Incidenta-what? An incidentaloma is an unexpected abnormality accidentally found on medical imaging studies done for an unrelated reason. It typically leads to further diagnostic tests, sometimes invasive ones like biopsies and surgery, and it almost always turns out to be something harmless. It would have been better not to know it was there. Sometimes ignorance is bliss. Maybe that ostrich with his head in the sand had the right idea.
Incidentalomas can kill
Even when the incidentaloma itself is harmless, the process of figuring out that it is harmless can be deadly. An incidentaloma killed my father-in-law. He had a routine chest x-ray on his retirement physical, and it showed a spot on his lung. They were afraid it was lung cancer, since he was a smoker and since his brother had died of lung cancer. They opened his chest and excised the lesion. He was a healthy man with no symptoms when he walked into the hospital, and he was a corpse when he was carried out two weeks later. One complication of surgery led to another, and after a third operation he developed peritonitis and died. The lesion wasn’t cancer; it turned out to be only a harmless scar from an old fungal infection, histoplasmosis. The chest x-ray wasn’t indicated in the first place; today we no longer do routine chest x-rays without a good reason.
Radiologist Jill Wruble recently addressed the incidentaloma problem in a Tedx talk available on YouTube.[1] She describes a hypothetical example of a competitive skier who takes a tumble on the slopes, falling on his left side. He goes to the ER and they quite reasonably do a CT scan to rule out damage to his spleen. The spleen is OK, but the CT scan shows two unexpected findings: a spot on his lung, and a lump on his kidney. That triggers an MRI of the kidney that shows a benign finding, and a CT scan of the chest that shows the lung spot is probably benign but that identifies a new abnormality: a thyroid lump. That finding triggers three more tests: an endocrinology consult, a thyroid ultrasound, and a biopsy. He has undergone a cascade of six tests, even though he has no symptoms. None of the findings is likely to ever have caused a problem.
This sequence is unfortunately all too common. It results in prolonged anxiety, expense, and danger of adverse effects. It’s nothing new; I remember the case of an Air Force general several decades ago who was evaluated for migraine headaches. If he had been an average patient, he would have been simply diagnosed on the basis of history and treated with anti-migraine medications. But since he was a VIP, he got a brain scan “just in case.” The radiologist couldn’t say that the scan was 100% normal, so they did an invasive imaging procedure (an angiogram with dye) that caused a fatal stroke. In the general’s case, RHIP (rank has its privileges) became “rank has its perils.”
With today’s modern imaging technologies, this kind of thing is happening more and more often. Less than 1% of incidentalomas are significant. In the other 99% of cases, patients get an unnecessary and sometimes harmful sequence of tests, and get to worry over nothing. The cost is astronomical: $200 billion a year in the US, equal to one-quarter of the total health care budget.
In 1980, 3 million CT scans were done; today that has risen to 80 million a year. When Wruble started out as a radiologist, one abdominal CT scan consisted of 50 separate images; today it consists of 300 to 3000 images with much finer detail, more likely to show small abnormalities that would have been missed on earlier scans. Dr. Wruble says she finds 2 to 3 abnormalities on every scan; she says, “I never see a normal patient.” If everyone has them, doesn’t that mean incidentalomas are “normal” findings? Why should we worry about them?
What to do?
Doctors are trapped in a dilemma. If the radiologist’s report says “Probably not significant but can’t exclude malignancy,” what are they to do? Watchful waiting is statistically the best course. They can monitor the patient for new symptoms and repeat the imaging procedure periodically to see if the lesion is enlarging. But for most patients and for most doctors, that course is unacceptable. Doctors are afraid that if the patient dies, they will be blamed for not acting sooner. Patients don’t want to live with the fear that they might have a deadly disease that won’t be treated in time. Uncertainty is hard to live with, even when studies show it is better to accept uncertainty than to rush into treatment that is probably not necessary. It’s hard to apply statistics to your own case. The emotional reflex is to do everything possible, even when you know better. Dr. Wruble confesses that even she herself was not immune from those very human emotions: irrational fear once led her to insist her husband have a biopsy that he didn’t need.
We believe tests save lives when they really don’t
We seek certainty, and we are certain that these tests are life saving. But that certainty is a false belief. There are thousands of people out there who falsely believe imaging studies have saved their lives. Cancer will eventually manifest symptoms that will lead to diagnosis. In most cases, earlier diagnosis in the asymptomatic stage doesn’t lead to improved outcomes; it doesn’t improve quality of life or life expectancy. Early detection has not reduced the death rate from cancer, except possibly for colon, cervical, and lung cancer. Screening mammograms have increased the number of women diagnosed with breast cancer, but they have not decreased the death rate.
Carotid ultrasound screening finds obstructions that can lead to strokes, and patients undergo risky surgery and think it has saved their life by preventing a stroke, when the truth is that ultrasound screening does more harm than good and the United States Preventive Services Task Force (USPSTF) recommends against it. A 60-year-old man has a screening ultrasound test and is diagnosed with an abdominal aortic aneurysm (AAA); he has surgery, and he gives the test credit for saving his life. The USPSTF says he should not have been screened, because screening men at his age results in unnecessary surgeries and does more harm than good. He almost certainly would have survived just as long if he had never had the test. A patient is diagnosed with an early, localized prostate cancer that would have been best managed with watchful waiting; he opts for surgery that leaves him incontinent and impotent, and he is grateful because he believes it saved his life. It didn’t; it only made him suffer.
We do too many imaging studies
Ultrasounds and MRIs don’t involve radiation, but CT scans do. And radiation is harmful. Researchers tracked the CT histories of more than 31,000 patients[2]
- 33% had 5 or more lifetime CT scans
- 5% had more than 22 CT scans
- 1% had more than 38
- One patient had a whopping 138
- 15% had radiation doses higher than the exposure from 1000 chest x-rays
- 4% had lifetime exposures equivalent to 2500 chest x-rays
- 7% had an elevated risk of cancer because of radiation from CT scans
One study predicted that up to 2% of all US cancers are caused by CT radiation.[3]
By another estimate, the extra risk of one person dying of cancer from a typical CT procedure is 1 in 2000.[4] Children are at higher risk, and so are women.
Doctors order CT scans routinely for many conditions. Patients can ask whether a scan is absolutely necessary, whether the results will change the treatment, and what’s the worst that could happen if the scan is not done. I’ve refused tests when my doctors admitted that the results would not affect my treatment and that they mainly wanted to satisfy their curiosity.
Treatment of incidentalomas does more harm than good. But imaging procedures as a whole do more good than harm. What can we do to minimize the harm? We can be more judicious about ordering them, we can raise awareness, we can do research to figure out which incidentalomas require further investigation, and we can learn to live with uncertainty.
The American College of Radiology has established recommendations for managing various types of incidentalomas.[5] In many cases, they recommend ignoring them. I have a couple of those, a small cyst on my liver and a benign lesion on my kidney that don’t even merit any kind of follow-up. I’m happy to ignore them. They might be as well be called “insignificantomas.”
Imaging is good; incidentalomas are not
Modern diagnostic imaging allows us to see anatomical details in incredible detail in the living body. I was thrilled to watch the valves opening and closing in my own heart during an echocardiogram; when I was in medical school that would have been unimaginable, tantamount to magic. We seldom do autopsies any more (except for forensic reasons) because today we can diagnose things in the living that we could once diagnose only by cutting into the body after death.
The problem is that we have become too good at finding things, and we are finding things that would be better left unfound. Technology is getting away from us like the broom in The Sorcerer’s Apprentice. We need to tame that broom.
[1] https://www.youtube.com/watch?v=vrjx8ikME7Q
[2] http://pubs.rsna.org/doi/10.1148/radiol.2511081296?url_ver=Z39.88-2003&rfr_id=ori%3Arid%3Acrossref.org&rfr_dat=cr_pub%3Dpubmed&
[3] Brenner DJ, Hall EJ. Computed tomography: an increasing source of radiation exposure. N Engl J Med 2007;357:2277–2284
[4] http://www.fda.gov/Radiation-EmittingProducts/RadiationEmittingProductsandProcedures/MedicalImaging/MedicalX-Rays/ucm115329.htm
[5] http://www.jacr.org/article/S1546-1440(10)00330-3/fulltext#sec17
This article was originally published as a SkepDoc column in Skeptic magazine.
