Aberdeen Biomedical Imaging Centre
Research Area: Imaging Technology Evaluation
Rapid developments in imaging technologies have created the need to evaluate the impact of these on clinical care. There is a requirement to establish whether new technologies are effective and whether there is a cost benefit for a healthcare system. From a diagnostic imaging perspective we have undertaken work in the following areas:
- Magnetic Resonance Imaging in low back pain
- Computer Aided Detection in mammography
- Magnetic Resonance Imaging breast screening
- Mammography screening in moderate risk women
- Magnetic Resonance Imaging of the knee
- Endoscopic Ultrasound in staging the oesophageal cancer
Magnetic resonance imaging in low back pain
We investigated the early use of MRI on the clinical management and outcome of patients with low back pain referred by their GP to a consultant orthopaedic specialist or neurosurgeon. The study was
a multicentre randomised trial conducted in collaboration with the Health Services Research Unit, and the Department of Orthopaedic Surgery, University of Aberdeen and funded by the NHS Research and Development Health Technology Assessment Programme (PI Professor Fiona Gilbert, trial coordinator Dr Maureen Gillan).
An ‘early imaging’ policy:
- Had no significant impact on clinical management overall
- Increased clinicians’ diagnostic confidence
- Resulted in a slight improvement in patient outcome but this was of questionable clinical importance.
The cost effectiveness of a policy of 'early imaging' is dependent on whether the small improvement observed in patient health outcome is considered to be clinically important
[Health Technol Assess 2004 8 (17) 1-144, Radiology 2004 231: 343-51, Radiology 2001 220: 393-399]
MRI L spine 
MRI and suspected internal derangement of the knee
In collaboration with the University of York, we participated in a multi centre pragmatic randomised trial (DAMASK) funded by the Medical Research Council. The trial evaluated the effect of GP direct access to MRI versus referral to orthopaedic surgeons for patients with suspected derangement of the knee in terms of clinical management, patient outcomes and cost effectiveness.
Compared to routine referral to an orthopaedic specialist, early access to MRI
- Did not significantly alter GPs’ diagnoses or their decision to refer for a specialist opinion
- Significantly increased GPs confidence in their decisions
- Resulted in an improvement in health-related quality of life but no significant improvement over time in patients’ physical functioning
- Represents a cost-effective use of healthcare resources
[Br J General Practice 2007 57: 622-629, Br J General Practice 2008 58:767-774, Br J General Practice 2008 58:775-778, Journal of Clinical Epidemiology.2007: 60(10):1046-51, BMC Medical Research Methodology.2007: 7:12]
Knee MRI
Computer aided detection (CAD) in mammography film reading in the NHS breast screening programme
We have evaluated the potential of CAD to enhance the performance of film readers and provide and alternative to the standard practice of double reading mammograms in breast screening mammography. A retrospective matched evaluation study (CADET I) was conducted in Aberdeen and Manchester funded by Cancer Research UK and the NHS Breast Screening Programme (PI Professor Fiona Gilbert). A random sample of over 10,000 routine screening mammograms, previously double read in 1996, were read again by single readers using CAD.
- The cancer detection rate using CAD was at least as good as double reading with only a small increase in recall rate.
[Radiology 2006 241 (1)47-53, Breast Cancer Research 2008 10:R64, Breast Cancer Research 2008 10:R64]
Image
Following on from CADET I, we recently completed a prospective randomised evaluation of CAD in the NHS Breast Screening Programme funded by Cancer Research UK and the NHS Cancer Screening Programme. The CADET II Trial was based in Aberdeen (PI Professor Fiona Gilbert, trial coordinator Dr Maureen Gillan) with collaborators in Coventry, Manchester, Nottingham and London.
CADET II was designed as an equivalence trial with matched pair comparisons of the cancer detection rate and recall rate of single reading using CAD with standard double reading of mammograms. Mammograms from 28,204 women attending routine mammography were read by both double reading and by another single reader using CAD.
- No significant difference in cancer detection rate between single reading with CAD and double reading
- Recall rate for single reading with CAD was 0.5 % higher than that for double reading (equivalent to a relative difference of 15%)
- Single reading with CAD could be an alternative to double reading or improve the cancer detection rate in screening mammograms read by a single reader
- Costs of the CAD equipment and the costs associated with an increase in recall must be balanced against any potential savings in reader time.
[NEJM 2008 359:1675-84]
MRI screening in high risk women
MARIBS - prospective multicentre cohort study of the sensitivity and specificity of contrast enhanced MRI and x-ray mammography for the screening of premenopausal women who are at high risk of developing breast cancer. The trial was funded by the Medical Research Council and Professor Fiona Gilbert was a local collaborator and steering committee member.
- Contrast enhanced MRI is almost twice as sensitive as mammography in identifying tumours in this cohort of women
- Recommendation was made for annual screening of high risk women with contrast enhanced MRI and mammography
- Collaboration with Dutch and Canadian studies showed that tumours in younger BRCA1 carriers had a more rapid doubling time compared to non carriers and older women.
- [Lancet, 2005: 1769 - 1778, Magnetic Resonance Imaging, 2006;24:917-929, Radiology 2009; Magnetic Resonance Imaging 2002; 20: 557-565; http://www.cancerimaging.org/articles/vol5iss1_022.htm; Radiology. 2005;236(3):779-788; Magnetic Resonance Imaging, 2006;24:917-929; Radiology 2006;239:677-685; British Journal of Cancer 2006:95;801-810; Clinical Cancer Research 2007;13:7357-7362
Breast MRI
Mammographic surveillance in women at moderate risk of breast cancer
The FH01 study (led by Profs James McKay and Stephen Duffy) is an evaluation of mammographic surveillance in women under 50 who have been referred to their local genetics centre because they have a family history of breast cancer,. The primary aim of the study is to evaluate the benefit, in terms of reduction in advanced tumours and the subsequent estimated reduction in breast cancer mortality and cost effectiveness, of mammographic screening in women with a moderate/high risk of developing breast cancer. The study is funded by Cancer Research UK and the NHS Research and Development Health Technology Assessment Programme. Recruitment of participants is now complete and the study is in the process of identifying screen detected and interval cancers in study women.
Journal of Medical Screening 2006:13:177-182
Endoscopic ultrasound in the in the diagnosis and staging of gastric and oesophageal cancer
In collaboration with the University of Wales, we are conducting a formal health technology evaluation (funded by the NHS Health Technology Assessment Programme) of endoscopic ultrasound in the diagnosis and staging of gastric and oesophageal cancer.
The COGNATE Trial is a pragmatic multi-centre randomised trial comparing the outcome of patients with gastric and oesophageal cancer who are staged with or without EUS. Recruitment of participants is now closed and data collection of patient management, healthcare resource use, and Quality of Life questionnaires is almost complete.
Contact: Professor Fiona J Gilbert, Dr Maureen Gillan