Abstract

The clinical success of osteochondral autografts is heavily reliant on their mechanical stability, as grafts which protrude above or subside below the native cartilage can have a negative effect on the tribological properties of the joint [1]. Furthermore, high insertion forces have previously been shown to reduce chondrocyte viability [2]. Commercial grafting kits may include a dilation tool to increase the diameter of the recipient site prior to insertion. The aim of this study was to evaluate the influence of dilation on the primary stability of autografts.

Six human cadaveric femurs were studied. For each femur, four 8.5 × 8mm autografts were harvested from the trochlear groove and implanted into the femoral condyles using a Smith & Nephew Osteochondral grafting kit. Two grafts were implanted into dilated recipient sites (n=12) and two were implanted with no dilation (n=12). Insertion force was measured by partially inserting the graft and applying a load at a rate of 1 mm/min, until the graft was flush with the surrounding cartilage. Push-in force was measured by applying the same load, until the graft had subsided 4mm below congruency. Significance was taken as (p<0.05).

Average maximum insertion force of dilated grafts was significantly lower (p<0.001) than their non-dilated equivalent [28.2N & 176.7N respectively]. There was no significant difference between average maximum push-in force between the dilated and non-dilated groups [1062.8N & 1204.2N respectively].

This study demonstrated that significantly less force is required to insert dilated autografts, potentially minimising loss of chondrocyte viability. However, once inserted, the force required to displace the grafts below congruency remained similar, indicating a similar degree of graft stability between both groups.

Introduction

PEEK-OPTIMA™ has been considered as an alternative to cobalt chrome in the femoral component of total knee replacements. Whole joint wear simulation studies of both the tibiofemoral and patellofemoral joints carried out to date have shown an equivalent wear rate of UHMWPE tibial and patella components against PEEK and cobalt chrome (CoCr) femoral components. In this study, the influence of third body wear on UHMWPE-on-PEEK was investigated, tests on UHMWPE-on-CoCr were carried out in parallel to compare PEEK to a conventional femoral component material.

Introduction

PEEK-OPTIMA™ has been considered as an alternative to cobalt chrome in the femoral component of total knee replacements. Wear simulation studies of both the tibiofemoral and patellofemoral joints carried out to date have shown an equivalent wear rate of UHMWPE tibial and patella components against PEEK and cobalt chrome (CoCr) femoral components implanted under optimal alignment conditions. In this study, fundamental pin-on-plate studies have been carried out to investigate the wear of UHMWPE-on-PEEK under a wider range of contact pressure and cross-shear conditions.

Objectives

Bone void fillers are increasingly being used for dead space management in arthroplasty revision surgery. The aim of this study was to investigate the influence of calcium sulphate bone void filler (CS-BVF) on the damage and wear of total knee arthroplasty using experimental wear simulation.

Introduction

PEEK-OPTIMA™ has been considered as an alternative bearing material to cobalt chrome in the femoral component of total knee replacements. To better understand the tribology of UHMWPE-on-PEEK-OPTIMA™ and to find the most appropriate environmental conditions under which to test this novel bearing material combination, a series of tests under different protein lubricant concentrations at rig (∼24°C) and elevated temperature (∼35°C) were carried out in simple geometry wear and friction rigs. Under all conditions, the wear of UHMWPE-on-PEEK-OPTIMA™ was compared to UHMWPE-on-cobalt chrome (CoCr).

Introduction

Experimental wear simulation of an all-polymer knee implant has shown an equivalent rate of wear of UHMWPE tibial components against PEEK-OPTIMA™ and cobalt chrome femoral components of a similar initial geometry and surface topography. However, when the patella is resurfaced with an UHMWPE patella button, it is important to also ascertain the wear of the patella. Wear debris from the patella contributes to the total volume of wear debris produced by the implant which should be minimised to reduce the potential for osteolysis and subsequent implant loosening. The aim of this study was to investigate the wear of the patellofemoral joint in an all-polymer knee implant. The wear of UHMWPE patellae articulating against PEEK-OPTIMA™ femoral components was compared to UHMWPE articulating against cobalt chrome femoral components.

Introduction

Calcium sulfate bone void fillers (CS-BVF) are increasingly being used for dead space management in infected arthroplasty revision surgery. The use of loose beads of CS-BVF close to the articulating surfaces of an implant means there is potential for them to migrate between the articulating surfaces acting as a third body particle. The aim of this study was to investigate the influence of CS-BVF on the third body wear of total knee replacements.

Introduction

There is a demand for longer lasting arthroplasty implants driving the investigation of novel material combinations. PEEK has shown promise as an arthroplasty bearing material, with potentially relatively bio inert wear debris [1]. When coupled with an all-polyethylene tibial component this combination shows potential as a metal-free knee. In this study, the suitability of PEEK Optima® as an alternative to cobalt chrome for the femoral component of total knee replacements was assessed using experimental knee wear simulation under two kinematic conditions.

Introduction

UHMWPE articulating against PEEK-OPTIMA® has the potential for use as a novel bearing couple in joint arthroplasty due to its potentially low wear rates and the bioinertness of its wear debris. The aim of this study was to investigate the role of protein in the lubricant on the wear of UHMWPE articulating against PEEK at both room and physiological temperature.

Introduction

When third body particles originating from bone cement or bone void fillers become trapped between articulating surfaces of joint replacements, contact surfaces may be damaged leading to accelerated wear and premature failure of the implant. In this study, the damage to cobalt chrome counterfaces by third body particles from PMMA bone cement (GMV, DePuy) and various bone void fillers was investigated; then wear tests of UHMWPE were carried out against these surfaces.

Extracorporeal irradiation of an excised tumour-bearing segment of bone followed by its re-implantation is a technique used in bone sarcoma surgery for limb salvage when the bone is of reasonable quality. There is no agreement among previous studies about the dose of irradiation to be given: up to 300 Gy have been used.

We investigated the influence of extracorporeal irradiation on the elastic and viscoelastic properties of bone. Bone was harvested from mature cattle and subdivided into 13 groups: 12 were exposed to increasing levels of irradiation: one was not and was used as a control. The specimens, once irradiated, underwent mechanical testing in saline at 37°C.

The mechanical properties of each group, including Young’s modulus, storage modulus and loss modulus, were determined experimentally and compared with the control group.

There were insignificant changes in all of these mechanical properties with an increasing level of irradiation.

We conclude that the overall mechanical effect of high levels of extracorporeal irradiation (300 Gy) on bone is negligible. Consequently the dose can be maximised to reduce the risk of local tumour recurrence.

Cite this article: Bone Joint J 2015;97-B:1152–6.

Extracorporeal irradiation and re-implantation of a bone segment is a technique employed in bone sarcoma surgery for limb salvage in the setting of reasonable bone stock. There is neither consensus nor rationale given for the dosage of irradiation used in previous studies, with values of up to 300Gy applied. We investigated the influence of extracorporeal irradiation on the elastic and viscoelastic properties of bone. Bone specimens were extracted from mature cattle and subdivided into thirteen groups; twelve groups exposed to increasing levels of irradiation and a control group. The specimens, once irradiated, underwent mechanical testing in saline at 37°C.

Mechanical properties were calculated by experimental means which included Young's Modulus, Poisson's Ratio, Dissipation Factor, Storage Modulus, Loss Modulus and Dynamic Modulus. These were all obtained for comparison of the irradiated specimens to the control group.

We found there to be a statistically significant increase in Poisson's ratio after increasing irradiation doses up to 300Gy were applied. However, there was negligible change in all other mechanical properties of bone that were assessed. Therefore, we conclude that the overall mechanical effect of high levels of extracorporeal irradiation (300Gy) is minute, and can be administered to reduce the risk of malignancy recurrence.

Unicompartmental knee replacement (UKR) is associated with higher revision rates than total knee replacement and it has been suggested that surgeons should receive specific training for this prosthesis. We investigated the outcome of all UKR in a district general hospital over ten years.

All patients who had received UKR from 2003 to 2013 were identified from theatre records, as were all revision knee arthroplasties. We contacted all patients (or their GP) with no known revision to ascertain UKR status. A life table analysis was used for three categories: all surgeons and types of UKR, Oxford UKR only and Oxford UKR by surgeons with specific training.

There were 319 UKR (one loss to follow up), four types of prosthesis, 21 failures and a 5-year cumulative survival rate of 91.54%. There were 310 Oxford UKR with 17 failures and 5-year survival 93.56%. Surgeons with training in use of Oxford UKR completed 242 replacements with 10 failures and 5-year survival of 95.68%.

In comparison with results for Oxford UKR in 11th annual NJR report, our results are satisfactory and support continued use of this prosthesis in a non-specialist centre. Our established programme of surveillance will monitor the survival of UKR in our hospital.

Extracorporeal irradiation and re-implantation of a bone segment is a technique employed in bone sarcoma surgery for limb salvage in the setting of reasonable bone stock. There is neither consensus nor rationale given for the dosage of irradiation used in previous studies, with values of up to 300Gy applied. We investigated the influence of extracorporeal irradiation on the elastic and viscoelastic properties of bone. Bone specimens were extracted from mature cattle and subdivided into thirteen groups; twelve groups exposed to increasing levels of irradiation and a control group. The specimens, once irradiated, underwent mechanical testing in saline at 37°C.

Mechanical properties were calculated by experimental means which included Young's Modulus, Storage Modulus and Loss Modulus. These were all obtained for comparison of the irradiated specimens to the control group.

There were non-significant negligible changes in all of the mechanical properties of bone that were assessed with increasing dosage of irradiation. Therefore, we conclude that the overall mechanical effect of high levels of extracorporeal irradiation (300Gy) is minute, and can be administered to reduce the risk of malignancy recurrence.

First-time anterior dislocation of the shoulder is associated with the development of recurrent instability. It is recognised that patients with recurrent instability often have osseous defects. Using 3D computerised tomography (3DCT) it is possible to quantify these defects. Whether these defects are present after the primary dislocation or occur progressively from multiple dislocations is unclear. We correlated the presence of Hill-Sachs lesions and anterior glenoid bone loss with evidence of recurrent dislocation and clinical outcomes.

78 patients were followed up for two years. All underwent a 3DCT within a week of injury. Standardised images of the humeral head and glenoid were produced. Using standardised digital techniques bone loss was measured.

39% of the patients developed further instability. Average Hill-Sachs circumferential length = 15.23%. Average Hill-Sachs surface area = 5.53%. The length and surface area of the Hill-Sachs lesions were significantly associated with further instability. (p=0.019 and p=0.003). Average en face glenoid surface area loss=1.30% with no association to instability (p=0.685). There was poor correlation between the size of the glenoid lesion and the size of the Hill-Sachs lesion.

Results showed that age and increasing size of the Hill-Sachs lesions result in a higher rate of instability. Interestingly glenoid bone loss was relatively low and did not predict recurrent instability. The size of the Hill-Sachs lesion does not have a linear relationship with glenoid bone loss. Further work defining the morphology of the Hill-Sachs lesion and its engagement with a glenoid defect is required.

Several human conditions have a tendency to affect one side of the body over the other. Do lumbar disc prolapses have such a tendency? We sought to answer this question by an analysis of operated cases.

Primary lumbar microdiscectomy cases were identified using the coding system. 1286 cases were identified and in 764 the laterality was not recorded. Electronic records were then examined to establish, where possible, the side of the procedure from the clinic letter or discharge summary. 22 cases were eliminated due to miscoding (laminectomy, instrumentation, revision) and in 24 the side of the operation could not be established. In the remaining 1240 cases (96.4%) the laterality was determined.

Patients who underwent primary lumbar microdiscectomy in a single neurosurgical unit over a 5-year period (2002-2007).

Procedures were bilateral, left or right.

126 cases were bilateral. There were 1114 defined lateral cases. 618 (55.5%) were on the left compared to 496 (44.5%) on the right. The ratio of left to right is close to 5:4. The null hypothesis was that the number of left and right sided operations would be equal. The findings of this study were statistically highly significant (p value < 0.001, binomial test) and the null hypothesis could be rejected.

There is a small but definite preponderance of left sided over right sided cases at a ratio of 5:4. This finding may have implications regarding our understanding of both the epidemiology and biomechanics of lumbar disc prolapse.

A Morton's neuroma is a painful forefoot disorder characterised by plantar pain and toe paraesthesia. Many treatments have been tried including modification of shoe wear, injections and surgery. The results from different treatment modalities are extremely variable in the scientific literature.

We reviewed 79 patients who presented with an ultrasound proven Morton's Neuromata at an average of 60 months following treatment in a patient blind randomised trial.

35 had surgery with resection of the Morton's Neuromata. We compared the results of the patients who had had a steroid injection and surgery, a steroid injection alone, a placebo injection and then surgery and a placebo injection alone.

We looked at the requirement for further consultations, pain scores, activity restrictions, footwear restriction, overall satisfaction with their treatment and overall quality of life scores. We found no significant difference between the groups. Overall 80% of the surgical patients had a good or excellent result and 67% of the non- operatively managed patients had a good or excellent result. There was no difference in their quality of life scores.

Previous studies have shown little benefit in steroid injections in the treatment of Morton's Neuromata. We conclude at the five year mark there is no significant differences between operating on Morton's Neuromata and treating them conservatively.

Further research should focus on comparing operative vs conservative treatment in the short to medium term. In order to justify the risks and costs of surgery.

Aim: To investigate factors that influence outcomes one year after distal radial fractures To investigate how deformity (radiologically), functional outcome and patient satisfaction affect one another.

Background: Identifying the factors that influence outcome in DRF is important in anticipating and treating patients with potentially correctable factors that may affect recovery. Previous studies have looked at different sub-sets of the DRF group most often with patient reported outcomes. We have reviewed a large consecutive group of DRF looking at which factors influenced the outcomes.

Methods: Data on 640 distal radial fractures was prospectively recorded over a 24 month period. The database was reviewed and validated. Mechanism of injury, hand dominance and occupation were noted. Initial, post reduction, one week, 6 week and one year x-rays were taken. The volar and dorsal shortening, tilt and angulation were recorded. Any operative intervention or complication was noted.

At one year follow up functional testing was performed including range of movement. This tested for grip strength, multiple postional strengths and a functional score looking at activities of daily living.

Results: Prediction of functional outcome was significantly associated with age, volar communition, dorsal angulation and pain. The grip strength after a distal radial fracture is significantly stronger in dominant side fractures compared with non dominant, in younger patients and those without dorsal communition.

We also showed that fractures that are most likely to malunite show a significantly poorer functional outcome and weaker grip strength.

Conclusion: This study identifies factors that predict the functional outcome in Distal Radial fractures. Although many assumptions are made that certain fractures lead to poorer results this has rarely been shown in such a large, diverse group of DRFs.

In an age where patients and practitioners strive to ever increasing levels of knowledge this study allows us to counsel patients in their likely functional outcomes more accurately.

Despite increasing scientific investigation, the best method for preventing post-operative deep vein thrombosis remains unclear in patients undergoing a total hip replacement. In the wake of the recent controversial National Institute for Health and Clinical Excellence Clinical Guidelines on the prevention of thrombo-embolism, we felt it was timely to survey current Scottish Surgeons thrombo-prophylactic practice when performing total hip replacements.

E-mail questionnaires were sent to all Scottish Orthopaedic consultants. They were asked about routine pharmacological and mechanical prophylaxis in patients undergoing total hip replacement. Comparison was made with a previous survey done in 2003.

The response rate was 75%. The survey showed an increased use of pharmacological prophylaxis from 93% to 100%. This was due to the increased use of aspirin, from 51% to 64%. The use of Low molecular weight heparin had remained the same at 51%. No surgeons routinely use Warfarin, low dose heparin or Fondaparinux. Use of graded compression stockings has increased from 59% to 70%.

There is increasing evidence that patients undergoing total hip replacement should receive extended prophylaxis for up to 35 days. This could explain why aspirin is commonly used. Most guidelines, however, do not recommend this drug. With the recent launch of two new oral agents we may see a huge change in prophylaxis in Scotland, as 94% of surgeons said they would use extended prophylaxis if a safe oral agent was found.

Aim: Controversy remains regarding the complications/necessity for removal of metalwork used in the fixation of paediatric orthopaedic surgery owing to potential complications of removal. It was the aim of this study to review all cases of metalwork removal performed in a prescribed year in order to ascertain the reasons for removal and complications of these surgeries.

Methods: All cases of metalwork removal performed under GA, from 1st January 2006 until 31st December 2006 in a regional paediatric orthopaedic unit, were reviewed. Data were obtained from case notes, computer and theatre records and radiographs. Details were obtained regarding demographic details, anatomic site, implant used, reasons for removal, problems of initial fixation and complications after removal.

Results: 34 buried kirschner wires were removed under general anaesthetic. Of other metal work, 38 implants were removed; 22 in males and 16 in females. The commonest age distribution was the 11–15 age groups. The most common anatomic site for removal was the forearm (42%) followed by the femur (18%) and ankle (18%). The most common implant removed was plate (53%), followed by Nancy nails (18%). Problems following initial insertion were found in 4 patients (10.5%) including re-fracture distal to forearm plates (both in patients with co-existing osteopaenia), bowing of forearm and decreased sensation over the thumb. Metal work was removed for symptoms in 13 cases (34%) and the rest for patient request or clinical indications. Complications after metalwork removal occurred in 8 (21%) patients and included re-fracture of forearm (2), hypertrophic scar, abscess, skin reaction, wound breakdown, excessive bruising and discomfort.

Conclusion: As routine removal of metalwork is generally considered not necessary, metalwork removal is often carried out for symptoms, at patient request as well as in patients with osteopaenia who have risk of fracture at the ends of metalwork. Parents need to be made aware of possible complications shown in the study, in order to give informed consent.

Background: Several recent studies have compared incorporation of autograft with that of allograft or synthetic bone substitutes in anterior cervical discectomy and fusion (ACDF). These studies have almost universally relied on plain radiography to assess bone incorporation despite the fact that we know, from similar lumbar spine studies, that bone ingrowth is over-estimated.

Our aim was to determine the exactly whether bone incorporation may be correctly assessed by this method by comparing the results to those obtained by spiral CT imaging.

Methods: 15 patients underwent ACDF. Helical CT scans were obtained. Fusion was defined as trabecular continuity across the disc space anterior, through and posterior to the cage proximally and distally and assessed by two of the authors independently.

Results: 14 of the 15 patients appeared to have solid incorporation of bone graft/substitute on plain radiography, 19 out of 20 cages. These findings were not however replicated on CT imaging. The autograft was not considered to have been incorporated proximally above the cage in 5 cases and distally in 6 cases.

Discussion: The implication of our results is that there is at least a false positive rate of bone incorporation of 20–25%. Pseudarthroses are generally painful and therefore we would recommend that spiral CT imaging is performed in patients who have ongoing pain following ACDF.

Ethics approval: COREC Ethics committee number 06/S1104/34

Interest Statement: None of the authors have received any grants to carry out this research.

Aim: To study the effect of smoking on the rate of complications and functional outcome after Total Hip Replacement (THR).

Methods: Data was prospectively collected for all patients undergoing a unilateral, primary THR in one hospital during the period 1998 to 2006 Data was collected at pre-operative assessment, in the peri-operative period and at 6 months, 18 months, 3 years and 5 years. Patients were placed in one of three groups; smokers, ex-smokers and those who had never smoked.

Outcome measures included rate of complications: infection, DVT, PE, length of hospital stay and Harris Hip Score (HHS).

Statistical analysis was undertaken to determine any correlation between smoking and these outcome measures using chi-squared tests, t-tests and multiple regression adjusting for confounding factors.

Results: 1765 patients underwent THR during the study period, of whom 635 were males and 1130 females, with a mean age at operation of 69.

268 patients (15%) were smokers, 582 patients (33%) were ex-smokers and 917 patients (52%) had never smoked. As there was little data available on when the ex-smokers had stopped smoking we studied current smokers compared to patients that had never smoked

There were no significant differences in complications such as DVT, PE, Deep infection and Superficial infection, these were all rare events. Neither was there any significant difference in hospital stay times. After adjusting for pre-operative HHS, age, sex and ASA status current smokers had significantly lower HHS at 6 months (p< 0.001, 95% confidence interval for effect size 1.6 to 5.3), and also showed a lower HHS at 18 months, 3 years and 5 years although not significantly so.

Conclusion: Current smokers have a significantly lower HHS post-operatively in comparison to non-smokers. In a health system where more and more patients wish to know the risk associated with potential operative interventions, we can now say that patients who smoke will have a poorer outcome post Total Hip Replacement. This may also allow us to discourage some patients wishing to have THR whom are poor operative candidates.

We plan to further analysis the data to try and ascertain why this is the case.

Introduction: The correction of anaemia prior to total hip arthroplasty reduces surgical risk, hospital stay and cost. This study considers the benefits of implementing a protocol of identifying and treating pre-operative anaemia whilst the patient is on the waiting list for surgery.

Methods: From a prospective series of 301 patients undergoing elective total hip arthroplasty (THA), patients identified as anaemic (Hb< 12 g/dl) when initially placed upon the waiting list were appropriately investigated and treated. Pre- and post- operative haemoglobin (Hb), need for transfusion, and length of hospital stay were collated for the entire patient cohort.

Result: 7.6% of patients were anaemic when initially placed upon the waiting list for THA and had a higher transfusion rate (25% to 4%, p< 0.05) and longer hospital stay (7.5 to 6.6 days, p< 0.05). Over 40% of these patients responded to investigation and treatment whilst on the waiting list, showing a significant improvement in Hb level (10.1 to 12.7 g/dl) and improved transfusion rate.

Discussion: Quantifying the haemoglobin level of patients when initially placed on the waiting list helps highlight those at risk of requiring a post-operative blood transfusion. Further, the early identification of anaemia allows for the utilization of the waiting list time to investigate and treat these patients. For patients who respond to treatment there is a significant reduction in the need for blood transfusion with its inherent hazards.

Introduction: Hypermobility is a common finding, however, it lacks diagnostic parameters and is poorly understood, especially in the foot.

Aim: To quantify medial column/first ray mobility in patients with midfoot arthritis and planovalgus feet.

Methods: We compared first ray mobility in patients with radiologically defined midfoot tarsometatarsal osteoarthritis, a radiologically normal first ray and planovalgus feet, with control subjects who had normal feet and first rays. An all female group of 20 patients (mean age of 70) and 20 controls (mean age of 53) met the criteria. Analysis of patients’ x-rays identified the site of their arthritis and allowed angular measurements of their flat foot deformity. Patient and control subjects underwent identical examinations, recording hindfoot correctability, medial longitudinal arch appearance, hindfoot prontion and supination, forefoot supination and degrees of flexion/extension and abduction/adduction with an electronic goniometer. Each subject was graded by the AOFAS and SF-36 outcome scores.

Results: There was a significant difference in first ray mobility between the patient and control subjects for all positions adopted (P=< 0.001), except when dorsiflexed and weight bearing (P=0.052). Patients with a neutral non-weight bearing ankle exhibited greatest mobility of 16.8 +/− 4.7 degrees compared to 9.4 +/− 2.6 degrees in controls. This was a significant difference, P=< 0.001, as was the difference between patients adopting the NWB plantarflexed, dorsiflexed and WB neutral positions. P=0.002, P=0.014, P=0.001 respectively. Patients’ median score for 5 out of 8 SF36 domains were considerably less than controls, as were patients’ AOFAS. Reduced physical and social functioning were shown to be linked to poor foot scores.

Conclusion: Patients with planovalgus feet and tarsometatarsal OA have greater first ray mobility than controls with normal feet. Recognising this may help plan orthotic or surgical treatment.

Introduction: The correction of anaemia prior to total hip arthroplasty reduces surgical risk, hospital stay and cost. This study considers the benefits of implementing a protocol of identifying and treating pre-operative anaemia whilst the patient is on the waiting list for surgery.

Methods: From a prospective series of 322 patients undergoing elective total hip arthroplasty (THA), patients identified as anaemic (Hb< 12 g/dl) when initially placed upon the waiting list were appropriately investigated and treated. Pre- and post-operative haemoglobin (Hb), need for transfusion, and length of hospital stay were collated for the entire patient cohort.

Result: 8.8% of patients were anaemic when initially placed upon the waiting list for THA and had a higher transfusion rate (23% to 3%, p< 0.05) and longer hospital stay (7.5 to 6.6 days, p< 0.05). Over 40% of these patients responded to investigation and treatment whilst on the waiting list, showing a significant improvement in Hb level (10.1 to 12.7 g /dl) and improved transfusion rate.

Discussion: Quantifying the haemoglobin level of patients when initially placed on the waiting list helps highlight those at risk of requiring a post-operative blood transfusion. Further, the early identification of anaemia allows for the utilization of the waiting list time to investigate and treat these patients. For patients who respond to treatment there is a significant reduction in the need for blood transfusion with its inherent hazards.

Bone-marrow transplantation has increased the survival of patients with mucopolysaccharidosis-I. We describe the spinal problems and their management in 12 patients with this disorder who have been followed up for a mean of 4.5 years since transplantation.

High lumbar kyphosis was seen in ten patients which was associated with thoracic scoliosis in one. Isolated thoracic scoliosis was seen in another. One patient did not have any significant problems in the thoracic or lumbar spine but had odontoid hypoplasia, which was also seen in three other children. Four of the eight patients in whom MRI of the cervical spine had been performed had abnormal soft tissue around the tip of the odontoid.

Neurological problems were seen in two patients. In one it was caused by cord compression in the lower dorsal spine 9.5 years after posterior spinal fusion for progressive kyphosis, and in the other by angular kyphosis with thecal indentation in the high thoracic spine associated with symptoms of spinal claudication.