Aims

Stiffness is a common complication after total knee arthroplasty (TKA). Pathogenesis is not understood, treatment options are limited, and diagnosis is challenging. The aim of this study was to investigate if MRI can be used to visualize intra-articular scarring in patients with stiff, painful knee arthroplasties.

Introduction

The primary aim of this study was to describe a baseline comparison of early knee-specific functional outcomes following revision total knee arthroplasty (TKA) using metaphyseal sleeves with a matched cohort of patients undergoing primary TKA. The secondary aim was to compare incidence of complications and length of stay (LOS) between the two groups.

Aims

The aim of this study was to give estimates of the incidence of component incompatibility in hip and knee arthroplasty and to test the effect of an online, real-time compatibility check.

Introduction. Tibiofemoral constraint in patients with total knee replacements (TKR) is dependent on both implant geometry and the surrounding soft tissue structures. Choosing more highly constrained geometries can reduce the contribution of soft tissue necessary to maintain joint stability [1]. Often when knee revision surgeries are required, the soft tissue and bone are compromised leading to the use of more constrained implants to ensure knee stability [2]. The current study quantifies the differences in varus-valgus (VV) and internal-external (IE) constraint between two types of total knee revision systems: SIGMA® TC3© and ATTUNE® REVISION. Methods. Nine cadaveric knees (7 male, age 64.0 ± 9.8 years, BMI 26.28 ± 4.92) were implanted with both fixed-bearing SIGMA TC3 and ATTUNE REVISION knee systems. Five knees received the TC3 implant first, while the remaining 4 received the ATTUNE implant first. The knees were mounted in an inverted position, and a six degree-of-freedom force-torque sensor (JR3, Woodland, CA) was rigidly secured to the distal tibia (Fig. 1). A series of manual manipulations applying IE and VV torques was performed through the flexion range [3]. Each specimen was then revised to the alternate revision system, and the manual manipulations were repeated. Joint loads were calculated, and tibiofemoral kinematics were described according to the Grood-Suntay definition [4]. VV and IE kinematics were calculated as a function of flexion angle, VV torque, and IE torque as has been described previously [3]. The knees were analysed at ±6 Nm VV and ±4 Nm IE, and the kinematics were normalized to the zero load path. A paired t-test (p < .05) was employed to identify significant differences between the kinematics of the two knee systems at 10º flexion increments. Results. Less VV motion was observed in the ATTUNE REVISION system compared to the TC3 system reaching statistical significance in mid-flexion. (Fig. 2). No significant differences were observed in IE rotation between the two designs, except in full extension where the SIGMA TC3 provided increased constraint (Fig. 3). Discussion. The ATTUNE REVISION System provided increased VV constraint compared to the TC3 design. The ATTUNE tibial post was more conforming to the femoral box throughout flexion, which contributed to the increased constraint. However, this increase was not concurrent with a reduction in IE rotational freedom as has been common with more constrained revision systems [5]. ATTUNE REVISION provides additional VV stability while retaining knee IE freedom and, therefore, may enable more natural knee kinematics for patients with MCL deficiency in need of a revision TKR. Future work will focus on how the increased levels of VV constraint affect weight-bearing knee kinematics in the presence of ligament deficiency. Acknowledgements. This work was supported by DePuy Synthes Joint Reconstruction, Warsaw, Indiana, a Johnson and Johnson Company. For figures and references – please contact authors directly

Introduction. Patellar crepitus and clunk are tendofemoral-related complications predominantly associated with posterior-stabilizing (PS) total knee arthroplasty (TKA) designs [1]. Contact between the quadriceps tendon and the femoral component can cause irritation, pain, and catching of soft-tissue within the intercondylar notch (ICN). While the incidence of tendofemoral-related pathologies has been documented for some primary TKA designs, literature describing revision TKA is sparse. Revision components require a larger boss resection to accommodate a constrained post-cam and stem/sleeve attachments, which elevates the entrance to the ICN, potentially increasing the risk of crepitus. The objective of this study was to evaluate tendofemoral contact in primary and revision TKA designs, including designs susceptible to crepitus, and newer designs which aim to address design features associated with crepitus. Methods. Six PS TKA designs were evaluated during deep knee bend using a computational model of the Kansas knee simulator (Figure 1). Prior work has demonstrated that tendofemoral contact predictions from this model can differentiate between TKA patients with patellar crepitus and matched controls [2]. Incidence of crepitus of up to 14% has been reported in Insall-Burstein® II and PFC® Sigma® designs [3]. These designs, in addition to PFC® Sigma® TC3 (revision component), were included in the analyses. Primary and revision components of newer generation designs (NexGen®, Attune® and Attune® Revision) were also included. Designs were evaluated in a patient model with normal Insall-Salvati ratio and a modified model with patellar tendon length reduced by two standard deviations (13mm) to assess worst-case patient anatomy. Results. During simulations with normal patellar tendon length, only PFC® Sigma® and PFC® Sigma® TC3 showed tendofemoral contact within the trochlea, and no design showed contact at the transition to the ICN (Figure 2). In simulations with patella baja, Insall-Burstein® II, PFC® Sigma®, and PFC® Sigma® TC3, demonstrated tendofemoral contact across the trochlea at the transition into the notch. In contrast, NexGen®, Attune® and Attune® Revision showed tendon contact for approximately half the width of the transition to the notch (Figure 3). PFC® Sigma® and Attune® demonstrated very similar tendofemoral contact to their equivalent revision components, although the shorter trochlear groove of Attune® Revision marginally increased contact at the transition. Discussion. Insall-Burstein® II, PFC® Sigma®, and PFC® Sigma® TC3 designs showed full contact with the quadriceps tendon at the anterior border of the ICN when combined with a short patellar tendon. NexGen®, Attune® and Attune® Revision had a more gradual transition between the trochlea and the notch, which resulted in less exposure to tendon contact. Even with the shorter trochlear groove required for revision components, Attune® Revision showed minimal difference in tendofemoral contact when compared with Attune®. There appears to be distinct benefit in a femoral design which reduces tendofemoral contact at the transition to the ICN; this may be of particular importance for patients with patella baja

INTRODUCTION. Modular knee implants are used to manage large bone defects in revision total knee arthroplasty. These implants are confronted with varying fixation characteristics, changes in load transfer or stiffen the bone. In spite of their current clinical use, the influence of modularity on the biomechanical implant-bone behavior (e.g. implant fixation, flexibility, etc.) still is inadequately investigated. Aim of this study is to analyze, if the modularity of a tibial implant could change the biomechanical implant fixation behavior and the implant-bone flexibility. MATERIAL & METHODS. Nine different stem and sleeve combinations of the clinically used tibial revision system Sigma TC3 (DePuy) were compared, each implanted standardized with n=4 in a total of 36 synthetic tibial bones. Four additional un-implanted bones served as reference. Two different cyclic load situations were applied on the implant: 1. Axial torque of ±7Nm around the longitudinal stem axis to determine the rotational implant stability. 2. Varus-valgus-torque of ±3,5Nm to determine the bending behavior of the stem. A high precision optical 3D measurement system allowed simultaneous measuring of spatial micromotions of implant and bone. Based on these micromotions, relative motions at the implant-bone-interface and implant flexibility could be calculated. RESULTS. Lowest relative micromotions were measured along the tibial base component and the sleeve; however, these motions varied depending on the implant construct used. Maximum relative micromotions were detected at the distal end of the implant for all groups, indicating a more proximal fixation of all modular combinations. Regarding varus-valgus-torque measurement, all groups showed a deviant flexibility behavior compared to the reference group. When referred to the un-implanted bone, implants without stems revealed the highest flexibility, whereas implants with shorter stems had lowest flexibility. DISCUSSION & CONCLUSION. All groups showed a more proximal fixation behavior; moreover, both extent and location of fixation could be influenced by varying the modular combination. Larger stems seemed to support a more distal fixation behavior, whereas the implant fixation moved proximal while extending the sleeve. Here the influence of the sleeve on fixation behavior seemed to be dominant compared to the influence of the stem. Concerning varus-valgus-torque, a strong connection between the used stem and implant-bone flexibility seemed to exist. In addition, the influence of the sleeve on flexibility seemed to be rather low. This study showed, that modularity can influence the biomechanical behavior of tibial implants. If these results can be transferred to other tibial implants still remains to be seen

Objectives

Initial stability of tibial trays is crucial for long-term success of total knee arthroplasty (TKA) in both primary and revision settings. Rotating platform (RP) designs reduce torque transfer at the tibiofemoral interface. We asked if this reduced torque transfer in RP designs resulted in subsequently reduced micromotion at the cemented fixation interface between the prosthesis component and the adjacent bone.

Purpose. Traditional total knee arthoplasty techniques have involved implantation of diaphyseal stems to aid in fixation expecially when using constrained polyethylene inserts. While the debate over cemented vs uncemented stems continues, the actual use of stems is considered routine. The authors' experience with cemented stemmed knee revisions in older patients with osteoporotic bone has been favorable. Our younger patients with press-fit stems from varying manufacturers have been plagued with a relatively high incidence of component loosening and stem tip pain in the tibia and occasionally thigh. We report the early results of the first 20 total knee revisions using press-fit metaphyseal filling sleeved stemless implants with constrained bearings. Methods. Twenty three patients with failed primary or revision total knees were assigned to receive stemless sleeved revision knee designs using the DePuy MBT/TC3 system. Reasons for revision included loosening, implant fracture, stiffness, instability, and stem pain. Twenty patients (ages ranging from 42–73) were successfully reconstructed without stems. Six knees with significant uncontained cavitary defects were included. Three patients with unexpectedly osteoporotic metaphyseal bone were revised with cemented stemmed implants and excluded. All cases used cement for initial fixation on the cut bone surface and fully constrained mobile bearing inserts. Results. Follow up ranged from six months to three years. All patients had radiographic evidence of well fixed stable implants on most recent examination. All four cases of revision for “end of stem pain” had complete resolution of symptoms within two weeks of revision surgery. Long leg anterior posterior mechanical alignment x-rays measured within two degrees of neutral in all cases. Knee Society Scores improved an average of 34 points. Clinical results for revision for stiffness had the lowest final scores post operatively. Conclusion. Stemmed total knee arthroplasty revision implants with or without cement are considered the standard for most revision reconstructions. Recently, primary total hip replacements using newer short metaphyseal stems have shown promising early clinical results. This case series of twenty total knee revisions using stemless press-fit metaphyseal sleeves shows similarly favorable outcomes. The complications of stemmed implants such as stem tip pain and difficulty of cemented stem removal can be avoided successfully in non-osteporotic bone reconstructions. With stable bony ingrowth visible on early post-operatyive radiographs, long term stable fixation even with constrained bearings is expected. Longer follow up will be needed to validate this technique for routine use

Stems may improve fixation and stability of components during revision total knee replacement. However, the choice between cemented and cementless stems is not a clear one. Cemented stems offer several advantages in terms of versatility, mechanical stability, surgical technique and clinical outcome over their cementless counterpart.

Cite this article: Bone Joint J 2014;96-B(11 Suppl A):115–7.

Bone loss in the proximal tibia and distal femur is frequently encountered in revision knee replacement surgery. The various options for dealing with this depend on the extent of any bone loss. We present our results with the use of cementless metaphyseal metal sleeves in 103 patients (104 knees) with a mean follow-up of 43 months (30 to 65). At final follow-up, sleeves in 102 knees had good osseointegration. Two tibial sleeves were revised for loosening, possibly due to infection.

The average pre-operative Oxford Knee Score was 23 (11 to 36) and this improved to 32 (15 to 46) post-operatively. These early results encourage us to continue with the technique and monitor the outcomes in the long term.

Cite this article: Bone Joint J 2013;95-B:1640–4.

Soft tissue balancing is critical to successful knee arthroplasty. Pre-operative planning ensures that the surgeon is prepared for any eventuality. We report a large femoral exostosis resulting in gross instability, requiring revision to a constrained implant. An 81 year old female presented with osteoarthritis of the left knee. Xray showed a medial bony mass. CT noted a large bony exostosis arising from the posteromedial femoral condyle. Review showed the exostosis was not related to the medial collateral ligament (MCL). At surgery, the exostosis was noted to be tenting the MCL – excision resulted in complete flaccidity. A trial of the Biomet AGC prosthesis revealed gross medial instability. The decision was taken to convert to a DePuy Sigma TC3 system. Whilst removing TC3 trial components, a lateral condyle fracture occurred. This was fixed with a 1/3 tubular plate and interfragmentary screw. The TC3 system and an AGC patellar button were found to be congruent. A small lateral release was performed, the deep MCL was replaced with tagging sutures through the MCL and the pes anserinus. At 9 weeks post operatively, the patient was pain free and mobilising independently. The knee was stable, with range of movement from 0 to 110 degrees. To our knowledge, this is the first report of such a complication in the literature. It highlights that despite optimal preoperative planning, the surgeon must be prepared to adapt to the situation at hand. It also highlights the importance of having ‘bail out’ options available on shelf when performing routine surgery

Introduction. Previous fluoroscopy studies have been conducted on numerous primary-type TKA, but minimal in vivo data has been documented for subjects implanted with revision TKA. If a subject requires a revision TKA, most often the ligament structures at the knee are compromised and stability of the joint is of great concern. In this present study, subjects implanted with a fixed or mobile bearing TC3 TKA are analyzed to determine if either provides the patient with a significant kinematic advantage. Methods. Ten subjects are analyzed implanted with fixed bearing PFC TC3 TKA and 10 subjects with a mobile bearing PFC TC3 TKA. Each subject underwent a fluoroscopic analysis during four weight bearing activities: deep knee bend (DKB), chair rise, gait, and stair descent. Fluoroscopic images were taken in the sagittal plane at 10 degree increments for the DKB, 30 degree increments for chair rise, and at heel strike, toe off, 33% and 66% cycle gait and stair descent. Results. The average weight bearing maximum flexion for the fixed bearing TKA group was 104 degrees (SD = 18.2 degrees). The average medial and lateral anterior-posterior (AP) translation for these subjects from full extension to maximum weight-bearing flexion was −6.74 mm and −8.0 mm in the posterior direction, respectively. The average femorotibial axial rotation was 1.27 degrees from full extension to maximum flexion. The average medial and lateral AP translations respectively from full extension to maximum flexion are shown in Figures 1 and 2 and the corresponding average femorotibial axial rotation pattern is shown in Figure 3. Subjects implanted with a mobile bearing device are presently being analyzed. Discussion. The fixed bearing device, on average, does not allow for much axial rotation when compared to less constrained or mobile bearing TKA designs. Previous studies have mobile bearing rotating platform primary posterior stabilized devices have documented that the bearing does rotate with the femur. Therefore, it is assumed subjects having a mobile bearing TC3 TKA may achieve greater axial rotation. Subjects having the fixed bearing TC3 TKA did achieve posterior femoral rollback of both condyles, revealing that a fixed bearing revision TKA may act more like a hinged device

Introduction. Electromyography (EMG) is the best known method in obtaining in vivo muscle activation signals during dynamic activities, and this study focuses on comparing the EMG signals of the quadriceps muscles for different TKA designs and normal knees during maximum weight bearing flexion. It is hypothesized that the activation levels will be higher for the TKA groups than the normal group. Methods. Twenty-five subjects were involved in the study with 11 having a normal knee, five a rotating platform (RP) posterior stabilized (PS) TKA, and nine subjects with a PFC TC3 revision TKA. EMG signals were obtained from the rectus femoris, vastus medialis, and vastus lateralis as the patients performed a deep knee bend from full extension to maximum flexion. The data was synchronized with the activity so that the EMG data could be set in flexion-space and compared across the groups. EMG signals were pre-processed by converting the raw signals into neural excitations and normalizing this data with the maximum voluntary contraction (MVC) performed by the subject. The signals were then processed to find the muscle activations which, normalized by MVC, range from 0 to 1. Results. The average muscle activations for each of the three groups are shown in Figures 1, 2, and 3 for the rectus femoris, vastus medialis, and vastus lateralis respectively. The vastus medialis had the highest activation of the muscles during the weight bearing activity from 0 to 90 degrees flexion. On average, the trend seen is that the normal group had lower muscle activation levels to perform the weight bearing activity as opposed to the TKA groups which supports our hypothesis. The PS RP TKA had lower peak values than the PFC TC3 TKA. Discussion. EMG analysis provides insight into muscle activation during dynamic activities. When designing TKA devices for implantation, the patient themselves must be taken into account. In a subject with deficient ligaments, a more constrained device can make day-to-day activities easier, but at the expense of extra effort in achieving higher flexion activities. The high constraints within the PFC TC3 may cause the patient to have to put more effort into the activity. The rotating platform TKA had closer to normal muscle activation levels for the maximum weight bearing knee flexion activity

Introduction. We aim to assess the functional outcome, patient perceived satisfaction and implant survival at a mean follow up of 13[10–16] years following revision knee replacement. Patients and Methods. Between 1995 and 2001, 243 revision knee replacements were performed in 230 patients using Endolink [Link, Hamburg] or TC3 [Depuy, Leeds] prosthesis at Wrightington hospital, Wrightington, were consented to take part in this study. Data was collected prospectively which includes complications and functional assessment by Oxford knee score, WOMAC, HSS, UCLA, SF12 scores, and patient satisfaction questioner. The scores were obtained pre-operatively and post-operatively at 1 year, 5 years and at the latest follow-up. The mean age was 69 yrs, 51% were males, TC3 prosthesis as used in 175 and Endolink in 68, the revision was for Infection in 71[29%], 53 patients had intra-operative positive culture, 35 had 2 stage revision. Results. At a mean follow up of 13 years [10–16] the survival of revision knee replacement in our patient group is 86%. Further surgery was performed in 35[14%], which includes 5 patients who had above knee amputation. The re-revision rate in the non-infected group [13%] was significantly lower compared to the re-revision in infected group [18%]. The Oxford scores improved at 1 year which continued to improve up to 5 years following which there was a gradual deterioration in the scores. There was no significant improvement in the generic HSQ, SF12 and UCLA scores following surgery. The functional scores improved to a lesser extent in patients with proven infection. Discussion and Conclusion. At a mean follow up of 13 years the implant survival for revision knee replacement using endolink or TC3 prosthesis is 86%. The non-infected group had lower re-revision rates. The improvement in functional scores is lower following revision for infection

This was a retrospective analysis of the medium- to long-term results of 46 TC3 Sigma revision total knee replacements using long uncemented stems in press-fit mode. Clinical and radiological analysis took place pre-operatively, at two years post-operatively, and at a mean follow-up of 8.5 years (4 to 12). The mean pre-operative International Knee Society (IKS) clinical score was 42 points (0 to 74), improving to 83.7 (52 to 100) by the final follow-up. The mean IKS score for function improved from 34.3 points (0 to 80) to 64.2 (15 to 100) at the final follow-up. At the final follow-up 30 knees (65.2%) had an excellent result, seven (15.2%) a good result, one (2.2%) a medium and eight (17.4%) a poor result. There were two failures, one with anteroposterior instability and one with aseptic loosening. The TC3 revision knee system, when used with press-fit for long intramedullary stems and cemented femoral and tibial components, in both septic and aseptic revisions, results in a satisfactory clinical and radiological outcome, and has a good medium- to long-term survival rate

Purpose of the study: Restrained implants with intrinsic stability guaranteed by a large central stem have been developed for revision knee arthroplasty, irrespective of the underlying cause. Successful restraint implies excellent fixation of the prosthetic implants which can be obtained using press-fit centromedullary stems. The purpose of this work was to assess the long-term results of this mode of fixation in this indication and to search for clinical correlations with potential radiological images around the stems. Material and methods: We report 46 cases of Sigma. ®. PFC TC3 revision total knee prostheses reviewed retrospectively at two years with a mean follow-up of eight years. Mean age at surgery was 68 years. Revision was indicated for aseptic loosening (n=24) and septic loosening (n=22). The knee society criteria were used for the clinical evaluation. Radiographic measurements were made semiautomatically with the Imagika. ®. software. Results: The mean clinical score improved from 42 points preoperatively to 84.7 at two years and 83.7 at last follow-up. Outcome was excellent (n=30), good (n=7), fair (n=1) and poor (n=1). The mean function score improved from 34.3 preoperatively to 69.1 at two years and 64.2 at last follow-up. Radiographic alignment was correct in all cases. The press-fit effect was observed for 63% of the femoral implants and 76% for the tibial implants. Around the stems, 57% of the implants exhibited condensation lines and 23% lucent lines measuring less than 2 mm. There was no relation between radiological findings and the clinical or functional scores. There were two failures, one for frontal instability at six years and the other for aseptic loosening at eight years. Excepting these two cases, there was one case of femoral implant migration with no clinical expression. For all other patients, the radiographic image remained unchanged between the two year check-up and the final follow-up. Discussion: Thee presence of lucent lines or condensation lines is well known for this type of implant with long centromedullary stems. Nevertheless, this is the first clinical series evaluating the clinical impact of these radiographic images. In our opinion, these images are related to the relative mobility of these implants which should be checked regularly, although no long-term clinical expression occurs. For us, this type of implant enables good function and long-term stability

Introduction: Fifteen revision knee arthroplasties were navigated (Praxim, Grenoble, France). The purpose was to investigate whether revision procedures would require a dedicated navigation system. Material: Two prostheses were revised for severe instability, two for catastrophic failure due to poly wear, 9 for loosening (1 femoral, 1 tibial and 3 both components in total knee arthroplasties, 3 femoral and 1 tibial in uni), two for late infection in extensively cemented prosthesis. The patients were revised using TC3 postero-stabilised revision implants (Depuy, Warsaw, Indiana) and one with hinge prosthesis. Method: Preoperatively, the rotation of the femoral component was investigated by scanner, and the ideal level of the joint line was determined relative to the fibula head on the healthy contralateral side. Navigation was performed with dependant bone cuts, tibia first. The tibial trackers were implanted distally, while the femoral trackers were implanted medially close to the joint line, to prevent impingement with the stems. Bone morphing was performed on the surface of the ancient prosthesis. The system showed the difference between the level of the ideal joint line and the real bone cut, thus indicating the height that had to be reconstructed. The provisional tibial plateau was assembled with its stem and its metallic augments and the knee was balanced with the new tibial component and the ancient femoral component still in place. Femoral reconstruction was finally performed based on an ideal position that had memorized by the computer. Bone reconstruction was required in 2 tibias (morselized allografts) and in one femur (structural femoral head allografts). Results: With a 6-months minimum follow-up, there was no postoperative complication. The HKA angles ranged from 176 to 185. The patella appeared centered on the Merchant view in 14 cases. The femoral rotation was contained between 0 and 5 degrees of external rotation. The 6 month ranges of flexion were 105° with an average gain of −4°. Conclusion: The navigation was a precious help to guide reconstruction of both extremities tibial and femoral, while maintaining the level of the joint line. The navigation system designed for primary surgery appeared convenient to perform revision surgery

We describe a cohort of patients with a high rate of mid-term failure following Kinemax Plus total knee replacement inserted between 1998 and 2001. This implant has been recorded as having a survival rate of 96% at ten years. However, in our series the survival rate was 75% at nine years. This was also significantly lower than that of subsequent consecutive series of PFC Sigma knee replacements performed by the same surgeon. No differences were found in the clinical and radiological parameters between the two groups. At revision the most striking finding was polyethylene wear. An independent analysis of the polyethylene components was therefore undertaken. Scanning electron microscopy revealed type 2 fusion defects in the ultra-high molecular weight polyethylene (UHMWPE), which indicated incomplete boundary fusion. Other abnormalities consistent with weak UHMWPE particle interface strength were present in both the explanted inserts and in unused inserts from the same period.

We consider that these type 2 fusion defects are the cause of the early failure of the Kinemax implants. This may represent a manufacturing defect resulting in a form of programmed polyethylene failure.

Between 1989 and 1992 we had 102 knees suitable for unicompartmental knee replacement (UKR). They were randomised to receive either a St Georg Sled UKR or a Kinematic modular total knee replacement (TKR). The early results demonstrated that the UKR group had less complications and more rapid rehabilitation than the TKR group. At five years there were an equal number of failures in the two groups but the UKR group had more excellent results and a greater range of movement.

The cases were reviewed by a research nurse at 8, 10 and 12 years after operation. We report the outcome at 15 years follow-up. A total of 43 patients (45 knees) died with their prosthetic knees intact. Throughout the review period the Bristol knee scores of the UKR group have been better and at 15 years 15 (71.4%) of the surviving UKRs and 10 (52.6%) of the surviving TKRs had achieved an excellent score. The 15 years survivorship rate based on revision or failure for any reason was 24 (89.8%) for UKR and 19 (78.7%) for TKR. During the 15 years of the review four UKRs and six TKRs failed.

The better early results with UKR are maintained at 15 years with no greater failure rate. The median Bristol knee score of the UKR group was 91.1 at five years and 92 at 15 years, suggesting little functional deterioration in either the prosthesis or the remainder of the joint. These results justify the increased use of UKR.

Fifty-seven revision total knee arthroplasties were performed in our hospital using the TC3 system between 1995 and 1997. Twelve patients died. Forty-five patients were followed up for an average of 5.6 years (range 4 – 7 years). No patients were lost to follow-up. All patients were clinically and radiologically evaluated. A postal patient satisfaction questionnaire was completed. Two patients were revised; one for infection and one for instability. Survivorship using revision as the end point was 93.3% at 7 years. Indications for revision were infection (4;9%), instability (38;84%), pain and stiffness (3;7%). 32 (71%) patients were satisfied with their outcome, 7 (16%) were noncommittal and 6 (13%) were disappointed at 5 years. We have analysed the 13 dissatisfied patients and highlight the lessons learned. Pain and stiffness are not good indications for revision; insert thickness of more than 17.5mm is suggestive of elevation of the joint-line; instead the femoral component should be distalised; step wedges should be used in preference to angular wedges; Always long stem the tibial implant if augments are used; stems should be canal filling with adequate grip on the diaphysis. We suggest the above lessons we have learned from our initial revision arthroplasty learning curve may correlate to the clinical outcome of this small group of dissatisfied patients