Dual-energy X-ray absorptiometry (DXA) forms an important component of fracture risk assessment. The objective of the report is to enable confident management by the non-expert recipient. A good report summarises the relevant information, provides a clear clinical interpretation and may also provide detailed management recommendations or refer to agreed management protocols. It needs to achieve a balance between over-simplification and inclusion of unnecessary detail.
Most referrers will have limited knowledge of the DXA technique and may not be aware of factors influencing the validity of the results such as osteoarthritic change in the spine causing artefactual overestimation of bone mineral density (BMD), which, if severe, may render the measurement uninterpretable. Artefacts may have a more limited impact and can be minimised by altering the scan analysis, for example to exclude a fractured vertebra from the region of interest. Decisions about the optimal scan analysis will be made by the DXA operator but the reporting clinician makes the final decision whether an unreliable result should be reported. It is important to highlight any unreliability and hence uncertainty in the report.
BMD measurements are interpreted in light of the clinical context, taking account of factors influencing bone health other than BMD including fracture history and independent risk factors for fracture. The history may highlight indications and contra-indications to specific therapeutic approaches. For example, evidence of poor visual acuity may suggest a need for assessment of falls risk and a history of oesophageal surgery or renal impairment would preclude bisphosphonate therapy.
DXA reporting should take place within a robust healthcare governance framework. All staff involved in scan acquisition, analysis and interpretation should undertake appropriate training and continuing professional development. The reporting process should be routinely monitored to ensure quality and consistency and ensure that the output is fit for purpose.