Radiation doses in computed tomography

BMJ 2000; 320 doi: http://dx.doi.org/10.1136/bmj.320.7235.593 (Published 04 March 2000) Cite this as: BMJ 2000;320:593

The increasing doses of radiation need to be controlled

  1. Madan M Rehani (mmrehani{at}vsnl.com), additional professor of medical physics.,
  2. Manorma Berry, professor of radiodiagnosis
  1. All India Institute of Medical Sciences, New Dehli 110029, India

    Computed tomography has made dramatic advances, both in its breadth of application and in its technological improvements. The advances are such that it is possible with the spiral technique to carry out an entire examination of the chest within a single breathhold as against a few minutes in earlier systems. Yet these advances have brought with them the potential for greatly increased doses of radiation to the patient.

    Until a few years ago computed tomography constituted about 2-3% of all radiological examinations but contributed about 20-30% of the total radiation load from medical use of ionising radiation.1 2 A recent report from the Royal College of Radiologists in the United Kingdom states, “CT now probably contributes almost half of the collective dose from all x ray examinations.”3 Although magnetic resonance imaging was expected to reduce the frequency of computed tomography, this has not happened. Indeed, the use of computed tomography has grown. It is now often used as an adjunct to radiotherapy or chemotherapy; interventional procedures use computed tomography for fluoroscopy and angiography;computed tomography equipment is available in operating theatres and postoperative areas; and the technique is increasingly used in children. All these contribute to an increased use of computed tomography and of high doses of radiation to patients. Europeans have long been concerned about these high doses—the recent European Union Euratom directive categorises computed tomography and interventional radiology as procedures that expose patients to high doses of radiation—but other parts of the world also need to take the risks seriously.

    Typical computed tomography of the chest gives a radiation dose equivalent to 400 chest radiographs (chest tomography ~8 mSv; chest radiography=0.02 mSv).3 Computed tomography of the thoracic spine, mediastinum, abdomen, liver, pancreas, kidney, lumbar spine, and pelvis is associated with effective doses of >5 mSv (equivalent to over 250 chest radiographs) and in some cases as high as 30 mSv (equivalent to 1500 chest radiographs). Furthermore, the dose to the breast in many thoracic examinations ranges from 18 to 33 mSv, 4 while the dose to the lens of the eye is around 30mSv in computed tomography of the head, about 70 mSv in scanning of sinuses, and about 10-130 mSv in scanning for orbital trauma.

    Wall and Hart reported a 30% reduction in doses of radiation from common radiological procedures compared with 10 years ago but an increase in radiation doses of about 35% for computed tomography of the abdomen and pelvis.5 This may be based on the collective dose, which depends on the frequency of examination, but individual doses are not reducing, because larger areas are being included in each examination. There is a common belief that the shorter the examination the lower the dose, but that is not so.

    What can be done to reduce these high doses? There may be alternative examinations. For example, Dixon has suggested that the role of computed tomography in following up of testicular cancer should be reconsidered.6 The abdomen could be examined with ultrasound and magnetic resonance imaging and the chest with low dose computed tomography, though it may seem more attractive in terms of speed and cost to perform the whole study involving chest and abdomen with computed tomography. Furthermore, many UK departments have already reduced to a minimum the number of computed tomography examinations for intra-abdominal disease alone.6 In recent years magnetic resonance imaging has superseded computed tomography for examining the head, neck, spine, and many parts of the musculoskeletal system, and it often offers an alternative for examining the abdomen and pelvis.

    Computed tomography, however, remains the technique of choice for evaluating head injury; assessing spinal, pelvic, or abdominal trauma; characterising parenchymal lung diseases; staging almost all solid tumours, including lymphoma; and treatment planning for most solid tumours. However, the use of pelvic computed tomography for clinical staging in patients with high levels of prostate specific antigen cannot be recommended because of limited utility and lack of cost effectiveness. 7 8

    Abdominal computed tomography in the absence of clinical or laboratory evidence of trauma, merely because of decreased sensation, or its use prophylactically before general anaesthesia for non-abdominal surgery yields few useful results.9 The Royal College of Radiologists recommends that referrals for computed tomography should be vetted by an experienced radiologist. It is not possible to achieve substantial dose reductions in abdominal computed tomography simply by technical factors, so the use of alternative methods of examination is preferable.10

    A promising approach to reducing doses of radiation is to shield superficial radiosensitive organs such as the breast, the lens of the eye and the thyroid, and the testes in computed tomography of the thorax, head, and pelvis respectively. Reductions in radiation doses to these organs of over 50% have been reported with the use of thinly layered bismuth radioprotective latex or leaded garments, 11 without affecting the display of other deeper structures. Such shielding is even more important in children. Although such developments are the responsibility of radiologists, referring clinicians should also remember that computed tomography examinations are not without risk and should ensure that the examinations they request are really necessary—and the most appropriate.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.