• Magnetic resonance imaging (MRI) is now established as a useful diagnostic technology in many examinations of the head and spine.
  • Hard evidence of the cost-effectiveness of MRI in informing clinical decision making remains limited. The technology has promise in many areas, but further work is needed to establish its relative advantage over other diagnostic technologies.
  • A variety of MRI scanners are available, differing in the type and field strength of the magnet used. The Panel considers that there is no optimum type of scanner and that decisions on procurement should be made on a case by case basis. The Panel considers that at present:
    • mid-field superconductive systems appear to offer advantages in terms of capital and operating costs and to give good image quality
    • high-field systems can provide high quality diagnostic services, but their superiority over mid field systems is not established in terms of influence on clinical decision making
    • lower field systems, some based on permanent magnets, offer the potential for reducing costs of MRI, but further information on their technical capability and clinical effectiveness is needed.
  • The Panel does not support the acquisition of MRI scanners with field strengths of greater than 1.5T, except in the context of medical research.
  • The performance of MRI is enhanced in some examinations by the use of contrast agents such as gadolinium-DTPA. The Panel supports the more general availability of such materials in Australia, but notes that criteria for their use will need to be developed and that they may increase the cost of MRI services.
  • The safety of MRI appears to be acceptable, if appropriate safeguards are taken to avoid exposure of ferromagnetic objects to the magnetic field of the scanner.
  • The costs of MRI examinations may decrease in future with the availability of cheaper magnets and faster scanning sequences. However, MRI can be expected to remain an expensive diagnostic imaging technology with higher costs than CT scanning.
  • It is necessary for units operating MRI scanners to acquire appropriate hardware and software updates to maintain acceptable standards of performance.
  • The Panel considers that expansion of MRI facilities in Australia should be cautious, and concentrated where the comparative advantage of the technology is best established.


The Panel recommends that:

  • Priority is given to providing access to MRI for hospitals with substantial neurosurgical responsibilities. This would imply a total of about 16 MRI units nationally in neurological teaching hospitals.
  • Those private sector units which are currently associated with public hospitals are brought into any national review of MRI facilities.
  • MRI is used by specialised personnel in the most cost-effective way. This implies some restriction on access to the technology to ensure its efficient usage.
  • The costs of upgrades to hardware and software are included in all operating budgets.
  • The development and use of MRI be kept under review, particularly in newer areas of application such as cardiovascular imaging.
  • All public sector MRI units be required to routinely collect information on the examinations they undertake so that a data base can be developed on the evolving role of the technology. Similar routine data collection at private MRI facilities should be encouraged.