly after the revision surgery (P b 0.001). Limb alignment (femorotibial alignment) averaged 13.5° varus before revision and 6.1° of valgus after revision surgery. Incidence of outlier (N3°) was 100% of 228 knees before revision and 48 of 228 knees (21%) after revision (Fig. 2) (Table 3). Of the 163 revised knees with hybrid fixation, 133 knees (82%) that had good cement interdigitation (no radiolucency between cement and bone interface, so-called “whiteout”) at both the femoral metaphysis and the femoral cut surface, none had a radiolucent line or aseptic loosening. Nine knees (26%) had a circumferential complete radiolucent line with a width N2 mm around the femoral component and these femoral components were loose. These 9 knees underwent a re-revision of the femoral components. All of these 9 knees had AORI type F2B bone defects. The interval between revision and re-revision surgery was 7.3 years (range, 4 to 11 years). These knees were included in the calculation of the survival analysis. Two of the 30 knees that were fixed with the use of cement in both in the femoral metaphysis and around the modular femoral stem had an incomplete radiolucent line of b1 mm, and the remaining 28 knees had no radiolucent line at the interface between the distal end of the femur and the cement. Of the 179 revised knees, 148 knees (83%) that had good cement interdigitation "(white-out)" at both the tibial metaphysis and the tibial cut surface, none had a radiolucent line or aseptic loosening. Nine knees (29%) had a circumferential complete radiolucent line with a width of N2 mm at the bone-cement interface of the proximal end of the tibia. These 9 knees underwent a re-revision of the tibial components. All of these 9 knees had AORI type T2B bone defects. Interval between revision and re-revision surgery averaged 6.5 years (range 3 to 9 years). These knees were included in the survival analysis. Eleven (22%) of the 49 knees that were fixed with the use of cement both in the metaphysis and around the modular tibial stem had an incomplete radiolucent line of b1 mm at the interface between the cement and the proximal end of the tibia. The remaining 38 of the 49 knees (78%) had no radiolucent line (Table 4). Patellae were resurfaced in 203 knees (89%) during the time of revisions, which were not resurfaced during the primary TKA, but 20 knees (9%) were not able to resurface because of insufficient bone sock and left alone. These unresurfaced patellae did not affect revision rate. Furthermore, these 20 knees with unresurfaced patellae had similar clinical results as in 203 knees that had resurfaced patella except less knee flexion. The mean score for patient satisfaction after the operation was 8.2 (and standard deviation of 1.8 points), as assessed on the 10-point visual analog scale. Eighteen of 194 patients (9%) were fully dissatisfied, 10 (5%) were somewhat dissatisfied, 85 (44%) were satisfied, and the remaining 81 (42%) were fully satisfied.
Complications Eighteen knees (8%) were re-revised for femoral and/or tibial components. Twelve (67%) of 18 knees were re-revised for aseptic loosening, four knees were re-revised for an infection, and the remaining two knees were re-revised for instability. All of these knees were rerevised with the use of LCCK prosthesis. Two knees with infection underwent two-stage re-re-revision operation. There was no recurrence of infection at 14.6 years of follow-up.
Please cite this article as: Kim Y-H, et al, Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.04.019
4
Y-H. Kim et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
Table 2 Clinical Results at Each Follow-Up.
Knee Society knee scores (points)a Knee Society knee function score (points)a Range of motion of knee (degree)a WOMAC score (total score)a Proportion of missing data a
Preoperative
1 Year Postoperative
5 Years Postoperative
10 Years Postoperative
15 Years Postoperative
43.5 ± 22.8 (0–59) 47.9 ± 26.0 (0–10) 94 ± 16 (55–115) 88 ± 15 (68–94) –
89.7 ± 15.6 (65–95) 68.5 ± 16.9 (15–82) 105 ± 13 (65–130) 25 ± 13 (24–31) 2 patients
87.9 ± 14.5 (65–94) 64 ± 15.1 (14–80) 105 ± 14 (65–130) 24 ± 11 (23–32) 11 patients
86.1 ± 13.1 (65–92) 60 ± 13.7 (15–80) 106 ± 15 (67–130) 23 ± 14 (21–31) 11 patients
85.6 ± 15.4 (16–91) 58.8 ± 13.1 (15–78) 106 ± 14 (69–131) 25 ± 15 (21–36) 11 patients
The values are given as the median, standard deviation, with range in parenthesis. WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index.
Kaplan–Meier Survivorship Kaplan–Meier survivorship analysis at 5-year, the rate of survival of the components was 100.0% (95% confidence interval, 94.3% to 100%) as the end point of loosening or revision. Survivorship analysis at 10-year, the rate of survival of the components was 97.8% (95% confidence interval, 92.5% to 99.0%) as the end point of loosening or revision. Kaplan– Meier survivorship analysis at 16-year, the rate of survival of the components was 94.7% (95% confidence interval, 91.2% to 98%) as the end point of loosening and 92.1% (95% confidence interval, 89.1% to 98%) as the end point of revision. The worst-case scenario survivorship analysis, with revision as the end point for failure, estimated the 16-year rate of survival of the components to be 87.3% (95% confidence interval, 81.3 to 96.4%) (Fig. 3).
The ideal means of fixation in revision TKA remain controversial. There have been a number of studies documenting the results after revision TKA in which the condylar and metaphyseal surfaces are cemented, whereas the diaphyseal-engaging stems are press-fit.
Discussion Numerous studies [4,17–22] have reported satisfactory results with stemmed constrained revision TKAs. However, most of these studies reported short-to medium-term results, and others had a heterogeneous case mix, including different modes of fixation or complex primary procedures. The goals of our study were to report the long-term clinical and radiological outcomes of a homogeneous series of LCCK total knee prosthesis. Hartford et al [4] published the results of sixteen LCCK total knee prostheses, which were used in patients with aseptic loosening. In their series, mean preoperative KS knee score of 37 points improved to 83 points at the five-year follow-up. Harwin18 published the results of 22 Kinemax plus Superstabilizer (Stryker Orthopaedics, Mahwah, New Jersey) revision TKAs, including those used for septic and aseptic loosening. They reported an improvement in the mean KS knee score from 40 points pre-operatively to 83 points at 11 years after the operation, with excellent and good results in 83%. Manopoulos et al [23] published the results of 46 PFC Sigma Total Condylar III components (DePuy, Warsaw, Indiana) revision TKAs, including those used for septic and aseptic loosening and reported an improvement in the mean KS knee scores from 42 points preoperatively to 83.7 points at 8.5 years, with excellent and good clinical results in 80.4%. In our series, the median KS knee scores improved from 43.5 points preoperatively to 85.6 points at 14.6 years, with excellent and good clinical results in 86%. These clinical results in our series are consistent with those of other authors [4,22,23]. The factors associated with good results in our series were: obtained accurate flexion and extension gap using femoral or tibial augmentation; use of proper size and length of medullary stem; adequate exposure by using a quadriceps snip or tibial tubercle osteotomy; good cementing technique around the articular surface as well as metaphysis of the distal femur and proximal tibia; and predominant female patients. Wilke et al [7] reported that male gender significantly increased the risk of re-revision after the use of Total Condylar III type prosthesis.
Fig. 2. Radiographs of a 60-year-old woman who had TKA for osteoarthritis of left knee. Patient had a global instability of left knee. (A) An anteroposterior and lateral radiograph of left knee reveals medial and hyperextension instability of the press fit condylar (PFC; DePuy, Warsaw, Indiana) knee prosthesis of left knee. (B) An anteroposterior standing radiograph of left knee made 14 years after revision TKAs reveals that femoral and tibial components of the left LCCK prosthesis are well fixed in a satisfactory position.
Please cite this article as: Kim Y-H, et al, Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.04.019
Y-H. Kim et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
5
Table 3 Radiographic Results of Component Position. Parameter
Limb alignment (°) Outlier N 3° Femoral angle (°) Coronal Sagittal Tibial angle (°) Coronal Sagittal Joint line (mm) Posterior condylar offset (mm) a
Perioda
P Value
Before Revision
After Revision
13.5 of varus (9–29 varus) 228 knees (100%)
6.1 of valgus (0.5 of varus to 7.9 of valgus) 48 of 228 knees (21%)
89 (81–100) −0.8 (−7.9 to 13.1)
94 (89–100) 2.1 (−0.7 to 4.9)
0.015 0.012
79.8 (71–93) 84.8 (79–94) 14.9 (9.4–20.1) 23.9 (13.1–28.3)
91 (86–93) 88 (84–94) 16.9 (2.9–29.8) 24.7 (18.9–33.9)
0.041 0.038 0.131 0.281
b0.001 b0.0001
The values are given as the mean, with the range in parentheses.
Manopoulos et al [23] found that use of a so-called hybrid stem fixation confers both primary and secondary stability. They had only a single case of aseptic loosening at 7.5 years post-operatively. They confirmed the benefits of fixation using a hybrid technique. Shannon et al [18] reported 63 knees treated with revision TKA using hybrid fixation. At a mean follow-up of 5.75 years, 10 (16%) knees had mechanically failed, as defined by revision for aseptic loosening (six knees) or radiographic loosening (four knees). Gofton et al [24] reported the results of 89 revision TKAs using hybrid fixation at a mean of 5.9 years follow-up. When revision was used as the endpoint, the survivorship at 8.6 years was 93.5% in their series. Haas et al [2] reported the results of 76 revision TKA using hybrid fixation. They found that the eight-year component survivorship using revision as an endpoint was 83%. In our study, 85% of 210 hybrid TKAs had good fixation. However, there was femoral stem tip pain in three patients (2%) and tibial stem tip pain in four patients (2%). There have been several studies documenting the results after revision TKA in which the condylar, metaphyseal surfaces and stems are cemented. Mabry et al [25] reported the long-term results of 72 revision TKAs using modular, fully cemented femoral and tibial stems. Five-year and 10-year implant survivorship free of revision for aseptic failure was 98% and 92%, respectively. Whaley et al [26] reviewed the long-term
Table 4 Radiographic Results of Component Fixation. Hybrid fixation (femur)
198 of 228 knees (87%)
Good interdigitation at both the femoral metaphysis and the femoral cut surface (163 knees [82%]) Good interdigitation between the cement and the femoral cut surface, but poor interdigitation in the femoral metaphysis (35 knees [18%]) Cement fixation in both femoral cut surface and metaphysis and stem (femur) (30 knees, [13%])
133 of 163 knees (82%): no radiolucent line of 198
Hybrid fixation (tibia) Good interdigitation between cement and bone in both tibial metaphysis and the tibial cut surface (179 of 228 knees [79%]) Good interdigitation between the cement and the tibial cut surface, but poor interdigitation in the tibial metaphysis (31 of 179 knees [31%]) Cement fixation in both tibial cut surface and metaphysis (tibia) (49 knees [21%])
results of cemented long-stemmed kinematic stabilizer revision TKA. The 10-year survival of the components free of revision for any reason or mechanical failure was 94%. In one of the few studies to directly compare methods of stem fixation, Fehring et al [27] reviewed their results of metaphyseal engaging stems fixed with or without cement. These authors noted a higher rate of radiographic and clinical loosening in the cohort treated with uncemented stems. Although our findings are consistent with those of other authors [1–6,14–24], we recommend to use a hybrid fixation technique, to reduce the technical difficulty in removing the implant in the re-revision circumstances. All of re-revision cases in our series had an AORI type F3 or T3 bone defects. Therefore, reconstitution of bone defect using allograft or metal block is essential to avoid rerevision surgery. Our study has some limitations. It was not randomized and all of the patients in this series received LCCK prosthesis. This may limit the applicability of the findings to other centers. Furthermore, we have no interobserver variability to ensure in interpreting knee scores and radiographic findings, including radiolucent line and loosening. These assessments are somewhat unreliable and therefore the knee scores were prone to some ranges, and we might have underestimated the incidence of radiolucent line or loosening. In conclusion, although constraint of this knee prosthesis can potentially lead to increased aseptic loosening in the long-term, good clinical and radiographic results of this constrained knee prosthesis were achieved in the current series. Adequate restoration of bone stock deficiency and flexion and extension gap are important goals. Long-term study is mandatory to document the incidence of loosening of this constrained knee prosthesis.
26 of 35 knees (74%): radiolucent line b 1 mm 9 of 35 knees (26%): N 2 mm complete radiolucent line 2 knees: b1 mm radiolucent line 28 knees: no radiolucent line 179 of 228 knees (79%) 148 of 179 knees: no radiolucent line 22 of 31 knees (71%): no radiolucent line 9 of 31 knees (29%): N2 mm complete radiolucent line 11 of 49 knees (22%): incomplete radiolucent line b1 mm 38 of 49 knees (78%): no radiolucent line
Fig. 3. Kaplan–Meier survivorship analysis curve.
Please cite this article as: Kim Y-H, et al, Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.04.019
6
Y-H. Kim et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
References 1. Vince KG, Long W. Revision knee arthroplasty. The limits of press fit medullary fixation. Clin Orthop Relat Res 1995;317:172. 2. Haas SB, Insall JN, Montgomery W, et al. Revision total knee arthroplasty with use of modular components with stems inserted without cement. J Bone Joint Surg Am 1995;77(11):1700. 3. Kim Y-H, Kim J-S. Revision total knee arthroplasty with use of a constrained condylar knee prosthesis. J Bone Joint Surg Am 2009;91(6):1440. 4. Hartford JM, Goodman SB, Schurman DJ, et al. Complex primary and revision total knee arthroplasty using the condylar constrained prosthesis. An average 5-year follow-up. J Arthroplasty 1998;13(4):380. 5. Lee J-K, Kim S-J, Choi C-H, et al. Revision total knee arthroplasty using a constrained condylar knee prosthesis in conjunction with a posterior stabilized articular polyethylene. J Arthroplasty 2013;28(4):566. 6. Rodriguez JA, Shahane S, Rasquinha VJ, et al. Does the total condylar 3 constrained knee prosthesis predispose to failure of revision total knee replacement? J Bone Joint Surg Am 2003;85(Suppl. 4):153. 7. Wilke BK, Wagner ER, Trousdale RT. Long-term survival of semiconstrained total knee arthroplasty for revision surgery. J Arthroplasty 2014;29(5):1005. 8. Hwang SC, Keng JY, Nam DC, et al. Revision total knee arthroplasty with a cemented posterior stabilized, condylar constrained or fully constrained prosthesis: a minimum 2-year follow-up analysis. Clin Orthop Surg 2010;2(2):112. 9. Insall JN, Dorr LD, Scott RD, et al. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res 1989;248:13. 10. Bellamy N, Buchanan WW, Goldsmith CH, et al. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to anti-rheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988;15(12):1833. 11. Engh GA, Ammeen DJ. Bone loss with revision total knee arthroplasty: defect classification and alternatives for reconstruction. Instr Course Lect 1998;48:167. 12. Insall JN, editor. Surgery of the knee. New York: Churchill Livingstone; 1984. p. 41. 13. Barrack RL, McClure JT, Burak CF, et al. Revision total hip arthroplasty: the patients’ perspective. Clin Orthop Relat Res 2006;453:173.
14. Kim Y-H, Kook H-K, Kim J-S. Comparison of fixed-bearing and mobile-bearing total knee arthroplasties. Clin Orthop Relat Res 2001;392:101. 15. Kaplan EL, Meire P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457. 16. Greenwood M. A report on the natural duration of cancer. Report on public health and medical subjects, no. 33. London: His Majesty’s Stationary Office; 1926 1. 17. Donaldson WF, Sculco TP, Insall JN, et al. Total condylar III knee prosthesis: long-term follow-up study. Clin Orthop Relat Res 1988;226:21. 18. Shannon BD, Klassen JF, Rand JA, et al. Revision total knee arthroplasty with cemented components and uncemented intramedullary stems. J Arthroplasty 2003; 18(7 Suppl. 1):27. 19. Lachiewicz PF, Falatyn SP. Clinical and radiographic results of the Total Condylar III and Constrained Condylar total knee arthroplasty. J Arthroplasty 1996; 11(8):916. 20. Peters CL, Hennessey R, Barden RM, et al. Revision total knee arthroplasty with a cemented posterior-stabilized or constrained condylar prosthesis: a minimum 3year and average 5-uear follow-up study. J Arthroplasty 1997;12(8):896. 21. Peters CL, Erickson J, Kloepper RG, et al. Revision total knee arthroplasty with modular components inserted with metaphyseal cement and stems without cement. J Arthroplasty 2005;20(3):302. 22. Harwin SF. Revision total knee arthroplasty using the Kinemax Plus Superstabilizer prosthesis: minimum 10-year follow-up. Knee 2006;13(2):93. 23. Manopoulos P, Havet E, Pearce O, et al. Mid-to long-term results of revision total knee replacement using press-fit intramedullary stems with cemented femoral and tibial components. J Bone Joint Surg (Br) 2012;94(7):937. 24. Gofton WT, Tsigaras H, Butler RA, et al. Revision total knee arthroplasty: fixation with modular stems. Clin Orthop Relat Res 2002;404:158. 25. Mabry TM, Vessely MB, Schleck CD, et al. Revision total knee arthroplasty with modular cemented stems. Long-term follow-up. J Arthroplasty 2007; 22(6 Suppl. 2):100. 26. Whaley AL, Trousdale RT, Rand JA, et al. Cemented long-stem revision total knee arthroplasty. J Arthroplasty 2003;18(5):592. 27. Fehring TK, Odum S, Olekson C, et al. Stem fixation win revision total knee arthroplasty: a comparative analysis. Clin Orthop Relat Res 2003;416:217.
Please cite this article as: Kim Y-H, et al, Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.04.019
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis Young-Hoo Kim, MD , Jang-Won Park, MD, Jun-Shik Kim, MD, Hyun-Keun Oh, MD Joint Replacement Center, Ewha Woman’s University School of Medicine, Seoul, Republic of Korea
a r t i c l e
i n f o
Article history: Received 19 January 2015 Accepted 18 April 2015 Available online xxxx Keywords: clinical outcomes radiographic outcomes revision total knee arthroplasty constrained condylar knee prosthesis
a b s t r a c t The purpose of this study was to determine long-term clinical and radiographic results. One hundred and ninetyfour patients (228 knees) underwent revision TKA with use of a constrained condylar knee prosthesis. The mean duration of follow-up was 14.6 years (range, 11 to 16 years). The mean pre-revision Knee Society knee scores (43.5 points) and function scores (47.0 points), and Western Ontario and McMaster Universities Osteoarthritis index scores (88 points) were improved significantly (P = 0.002) to 85.6, 68.5, and 25 points, respectively, at 14.6 years follow-up. Eighteen knees (8%) had re-revision. Four knees were re-revised for infection. Kaplan– Meier survivorship analysis revealed that the 16-year rate of survival of the components was 94.7% as the end point of loosening and 92% as the end point of revision. © 2015 Elsevier Inc. All rights reserved.
Revision total knee arthroplasty (TKA) presents numerous challenges including stabilization of ligamentous laxity and marked bone loss. To manage some of these challenges, the legacy constrained condylar knee system (modification of the original Total Condylar III design) [1,2] was developed to resist coronal plane moments due to deficient soft-tissue constraints. The issue with constrained condylar total knee prosthesis is the advantage of added stability that comes at the price of potentially loosening due to transfer of forces to the other interfaces. In recent years, authors of several studies have investigated the clinical and radiographic results of revision TKAs with the use of legacy constrained condylar knee prosthesis (LCCK; Zimmer, Warsaw, Indiana) [3–6]. However, most of these studies had less than 10 years of follow-up and included a small number of patients [7,5,8]. The purpose of the current retrospective study was to determine the long-term clinical results of the patients after revision with these constrained prostheses, including the Knee Society knee and function scores [9], the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores [10] and radiographic results including fixation of the components and osteolysis. Materials and Methods From January 1995 to February 2003, 424 revision TKAs in 355 patients were carried out by a senior surgeon (YHK). Of the 355 patients, No author associated with this paper has disclosed any potential or pertinent conflicts which may be perceived to have impending conflict with this work. For full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2015.04.019. Reprint requests: Young-Hoo Kim, MD, The Joint Replacement Center, Ewha Woman’s University, MokDong Hospital, 911-1, MokDong, YangChun-Ku, Seoul, Republic of Korea 158-710.
107 patients were excluded because the constrained condylar knee prosthesis was not used for the revision (107 patients had a revision using a posterior cruciate ligament substitute prosthesis), 28 patients were excluded because they declined to participate, and 15 patients were excluded because they had infected knees. Two hundred and five patients were enrolled and all of these 205 patients required condylar constrained total knee prosthesis to substitute the distal femoral and/or proximal tibial bone defects. Data on 205 patients were entered into the database. However, eleven of the 205 patients were lost to follow-up or died. Therefore, only 194 patients (228 knees) comprised the study (Fig. 1). The study was approved by the institutional review board, and all patients provided informed consent before undergoing revision TKA. We retrospectively reviewed the data for prospectively followed patients. Ninety-seven of the 194 patients (50%) were reported previously, at a mean of 7.2 years of follow-up [3]. The current study group included 168 women and 26 men who had a mean age (and standard deviation) of 65.0 ± 10.4 years (range, 26 to 86 years) at the time of revision surgery. The mean body mass index (and standard deviation) was 26.9 ± 4.1 kg/m 2 (range, 19.1 to 40.5 kg/m 2). The mean follow-up was 14.6 years (range, 11 to 16 years). Previous surgeries were performed once in 110 knees (48%), two times in 88 knees (39%), and three times in 30 knees (13%). Evaluation of bone defect was performed using the Anderson Orthopaedic Research Institute (AORI) bone defect classification [11]. The reason for revision was aseptic loosening with AORI type F2B and/or T2B bone loss in 98 knees (43%), aseptic loosening with AORI type F2A bone defect and/or T2A in 68 knees (30%) and wear of tibial polyethylene with instability in 62 knees (27%). The demographic data on the patients are summarized in Table 1. Epidural normotensive anesthesia was used for 146 patients (75%), general anesthesia in 41 (21%), and spinal anesthesia in seven (4%). All
http://dx.doi.org/10.1016/j.arth.2015.04.019 0883-5403/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Kim Y-H, et al, Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.04.019
2
Y-H. Kim et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx Table 1 Demographic Data on 194 Patients (228 Knees).
Fig. 1. CONSORT flow diagram.
procedures were performed with tourniquet inflation to 250 mmHg. One hundred and 41 knees (62%) were approached through a previous anterior midline skin incisional scar, and the remaining 87 knees (38%) were approached through a previous medial parapatellar skin incisional scar. All patients received medial parapatellar capsulotomy. A quadriceps snip [12] was required in 164 knees (72%) for adequate exposure. Two knees (0.9%) had a tibial tubercle osteotomy. All knees were revised with the use of LCCK prosthesis. The indication for using this prosthesis was a bone stock deficiency with insufficient medial or lateral collateral ligaments but an intact, quadriceps mechanism. All of 42 posterior cruciate retaining TKA before revision had an insufficient posterior cruciate ligament and medial or lateral collateral ligaments because of bone defects. The medullary canal of femur was reamed to the point of mild resistance but not to so-called cortical chatter. A modular stem with distal and/or posterior femoral augments was used to fill the flexion space and to achieve balance with the extension gap. Femoral (lateral and/or medial condyles) augmentation was required in 145 knees (64%) and femoral cone was required in one knee (0.4%). Twenty centimeters long femoral medullary stem was used in 198 knees (87%) and 14.5 cm long femoral medullary stem was used in 30 knees (13%). The osseous surfaces were meticulously prepared with a saw or a highspeed burr to increase surface contact area. Simplex P cement (Howmedica, Rutherford, New Jersey) was digitally pressurized into the metaphyseal bone while the modular femoral component was press-fit in 198 knees (87%) (in these knees, metaphysis and the joint surfaces were cemented and medullary stem was press fitted). In the remaining 30 knees, with severe osteoporosis were fixed using cement in both stem and the femoral cut surface (in these knees, joint surfaces, metaphysis and medullary stem were cemented). The medullary canal of tibia was reamed to the point of mild resistance but not so-called cortical chatter. Proximal tibial augmentation (medial or lateral plateau) was required in 135 knees (59%). Twenty
Gender (M/F)
26/168
Agea (years) Heighta (cm) Weighta (kg) Body mass indexa (kg/m2) Primary diagnosis (no. [%] of patients/no. of knees) Osteoarthritis Osteonecrosis of the medial femoral condyle Childhood tuberculous arthritis Previous surgeries per knee Previous surgeries 1 time 2 times 3 times Preoperative knee laxity Medial collateral ligament laxity Lateral collateral ligament laxity Reasons for revision (no. [%] of knees) Aseptic loosening with AORI type F2B and/or T2Bb bone loss Aseptic loosening with AORI type F2A and/or T2A bone defects Wear of tibial polyethylene with instability Mean duration between primary and revision total knee arthroplasties (range) Prosthesis before revisionc (no. [%] of knees) Cementless Miller-Galante Cementless porous-coated anatomic Cemented low contact stress Cemented press-fit condylar Cemented anatomic modular Cemented NexGen cruciate-retaining flex Cemented Scorpio Cemented Advance Cemented Omnifit Duration of follow-upa (years)
65.0 ± 10.4 (26–86) 154.8 ± 8.4 (139–178) 64 ± 10 (43–92) 26.9 ± 4.1 (19.1–40.5)
186 (96)/216 4 (2)/8 4 (2)/4 1.5 times (1 to 3 times) 110 knees (48%) 88 knees (39%) 29 knees (13%) 228 knees 201 knees (88%) 27 knees (12%) 98 (43)
68 (30)
62 (27) 12.8 years (8 years to 28 years)
50 (22) 48 (21) 28 (12) 24 (11) 24 (11) 18 (8) 14 (6) 12 (6) 10 (4) 14.6 (11–16)
a
The values are given as the mean, with the range in parentheses. AORI: Anderson Orthopaedic Research Institute. c The NexGen and Miller-Galante Prostheses are manufactured by Zimmer (Warsaw, Indiana); the Scorpio and Omnifit prostheses, by Stryker (Mahwah, New Jersey); the Advance prosthesis is manufactured by Wright Medical (Arlington, Tennessee); cemented porous-coated anatomic prosthesis was manufactured by Howmedica (Mahwah, New Jersey); and the cemented low contact stress, cemented press-fit condylar and anatomic modular prostheses are manufactured by DePuy, Warsaw, Indiana. b
centimeters long tibial medullary stem was used in 210 knees (92%) and 14.5 cm long tibial stem was used in 18 knees (8%). Two hundred and ten knees (92%) were fixed with so-called hybrid fixation and the remaining 18 knees (8%) with severe osteoporosis were fixed using Simplex P cement in both tibial stem and the proximal cut surface. Average polyethylene thickness was 14.5 mm (range, 10 to 25 mm). The patellae were resurfaced in 203 knees (89%) during the time of revisions, which were not resurfaced during the primary TKA. Twenty knees (9%) were not able to be resurfaced and left alone because of insufficient bone stock. In the remaining five knees (2%), a trabecular patellar prosthesis was used to resurface the patella. All of 203 unresurfaced patellae during the time of primary surgery were found to have denuded cartilage at the time of revision surgery. Twenty knees that were resurfaced during the time of primary TKA were found to have worn of polyethylene patellar prosthesis. Mean tourniquet time was 71 minutes (range, 45 to 101 minutes) and mean surgical time from skin to skin was 99 minutes (range, 81 to 124 minutes). A splint was applied with the knee in 15° of flexion and was maintained for the first 24 hours after the operation to relieve pain. Subsequently, the knee then was placed on a continuous passive motion machine twice daily for 30 minutes each time for 10 to 15 days. No
Please cite this article as: Kim Y-H, et al, Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.04.019
Y-H. Kim et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
patient had continuous-passive-motion therapy at home. All of the patients began standing at the bedside or walking with crutches or a walker twice daily for 30 minutes each time under the supervision of a therapist on the second postoperative day. The patients used crutches or a walker with full weight-bearing for six weeks and used a cane when needed thereafter. The average length of stay in hospital was 16 days (range, 14 to 21 days). The readmission rate was 8%. The records of the 194 patients were entered into an ongoing computerized registry that was updated continuously. All of the knees were reviewed by a research associate not connected with the original surgery, and the data were entered into a computerized record. Follow-up evaluation was scheduled at postoperative intervals of three months, one year, and yearly thereafter. At these intervals, the patients were evaluated and radiographs were made. Preoperative and postoperative review data were recorded according to the scoring systems of the Knee Society [9], the WOMAC [10], and a 10-point visual analog scale for patient satisfaction. We grouped the visual analog scale responses into four categories [13]: ≤ 2 indicated fully dissatisfied; 3 to 5, somewhat dissatisfied; 6 to 8, satisfied, and 9 to 10, fully satisfied. The grouping is somewhat arbitrary and has not been validated, but we believed that it provides a subjective impression of the degree of patient satisfaction. Standing anteroposterior radiographs, including the femoral head and ankle, as well as supine, lateral, and skyline patellar radiographs of the knee were obtained under fluoroscopic control at each followup. Two observers, who were not involved with the surgery, assessed the radiographs for the alignment of the limb, the position of the components, and the presence and location of all radiolucent lines at the cement-bone or cement-implant interface, according to the recommendations of the Knee Society [9]. The skyline patellar radiographs were examined for patellar tilt, subluxation, or dislocation. If there was no alignment or position change of the components and a radiolucent line was incomplete and b 1 mm in width, the radiographic results were defined as good. If there was no circumferential radiolucent line or component migration, the component was defined as well fixed. Good interdigitation between cement and bone was defined when there was no radiolucent line between cement and bone. The joint line was determined on anteroposterior radiographs, made with the patient supine before and after revision by measuring the distance between the tip of the fibular head and the distal margin of the lateral femoral condyle preoperatively and the tip of the fibular head and the distal margin of the lateral femoral component postoperatively [14]. Statistical Analysis The normality of the distributions of the variables was tested with use of the Shapiro–Wilk test. Clinical variables measured preoperatively and at the final follow-up were compared by using a paired t test. A P value of b0.05 was considered to be significant. Kaplan–Meier survivorship analyses [15] were used in the analysis, with a re-revision of TKA as end point and the confidence interval at certain time points was calculated with the formula of Greenwood [16]. Results The median Knee Society knee and function scores were 43.5 points (range, 0 to 59 points) and 47.9 points (range 0 to 60 points) points, respectively, before the revision and improved to 85.6 points (range 16 to 91 points) and 58.5 points (range, 15 to 78 points), respectively, at final follow-up. These improvements were significant (P b 0.001). The Knee society knee and function scores were decreased gradually at each follow-up because of patients’ ages. The median preoperative pain score derived from KS knee score was 10.4 points (and standard deviation of 18.2 points) and 45.1 points (and standard deviation of 8.1 points) at the final follow-up. The median total preoperative and postoperative WOMAC scores were 88 points (and standard deviation of
3
15 points) and 28 points (and standard deviation of 15 points), respectively (P b 0.001). The difference in mean range of motion between preoperatively and postoperatively was not significant (P = 0.145). The mean range of motion was 94° preoperatively and 108° at the time of the final follow-up (Table 2). Three patients (2%) had femoral stem tip pain and four patients (2%) had tibial stem tip pain during weight bearing. Stem tip pain was mild in all patients, and did not require revision for stem tip pain. Mean joint line elevation was 2.0 mm in the current series. It did not affect on the revision rate. The position of the femoral and tibial components improved significantly after the revision surgery (P b 0.001). Limb alignment (femorotibial alignment) averaged 13.5° varus before revision and 6.1° of valgus after revision surgery. Incidence of outlier (N3°) was 100% of 228 knees before revision and 48 of 228 knees (21%) after revision (Fig. 2) (Table 3). Of the 163 revised knees with hybrid fixation, 133 knees (82%) that had good cement interdigitation (no radiolucency between cement and bone interface, so-called “whiteout”) at both the femoral metaphysis and the femoral cut surface, none had a radiolucent line or aseptic loosening. Nine knees (26%) had a circumferential complete radiolucent line with a width N2 mm around the femoral component and these femoral components were loose. These 9 knees underwent a re-revision of the femoral components. All of these 9 knees had AORI type F2B bone defects. The interval between revision and re-revision surgery was 7.3 years (range, 4 to 11 years). These knees were included in the calculation of the survival analysis. Two of the 30 knees that were fixed with the use of cement in both in the femoral metaphysis and around the modular femoral stem had an incomplete radiolucent line of b1 mm, and the remaining 28 knees had no radiolucent line at the interface between the distal end of the femur and the cement. Of the 179 revised knees, 148 knees (83%) that had good cement interdigitation "(white-out)" at both the tibial metaphysis and the tibial cut surface, none had a radiolucent line or aseptic loosening. Nine knees (29%) had a circumferential complete radiolucent line with a width of N2 mm at the bone-cement interface of the proximal end of the tibia. These 9 knees underwent a re-revision of the tibial components. All of these 9 knees had AORI type T2B bone defects. Interval between revision and re-revision surgery averaged 6.5 years (range 3 to 9 years). These knees were included in the survival analysis. Eleven (22%) of the 49 knees that were fixed with the use of cement both in the metaphysis and around the modular tibial stem had an incomplete radiolucent line of b1 mm at the interface between the cement and the proximal end of the tibia. The remaining 38 of the 49 knees (78%) had no radiolucent line (Table 4). Patellae were resurfaced in 203 knees (89%) during the time of revisions, which were not resurfaced during the primary TKA, but 20 knees (9%) were not able to resurface because of insufficient bone sock and left alone. These unresurfaced patellae did not affect revision rate. Furthermore, these 20 knees with unresurfaced patellae had similar clinical results as in 203 knees that had resurfaced patella except less knee flexion. The mean score for patient satisfaction after the operation was 8.2 (and standard deviation of 1.8 points), as assessed on the 10-point visual analog scale. Eighteen of 194 patients (9%) were fully dissatisfied, 10 (5%) were somewhat dissatisfied, 85 (44%) were satisfied, and the remaining 81 (42%) were fully satisfied.
Complications Eighteen knees (8%) were re-revised for femoral and/or tibial components. Twelve (67%) of 18 knees were re-revised for aseptic loosening, four knees were re-revised for an infection, and the remaining two knees were re-revised for instability. All of these knees were rerevised with the use of LCCK prosthesis. Two knees with infection underwent two-stage re-re-revision operation. There was no recurrence of infection at 14.6 years of follow-up.
Please cite this article as: Kim Y-H, et al, Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.04.019
4
Y-H. Kim et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
Table 2 Clinical Results at Each Follow-Up.
Knee Society knee scores (points)a Knee Society knee function score (points)a Range of motion of knee (degree)a WOMAC score (total score)a Proportion of missing data a
Preoperative
1 Year Postoperative
5 Years Postoperative
10 Years Postoperative
15 Years Postoperative
43.5 ± 22.8 (0–59) 47.9 ± 26.0 (0–10) 94 ± 16 (55–115) 88 ± 15 (68–94) –
89.7 ± 15.6 (65–95) 68.5 ± 16.9 (15–82) 105 ± 13 (65–130) 25 ± 13 (24–31) 2 patients
87.9 ± 14.5 (65–94) 64 ± 15.1 (14–80) 105 ± 14 (65–130) 24 ± 11 (23–32) 11 patients
86.1 ± 13.1 (65–92) 60 ± 13.7 (15–80) 106 ± 15 (67–130) 23 ± 14 (21–31) 11 patients
85.6 ± 15.4 (16–91) 58.8 ± 13.1 (15–78) 106 ± 14 (69–131) 25 ± 15 (21–36) 11 patients
The values are given as the median, standard deviation, with range in parenthesis. WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index.
Kaplan–Meier Survivorship Kaplan–Meier survivorship analysis at 5-year, the rate of survival of the components was 100.0% (95% confidence interval, 94.3% to 100%) as the end point of loosening or revision. Survivorship analysis at 10-year, the rate of survival of the components was 97.8% (95% confidence interval, 92.5% to 99.0%) as the end point of loosening or revision. Kaplan– Meier survivorship analysis at 16-year, the rate of survival of the components was 94.7% (95% confidence interval, 91.2% to 98%) as the end point of loosening and 92.1% (95% confidence interval, 89.1% to 98%) as the end point of revision. The worst-case scenario survivorship analysis, with revision as the end point for failure, estimated the 16-year rate of survival of the components to be 87.3% (95% confidence interval, 81.3 to 96.4%) (Fig. 3).
The ideal means of fixation in revision TKA remain controversial. There have been a number of studies documenting the results after revision TKA in which the condylar and metaphyseal surfaces are cemented, whereas the diaphyseal-engaging stems are press-fit.
Discussion Numerous studies [4,17–22] have reported satisfactory results with stemmed constrained revision TKAs. However, most of these studies reported short-to medium-term results, and others had a heterogeneous case mix, including different modes of fixation or complex primary procedures. The goals of our study were to report the long-term clinical and radiological outcomes of a homogeneous series of LCCK total knee prosthesis. Hartford et al [4] published the results of sixteen LCCK total knee prostheses, which were used in patients with aseptic loosening. In their series, mean preoperative KS knee score of 37 points improved to 83 points at the five-year follow-up. Harwin18 published the results of 22 Kinemax plus Superstabilizer (Stryker Orthopaedics, Mahwah, New Jersey) revision TKAs, including those used for septic and aseptic loosening. They reported an improvement in the mean KS knee score from 40 points pre-operatively to 83 points at 11 years after the operation, with excellent and good results in 83%. Manopoulos et al [23] published the results of 46 PFC Sigma Total Condylar III components (DePuy, Warsaw, Indiana) revision TKAs, including those used for septic and aseptic loosening and reported an improvement in the mean KS knee scores from 42 points preoperatively to 83.7 points at 8.5 years, with excellent and good clinical results in 80.4%. In our series, the median KS knee scores improved from 43.5 points preoperatively to 85.6 points at 14.6 years, with excellent and good clinical results in 86%. These clinical results in our series are consistent with those of other authors [4,22,23]. The factors associated with good results in our series were: obtained accurate flexion and extension gap using femoral or tibial augmentation; use of proper size and length of medullary stem; adequate exposure by using a quadriceps snip or tibial tubercle osteotomy; good cementing technique around the articular surface as well as metaphysis of the distal femur and proximal tibia; and predominant female patients. Wilke et al [7] reported that male gender significantly increased the risk of re-revision after the use of Total Condylar III type prosthesis.
Fig. 2. Radiographs of a 60-year-old woman who had TKA for osteoarthritis of left knee. Patient had a global instability of left knee. (A) An anteroposterior and lateral radiograph of left knee reveals medial and hyperextension instability of the press fit condylar (PFC; DePuy, Warsaw, Indiana) knee prosthesis of left knee. (B) An anteroposterior standing radiograph of left knee made 14 years after revision TKAs reveals that femoral and tibial components of the left LCCK prosthesis are well fixed in a satisfactory position.
Please cite this article as: Kim Y-H, et al, Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.04.019
Y-H. Kim et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
5
Table 3 Radiographic Results of Component Position. Parameter
Limb alignment (°) Outlier N 3° Femoral angle (°) Coronal Sagittal Tibial angle (°) Coronal Sagittal Joint line (mm) Posterior condylar offset (mm) a
Perioda
P Value
Before Revision
After Revision
13.5 of varus (9–29 varus) 228 knees (100%)
6.1 of valgus (0.5 of varus to 7.9 of valgus) 48 of 228 knees (21%)
89 (81–100) −0.8 (−7.9 to 13.1)
94 (89–100) 2.1 (−0.7 to 4.9)
0.015 0.012
79.8 (71–93) 84.8 (79–94) 14.9 (9.4–20.1) 23.9 (13.1–28.3)
91 (86–93) 88 (84–94) 16.9 (2.9–29.8) 24.7 (18.9–33.9)
0.041 0.038 0.131 0.281
b0.001 b0.0001
The values are given as the mean, with the range in parentheses.
Manopoulos et al [23] found that use of a so-called hybrid stem fixation confers both primary and secondary stability. They had only a single case of aseptic loosening at 7.5 years post-operatively. They confirmed the benefits of fixation using a hybrid technique. Shannon et al [18] reported 63 knees treated with revision TKA using hybrid fixation. At a mean follow-up of 5.75 years, 10 (16%) knees had mechanically failed, as defined by revision for aseptic loosening (six knees) or radiographic loosening (four knees). Gofton et al [24] reported the results of 89 revision TKAs using hybrid fixation at a mean of 5.9 years follow-up. When revision was used as the endpoint, the survivorship at 8.6 years was 93.5% in their series. Haas et al [2] reported the results of 76 revision TKA using hybrid fixation. They found that the eight-year component survivorship using revision as an endpoint was 83%. In our study, 85% of 210 hybrid TKAs had good fixation. However, there was femoral stem tip pain in three patients (2%) and tibial stem tip pain in four patients (2%). There have been several studies documenting the results after revision TKA in which the condylar, metaphyseal surfaces and stems are cemented. Mabry et al [25] reported the long-term results of 72 revision TKAs using modular, fully cemented femoral and tibial stems. Five-year and 10-year implant survivorship free of revision for aseptic failure was 98% and 92%, respectively. Whaley et al [26] reviewed the long-term
Table 4 Radiographic Results of Component Fixation. Hybrid fixation (femur)
198 of 228 knees (87%)
Good interdigitation at both the femoral metaphysis and the femoral cut surface (163 knees [82%]) Good interdigitation between the cement and the femoral cut surface, but poor interdigitation in the femoral metaphysis (35 knees [18%]) Cement fixation in both femoral cut surface and metaphysis and stem (femur) (30 knees, [13%])
133 of 163 knees (82%): no radiolucent line of 198
Hybrid fixation (tibia) Good interdigitation between cement and bone in both tibial metaphysis and the tibial cut surface (179 of 228 knees [79%]) Good interdigitation between the cement and the tibial cut surface, but poor interdigitation in the tibial metaphysis (31 of 179 knees [31%]) Cement fixation in both tibial cut surface and metaphysis (tibia) (49 knees [21%])
results of cemented long-stemmed kinematic stabilizer revision TKA. The 10-year survival of the components free of revision for any reason or mechanical failure was 94%. In one of the few studies to directly compare methods of stem fixation, Fehring et al [27] reviewed their results of metaphyseal engaging stems fixed with or without cement. These authors noted a higher rate of radiographic and clinical loosening in the cohort treated with uncemented stems. Although our findings are consistent with those of other authors [1–6,14–24], we recommend to use a hybrid fixation technique, to reduce the technical difficulty in removing the implant in the re-revision circumstances. All of re-revision cases in our series had an AORI type F3 or T3 bone defects. Therefore, reconstitution of bone defect using allograft or metal block is essential to avoid rerevision surgery. Our study has some limitations. It was not randomized and all of the patients in this series received LCCK prosthesis. This may limit the applicability of the findings to other centers. Furthermore, we have no interobserver variability to ensure in interpreting knee scores and radiographic findings, including radiolucent line and loosening. These assessments are somewhat unreliable and therefore the knee scores were prone to some ranges, and we might have underestimated the incidence of radiolucent line or loosening. In conclusion, although constraint of this knee prosthesis can potentially lead to increased aseptic loosening in the long-term, good clinical and radiographic results of this constrained knee prosthesis were achieved in the current series. Adequate restoration of bone stock deficiency and flexion and extension gap are important goals. Long-term study is mandatory to document the incidence of loosening of this constrained knee prosthesis.
26 of 35 knees (74%): radiolucent line b 1 mm 9 of 35 knees (26%): N 2 mm complete radiolucent line 2 knees: b1 mm radiolucent line 28 knees: no radiolucent line 179 of 228 knees (79%) 148 of 179 knees: no radiolucent line 22 of 31 knees (71%): no radiolucent line 9 of 31 knees (29%): N2 mm complete radiolucent line 11 of 49 knees (22%): incomplete radiolucent line b1 mm 38 of 49 knees (78%): no radiolucent line
Fig. 3. Kaplan–Meier survivorship analysis curve.
Please cite this article as: Kim Y-H, et al, Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.04.019
6
Y-H. Kim et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
References 1. Vince KG, Long W. Revision knee arthroplasty. The limits of press fit medullary fixation. Clin Orthop Relat Res 1995;317:172. 2. Haas SB, Insall JN, Montgomery W, et al. Revision total knee arthroplasty with use of modular components with stems inserted without cement. J Bone Joint Surg Am 1995;77(11):1700. 3. Kim Y-H, Kim J-S. Revision total knee arthroplasty with use of a constrained condylar knee prosthesis. J Bone Joint Surg Am 2009;91(6):1440. 4. Hartford JM, Goodman SB, Schurman DJ, et al. Complex primary and revision total knee arthroplasty using the condylar constrained prosthesis. An average 5-year follow-up. J Arthroplasty 1998;13(4):380. 5. Lee J-K, Kim S-J, Choi C-H, et al. Revision total knee arthroplasty using a constrained condylar knee prosthesis in conjunction with a posterior stabilized articular polyethylene. J Arthroplasty 2013;28(4):566. 6. Rodriguez JA, Shahane S, Rasquinha VJ, et al. Does the total condylar 3 constrained knee prosthesis predispose to failure of revision total knee replacement? J Bone Joint Surg Am 2003;85(Suppl. 4):153. 7. Wilke BK, Wagner ER, Trousdale RT. Long-term survival of semiconstrained total knee arthroplasty for revision surgery. J Arthroplasty 2014;29(5):1005. 8. Hwang SC, Keng JY, Nam DC, et al. Revision total knee arthroplasty with a cemented posterior stabilized, condylar constrained or fully constrained prosthesis: a minimum 2-year follow-up analysis. Clin Orthop Surg 2010;2(2):112. 9. Insall JN, Dorr LD, Scott RD, et al. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res 1989;248:13. 10. Bellamy N, Buchanan WW, Goldsmith CH, et al. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to anti-rheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988;15(12):1833. 11. Engh GA, Ammeen DJ. Bone loss with revision total knee arthroplasty: defect classification and alternatives for reconstruction. Instr Course Lect 1998;48:167. 12. Insall JN, editor. Surgery of the knee. New York: Churchill Livingstone; 1984. p. 41. 13. Barrack RL, McClure JT, Burak CF, et al. Revision total hip arthroplasty: the patients’ perspective. Clin Orthop Relat Res 2006;453:173.
14. Kim Y-H, Kook H-K, Kim J-S. Comparison of fixed-bearing and mobile-bearing total knee arthroplasties. Clin Orthop Relat Res 2001;392:101. 15. Kaplan EL, Meire P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457. 16. Greenwood M. A report on the natural duration of cancer. Report on public health and medical subjects, no. 33. London: His Majesty’s Stationary Office; 1926 1. 17. Donaldson WF, Sculco TP, Insall JN, et al. Total condylar III knee prosthesis: long-term follow-up study. Clin Orthop Relat Res 1988;226:21. 18. Shannon BD, Klassen JF, Rand JA, et al. Revision total knee arthroplasty with cemented components and uncemented intramedullary stems. J Arthroplasty 2003; 18(7 Suppl. 1):27. 19. Lachiewicz PF, Falatyn SP. Clinical and radiographic results of the Total Condylar III and Constrained Condylar total knee arthroplasty. J Arthroplasty 1996; 11(8):916. 20. Peters CL, Hennessey R, Barden RM, et al. Revision total knee arthroplasty with a cemented posterior-stabilized or constrained condylar prosthesis: a minimum 3year and average 5-uear follow-up study. J Arthroplasty 1997;12(8):896. 21. Peters CL, Erickson J, Kloepper RG, et al. Revision total knee arthroplasty with modular components inserted with metaphyseal cement and stems without cement. J Arthroplasty 2005;20(3):302. 22. Harwin SF. Revision total knee arthroplasty using the Kinemax Plus Superstabilizer prosthesis: minimum 10-year follow-up. Knee 2006;13(2):93. 23. Manopoulos P, Havet E, Pearce O, et al. Mid-to long-term results of revision total knee replacement using press-fit intramedullary stems with cemented femoral and tibial components. J Bone Joint Surg (Br) 2012;94(7):937. 24. Gofton WT, Tsigaras H, Butler RA, et al. Revision total knee arthroplasty: fixation with modular stems. Clin Orthop Relat Res 2002;404:158. 25. Mabry TM, Vessely MB, Schleck CD, et al. Revision total knee arthroplasty with modular cemented stems. Long-term follow-up. J Arthroplasty 2007; 22(6 Suppl. 2):100. 26. Whaley AL, Trousdale RT, Rand JA, et al. Cemented long-stem revision total knee arthroplasty. J Arthroplasty 2003;18(5):592. 27. Fehring TK, Odum S, Olekson C, et al. Stem fixation win revision total knee arthroplasty: a comparative analysis. Clin Orthop Relat Res 2003;416:217.
Please cite this article as: Kim Y-H, et al, Long-Term Clinical Outcomes and Survivorship of Revision Total Knee Arthroplasty with Use of a Constrained Condylar Knee Prosthesis, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.04.019
