Avascular necrosis (AVN) is a serious complication of high-dose chemotherapy for haematological malignancy in childhood. In order to describe its incidence and main risk factors and to evaluate the current treatment options, we reviewed 105 children with a mean age of 8.25 years (1 to 17.8) who had acute lymphoblastic or acute myeloid leukaemia, or a non-Hodgkin’s lymphoma. Overall, eight children (7.6%) developed AVN after a mean of 16.8 months (8 to 49). There were four boys and four girls with a mean age of 14.4 years (9.8 to 16.8) and a total of 18 involved sites, 12 of which were in the femoral head. All these children were aged > nine years (p < 0.001). All had received steroid treatment with a mean cumulative dose of prednisone of 5967 mg (4425 to 9599) compared with a mean of 3943 mg (0 to 18 585) for patients without AVN (p = 0.005). No difference existed between genders and no thrombophilic disorders were identified. Their initial treatment included 11 core decompressions and two bipolar hip replacements. Later, two salvage osteotomies were done and three patients (four hips) eventually needed a total joint replacement. We conclude that AVN mostly affects the weight-bearing epiphyses. Its risk increases with age and higher steroid doses. These high-risk patients may benefit from early screening for AVN.

Cite this article: Bone Joint J 2013;95-B:1708–13.

Introduction: The aim of this study is long term comparison of hydroxyapatite (HA) coating and porous (PO) coating in an identical stem design.

Material and Methods: 100 consecutive hips from 86 patients scheduled for uncemented primary arthroplasty were quasi-randomized to receive a titanium-alloy anatomic PROFILE stem (DePuy, Warsaw, IN), HA coated in every second hip and PO coated in the remainder 50 hips. Patients receiving bilateral total hip arthroplasty were implanted an HA coated stem in one and a PO coated in the contralateral hip. All surgeries were performed by the posterolateral approach by 2 surgeons (the author PBT being one of them). The acetabular component was either an AML+ or a Duraloc 300 implant with a cobalt chrome 28mm head (DePuy, Warsay, IN).

15 to 18 years postoperatively these patients and their hip radiographs were examined by an independent observer (WD) to establish the long term survival, clinical and radiographical performance of this stem and especially to compare the two coatings. The Harris hip score was employed for clinical evaluation

Results: 16 years and 3 months (range 15y 0m – 17y 8m) postoperatively 21 patients with 23 hips had died and 4 hips (2HA/2 PO) had been revised. This leaves 62 patients with 73 hips (34 HA/39 PO) for evaluation. Life tables showed 18 years cumulative survival (free of revision of any reason) of 95.8% for HA and 95.5% for PO. 18 years cumulative survival (free of revision for aseptic loosening) were 95.8% for HA and 97.5% for PO. There were no infections in either group. Mean HHS was 83.4 ± 14.8 in HA and 86.8 ± 11.5 in PO (P = 0.32). Mean pain score was 39.1 ± 9.4 in HA and 40.9 ± 6.7 in PO (P = 0.69). Radiographs showed osseointegration of the stem except in 1 HA and in 2 PO stems. Lucencies/osteolytic scalloping were only seen in zone 1 and 7. Significant bone remodeling changes were seen. Ectopic ossification developed in 1/3 of hips non progressive stem subsidence of 2 mm were measured in a few hips.

Conclusion: This prospective quasi-randomized study show excellent long term performance and survival of the titanium-alloy anatomic PROFILE stem in primary hip arthroplasty in patients < 66 years with no significant differences between HA and PO coating.

Introduction: The aim of this study was to investigate if steroids enhance the vasoconstrictive effect of endothe-lin-1 (ET-1) on femoral arteries.

Materials and Methods: Ten female Wistar rats 59 to 88 days of age and 238 to 310 g of body weight, were used. Forty femoral artery segments were harvested. These arterial segments were mounted as ring preparations on a small vessel myograph. Two vessels from each animal were randomized to incubation with methylprednisolone 5 μg/ml [1] while the other 2 vessels were incubated with placebo. The arteries were stimulated cumulatively with endothelin-1. Isometric wall tension was quantified by the EC50; the vasoconstrictor concentration resulting in half maximal contraction.

Results: Thirty-eight arteries could be harvested in total; 20 were randomized to steroid treatment while 18 served as controls. The endothelin-1 dose-response curve displayed a stronger contraction for the steroid group in relation to the controls with increasing doses of ET-1. The EC50 of 4.4*10⁻⁸ M ± 1.8*10⁻⁸ M for the steroid vessels was lower compared to 5.9*10⁻⁸ M ± 3.4*10⁻⁸ M for the controls (mean ±SD; n.s.).

Discussion: Endothelin-1 is a potent vasoconstrictor. This study showed that incubation with methylprednisolone enhanced ET-1 mediated contraction of femoral arteries which can diminish blood flow within the vascular bed supplying the femoral head. This may be a relevant cofactor in the early pathogenesis of steroid-associated femoral head necrosis.

Introduction: The aim of this study was to investigate if steroids enhance the vasoconstrictive effect of nor-adrenaline on femoral arteries, which may result in femoral head blood flow reduction.

Materials and Methods: Ten male Wistar rats 62 to 88 days of age, 254 to 318 g of body weight, were used. Twenty femoral artery segments were harvested. These arterial segments were mounted as ring preparations on a small vessel myograph for isometric force measurements. The arteries were stimulated cumulatively with noradrenaline before and after incubation with methylprednisolone (5 μg/ml). Isometric wall tension was plotted and quantified by the EC50, the vasoconstrictor concentration resulting in halfmaximal contraction.

Results: The noradrenaline dose-response curve displayed a shift to the left for the steroid group in relation to the controls. This was reflected by a significantly lower EC50 of 9.5*10⁻⁷ M ± 5.1*10⁻⁷ M for the steroid vessels compared to 2.5*10⁻⁶ M ± 1.1*10⁻⁶ M for the control vessels (mean ± SD; p< 0.005).

Discussion: This study showed that incubation with methylprednisolone enhanced noradrenaline-mediated contraction of femoral arteries. Enhanced contraction of femoral arteries can diminish blood flow within the vascular bed supplying the femoral head. This may be a relevant cofactor in the early pathogenesis of steroid-associated femoral head necrosis.

Aims: The present experimental study raised the question whether corticosteroid therapy inßuences the sensitivity of the femoral head circulation to ischemia induced by hip joint tamponade. Methods: 31 Landrace pigs were treated in 4 groups. 12 animals received a 14 day methylprednisolone intramuscular application before hip joint tamponade. 11 pigs underwent hip joint tamponade without previous medication. Control groups comprised 4 animals not undergoing hip joint tamponade. Blood ßow measurement was undertaken in predeþned regions by radioactive microsphere technique. Results: Epiphyseal blood ßow decreased signiþcantly during hip joint tamponade. Reperfusion occurred to a level not signiþcantly differing from that before ischemia, whereas epiphyses remained ischemic in 2 pigs. In the steroid treated animals, the basic blood ßow appeared 2–3 times lower than that of the non medicated pigs. Also in the steroid group 2 epiphyseal remained ischemic. The metaphyseal corticalis in the steroid treated animals revealed signiþcant hyperperfusion. Conclusions: Ischemia by hip joint tamponade in a porcine model was produced quantitatively for the þrst time. The majority of femoral head epiphyses was reperfused on steady state blood ßow level. Nonreperfusion of 2 epiphyses in each group indicated that 6 hours of ischemia might be just below the minimum stress in order to produce necrosis of the femoral head. High dose steroid medication reduced the steady state blood ßow level of the femoral head 2–3 times but did not inhibit or disturb reperfusion.

Aims: The present experiment addressed the question whether lipopolysaccharides (LPS), hip joint tamponade or their combination modulate hip perfusion. Methods: 16 immature Danish Landrace pigs of both genders were treated in 3 groups. 4 animals received LPS from escherichia coli intravenously 4 hours previous to hip joint tamponade. 8 pigs underwent the hip operation without previous medication. 4 animals without treatment served as control group. Blood ßow measurement was done by the Radioactive Tracer Microspheres technique. Results: Femoral head epiphyseal blood ßow decreased signiþcantly during hip joint tamponade. Reperfusion occurred to a level not signiþcantly differing from that before ischemia, whereas epiphyses remained ischemic in 2 pigs. The hip joint capsule showed signiþcant hyperperfusion during and after joint tamponade. No signiþcant difference was revealed comparing the LPS-treated and non-treated groups of pigs in all hip regions (p = 0.79, U-test). In addition, in the LPS-group, none of the femoral head epiphyses remained ischemic. Conclusions: LPS and hip joint tamponade, which have separately been discussed as pathomechanic factors of Non Traumatic Femoral Head Necrosis, have been combined in a bifactorial porcine model. Systemic lipopolysacchrides as bacterial endotoxin have no acute effect on regional hip perfusion which would make a consequent osteonecrosis probable. 6hourly hip joint tamponade alone evoked non reperfusion in 2 out of 8 pigs and a prolongation of the 6 hours ischemia might evoke more cases of non reperfusion.

Aims: The pathomechanism of avascular necrosis of the femoral head (AVN) is still debated. Hip joint synovitis and effusion may impair blood ßow to the femoral head. The critical ischemia time is around 6 hours, but repeated ischemic episodes may impair reperfusion and elicit AVN. The aims of this study were to investigate the value of dynamic MRI in femoral head after ischemia and during reperfusion. Methods: In 15 domestic pigs, 3–4 months old, femoral head ischemia was achieved by raising the joint pressure to 250 mmHg by dextran infusion through a hole in the acetabular wall. MRI was performed (Philips gyroscan S15, 1.5 T, Gd-DTPA enhancement, dynamic imaging interval 39 sec.) before ischemia, after 6 hours of ischemia, and again after 4 hours of reperfusion. Results: Signal intensity versus time (SI/t) plots were constructed from 347 MR studies. By regression analysis of SI/t curves an index (enhancement/decrease) was developed as criterion for arterial or venous circulatory disturbance. Index values < 1.1 signiþed arterial impairment, > 100 venous disturbance. Values between 1.1 and 100 were considered normal. The positive predictive value for disturbed osseous blood ßow was 96%. Conclusions: Early detection of intraosseous circulatory disturbance was possible with a mathematical model for dynamic MRI results. The method is reproducible and may be employed in the early diagnosis of AVN and during treatment for monitoration of revascularisation.

Aims: Increased intraarticular hip joint pressure has been considered a pathomechanism in femoral head necrosis. The aim of this study was to investigate histopathological femoral head changes in an immature big animal model of arterial hip joint tamponade. Methods: Out of a total of 15 domestic pigs, 11 animals were randomly chosen to undergo 6h unilateral hip joint tamponade at an intraarticular pressure of 250mmHg while 4 animals underwent a unilateral sham operation serving as controls. 4h after the end of hip joint tamponade, the animals were killed with potassium chloride, the femoral heads were excised, and þxed in Schafferñs solution for undecalciþed Goldnerñs, alcianblue-PAS, Krutsay, methyl green pyronine, toluidin blue O, and standard Giemsa staining. Results: A great number of congested sinus and vessels of the terminal vascular system showed inclusions of blood cells dominantly in the tamponaded hip side. Congestion also be documented by dilated sinus with deformed blood cells. Bone remodeling of normal osteoblast and osteoclastic lacuna activity was noted in all trabeculae. None of the known signs of osteonecrosis were found. Conclusions: Our acute histological study in immature pigs shows the early microcirculatory disturbances which may precede femoral head necrosis. Future research is needed to investigate histologic changes after a longer time interval following hip joint tamponade, and into the duration of the joint tamponade.

The treatment of osteonecrosis of the femoral head (FHN) is controversial. It mainly occurs in young patients in whom total hip replacement is best avoided because of an increased risk of revision. The objective of this long-term follow-up study was to evaluate the outcome of intertrochanteric flexion osteotomy as a hip joint preserving operation for FHN.

Over a 19-year period we carried out 70 intertrochanteric flexion osteotomies for FHN in 64 patients. The mean follow-up was 10.4 years (3.0 to 20.3). The overall mean Harris hip score increased from 51 points preoperatively to 71 points postoperatively. Six patients (9%) developed early postoperative complications. A total of 19 hips (27%) underwent total hip arthroplasty at a mean of 8.7 years after osteotomy. The five-year survival rate was 90%. Survival rates of hips in Ficat stage 2 were higher than those in stages 3 or 4. Hips with a preoperative necrotic angle of < 200° had a better survival probability than those with a necrotic angle > 200°. Our findings suggest that flexion osteotomy is a safe and effective procedure in Ficat stage 2 and 3 FHN, preferably with a necrotic angle of < 200°.

Introduction

Synovectomy in children with juvenile rheumatoid arthritis (JRA) and psoriatic arthritis (PSA) is still subject of controversial discussion. Our results of arthroscopic synovectomy of the knee in children with chronic inflammatory joint disease are presented.

Material

From 1989–1997 27 synovectomies were performed in 27 children with inflammatory arthritis (15 JRA, 12 PSA). Average age at surgery was 12. 5 y (2. 9–17. 8 y). Mean follow-up was 4. 9 years.

Methods

Onset of disease and conservative therapy was documented. Each patient was physically and radiologically examined preoperatively and 24 children postoperatively (mean follow-up 4. 9 years). For arthroscopic shaver-assisted synovectomy of the knee we used minimum 4 portals and normally 6 portals (2 anterior, 2 suprapatellar and 2 posterior portals). In addition to the physical examination we used a special clinical score (Laurin 1974). We compared the pre- and postoperative limits of active and passive knee movement. We performed sonographs and radiographs of the infected joint. Radiography was classified following the Larsen-Scale. Patient and parents gave their opinion whether the operation was successful. Before surgery all children had intensive drug and physical therapy for 8–62 months (42 month). In the course of conservative treatment, knees had local joint treatment with triamcinolonhex-acetomid (THA), normally for three times before surgery. Preoperative X-rays showed Larsen stage I in 3 knee joints and Larsen stage 0 in the other knees.

Results

In 85% of the children, we found good or excellent surgical outcome. 2 joints achieved fair and 2 joints poor outcome. Concerning subjective outcome 22 (82%) children had been very satisfied (56%) or satisfied (26%). 25 of the children’s parents would agree in the same surgical procedure again. In 6 knee joints we found recurrent synovitis. 2 of these knee joints were reoperated (30 and 22 month postoperatively with satisfying result), the other 4 joints were treated with THA i. a.. The 2 reoperations were regarded as poor result. We had no postoperative complications and the normal hospitalisation was 15 days. Prior to surgery, 12 knee joints had an average deficit of active knee extension of 10° (5–20°). Postoperatively, no extension deficit was found anymore in 25 of the knees. Compared to the contralateral knees, a flexion deficit of 10° (5–15°) was obtained postoperatively. At sonography, no joint effusion could be revealed. Postoperative X-rays showed no progression in Larsen stage. Outcome in children with oligoarthritis was better than in those with polyarthritic disease.

Discussion

Early arthroscopic synovectomy of the knee in children with chronic inflammatory joint disease is, in case of failure of conservative treatment, a useful method of treatment. We propose early synovectomy of the knee joint as an essential part of the treatment scheme for children with inflammatory joint disease.

Treatment with corticosteroids is a risk factor for non-traumatic avascular necrosis of the femoral head, but the pathological mechanism is poorly understood. Short-term treatment with high doses of methylprednisolone is used in severe neurotrauma and after kidney and heart transplantation. We investigated the effect of such treatment on the pattern of perfusion of the femoral head and of bone in general in the pig.

We allocated 15 immature pigs to treatment with high-dose methylprednisolone (20 mg/kg per day intramuscularly for three days, followed by 10 mg/kg intramuscularly for a further 11 days) and 15 to a control group. Perfusion of the systematically subdivided femoral head, proximal femur, acetabulum, humerus, and soft tissues was determined by the microsphere technique. Blood flow in bone was severely reduced in the steroid-treated group. The reduction of flow affected all the segments and the entire epiphysis of the femoral head. No changes in flow were found in non-osseous tissue. Short-term treatment with high-dose methylprednisolone causes reduction of osseous blood flow which may be the pathogenetic factor in the early stage of steroid-induced osteonecrosis.