Vibration as a Method of Fracture Screening in Children

When assessing an injured child, doctors must decide whether or not there is an underlying bone fracture. The best way of doing this is to take an x-ray. In 2011, the 46,000 children attending Sheffield Children's Hospital Emergency Department had 10,400 x-rays mainly to help diagnose fractures. Taking just the foot and wrist, 2,215 x-rays were normal with no fracture, at a cost of £119,610 for the Sheffield community alone (at tariff £54 per x-ray). This works out as a cost of approximately £12 million per year across England and Wales. Additionally, although the radiation dose is quite small, given that x-rays can cause cancer, no radiation is better than some radiation. A fracture screening method is needed that will help doctors, schoolteachers and others more reliably decide which children should have an x-ray. Vibration is reliably used in industry to find defects such as cracks in machines and other structures. The researchers believe that vibration can similarly find fractures in bones in children. The team has recently demonstrated the ability of vibration to correctly pick the 3 x-ray confirmed cases out of 13 adults who had a wrist fracture (7 healthy adults and 6 with wrist injury). None of the 6 injured adults felt that vibration would be too painful to use on injured children. The proposal is now to compare the vibration patterns of the bones of about 150 children over 10 years of age attending the researchers' Emergency Department with their fracture positive or fracture negative x-rays. The researchers also propose to assess any differences in the vibration patterns between left and right wrist and ankles in 50 healthy school children with no injury. Should vibration analysis for fracture screening prove sufficiently accurate, further larger studies shall be conducted, with the aim of developing an instrument that will reduce the number of injured children having unnecessary x-rays. On completion of this study, the plan is to extend the study to include younger children. This will lead to cost savings for the NHS and less inconvenience for patients and their families, with shorter stays in Emergency Departments and reduced population exposure to harmful ionising radiation.

The main research question underpinning this pilot study is, "Can vibration spectra reliably distinguish fractured from non-fractured bones in children?" The Problem: A significant number of unnecessary radiographs are performed to exclude the presence of fractures. In 2011, 46,000 children attended Sheffield Children's Hospital Emergency Department (ED), for whom 10,400 radiographs were performed. 2,215 foot/ankle and wrist radiographs alone proved unnecessary. This translates to a potential cost saving of £119,610/year in Sheffield and approximately £12M/year in England and Wales. Clearly there is the need for a non-invasive, portable, cheap and rapid method of screening long bones for fractures to establish the probability of whether the bone is fracture negative or positive; those with a high probability of fracture then going on to be radiographed, to establish presence, type and position of fracture. A Possible Solution: Vibration analysis involves the mechanical excitation of an object, then recording and analysing the subsequent responses . When a structure is allowed to move freely after an excitation, it will vibrate at its specific natural frequencies where the inertial forces are in balance with the elastic forces, depending upon the stiffness of that structure. Vibration spectral analysis is reliably used in industry to detect cracks in machinery. The researchers tested the hypothesis that fractured and non-fractured bones return different vibration spectra on 13 adults; 7 healthy volunteers from the Engineering Department of Sheffield Hallam University and 6 patients following injury at Northern General Hospital, 3 of whom had fractures , with 100% correlation of vibration spectra with presence/absence of fracture and no pain. Given these encouraging results, the proposal is to extend the study to children. Research Theory: The researchers postulate that there are detectable differences in the natural vibration frequencies obtained from bones with and without fracture. Each child's uninjured contralateral bone will serve as a comparison for the injured side. The application of vibration analysis involves mechanical excitation of an object and recording and analysing the subsequent responses. In theory, by applying a stimulus to the bone and analysing the resulting response, it is possible to monitor various pathological and trauma-induced conditions. Vibrations may be induced either by an impulse, which yields a response consistent with the inputting of a number of independent fundamental frequencies, or by variable frequency cycles of pure sine waves. Having induced the vibrations by a small purpose-built computer-controlled tapper, the commonest method of recording them uses piezoelectric gauges on bony prominences (minimising the effect of underlying soft tissue). For example, when analysing tibial fracture healing, the lateral or medial malleolus are prime sites . Successful operation of piezoelectric gauges depends on the coupling of the gauge to the structure under investigation, such that the induced stresses produce a net change in charge of the transducer terminals due to the piezoelectric effect. Piezoelectric crystals are relatively inexpensive and their properties well understood. Both piezoresistive and piezoelectric materials are commonly used to detect strain caused by vibrations in macro-scale structures. However, they often require challenging signal processing and interpretation. The research group has experience working with complex medical signals (e.g. electroencephalograms) to extract relevant information and will bring this experience to bear in this project. Previous feasibility studies provide a sound basis for the proposed study. There is extensive information on theory, techniques for vibration analysis of physical structures are well established and furthermore, a number of studies have explored low-frequency vibration analysis as a possible diagnostic tool in the orthopaedic field. For example, in one study it was reported that the vibration transmission across a fracture is affected by the stages of healing of the fracture callus. As far as the researchers are aware, no study has used the precise methodology proposed in this current study to screen for fractures in children.

Child with trauma needing wrist or ankle x-ray: Will receive vibration analysis blinded to x-ray findings.

Number of patients correctly stratified by vibration analysis into fracture and non-fracture groups (reference standard = wrist/ankle radiographs)

Patients: We shall recruit children attending Sheffield Children's Hospital Emergency Department and requiring imaging (usually AP and lateral radiographs) of ankle or wrist for suspected fracture. Exclusion criteria: Age 9 years and 364 days or less. Age 16 years or more Injury sustained more than 24 hours before attendance Informed consent withheld Distressed child or family (most children will receive pain relief in the Emergency Department prior to having their radiographs and vibration analysis) Both arms or both legs thought to be fractured (the non-fractured contralateral limb will act as control for each child) Inability to access bony prominence Presence of metal work (e.g. internal or external fixators), plaster of Paris or other cast Previously recorded fracture or orthopaedic surgery (this might affect vibration spectra). Healthy Controls We have previously obtained a letter of support from the Deputy Head of King Edward's VII Language College, Sheffield, to recruit 20 healthy children and perform vibration analysis at a convenient time (11.00am to 1.00pm on Saturdays). We have so far recruited 15 healthy children and will seek permission from the Head/Deputy Head to extend the original 20 to 50. Exclusion criteria: Age 9 years and 364 days or less. Age 16 years or more Previous fracture to any wrist or ankle Previous surgery to any wrist or ankle Known chronic disease (any long-term disease on-going for over one year) Informed consent withheld Inability to access bony prominence Presence of metal work (e.g. internal or external fixators), plaster of Paris or other cast, limb prosthesis