The outcome of revision total hip arthroplasty for instability.

Aims: Instability is a common indication for revision total hip arthroplasty (THA). However, even after the initial revision, some patients continue to have recurrent dislocation. The aim of this study was to assess the risk for recurrent dislocation after revision THA for instability. Methods: Between 2009 and 2019, 163 patients underwent revision THA for instability at Stanford University Medical Center. Of these, 33 (20.2%) required re-revision due to recurrent dislocation. Cox proportional hazard models, with death and re-revision surgery for periprosthetic infection as competing events, were used to analyze the risk factors, including the size and alignment of the components. Paired t-tests or Wilcoxon signed-rank tests were used to assess the outcome using the Veterans RAND 12 (VR-12) physical and VR-12 mental scores, the Harris Hip Score (HHS) pain and function, and the Hip disability and Osteoarthritis Outcome score for Joint Replacement (HOOS, JR). Results: The median follow-up was 3.1 years (interquartile range 2.0 to 5.1). The one-year cumulative incidence of recurrent dislocation after revision was 8.7%, which increased to 18.8% at five years and 31.9% at ten years postoperatively. In multivariable analysis, a high American Society of Anesthesiologists (ASA) grade (hazard ratio (HR) 2.72 (95% confidence interval (CI) 1.13 to 6.60)), BMI between 25 and 30 kg/m2 (HR 4.31 (95% CI 1.52 to 12.27)), the use of specialized liners (HR 5.39 (95% CI 1.97 to 14.79) to 10.55 (95% CI 2.27 to 49.15)), lumbopelvic stiffness (HR 6.03 (95% CI 1.80 to 20.23)), and postoperative abductor weakness (HR 7.48 (95% CI 2.34 to 23.91)) were significant risk factors for recurrent dislocation. Increasing the size of the acetabular component by > 1 mm significantly decreased the risk of dislocation (HR 0.89 (95% CI 0.82 to 0.96)). The VR-12 physical and HHS (pain and function) scores improved significantly at mid term. Conclusion: Patients requiring revision THA for instability are at risk of recurrent dislocation. Higher ASA grades, being overweight, a previous lumbopelvic fusion, the use of specialized liners, and postoperative abductor weakness are significant risk factors.

Primary Total Hip Arthroplasty in Juvenile Idiopathic Arthritis: Survivorship After a Median Follow-Up of 12 Years.

BACKGROUND: Juvenile idiopathic arthritis (JIA) is a chronic inflammatory condition of childhood that frequently affects the hip. Total hip arthroplasty (THA) in JIA can be challenging due to the patient's young age, small proportion, complex anatomy, and bone loss. Outcome data are limited. METHODS: We reviewed prospectively collected data in 57 JIA patients (83 hips) who underwent THA between 1986 and 2020 by a single surgeon. The median patient age at surgery was 26 years (range, 14 to 62). Reoperation-free survival was assessed via the cumulative incidence function, accounting for the competing risk of death. Relationships between patient and implant factors and survivorship were evaluated by stratification of the cumulative incidence function and Gray's tests. Wilcoxon signed rank tests were used to assess the preoperative to latest postoperative change in patient-reported outcome measures. RESULTS: At a median (interquartile range) of 12 (4, 20) years of follow-up, 13 (16%) patients underwent reoperation, most commonly for polyethylene wear and osteolysis (7 hips). The estimated incidence of 10-year, 20-year, and 30-year revision (95% confidence interval) were 11.3% (4.5, 21.6%), 18.5% (8.9, 30.9%), and 40.6% (19.4, 60.9%), respectively. There were no differences in survival based on patient age, sex, implant fixation method, polyethylene type, or thickness. All patient-reported outcome measures improved from preoperative to latest follow-up. CONCLUSIONS: Primary THA is a durable and effective treatment for JIA patients with severe hip involvement and results in major improvements in pain and function. We did not identify any factors predictive of failure.

Complications, Implant Survivorships, and Functional Outcomes of Conversion Total Knee Arthroplasty With Prior Hardware.

BACKGROUND: End-stage knee osteoarthritis with retained periarticular hardware is a frequent scenario. Conversion total knee arthroplasty (TKA) leads to excellent outcomes, but poses unique challenges. The evidence supporting retention versus removal of hardware during TKA is controversial. METHODS: Patients who underwent TKA with prior hardware between January 2009 and December 2019 were identified. A total of 148 patients underwent TKA with prior hardware. The mean follow-up was 60 months (range, 24-223). Univariate and multivariable analyses were used to study correlations among factors and surgical-related complications, prosthesis failures, and functional outcomes. RESULTS: The complication rate was 28 of 148 (18.9%). The use of a quadriceps snips in addition to a medial parapatellar arthrotomy was associated with a higher complication (odds ratio: 20.7, P < .05), implant failures (odds ratio: 13.9, P < .05), and lower the Veterans Rand 12 Mental Score (VR-12 MS) (-14.8, P < .05). Hardware removal versus retention and use of single versus multiple incisions were not associated with complications or prosthesis failures. Removal of all hardware was associated with significantly higher (+7.3, P < .05) VR-12 MS compared to retention of all hardware. CONCLUSIONS: TKA with prior hardware was associated with more complications, implant failures, and lower VR-12 MS when a more constrained construct or quadriceps snip was performed. This probably reflects the level of difficulty of the procedure rather than the surgical approach used. Hardware removal or retention was not associated with complications or implant failures; however, removal rather than retention of all prior hardware is associated with increased general health outcomes. LEVEL OF EVIDENCE: IV, cohort without control.

Revision Hip Arthroplasty Using a Modular, Cementless Femoral Stem: Long-Term Follow-Up.

BACKGROUND: As the number of primary total hip arthroplasty (THA) cases increase, so does the demand for revision operations. However, long-term follow-up data for revision THA is lacking. METHODS: A retrospective review was completed of patients who underwent revision THA at a single institution between January 2002 and October 2007 using a cementless modular stem. Patient demographic, clinical, and radiographic data was collected. Preoperative and postoperative patient-reported outcome scores were compared at a minimum of fourteen-year follow-up. RESULTS: Eighty-four patients (89 hips) with a median age of 69 years (range, 28 to 88) at operation were included. Indications for revision included aseptic loosening (84.2%), infection (12.4%), and periprosthetic fracture (3.4%). Twenty-two hips sustained at least 1 complication: intraoperative fracture (7.9%), dislocation (6.7%), prosthetic joint infection (4.5%), deep venous thrombosis (3.4%), and late periprosthetic fracture (2.2%). There were no modular junction complications. Eight patients underwent reoperations; only three involved the stem. Thirty-eight patients (45%) were deceased prior to final follow-up without known reoperations. Twenty-seven patients (32%) were lost to follow-up. Twenty-one patients (23%) were alive at minimum fourteen-year follow-up. Complete patient-reported outcomes were available for nineteen patients (range, 14 to 18.5 years of follow-up). Significant improvement was seen in UCLA activity, VR-12 physical, hip disability and osteoarthritis outcome score, joint replacement., and Harris Hip score pain and function scores. CONCLUSION: Challenges of long-term follow-up include patient migration, an unwillingness to travel for re-examination, medical comorbidities, advanced age, and death. The cementless modular revision stem demonstrated long-term clinical success and remains a safe and reliable option for complex revision operations.

Outcome of the Wagner Cone femoral component for difficult anatomical conditions during total hip arthroplasty.

PURPOSE: Total hip arthroplasty (THA) in patients with small or unusual proximal femoral anatomy is challenging due to sizing issues, control of version, and implant fixation. The Wagner Cone is a monoblock, fluted, tapered stem with successful outcomes for these patients; however, there is limited information on subsidence, a common finding with cementless stems. METHODS: We retrospectively reviewed our cases using the modified Wagner Cone (Zimmer, Warsaw, IN) implanted over a 13-year period (2006-2019) in patients with small or abnormal proximal femoral anatomy. We performed 144 primary THAs in 114 patients using this prosthesis. Mean follow-up was 4.5 ± 3.4 years (range, 1-13 years). Common reasons for implantation were hip dysplasia (52%) and osteoarthritis in patients with small femoral proportions (22%). Analysis of outcomes included assessment of stem subsidence and stability. RESULTS: Survival was 98.6% in aseptic cases; revision-free survival was 97.9%. Femoral subsidence occurred in 84 cases (58%). No subsidence progressed after 3 months. Of those that subsided, the mean distance was 2.8 ± 2.0 mm. There was less subsidence in stems that stabilized prior to six weeks (2.2 ± 1.4 mm) compared to those that continued until 12 weeks (3.9 ± 1.6, p = 0.02). Harris Hip, UCLA, and WOMAC scores significantly improved from pre-operative evaluation (p < 0.001*, p < 0.003*, p ≪ 0.001*); there was no difference in outcome between patients with and without subsidence (p = 0.430, p = 0.228, p = 0.147). CONCLUSION: The modified Wagner Cone demonstrates excellent clinical outcomes in patients with challenging proximal femoral anatomy. Subsidence is minor, stops by 3 months, and does not compromise clinical outcome.

Isolated Versus Full Component Revision in Total Knee Arthroplasty for Aseptic Loosening.

BACKGROUND: Revision of both femoral and tibial components of a total knee arthroplasty (TKA) for aseptic loosening has favorable outcomes. Revision of only one loose component with retention of others has shorter operative time and lower cost; however, implant survivorship and clinical outcomes of these different operations are unclear. METHODS: Between January 2009 and December 2019, a consecutive cohort of revision TKA was reviewed. Univariate and multivariable analyses were used to study correlations among factors and surgical related complications, time to prosthesis failure, and functional outcomes (University of California Los Angeles, Knee Society functional, knee osteoarthritis and outcome score for joint replacement, Veterans RAND 12 (VR-12) physical, and VR-12 mental). RESULTS: A total of 238 patients underwent revision TKA for aseptic loosening. The mean follow-up time was 61 months (range 25 to 152). Ten of the 105 patients (9.5%) who underwent full revision (both femoral and tibial components) and 18 of the 133 (13.5%) who underwent isolated revision had subsequent prosthesis failure [Hazard ratio (HR) 0.67, P = .343]. The factor analysis of type of revision (full or isolated revision) did not demonstrate a significant difference between groups in terms of complications, implant failures, and times to failure. Metallosis was related to early time to failure [Hazard ratio 10.11, P < .001] and iliotibial band release was associated with more complications (Odds ratio 9.87, P = .027). Preoperative symptoms of instability were associated with the worst improvement in University of California Los Angeles score. Higher American Society of Anesthesiologists status and higher Charlson Comorbidity Index were related with worse VR-12 physical (-30.5, P = .008) and knee osteoarthritis and outcome score for joint replacement (-4.2, P = .050) scores, respectively. CONCLUSION: Isolated and full component revision TKA for aseptic loosening does not differ with respect to prosthesis failures, complications, and clinical results at 5 years. Poor American Society of Anesthesiologists status, increased comorbidities, instability, and a severe bone defect are related to worse functional improvement. LEVEL OF EVIDENCE: III, cohort with control.

Reconstruction of Disrupted Extensor Mechanism After Total Knee Arthroplasty.

BACKGROUND: Disruption of the extensor mechanism after total knee arthroplasty (TKA) is a debilitating complication that results in extension lag, limited range of motion, difficulty in walking, frequent falls, and chronic pain. This study presents the clinical and radiographic results of reconstruction after extensor mechanism disruption in TKA patients. METHODS: Consecutive patients with allograft reconstruction of extensor mechanism after TKA were identified retrospectively from an academic tertiary center for revision TKA. RESULTS: Sixteen patients with a mean age of 61 ± 14 years at extensor mechanism reconstruction with a minimum of 2-year follow-up were included. The mean follow-up was 3.3 ± 2.2 years. Knee Society score (KSS), before and at final follow-up extension lag, range of motion, and radiographic change in patellar height were reviewed. There were statistically significant improvements between preoperative and final follow-up KSS (P < .001; KSS for pain, preoperative 40 ± 14 points to final follow-up 67 ± 15 points [P < .001]; KSS for function, preoperative 26 ± 21 points to final follow-up 48 ± 25 points [P < .001]). The extension lag was also reduced from 35° ± 16° preoperatively to 14° ± 18° (P < .001) at final follow-up. There was an average proximal patellar migration of 8 ± 10 mm. Five (31%) cases had an extensor lag of >30° or revision surgery for repeat extensor mechanism reconstruction, infection, or arthrodesis. CONCLUSION: Our 10-year experience using allografts during extensor mechanism reconstruction demonstrates reasonable outcomes, but failures are to be anticipated in approximately one-third of patients.

Cortical Strut Allograft Support of Modular Femoral Junctions During Revision Total Hip Arthroplasty.

BACKGROUND: There is risk of junction failure when using modular femoral stems for revision total hip arthroplasty (THA), especially with loss of bone stock in the proximal femur. Using a cortical strut allograft may provide additional support of a modular femoral construct in revision THA. METHODS: We reviewed prospectively gathered clinical and radiographic data for 28 revision THAs performed from 2004 to 2014 using cementless modular femoral components with cortical strut allograft applied to supplement proximal femoral bone loss: 5 (18%) were fluted taper designs and 23 (82%) were porous cylindrical designs All the patients had a Paprosky grade IIIA or greater femoral defect. The mean follow-up was 5.4 ± 3.9 years. RESULTS: The Harris Hip Scores improved from 26 ± 10 points preoperatively to 71 ± 10 points at final follow-up (P < .001). The Western Ontario McMaster Universities Osteoarthritis Index scores improved from 45 ± 12 points preoperatively to 76 ± 12 points at final follow-up (P < .001). Eighty-nine percent (25 hips) of all revision or conversion THAs were in place at final follow-up. Three (11%) patients underwent reoperations, 2 for infection and 1 for periprosthetic fracture. There was no statistical significant change in femoral component alignment (P = .161) at final follow-up. Mean subsidence was 1.8 ± 1.3 mm at final follow-up. Femoral diameter increased from initial postoperative imaging to final follow-up imaging by a mean of 9.1 ± 5.1 mm (P < .001) and cortical width increased by a mean of 4.5 ± 2.2 mm (P < .001). Twenty-seven hips (96%) achieved union between the cortical strut allograft and the host femur. CONCLUSION: The use of a modular femoral stem in a compromised femur with a supplementary cortical strut allgraft is safe and provides satisfactory clinical and radiological outcomes.

Use of Cortical Strut Allograft After Extended Trochanteric Osteotomy in Revision Total Hip Arthroplasty.

BACKGROUND: Cortical strut allografts restore bone stock and improve postoperative clinical scores after revision total hip arthroplasty (THA). However, use of a cortical strut allograft is implicated in delayed healing of an extended trochanteric osteotomy (ETO). To date, there are no reports directly comparing ETO with or without cortical strut allografts. METHODS: We reviewed prospectively gathered data on 50 revision THAs performed from 2004-2014 using an ETO. We compared the demographic, radiological, and clinical outcome of patients with (16 hips) and without (34 hips) cortical strut allograft after an ETO. RESULTS: There were no significant differences in Western Ontario McMaster Universities Osteoarthritis Index or Harris Hip Score between the ETOs with and without a cortical strut allograft. Fifteen of the ETOs (94%) with a cortical strut allograft and 31 of the ETOs (91%) without a cortical strut allograft were in situ at final follow-up (P = 1.000). A higher proportion hips with cortical strut allograft (100%, 16 patients) had preoperative Paprosky grade bone loss more than IIIA compared to those without allograft (29%, 10 patients) (P < .001). There were no differences in femoral stem subsidence (P = .207), alignment (P = .934), or migration of the osteotomized fragment (P = .171). Fourteen of the ETOs (88%) in patients with cortical strut allograft united compared to 34 ETOs (100%) in patients without allograft (P = .095). CONCLUSION: Our study shows that the use of cortical strut allograft during revision THA with ETO does not reduce the rate of union, radiological or clinical outcomes.

The Direct Anterior Approach is Associated With Early Revision Total Hip Arthroplasty.

BACKGROUND: The direct anterior approach for total hip arthroplasty (THA) has generated increased interest recently. The purpose of this study was to compare the duration to failure and reasons for revision of primary THA performed elsewhere and subsequently revised at our institution after the direct anterior vs other nonanterior surgical approaches to the hip. METHODS: All primary THAs performed elsewhere and referred to our institution for revision were divided into the direct anterior approach (30 cases) or nonanterior approach groups (100 cases, randomly selected from 453 cases) based on the original surgical approach. Because all primary direct anterior THAs were originally performed after 2004 to eliminate temporal bias, we identified a subset of the nonanterior group in which the primary THA was performed after 2004 (known as the recent nonanterior group, 100 cases, randomly selected from 169 available cases). RESULTS: The mean duration from primary to revision THA was 3.0 ± 2.7 years (direct anterior approach), 12.0 ± 8.8 years (nonanterior approach), and 3.6 ± 2.8 years (recent nonanterior), respectively. There was a significant difference in time to revision between the direct anterior and nonanterior approach groups (P < .001). Aseptic loosening of the stem was significantly more frequent with the direct anterior approach group (9/30, 30.0%) when compared with the nonanterior group (8/100, 8.0%, P = .007) and the recent nonanterior group (7/100, 7.0%, P = .002). CONCLUSION: Revision of the femoral component for aseptic loosening is more commonly associated with the direct anterior approach in our referral practice.

Treatment of Secondary Osteonecrosis of the Knee With Local Debridement and Osteoprogenitor Cell Grafting.

Secondary osteonecrosis of the knee (SOK) affects young individuals with chronic diseases and corticosteroid use. We report a series of young patients in whom the osteonecrotic lesion was openly debrided, and concentrated bone marrow osteoprogenitor cells (OPCs) harvested from the iliac crest were placed in the defect. Twelve patients (fourteen knees) have undergone debridement and grafting of distal femoral osteonecrotic lesions. Age at surgery averaged 23 years. Follow-up averaged 5 years. None of the patients have undergone further surgery, or were taking medications for ipsilateral knee pain. Knee Society Score and Knee Function Score averaged 87 and 85 respectively. The technique of open debridement and osteoprogenitor cell grafting for SOK is relatively simple, efficacious, has low morbidity, and does not preclude future interventions.

Polyethylene wear and osteolysis is associated with high revision rate of a small sized porous coated THA in patients with hip dysplasia.

The outcome of 25 primary THAs in patients with hip dysplasia using the AML Bantam femoral stem (DePuy) is reported. Age at operation averaged 43 ± 10 years. Twenty-two of 25 stems were cementless. All cementless acetabular components had conventional or cross-linked polyethylene and screws. Follow-up averaged 11 ± 5 years (range 4-18). Four cementless stems were revised after 3, 4, 8, and 9 years; 2/3 cemented stems were revised at 8 and 18 years. Femoral revisions demonstrated extensive conventional polyethylene wear, periprosthetic osteolysis and loosening. Five entire cups were revised for wear and loosening; four liners were replaced. Harris Hip Scores for patients with retained stems went from 43 ± 12 to 85 ± 13. High revision rates with the proximally porous coated Bantam stem are due to loss of fixation, often associated with polyethylene wear and osteolysis.

High complication rate in revision total hip arthroplasty in juvenile idiopathic arthritis.

BACKGROUND: Revision total hip arthroplasty (THA) in patients with juvenile idiopathic arthritis (JIA) is challenging as a result of the patient's young age, systemic disease, multiple affected joints, small proportions, and bone loss. The intermediate- to long-term results of these surgeries remain unknown. QUESTIONS/PURPOSES: The purpose of this study is to determine the (1) functional outcomes; (2) surgical complications; and (3) frequency of reoperation or revision after revision THA for JIA. METHODS: We reviewed the records of all patients from one center who underwent revision THA for JIA who had a minimum of 5 years of followup (mean, 9 years; range, 5-19 years). This resulted in a series of 24 revision THAs in 15 patients. All patients were Charnley Class C. Age at revision averaged 35 years (range, 21-53 years). The 20 acetabular and 12 femoral revision components included 15 cementless cups, five reconstruction/roof rings with a cemented cup, and four cemented and eight cementless femoral stems. RESULTS: The Harris hip scores improved from 54 (range, 34-85) to 77 (range, 37-100) (p < 0.001). Complications included two proximal femoral fractures associated with severe osteolysis and one sciatic nerve palsy in a patient with severe acetabular deficiency. A total of seven hips (29%) required reoperation or revision surgery, including three for infection (one early and two late) and four for mechanical loosening. CONCLUSIONS: Revision THA in JIA is very challenging owing to patients' small proportions and compromised bone stock. The intraoperative and early complication rates are relatively high. Prognosis for long-term survivorship is guarded; limiting factors include periprosthetic osteolysis associated with older implants that used conventional polyethylene and cemented stems.

Cancellous impaction bone grafting of acetabular defects in complex primary and revision total hip arthroplasty.

The reconstruction of major acetabular bone defects during revision, conversion, and primary total hip arthroplasties (THAs) is challenging. We reviewed a consecutive series of 168 THAs (108 revisions, 8 conversions, and 52 primary THAs) performed by 1 surgeon (S.B.G.) between 1997 and 2008 using impaction bone grafting for acetabular reconstruction. Autograft, cancellous allograft croutons, and demineralized bone matrix were used to fill bone defects as needed. The acetabular bone deficiency was classified according to the American Academy of Orthopaedic Surgeons: type I, segmental deficiency with significant rim defect; type II, cavitary defects medially or posteriorly; type III, combined cavitary and segmental deficiency; type IV, pelvic discontinuity; and type V, arthrodesis. According to this method, 56 hips had type I, 31 hips had type II, 48 hips had type III, and 27 hips had type IV deficiencies. Of the 168 patients, 19 subsequently died of causes unrelated to the THA, and 11 were lost to follow-up. All patients had at least 2 years of follow-up. Average Harris Hip Score improved from 45.5±17.9 preoperatively to 81.1±16.5 postoperatively (P<.05) for revision THAs, from 40.0±11.3 preoperatively to 85.0±12.8 postoperatively (P<.05) for conversion THAs, and from 42.3±14.9 preoperatively to 85.0±12.0 postoperatively (P<.05) for primary THAs. All impaction grafted bone (allograft, autograft, or a combination) incorporated radiographically, thus restoring bone stock. Complications included 1 early infection, which was managed successfully with debridement and liner exchange, and 2 late infections that were managed successfully with staged revision. Two revisions required subsequent re-revision for late loosening. Two hip dislocations occurred, 1 of which required surgical treatment to place a constrained liner.

Outcome of primary total hip arthroplasty in Charnley Class C patients with juvenile idiopathic arthritis: a case series.

The outcome and complications of 37 primary total hip arthroplasties by one surgeon in 24 patients with Charnley Class C juvenile idiopathic arthritis with up to 19.6 years follow-up are reported. Twenty-six femoral components were cementless; all acetabular components were cementless with screws. Age at operation averaged 22.6 years. Two patients (3 hips) have died. Twelve hips in 9 patients have failed. Six cementless acetabular components with conventional polyethylene were revised because of osteolysis after 5.5 to 14.5 years. All 3 cementless C2 femoral stems with minimal porous coating failed. One of eight cemented AML Bantam stems loosened at 3.5 years; 2 of 23 cementless AML Bantam stems loosened at 9.5 and 19.6 years. Pain relief and functional improvement are dramatic after total hip arthroplasty in juvenile idiopathic arthritis; however, the long-term outcome is guarded.

Cementless femoral prostheses cost more to implant than cemented femoral prostheses.

UNLABELLED: Prosthetic cost contributes greatly to the overall expense of THA. A key question, therefore, in the selection of implant technique is whether any price difference exists between a cementless and a cemented femoral prosthesis. We evaluated the price difference between the most commonly used cemented and cementless femoral stems at three high-volume academic medical centers. Each hospital's costs for prostheses from the manufacturers were recorded. The average cost of implanting a cementless femoral prosthesis was $296 more than the average cost of implanting a cemented femoral stem, even with the additional expense of two batches of bone cement and the accessories commonly used to achieve a third-generation cementing technique. The price difference was less variable if the cost of the prostheses from only the primary implant supplier for each institution was considered. As the number of THAs performed per year continues to escalate, implantation of a cemented femoral component remains an attractive method of fixation from an economic standpoint. LEVEL OF EVIDENCE: Level III, economic and decision analysis. See the Guidelines for Authors for a complete description of levels of evidence.

Revision total hip arthroplasty in juvenile chronic arthritis: 17 revisions in 11 patients followed for 4-12 years.

BACKGROUND: Revision total hip arthroplasty (THA) in patients with juvenile chronic arthritis (JCA) is complicated by the young age of the patient, poor bone stock and small physical proportions. We report the complications and outcome of a prospective series of 17 revision THAs in Charnley class C JCA patients. METHODS: 15 acetabular components and 10 femoral components were revised. 13 cementless cups, 2 reconstruction/roof rings and cemented cups, and 4 cemented and 6 cementless femoral stems were implanted. 2 proximal femoral allografts and 1 strut allograft were used. Age at revision was 32 (21-53) years. Follow-up averaged 7 (4-12) years. RESULTS: 2 patients with cemented femoral stems developed loosening, osteolysis and fracture. Both were successfully revised to long-stem cementless implants with strut/proximal femoral allografts. 1 loose, worn cementless cup with osteolysis was revised. 1 patient with a peri-operative infection and late acetabular fracture had a loose, non-revised cementless cup. 1 case of sciatic nerve palsy occurred after revision using a reconstruction ring. 1 late infection necessitated resection arthroplasty. Harris hip scores improved from 53 (34-85) to 76 (47-96). INTERPRETATION: Revision THA in JCA has a substantial complication rate, even in experienced hands. The problem of obtaining long-term stable fixation, osteolysis, and replenishment of lost bone stock are major difficulties.