Maximum exposure – radiation risks during CT imaging

Ionising radiation saves countless lives every year. Its ability to detect all manner of conditions and provide help for all kinds of diagnoses has made it an indispensable tool in medical imaging, but its use is not without risk, particularly in the area of computed tomography (CT). Public concern about the overuse of the technique in healthcare is growing in Europe and the US, with the American National Cancer Institute in 2013 reporting that 2% of the country's cancer diagnoses every year can be attributed to CT scans.

Dr Sebastian Schindera has worked in the field of CT radiation dose management at various university hospitals in Switzerland, the US and Canada for more than ten years. He is also the project leader for CT Consulting Service, an organisation that advises radiological institutes in Switzerland on best practice with CT scanners, and has recently focused his efforts on arguing for stricter dose tracking and the wiser use of big data in radiology.

Risk factor

Health problems occur in cases in which clinicians neglect to control CT dose properly, and Schindera is emphatic in his belief that investment in state-of-the art tools for dose reduction and IT solutions for radiation dose monitoring are essential.

Most of the damage done by ionising radiation is rapidly repaired by the body, but in rare cases the DNA mutations it can cause lead to cancer, especially in younger patients.

"Those at highest risk are children and young adults," says Schindera. "Children's organs are more radiosensitive, and young people have higher life expectancies, so it sometimes takes decades to develop these cancers.

"Female patients are also at higher risk, due to the radiosensitivity of breast tissue. This doesn't mean that young male patients are not at risk, though"

A second major danger associated with ionising radiation is cell death, which only occurs when very high doses are applied with CT scans, usually caused by an error by the operator. While those incidents are rare, they do occur, and lack of training and effective technology often means that technicians are unaware of the damage they are doing.

Schindera points to a blunder in California a few years ago in which systematic overdoses took place because wrong technical settings were applied. "Up to eight times the usual dose was used for brain perfusion for CT, and patients came back with hair loss," he says.

"That's how they found out that the settings were wrong; it happened to almost 270 patients, and it was more than a year before they found out.

"Dose management software would have prevented this accident, because it immediately alerts radiologists and technicians that an error that has caused an overdose has occurred, thereby improving patient safety."

Schindera argues that the industry has only recently recognised problems with monitoring radiation levels in CT, but says that, despite efforts to cut down doses, many radiologists still do not track and collect data in the correct fashion, which is through systematic tracking and analysis.

"In the last ten or 15 years the radiology community has acknowledged that increased use of CT has resulted in more of the population being exposed to it, and a lot of effort has gone into reducing radiation doses.

"We put a lot of work into that subject, but don't know the exact figures, because we didn't collect the doses. A lot of institutions don't pay enough attention to dosing because it's time consuming: if you want to do it comprehensively and systematically, you must input the numbers manually."

He says he frequently meets radiologists, asks them what their analysis of their long-term dosing patterns in CT is and is frustrated by the uniformity of their answers.

"They say they don't know exactly, and I respond by saying that I understand why they don't know: it is because they are not tracking dosing comprehensively. I like to compare it to driving old cars that didn't allow you to monitor fuel consumption."

Schindera also believes that radiologists should not only reduce the doses they use, but also be encouraged to collect and analyse data, allowing them to see how dose-efficient their CT practice is compared with other institutions with the same sort of equipment. He describes this is a benchmark for doses.

Cumulative effect

Dose tracking also has important ramifications for patients who, because of ongoing health problems, have to take multiple CT scans over extended periods of time and build up a high cumulative dose. This is something that, surprisingly, has been neglected in the past.

"We get alerted by our dose tracking software when patients reach a high cumulative number," Schindera says. "When a patient comes back after numerous CT scans, we consider - while bearing in mind their disease and age - if they should get be imaged by techniques that don't use ionising radiation, such as MRI or ultrasound."

Good dose management also involves understanding that there are alternatives to ionising radiation. 'Dose justification' encourages clinicians and radiologists to think carefully about whether a CT scan is necessary and IT is at the forefront of this approach, both in terms of the dose tracking and through the use of clinical decision support systems, which use referral guidelines based on international standards. "The clinician enters the findings of the patient and any clinical questions, and the software refers him or her to the best imaging study with the lowest ionising radiation," says Schindera.

He also believes that radiology departments should take a multifaceted approach to CT dose optimisation. "First, we do the dose tracking," he says. "We want to know how good, or bad, we are, so we need some data, and this is based on the mean values of hundreds, or even thousands, of CT studies. We compare ourselves with other institutions in Switzerland, or internationally; you can look up the information in publications, and there are also some national diagnostic reference levels."

All this lets radiologists find the right level, or balance between the optimal dose and image quality. This is called the ALARA (as low as reasonably achievable) principle in radiology: to keep dosage as low as possible while ensuring a diagnosis can be made.

Schindera argues that the dose management software should be easy to use and robust functionally. "This is the main selling point for quality improvement; we can improve our quality with this data, which was not used in the past," he says. "I strongly believe that, in the near future, the average applied radiation doses in CT will demonstrate a key quality indicator in radiology.

"Departments that already use dose tracking demonstrate they are forward-thinking, and are in a position to tell their patients that they do more than their competitors in the market."

Marker of success

Schindera is quick to stress that software does not lead automatically to dose reduction, and argues that, where diagnostic accuracy is concerned, it is vitally important to strike a balance between reducing dosage and maintaining the requisite image quality.

"With dose optimisation, the radiologist must figure out how low they can go with regards to the dose without missing important information in the CT scan. We need some kind of marker, a measurement of image quality that tells the radiologist what level of dose reduction can be tolerated without any loss of data. Implementing image quality into dose monitoring solutions is very important in order to support radiologists in their dose optimisation efforts.

"But this is a big task, since image quality is affected by many different parameters. The vendors know that, and I am sure they are also working for future improvement on this topic."

Improving the way data is stored and managed is also vital, and Schindera claims hospitals and healthcare systems should work together to develop digital patient dose records - which is already technically possible - to minimise the risk of overdose.

"If a patient goes from, say, a hospital in Manchester to another in London, clinicians should be able to obtain the information from any previous imaging studies, including their results and the patient's cumulative dose," he says.

"There have been cases of patients receiving unnecessary CT scans, where the same procedure has been repeated in a short period of time because the recent studies were not available, or the clinicians did not know about them. It ought to be the case that all the information for a patient is stored on a central national server, or the patient's healthcare insurance card."

While the industry is largely self-regulating, new legislation introduced by the European Union in 2013 (coming into force in 2018), will mean that radiologists and clinicians will be obligated to document patient dose for every interventional study and CT study. This means that post-2018, manufacturers of CT scanners and fluoroscopy systems will have to make sure that all their devices track patient dosage, something that many vendors already do, according to Schindera. Now the technology is so widely available, though, there can no longer be any excuses for failing to control doses and giving this aspect
of patient care the attention it so obviously, and urgently, deserves.

Bayer introduced Dr Schindera to Medical Imaging Technology, but the opinions expressed in this article are his alone.