RPS 1012a - Paediatric musculoskeletal imaging: what's new?

Purpose:

WBMRI enables the depiction and characterisation of diseases at an early stage and is embraced for use in children by many centres, although no studies addressing its precision, accuracy, and clinical validity are published. The signal in the bone marrow that may be interpreted as pathology is shown to be present in healthy children. We set out to examine healthy, asymptomatic children with WBMRI to establish normal reference standards for the skeleton in children.

Methods and materials:

From 2018-2019, 160 volunteers 6-18 years were recruited for research-purposes only. Coronal T2dixon, T1tse, and DWI (b50 and b1000) from skull base to toes, on state of the art 1.5T scanners (Siemens and Philips), whole body-coil, no sedation, were performed. The scantime was 40-45 minutes. Scoring was performed by two radiologists in consensus, assessing the presence of signal changes suggestive of bone marrow oedema.

Results:

All examinations were successful and high-quality images were obtained. Preliminary results of 80 children showed several expected and surprising signal changes that could be mistaken for pathology in a clinical setting.

Conclusion:

Signal change is frequently seen on WBMRI in the skeleton of healthy children. This knowledge is crucial to avoid the overstaging of disease, particularly when assessing for lesions at a preclinical stage.

Limitations:

Norway is a relatively homogeneous society regarding social status, physical activity, and body-habitus. The definition of “healthy” is challenging. The majority of children report minor intermittent musculoskeletal symptoms. We defined “asymptomatic” as no symptoms impairing everyday activity or necessitating consulting a doctor in the last 6 months. No histological correlation could be made.

Ethics committee approval

Approved by the Regional Committees for Medical Research Ethics, South East Norway, informed consent obtained from all participants/carers.

Funding:

National grant (Helse Sør-Øst) for a three year PhD-program.

Purpose:

To assess bone manifestations in patients with Gaucher disease type 1 (GD1) with whole-body magnetic resonance imaging (WB-MRI) and to evaluate the effects of different timing in initiating long-term enzyme replacement therapy.

Methods and materials:

In 17 long-term treated (enzyme replacement therapy with alglucerase/imiglucerase for a median of 13 years) patients with GD1, we performed 2 WB-MRI examinations at a median interval of 13 months. MRI results were retrospectively stratified based on treatment initiation into two groups: “early” (age ≤ 12 years) and “late” (during adulthood). We assessed the occurrence of irreversible avascular necroses (AVN) and applied several semi-quantitative scores: the bone-marrow-burden (BMB) score, the Düsseldorf-Gaucher score (DGS), the vertebra-disc-ratio (VDR), and the Gaucher disease type 1 severity scoring system (GD-DS3).

Results:

MRI assessments showed no AVN in the “early” treatment group. AVN were observed in 2 patients of the “late” group. The follow-up examinations showed slight improvements in the BMB-score (P=0.039), DGS (P=0.011), and VDR (P=0.047), with similar tendencies in both treatment groups. The GD-DS3 score only improved in the “late” treatment group (P=0.043).

Conclusion:

The ongoing clinical value of enzyme replacement therapy with alglucerase/imiglucerase is supported by our data as WB-MRI-based scores stayed constant or slightly improved even after long-term treatment. Complications (i.e. AVN) were only observed in the "late" treatment group. The early initiation of enzyme replacement therapy may protect the bone.

Limitations:

The relatively small study cohort and retrospective study design. The interval between MR examinations was rather short. However, we provided data from patients with very long treatment courses.

Ethics committee approval

Approval of the local independent ethics committee was not necessary due to the retrospective study design.

Funding:

This study was financially supported by Genzyme GmbH.

Purpose:

Radiological investigation should be performed where physical abuse of a child is suspected. Appropriate timely imaging is vital in view of the social and legal implications of the accompanying radiological reports. The aim of this work is to audit practice in a UK district general hospital and evaluate its adherence to national standards. These standards are set by the joint Royal College of Radiologists (RCR) and Royal College of Paediatrics and Child Health (RCPCH) publication: ‘The radiological investigation of suspected physical abuse in children’.

Methods and materials:

Patients who received an x-ray skeletal survey for suspected non-accidental injury from the 1st February 2017-26th March 2019 (n=27) were identified using the software Sunquest Integrated Clinical Environment (ICE). The mean age was 10 months, with a range of 11 days to 3 years and 5 months. The radiological reports and notes of these patients were analysed and used to evaluate adherence to standards.

Results:

Good practice identified in this audit was that skeletal surveys were double reported 92.6% of the time. Areas requiring improvement included the speed of double reporting, attendance for the follow-up skeletal survey, and performance of CT head in the under 1-year-old group, with a follow-up MRI head and spine where indicated.

Conclusion:

Improving adherence to standards requires co-operation between radiological and paediatric departments. Additional qualified radiologists can improve reporting speeds, as can appropriate outsourcing to local hospitals. Where CT or MR scans are decided against for clinical reasons, this should be clearly documented in the notes, and the integrity of ‘the net’ should be improved to avoid a loss to follow-up in children who are potentially at risk.

Limitations:

n/a

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

To evaluate the diagnostic accuracy of morphometric vertebral fracture analysis (MXA) using a 33-point software program designed for adults on dual-energy x-ray absorptiometry (DXA) images of children.

Methods and materials:

Lateral spine DXA images of 420 children aged 5-18 years were retrospectively reviewed. VFA by an expert paediatric radiologist using Genant's SQ scoring system served as the gold standard. All 420 DXA scans were analysed by a trained radiographer, using semi-automated software (33-point morphometry). VFA of a random sample of 100 DXA was performed by an experienced paediatric clinical scientist. MXA of a random sample of 30 DXA images were analysed by 4 additional observers. Diagnostic accuracy and inter and intraobserver agreement (kappa statistics) were calculated.

Results:

Overall sensitivity, specificity, false-positive (FP) and false-negative (FN) rates for the radiographer using the MXA software were 80%, 90%, 10%, and 20%, respectively, and for mild fractures alone were 46%, 92%, 8%, and 54%, respectively. Overall sensitivity, specificity, FP, and FN rates for the 4 additional observers using MXA were 89%, 79%, 21%, and 11%, respectively, and for mild fractures alone were 36%, 86%, 14%, and 64%, respectively. The agreement between the two expert observers was fair to good for VFA and MXA [kappa=0.29 to 0.76 (95% CI: 0.17-0.88) and 0.29 to 0.69 (95% CI: 0.17-0.83)], respectively.

Conclusion:

MXA using a 33-point technique developed for adults is not a reliable method for the identification of mild vertebral fractures in children. A paediatric standard is required which not only incorporates specific vertebral body height ratios but also the age-related physiological changes in vertebral shape that occur throughout childhood.

Limitations:

n/a

Ethics committee approval

n/a

Funding:

This study funded as part of PhD project by Najran University.

Purpose:

Atlantoaxial rotary subluxation/fixation (AARS/F) is a rare cause of torticollis in children. Patients usually present with their head fixed in rotation with some degrees of flexion (‘cock-robin’ head position). CT permits the diagnosing and grading of the disease.

The purpose of our study was to evaluate the correlations between radiological and clinical factors and patients' prognosis.

Methods and materials:

We retrospectively reviewed all CT studies and all clinical reports from our hospital archive of children diagnosed with AARS/F in the last 15 years.

Two expert musculoskeletal radiologists reviewed all CT studies, grading AARS/F according to the Fielding-Hawkins system. Moreover, C1-C2 rotation-degrees were calculated.

The outcome assessments were positive (complete clinical/radiological recovery at three months follow-up), and negative (disease persistence or relapse at three months follow-up).

All patients were treated conservatively after diagnosis (cervical traction/neck collar).

Results:

We included 53 patients (mean age=8.7y – 26F, 27M) diagnosed with AARS/F in the study. In 7/53 subjects (13.2%), a negative clinical outcome was observed.

The presence of concomitant infection or inflammation at the head and neck region (Griesel’s syndrome) was significantly associated with a negative outcome (p=0.001). Fielding-Hawkins grading was also well correlated with the patients' outcome (p=0.01). C1-C2 rotation-degrees was rather associated with the outcome, without reaching statistical significance (p=0.056). The presence/absence of a traumatic event that triggered AARS/F was not correlated with the patients’ outcome (p=0.171).

Conclusion:

The presence of concomitant infection/inflammation at head and neck regions in patients with AARS/F is the most important prognostic factor and is associated with poorer outcomes. The Fielding-Hawkins radiological grading system well correlates with the patients’ outcome.

Limitations:

A retrospective study.

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

Post-reduction perfusion MRI has been previously introduced to assess femoral head perfusion but its prognostic value is unclear. We sought to determine whether globally-decreased femoral head perfusion predicts femoral growth disturbance at a minimum of 5 years after closed/open reduction for DDH.

Methods and materials:

A retrospective analysis of patients undergoing closed/open reduction for DDH and post-reduction perfusion MRI within 24 hours was conducted. 94 hips (83 patients, mean 11±4 year follow-up) had a minimum 5-year radiographic follow-up. Of these, 45/49 hips had open/closed reduction. The Kalamchi/MacEwen classification of the proximal femoral deformity was used (grade 4; total damage to femoral epiphysis, -physis, -neck) at latest the radiographic follow-up. 1.5T MRI was performed including 2D coronal-, axial PD-w, and T1-w TSE images, before/after i.v. contrast injection (0.2 mmol/l Gd-DTPA2-). Femoral head enhancement was graded as normal/asymmetric/focally/globally decreased. Multivariate regression analysis was performed and the diagnostic performance calculated.

Results:

At follow-up, 11 hips (12%) had femoral growth disturbance. The prevalence of globally-decreased perfusion was higher (p=0.016) in hips with (36%, 4/11) compared to hips without growth disturbance (10%; 8/75) and was independently associated with femoral growth disturbance (OR: 6.2; 95%CI, 2 to 27; p=0.014). While the negative predictive value of globally-reduced perfusion was high (92%), its positive predictive value was low (36%) in predicting femoral growth disturbance.

Conclusion:

Globally-decreased perfusion on post-reduction MRI does not reliably predict proximal femoral growth disturbance in DDH at mid-term. The aetiology of femoral growth disturbance after treatment of DDH is likely a multifactorial problem not solely attributed to decreased femoral head perfusion after closed/open reduction.

Limitations:

A singe-centre study at a tertiary centre.

Ethics committee approval

IRB approaval was granted under a waiver for informed consent.

Funding:

No funding was received for this work.

Purpose:

To investigate which risk factors are predictive of developmental dysplasia of the hip (DDH) or hip instability (HI) in newborns using hip ultrasound (HUS)-based diagnosis.

Methods and materials:

We retrospectively analysed the clinical data of 1,017 newborns who underwent HUS from August 2016-December 2018. Included patients showed dependent risk factors (familiarity, macrosomia, twins, and podalic presentation), independent risk factors (doubt or positive Ortolani manoeuvre, hip click, skin folds asymmetry, and others), or coexistence of two or more risk factors (dependent or independent) for DDH. Each hip was analysed by one of two operators with 3- and 7-years’ experience, respectively, using a high-frequency linear probe and the graft technique. Patients were divided into three groups according to HUS-related categorisation (1a-1b versus 2a-2b versus >2c).

We used the χ2 test and built two different logistic regression models to asses which risk factors were predictive of HUS categorisation 2a-2b (HI) or >2c group (DDH).

Results:

HUS categorised patients as 1a-1b in 872/1017 (85.7%), HI in 126/1017 (12.4%), and DDH in 19/1017 (1.9%), respectively. On multivariate analysis, HI was independently predicted by positive Ortolani manoeuvre (OR=10.16; p<0.0001) and the coexistence of two risk factors (OR= 2.64; p=0.014), which occurred in 15/126 (11.9%) and 10/126 (7.9%) cases, respectively. DDH was independently predicted by positive Ortolani manoeuvre (OR=74.64; p<0.0001), coexistence of > 3 risk factors (OR=186.6; p<0.0001), and skin folds asymmetry (OR=10.36; p=0.037), which occurred in 8/19 (42.1%), 2/19 (10.5%), and 1/19 (5.3%) cases, respectively.

Conclusion:

HI and DDH are predicted by a limited number of risk factors, suggesting that indication of HUS should be restricted accordingly.

Limitations:

A monocentric and retrospective design.

Ethics committee approval

Informed consent waived because of the retrospective design.

Funding:

No funding was received for this work.

Purpose:

To define normal age-related triangular fibrocartilage complex (TFCC) morphology, homogeneity, and signal intensity on high-resolution 3.0 Tesla (3T) magnetic resonance imaging (MRI) in children.

Methods and materials:

This cross-sectional study included 24 healthy children, aged 12-18 years, that did not participate in wrist-loading sports more than two times a week. They underwent a 3T MRI of one wrist, including proton-density (PD)-weighted images in three planes without fat saturation and PD-weighted coronal and T2-weighted axial images with fat saturation. Morphology, homogeneity, and signal intensity of TFCC components were assessed by 3 radiologists and 1 hand surgeon independently using a standardised score form that was developed by 1 physician experienced in TFCC-research and the 4 observers with several consensus meetings and test rounds.

Results:

The cohort consisted of 12 girls and 12 boys with a mean age of 13.5 (SD, 1.1). Observers were consistent in defining anatomical variants and measurements. The observers reported anatomical variations in a substantial number of children. For instance, linear vertical increased the signal with disruption of the articular disc and diffuse increased the signal of the disc not extending to the joint surface. Also, general TFCC disc morphology, proximal and distal lamina, radio-ulnar ligaments, prestyloid recess, and meniscus homologue showed anatomical variants.

Conclusion:

This study shows normal TFCC variation on high-resolution 3T MRI in healthy children, such as focal articular disc perforations. When interpreting MRI of the wrist in children, it is important to be aware of the TFCC anatomy and its normal variations.

Limitations:

The sample size. Observers were aware that children were healthy and asymptomatic.

Ethics committee approval

Approved by the local medical ethical committee (reference number 2014_382#B2015303).

Funding:

No funding was received for this work.