The risks of radiation exposure related to diagnostic imaging and how to minimise them
BMJ 2011; 342 doi: https://doi.org/10.1136/bmj.d947 (Published 25 February 2011) Cite this as: BMJ 2011;342:d947
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Rapid response to Davis et al. BMJ.
Davis and colleagues[1] have eloquently drawn attention to the
topical and important, but frequently overlooked, and poorly understood
subject of radiation dose and radiation protection. Sadly nuclear power
plant disasters capture headlines but collective dose to the population
from medical imaging is 3 orders of magnitude greater than from nuclear
energy production[2].
Perhaps the two principles of radiation protection viz. optimisation
and justification should be reiterated and the impact of measures to
support these outlined. Optimisation or the reduction of medical exposure
dose to as low as reasonably achievable (ALARA principle) is promoted in
the UK through use of Diagnostic Reference Levels and the National Patient
Dose Database. Justification (a legal requirement balancing benefit and
risk of a medical exposure) requires guidance to referrer and radiology
practitioner.
Over the last 20 years, the Royal College of Radiologists (RCR) has
published guidance to help clinicians select the appropriate imaging
investigation. Making the best use of clinical radiology services (sixth
edition) is available through NHS networks in the UK
(http://mbur.nhs.uk/index.asp ) and from the RCR (www.rcr.ac.uk ).
Reduction in medical exposures by 20% has been estimated[3]. Although
difficult to find a surrogate for appropriate imaging, the UK has the
lowest per caput collective dose in a dozen European countries lending
some weight to the value of justification[4].
The forthcoming seventh edition of the NHS Evidence-accredited RCR
imaging referral guidelines will be published in the autumn of this year
and aims to be concordant with guidance in many validated clinical
guidelines eg NICE guidance on low back pain. Although the pocket-sized
format of the print version has changed little in outward appearance and
will still fit in a now little used white coat, the size belies the value
for justification of medical exposures and more importantly bridges the
gap between clinician, radiologist and quite possibly the patient,
promoting the best test first.
References:
1. Davis H et al. Risks of exposure to radiological imaging and how
to minimise them. BMJ 2011;342:d947
2. Mettler F et al. Radiologic and Nuclear Medicine Studies in the
United States and Worldwide: Frequency, Radiation Dose, and Comparison
with Other Radiation Sources 1950-2007. Radiology. 2009; 253: 520-531.
doi: 10.1148/radiol.2532082010
http://radiology.rsna.org/content/253/2/520.full (Accessed 21.3.11)
3. Royal College of Radiologists Working Party. Influence of Royal
College of Radiologists' guidelines on referral from general practice.
BMJ. 1993 Jan 9; 306(6870):110-1.
http://www.ncbi.nlm.nih.gov/pubmed/8435606 (Accessed 21.3.11)
4. Hart et al. Frequency and collective dose for Medical and Dental X
-ray examinations in the UK. Dec 2010.
http://www.hpa.org.uk/web/HPAwebFile/HPAweb_C/1287148001641 (Accessed
21.3.11)
Competing interests: No competing interests
Expressing relative risk from exposure to ionising radiation is
frequently difficult -I thought the link to this website would interest
readers
http://www.xkcd.com/radiation/
Competing interests: No competing interests
Davies et al. have clearly identified key factors regarding ionizing
radiation utility and touched upon risk management. There is no doubt that
extra care can always be given to avoid inappropriate use of ionising
radiation studies. The authors emphasise that it is the responsibility of
the clinician to provide patients with the information needed to ensure an
informed decision is made before high radiation dose imaging tests are
performed.
A fundamental right of every patient is to know the risks versus of
benefit of any test or intervention. The risk of radiation-related disease
is generally perceived as being high. The recent tragedy in Japan is a
harrowing reminder of how people react to the threat of radiation
exposure.
Recognising the risk and communicating it to patients is vital. This
is rarely a significant part of the undergraduate curriculum - how good
are we at translating mathematical odds to patients in a way that they
understand, especially when the denominators are in the thousands or tens
of thousands?
Furthermore, risk perception undoubtedly links medical management and
health behaviour by patients.
Examining lay risk perception, it has been shown that people find it
difficult to process mathematical risk [1]. Assuming that even highly
educated patients can convert a risk into a percentage (eg 1:10000 to
0.01%) may not always be appropriate [2,3].
Furthermore, we often focus on the numerator when considering the risk of
an event: the 1 in a 100 chance is not always a consolation about the 99
who are absolutely fine.
The current average lifetime risk of dying from cancer is 1 in 4
(25%) [4]. The Health and Safety Executive (HSE) have examined levels of
risk, mainly in terms of tolerability and acceptability of different
levels of risk. Broadly speaking, as a risk increases the tolerability of
that risk moves from acceptable, to tolerable to unacceptable (HSE, 2001).
An individual risk of 1 in 1,000,000 (0.0001%) per year is considered
acceptable, and 1 in 100,000 (0.001%) per year is considered tolerable
[5].
It is reasonable to communicate radiation risk to patients by
describing risk as negligible (< 1 in a million), minimal (1 in a
million to 1 in 100,000), very low (1 in 100,000 to 1 in 10,000) and low
(1 in 10,000 to 1 in 1000). By using these terms in this context, and
knowing the mathematical risk to which they refer, a clinician may offer
justified reassurance to a patient when considering the risk of fatal
malignancy (adapted from HPA-RPD and the IRMER (2000) course).
Risk may also be compared to events that we all assume are unlikely,
for example a risk of 1 in 1000 is about the same as the lifetime risk of
a pedestrian being killed in a road traffic accident in the UK (National
Mortality Statistics, 2006). Alternatively, medical radiation risk can be
compared to cosmic radiation when taking a flight - an activity these days
of which most of our UK population will have some experience. A return
flight to New York is roughly the same radiation exposure as 5 chest x-
rays or about 2 weeks background radiation.
Comparing the risk of medical radiation exposure to background
radiation is useful. As malignancy will take several years to develop, a
patient over the age of sixty may not have enough time in their remaining
lifespan for a radiation-induced cancer to occur. Clearly the converse is
true for the paediatric patient. The Health Protection Agency (HPA) have
estimated that for patients under sixteen, the risk may be roughly
doubled, but in those over seventy the risk may be divided by about 5 for
radiation-induced cancer specific death [6].
The Ionising Radiation (Medical Exposure) 2000 (IRMER) guidelines [7]
should be referred to by readers of this article. The summary is available
online in PDF format [8]. IRMER is clear about the need for justifying
each medical exposure, and should arguably make up part of every medical
undergraduate curriculum.
For the time being, we are stuck with CT and ionising radiation studies.
Educating ourselves on how to communicate risk will help us to help our
patients make informed decisions.
References
[1] Paulos JA. Innumeracy: Mathematical illiteracy and its
consequences. New York, NY: Hill and Wang, 1988
[2] Woloshin S, Schwartz LM, Moncur M, et al. Assessing values for
health: numeracy matters. Med Decis Making 2001;21:382-390
[3] Lipkus IM, Samsa G, Rimer BK. General performance on a numeracy
scale among highly educated samples. Med Decis Making 2001;21:37-44
[4] Cancer Research, 2008
[5] Mobbs S, Watson S, Harrison J, Muirhead C and Bouffler S. HPA-RPD
-055 - An Introduction to the Estimation of Risks Arising from Exposure to
Low Doses of Ionising Radiation. June 2009
http://bit.ly/ig7W1l
[6] HPA Medical radiation FAQs
http://bit.ly/fxaTyN
[7] The Ionising Radiation (Medical Exposure) Regulations 2000
http://bit.ly/hdZnLD
[8] The Ionising Radiation (Medical Exposure) Regulations 2000
http://bit.ly/f1KgBA
Competing interests: No competing interests
Davies et al are right to warn of the health risks to patients from
their cumulative exposure to radiation during radiological imaging
investigations ordered by the doctors looking after them(1). BMJ
Editorialists warned us about this growing problem over a decade ago (2).
Part of the problem is ignorance on the part of doctors of the
radiation doses involved in the investigations we order (3) but another
part is ignorance of a patient's cumulative dose so far. In any one
institution it would be possible to calculate this from the time of the
first investigation recorded on PACS but it would be very time-consuming
to do so.
It should be possible to overcome the latter problem by modifying the
PACS system so that each radiological investigation undertaken results in
the effective radiation dose (e.g. 0.02 mSv for a chest x-ray) being
assigned to that patient and weighted for the patient's age and sex at the
time, resulting in a "risk score" which was displayed prominently in PACS.
Each time that patient had another radiological investigation the "risk
score" would be added to and updated. Patients commonly have
investigations in more than one hospital around the country and so ideally
patients' up-to-date risk scores from radiological investigations (and
treatment) undertaken nationwide would be displayed on the PACS system of
every hospital.
In addition, ideally such a system should eventually also take
account of radiation exposure during procedures not currently universally
listed on PACS such as coronary angiography and intervention, temporary
and permanent cardiac pacing, and orthopaedic surgery.
Risk scores could be assigned "green", "amber" and "red" status.
"Amber" status might mean the doctor ordering the investigation was
somehow required to answer the question: "Does this patient really require
this investigation?" before it could be undertaken. "Red" status might
mean, for example, that a request for a radiological investigation or
treatment must be ordered by a Consultant, discussed first with a
Consultant Radiologist, or approved at an MDT.
Such a system would need to be trialled. In simple terms I would
envisage a trial region; an invitation to all patients already on PACS to
participate; calculation of "risk scores" for all those who agree and
their random allocation to the current system or the new system; and a
comparison of two main short term (say 5 years) outcomes namely 1)
cumulative radiation dose, and 2) morbidity and mortality from conditions
in which radiological investigations and treatments contribute
significantly to diagnosis or management.
Yours sincerely,
A J M Broadley
Consultant Cardiologist
Yeovil District Hospital
1. Davies HE, Wathen CG, Gleeson FV. Risks of exposure to radiation
exposure to radiological imaging and how to minimise them. BMJ
2011;342:589-593.
2. Rehani MM, Berry M. Radiation doses in computed tomography: The
increasing doses of radiation need to be controlled. BMJ 2000;320:593-594.
3. Shiralkar S, Rennie A, Snow M, Galland RB, Lewis MH, Gower-Thomas
K. Doctors' knowledge of radiation exposure: questionnaire study. BMJ
2003;325:371-372.
Competing interests: No competing interests
I read with interest the timely article by Davies et al on the risks
of exposure to radiological imaging. The article provides an excellent
review of an important subject. One minor point I had some concern
regarding was their recommendation to perform ventilation-perfusion (V/Q)
scanning as the modality of choice to investigate pulmonary
thromboembolism (PTE) in pregnant women. This recommendation is at odds
with the advice from the Royal College of Obstetricians and
Gynaecologists[1]. They recommend chest radiography and compression duplex
Doppler should be performed as an initial investigation. If both tests are
negative with persistent clinical suspicion of PTE, then either a V/Q lung
scan or a computed tomography pulmonary angiogram (CTPA) is recommended,
without putting specific recommendation upon either investigation. They
offer balanced reasoning on the advantages and disadvantages of each, and
highlight that the decision to choose the modality should ideally involve
the patient. The guidelines state:
"(Pregnant) women with suspected PTE should be advised that V/Q
scanning carries a slightly increased risk of childhood cancer compared
with CTPA (1/280,000 versus less than 1/1,000,000) but carries a lower
risk of maternal breast cancer (lifetime risk increased by up to 13.6%
with CTPA, background risk of 1/200 for study population).
Where feasible, women should be involved in the decision to undergo CTPA
or V/Q scanning. Ideally, informed consent should be obtained before these
tests are undertaken."
This last point would surely be in line with the entire message of
the article by Davies et al. I suspect that many expectant mothers might
not choose to protect themselves over their unborn child.
1. Greer ITA. Thrombosis and Embolism During Pregnancy and the
Puerperium, the Acute Management of (Green-top 37b) British Society of
Obstetrics and Gynaecology Guidelines 2007.
Competing interests: No competing interests
Dear editor,
We read with interest the comprehensive review by Davies et al.(1)
which promoted medical imaging radiation dose awareness. It is imperative
that we do not lose sight of the fact that many more patients will receive
timely, effective and in some instances lifesaving treatment due to the
appropriate utilisation of X-ray imaging than will develop iatrogenic
induced cancer.
The article failed to highlight the assumptions that are used in
current epidemiologic models that estimate stochastic risk. There is no
conclusive evidence that cancers are induced by the level of exposure
associated with diagnostic radiology. Furthermore it is uncertain if the
radiation dose response curve at low level exposure is linear as largely
assumed or if there is a threshold effect for adverse events.
Increased awareness of cumulative radiation exposure, thoughtful
physician examination requisition and ALARA principles will reduce medical
exposure. CT test characteristics mean that it is unlikely to be surpassed
by MRI as the cross-sectional diagnostic workhorse of the NHS for some
time particularly with the introduction of newer applications such as
cardiac CT angiography.
The effective radiation exposure quoted for cardiac CT angiography is
not without concern. The authors fail to acknowledge the introduction of
patient specific protocols, tube dose modulation algorithms and iterative
reconstruction techniques that facilitate ultralow dose cardiac CTA
(<2mSv). In addition, the application of bismuth breast shields reduces
the organ specific dose to the breast by 50%. (2)
Surely an informed patient decision is only as good as the
information provided. The information we have is far from substantive and
at best speculative. Would the authors support accurate quantification of
cancer induction in in-vivo prospective longitudinal trials by DNA strand
damage and repair before we give patients a "lifetime of worry" about a
potential cancer.
(1) Davies HE, Wathen CG, Gleeson FV. The risks of radiation exposure
related to diagnostic imaging and how to minimise them. BMJ 2011;342:d947
(2) Bennett J, McKavanagh P, Doyle P, Ball P, Higginson D, Donnelly P.
Cardiac CT dose radiation using Bismuth Breast Shields. Irish Journal of
Medical Science. 2010;179(10):33.
Competing interests: No competing interests
I found the insightful review article by Davies et al extremely
useful as a junior doctor. Whilst we are reminded of the radiation risks
and contra-indictions in medical school, and sometimes asked about
radiation doses in membership exams, modern medicine now allows us to
request ad lib and view chest x-rays as an essential of any admission. In
addition, the CT from head to toe is often a pre-requisit of any fall or
trauma.
However, CT's or even the humble x-ray or Ultrasound is not always as
easy to access as in a DGH or teaching hospital. My most recent 2
rotations have been in a small DGH with limited x-ray facilities followed
by a cancer hospice with no on-site facilities. Using clinical reasoning,
deciding on why and how the investigation would change your management and
relying on your knowledge instead of working on auto-pilot has made me
appreciate the benefits and downsides of easy access imaging.
However, at times when I have requested less invasive and risky
imaging in the past, particularly MR there is more often than not a
barricade put up as the waiting lists are too long, or the imaging is more
precious and not first line. In addition, I feel the review almost
suggests we should be taking consent before requesting x-rays etc. If this
is to occur then I would suggest that paperwork will increase, bureaucracy
will increase and thinking about radiation will just become another tick
box exercise. What is needed is a change in culture from Consultant down
to encourage us to think before asking for imaging. After all, a good
author never reveals the culprit before you've read the whole book!
Competing interests: No competing interests
Davies and colleagues have given us an extremely useful and timely
review of the risks of exposure to radiation during radiological imaging
1.
However, it is really disappointing that in the Editors Choice text
referring to this article in the same edition of the BMJ 2, there is a
major error of fact. The Editors Choice states that " A CT scan of the
abdomen delivers eight times the dose of a chest radiograph. So does a
barium enema." As the article by Davies et al. has indicated in table 1,
the true average dose from a CT abdomen is 400 times the dose from a chest
radiograph and the average dose from a barium enema is also 400 times the
dose from a chest radiograph.
An error of interpretation perhaps occurred because the average dose
from a CT abdomen and a barium enema are each 8 mSv ( column one). The
same column of the table however indicates the dose of a chest radiograph
is 0.02 mSv. The number of equivalent chest radiographs for dose is given
in column 2.
I believe that radiologists will have been delighted that the
potential risks of radiation from diagnostic imaging were highlighted by
the original article and the topic included in the cover image of the BMJ.
It is most important that our clinical colleagues and patients have an
accurate understanding of those risks.
1 Davies HE, Wathen CG, Gleeson FV. Risks of exposure to radiological
imaging and how to minimise them. BMJ 2011;342:d947
2 Goodlee F Editor's choice Iatrogenic radiation, and unethical
reforms. BMJ 2011;342:d1551
Competing interests: No competing interests
Davies and colleagues highlight the risks of exposure to radiation
from diagnostic imaging techniques and recommend that as in any other
procedure, the benefits should outweigh the risks and the patients should
be informed of the potential complications.[1]
However, they focused on diagnostic tests. Fluoroscopy is another
imaging technique delivering ionising radiation used to assist therapeutic
procedures. This method has become mandatory for several procedures,
especially interventional pain management as it provides precise location
of anatomic target areas and facilitates accurate delivery of injection
therapy.[2]
The commonest procedures for which fluoroscopy is required in pain
management involve spinal injection therapies (e.g. lumbar medial branch
block, facet joint injection or lumbar rhizotomy). When assessing 137
patients undergoing fluoroscopically guided spinal procedures we found
that overall the radiation exposure is significantly lower than for other
interventional procedures, such as endovascular and cardiology procedures.
However, radiation doses are cumulative and this may be relevant in
patients undergoing repeated procedures. The practical experience of the
radiographer may also contribute to the patients' exposure to the
radiation, as screening time (34 ? 27 seconds (range 3-218)) is associated
with higher dose area product (r = 0.551, p<0.01). The associations of
dose area product with weight (r = 0.23, p<0.05) and spinal pathology
(r = 0.281, p<0.01), which interferes with landmark recognition, raise
the spectre that these are additional risk factors for increased risk of
radiation damage.
When considering interventional procedures under fluoroscopy imaging,
the consultant should take into account not only the average radiation
from a procedure but also specific characteristics, such as obesity,
spinal pathology and experience of the radiographer.
1. Davies HE, Wathen CG, Gleeson FV. Risks of exposure to
radiological imaging and how to minimise them. BMJ 2011;342:d947.
2. Manchikanti L, Cash KA, Moss TL, Pampati V. Effectiveness of protective
measures in reducing risk of radiation exposure in interventional pain
management: a prospective evaluation. Pain Physician 2003;6(3):301-5.
Competing interests: No competing interests
Informed consent for radiological investigation in pregnancy
Last year we carried out a national anonymous web based survey on the investigation of PE in pregnancy. 131 Acute Physicians and 64 Obstetricians responses were analysed. Physicians appeared somewhat better at involving women in the decision to have a V/Q scan or a CTPA at 52%, compared to 30% of Obstetricians. 24% of obstetricians and 11% physicians used consent forms for these investigations in pregnancy. Only 9% of Physicians and 11% of Obstetricians had information sheets explaining the benefits and potential risks to the mother and her baby from these diagnostic radiological procedures. Since this survey our Trust (Portsmouth Hospitals NHS Trust) has designed a patient information leaflet for investigating PE in pregnancy which we are currently trialling.
References
1)Davies HE, Wathen CG, Gleeson FV. The risks of radiation exposure related to diagnostic imaging and how to minimise them BMJ 2011 march, 342:589--5921
2)Matthews S. Short communication: imaging pulmonary embolism in pregnancy: what is the most appropriate imaging protocol? Br J Radiol 2006 May;79(941):441-4.
3)Thromboembolic disease in pregnancy and the puerperium: Acute management, Royal College of Obstetrician and Gynaecologists. Green-top guidelines No 28 Feb 2007.
4) Rajaraman P, Simpson J, Neta G, Berrington de Gonzalez A, Ansell P, Linet MS, et al. Early life exposure to diagnostic radiation and ultrasound scans and risk of childhood cancer: case-control study. BMJ2011;342:d472. (10 February.)
Competing interests: No competing interests