Statistics from Altmetric.com
Population-based cohort study, linking data from national Medicare and Cancer Database records.
10.9 million patients aged 0–19 years from 1985 to 2005. 681 211 patients had a Medicare-funded CT scan during the study period and were classed as exposed. Patients whose scans were in state-based tertiary hospitals (not Medicare funded) were not identifiable and were not moved to the exposed group.
3150 cancers were identified in patients who had a CT scan at least a year before the cancer was detected. Cancers detected within a year of the CT scan were not included to minimise the potential reverse causation of CT scans performed for cancer-related symptoms.
Time since exposure, year of exposure, age at exposure, sex, socioeconomic status, site of CT scan and estimated radiation dose.
Cancer incidence rates in CT-exposed versus non-CT-exposed patients.
After a mean of 9.5 years follow-up, the exposed group of patients had 24% greater cancer incidence with an incidence rate ratio (IRR) of 1.24 (95% CI 1.20 to 1.29). Factors associated with increased risk were multiple CT scans, exposure under the age of 5 years and scans of the chest, abdomen or pelvis. The absolute excess incidence rate was calculated as 9.38 per 100 000 person years at risk.
When the results were analysed introducing lag periods of 5 and 10 years instead of 1 year, the IRR remained significantly increased at 1.21 (95% CI 1.16 to 1.26) for 5 years and 1.18 (95% CI 1.11 to 1.24) for 10 years.
Diagnostic CT scans in children and young people increase the risk of radiation-related malignancy.
CT scans are invaluable medical tools aiding diagnosis and supporting patient management. However, a single CT scan can deliver a radiation dose equivalent to 1000 X-ray films. Indeed, a neonate can receive an effective dose of 20 mSv from a single abdominal scan—the equivalent of the 1-year radiation advisory limit for radiation workers.1
Radiation exposure is hazardous to health: at high doses it causes cancer and other diseases.2 However, the effect of lower doses has been difficult to assess, especially as long the latency between exposure and disease makes follow-up difficult. However, it is now possible to link electronic records on large patient populations over many years, and using this approach Mathews’ group in Australia was able to follow CT scan and cancer history in 11 000 000 people over 10 years. The study found a 24% increase in cancer in the CT scan-exposed group, consistent with a similar, although smaller, study of cancer risk in over 120 000 young people who received CT scans in the UK.3
The conclusion is inescapable: radiation causes cancer and CT scans expose patients to radiation and cause cancer in patients. The younger the child, the more numerous the scans, the greater the risk.
Radiological organisations have embraced the ALARA (As Low As Reasonably Achievable) approach for both occupational and medical exposures. Paediatricians should also embrace the ALARA concept: fewer children having fewer CT scans. For the paediatrician, this means considering the necessity of each radiological image ordered. Will the results of the scan make a direct difference to the management of this patient? Are CT images essential? Would a plain X-ray or ultrasound or MRI be possible alternatives?
Like many medical tools, CT scans are not risk free. It is the physician's responsibility to ensure that the benefit of the information provided by a CT scan outweighs the risk of the procedure, every time.
Competing interests None.
Provenance and peer review Not commissioned; internally peer reviewed.