TCS Daily

Unnecessary X-Rays Should Be Avoided (But Which are Unnecessary?)

By Gilbert Ross - December 6, 2007 12:00 AM

American consumers should not be alarmed or confused by two recent studies that call our attention to the difficulty in deciding who should -- and who should not -- be exposed to radiation from medical X-rays. In fact, all you really need to sort through this problem is a dose of common sense.

As with much medical decision-making, the final analysis involves a delicate weighing of benefits -- the diagnostic value -- of a particular treatment plan against the risks. In this instance, the risk is of marginally increased radiation exposure. The risk-benefit equation when deciding to order (or have) a diagnostic CT scan is even more nuanced than many other medical questions, as the potential benefits are immediate, while the dangers lie in the distant future.

Recently, researchers at the VA Outcomes Group in Vermont and the Dartmouth Medical School, led by Dr. Gilbert Welch, published a report provocatively titled "Overstating the Evidence for Lung Cancer Screening" in the Archives of Internal Medicine. The authors took strong issue with an earlier New England Journal of Medicine paper advocating the wide use of spiral CT scan screening for smokers to detect lung cancer earlier.

The authors of the new study pointed out four specific reasons the "obvious" rationale for such screening would, in fact, save few lives while drastically increasing costs for the health care system.

Worse, the detection of minor and non-progressive abnormalities would inevitably lead to biopsies and other surgical procedures, in effect resulting in a net harm to the population screened. Further, they remind us that we should give due consideration to the terrible anxiety invoked by finding such "false positives." These expert epidemiologists strongly advocate reliance on controlled studies, which are now in progress, using non-scanned cohorts as a baseline to truly gauge the real lifesaving benefits -- if any -- of CT scanning of all smokers (not to mention former smokers).

Recently the NEJM published an overview of CT scanning overuse and its potential dangers, authored by experts in radiology at the Columbia University Medical Center in New York. Their analysis used paradigms based on the rate of cancer in Japanese A-bomb survivors to show that "excessive" exposure to radiation from CT scans might lead to a marked upsurge in radiation-induced cancers over the course of the next thirty years or so. They go so far as to estimate a worst- case scenario of X-ray-induced cancers comprising up to 2% of all cancers in some vaguely-described future decade. (Some experts stated that this is a gross over-estimate of the actual risks.)

The authors point, especially, to emergency room physicians, who -- faced with urgent and overcrowded conditions -- sometimes give in to their clientele's demands for rapid diagnosis. Many unnecessary CT scans are ordered in the ER, the Columbia authors claim. The real problem, of course, is that the labeling of a CT scan as unnecessary is often only possible using that favorite tool of Monday-morning quarterbacks: the "20/20 retroscope." When working "in the trenches" -- especially in a busy emergency room -- the decision to scan or not is often a difficult but crucial one.

The radiologists, in their NEJM article, recommend three simple measures to address the problem: reduce the radiation dose per CT; replace CT scanning with tests requiring less, or no, radiation, such as sonograms or MRIs; and, most important, don't order CTs when they are not really necessary. They estimate that one-third of all CT scans are unnecessary. If that's true, then 20+ million adults and over one-million children are being radiated unnecessarily.

So how afraid of CT scans should we be? The simple dictum applies: if a doctor believes that you need a CT scan, by all means get one. The benefit-risk equilibrium is strongly -- I'd say completely -- dictated by the medical necessity of getting the test. If a substantial diagnostic dilemma exists, or a need to evaluate the results of a therapy, don't worry about the small dose of radiation involved in one CT scan. The concern should kick in when considering a "whole-body" scan for no particular reason -- or if you are part of that small segment of the population whose chronic diseases warrant repetitive scanning (the danger is cumulative, so it's the X-rays absorbed over one's lifetime that really matter). It is true that occasionally a "screening" CT scan may reveal a hidden danger that can be treated in time to avert disaster. However, all too often, an insignificant nodule or calcification (an "incidentaloma," in medicalese) will lead down the path of unnecessary surgery and its consequences. Often, a different type of diagnostic test using no radiation would suffice -- e.g., using ultrasound to detect a potentially lethal aneurysm in the abdominal or thoracic aorta.

The age-old dictum doctors strive to live by still applies here: first, do no harm. If a test has a very small chance of helping a patient, it is likely to do more harm than good. And for the public, my advice would be: when you see a solicitation for a total body scan, you should, so to speak, see right through such marketing ploys.

Gilbert Ross, M.D., is Executive and Medical Director of the American Council on Science and Health (,


The lesson here
is that "expert epidemiologists" have utterly no business in being allowed anywhere near the clinical treatment of patients. Moreover, the reference used by the epidemiologists, namely the Hiroshima survivors, is utterly irrelevant, because it was a large, prompt dose as opposed to the chronic repetition of CT scans of which they complain. Finally, if the epidemiologists based their claims of harm on LNT then they again have no case because low dose based on LNT has never been shown to have any effect whatsoever.

The lesson is
The patient must be informed and take an active role in their personal health care. I have had doctors get angry because, as a scientist, I do not take advice without question.

The only reason Nationalized Health Care is a thought is so few people do.

TCS Daily Archives