MRI Brain Roulette

Cut-away MRI of brain - level of thalamus

Back in the summer, clinicians and scientists shunned the sunny weather to consider something more insular: brain imaging. Specifically, they gathered at a one day meeting at the Wellcome Trust to explore the thorny issue of what researchers should do when scans of volunteers – taken purely for research purposes – reveal ‘incidental abnormalities’, that is physical anomalies that are outside the scope of the study, which may or may not have health implications for the volunteer.

The problem isn’t explicitly addressed in existing guidelines, although these kinds of findings are surprisingly common. A neurologist at the meeting noted that one in 37 neurologically asymptomatic people going into a brain MRI scanner have an incidental abnormality. Since a roulette wheel has 37 pockets, he likened taking part in this kind of research as a form of ‘brain MRI roulette’.

Delegates at the conference were aiming to reach some sort of consensus on how to fairly treat normal subjects undergoing scans for research, and to establish some sort of guidance for researchers faced with incidental findings in their volunteers’ scans.

The big question is whether – and to what extent – researchers should follow up an incidental finding.

Is it outside their responsibility to do anything at all about an anomaly they find during their research? Or should they tell the volunteer, inform his or her GP, and facilitate or oversee the referrals process for further clinical investigation and even treatment?

More harm than good?

The lack of clear boundaries is compounded by the fact that reporting and acting on an incidental finding may not always be the best thing for the volunteer.

Of course, on the one hand, finding out about these findings can save lives. A cardiologist at the meeting described how a scan he was given as part of a research study unexpectedly revealed an aortic aneurysm with a high risk of sudden death. Finding it significantly increased his chance of survival because he is now being treated and monitored.

Quite often, however, further investigation reveals the finding to be a ‘false positive’ (clinically insignificant). That means the volunteer has needlessly endured a period of stress and uncertainty, which in itself takes a physical toll.

Even if the finding does turn out to be clinically important, there is still the caveat that no medical treatment is 100 per cent safe. They all have their risks and side effects, and sometimes treatment can even do more harm than good – particularly if the volunteer has no symptoms or obvious problems.

Other impacts may include the shock and disempowerment of suddenly, out of the blue, becoming a patient instead of simply a research participant. Job and career loss is possible (if, for example, a scan shows that a train driver or pilot has a brain anomaly), and if the finding is significant, there may be problems getting health and travel insurance.

So in a nutshell, if you volunteer for a research study and get told about a finding, it could be life-saving – but it could also be pretty life-damaging. And it’s not always possible to know which it will be until the finding has been investigated (which means it will also be recorded on your medical records).

The right to remain ignorant

Some volunteers state plainly that they don’t want to know the results of their scans, and they don’t want their GPs to be told about them either. As one speaker at the conference pointed out, the right to personal autonomy includes the right of a research volunteer to choose whether or not to be told about any findings – and whether or not to act on them.

On the other hand, a participant’s right to remain ignorant of research findings has to be balanced against public safety. If a pilot or train driver turns out to have a possible brain tumour, should they be allowed to continue in their jobs unaware of their condition?

Different research centres have different policies. Some allow volunteers to opt out of being told their findings and agree not to notify their GPs. Other centres don’t allow people to take part in research unless they agree to their GP being notified of significant findings.

These differences were reflected in mixed feelings among the conference delegates. Some felt that informing the GP served as a minimum standard or safety net to ensure volunteers with findings could be followed up. Others were concerned about breach of confidentiality, the possibility of compromising the volunteer (who may face future difficulties getting insurance, for example), and infringement of autonomy personal if external regulatory guidance determines when someone should go to their GP and become a patient.

Another important question is whether the principle investigator of the research study should also be notified of an incidental finding, bearing in mind that it might affect the validity of the study. Or again, would this be a breach of confidentiality?

Great expectations

Some of the confusion over what, if anything, to do about incidental findings lies in the expectations of volunteers taking part in research. Some people assume that the scan will be completely anonymous and they won’t be hearing any more about it. Others seem to associate getting into a scanner with medicine. Although it is explained to them during the consent process that the scan is not medical or diagnostic in nature, they expect to get clinical information back from the scan.

Some even volunteer for the study because they have symptoms but are unwilling or afraid to approach the doctor about them, and come for a research scan to get reassurance. There is a real danger of false reassurance, if they haven’t understood that the scan may not be reviewed by a radiologist or clinician.

To screen or not to screen?

One possible way to circumvent the confusion is to include a clinician or radiologist on the research team. They would then be responsible for reviewing scans and following up incidental findings.

Or potential participants could be offered a diagnostic scan prior to the research, which would be examined by an experienced radiologist or clinician, who would decide whether any finding is clinically important. The volunteer could opt to ignore the result – but may be refused participation in the study.

This would solve the problem of false positives – and would also guarantee for the researcher that the volunteer’s body or brain region is in the normal range for the study, with no anomalies that might influence the results.

However, both these measures would significantly increase the costs of research and there simply isn’t enough manpower in terms of the clinical expertise needed to review the scans. As one delegate pointed out, there is simply more research going on than there are expert radiologists. One way around this, which was briefly discussed, may be to have a centralized resource for radiological review of research scans.

Discussing the issues certainly helped elucidate the issue, and as a next step, the organisers and delegates are working together to produce a set of standard procedures – or reasonable ways of doing things – that will help researchers faced with similar dilemmas in the future. This will encourage best imaging research practice in the UK.

There are clearly no easy solutions. However the aim is to incorporate the many different viewpoints heard at the meeting – including the complex responsibility of both researcher and volunteer – within the guidance, and attempt to set a framework within practical limitations.

Penny Bailey

Image credit: Mark Lythgoe & Chloe Hutton. Wellcome Images