Dr. Len's Cancer Blog

Expert perspective, insight and discussion

Dr. Len's Cancer Blog

The American Cancer Society

CT Scans and Cancer Risk: Been There, Done That

by Dr. Len December 15, 2009

Two articles and an editorial in this week’s issue of the Archives of Internal Medicine should give all of us pause about the potential risk of increased cancer cases and deaths caused by the overuse and inappropriate use of CT scans.


According to this research, it is possible that 1-2% of cancer deaths every year in the future may be due to a cancer caused by a CT scan performed years previously.  In addition, the researchers found that the amount of radiation per CT scan differed substantially for the same type of scan performed on different machines in the same or other institutions.


For me, this is not some abstract discussion.  Two years ago I had to decide between getting several CT scans recommended by a radiologist or avoid the scans and take the chance I had a serious disease that might progress undetected.  I decided against the scans for the very reasons noted by these researchers. 


Looking back, it was clearly the right decision.


In the first study as reported in the Archives, the authors determined the number of CT scans performed currently in the United States. Then, they estimated the number of cancers that could eventually result from these scans. 


They estimated that 72 million scans were performed in this country in 2007.  The highest cancer risks were for chest or abdomen CT angiography (a study looking at blood vessels in the heart or aorta, which is a large blood vessel in the chest and abdomen), and whole body CT scans. 


The researchers also noted that the risk of cancer caused by CT scans declined as the patients got older.


They came to the conclusions that approximately 29,000 future cancers could be related to CT scans performed in 2007.  Most of this risk would come from the scans performed most often, namely CT scans of the chest, abdomen and pelvis and head, as well as CT angiography of the chest which looks for coronary artery disease.


1/3 of these projected cancers would come from scans performed in people between the ages of 35 and 54 years old, while an additional 15% were from scans performed before the age 18.  The most common cancers were lung cancer, followed by colon cancer and leukemia.


The second paper in the Archives took a close look at the actual amount of radiation that was received by patients who had CT scans at four San Francisco area hospitals. 


What was troubling about this study was the fact that the authors found essentially no standardization for the way the CT scans were done, resulting in wildly different radiation doses for the different types of scans performed and depending on where they were performed.


For example, the radiation dose for a CT angiogram of the heart was almost 3 times greater than for a routine CT scan of the chest.  The radiation dose was 7 times greater for a patient who had a CT scan of the head to look for as stroke as opposed to a routine head CT scan for other causes.


The researchers also found an average 13 times variation between the highest and lowest radiation exposures for each type of CT study they examined.  This difference occurred not only between different hospitals, but also within the same hospital.


Then there is the question of how many CT scans it would take to cause one additional cancer to develop in the future. 


For 40 year old women who had CT angiograms, that number is 270.  For those same 40 year old women who had head CT scans, there would be one additional cancer caused at some time in the future for every 8105 women who were scanned.  The authors also estimate that for a 20 year old woman who needed a CT scan for a possible pulmonary embolism (blood clot in the lung), a CT coronary angiogram, or a CT scan of the abdomen and pelvis, the risk of developing a cancer in the future as a result of the CT scan could be as high as 1 in 80.


There is a comment in the article that I think is worthy of highlighting:


“CT is generally considered to have a very favorable risk to benefit profile among symptomatic patients.  However, the threshold for using CT has declined so that it is no longer used only in very sick patients but also in those with mild, self-limited illnesses who are otherwise healthy.  In these patients, the value of CT needs to be balanced against this small but real risk of carcinogenesis resulting from its use.  Neither physicians nor patients are generally aware of the radiation associated with CT, its risk of carcinogenesis, or the importance of limiting exposure among younger patients,   It is important to make both physicians and patients aware that this risk exists.” (emphasis mine)


These researchers also call on the profession to adopt and put in place standards similar to those developed by the Food and Drug Administration to monitor the performance of mammography machines to assure patients and physicians that the doses being used are in fact the correct and lowest dose needed for the CT scan.  There is currently no regulation of CT scans “in the field” at this time by the FDA.


In the editorial that accompanies these papers, the author points out that every day there are 19,500 CT scans performed in the United States, which subjects patients to a radiation dose equal to anywhere from 30 to 442 chest x-rays.  Also, 70% of adults in this country (including me) had a CT scan between 2005 and 2007. 2% of these patients received high to very high doses of radiation from their CT scan.


The editorialist goes on to write:


“A popular current paradigm for health care presumes that more information, more testing and more technology inevitably leads to better care.  (These studies) counsel a reexamination of that paradigm for nuclear imaging.  In addition, it is certain that a significant number of CT scans are not appropriate.  A recent Government Accountability Office report on medical imaging, for example, found an 8-fold variation between states on expenditures for in-office medical imaging; given the lack of data indicating that patients do better in states with more imaging and given the highly profitable nature of diagnostic imaging, the wide variation suggest that there may be significant overuse in parts of the country.”


I can recall a day when CT scans were actually hard to get.  Now, everyone has one—including many doctors and practices in their own private offices. 


CT scans have become the new chest x-ray.  They have replaced the history and physical.  They have become the “defensive medicine fallback,” since doctors tell me frequently that they have to get the scan to protect themselves on the very outside possibility that—for example—the patient with a headache may have a brain tumor, or the pneumonia may be caused by a cancer.


And then there was my own experience with the benefits/risks “equation” of getting a CT scan.


Two years ago—at the urging of my wife (who is a doctor) and my physician--I had a chest CT to look at the amount of calcium in my coronary arteries.  Given my underlying medical problems, which include hypertension and elevated cholesterol as well as a reasonably stressful job (which I love, by the way—it’s the travel that sometimes becomes a bit too much), they thought that even though I had no symptoms of heart disease and was reasonably physically fit, I should have my arteries checked.  (The scan was cheap, by the way—costing about $150.  The hospital had recently discounted the price from the original quote of $200, which was considerably less than the $1400 they subsequently charged me for a routine follow-up chest CT.)


The good news was that there was no calcium in the arteries.  But there was a very small lesion in my chest which did not have any calcium, and which could have been a very early lung cancer.


Never mind that the medical literature suggests that these types of lesions are very common in people like me, especially those who live in the South. Never mind that when seen on a routine chest CT in a non-smoker they are rarely if ever a cancer. 


None of that mattered.  The radiologist recommended serial CT scans with intravenous contrast every 6 months for two years.  I did get the first follow-up scan at six months—without the contrast—and everything was stable. 


I finally took my own health into my own hands and said “No more!”  I knew the research data, knew the recommendations of the experts, and had discussions with other radiologists who were familiar with the literature.  I concluded that my risk of getting cancer from the scans was greater than the risk of having lung cancer in that nodule. 


Two years later, and still no problem.


I guess the message of my own experience was that I took responsibility for my own health.  But let’s face a little reality here: I am a doctor who happens to work with experts who know about these things.  It was hard to beat having access to the “best in the world” when it came to making that decision. 


My problem is that too often doctors don’t know their patients, don’t have time for a conversation about the benefits, indications and risks of a particular CT scan, and feel they will be sued even if they miss something—even if the chance of that “something” is minimal at best.  They don’t have the time or the inclination to have a conversation that might outline an alternative path consistent with reasonable medical judgment (like, “here are the things you need to know and need to do if this or that happens after you leave my office”).  It’s a lot simpler to just go ahead and order the CT.  (And, if they happen to own the machine and can be paid by the insurance company, the decision gets even easier.)


Too many CT scans are not medically necessary, and won’t impact the course of treatment for the patient.  Too many CT scans replace the history and physical and talking with the patient.  Too many CT scans are done because doctors are worried that they may be sued if they don’t do it and something rare shows up later.  Too many CT scans are done because patients aren’t willing to take some responsibility for their health and participate in the decision-making process.


All this “avoidance,” unfortunately, has now been shown through this research and other similar reports to have a very real cost, which is not just financial.  It could be the cause of a future cancer or even a death.


Doctors need to lead the way in reducing the risks of these CT scan related problems. 


They need to be certain the scan is truly needed.  They need to be certain that the CT scan machines are monitored carefully for the amount of radiation they produce.  They need to adhere to standards to be certain that the dose of radiation used is the least required to get an adequate study.


My friends, this is a serious problem.  The awareness of doctors and patients about the problem is long overdue. 


Our technology can be terrific and can be lifesaving, but only if used properly and carefully.  It is critical that we be certain that the CT scans we recommend and the CT scans we undergo be done only for appropriate conditions and circumstances, where the benefits clearly outweigh the risks.


Medical technology can be a two-edged sword.  In the case of CT scans, these reports are a clear indication that the sword may just turn out to be the Grim Reaper’s scythe when not used properly.


We simply cannot stand-by as patients or professionals and let that happen.  We must address the issues and find solutions, or the consequences may be enormous.

Filed Under:

Cancer Care | Treatment

Mammography Guidelines: You Can't Dress It Up

by Dr. Len December 10, 2009

I don’t like to keep kicking the proverbial can down the road, but a column in yesterday’s Wall Street Journal about the statistics in the recent mammogram guideline recommendations from the U S Preventive Services Task Force is worthy of comment.


Aside from “getting it right” in my opinion, Carl Bialik’s (“The Numbers Guy”) discussion highlights the imperfection of the statistics that the Task Force relied on in making a recommendation to the public that we should abandon a long standing (and, in my opinion, effective) public health recommendation that women at average risk of breast cancer get a screening mammogram every year beginning at age 40.


The reality is that the fall out from the Task Force recommendations is just beginning, with one state government cutting mammograms for women in their 40's and insurers concerned that their contracts with companies to provide health insurance benefits for employees will require them to do the same.


Mr. Bialik points out that one of the key statistics was not as perfect as one might think.  In fact, the statistic is so imperfect in this case as to make it meaningless. 


As Mr. Bialik reported, the “number needed to screen to save one life”—which the task force said was 1904 for women in their 40’s and 1339 for women in their 50’s—could have in fact been much lower or much larger both for 40-49 year old women and the 50-59 year old age group (see my initial blog on the Task Force report for further discussionon the confidence intervals for these numbers).


If you look at the number and understand statistics, you realize that the Task Force concluded there was a real difference in the impact of screening mammography in saving lives from breast cancer between the 40 and 50 year old age groups. That led the Task Force to conclude there was sufficient benefit in the older women but not in the younger women.


From a science standpoint, as pointed out in the column, that simply isn’t true.  There was so much overlap in the statistic between the two groups as to make any reliable difference in benefit impossible to detect with any degree of certainty.


Then there was the issue of the computer model and the role of value judgments comparing years-of-life-lost to how many “extra” mammograms would be required to save those years of life (as though the computer models all agreed on the same numbers, which they don’t).


Would I be intemperate to suggest that if given a choice, women would rather have the extra mammogram than die? 


Putting an extra mammogram up against losing your life--and coming to the conclusion that women would prefer to avoid that extra mammogram as opposed to saving their life--is an incredibly naïve conclusion in my opinion.  And that’s assuming the computer model is perfect in its conclusions, which is probably not the case.


A comment in the article from a Task Force member also has me concerned: 

“Diana Petitti, a professor in biomedical informatics at Arizona State University and vice chairwoman of the panel, said the task force looked at a range of evidence in making its recommendation. ‘This is purposely a qualitative assessment and not an assessment based on some magic number,’ she said in an email.” 

The Task Force, on its website says it is the “gold standard” in determining what medical screening interventions are effective and which are not.   

I have sat through a number of presentations from the Task Force—including one from its current chairman—where they have emphasized the impartiality of the task force, and the fact that they use a strict “evidence based” standard to make their recommendations.  No evidence or insufficient evidence means no recommendation for or against a particular screening test. 

In this current situation, their vice-chair is now saying that they made “qualitative assessments.”  My interpretation of that statement is that they applied their own values to the interpretation of the evidence, which is what I wrote when their report first came out. They have ignored the valid scientific/data based assessments of others, who have looked at the same data and came to different conclusions about the value of screening mammograms in women ages 40-49.

 That is not what I thought the Task Force was supposed to do.   

I went to the Task Force website today, and here is what I found for their current recommendation language: 

“The decision to start regular, biennial screening mammography before the age of 50 years should be an individual one and take patient context into account, including the patient's values regarding specific benefits and harms. Grade: C recommendation. 

“On December 4, 2009, the USPSTF unanimously voted to update the language of their recommendation regarding women under 50 years of age to clarify their original and continued intent.” (emphasis mine) 

Follow the embedded link for “C recommendation” and this is what you will find:

 Definition:                                                               Suggestions for practice: 

The USPSTF recommends against routinely providing the service. There may be considerations that support providing the service in an individual patient. There is at least moderate certainty that the net benefit is small.

Offer or provide this service only if other considerations support the offering or providing the service in an individual patient.

 Tap dance all you want, the bottom line is that the task force still recommends against routine screening mammography. 

What’s worse in my opinion is that there is no evidence based research to show that following the Task Force recommendation will in fact save lives from breast cancer for women in their 40’s.  None, nada, nothing. 

This is beginning to look like an effort to put “lipstick on the pig”.   

What happened to evidence based recommendations?   

Clearly their “suggestions for practice” has no evidence base whatsoever for women at average risk of breast cancer.  No study has ever been done which confirms that physicians and other health care professionals can accurately predict which women in their 40’s are at greater risk of getting breast cancer and should therefore be advised to get a screening mammogram.  That’s important when you consider the fact that most women who get breast cancer in fact have no specific risk factors for the disease, and that there is a not inconsequential incidence of breast cancer in this age group. 

All of this discussion would be nothing more than interesting chatter and disagreement among scientists if it did not have practical implications. 

Already, we are hearing that insurance companies are concerned they may have to change their policies based on contracts that require them to follow the Task Force recommendations.  And the state of California has taken advantage of the dispute to restrict screening mammograms to disadvantaged women under 50 because of the state’s budget crisis.  The guideline is the reason they have made this decision. 

So what is the solution? 

A respected, evidence-based organization comes out with a guideline that is formed in no small part by opinion rather than solid evidence.  They fail to provide compelling evidence that would support the need for them to change their prior recommendations, which in fact were fairly consistent with other organizations.  They try to “dress up” their recommendation, but when you look behind the screen you see that in fact they still haven’t changed their recommendation, which is against routine mammography for women in their 40’s.  State governments and perhaps insurance companies start to restrict their coverage, in accordance with insurance contracts and other political considerations. 

Maybe it is time for some further action.  Maybe it is time for the Task Force to at least admit that the evidence is inconclusive one way or the other.  Although many of us don’t agree with that, at least it would move us in the right direction. 

The organizations that have supported the USPSTF because those organizations believe  the Task Force relied on solid evidence may not be so comfortable when they actually look at the evidence, or consider the comments from the Task Force.  And I suspect they may be a bit concerned that the task force now admits they made a judgment call, and not one based only on what the science was clearly telling them. 

And let’s also note that supporters of the Task Force—invoking the “evidence standards” of the Task Force—suggested that those who disagreed with the Task Force did so only because of their uninformed or otherwise conflicted reasoning, as though organizations like the American Cancer Society don’t consider the evidence before coming out with their own recommendations or comments. 

It appears that the Task Force feels that decreasing the number of mammograms is a better deal than saving lives.  All one has to do is listen to the outcry from other scientists, physicians, and the public to figure out that there are a couple of folks out there who don’t agree with them. 

It’s time they set the record straight, even if it means changing their recommendation. 

Too many lives may hang in the balance.




Cancer Incidence and Deaths Continue Decline

by Dr. Len December 07, 2009

If you hear good news often enough, does it become “no news?” 


That’s my concern with today’s release of the 2009 version of the Annual Report to the Nation on the status of cancer incidence and deaths in the United States from 1975 to 2006.  The “news” is still good—with some caveats—but will people still pay attention to the message? I certainly hope so. 


If someone had told me years ago about the successes we could achieve in reducing cancer incidence and deaths I would have had serious reservations about our ability to accomplish that task.  Here we are 35 years later, and we have done just that.


This report—which was published in the journal Cancer and is a collaborative effort between the American Cancer Society, the National Cancer Institute, the Centers for Disease Control and Prevention and North American Association of Central Cancer Registries--brings us the same message as last year: overall the rates of cancer incidence and death continue to decline in men and women.  At the same time, some types of cancer are increasing in frequency, and while death rates from lung cancer are dropping in men, the same cannot be said for women.


The report also highlights the incredible decline in cancer deaths from colorectal cancer in the United States, and emphasizes the opportunity for this country to reduce cancer deaths from this disease by half by 2020 if we did everything we could to get people screened and properly treated.  But there may be warning signs of things to come:  there is a disturbing trend of increasing deaths from colorectal cancer in people younger than age 50, which is the age when everyone recommends starting routine screening for people at average risk of getting colorectal cancer.


The report is full of complicated data and information.  Let me offer some of the statistical highlights:


1)      From 1999 through 2006, incidence rates for all cancers in both sexes combined declined 0.7% per year.  For men, the decline was 1.3% per year, and for women the decrease was 0.5% per year.


2)      From 2001 to 2006, the annual decline in cancer death rates in both sexes combined was 1.6% per year.  For men from 2001-2006, the decline was 2% per year, and for women from 2002-2006 the decrease was 1.5% per year.


3)      In men, the decline in incidence and death rates was primarily due to decreases in three major cancers: lung, prostate, and colorectal.  In women, the declines were primarily due to decreases in breast and colorectal cancers. 


4)      There is a glimmer of hope that lung cancer deaths may be declining in women (they have been going down in men for a number of years), but unfortunately the actual number (0.9% decrease per year from 2003-2006) is not significant.  And, the number year of new lung cancer cases in women measured from year to year has been increasing 0.4% per year from 2002-2006.


5)      There are still cancers that are increasing in frequency. For men, among the top 15 cancers, from 2002-2006 the researchers report significant average annual increases each year in melanoma (3.1%), kidney (1.8%), liver (2.6%), esophagus (0.7%) and myeloma (0.7%).


For women, the leading increases in cancer incidence include lung cancer (as noted above), melanoma (3.0%), non-Hodgkin lymphoma (1.1%), thyroid (6.3%), pancreas (1.7%), leukemia (0.3%), kidney (2.4%), and bladder (0.2%).


The researchers continue to highlight the critical fact that there continues to be substantial disparities in cancer incidence and deaths in this country.


For example, black men have the highest incidence of cancer in the country.  Black men also have had a significant decline in the incidence of prostate cancer from 1997-2006.  When looking at death rates, black men and women had the highest rates from 2002-2006 and they were lowest for Asian and Pacific Islanders.


As has been the custom in these annual reports, the researchers concentrate on a specific area of interest for further, in-depth analysis.  In this year’s report, the topic was colorectal cancer, its declining mortality, and the opportunities over the next decade to reduce deaths even further.


The good news here is that the death rates for colorectal cancer are declining dramatically in the United States among all ethnic groups except for American Indian/Alaskan Natives.  The decline is most dramatic for people age 65 and over, possibly due to the fact that they have access to Medicare to cover screening and treatment costs.


However—and this is one of the most concerning aspects of this report—the researchers also report that there is a short term increase in the incidence of colorectal cancer in people less than 50, possibly due to a long term impact of changing risk factors including smoking, overweight/obesity and eating a diet high in red meat consumption all of which are known to be associated with increasing colorectal cancer rates (as recently reported in several recent studies, including one from the American Cancer Society).


The report notes that from 1975 to 2000, the incidence of colorectal cancer has declined 22%.  Based on a computer model, screening has accounted for 50% of that decline.  Treatment and changing risk factors (for the better, during that time period) accounted for the rest of the benefit that was observed.


To me, this is the “kicker” statement of the paper, one that we should all pay close attention to:


“If we can accelerate the projected trends, then an overall mortality reduction of 50% by 2020 is possible….Risk factor modifications, although they require the longest time to produce an impact, will have a sizable effect  by 2020.  Increases in the proportion of adults screened and in the use of endoscopic CRC screening will provide the largest reduction in future death rates with application of current state-of-screening technologies, risk factor modification, and use of current treatment practices.”


Of course, if you are a regular reader of this blog, you know that I am a strong proponent of colorectal cancer screening and treatment.  And, you also know that the comment above—although important and relevant—really isn’t new.  That’s because colorectal cancer is one of the cancers that can frequently be prevented or caught early and treated effectively, if we only applied the basic knowledge that we already have.


To give you an idea of the impact of our capabilities to reduce the burden and suffering from this cancer, consider the above in light of the following statistic:


·        In 2009, the American Cancer Society estimates there will be 49,920 deaths from colorectal cancer in the United States. 


·        If you reduce that by half—as the paper suggests could be done if we did everything right—then by the year 2020 we would have about 25,000 (or more) fewer deaths from colorectal cancer.


Can you imagine saving 25,000 lives from colorectal cancer every year?  Talk about impact! I hope you agree that would be dramatic.


But then we have to look at the context of the current situation, especially when it comes to getting effective screening tests into more widespread use. 


As this current report highlights, there are new ways to screen for colorectal cancer such as CT colonography.  But, even though the burden of this disease is greatest in the Medicare population, the Centers for Medicare and Medicaid Services has declined to cover the test because they claim their analysis shows it isn’t cost effective and because it hasn’t been carefully studied in the 65-and-over population (which is also true for many other things we do in medicine).


I am not going to go into detail on that issue here, and you can read my previous blog posts on the topic—especially the one from June 17, 2009 where the staff from CMS that made the decision to deny coverage talked about how proud they were of their decision in holding down costs and standing up against “special interests.”


When you look at the total picture, it would seem that we should be doing everything we can for all of our people to reduce the risk of death from this disease (which can frequently be prevented in the first place).


That would mean emphasizing the need to “stick to the program” of eating well, avoiding tobacco, maintaining a healthy body weight, and getting screened.  And we should be emphasizing getting screened with the best available methods, which the American Cancer Society believes should include CT colonography.


If we don’t do what we need to do, we certainly will not meet a goal of a 50% reduction in colorectal cancer deaths by 2020.  At the same time, we run the risk of seeing an actually increase in the disease if we don’t get our arms around reducing the risk factors noted above.


In the end, I suspect the most important message of the day is that we have made true, measureable progress in reducing the incidence and deaths from many cancers (some of the recent and continuing comments in a leading newspaper notwithstanding).  This is due to a combination of many factors, including better screening, better treatment, and more awareness and communication.


But we should not fail to recognize that there are some cancers where we could do much better, and some cancers where the rates of incidence and deaths are actually increasing.


Our research has brought us a long way, and we have much further to go.  These annual reports offer us the opportunity to take a true measure of our successes and our failures.


So, here is to the hope of more successes, our opportunity to do even better, and the willingness to squarely address issues which prevent us from becoming all that we can be when it comes to reducing the burden and suffering of the many diseases we call cancer.

Filed Under:

Prevention | Screening | Treatment

About Dr. Len

Dr. Len

J. Leonard Lichtenfeld, MD, MACP - Dr. Lichtenfeld is Deputy Chief Medical Officer for the national office of the American Cancer Society.