The aim of this study was to report the incidence of arthrofibrosis of the knee and identify risk factors for its development following a fracture of the tibial plateau. We carried out a retrospective review of 186 patients (114 male, 72 female) with a fracture of the tibial plateau who underwent open reduction and internal fixation. Their mean age was 46.4 years (19 to 83) and the mean follow-up was16.0 months (6 to 80).

A total of 27 patients (14.5%) developed arthrofibrosis requiring a further intervention. Using multivariate regression analysis, the use of a provisional external fixator (odds ratio (OR) 4.63, 95% confidence interval (CI) 1.26 to 17.7, p = 0.021) was significantly associated with the development of arthrofibrosis. Similarly, the use of a continuous passive movement (CPM) machine was associated with significantly less development of arthrofibrosis (OR = 0.32, 95% CI 0.11 to 0.83, p = 0.024). The effect of time in an external fixator was found to be significant, with each extra day of external fixation increasing the odds of requiring manipulation under anaesthesia (MUA) or quadricepsplasty by 10% (OR = 1.10, p = 0.030). High-energy fracture, surgical approach, infection and use of tobacco were not associated with the development of arthrofibrosis. Patients with a successful MUA had significantly less time to MUA (mean 2.9 months; sd 1.25) than those with an unsuccessful MUA (mean 4.86 months; sd 2.61, p = 0.014). For those with limited movement, therefore, performing an MUA within three months of the injury may result in a better range of movement.

Based our results, CPM following operative fixation for a fracture of the tibial plateau may reduce the risk of the development of arthrofibrosis, particularly in patients who also undergo prolonged provisional external fixation.

Cite this article: Bone Joint J 2015;97-B:109–14.

A small proportion of children with Gartland type III supracondylar humeral fracture (SCHF) experience troubling limited or delayed recovery after operative treatment. We hypothesised that the fracture level relative to the isthmus of the humerus would affect the outcome.

We retrospectively reviewed 230 children who underwent closed reduction and percutaneous pinning (CRPP) for their Gartland type III SCHFs between March 2003 and December 2012. There were 144 boys and 86 girls, with the mean age of six years (1.1 to 15.2). The clinico-radiological characteristics and surgical outcomes (recovery of the elbow range of movement, post-operative angulation, and the final Flynn grade) were recorded. Multivariate analysis was employed to identify prognostic factors that influenced outcome, including fracture level. Multivariate analysis revealed that a fracture below the humeral isthmus was significantly associated with poor prognosis in terms of the range of elbow movement (p < 0.001), angulation (p = 0.001) and Flynn grade (p = 0.003). Age over ten years was also a poor prognostic factor for recovery of the range of elbow movement (p = 0.027).

This is the first study demonstrating a subclassification system of Gartland III fractures with prognostic significance. This will guide surgeons in peri-operative planning and counselling as well as directing future research aimed at improving outcomes.

Cite this article: Bone Joint J 2015;97-B:134–40.

The Essex-Lopresti injury (ELI) of the forearm is a rare and serious condition which is often overlooked, leading to a poor outcome.

The purpose of this retrospective case study was to establish whether early surgery can give good medium-term results.

From a group of 295 patients with a fracture of the radial head, 12 patients were diagnosed with ELI on MRI which confirmed injury to the interosseous membrane (IOM) and ligament (IOL). They were treated by reduction and temporary Kirschner (K)-wire stabilisation of the distal radioulnar joint (DRUJ). In addition, eight patients had a radial head replacement, and two a radial head reconstruction.

All patients were examined clinically and radiologically 59 months (25 to 90) after surgery when the mean Mayo Modified Wrist Score (MMWS) was 88.4 (78 to 94), the mean Mayo Elbow Performance Scores (MEPS) 86.7 (77 to 95) and the mean disabilities of arm, shoulder and hand (DASH) score 20.5 (16 to 31): all of these indicate a good outcome.

In case of a high index of suspicion for ELI in patients with a radial head fracture, we recommend the following: confirmation of IOM and IOL injury with an early MRI scan; early surgery with reduction and temporary K-wire stabilisation of the DRUJ; preservation of the radial head if at all possible or replacement if not, and functional bracing in supination. This will increase the prospect of a good result, and avoid the complications of a missed diagnosis and the difficulties of late treatment.

Cite this article: Bone Joint J 2014;96-B:1385–91.

We retrospectively reviewed 89 consecutive patients (45 men and 44 women) with a mean age at the time of injury of 58 years (18 to 97) who had undergone external fixation after sustaining a unilateral fracture of the distal humerus. Our objectives were to determine the incidence of heterotopic ossification (HO); identify risk factors associated with the development of HO; and characterise the location, severity and resultant functional impairment attributable to the presence of HO.

HO was identified in 37 elbows (42%), mostly around the humerus and along the course of the medial collateral ligament. HO was hazy immature in five elbows (13.5%), mature discrete in 20 (54%), extensive mature in 10 (27%), and complete bone bridges were present in two elbows (5.5%). Mild functional impairment occurred in eight patients, moderate in 27 and severe in two. HO was associated with less extension (p = 0.032) and less overall flexion-to-extension movement (p = 0.022); the flexion-to-extension arc was < 100º in 21 elbows (57%) with HO compared with 18 elbows (35%) without HO (p = 0.03). HO was removed surgically in seven elbows.

The development of HO was significantly associated with sustaining a head injury (p = 0.015), delayed internal fixation (p = 0.027), the method of fracture fixation (p = 0.039) and the use of bone graft or substitute (p = 0.02).HO continues to be a substantial complication after internal fixation for distal humerus fractures.

Cite this article: Bone Joint J 2014;96-B:1681–7.

Our aim was to compare the one-year post-operative outcomes following retention or removal of syndesmotic screws in adult patients with a fracture of the ankle that was treated surgically. A total of 51 patients (35 males, 16 females), with a mean age of 33.5 years (16 to 62), undergoing fibular osteosynthesis and syndesmotic screw fixation, were randomly allocated to retention of the syndesmotic screw or removal at three months post-operatively. The two groups were comparable at baseline.

One year post-operatively, there was no significant difference in the mean Olerud–Molander ankle score (82.4 retention vs 86.7 removal, p = 0.367), the mean American Orthopedic Foot and Ankle Society ankle-hindfoot score (88.6 vs 90.1, p = 0.688), the mean American Academy of Orthopedic Surgeons foot and ankle score (96.3 vs 94.0, p = 0.250), the mean visual analogue pain score (1.0 vs 0.7, p = 0.237), the mean active dorsiflexion (10.2° vs 13.0°, p = 0.194) and plantar flexion (33.6° vs 31.3°, p = 0.503) of the ankle, or the mean radiological tibiofibular clear space (5.0 mm vs 5.3 mm, p = 0.276) between the two groups. A total of 19 patients (76%) in the retention group had a loose and/or broken screw one year post-operatively.

We conclude that removal of a syndesmotic screw produces no significant functional, clinical or radiological benefit in adult patients who are treated surgically for a fracture of the ankle.

Cite this article: Bone Joint J 2014;96-B:1699–1705.

We assessed hyperextension of the knee and joint laxity in 169 consecutive patients who underwent an anterior cruciate ligament reconstruction between 2000 and 2002 and correlated this with a selected number of age- and gender-matched controls. In addition, the mechanism of injury in the majority of patients was documented. Joint laxity was present in 42.6% (72 of 169) of the patients and hyperextension of the knee in 78.7% (133 of 169). All patients with joint laxity had hyperextension of their knee. In the control group only 21.5% (14 of 65) had joint laxity and 37% (24 of 65) had hyperextension of the knee. Statistical analysis showed a significant correlation for these associations. We conclude that anterior cruciate ligament injury is more common in those with joint laxity and particularly so for those with hyperextension of the knee.

Haematomas, drainage, and other non-infectious wound complications following total knee replacement (TKR) have been associated with long-term sequelae, in particular, deep infection. However, the impact of these wound complications on clinical outcome is unknown. This study compares results in 15 patients re-admitted for wound complications within 90 days of TKR to 30 matched patients who underwent uncomplicated total knee replacements. Patients with wound complications had a mean age of 66 years (49 to 83) and mean body mass index (BMI) of 37 (21 to 54), both similar to that of patients without complications (mean age 65 years and mean BMI 35). Those with complications had lower mean Knee Society function scores (46 (0 to 100 vs 66 (20 to 100), p = 0.047) and a higher incidence of mild or greater pain (73% vs 33%, p = 0.01) after two years compared with the non re-admitted group. Expectations in patients with wound complications following TKR should be tempered, even in those who do not develop an infection.

Cite this article: Bone Joint J 2014;96-B:619–21.

Medium-term results of the Discovery elbow replacement are presented. We reviewed 51 consecutive primary Discovery total elbow replacements (TERs) implanted in 48 patients. The mean age of the patients was 69.2 years (49 to 92), there were 19 males and 32 females (37%:63%) The mean follow-up was 40.6 months (24 to 69). A total of six patients were lost to follow-up. Statistically significant improvements in range movement and Oxford Elbow Score were found (p < 0.001). Radiolucent lines were much more common in, and aseptic loosening was exclusive to, the humeral component. Kaplan–Meier survivorship at five years was 92.2% (95% CI 74.5% to 96.4%) for aseptic loosening. In four TERs, periprosthetic infection occurred resulting in failure. A statistically significant association between infection and increased BMI was found (p = 0.0268). Triceps failure was more frequent after the Mayo surgical approach and TER performed after previous trauma surgery. No failures of the implant were noted.

Our comparison shows that the Discovery has early clinical results that are similar to other semi-constrained TERs. We found continued radiological surveillance with particular focus on humeral lucency is warranted and has not previously been reported. Despite advances in the design of total elbow replacement prostheses, rates of complication remain high.

Cite this article: Bone Joint J 2014;96-B:1359–65

We hypothesised that the use of pulsed electromagnetic field (PEMF) bone growth stimulation in acute scaphoid fractures would significantly shorten the time to union and reduce the number of nonunions in a randomised, double-blind, placebo-controlled multicentre trial. A total of 102 patients (78 male, 24 female; mean age 35 years (18 to 77)) from five different medical centres with a unilateral undisplaced acute scaphoid fracture were randomly allocated to PEMF (n = 51) or placebo (n = 51) and assessed with regard to functional and radiological outcomes (multiplanar reconstructed CT scans) at 6, 9, 12, 24 and 52 weeks. The overall time to clinical and radiological healing did not differ significantly between the active PEMF group and the placebo group. We concluded that the addition of PEMF bone growth stimulation to the conservative treatment of acute scaphoid fractures does not accelerate bone healing.

Cite this article: Bone Joint J 2014; 96-B:1070–6.

The relationship between post-operative bone density and subsequent failure of total knee replacement (TKR) is not known. This retrospective study aimed to determine the relationship between bone density and failure, both overall and according to failure mechanism. All 54 aseptic failures occurring in 50 patients from 7760 consecutive primary cemented TKRs between 1983 and 2004 were matched with non-failing TKRs, and 47 failures in 44 patients involved tibial failures with the matching characteristics of age (65.1 for failed and 69.8 for non-failed), gender (70.2% female), diagnosis (93.6% OA), date of operation, bilaterality, pre-operative alignment (0.4 and 0.3 respectively), and body mass index (30.2 and 30.0 respectively). In each case, the density of bone beneath the tibial component was assessed at each follow-up interval using standardised, calibrated radiographs. Failing knees were compared with controls both overall and, as a subgroup analysis, by failure mechanism. Knees were compared with controls using univariable linear regression.

Significant and continuous elevation in tibial density was found in knees that eventually failed by medial collapse (p < 0.001) and progressive radiolucency (p < 0.001) compared with controls, particularly in the medial region of the tibia. Knees failing due to ligamentous instability demonstrated an initial decline in density (p = 0.0152) followed by a non-decreasing density over time (p = 0.034 for equivalence). Non-failing knees reported a decline in density similar to that reported previously using dual-energy x-ray absorptiometry (DEXA). Differences between failing and non-failing knees were observable as early as two months following surgery. This tool may be used to identify patients at risk of failure following TKR, but more validation work is needed.

Cite this article: Bone Joint J 2014;96-B:1503–9.

Although it is clear that opening-wedge high tibial osteotomy (HTO) changes alignment in the coronal plane, which is its objective, it is not clear how this procedure affects knee kinematics throughout the range of joint movement and in other planes.

Our research question was: how does opening-wedge HTO change three-dimensional tibiofemoral and patellofemoral kinematics in loaded flexion in patients with varus deformity?Three-dimensional kinematics were assessed over 0° to 60° of loaded flexion using an MRI method before and after opening-wedge HTO in a cohort of 13 men (14 knees). Results obtained from an iterative statistical model found that at six and 12 months after operation, opening-wedge HTO caused increased anterior translation of the tibia (mean 2.6 mm, p <  0.001), decreased proximal translation of the patella (mean –2.2 mm, p <  0.001), decreased patellar spin (mean –1.4°, p < 0.05), increased patellar tilt (mean 2.2°, p < 0.05) and changed three other parameters. The mean Western Ontario and McMaster Universities Arthritis Index improved significantly (p < 0.001) from 49.6 (standard deviation (sd) 16.4) pre-operatively to a mean of 28.2 (sd 16.6) at six months and a mean of 22.5 (sd 14.4) at 12 months.

The three-dimensional kinematic changes found may be important in explaining inconsistency in clinical outcomes, and suggest that measures in addition to coronal plane alignment should be considered.

Cite this article: Bone Joint J 2014; 96-B:1214–21.

We reviewed the long-term clinical and radiological results of 63 uncemented Low Contact Stress (LCS) total knee replacements (TKRs) in 47 patients with rheumatoid arthritis. The mean age of the patients at the time of surgery was 69 years (53 to 81). At a mean follow-up of 22 years (20 to 25), 12 patients were alive (17 TKRs), 27 had died (36 TKRs), and eight (ten TKRs) were lost to follow-up.

Revision was necessary in seven patients (seven TKRs, 11.1%) at a mean of 12.1 years (0 to 19) after surgery. In the surviving ten patients who had not undergone revision (15 TKRs), the mean Oxford knee score was 30.2 (16 to 41) at a mean follow-up of 19.5 years (15 to 24.7) and mean active flexion was 105° (90° to 150°). The survival rate was 88.9% at 20 years (56 of 63) and the Kaplan–Meier survival estimate, without revision, was 80.2% (95% confidence interval 37 to 100) at 25 years.

Cite this article: Bone Joint J 2013;95-B:1497–9.

We have previously reported the short-term radiological results of a randomised controlled trial comparing kinematically aligned total knee replacement (TKR) and mechanically aligned TKR, along with early pain and function scores. In this study we report the two-year clinical results from this trial. A total of 88 patients (88 knees) were randomly allocated to undergo either kinematically aligned TKR using patient-specific guides, or mechanically aligned TKR using conventional instruments. They were analysed on an intention-to-treat basis. The patients and the clinical evaluator were blinded to the method of alignment.

At a minimum of two years, all outcomes were better for the kinematically aligned group, as determined by the mean Oxford knee score (40 (15 to 48) versus 33 (13 to 48); p = 0.005), the mean Western Ontario McMaster Universities Arthritis index (WOMAC) (15 (0 to 63) versus 26 (0 to 73); p = 0.005), mean combined Knee Society score (160 (93 to 200) versus 137 (64 to 200); p= 0.005) and mean flexion of 121° (100 to 150) versus 113° (80 to 130) (p = 0.002). The odds ratio of having a pain-free knee at two years with the kinematically aligned technique (Oxford and WOMAC pain scores) was 3.2 (p = 0.020) and 4.9 (p = 0.001), respectively, compared with the mechanically aligned technique. Patients in the kinematically aligned group walked a mean of 50 feet further in hospital prior to discharge compared with the mechanically aligned group (p = 0.044).

In this study, the use of a kinematic alignment technique performed with patient-specific guides provided better pain relief and restored better function and range of movement than the mechanical alignment technique performed with conventional instruments.

Cite this article: Bone Joint J 2014;96-B:907–13.

We present the long-term results of open surgery for internal shoulder rotational deformity in brachial plexus birth palsy (BPBP).

From 1997 to 2005, 207 patients (107 females, 100 males, mean age 6.2 (0.6 to 34)) were operated on with subscapularis elongation and/or latissimus dorsi to infraspinatus transfer. Incongruent shoulder joints were relocated. The early results of these patients has been reported previously. We analysed 118 (64 females, 54 males, mean age 15.1 (7.6 to 34)) of the original patient cohort at a mean of 10.4 years (7.0 to 15.1) post-operatively. A third of patients with relocated joints had undergone secondary internal rotational osteotomy of the humerus.

A mixed effects models approach was used to evaluate the effects of surgery on shoulder rotation, abduction, and the Mallet score. Independent factors were time (pre-and post-surgery), gender, age, joint category (congruent, relocated, relocated plus osteotomy) and whether or not a transfer had been performed. Data from a previously published short-term evaluation were reworked in order to obtain pre-operative values.

The mean improvement in external rotation from pre-surgery to the long-term follow-up was 66.5° (95% confidence interval (CI) 61.5 to 71.6). The internal rotation had decreased by a mean of 22.6° (95% CI -18.7 to -26.5). The mean improvement in the three-grade aggregate Mallet score was 3.1 (95% CI 2.7 to 3.4), from 8.7 (95% CI 8.4 to 9.0) to 11.8 (11.5 to 12.1).

Our results show that open subscapularis elongation achieves good long-term results for patients with BPBP and an internal rotation contracture, providing lasting joint congruency and resolution of the trumpet sign, but with a moderate mean loss of internal rotation.

Cite this article: Bone Joint J 2014;96-B:1411–18

We retrospectively assessed the value of identifying impinging osteophytes using dynamic computer simulation of CT scans of the elbow in assisting their arthroscopic removal in patients with osteoarthritis of the elbow. A total of 20 patients were treated (19 men and one woman, mean age 38 years (19 to 55)) and followed for a mean of 25 months (24 to 29). We located the impinging osteophytes dynamically using computerised three-dimensional models of the elbow based on CT data in three positions of flexion of the elbow. These were then removed arthroscopically and a capsular release was performed.

The mean loss of extension improved from 23° (10° to 45°) pre-operatively to 9° (0° to 25°) post-operatively, and the mean flexion improved from 121° (80° to 140°) pre-operatively to 130° (110° to 145°) post-operatively. The mean Mayo Elbow Performance Score improved from 62 (30 to 85) to 95 (70 to 100) post-operatively. All patients had pain in the elbow pre-operatively which disappeared or decreased post-operatively. According to their Mayo scores, 14 patients had an excellent clinical outcome and six a good outcome; 15 were very satisfied and five were satisfied with their post-operative outcome.

We recommend this technique in the surgical management of patients with osteoarthritis of the elbow.

Cite this article: Bone Joint J 2014;96-B:237–41.

We retrospectively evaluated the clinical and radiological outcomes of a consecutive cohort of patients aged > 70 years with a displaced fracture of the olecranon, which was treated non-operatively with early mobilisation. We identified 28 such patients (27 women) with a mean age of 82 years (71 to 91). The elbow was initially immobilised in an above elbow cast in 90° of flexion of the elbow for a mean of five days. The cast was then replaced by a sling. Active mobilisation was encouraged as tolerated. No formal rehabilitation was undertaken. At a mean follow-up of 16 months (12 to 26), the mean ranges of flexion and extension were 140° and 15° respectively. On a visual analogue scale of 1 (no pain) to 10, the mean pain score was 1 (0 to 8). Of the original 28 patients 22 developed nonunion, but no patients required surgical treatment.

We conclude that non-operative functional treatment of displaced olecranon fractures in the elderly gives good results and a high rate of satisfaction.

Cite this article: Bone Joint J 2014;96-B:530–4.

Between 2003 and 2007, 99 knees in 77 patients underwent opening wedge high tibial osteotomy. We evaluated the effect of initial stable fixation combined with an artificial bone substitute on the mid- to long-term outcome after medial opening-wedge high tibial osteotomy (HTO) for medial compartmental osteoarthritis or spontaneous osteonecrosis of the knee in 78 knees in 64 patients available for review at a minimum of five years (mean age 68 years; 49 to 82). The mean follow-up was 6.5 years (5 to 10). The mean Knee Society knee score and function score improved from 49.6 (sd 11.4, 26 to 72) and 56.6 (sd 15.6, 5 to 100) before surgery to 88.1 (sd 12.5, 14 to 100) and 89.4 (sd 15.6, 5 to 100) at final follow-up (p <  0.001) respectively. There were no significant differences between patients aged ≥ 70 and < 70 years. The mean standing femorotibial angle was corrected significantly from 181.7° (sd 2.7°, 175° to 185°) pre-operatively to 169.7° (sd 2.4°, 164° to 175°) at one year’s follow-up (p < 0.001) and 169.6° (sd 3.0°, 157° to 179°) at the final follow-up (p = 0.69 vs one year).

Opening-wedge HTO using a stable plate fixation system combined with a bone substitute is a reliable procedure that provides excellent results. Although this treatment might seem challenging for older patients, our results strongly suggest that the results are equally good.

Cite this article: Bone Joint J 2014;96-B:339–44.

Inherent disadvantages of reverse shoulder arthroplasty designs based on the Grammont concept have raised a renewed interest in less-medialised designs and techniques. The aim of this study was to evaluate the outcome of reverse shoulder arthroplasty (RSA) with the fully-constrained, less-medialised, Bayley–Walker prosthesis performed for the treatment of rotator-cuff-deficient shoulders with glenohumeral arthritis. A total of 97 arthroplasties in 92 patients (53 women and 44 men, mean age 67 years (standard deviation (sd) 10, (49 to 85)) were retrospectively reviewed at a mean follow-up of 50 months ((sd 25) (24 to 96)). The mean Oxford shoulder score and subjective shoulder value improved from 47 (sd 9) and 24 points (sd 18) respectively before surgery to 28 (sd 11) and 61 (sd 24) points after surgery (p <  0.001). The mean pain at rest decreased from 5.3 (sd 2.8) to 1.5 (sd 2.3) (p < 0.001). The mean active forward elevation and external rotation increased from 42^°(sd 30) and 9^° (sd 15) respectively pre-operatively to 78^° (sd 39) and 24^° (sd 17) post-operatively (p < 0.001). A total of 20 patients required further surgery for complications; 13 required revision of components. No patient developed scapular notching.

The Bayley–Walker prosthesis provides reliable pain relief and reasonable functional improvement for patients with symptomatic cuff-deficient shoulders. Compared with other designs of RSA, it offers a modest improvement in forward elevation, but restores external rotation to some extent and prevents scapular notching. A longer follow-up is required to assess the survival of the prosthesis and the clinical performance over time.

Cite this article: Bone Joint J 2014;96-B:936–42.

We used immediate post-operative in vivo three-dimensional computed tomography to compare graft bending angles and femoral tunnel lengths in 155 patients who had undergone single-bundle reconstruction of the anterior cruciate ligament using the transtibial (n = 37), anteromedial portal (n = 72) and outside-in (n = 46) techniques.

The bending angles in the sagittal and axial planes were significantly greater but the coronal-bending angle was significantly less in the transtibial group than in the anteromedial portal and outside-in groups (p < 0.001 each). The mean length of the femoral tunnel in all three planes was significantly greater in the transtibial group than the anteromedial portal and outside-in groups (p < 0.001 each), but all mean tunnel lengths in the three groups exceeded 30 mm. The only significant difference was the coronal graft- bending angle in the anteromedial portal and outside-in groups (23.5° vs 29.8°, p = 0.012).

Compared with the transtibial technique, the anteromedial portal and outside-in techniques may reduce the graft-bending stress at the opening of the femoral tunnel. Despite the femoral tunnel length being shorter in the anteromedial portal and outside-in techniques than in the transtibial technique, a femoral tunnel length of more than 30 mm in the anteromedial portal and outside-in techniques may be sufficient for the graft to heal.

Cite this article: Bone Joint J 2014;96-B:743–51.

In 2012 we reviewed a consecutive series of 92 uncemented THRs performed between 1986 and 1991 at our institution using the CLS Spotorno stem, in order to assess clinical outcome and radiographic data at a minimum of 21 years. The series comprised 92 patients with a mean age at surgery of 59.6 years (39 to 77) (M:F 43;49).

At the time of this review, seven (7.6%) patients had died and two (2.2%) were lost to follow-up. The 23-year Kaplan–Meier survival rates were 91.5% (95% confidence intervals (CI) 85.4% to 97.6%; 55 hips at risk) and 80.3% (95% CI, 71.8% to 88.7%; 48 hips at risk) respectively, with revision of the femoral stem or of any component as endpoints. At the time of this review, 76 patients without stem revision were assessed clinically and radiologically (mean follow-up 24.0 years (21.5 to 26.5)). For the 76 unrevised hips the mean Harris hip score was 87.1 (65 to 97). Femoral osteolysis was detected in five hips (6.6%) only in Gruen zone 7. Undersized stems were at higher risk of revision owing to aseptic loosening (p = 0.0003). Patients implanted with the stem in a varus position were at higher risk of femoral cortical hypertrophy and thigh pain (p = 0.0006 and p = 0.0007, respectively).

In our study, survival, clinical outcome and radiographic data remained excellent in the third decade after implantation. Nonetheless, undersized stems were at higher risk of revision owing to aseptic loosening.

Cite this article: Bone Joint J 2014;96-B:455–61.