The Wall Street Journal reports on a recent New England Journal of Medicine study which concludes that doctors are over-utilizing CT scans, exposing their patients to excessive, and potentially harmful, radiation doses:
Doctors are ordering too many unnecessary diagnostic CT scans, exposing their patients to potentially dangerous levels of radiation that could increase their risk of cancer, according to Columbia University researchers.
The researchers, writing in this week’s New England Journal of Medicine, conclude that in the coming decades up to 2% of all cancers in the United States may be caused by radiation from computed tomography scans performed now. Children face the most danger, they said.
In ordering CT scans, doctors are underestimating the radiation danger … In many cases, the researchers say, older technologies like X-rays and ultrasound that expose patients to lower radiation doses or no radiation at all would work just as well.
Since CT scans were introduced in the 1970s, their use has grown to an estimated 62 million annually. An estimated four million to five million scans are ordered for children, Mr. Brenner said. Adults receive scans for diseases of the stomach, colon, breast and other areas. Children most often are scanned for appendicitis. It has become a favored technology because it provides detailed information about patients’ bodies, is noninvasive and typically is covered by health insurance.
While the scans save lives, the authors say, doctors are leaning on them over safer diagnostic tools because they underestimate the levels of radiation people receive from the scans.
The authors measured typical levels of radiation that CT scans emit. They found levels they say were comparable to that received by some people miles from the epicenters of the 1945 atomic blasts over Hiroshima and Nagasaki, Japan.
There can be little doubt that CAT scans, as well as other expensive medical imaging studies, are overutilized in medicine today. There is also no doubt that the overutilization of CAT scans in particular, with their ionizing radiation, does expose patients to significantly more radiation. It may be worthwhile to pause and think about why so many CAT scans are being performed.
Hint: It’s not because doctors don’t know that CAT scans deliver more radiation.
Now, to be sure, there are no doubt some dim bulbs in the medical profession who don’t realize that a CAT scan delivers quite a bit of radiation. In fact, it delivers about 500 times as much as a chest x-ray. (To keep this in perspective, a chest x-ray uses an extremely small amount of radiation because the chest is mostly air). And I’ve seen my share of abuses — like the physician whose daughter underwent 3 CT scans in a week to monitor her small kidney stone, which was going to pass no matter what. But in reality, the physicians ordering these studies generally recognize that CT scans expose their patients to higher radiation than other studies. So why are so many of these studies performed?
Well, the easy answer is this: CT scans are an excellent diagnostic tool. The scanner works by taking multiple cross-sectional x-rays at slightly different angles, like spokes of a wheel. Computer software then generates an image with extraordinary detail of the internal organs, which can be digitally manipulated to enhance contrast, measure density, or view internal structures and organs from different perspectives. Three-dimensional viewing is even possible. While they are not capable of diagnosing every serious problem, they are excellent at detailing anatomical changes and identifying potentially serious pathology.
When I was a surgical resident, before the regular use of CT scans (and yes, I am dating myself — we also wrote our progress notes on papyrus), standard surgical teaching on acute appendicitis and appendectomy was this: If 10% of a general surgeon’s appendixes removed at surgery were not normal, he or she was underdiagnosing appendicitis. Since the consequences of missing appendicitis can be dire, surgeons would err on the side of operating too frequently rather than not frequently enough.
The introduction of CT scanning has eliminated this calculus, as appendicitis can be ruled out with near 100% confidence on CT scans. Similarly, the frequent use of CT scanning for diagnosis of other abdominal problems often leads to incidental findings of serious problems which would have gone undiagnosed. I have personally in the past year operated on 4 or 5 renal cancers which were far too small to be detected by any other means, and which were totally asymptomatic. Without CT many of these would have had the potential to spread and be potentially lethal.
While it is certainly true that in some circumstances alternative imaging studies, such as ultrasound, may be a safer and entirely appropriate choice, these other studies often have significant limitations. Ultrasound simply cannot provide the level of anatomic detail which CT provides, although it does not use any ionizing radiation and is much less expensive to perform. MRI scans, which use high-power magnetic fields, can provide similar quality imaging, but are substantially more expensive than CT scans, and cannot be used in certain patients, such as those who have pacemakers or metallic prostheses.
So the increasing frequency of CT scanning is due in no small part to its excellence as a diagnostic tool. But there is another, very large elephant sitting on the couch, which is not mentioned in this study. That large, gray animal with floppy ears and a long nose, sitting on your finest Chippendale while snarfing down peanuts, is the threat of medical malpractice lawsuits.
Far and away the most common reason for the marked increasing utilization of CAT scans and other expensive and potentially risky studies is the perceived threat of a medical malpractice lawsuit. Medical economists and social engineers love to point out the statistical rareness of certain diseases, and thereby castigate the health care system for recommending such expensive and potentially risky tests in many patients at low risk for them. These same facts and statistics are well-known to most physicians — the medical literature is replete with them. Performing a CAT scan on many patients who present with chronic abdominal pain, for example, is genuinely unnecessary in a substantial number of these patients, as the yield is very low.
However, a physician need miss only one ruptured appendix, or one abdominal cancer, or one impending ruptured aortic aneurysm, to find themselves in a multimillion-dollar lawsuit which is indefensible. Every physician, when he sees such a patient, hears in the depths of his subconscious the cross-examination of a plaintiffs lawyer, asking him why he neglected to perform such a simple diagnostic test which would have saved his client’s life or prevented an extensive and complicated illness. I assure you that this tape plays in the mind of virtually every physician practicing medicine today, and this mental image inculcates the philosophy that one can never go wrong by ordering another test. Perhaps we should rename CAT scans, “CYA scans.”
Much like a stockbroker who makes money whether the market is gaining or losing, the trial lawyers will make money whether physicians are doing too many CT scans or not enough. The study cited by the Wall Street Journal will likely provide additional fodder for even more lawsuits. If you are medically careful and financially prudent, and do not order the “unnecessary” follow-up CAT scan (after the first one was normal), you will be sued when your patient is later found to have a cancer. If you perform too many CAT scans, you will be sued because your CT scans caused the cancer — complete with theatrical courtroom scenes about physicians inflicting the effects of “Nagasaki and Hiroshima” on their injured client. Heads you win, tails I lose.
And do excess CT scans really increase the risk of cancer by 2%? In point of fact, estimating the risk of increased cancer from diagnostic imaging studies is a “science” not unlike that behind global warming: much statistical sleight-of-hand with little hard evidence for real-life risk. It is certainly true that high doses of radiation can cause cancer; the incidence of leukemia and thyroid cancer was markedly higher in those exposed to nuclear weapons than in the general population. But it is impossible to extrapolate that to diagnostic imaging studies which are far more limited in body parts exposed and tightly controlled in dosing, and which use far less dangerous radioisotopes for visualization. Nevertheless, this kind of junk science statistical analysis will no doubt become “fact”, replayed endlessly by the media, consumer groups — and yes, trial attorneys.
As much as we would like to believe that medicine is a pure science, and that all medical decisions can be reduced to statistical risks and cost-benefit analysis, the fact remains that medicine is a human undertaking — both in its practitioners and those who are its beneficiaries. There is a constant tension between risk and benefit — and a million variables which influence outcomes. This delicate balancing act is wildly distorted by an out-of-control tort system which, despite its marketing blather about “standing up for patients,” in fact harms them far more than it helps.