Sophisticated medical scans such as MRIs are getting better at finding things potentially wrong with us. But sometimes it’s better not to know. By Ray Moynihan.

Incidentalomas make for testing times

A patient being positioned in  a CT scanner.
A patient being positioned in a CT scanner.

Ultrasounds, CTs and MRI scans are surely among the great medical miracles of our time. They help doctors diagnose potentially deadly disease and, as we all know, early detection can lead to successful treatment.

So when I ask Professor Rustam Salman, chair of clinical neurology at the University of Edinburgh, whether he’d have an MRI scan to check his brain for any suspicious abnormalities, his response comes as something of a shock.

“I would not have a scan,” he says calmly. “My personal and professional view is that ignorance is bliss.”

Why would a brain specialist not advocate using the most sophisticated high-tech equipment available to scan for any early signs of trouble?

The answer lies in his research on “incidentalomas”, which Salman is speaking about during a visit to Australia.

Incidentalomas are abnormalities picked up incidentally when healthy people are screened with ultrasounds, magnetic resonance imaging (MRI) or computed tomography (CT) scans, or found accidentally when we’re being tested for some other problem. The term might sound like a joke – it’s derived from “incidental” and “-oma”, meaning tumour – but the threat posed to our health by discovering so many of them is deadly serious.

As a neurologist, Salman sees firsthand the devastation a brain malformation can wreak on someone – the bleed from a burst aneurysm, for instance, can cause permanent disability or death. So his desire to remain blissfully ignorant seems rather odd. 

Indeed, statistics on scans suggest many Australians don’t share his view. Driven by a deep faith in early detection and enthusiastic referral from litigation-averse doctors, we line up for X-rays and ultrasounds and lie down for CTs and MRIs at an extraordinary rate.

Medicare figures show that in the past seven years the total number of scans has jumped 40 per cent, to more than 22 million this past year, at a yearly cost of $3.3 billion.

Undoubtedly, many are necessary, aiding diagnosis and guiding therapy, and sometimes even averting life-threatening illness. But there’s a strong view many are not needed. And because we’re looking inside ourselves so often, with ever-more sophisticated technology, we’re naturally finding more and more small “abnormalities”, which may never have caused us any problem if we didn’t know they were there. Once found though, the clock can’t be wound back, and the cascade of unintended consequences flows with full force – fear, anxiety, labels, more tests and, in some cases, lifelong treatments. The obsession with scanning is causing an epidemic of incidentalomas in many of our organs, including the thyroid, liver and brain.

“The popular idea is ‘have technology, will use it’ and that has inherent appeal,” says Professor Salman, “but we need to know if it does more good than harm.” This is exactly what the British neurologist and his colleagues are trying to find out. One key area of their research is aneurysms, the abnormal pouches that can form on the side of a blood vessel.

The evidence around aneurysms is complex and uncertain but, as Salman explains it, it goes something like this. About one in every 50 people has an aneurysm in their brain, without them ever knowing it. Among those who do have one, there’s approximately a one in 15 chance it will rupture and bleed at some point in their lifetime. Operating to remove an aneurysm can lead to complications, and comes with a risk of stroke or even death. So if an aneurysm is found on a scan, the awkward question emerges: how does the risk of leaving it untreated compare with the risk of treatment to remove it? Put another way: is the treatment worse than the disease?

According to Salman’s analysis the five-year risk of a bleed if the aneurysm is left untreated is roughly equivalent to the risk of stroke or death that comes with the surgery to remove it. So the dilemma is dire: live with the ticking time bomb, or face a not insignificant chance of serious harm from defusing it. With this in mind, the “ignorance is bliss” approach starts to make a lot of sense. And all those quantified-self junkies who want to measure everything and the techno-fantasists who dream of hand-held MRIs should be careful what they wish for.

Another example of a brain incidentaloma is the tangle of blood vessels known as an arteriovenous malformation. In this case, researchers including Salman ran a randomised trial with more than 200 people with brain tangles. Half the participants received low-tech “medical management”– consisting mainly of medication to help with symptoms if they emerged – and the other half received the more aggressive “interventional therapy”, which could involve surgery or radiotherapy to eradicate the tangle.

The results published this February in The Lancet are sobering. The trial was stopped early because “the risk of death or stroke was significantly lower in the medical management group than in the interventional therapy group”. The implication is that if one of these rare brain malformations is picked up on a scan, it may be safer to live with it, with all the anxiety the diagnosis brings, than to try to have it removed.

The American College of Radiology has also released a policy paper on this phenomenon. Radiologists know the problem better than most, because they’re often the ones deciding whether that blur on your scan is worth reporting or not, with litigation fears ensuring they err on the side of reporting everything. Once the radiologist reports the “abnormality” to your GP, it’s incumbent on them to tell you, and as the policy paper points out, further investigation becomes virtually inevitable: “Although it is known that most incidental findings are likely benign and often have little or no clinical significance, the inclination to evaluate them is often driven by physician and patient unwillingness to accept uncertainty, even given the rare possibility of an important diagnosis.”

Two years ago a surgeon from Canberra Hospital, James Fergusson, published an academic paper on incidentalomas found in the liver. “With the widespread use of medical imaging,” Fergusson wrote, “has come the detection of incidental liver lesions that are, by and large, asymptomatic prior to their discovery.”

Fergusson argues these incidentalomas create anxiety and fuel further tests as doctors attempt to reassure people of the “usually benign nature” of the incidental finding. And that is the catch-22: to get a definitive diagnosis and determine whether your problem is benign or life-threatening, you may need surgery to remove a portion of your liver, which is a costly procedure that can cause significant harm and even death.

The explosion of thyroid incidentalomas provides further reason to rethink our blind faith in early detection. Developed countries including Australia and the US have in recent times experienced a more than doubling in the numbers of people diagnosed every year with cancer of the thyroid – the all-important gland that sits in our neck. Yet over that same period there’s been very little change in the rate of death from the disease. One response is to hail the success of thyroid cancer treatment. Another interpretation is that many people are being diagnosed with thyroid incidentalomas that would never have harmed them.

As it happens, most of the increase is in diagnosis of the small papillary thyroid cancers, which carry the least chance of causing you harm. Yet treating thyroid cancer carries risks that can include nerve damage from surgery and lifelong thyroid pills. Writing in BMJ last year, Mayo Clinic researchers revealed that more thyroid cancers were now being discovered via  incidental findings than because someone presented with symptoms. They concluded that overdiagnosis and overtreatment of this cancer need to be fully recognised.

Whether it’s incidentalomas of the brain, liver, thyroid or any other organ, the risks of overdiagnosis and overtreatment need a lot more public and professional attention. Scans will continue to be an important diagnostic tool, but winding back their excessive use can be a genuine win-win. By being more sceptical and avoiding unwarranted scans we’ll not only help protect ourselves and our loved ones from the epidemic of incidentalomas, we’ll help preserve precious health resources for those who really need them.

This article was first published in the print edition of The Saturday Paper on May 31, 2014 as "Testing times".

A free press is one you pay for. Now is the time to subscribe.

Ray Moynihan is an author and BMJ columnist, and a senior research fellow at Bond University.

Sharing credit ×

Share this article, without restrictions.

You’ve shared all of your credits for this month. They will refresh on August 1. If you would like to share more, you can buy a gift subscription for a friend.