Better Health: Smart Health Commentary Better Health (TM): smart health commentary

Article Comments (1)

The Evidence For Colonoscopies As A Cancer Screening Test

Everybody knows that colonoscopy is the best test to screen for colorectal cancer and that colonoscopies save lives. Everybody may be wrong. Colonoscopy is increasingly viewed as the gold standard for colorectal cancer screening, but its reputation is not based on solid evidence. In reality,  it is not yet known for certain whether colonoscopy can help reduce the number of deaths from colorectal cancer. Screening with fecal occult blood testing (FOBT) and flexible sigmoidoscopy are supported by better evidence, but questions remain.  It seems our zeal for screening tests has outstripped the evidence.

Statistics show that the life-time risk for an adult American to develop colorectal cancer (CRC) is approximately 6%. Colorectal cancer is the second leading cause of cancer deaths in the United States. In the US there are currently 146,970 new cases   and 50,630 deaths each year. Between 1973 and 1995, mortality from CRC declined by 20.5%, and incidence declined by 7.4% in the United States. 

The US Preventive Services Task Force (USPSTF) recommends screening for colorectal cancer (CRC) using fecal occult blood testing, sigmoidoscopy, or colonoscopy, in adults, beginning at age 50 years and continuing until age 75 years.

The American Cancer Society divides the available tests into these two categories and makes these recommendations for frequency of testing:

Tests that find polyps and cancer

  • flexible sigmoidoscopy every 5 years*
  • colonoscopy every 10 years
  • double contrast barium enema every 5 years*
  • CT colonography (virtual colonoscopy) every 5 years*

Tests that mainly find cancer

  • fecal occult blood test (FOBT) every year*,**
  • fecal immunochemical test (FIT) every year*,**
  • stool DNA test (sDNA), interval uncertain*

Screening by colonoscopy seems to make more sense than other screening methods, because you can actually see the entire inside of the colon. Colon cancer is preceded by polyps and adenomas that progress to cancer. When a polyp is seen, it can be removed during the procedure. In this study, colonoscopic polypectomy resulted in a lower than expected incidence of colorectal cancer. But other studies suggest that the progression to cancer is not a steady process, and that adenomas may regress.

There is good evidence here and here that any benefit of colonoscopy is restricted to left-sided colon cancers, with no impact on right-sided colon cancer; we don’t understand why. Some possible explanations are discussed here.

There are pros and cons to each of the different screening tests.  Barium enemas and CT virtual colonoscopy involve significant doses of radiation. Colonoscopy only needs to be done every 10 years, but it involves an uncomfortable bowel prep, requires sedation, can cause serious complications like bowel perforation, and is unacceptable to some patients. FOBT screening is painless and harmless but has a lot of false positives and requires annual testing. Getting patients to come back every year for FOBT is problematic. Compliance and cost must be considered. Colonoscopy is expensive and there are not enough colonoscopists to screen everyone.

Apart from all those peripheral considerations, what do we know about the bottom line: the ability of each screening method to prevent deaths from colon cancer?  According to the National Cancer Institute,

  • Studies have shown that FOBT, when performed every 1 to 2 years in people ages 50 to 80, can help reduce the number of deaths due to colorectal cancer by 15 to 33 percent.
  • Studies suggest that regular screening with sigmoidoscopy after age 50 can help reduce the number of deaths from colorectal cancer, perhaps by as much as 50%, but the quality of evidence is not as good as for FOBT.
  • it is not yet known for certain whether colonoscopy can help reduce the number of deaths from colorectal cancer.

No randomized controlled trials have tested whether colonoscopy reduces the incidence of CRC. Support for the role of colonoscopy in CRC prevention derives from indirect evidence and observational studies.

There is an excellent review of all the pertinent studies here.  Even though studies show that screening can reduce disease-specific mortality from colorectal cancer, there is little evidence that it reduces all-cause mortality. So as far as we know, screening probably won’t prolong your life. It seems like it should: I don’t understand why it doesn’t, and it bothers me. This certainly isn’t the message we’re getting from the media and from the medical profession.

I’m guessing that if the appropriate studies were done and the technique of colonoscopy were optimized, a reduction in colon cancer deaths would be demonstrated. I’m guessing that colonoscopy would detect more cancers and precancerous lesions than FOBT or sigmoidoscopy, but I’m wondering whether the benefits of colonoscopy would outweigh the additional cost and risks compared to other screening methods. And I’m disturbed that a reduction in all-cause mortality has not yet been clearly shown for any screening method. We need more research to help us understand these issues.

Pending better evidence, I support the current USPSTF recommendations. I think patients should be told the pros and cons and choose which screening test they prefer. I have chosen annual FOBT for myself.

*This blog post was originally published at Science-Based Medicine*

You may also like these posts

    None Found

Read comments »

One Response to “The Evidence For Colonoscopies As A Cancer Screening Test”

  1. I’ve performed around 15,000 colonoscopies in the past 20 years. The author has summarized the issue very well. There has been no randomized controlled trial (gold standard) demonstrated proof that colonoscopy works; and it is unlikely that any such trial will be forthcoming. The test is endorsed by the skeptical and conservative USPSTF, as the author points out. When physicians reach the age of 50, the overwhelming majority seek a colonoscopy for themselves. In my practice, admittedly not a scientific experiment, I cannot recall finding a colon cancer in an individual who has had recommended colonoscopies. Conversely, the colon cancers I discover are generally folks in their 60s or 70s who have symptoms and have never been screened. I agree that it’s not perfect and it’s not proven, but I believe it is the best screening option we have at this time.

Return to article »

Latest Interviews

IDEA Labs: Medical Students Take The Lead In Healthcare Innovation

It’s no secret that doctors are disappointed with the way that the U.S. healthcare system is evolving. Most feel helpless about improving their work conditions or solving technical problems in patient care. Fortunately one young medical student was undeterred by the mountain of disappointment carried by his senior clinician mentors…

Read more »

How To Be A Successful Patient: Young Doctors Offer Some Advice

I am proud to be a part of the American Resident Project an initiative that promotes the writing of medical students residents and new physicians as they explore ideas for transforming American health care delivery. I recently had the opportunity to interview three of the writing fellows about how to…

Read more »

See all interviews »

Latest Cartoon

See all cartoons »

Latest Book Reviews

Book Review: Is Empathy Learned By Faking It Till It’s Real?

I m often asked to do book reviews on my blog and I rarely agree to them. This is because it takes me a long time to read a book and then if I don t enjoy it I figure the author would rather me remain silent than publish my…

Read more »

The Spirit Of The Place: Samuel Shem’s New Book May Depress You

When I was in medical school I read Samuel Shem s House Of God as a right of passage. At the time I found it to be a cynical yet eerily accurate portrayal of the underbelly of academic medicine. I gained comfort from its gallows humor and it made me…

Read more »

Eat To Save Your Life: Another Half-True Diet Book

I am hesitant to review diet books because they are so often a tangled mess of fact and fiction. Teasing out their truth from falsehood is about as exhausting as delousing a long-haired elementary school student. However after being approached by the authors’ PR agency with the promise of a…

Read more »

See all book reviews »