A total of 187 patients with primary osteoarthritis (OA) of the knee undergoing total knee replacement (TKR) were randomly divided into two groups, one of which underwent synovectomy. The patients and assessors were blinded to the randomisation both before and after surgery. The duration of surgery, hospitalisation period, concealed bleeding, drainage volume, blood transfusion rate and range of movement of the knee at three days after the operation were analysed. Patients were followed up at four weeks and 12 months after their operation, and a visual analogue score (VAS) for pain, Knee Society score (KSS) and a patellar ballottement test were compared between the groups.

The mean amount of concealed bleeding was higher in the synovectomy group compared with the control group (1.24 l (0.08 to 3.28) vs 1.03 l (0.16 to 2.94); p = 0.042), as was the mean drainage volume (0.90 l (0.35 to 1.81) vs 0.81 (0.25 to 1.65); p = 0.030). The mean operating time was also higher in the synovectomy group compared with the controls (1.50 hours (1.34 to 1.75) vs 1.41 hours (1.21 to 1.79); p = 0.006). There were no significant differences in blood transfusion rate (p = 0.882), hospital stay (p = 0.805) or range of movement of the knee (p = 0.413) between the two groups. At four weeks and 12 months post-operatively there were no statistically significant differences in any of the measured parameters. We concluded that synovectomy confers no clinical advantages in TKR for primary OA while subjecting patients to higher levels of bleeding and longer operating times.

Cite this article: Bone Joint J 2013;95-B:1197–200.

Fractures of the tibial shaft are common injuries, but there are no long-term outcome data in the era of increased surgical management. The aim of this prospective study was to assess the clinical and functional outcome of this injury at 12 to 22 years. Secondary aims were to determine the short- and long-term mortality, and if there were any predictors of clinical or functional outcome or mortality. From a prospective trauma database of 1502 tibial shaft fractures in 1474 consecutive adult patients, we identified a cohort of 1431 tibial diaphyseal fractures in 1403 patients, who fitted our inclusion criteria. There were 1024 men, and mean age at injury was 40.6 years. Fractures were classified according to the AO system, and open fractures graded after Gustilo and Anderson. Requirement of fasciotomy, time to fracture union, complications, incidence of knee and ankle pain at long-term follow-up, changes in employment and the patients’ social deprivation status were recorded. Function was assessed at 12 to 22 years post-injury using the Short Musculoskeletal Function Assessment and short form-12 questionnaires. Long-term functional outcome data was available for 568 of the surviving patients, 389 were deceased and 346 were lost to follow-up. Most fractures (90.7%, n = 1363) united without further intervention. Fasciotomies were performed in 11.5% of patients; this did not correlate with poorer functional outcome in the long term. Social deprivation was associated with a higher incidence of injury but had no impact on long-term function. The one-year mortality in those over 75 years of age was 29 (42%). At long-term follow-up, pain and function scores were good. However, 147 (26%) reported ongoing knee pain, 62 (10%) reported ankle pain and 97 (17%) reported both. Such joint pain correlated with poorer functional outcome.

Cite this article: Bone Joint J 2014;96-B:1370–7.

Patients who have limb amputation for musculoskeletal tumours are a rare group of cancer survivors. This was a prospective cross-sectional survey of patients from five specialist centres for sarcoma surgery in England. Physical function, pain and quality of life (QOL) outcomes were collected after lower extremity amputation for bone or soft-tissue tumours to evaluate the survivorship experience and inform service provision.

Of 250 patients, 105 (42%) responded between September 2012 and June 2013. From these, completed questionnaires were received from 100 patients with a mean age of 53.6 years (19 to 91). In total 60 (62%) were male and 37 (38%) were female (three not specified). The diagnosis was primary bone sarcoma in 63 and soft-tissue tumour in 37. A total of 20 tumours were located in the hip or pelvis, 31 above the knee, 32 between the knee and ankle and 17 in the ankle or foot. In total 22 had hemipelvectomy, nine hip disarticulation, 35 transfemoral amputation, one knee disarticulation, 30 transtibial amputation, two toe amputations and one rotationplasty. The Toronto Extremity Salvage Score (TESS) differed by amputation level, with poorer scores at higher levels (p < 0.001). Many reported significant pain. In addition, TESS was negatively associated with increasing age, and pain interference scores. QOL for Cancer Survivors was significantly correlated with TESS (p < 0.001). This relationship appeared driven by pain interference scores.

This unprecedented national survey confirms amputation level is linked to physical function, but not QOL or pain measures. Pain and physical function significantly impact on QOL. These results are helpful in managing the expectations of patients about treatment and addressing their complex needs.

Cite this article: Bone Joint J 2015;97-B:1284–90.

Periprosthetic femoral fracture (PFF) is a potentially devastating complication after total hip arthroplasty, with historically high rates of complication and failure because of the technical challenges of surgery, as well as the prevalence of advanced age and comorbidity in the patients at risk.

This study describes the short-term outcome after revision arthroplasty using a modular, titanium, tapered, conical stem for PFF in a series of 38 fractures in 37 patients.

The mean age of the cohort was 77 years (47 to 96). A total of 27 patients had an American Society of Anesthesiologists grade of at least 3. At a mean follow-up of 35 months (4 to 66) the mean Oxford Hip Score (OHS) was 35 (15 to 48) and comorbidity was significantly associated with a poorer OHS. All fractures united and no stem needed to be revised. Three hips in three patients required further surgery for infection, recurrent PFF and recurrent dislocation and three other patients required closed manipulation for a single dislocation. One stem subsided more than 5 mm but then stabilised and required no further intervention.

In this series, a modular, tapered, conical stem provided a versatile reconstruction solution with a low rate of complications.

Cite this article: Bone Joint J 2015;97-B:1031–7.

We undertook a retrospective study investigating the accuracy and safety of percutaneous pedicle screws placed under fluoroscopic guidance in the lumbosacral junction and lumbar spine. The CT scans of patients were chosen from two centres: European patients from University Medical Center Hamburg-Eppendorf, Germany, and Asian patients from the University of Malaya, Malaysia. Screw perforations were classified into grades 0, 1, 2 and 3. A total of 880 percutaneous pedicle screws from 203 patients were analysed: 614 screws from 144 European patients and 266 screws from 59 Asian patients. The mean age of the patients was 58.8 years (16 to 91) and there were 103 men and 100 women. The total rate of perforation was 9.9% (87 screws) with 7.4% grade 1, 2.0% grade 2 and 0.5% grade 3 perforations. The rate of perforation in Europeans was 10.4% and in Asians was 8.6%, with no significant difference between the two (p = 0.42). The rate of perforation was the highest in S1 (19.4%) followed by L5 (14.9%). The accuracy and safety of percutaneous pedicle screw placement are comparable to those cited in the literature for the open method of pedicle screw placement. Greater caution must be taken during the insertion of L5 and S1 percutaneous pedicle screws owing to their more angulated pedicles, the anatomical variations in their vertebral bodies and the morphology of the spinal canal at this location.

Cite this article: Bone Joint J 2015; 97-B:1111–17.

We describe the impact of a targeted performance improvement programme and the associated performance improvement interventions, on mortality rates, error rates and process of care for haemodynamically unstable patients with pelvic fractures. Clinical care and performance improvement data for 185 adult patients with exsanguinating pelvic trauma presenting to a United Kingdom Major Trauma Centre between January 2007 and January 2011 were analysed with univariate and multivariate regression and compared with National data. In total 62 patients (34%) died from their injuries and opportunities for improved care were identified in one third of deaths.

Three major interventions were introduced during the study period in response to the findings. These were a massive haemorrhage protocol, a decision-making algorithm and employment of specialist pelvic orthopaedic surgeons. Interventions which improved performance were associated with an annual reduction in mortality (odds ratio 0.64 (95% confidence interval (CI) 0.44 to 0.93), p = 0.02), a reduction in error rates (p = 0.024) and significant improvements in the targeted processes of care. Exsanguinating patients with pelvic trauma are complex to manage and are associated with high mortality rates; implementation of a targeted performance improvement programme achieved sustained improvements in mortality, error rates and trauma care in this group of severely injured patients.

Cite this article: Bone Joint J 2014;96-B:1090–7.

The incidence of periprosthetic fractures of the ankle is increasing. However, little is known about the outcome of treatment and their management remains controversial. The aim of this study was to assess the impact of periprosthetic fractures on the functional and radiological outcome of patients with a total ankle arthroplasty (TAA).

A total of 505 TAAs (488 patients) who underwent TAA were retrospectively evaluated for periprosthetic ankle fracture: these were then classified according to a recent classification which is orientated towards treatment. The outcome was evaluated clinically using the American Orthopedic Foot and Ankle Society (AOFAS) score and a visual analogue scale for pain, and radiologically.

A total of 21 patients with a periprosthetic fracture of the ankle were identified. There were 13 women and eight men. The mean age of the patients was 63 years (48 to 74). Thus, the incidence of fracture was 4.17%.

There were 11 intra-operative and ten post-operative fractures, of which eight were stress fractures and two were traumatic. The prosthesis was stable in all patients. Five stress fractures were treated conservatively and the remaining three were treated operatively.

A total of 17 patients (81%) were examined clinically and radiologically at a mean follow-up of 53.5 months (12 to 112). The mean AOFAS score at follow-up was 79.5 (21 to 100). The mean AOFAS score in those with an intra-operative fracture was 87.6 (80 to 100) and for those with a stress fracture, which were mainly because of varus malpositioning, was 67.3 (21 to 93). Periprosthetic fractures of the ankle do not necessarily adversely affect the clinical outcome, provided that a treatment algorithm is implemented with the help of a new classification system.

Cite this article: Bone Joint J 2015;97-B:950–6.

The aim of this study was to define the incidence of venous thromboembolism (VTE) and risk factors for the development of deep-vein thrombosis (DVT) after the resection of a musculoskeletal tumour. A total of 94 patients who underwent resection of a musculoskeletal tumour between January 2003 and December 2005 were prospectively studied. There were 42 men and 52 women with a mean age of 54.4 years (18 to 86). All patients wore intermittent pneumatic compression devices and graduated compression stockings. Ultrasound examination of the lower limbs was conducted to screen for DVT between the fifth and ninth post-operative days.

DVT was detected in 21 patients (22%). Of these, two were symptomatic (2%). One patient (1%) had a fatal pulmonary embolism. Patients aged ≥ 70 years had an increased risk of DVT (p = 0.004).

The overall incidence of DVT (both symptomatic and asymptomatic) after resection of a musculoskeletal tumour with mechanical prophylaxis was high. It seems that both mechanical and anticoagulant prophylaxis is needed to prevent VTE in patients who have undergone the resection of a musculoskeletal tumour.

Cite this article: Bone Joint J 2013;95-B:1280–4.

Tranexamic acid (TXA), an inhibitor of fibrinolysis, reduces blood loss after total knee arthroplasty. However, its effect on minimally invasive total hip arthroplasty (THA) is not clear. We performed a prospective, randomised double-blind study to evaluate the effect of two intravenous injections of TXA on blood loss in patients undergoing minimally invasive THA.

In total, 60 patients (35 women and 25 men with a mean age of 58.1 years; 17 to 84) who underwent unilateral minimally invasive uncemented THA were randomly divided into the study group (30 patients, 20 women and ten men with a mean age of 56.5 years; 17 to 79) that received two intravenous injections 1 g of TXA pre- and post-operatively (TXA group), and a placebo group (30 patients, 15 women and 15 men with a mean age of 59.5 years; 23 to 84). We compared the peri-operative blood loss of the two groups. Actual blood loss was calculated from the maximum reduction in the level of haemoglobin. All patients were followed clinically for the presence of venous thromboembolism.

The TXA group had a lower mean intra-operative blood loss of 441 ml (150 to 800) versus 615 ml (50 to 1580) in the placebo (p = 0.044), lower mean post-operative blood loss (285 ml (120 to 570) versus 392 ml (126 to 660) (p = 0.002), lower mean total blood loss (1070 ml (688 to 1478) versus 1337 ml (495 to 2238) (p = 0.004) and lower requirement for transfusion (p = 0.021). No patients in either group had symptoms of venous thromboembolism or wound complications.

This prospective, randomised controlled study showed that a regimen of two intravenous injections of 1 g TXA is effective for blood conservation after minimally invasive THA.

Cite this article: Bone Joint J 2015;97-B:905–10.

Oxidised zirconium (OxZi) has been developed as an alternative bearing surface for femoral heads in total hip arthroplasty (THA). This study has investigated polyethylene wear, functional outcomes and complications, comparing OxZi and cobalt–chrome (CoCr) as part of a three-arm, multicentre randomised controlled trial. Patients undergoing THA from four institutions were prospectively randomised into three groups. Group A received a CoCr femoral head and highly cross-linked polyethylene (XLPE) liner; Group B received an OxZi femoral head and XLPE liner; Group C received an OxZi femoral head and ultra-high molecular weight polyethylene (UHMWPE) liner. At five years, 368 patients had no statistically significant differences in short-form-36 (p = 0.176 mental, p = 0.756 physical), Western Ontario and McMaster Universities Osteoarthritis Index (p = 0.847), pain scores (p = 0.458) or complications. The mean rate of linear wear was 0.028 mm/year (standard deviation (sd) 0.010) for Group A, 0.023 mm/year (sd 0.010) for Group B, and 0.09 mm/year (sd 0.045) for Group C. Penetration was significantly higher in the UHMWPE liner group compared with both XLPE liner groups (p < 0.001) but no significant difference was noted between CoCr and OxZi when articulating with XLPE (p = 0.153). In this, the largest randomised study of this bearing surface, it appears that using a XLPE acetabular liner is more important in reducing THA component wear than the choice of femoral head bearing, at mid-term follow-up. There is a non-significant trend towards lower wear, coupling OxZi rather than CoCr with XLPE but long-term analysis is required to see if this observation changes with time and becomes significant.

Cite this article: Bone Joint J 2015;97-B:883–9.

Instability is the reason for revision of a primary total knee replacement (TKR) in 20% of patients. To date, the diagnosis of instability has been based on the patient’s symptoms and a subjective clinical assessment. We assessed whether a measured standardised forced leg extension could be used to quantify instability.

A total of 25 patients (11 male/14 female, mean age 70 years; 49 to 85) who were to undergo a revision TKR for instability of a primary implant were assessed with a Nottingham rig pre-operatively and then at six and 26 weeks post-operatively. Output was quantified (in revolutions per minute (rpm)) by accelerating a stationary flywheel. A control group of 183 patients (71 male/112 female, mean age 69 years) who had undergone primary TKR were evaluated for comparison.

Pre-operatively, all 25 patients with instability exhibited a distinctive pattern of reduction in ‘mid-push’ speed. The mean reduction was 55 rpm (sd 33.2). Post-operatively, no patient exhibited this pattern and the reduction in ‘mid-push’ speed was 0 rpm. The change between pre- and post-operative assessment was significant (p < 0.001). No patients in the control group exhibited this pattern at any of the intervals assessed. The between-groups difference was also significant (p < 0.001). This suggests that a quantitative diagnostic test to assess the unstable primary TKR could be developed.

Cite this article: Bone Joint J 2014;96-B:1339–43.

The effects of surgical approach in total hip replacement on health-related quality of life and long-term pain and satisfaction are unknown. From the Swedish Hip Arthroplasty Register, we extracted data on all patients that had received a total hip replacement for osteoarthritis through either the posterior or the direct lateral approach, with complete pre- and one-year post-operative Patient Reported Outcome Measures (PROMs). A total of 42 233 patients met the inclusion criteria and of these 4962 also had complete six-year PROM data. The posterior approach resulted in an increased mean satisfaction score of 15 (sd 19) vs 18 (sd 22) (p <  0.001) compared with the direct lateral approach. The mean pain score was 13 (sd 17) vs 15 (sd 19) (p < 0.001) and the proportion of patients with no or minimal pain was 78% vs 74% (p < 0.001) favouring the posterior approach. The patients in the posterior approach group reported a superior mean EQ-5D index of 0.79 (sd 0.23) vs 0.77 (sd 0.24) (p < 0.001) and mean EQ score of 76 (sd 20) vs 75 (sd 20) (p < 0.001). All observed differences between the groups persisted after six years follow-up. Although PROMs after THR in general are very good regardless of surgical approach, the results indicate that some patients operated by the direct lateral approach report an inferior outcome compared with the posterior approach. The large number of procedures and the seemingly sustained differences make it likely these findings are clinically relevant.

Cite this article: Bone Joint J 2014;96-B:590–6.

Rebound growth after hemiepiphysiodesis may be a normal event, but little is known about its causes, incidence or factors related to its intensity. The aim of this study was to evaluate rebound growth under controlled experimental conditions.

A total of 22 six-week-old rabbits underwent a medial proximal tibial hemiepiphysiodesis using a two-hole plate and screws. Temporal growth plate arrest was maintained for three weeks, and animals were killed at intervals ranging between three days and three weeks after removal of the device. The radiological angulation of the proximal tibia was studied at weekly intervals during and after hemiepiphysiodesis. A histological study of the retrieved proximal physis of the tibia was performed.

The mean angulation achieved at three weeks was 34.7° (standard deviation (sd) 3.4), and this remained unchanged for the study period of up to two weeks. By three weeks after removal of the implant the mean angulation had dropped to 28.2° (sd 1.8) (p < 0.001). Histologically, widening of the medial side was noted during the first two weeks. By three weeks this widening had substantially disappeared and the normal columnar structure was virtually re-established.

In our rabbit model, rebound was an event of variable incidence and intensity and, when present, did not appear immediately after restoration of growth, but took some time to appear.

Cite this article: Bone Joint J 2015;97-B:862–8.

Acetabular labral tears and associated intra-articular pathology of the hip have been recognised as a source of symptoms. However, it is now appreciated that there is a relatively high prevalence of asymptomatic labral tears. In this study, 70 young asymptomatic adult volunteers with a mean age of 26 years (19 to 41) were recruited and underwent three tesla non-arthrographic MR scans. There were 47 women (67.1%) and 23 men (32.9%).

Labral tears were found in 27 volunteers (38.6%); these were an isolated finding in 16 (22.9%) and were associated with other intra-articular pathology in the remaining 11 (15.7%) volunteers. Furthermore, five (7.1%) had intra-articular pathology without an associated labral tear.

Given the high prevalence of labral pathology in the asymptomatic population, it is important to confirm that a patient's symptoms are due to the demonstrated abnormalities when considering surgery.

Cite this article: Bone Joint J 2015;97-B:623–7.

We describe the outcome of tibial diaphyseal fractures in the elderly (≥ 65 years of age). We prospectively followed 233 fractures in 225 elderly patients over a minimum ten-year period. Demographic and descriptive data were acquired from a prospective trauma database. Mortality status was obtained from the General Register Office database for Scotland. Diaphyseal fractures of the tibia in the elderly occurred predominantly in women (73%) and after a fall (61%). During the study period the incidence of these fractures decreased, nearly halving in number. The 120-day and one-year unadjusted mortality rates were 17% and 27%, respectively, and were significantly greater in patients with an open fracture (p < 0.001). The overall standardised mortality ratio (SMR) was significantly increased (SMR 4.4, p < 0.001) relative to the population at risk, and was greatest for elderly women (SMR 8.1, p < 0.001). These frailer patients had more severe injuries, with an increased rate of open fractures (30%), and suffered a greater rate of nonunion (10%).

Tibial diaphyseal fractures in the elderly are most common in women after a fall, are more likely to be open than in the rest of the population, and are associated with a high incidence of nonunion and mortality.

Cite this article: Bone Joint J 2013;95-B:1255–62.

Deep vein thrombosis is a common complication of immobilising the lower limb after surgery. We hypothesised that intermittent pneumatic compression (IPC) therapy in outpatients who had undergone surgical repair of acute ruptures of the Achilles tendon could reduce the incidence of this problem.

A total of 150 patients who had undergone surgical repair of the Achilles tendon were randomised to either treatment with IPC for six hours per day (n = 74) under an orthosis or treatment as usual (n = 74) in a plaster cast without IPC. At two weeks post-operatively, the incidence of deep vein thrombosis was assessed using blinded, double-reported compression duplex ultrasound. At this point, IPC was discontinued and all patients were immobilised in an orthosis for a further four weeks. At six weeks post-operatively, a second compression duplex ultrasound scan was performed.

At two weeks, the incidence of deep vein thrombosis was 21% in the treated group and 37% in the control group (p = 0.042). Age over 39 years was found to be a strong risk factor for deep vein thrombosis (odds ratio (OR) = 4.84, 95% confidence interval (CI) 2.14 to 10.96). Treatment with IPC, corrected for age differences between groups, reduced the risk of deep vein thrombosis at the two-week point (OR = 2.60; 95% CI 1.15 to 5.91; p =0.022). At six weeks, the incidence of deep vein thrombosis was 52% in the treated group and 48% in the control group (OR 0.94, 95% CI 0.49 to 1.83). IPC appears to be an effective method of reducing the risk of deep vein thrombosis in the early stages of post-operative immobilisation of outpatients. Further research is necessary to elucidate whether it can confer similar benefits over longer periods of immobilisation and in a more heterogeneous group of patients.

Cite this article: Bone Joint J 2015;97-B:675–80.

To confirm whether developmental dysplasia of the hip has a risk of hip impingement, we analysed maximum ranges of movement to the point of bony impingement, and impingement location using three-dimensional (3D) surface models of the pelvis and femur in combination with 3D morphology of the hip joint using computer-assisted methods. Results of computed tomography were examined for 52 hip joints with DDH and 73 normal healthy hip joints. DDH shows larger maximum extension (p = 0.001) and internal rotation at 90° flexion (p < 0.001). Similar maximum flexion (p = 0.835) and external rotation (p = 0.713) were observed between groups, while high rates of extra-articular impingement were noticed in these directions in DDH (p < 0.001). Smaller cranial acetabular anteversion (p = 0.048), centre-edge angles (p < 0.001), a circumferentially shallower acetabulum, larger femoral neck anteversion (p < 0.001), and larger alpha angle were identified in DDH. Risk of anterior impingement in retroverted DDH hips is similar to that in retroverted normal hips in excessive adduction but minimal in less adduction. These findings might be borne in mind when considering the possibility of extra-articular posterior impingement in DDH being a source of pain, particularly for patients with a highly anteverted femoral neck.

Cite this article: Bone Joint J 2014;96-B:580–9.

Orthopaedic surgeons have accepted various radiological signs to be representative of acetabular retroversion, which is the main characteristic of focal over-coverage in patients with femoroacetabular impingement (FAI). Using a validated method for radiological analysis, we assessed the relevance of these signs to predict intra-articular lesions in 93 patients undergoing surgery for FAI. A logistic regression model to predict chondral damage showed that an acetabular retroversion index (ARI) > 20%, a derivative of the well-known cross-over sign, was an independent predictor (p = 0.036). However, ARI was less significant than the Tönnis classification (p = 0.019) and age (p = 0.031) in the same model. ARI was unable to discriminate between grades of chondral lesions, while the type of cam lesion (p = 0.004) and age (p = 0.047) were able to. Other widely recognised signs of acetabular retroversion, such as the ischial spine sign, the posterior wall sign or the cross-over sign were irrelevant according to our analysis. Regardless of its secondary predictive role, an ARI > 20% appears to be the most clinically relevant radiological sign of acetabular retroversion in symptomatic patients with FAI.

Cite this article: Bone Joint J 2013;95-B:893–9.

The aim of this study was to assess the effect of injecting genetically engineered chondrocytes expressing transforming growth factor beta 1 (TGF-β1) into the knees of patients with osteoarthritis. We assessed the resultant function, pain and quality of life.

A total of 54 patients (20 men, 34 women) who had a mean age of 58 years (50 to 66) were blinded and randomised (1:1) to receive a single injection of the active treatment or a placebo. We assessed post-treatment function, pain severity, physical function, quality of life and the incidence of treatment-associated adverse events. Patients were followed at four, 12 and 24 weeks after injection.

At final follow-up the treatment group had a significantly greater improvement in the mean International Knee Documentation Committee score than the placebo group (16 points; -18 to 49, vs 8 points; -4 to 37, respectively; p = 0.03). The treatment group also had a significantly improved mean visual analogue score at final follow-up (-25; -85 to 34, vs -11 points; -51 to 25, respectively; p = 0.032). Both cohorts showed an improvement in Western Ontario and McMaster Osteoarthritis Index and Knee Injury and Osteoarthritis Outcome Scores, but these differences were not statistically significant. One patient had an anaphylactic reaction to the preservation medium, but recovered within 24 hours. All other adverse events were localised and resolved without further action.

This technique may result in improved clinical outcomes, with the aim of slowing the degenerative process, leading to improvements in pain and function. However, imaging and direct observational studies are needed to verify cartilage regeneration. Nevertheless, this study provided a sufficient basis to proceed to further clinical testing.

Cite this article: Bone Joint J 2015;97-B:924–32.

Corticosteroid use has been implicated in the development of osteonecrosis of the femoral head (ONFH). The exact mechanism and predisposing factors such as age, gender, dosage, type and combination of steroid treatment remain controversial. Between March and July 2003, a total of 539 patients with severe acute respiratory syndrome (SARS) were treated with five different types of steroid. There were 129 men (24%) and 410 women (76%) with a mean age of 33.7 years (21 to 59). Routine screening was undertaken with radiographs, MRI and/or CT to determine the incidence of ONFH.

Of the 129 male patients with SARS, 51 (39.5%) were diagnosed as suffering from ONFH, compared with only 79 of 410 female patients (19.3%). The incidence of ONFH in the patients aged between 20 and 49 years was much higher than that of the group aged between 50 and 59 years (25.9% (127 of 491) versus 6.3% (3 of 48); p = 0.018). The incidence of ONFH in patients receiving one type of steroid was 12.5% (21 of 168), which was much lower than patients receiving two different types (28.6%; 96 of 336) or three different types of steroid (37.1%; 13 of 35).

Cite this article: Bone Joint J 2014;96-B:259–62.