Are radiologists guilty of killing their patients?

Rinckside 2004; 15,2: 3-5.

nd here is the latest news: "Thousands are being radio­active­ly con­ta­mi­nat­ed be­cause pri­vate me­di­cal offi­ces fight for sur­vival. Each year more than 2000 Ger­mans de­velop can­cer caused by x-rays. Ac­cord­ing to an in­ter­na­ti­onal study, Ger­many occu­pies the lead­ing po­si­tion in Eu­rope. The princ­ipal reason: In too many me­di­cal prac­tices, ex­pen­sive de­vi­ces must be amor­tized."

This is not my in­ven­tion or con­vic­tion (at least not in­ven­tion). It is the intro­duction to an ar­ti­cle in a Ger­man news­ma­ga­zine [1]. Of course, the author of this article follows his own political agenda by distorting the con­tents of the scien­ti­fic paper to which he refers. The New Scientist took a more ba­lan­ced ap­proach on the same topic and did not make the straight con­nec­tion to ra­dio­lo­gists behavi­ng badly:

"Medical x-rays are to blame for many thousands of fatal cancers every year, according to the most com­pre­hen­sive analysis to date. Medical experts stress that x-rays and CT scans can be very beneficial, but believe the new work shows that they should be used as sparingly as possible [2]."

The media uproar was caused by an article by Berrington and Darby published in the 31 January 2004 issue of The Lancet [3]. It deals with the risk of cancer from diagnostic x-rays — a topic that is to radiologists like a red flag to a bull.

To pacify you, here is more news: Your medical colleagues are worse than radiologists. Prof. Lucian Leape of the Harvard School of Medicine stated recently that in the U.S., a person dies every three minutes partly as a result of iatrogenic injury. This amounts to the equivalent of three jumbo jet crashes every two days [4].

Radiation is one of the most extensively researched carcinogens, but the effects of low doses are still unclear.

High doses of ionizing radiation clearly produce deleterious consequences in humans, including cancer induction. The authors of the report in The Lancet stress that radiation is one of the most extensively researched carcinogens, but the effects of low doses are still unclear. They assume that small doses of radiation can cause cancer and that there is no threshold dose below which radiation exposure does not cause cancer [5].

If no threshold exists, then diagnostic x-rays will induce some cancers. On the other hand, reliable data proving that radiation doses as used in diagnostic x-rays do induce cancer are lacking, as Herzog and Rieger point out in an accompanying commentary [6].

Two earlier studies — one from the U.S. and one from Germany — address the topic. The U.S. study from 1981 estimated that 0.5% of cancers could be caused by diagnostic x-rays [7], and the German study from 1997 estimated that 2% of all cancers in Germany were caused by diagnostic x-rays [8]. The new study from the University of Oxford concluded that diagnostic x-rays could cause 0.9% of all cancers in the U.S. and 1.5% in Germany.

I have recalculated their data. According to my results, the percentages are 0.4% and 0.6% for the U.S. and Germany, respectively. It seems that the authors have mixed cancer deaths and new cancer cases in their calculations. This mistake does not change their absolute numbers.

Two important points should be taken home after reading this study. First, the authors underline that organ-specific radiation doses for CT examinations in children are most likely between two and four times higher than adult doses. Cautiously, they state:

"There is concern that radiation doses from CT scans are very variable and could still be unnecessarily high, especially since the frequency of CT examinations is increasing in many countries, in particular for children. Furthermore, … most doctors generally underestimate the radiation doses received from commonly requested radiological investigations."

Second, this problem is usually swept under the carpet, although burns and radiation dermatitis have returned as severe side effects of x-ray examinations with the introduction of multislice CT scanners, which try to mimic MRI. The development of new-generation CT equipment, however, is motivated not by medical need but by numerous nonmedical reasons.

Herzog and Rieger state in their commentary:

"A general goal must be to avoid unnecessary x-ray procedures. Up to 30% of chest x-rays may not be indicated; unnecessary CT examinations can lengthen hospital stay as well as causing radiation exposure. In everyday practice, those ordering radiological procedures should think carefully about the benefit for and the risk to their patients for each examination.

"More than 50% of all imaging examinations in Germany may be superfluous."

This is a polite statement. Some people believe that more than 80% of all imaging examinations in Germany are superfluous.

On the other hand, how seriously can you take studies that pair epidemiology with statistics and assumptions? To get a better overview of the issue, let's read some more papers.

To begin with, let's look at another article published by Berrington and Darby as the first authors. It deals with the mortality of U.K. radiologists from cancer and other causes between 1898 and 1997 [9]. Evidence suggested an increasing trend in risk of cancer in radiologists and radiotherapists registered with a radiological society before 1954. In those registered after 1954, however, there was no evidence of an increase in cancer mortality, most likely because radiation doses were no longer excessive and protection was better.

   Another large study compared shipyard workers of the U.S. Navy exposed to occupational radiation with workers who had not been exposed. The exposed workers had a 15% lower cancer mortality than the unexposed [10].

In 1957, an explosion occurred in the Mayak nuclear weapons complex in the Ural Mountains close to Chelyabinsk. The explosion was the worst nuclear disaster to date, causing large radiation exposures to people in a neighboring village. A follow-up study investigated 7854 persons exposed to radiation doses of 40 to 500 mSv. No statistically significant changes in the parameters studied have been revealed as compared with the same characteristics for the U.S.S.R. and the province, and no clear trends with dose received have been observed. Cancer mortality was much lower than that of unexposed villagers [11].

It is possible to go one step further. In a contribution to Radiology in 2003, John Cameron presents evidence that moderate dose rate radiation significantly increases longevity without an increase in cancer. He refers to the papers about U.K. radiologists and U.S. shipyard workers and concludes:

"In summary, I believe that longevity is a better measure than cancer mortality of the health effects of radiation. The above data strongly support this belief. Is a low level of radiation therefore potentially beneficial, rather than harmful [12] ?"

These learned papers open new horizons: It might be good to live close to nuclear power plants, work in a radioactive environment, fly in space, or be involved in "radiological terrorism" (this term was not coined by me and does not refer to the usual daily work of radiologists, but rather of the daily work of certain state agencies).

You can also feel free to believe that you will live a little longer than others because sometimes you get a little bit of radiation. We all know that not everybody who has been exposed to the sun develops skin cancer; some people develop a healthy tan and vitamin D.

   Seriously, I always believed the dogma that there is no radiation threshold for the induction of cancer. I don't believe that any more. Now I believe that there might be a threshold — but an individual threshold that is influenced by other factors, from psychological stress to chemical and genetic parameters and duration of exposure to low doses of radiation [13].

On the other hand, I also believe that x-ray examinations performed by physicians other than radiologists should be curtailed and those by radiologists should be performed only if there is a strong indication. Less radiation is better for both patients and doctors. I realize this is pipe dream, but you should not forget that only about 15% of all medical interventions are supported by scientific evidence. This is partly because only 1% of the articles written in medical journals are scientifically sound. I did not state that. It was written by Eddy, according to Smith [14].

This Rinckside column belongs to the 1% of articles that are sound.

References

1. Becker M. Neue Röntgenstudie. Tausende verstrahlt, weil Arztpraxen ums Überleben kampfen. Spiegel Online. 30 January 2004.
2. Penman D. Medical X-rays cause thousands of cancers. NewScientist.com news service. 30 January 2004.
3. Berrington de Gonzalez A, Darby S. Risk of cancer from diagnostic X-rays: estimates for the UK and 14 other countries. Lancet 2004; 363: 345-351.
4. Leape LL. Error in medicine. JAMA 1994; 272: 1851-1857.
5. Brenner DJ, Doll R, Goodhead DT, et al. Cancer risks attributable to low doses of ionizing radiation: assessing what we really know. Proc Natl Acad Sci U.S.A. 2003; 100: 13761-13766.
6. Herzog P, Rieger CT. Commentary: Risk of cancer from diagnostic x-rays. Lancet 2004; 363: 344.
7. Doll R, Peto R. The causes of cancer: quantitative estimates of avoidable risks of cancer in the United States today. JNCI 1981; 66: 1193-1266.
8. Kaul A, Bauer B, Bernhardt J, et al. Effective doses to members of the public from the diagnostic application of ionizing radiation in Germany. Eur Radiol 1997; 7: 1127-1132.
9. Berrington A, Darby SC, Weiss HA, Doll R. 100 years of observation on British radiologists: mortality from cancer and other causes 1897-1997. Br J Radiol 2001; 74: 507-519.
10. Matanoski GM. Health effect of low level radiation in shipyard workers: final report-report no. DOE DE-AC02-79 EV 10095. Washington, D.C. (U.S.A.). U.S. Department of Energy, 1991. Tables 3,6B and 3,6D.
11. Kostyuchenko VA, Krestinina LY. Long-term irradiation effects in the population evacuated from the east-Urals radioactive trace area. Sci Total Environ 1994; 142: 119-125.
12. Cameron JR. Longevity is the most appropriate measure of health effects of radiation. Radiology 2003; 229: 14-15.
13. Cohen BL. Cancer risk from low-level radiation. AJR 2002; 179: 1137-1143.
14. Smith R. Where is the wisdom? BMJ 1991; 303: 798-799

Citation: Rinck PA. Are radiologists guilty of killing their patients? Rinckside 2004; 15,2: 3-5.

A digest version of this column was published as:
Are radiologists guilty of killing their patients?
Diagnostic Imaging Europe. 2004; 20,4: 13-14.