Sections & Programs

Virtual Colonoscopy

Abraham H. Dachman, MD oversees the Virtual Colonoscopy program here at the University of Chicago. Dr. Dachman is a member of the American College of Radiology and the subcommittee on Colon Cancer Screening on Virtual Colonoscopy. He is an expert in the field of abdominal imaging who is accomplished and respected in the field of Virtual Colonoscopy. Dr. Dachman has edited a textbook, The Atlas of Virtual Colonoscopy, to help train radiologists and inform clinicians and interested patients about this test.
 

Abraham H. Dachman, MD

Professor of Radiology

 

 

CT Virtual Colonoscopy is a CT scan done in a special way that allows a radiologist (specialized physician trained in x-ray imaging) to look at the large bowel (colon) to detect polyps and cancers. Polyps are small growths in the colon. They include benign lesions, some of which may become cancerous if not removed. CT Virtual Colonoscopy is a technique that uses a CT scanner and computer virtual reality software to look inside the body without having to use sedation and without having to insert a long tube (Conventional Colonoscopy) into the colon or without having to fill the colon with liquid barium (Barium Enema). Additionally, the CT scan images allow a radiologist to see other areas in the abdomen that are not visualized by the other tests. In this manner, a potentially life threatening problem like cancer may be detected early, and possibly treated. At The University of Chicago, we have been doing research on virtual colonoscopy since 1996. We believe that CT Virtual Colonoscopy is safer, more comfortable, less expensive and faster than conventional colonoscopy. We know that it’s not as accurate a test as conventional colonoscopy, but data now show that it is a very good screening test for colon cancer.

Here you can review an American College of Radiology published brochure with additional information on what a virtual colonoscopy is.

 

Colorectal cancer (CRC) is a leading cause of cancer-related death in the United States. In 2007 an estimated 112,340 new cases of colon cancer will be diagnosed in the United States, and 52,180 patients are expected to die of the disease. Colorectal cancer is the third most common cancer in men and women (and second most common when combining data for men and women). It is the most common cause of cancer deaths in non-smokers.

Research suggests that the incidence is declining due to increased screening and polyp removal, which prevents the progression of polyps into invasive cancers. Colon cancer can be prevented if polyps are discovered and removed early. Tumors, masses of abnormal cells, take years to develop. Initially, a cell from the colon starts to multiply abnormally and forms a benign (non-cancerous) polyp, which can remain harmless for a long time before becoming an aggressive cancer. Polyps, when detected early can often be removed during colonoscopy. Nevertheless, individuals at greatest risk of developing colorectal cancer remain largely under screened. This is due, in part, to poor public awareness and acceptance of current screening techniques. In fact, about half the patients that are eligible to be screened do no get screened.

Risk Factors: A personal or family history of colorectal cancer or polyps, and inflammatory bowel disease have been associated with increased colorectal cancer risk. Other possible risk factors include physical inactivity, high-fat and/or low-fiber diet, as well as inadequate intake of fruits and vegetables. Recent studies have suggested that estrogen replacement therapy and non-steroidal anti-inflammatory drugs such as aspirin may reduce colorectal cancer risk. The risk for colon cancer increases with age and screening is recommended for anyone 50 years or older.

Early Detection: The American Cancer Society recommends screening beginning at age 50 for both men and women. As of March 2008, the ACS includes virtual colonoscopy (VC) as one of the recommended colon cancer screening tests. Virtual colonoscopy should be repeated every 5 years if normal and a conventional colonoscopy should be performed if VC is abnormal (ACS Screening Recommendations). The other recommended tests are the following: a fecal occult blood test (FOBT) and flexible sigmoidoscopy (if normal, repeat FOBT annually, and flexible sigmoidoscopy every 5 years), double contrast barium enema (if normal, repeat every 5 years), or colonoscopy (if normal, repeat every 10 years). A digital rectal examination should be done at the same time as VC, sigmoidoscopy, colonoscopy, or double-contrast barium enema. These tests offer the best opportunity to detect colorectal cancer at an early stage when successful treatment is likely, and to prevent some cancers by detection and removal of polyps.

People should begin colorectal cancer screening earlier and/or undergo screening more often if they have a personal history of colorectal cancer or adenomatous polyps, a strong family history of colorectal cancer or polyps, a personal history of chronic inflammatory bowel disease, or if they are a member of a family of a hereditary colorectal cancer syndromes.

Some of the symptoms of colon cancer include blood in the stool, changes in bowel habits, decreased appetite, or fatigue. Although people with a family history may be at risk of colon cancer, many cases of cancer arise in people who have NO family history and NO noticeable symptoms. This is another reason why screening is so very important. For further information pertaining to colon cancer and related topics, please follow this link to the Gastroenterology department at the University of Chicago.

 

Although the Barium Enema and Conventional Colonoscopy are exams of the entire colon, the sensitivity of barium enema varies greatly with the experience of the radiologist and is uncomfortable. Conventional Colonoscopy is the most accurate test for detection of polyps, but is associated with increased risk of complications and expense. About 2-5% of the time conventional colonoscopy cannot see the entire length of the colon. It may also cause an injury to the colon (tear or perforation) in one out of 1500 patients. Conventional Colonoscopy requires medication be given into the vein to make you sleepy and make the exam tolerable, therefore an hour or so of rest is needed after the exam and you should not drive yourself home.

Virtual colonoscopy is considered safe, non-invasive and a good screening test. It is well accepted by patients. They like the fact that the test takes 15 minutes and the computer does the rest. You can get up and drive home afterwards. This is why we have vigorously pursued research on virtual colonoscopy and offer it to patients for colorectal cancer screening. We still recommend conventional colonoscopy for patients with a high risk of colon cancer. High risk individuals who can not undergo conventional colonoscopy or whose physician recommends virtual colonoscopy for some reason, may also be screened with CT virtual colonoscopy. High risk patients include those with a personal history of colon cancer, a close relative who had colorectal cancer before age 60, a history of pre-malignant adenomatous polyps in the colon, a history of ulcerative colitis and certain family syndromes.

While there are many benefits to using virtual colonoscopy to screen for colon cancer, there are some limitations of this technique. Since it is a non-invasive test, it cannot be used to remove any polyps that are found. This is why at the University of Chicago patients are often offered same day conventional colonoscopy if their virtual colonoscopy shows polyps that need removal. Virtual colonoscopy is also not perfect at detecting small polyps less than 6mm in size. However, the vast majority of these small polyps (more than 99%) are not pre-cancerous and do not need to be removed. The cancerous potential of a polyp is directly related to its size (the bigger the polyp the higher the chances of it being cancerous). Also, polyps tend to grow very slowly over years. This is why small polyps can safely be monitored for growth with regular follow-up screening without having to remove them with the more invasive conventional colonoscopy. Flat polyps that do not protrude into the bowel lumen may also be difficult to find on virtual colonoscopy. However, many of these flat and small polyps are also missed during conventional colonoscopy. In fact, because of advances in technology and software used in virtual colonoscopy, it may be better than conventional colonoscopy in finding these polyps.

Follow-up screening after a virtual colonoscopy is similar to that after a conventional colonoscopy and depends on the findings. If no polyps are found, patients are asked to repeat the test in 5-10 years. If 1 or 2 small polyps are found, it is considered safe to monitor them with a repeat screening in about 3 years. If more small polyps or a single large polyp are detected, patients undergo a conventional colonoscopy to remove them and can have a follow-up virtual colonoscopy in 5 years. Since virtual colonoscopy is a relatively new procedure, the standards for follow-up have not been universally adopted and may vary with each individual patient case.

 

Virtual Colonoscopy uses a low radiation dose helical (or spiral) computed tomography (CT or CAT scan) of the abdomen, which allows a radiologist to create pictures on the computer that look similar to those seen by conventional colonoscopy. Patients need a cleansing preparation of their bowel prior to the test and several mild preparation choices are available. The day of the test, they come to the radiology department for a CT scan. The actual virtual colonoscopy procedure will begin by having a small, thin, flexible plastic tube placed in the rectum, so that carbon dioxide gas (or room air in some cases) can be introduced in a safe manner using a pressure controlled mechanical pump. A very low x-ray dose CAT scan is then performed while the patient lies on their back and then again on their stomach. The patient needs to remain motionless and hold their breath for about 15 seconds in each position to improve picture quality. The total time required for the study is approximately 15 minutes of which about 5-7 minutes are uncomfortable due to the filling of the colon with gas. There may be mild cramping while the colon is filled and a few patients have severe cramping which goes away immediately on completion of the exam. Because sedation is not required, patients are free to leave the CT suite immediately without the need for observation or recovery. Patients can resume normal activities after the procedure. We offer the choice of remaining in the department for about 1-2 hours while the scan is read by the radiologist. If the scan shows a large polyp or mass, we offer same day referral to conventional colonoscopy, performed by the gastroenterologist (a bowel specialist) for biopsy or removal of the polyp(s) or mass.

 

When gas or air is introduced in the colon some patients experience minimal temporary abdominal cramping or “gas pain”. We generally use carbon dioxide which is absorbed faster than air and makes you comfortable as soon as the exam is over. An intravenous injection of glucagon, a widely used medicine to relax the bowel can also be given to help reduce gas pain. We ordinarily do not use glucagon since the discomfort is brief and mild to moderate for most patients.

 

Virtual colonoscopy is a very safe procedure and is considered safer than conventional colonoscopy but has some small risks. Some patients may experience a brief feeling of sweating and feeling faint during the distention phase of the exam. This is called a “vasovagal reaction” and occurs about 1% of the time. The feeling goes away after lying down for a few minutes.

Perforation of the colon is rare (0.0046% in a large survey of experts doing virtual colonoscopy) and is usually detected by the CT scan during the exam. Some of these rare perforations give no symptoms at all and patients are merely observed overnight for their safety. Symptomatic perforations are very rare and could require surgical repair.

The radiation dose for a virtual colonoscopy is much less than for a routine CT scan. However, pregnant women should not undergo virtual colonoscopy. The risk of not screening for colorectal cancer is much greater than the radiation risk of a low dose CT scan. For patients 50 years of age or older the use of CT is appropriate for colorectal cancer screening.

The preparation for virtual colonoscopy includes a colon cleansing. We prefer to use a saline cathartic in the form of magnesium citrate or phosphosoda. People with impaired renal function or cardiac disease should ask about an alternate preparation with a reduced volume of polyethylene glycol instead. Current research in virtual colonoscopy is exploring “fecal tagging” techniques with contrast material that may eliminate the need for colon cleansing in the future. This method will involve an alternate bowel prep strategy with the patient consuming several small doses of the contrast material solution, either iohexol or barium-sulfate, during the two days just prior their virtual colonoscopy. Even though using oral contrast for virtual colonoscopy is a relatively new method, oral contrast material has been used in radiology for many years and is safe and well tolerated by patients

 

There are several ways to get a colonoscopy test, as described below. If you require any further assistance or have any questions, please feel free to Contact Us.
 

Screening (Self-Pay) Virtual Colonoscopy

For patients who meet American Cancer Society guidelines of-average risk-for colon cancer, you may self-pay ($3,084, of which a 25% discount is offered) and obtain a virtual colonoscopy. Generally this means you:

  • are 50 years of age or older;
  • do not have symptoms of colon cancer;
  • do not have documented occult blood in the stool (you may have visible blood from hemorrhoids); and
  • do not have a brother, sister or parent who had colon cancer.

Call our registration team at 773-795-9723, who will go through a checklist with you and answer your questions. They will send you a card to test your stool for blood. You will be given instructions on how to pay and how to schedule your tests.
 

Insurance-Billed VC for Incomplete Colonoscopy Or Patients Who Cannot Undergo Colonoscopy

Patients who have an incomplete colonoscopy at The University of Chicago, are offered a same-day VC if their colonoscopy is incomplete. The patient is already prepared so by doing it on the same day, the patient avoids the need for another prep and another day off from work. This is a particular advantage to having your conventional colonoscopy done at the University of Chicago. For more information about how to make an appointment with a gastroenterologist please see the Gastroenterology at the University of Chicago website. We believe that in our hands, virtual colonoscopy is the best test to evaluate for polyps and masses in patients who have an incomplete colonoscopy or who cannot undergo colonoscopy or who refuse colonoscopy. Insurance is billed for these VC exams, as a non-infused CT of the abdomen and pelvis. If you have had a VC here under these circumstances and insurance refuses to pay, please contact our registration team at 773-795-9723. Traditionally, patients who have had an incomplete colonoscopy would undergo a barium enema.
 

Will My Insurance Pay For Virtual Colonoscopy?

Virtual Colonoscopy for screening is not currently reimbursed by most insurance companies, but some do. Some companies pay for virtual colonoscopy if it is done for diagnosis (signs or symptoms) especially after an incomplete optical colonoscopy. As a result, patients calling for a screening virtual colonoscopy will have to assume the cost of the procedure themselves (unless they can determine that their insurer does cover it) – a $3,084 charge. This charge has 2 components a) the hospital charge (payable to “The University of Chicago Hospital”) and b) the radiologist charge (payable to “The University of Chicago Physicians Group”).
 

Will My Insurance Pay For Conventional Colonoscopy Done To Evaluate An Abnormality Found On The Virtual CT Colonoscopy Exam?

If you choose to have a same-day optical colonoscopy, we need to know some additional information beforehand. You will be given information about some additional restrictions during the colon preparation (e.g. no colored liquids) and the need to stop some medications like blood thinners. A request form will be sent (usually faxed) to your doctor to fill out. This is needed to gain information about things that might affect the colonoscopy, such as need for prophylactic antibiotics (e.g., patients with a heart valve or artificial joint) or whether you have a pacemaker or defibrillator (that affect use of devices to remove polyps). Our coordinator will give your insurance information to the gastroenterology clinic to help expedite pre-approval by insurance, if possible. Additional questions about insurance for conventional colonoscopy can be directed to the Gastroenterology Business Representative (773-702-2122).

You can always choose to have optical colonoscopy scheduled at a future date. That exam does involve sedation and normally you will need someone else to take you home after the exam since you may not drive or operate machinery on the same day that you have sedation. The department of gastroenterology will give you more information about that.

Medicare patients. Patients with Medicare will be required to sign a waiver informing you that virtual colonoscopy is a non-covered service (when done for screening), and you will accept responsibility for the payment in full. Your Medicare carrier will be initially billed and upon denial of the claim, which we expect, you will be billed directly. Medicare often does pay for virtual colonoscopy when done to evaluate an abnormality (such as blood in the stool or a low blood count) if performed after a failed attempted conventional colonoscopy.

Medicare may start covering virtual colonoscopy for screening in the near future. On December 20, 2007, U.S. Representative from Wyoming Barbara Cubin introduced legislation (H.R. 4879) to promote access to this form of colon cancer screening and mandate coverage of the test by Medicare. The bill also proposes to waive patient co-pays for VC if the test is done within 6 months of enrollment in Medicare, just like what is currently done for colonoscopies and mammograms (Full story).
 

Whom Do I Call To Schedule A Virtual Colonoscopy At The University Of Chicago?

Call our registration team at 773-795-9723, who will go through a checklist with you and answer your questions. They will send you a card to test your stool for blood. You will be given instructions on how to pay and how to schedule your tests.

Before you call please prepare the following information to expedite your call:

  • Your medical and family history
  • Medication list
  • Know whether you are undergoing screening or diagnostic virtual colonoscopy
  • If you are undergoing diagnostic virtual colonoscopy, you will need a prescription from your referring doctor. Please have that doctor’s contact information available.
  • Insurance or Medicare information

If you do not know some of the above, please call anyway and we will try to assist you.

 

As VC continues to evolve and become a more prevalent exam across the country, more patients will undoubtedly look to their physicians as a source of information. Our growing website will serve as an excellent resource for physicians to provide their patients with accurate and easy to understand information about VC. The information may be viewed privately in the doctor’s office or in the patient’s home, thus reducing the level of embarrassment many patients experience when trying to obtain information they consider "sensitive." Physicians will find that they can devote more time to individual patient care in lieu of explaining a single procedure repeatedly. We aim to answer many of the questions patients will have concerning VC and at the same time educate the patient about CRC and screening in general.

 

Conventional colonoscopy (CC) is considered the "gold standard" by which we measure any other CRC screening test. It offers direct visualization of the colon and concurrent removal of potentially cancerous lesions. In terms of performance this is a very effective method to detect cancer. However, it is a time consuming procedure that many patients consider to be uncomfortable and thus undesirable. Sedation is often required, which is associated with negative side effects, and often requires patients to take a day off from work or other activities. Also, conditions may exist that do not permit visualization of the entire colon, leaving the patient with an "incomplete" evaluation. Virtual colonoscopy (VC) is a relatively new exam for polyp and CRC detection that shares many of the benefits of CC, yet has particular advantages in areas where CC falls short.

A VC involves colonic insufflation with carbon dioxide via a short rectal tube and a subsequent helical CT of the abdomen and pelvis (Optimizing Colonic Distension). A pressure sensitive mechanical pump is used which adds a safety factor to prevent or minimize the risk of perforation. Specialized software is used to produce 2D and 3D reconstructions of the colon for radiological interpretation. For the patient, an exam typically lasts about 15 minutes and is performed without sedation allowing an immediate return to one’s daily activities. Moreover, it has been suggested that patients find VC a more desirable procedure to undergo than CC. Visualization of the entire colon is usually possible using VC which permits assessment of colonic segments that are rendered inaccessible to CC as in the case of an obstructing lesion or patient intolerance. Likewise, the modality of CT in the performance of a VC allows detection of potentially important extra-intestinal findings (e.g. tumor in another solid organ) that would not be found by any of the other screening techniques. The incidence of “significant” extra-intestinal findings have been reported at 9%-12% [17-19]. The cost to work-up extra-colonic findings on VC was estimated by Gluecker et al. [20] to be from $28-34 per exam.

The limitations of VC are largely due to both technical factors and a lack of experienced interpreters. False positive readings may result from misinterpretation of colonic folds or retained stool as polyps. Conversely, polyps or cancerous lesions may be mistaken for normal anatomic structures such as the ileocecal valve or normal colonic folds. Although current research shows VC to be inadequate in detecting polyps less than 6mm in size, fewer than 1% of such lesions are precancerous. Some of the parameters involved in the performance of a VC, such as optimal colon preparation, and computer based interpretive methods are currently being researched and will no doubt be improved upon. As radiologists and radiology residents continue to receive formal training in VC, we can expect more reliable interpretations of the studies. The opportunity for radiologists to gain more practical and applied experience in VC will depend a great deal on how willing patients are to undergo the procedure.
 

In a recent study by Kim et al. [18] involving 6,283 screening subjects, the diagnostic yield and use of resources for screening purposes compared VC to CC. The study found that VC and CC had similar rates for detecting advanced neoplasia, but CC had considerably greater complication and polypectomy numbers. Pickhardt et al. [19] reported a test positive rate of 12% based on screening 4,000 asymptomatic adults.

The American College of Radiology Imaging Network trial (ACRIN 6664), a large multi-center study involving 2,531 screening patients and well-trained interpreters, recently unveiled its results. Preliminary data are promising, with a per-patient sensitivity of 90% for polyps 10mm or larger, on par with CC performance. This trial supports more widespread use of VC in the screening population and reimbursement of this method for screening purposes. While CC does have the advantage of being both a screening and therapeutic tool, VC appears to be an attractive screening alternative due to its good performance in recent studies, less invasive method, and more efficient use of resources.

Table 1. Preliminary ACRIN data. 2,531 screening patients in 15 medical centers.

  ≥ 5 mm ≥ 6 mm ≥ 7 mm ≥ 8 mm ≥ 9 mm ≥ 10 mm
Sensitivity 65% 78% 84% 87% 90% 90%
Specificity 89% 88% 87% 87% 86% 86%
PPV 45% 40% 35% 31% 25% 23%
NPV 95% 98% 99% 99% 99% 99%

The results of the Munich Colorectal Cancer Prevention Trial were presented at the 2007 International Symposium on Virtual Colonoscopy. It involved 307 screening patients and prospectively tested five different colorectal cancer screening modalities: VC, CC, flexible sigmoidoscopy, and blood tests like fecal occult blood test (FOBT) and imHb immunochemical fecal occult blood testing (FIT). For lesions >6mm, sensitivity of VC was nearly as good as CC, 93.7% vs. 95%, respectively. Specificity of VC was 98.6% and 96.9% for large and medium polyps, respectively. VC and CC both did better than sigmoidoscopy. FIT and FOBT yielded sensitivities of 22.2% and 17.8%, respectively, for adenomas 10mm or larger.

Virtual colonoscopy in higher risk patients has also had positive results. An Italian study lead by Dr. Daniele Regge was presented at the 2007 International Symposium on Virtual Colonoscopy and included 934 asymptomatic but higher risk patients due to personal history of polyps, family history of colon cancer, or positive FOBT. Per-patient sensitivity/specificity for lesions 6mm or larger was 84.2%/90.4%. A British study with 100 symptomatic or higher-risk patients found VC equivalent to optical colonoscopy for diagnosing colon cancer and clinically significant polyps [22]. VC can be a useful alternative in symptomatic adults where the clinician wishes to exclude polyps, cancer, strictures, and diverticulosis as a cause of symptoms.

The performance of VC has been demonstrated in several meta-analyses. Overall, these older studies show VC is less sensitive and more variable than CC in diagnosing medium polyps, does well in identifying larger polyps and is very specific in polyp detection. The most recent meta-analysis by Rosman et al. [1] included 30 studies of VC performance results. Pooled per-patient sensitivity of VC was 82% for polyps greater than 10 mm, 63% for polyps 6 to 10 mm, 56% for polyps 0 to 5 mm. Per-patient specificity was 96% for polyps 10mm or larger and 84% for polyps 6mm or larger. Three other meta-analyses have shown better sensitivity results (table 2). The poor results from some of the studies included in these meta-analyses have been blamed on poor study design and insufficient interpreter training. These potential pitfalls have been identified and are corrected in any ongoing studies or clinical practice.

Table 2. Meta-analysis sensitivity results for virtual colonoscopy.

Study Patients >10mm per patient sensitivity % 6-9mm per patient sensitivity (%)
Sosna et al. (2) 1,324 88 84
Mulhall et al. (3) 6,393 85 70
Halligan et al. (4) 4,181 93 Large + medium polyps 86

(Reproduced from Lefere et al. [8])

 

Based on positive performance results from multiple trials, the American Cancer Society and American College of Radiology currently recommend virtual colonscopy as a screening exam (Screening Guidelines Summary). Colorectal cancer screening with VC is also endorsed by the three major gastrointestinal societies: the American Gastroenterological Association (AGA), the American Society for Gastrointestinal Endoscopy (AGSE), and the American College of Gastroenterology (ACG).

In addition to screening, VC is also indicated after incomplete conventional colonoscopy, either due to an obstructing process or colonic tortuosity. In a recent study, lesions 6mm or larger were found in 13.2% of patients that had an incomplete colonoscopy [21]. Furthermore, VC is a better option than barium-enema after incomplete CC due to its increased sensitivity and ability to be performed on the same day as CC. VC can also be utilized when conventional colonoscopy is relatively contraindicated. This would include patients with severe cardio-pulmonary disease, patients who must remain anticoagulated, and those simply unwilling to undergo CC. The unwillingness to undergo CC may be one of the reasons why about 50% of the eligible population has not been screened for colon cancer. VC may be particularly useful in this population due to its less invasive nature. Current research is also exploring fecal tagging techniques with contrast material to try and eliminate the need for extensive bowel preparation which would make VC an even less uncomfortable test.

 

Clear guidelines as to the treatment of lesions found on virtual colonoscopy have not yet been established and opinion differs among radiologists and gastroenterologists. Polyps 10mm or larger always warrant a colonoscopy for removal. The recommendations on 6-9mm polyps vary and depend on the number of these polyps found (Table 3). Polyps 5mm or less in size can be followed with regular follow-up in 5-10 years, although the American College of Gastroenterology (ACG) and Rex have a more aggressive approach and recommend removing <5mm polyps if there are more than three or two present, respectively.

The Working Group on Virtual Colonoscopy has created a C-RADS system for reporting virtual colonoscopy findings (Table 4). This system will help generalize results among institutions and assist in making clinical decisions regarding treatment and follow-up.

The risk of finding advanced adenoma in a polyp is directly proportional to its size. Kim et al.[16] have suggested referral for polypectomy only in the case of polyps 10mm or larger and surveillance with follow-up VC for all smaller lesions. This conclusion is based on the low rate of advanced adenoma and high-grade displasia found in polyps smaller than 10mm. The rate of conventional colonoscopy done after VC for polyp removal has been reported to be 6.4% by Pickhardt et al. [17], 7.8% by Pickhardt et al [unpublished data], and 7.9% by Kim et al. [18]. If all the patients with either a polyp 6mm or larger underwent CC, the maximal referral rate would have been 11.9% in the Pickhardt study [17] and 12.9% in the Kim study. The actual CC rates performed were smaller because many patients with 6-9mm polyps opted for short term VC surveillance instead of CC. In the study by Pickhardt et al. [17], the concordance rate of polyp findings VC and CC was 91.5%( per-patient PPV ) and 86% of the patients undergoing CC were able to do it same day as VC. Most recently, the ACRIN trial showed a referral rate of only 8.3% in a screening population. This is low enough for VC to be efficient and cost-effective enough for screening purposes. Patient convenience and efficiency was emphasized, with 99% of patients undergoing same day colonoscopy.

The current general consensus is that polyps 5mm or smaller should not be reported on VC and a study with such a polyp is considered negative. One reason for this is that the risk of advanced disease in these lesions is very small and reporting them could cause unnecessary anxiety in patients and their referring doctors. Another reason is that most of these diminutive lesions are false negatives and cannot be found at subsequent colonoscopy. Patients with a negative study are asked to repeat colorectal screening in 5 years. Clinical management of polyps 6-9mm is more controversial and includes either removal with CC or short-term VC surveillance. Surveillance offers the benefit of less invasive tests and more efficient use of resources, however, it involves the small risk of following a polyp that contains a focus of cancer. Patients with 6-9mm polyps are always offered both removal and surveillance options and a decision is made based on the patient’s preference and their particular situation. Again, any polyps 10mm or larger are referred for CC removal.

Table 3. Recommendations for follow-up and removal of lesions found at VC.

 

Routine screening 5 years-10 years

Screening 3 years Removal of lesions
Working group on Virtual Colonoscopy [5] All lesions ≤5mm <3 Lesion 6 mm-9 mm Lesion ≥1 cm ≥3 lesions 6 mm-9 mm
American College of Gastroenterology [6] One lesion ≤5 mm   One lesion ≥6mm three lesions any size
Rex [7] One lesion ≤5mm   One lesion ≥6mm two lesions any size
Ransohoff [7] All lesions ≤5 mm 6-9 mm (watchful waiting) Lesion ≥1 cm ≥3 Lesions 6-9 mm

(Reproduced from Lefere et al. [8])

 

Radiation Exposure

One of the concerns about using virtual colonoscopy to find colon polyps is the exposure to radiation during the test and potential to induce cancer. However, very low dose scanning protocols have been optimized to reduce radiation exposure while still maintaining image quality. Even with the use of multi-slice scanners, radiation dose is kept to a minimum by reducing the tube current used (mAs). In a study by Brenner et al. [9], the absolute life time risk for cancer induction from a normal dose virtual colonoscopy is 0.14% in a 50 year-old patient and about 0.07% in a 70 year-old patient. Overall, the 6% mortality risk from colon cancer outweighs the small cancer risk from VC radiation exposure.
 

Perforation Risk

Most of the data on perforation is weighted to older cases performed without the current version of the carbon dioxide pump. There is a risk of perforation, albeit very small. The Boston Working Group questionnaire indicated a 0.005% incidence of symptomatic perforation [15]. In a survey from Israel, Sosna reported to be 0.059% by Sosna et al.[10] and 0.052% by Burling et al. [11]. The overall symptomatic perforation rate from these studies combined was 0.005-0.03%. Manual insufflation was used in 16 of the 18 patients with perforations in these studies. This is why we use automatic pump insufflation with constant monitoring of intra-luminal pressure. The majority of perforations occur in symptomatic or high-risk patients and is very rare in the asymptomatic, screening population. Even with this potential risk to undergoing virtual colonoscopy, the VC perforation risk is much lower than CC (0.1-0.2% perforation risk) and more perforation cases after CC result in surgery or even death [12].

Patients who should not get barium enema because of risk of perforation (e.g. acute abdomen, peritoneal signs) should not get a VC. Patients who had a recent colonoscopic biopsy should not get a same day VC if the biopsy was a snare polypectomy or “well biopsy” in which deep tissue is obtained. There are no standard guidelines, but to be safe these patients should wait at least a couple of weeks or more. Patients who had a superficial biopsy could get same day VC if necessary.

 

We believe that VC for screening is a “win-win” for the gastroenterologist, the patient and the radiologist. Since most screening patients have a normal exam, it makes sense to use a cheaper, less invasive exam and reserve the busy optical colonoscopy schedule for high risk patients who are more likely to harbor a polyp or a mass.

Data on VC cost effectiveness is lacking. Arnesen et al. [14] showed that VC can be cost-effective for medium sized and larger polyps followed by CC for polypectomy, but depends on an institution’s hardware and organization. A study using 100,000 hypothetical patients found that VC is the cheapest and safest way to reduce colon cancer mortality in patients 50 years or older [23]. Vijan et al.[13] concluded that VC is cost-effective compared to no screening at all based on the results of a meta-analysis and the natural history of colorectal cancer. However, VC is not cost-effective compared to CC when used for screening.

Virtual colonoscopy is currently not covered by third-party payers for screening purposes. However, one group at the University of Wisconsin was able to convince local managed care organizations to approve reimbursement for VC colorectal screening [19]. This favorable outcome along with further positive results from VC trials, including the ACRIN trial, will pave the road for universal VC screening coverage.

Medicare may start covering virtual colonoscopy for screening in the near future. On December 20, 2007, U.S. Representative from Wyoming Barbara Cubin introduced legislation (H.R. 4879) to promote access and mandate coverage of VC screening by Medicare. The bill also proposes to waive patient co-pays for VC if the test is done within 6 months of enrollment in Medicare, just like what is currently done for colonoscopies and mammograms.

 

Most patients experience mild or moderate pain. Severe pain is very brief for some patients. We employ a “deflation maneuver” in which the colon is partly deflated between the supine and prone scans. In this way, pain is brief, about 7 minuets total for both scans together. Furthermore, since carbon dioxide is used and not room air (except in rare instances), the gas is quickly resorbed. Most patients are pain free when leaving the CT suite.

The greatest discomfort caused by VC is due to bowel preparation and colonic distension. Studies have not shown a clear patient preference of VC vs. CC. The fact that patients are under conscious sedation during most CC procedures may lead to patients preferring CC while not considering the inconveniences of sedation. Ongoing research is developing fecal-tagging methods to decrease the need for extensive bowel prep and perhaps eliminate the need for preparation all together. These new methods will enhance the VC experience and potentially increase colon cancer screening compliance.

Prepless and mild prep options would be of great use in elderly or frail patients that cannot tolerate laxatives or conventional colonoscopy. Virtual colonoscopy in this population would be especially useful since they have a higher threshold for intervention and need for referral to conventional colonoscopy would be minimized. A study by Dr. Iafrate presented at the 2007 RSNA consisted of 132 patients with an average age of 81 who had an incomplete CC. The subjects ate a low-fiber diet along with Gastrograffin tagging agent for two days prior to exam. No actual laxative was used although Gastrograffin is a mild cathartic. All subjects completed VC and 14% had additional findings. The prep was well tolerated with no major side effects and only 25% experiencing mild side effects like diarrhea. Another benefit of VC in the elderly may be finding of extracolonic malignancies. A study by Tolan et al. [21] found that 6% of symptomatic patients undergoing VC had extrocolonic malignancies.

 

CAD and Virtual Colonoscopy

Computer-Aided Detection (CAD) is currently an area of significant research in the world of virtual colonoscopy. In basic terms, CAD works by filtering areas of colon surface that have dome-like shape properties to identify potential polyp candidates (Click for more information on CAD techniques). The main goals of CAD are to increase sensitivity, decrease reading time, and decrease inter-interpreter variability. A study by Yashida et al. (presented at the 2007 International Symposium for Virtual Colonoscopy in Boston) showed CAD had greater sensitivity than human readers (Table 5). While CAD is not yet used in clinical practice, it is certainly a tool that will be implemented to further improve VC performance.

Table 5. Human vs. CAD Sensitivity

Per-polyp sensitivity            
Size Combined   Duke   DoD  
  human CAD human CAD human CAD
≥ 6 mm 70% 76% 65% 76% 81% 80%
≥ 8 mm 69% 88% 63% 85% 85% 96%
≥ 10 mm 69% 90% 65% 88% 78% 94%

Resection vs. Surveillance of Small Polyps

In a January, 2008 AJR article, Pickhardt et al. quantified the benefits, risks, and costs of referring patients for polypectomy for different sized polyps. Based on their results, the 10-year risk of colorectal cancer for unresected polyps was 0.08% for diminutive lesions (<5mm), 0.7% for small polyps (6-9mm), and 15.7% for large polyps (>10mm). This calculation included the conservative assumption that large and small advanced adenomas harbored the same cancer risk. The number of diminutive, small, and large polyps needed to be removed to prevent one carcinoma was 2,352, 297, and 10.7; to prevent leaving behind one advanced adenoma it was 2,352, 297, and 10.7 polypectomies, respectively. Additionally, the cost-effectiveness ratio of resecting all diminutive and small polyps was $464,407 and $59,015 per life year gained. Polypectomy for large polyps saved $151 per patient screened. Based on the low yield and high cost of resecting all small polyps in this study, small polyp surveillance with follow-up virtual colonoscopy appears to be a reasonable alternative and should be strongly considered when making clinical decisions.