Sunday, February 2, 2014

Are we really giving ourselves cancer, as the New York Times Op Ed claims?

Line-by-line criticism of the Op-Ed in New York Times entitled 
"We Are Giving Ourselves Cancer"

The lines from Op-Ed are gives in quotes "   ".
My comments are in boldface.

The Title of the Op-Ed:
 "We Are Giving Ourselves Cancer"
Really?
This article does not represent the current state of knowledge in this field.  The concerns raised by the authors regarding CT scans are based on the concept called the linear no-threshold (LNT) model according to which even the smallest amount of radiation (since there is no-threshold) can increase the mutations and the risk of cancer.  The current state of knowledge on the LNT model can be assessed by reading recent peer-reviewed publications on the subject.  If you search for "Radiation LNT Model" in Pubmed, the vast majority of recent publications have questioned the validity of the LNT model giving many reasons, and showing evidence for the opposite of the LNT model, i.e. the radiation hormesis model,  according to which small amounts of radiation can boost the defenses in our body including the immune system reducing cancers and other diseases.  The scientists who advocate the LNT model routinely avoid discussing any of the ideas expressed in such articles, and have not refuted the arguments presented.  The present Op-Ed article does the same, making claims of low dose radiation cancer risks using older articles whose foundations have been discredited.  The misleading information presented in this article, since it has been given widespread coverage by a leading newspaper, may  have severe consequences to members of the public who refuse to have CT scans based on such articles when the CT scans are prescribed for proper diagnosis of their conditions.

Op Ed: "By RITA F. REDBERG and REBECCA SMITH-BINDMAN JAN. 30, 2014

DESPITE great strides in prevention and treatment, cancer rates remain stubbornly high and may soon surpass heart disease as the leading cause of death in the United States."

    There have not been great strides in prevention of cancer.  There have indeed been great improvements in early detection and treatment of cancer.   

    Cancer rates being stubbornly high is because we (the scientific community) have not understood the primary cause of cancer, and so we have not found a way of preventing it. 


Op Ed: "Increasingly, we and many other experts believe that an important culprit may be our own medical practices: We are silently irradiating ourselves to death."

    The above statements are patently false. Assuming the authors' estimate below of 3-5% cancers due to CT scans is correct, to blame CT scans for the cancer rates being stubbornly high doesn't make any sense, as the CT scans contribute a maximum of only 5% of the cancers. Their implication is: if we remove these 3-5% excess cancers, the cancer rates would not be stubbornly high. Not true.

     Re: "many other experts believe":  I wonder which other experts believe the culprit for the stubbornly high cancer rates is CT scans.  How many cancers have been found by CT scans and treated to extend people's lives? How about the diagnoses of all the other illnesses and conditions using CT scans? Aren't these diagnoses without exploratory surgeries saving lives?  Is the net result of all the CT scans increased deaths? 


Op Ed: "The use of medical imaging with high-dose radiation — CT scans in particular — has soared in the last 20 years."

     Do CT scans use high-dose radiation?  A CT scan typically gives radiation dose of about 1 cSv.  In some areas of the world, people receive such an amount of radiation from natural causes every year, and no illnesses or increased cancers have been observed in studies of such populations.  Animal studies have shown repeated CT scans result in reduced mutations.  CT scans do not result in high-dose radiation exposure.  CT scans give low-dose radiation exposures.  

     As the public and physicians have found the CT scans to be beneficial for improved diagnosis and health, the use of CT scans has soared.   Nothing wrong with it.


Op Ed: "Our resulting exposure to medical radiation has increased more than sixfold between the 1980s and 2006, according to the National Council on Radiation Protection & Measurements."
     Of course.  It is a direct consequence of increased use of CT scans.


Op Ed: "The radiation doses of CT scans (a series of X-ray images from multiple angles) are 100 to 1,000 times higher than conventional X-rays."
     Why is conventional X-ray a benchmark, and why does it matter that CT scans are 1000 times higher than X-rays?  What if it were 10000 times higher or 10 times higher?

Op Ed: "Of course, early diagnosis thanks to medical imaging can be lifesaving."
     Not always true, as there can be overdiagnosis, which can result in harming the patient from the unneeded treatments.

Op Ed: "But there is distressingly little evidence of better health outcomes associated with the current high rate of scans."
     There may be overuse of diagnostic CT scans, but that needs to be reduced for cost and resource reasons, not by making false claims of cancer risk from CT scans.

Op Ed: "There is, however, evidence of its harms."
    Where is the evidence of its harms? Animal studies have shown repeated CT scans reduce DNA damage/ double strand breaks:   http://www.ncbi.nlm.nih.gov/pubmed/22059980
    Radiation dose equivalent to several CT scans has led to reduced breast cancers, as seen Miller's data shown below.

Op Ed: "The relationship between radiation and the development of cancer is well understood:"
     For high doses of radiation, greater than the equivalent of several tens of CT scans per year, there is evidence for increased cancers.  For low doses of radiation, there is in fact increasing evidence in recent years for its cancer-preventive effect.  The official view on low dose radiation as well as public perception  is that it increases the risk of cancer.  Since this is contrary to experimental evidence, I would not consider the radiation effects to be well understood by the scientific community or the public.

Op Ed: "A single CT scan exposes a patient to the amount of radiation that epidemiologic evidence shows can be cancer-causing."
     Where is the evidence?  There are plenty of evidence that shows otherwise.  See my blogs and the referenced articles, for example, for contradicting evidence: Atomic bomb survivor data no longer support this claim  http://lss-14-report-analysis.blogspot.com/   The 15-country study of radiation worker study also no longer shows increased risk of cancers from low dose radiation http://15-country-study-rad-workers-analysis.blogspot.com/    Taiwan apt study shows reduction in cancers from low dose radiation http://taiwan-apt-cancer-data-analysis.blogspot.com/

Op Ed: "The risks have been demonstrated directly in two large clinical studies in Britain and Australia. In the British study, children exposed to multiple CT scans were found to be three times more likely to develop leukemia and brain cancer."
     There is a major problem of reverse causation in these two studies, i.e. their illnesses were the reason they had increased number of CT scans. Using the argument of these publications, we could rightly claim hospitals are the reason for deaths of most people because the chance of dying is higher in hospitals compared to those who are not in hospitals.  Not true.  Association is not causation. 
 Op Ed: "In a 2011 report sponsored by Susan G. Komen, the Institute of Medicine concluded that radiation from medical imaging, and hormone therapy, the use of which has substantially declined in the last decade, were the leading environmental causes of breast cancer, and advised that women reduce their exposure to unnecessary CT scans."
     This report refers to IARC 2011 Report.  The IARC report refers to articles on second cancers following radiation therapy, but these relate to high dose radiation.  The report also refers to Ronckers 2008 which shows:


     This data does not show increased risk of cancer from 1 cGy of radiation (1 CT scan) but only for much higher doses.  Miller et al. Canadian study of breast cancers in TB patients, with much better statistics, shows a decrease in breast cancers for the equivalent of several CT scans.



Thus, the warnings of the 2011 report sponsored by Susan G. Komen about radiation dose from medical imaging does not have any justification from data.

Op Ed: "CTs, once rare, are now routine. One in 10 Americans undergo a CT scan every year, and many of them get more than one. This growth is a result of multiple factors, including a desire for early diagnoses, higher quality imaging technology,"
These are good reasons for the growth in the number of CT scans.


Op Ed: "direct-to-consumer advertising and the financial interests of doctors and imaging centers. CT scanners cost millions of dollars; having made that investment, purchasers are strongly incentivized to use them."
         These reasons for growth in CT scans should be discouraged by making changes in the present system of healthcare.  CT scans or other testing should only be done to improve the health of the patients, not for financial reasons.

Op Ed: "While it is difficult to know how many cancers will result from medical imaging,
          Because none are caused by medical imaging.

Op Ed: "a 2009 study from the National Cancer Institute estimates that CT scans conducted in 2007 will cause a projected 29,000 excess cancer cases and 14,500 excess deaths over the lifetime of those exposed. Given the many scans performed over the last several years, a reasonable estimate of excess lifetime cancers would be in the hundreds of thousands. According to our calculations, unless we change our current practices, 3 percent to 5 percent of all future cancers may result from exposure to medical imaging."
     This type of estimate is based on the BEIR VII report model based on the LNT model.  BEIR VII report based its model on atomic bomb survivor data.  The updated data for atomic bomb survivors does not show evidence for LNT model.  BEIR VII report also claimed risk of cancer from low dose radiation based on the 15 country radiation worker study, the conclusions of which have been discredited.  There is no credible study that shows increased risk of cancers from low dose radiation.   



Op Ed: "We know that these tests are overused."
     If the tests are overused, such overuse should be prevented through changes to the current system of healthcare.  

Op Ed: "But even when they are appropriately used, they are not always done in the safest ways possible."
   The authors need to show some credible evidence that radiation at the level of a few CT scans is causing cancer.  Without such evidence, the above statement is simply incorrect.

Op Ed: "The rule is that doses for medical imaging should be as low as reasonably achievable.
      This is not a rule.  The as low as reasonably achievable (ALARA) was based on the misguided concept of LNT model.  Since LNT model is invalid, the ALARA concept has no merit.

Op Ed: "But there are no specific guidelines for what these doses are, and thus there is considerable variation within and between institutions. The dose at one hospital can be as much as 50 times stronger than at another."
     So what?  There may be different machines of different ages and technologies which would justify the differences in doses to obtain good quality images.  There may be patient weight differences that may require different doses.      

Op Ed: "A recent study at one New York hospital found that nearly a third of its patients undergoing multiple cardiac imaging tests were getting a cumulative effective dose of more than 100 millisieverts of radiation — equivalent to 5,000 chest X-rays."
     Why compare to X-rays?  What if the dose is equivalent to 100 chest X-rays?  100 millisieverts has shown a cancer preventive effect in many studies.  This much of radiation dose is not a concern.

Op Ed: "And last year, a survey of nuclear cardiologists found that only 7 percent of stress tests were done using a “stress first” protocol (examining an image of the heart after exercise before deciding whether it was necessary to take one of it at rest), which can decrease radiation exposure by up to 75 percent."
     Unnecessary testing should be avoided, from the point of view of costs, and for judicious use of resources.  No need to mention radiation exposure as it has no relevance to health at these low levels.

Op Ed: "In recent years, the medical profession has made some progress on these issues. The American College of Radiology and the American College of Cardiology have issued “appropriateness criteria” to help doctors consider the risks and benefits before ordering a test. "
     The professional organizations take their cue from reports from advisory bodies such as BEIR VII report in evaluating the risk of low dose radiation.  BEIR VII report based its cancer risk estimates on atomic bomb survivors and claimed support for the concept of low dose radiation cancer risk from the 15 country study of radiation workers.  The updated data for these no longer show evidence for cancer risk at low doses (see above).  Search on Google or Pubmed for LNT model, and you will see plenty of evidence against the LNT model.  There have been many articles showing invalidity of LNT model, and the supporters of LNT model have not responded to such articles with rebuttals.  Advisory bodies such as NAS that produced BEIR VII report have backed the LNT model for so long that they would have no credibility if they reverse their stand now.  
      Since the recommendations of ACR and ACC are based on the reports like BEIR VII that have been discredited, such recommendations do not have any validity.

Op Ed: "And the insurance industry has started using radiology benefit managers, who investigate whether an imaging test is necessary before authorizing payment for it. Some studies have shown that the use of medical imaging has begun to slow.

But we still have a long way to go."
     Using insurance company bureaucrats to deny coverage of studies ordered by physicians is a wasteful interference in the healthcare of the patient.   Changes should be made in the healthcare system so that the doctors prescribe tests only when necessary, by eliminating the profit motive.

Op Ed: "Fortunately, we can reduce the rate of medical imaging by simply avoiding unnecessary scans and minimizing the radiation from appropriate ones."
     The second part of the statement is simply wrong.  We cannot reduce the rate of medical imaging by minimizing the radiation from appropriate scans.  
          Op Ed: "For example, emergency room physicians routinely order multiple CT scans even before meeting a patient."
    Many emergency room physicians have contested this statement, that they routinely order multiple CT scans even before meeting a patient.  If the physicians know the history of the patient before their arrival through phone communication, etc., CT scans can indeed be justifiably ordered beforehand, to speed the diagnosis and emergency treatment when seconds can be important.  

Op Ed: "Such practices, for which there is little or no evidence of benefit, should be eliminated."
      There are benefits, like saving lives.  Seconds count in emergencies. 


Op Ed: "Better monitoring and guidelines would also help. The Food and Drug Administration oversees the approval of scanners, but does not have regulatory oversight for how they are used. We need clear standards, published by professional radiology societies or organizations like the Joint Commission or the F.D.A. In order to be accredited for CT scans, hospitals and imaging clinics should be required to track the doses they use and ensure that they are truly as low as possible by comparing them to published guidelines."
     When doses are reduced, there is a risk of having poorer image quality and misdiagnosis.  This may not happen for the average patient, but in a small fraction of patients this would happen.  For what percentage of patients are the authors willing to accept misdiagnosis?  Since there is no evidence for cancers from doses equivalent to CT scans, there is no justification for ALARA, as that would harm patients' health. 

Op Ed: "Patients have a part to play as well. Consumers can go to the Choosing Wisely website to learn about the most commonly overused tests. Before agreeing to a CT scan, they should ask: Will it lead to a better treatment and outcome?"
     This may not be a fair question to ask, as the doctor would not be able to guarantee a better outcome to an individual patient based on a CT scan or other testing. Patients should ask if the CT scan is likely to improve their treatment and outcome.

Op Ed: "Would they get that therapy without the test?"
     It would be valid to ask if the results of the CT scan would affect the course of treatment.  If the CT scan would not have any effect on the treatment, there would be no benefit from the scan, and the scan can be avoided.  This is true for any test, not just CT scans.  

Op Ed: "Are there alternatives that don’t involve radiation, like ultrasound or M.R.I.? When a CT scan is necessary, how can radiation exposure be minimized?"
     These questions would be valid if CT scans were shown to be causing cancers.  In the absence of any such evidence, these questions are of no use.

Op Ed: "Neither doctors nor patients want to return to the days before CT scans. But we need to find ways to use them without killing people in the process."
     Pretty strong statement, claiming we (the staff in diagnostic imaging) are killing patients with CT scans in the process of diagnosis, without a shred of evidence.  

Op Ed: "Rita F. Redberg is a cardiologist at the University of California, San Francisco Medical Center, where Rebecca Smith-Bindman is a radiologist. "
      Surprised to see such an article from a respected University Hospital. 

Disclaimer: The opinions expressed in this blog are my personal, professional opinion.

                                       Mohan Doss
                                       Medical Physicist, Diagnostic Imaging
                                                          Email: mohandoss99@gmail.com