Computed Tomography Colonography and Evaluation of the Colon

Computed tomography colonography (CTC) is a low-dose, cross-sectional imaging examination optimized for the detection of colorectal polyps and masses. As a result of advanced computer three-dimensional (3D) post-processing, a popular misperception is one of an imaging analogue to optical colonoscopy in which a 3D model alone is used to view the mucosal surface of the colon. However, CTC is fundamentally a different examination, using both 2D cross-sectional CT images and 3D datasets in the interpretation of mucosal-based processes, deeper intramural structures and surrounding pericolonic fat. Due to its CT underpinnings, CTC also allows a limited evaluation of the abdominal and pelvic contents besides the colon. CTC has now been in general clinical use since the early 2000s. Performance in low-polyp prevalence screening populations has been validated in a number of prospective clinical trials. There is now general consensus that CTC performs substantially better than double-contrast barium enema (DCBE) and is equivalent to optical colonoscopy for clinically relevant polyps and masses, including flat serrated lesions.

This chapter will provide a comprehensive overview of this modality. The various indications and uses for this exam will be discussed. Major trials and studies will be highlighted. The technical components that comprise CTC including bowel preparation, colonic distention, and image acquisition will be discussed. Interpretation will be covered in detail, including the mechanics of the interpretative process. Common and important pitfalls that can affect the accuracy of this process will be highlighted. Finally, pertinent issues associated with CTC will be examined, including radiation dose, complications, and extracolonic (incidental) findings.

Indications and Use

There are three broad categories of use for CTC: (1) screening for colorectal cancer (CRC); (2) focused evaluation for various diagnostic reasons, including incomplete colonoscopic screening; and (3) simultaneous colonic evaluation and distant staging of initially diagnosed CRC or surveillance of known CRC. Many of the indications previously undertaken by DCBE have been now supplanted by CTC. On the other hand, CTC is not intended for situations better handled by single column barium or water-soluble contrast examinations, such as postoperative anastomotic leak checks or evaluation for fistulas related to disease processes such as diverticulitis. Such indications remain better evaluated by fluoroscopic means.

SCREENING FOR COLORECTAL CANCER

Screening for CRC is perhaps the most important indication for CTC, for which its use could have a major public health impact. CRC is currently the second leading cause of cancer deaths in the United States; there are approximately 143,000 new cases/year, leading to almost 52,000 deaths annually. Screening is particularly effective in decreasing CRC mortality because of several favorable factors in its biology: (1) CRC has a stepwise progression, from a benign precursor lesion to cancer; (2) the benign precursor can be identified and removed; and (3) the benign precursor has an extended latent period (10–15 years) prior to transformation to cancer, leading to a wide window for detection and intervention. As opposed to other cancers, such as breast cancer, in which the intent is to detect early cancer, CRC screening represents true primary prevention, in which the removal of a benign precursor lesion disrupts the future development of a cancer.

However, it is estimated that of the 80 million U.S. adults eligible for screening, only slightly over 50% participate with any of the traditional options, including optical colonoscopy, fecal occult blood test (FOBT)–fecal immunochemical test (FIT), flexible sigmoidoscopy, and DCBE. ^, In addition, there are concerns regarding population screening capacity for options such as optical colonoscopy. This is further accentuated by the growing awareness that specialized gastrointestinal (GI) capacity is uneven and is dependent on geographic region, which is particularly limited in rural areas. Thus, CTC has the potential to increase screening adherence substantially by appealing to currently noncompliant persons ^, and by increasing the screening capacity, particularly in rural underserved areas, where examinations can be conducted at the local hospital and networked to central areas of expertise for interpretation.

There is currently a menu of screening options for CRC. However, a case could be made that CTC represents the optimal choice among all options. CTC has a similar performance profile to colonoscopy. CTC detects both target polyp precursors and cancers with a sensitivity/specificity of 90%–94% and 86%–96%, respectively, for large polyps (≥10 mm) ^, and sensitivity of 96.1% for cancers. In contrast, stool studies (such as FOBT, FIT, and stool DNA) are largely limited to the detection of early cancers and miss most benign precursor lesions which do not typically bleed or shed abnormal DNA. Sensitivity for large polyps is poor, ranging from 22% to 40% for FOBT/FIT and 42% for stool DNA. The case for CTC is strengthened due to a much safer risk profile than colonoscopy despite similar performance. Whereas the perforation rate is 0.1% for screening colonoscopy, it is near nonexistent for CTC at 0.009%. In addition, because sedation is not required at CTC, additional complications related to these medications are obviated.

Screening by CTC is appropriate for average-risk individuals (≥50 years old), which represent the largest cohort. CTC-based screening is also appropriate for individuals of this age group who have a somewhat elevated risk because of a positive family history. Family history can be defined as a history of cancer in one first-degree relative or two second-degree relatives. Although somewhat controversial, CTC can also be used in patient with a positive stool-based test. Lesion detection and negative predictive value are high, but questions have been raised about cost-effectiveness because of the higher underlying polyp and cancer prevalence in this group. ^, ^, Finally, CTC can be substituted when screening colonoscopy is contraindicated in a patient because of comorbidities and sedation issues or if the patient is anticoagulated. Particularly in the setting of warfarin use, CTC is an excellent alternative for screening because the anticoagulation therapy does not have to be interrupted for the exam. For the small percentage of positive patients who require removal of a large polyp, the anticoagulation can be reversed at a future date for the therapeutic polypectomy.

In contrast, screening by CTC is not indicated for individuals at high risk for colorectal carcinoma and polyps for whom the pretest probability for a positive examination is substantially increased. These include patients with a polyposis syndrome (e.g., familial adenomatous polyposis and hereditary nonpolyposis colorectal cancer syndrome) and inflammatory bowel disease. Patients with these underlying diagnoses are better evaluated by optical colonoscopy.

DIAGNOSTIC INDICATIONS

CTC can also be used for the evaluation of various diagnostic situations. It represents a less invasive alternative to colonoscopy for the workup of clinical scenarios, such as altered bowel habits, iron deficiency anemia, and nonspecific weight loss, in which one of the concerns is cancer. In terms of cancer detection, a large meta-analysis (49 studies; compiled cohort N = 11,151) demonstrated that CTC had a high sensitivity of 96.1%. Unlike colonoscopy, there is less concern that detection ability with CTC decreases for right-sided cancers. Several population-based studies have indicated decreased detection or protective effects of colonoscopy for proximal CRCs.

In addition to symptomatology, CTC is often used to complete colonic evaluation in the situation of an incomplete colonoscopy. ^, Whereas same-day barium enema following incomplete colonoscopy may have been difficult because of incomplete colonic coating caused by air block, this is not a concern with CTC. Typically, same-day CTC examinations of good diagnostic quality can be performed. The bowel preparation is often supplemented by oral contrast tagging agents prior to scanning. A typical protocol involves administering 30 mL of diatrizoate after the patient recovers from sedation to minimize aspiration risk. Scanning occurs approximately 2 hours after contrast administration, allowing for complete colonic fluid tagging in most patients. However, some concerns exist that serrated polyps detection may be compromised with a same-day CTC without a dual tagging regimen that also includes barium. ^,

STAGING AND SURVEILLANCE

CTC offers a unique opportunity for the initial staging of CRC and for surveillance following surgical resection. Because of its cross-sectional nature, CTC can assess the colon and extracolonic structures. In both situations, the standard protocol is typically altered; the prone series is undertaken first at low dose, without intravenous (IV) contrast. This is followed by the supine series, in which a standard-dose CT with IV contrast is performed to allow optimal extracolonic evaluation for metastatic disease.

In staging, CTC excels in the localization of the primary cancer, particularly in tortuous and elongated colons, in which such localization is difficult for colonoscopy, and for the detection of synchronous polyps and cancers. Although several studies have reported decent discriminatory ability in local tumor staging, ^, ^, there is general consensus that this evaluation is better undertaken by magnetic resonance, which can better resolve extramural extension. In addition to colorectal tumor evaluation, CTC can also evaluate for regional lymph nodes and distant metastases. With minor image parameter changes, CTC is essentially a contrast CT examination, which currently is the modality of choice for metastatic evaluation.

Surveillance after CRC treatment is an emerging application for CTC. ^, Similar to colonoscopy, CTC allows for intraluminal evaluation for anastomotic recurrences as well as for metachronous polyps and cancers, similar to colonoscopy. Unlike colonoscopy, however, CTC is not limited to the colorectum, where its cross-sectional nature allows for extraluminal evaluation, including extraluminal pericolonic recurrences, regional adenopathy, and distant metastases. Kim and colleagues reported a large Asian, single-institution experience. CTC detected six metachronous cancers, including one anastomotic recurrence in 548 patients undergoing routine surveillance without clinical suspicion or increasing carcinoembryonic antigen levels following prior CRC resection and treatment. The per-patient sensitivity was 81.8% (18 of 22) for advanced adenomas and 100% for cancer. CTC was also able to detect extracolonic recurrence in an additional 11 patients, which could not be seen at colonoscopy.

History of Computed Tomography Colonography Trial Results and Supporting Evidence

A number of trials in high-polyp prevalence groups and low-prevalence screening cohorts have been conducted to evaluate CTC performance. The high-polyp prevalence population studies represented initial proof of concept studies. Typically performed as single-institution series, these studies showed that CTC could detect polyps with high sensitivities, albeit in the least difficult situation, in which a relatively large percentage of examinations were positive. One notable study from this group was a prospective trial from Boston University published in the New England Journal of Medicine in 1999. Fenlon and colleagues enrolled 100 high-risk individuals to undergo both CTC and colonoscopy evaluation. Despite using now outdated technology and protocols (including no fecal tagging, room air distention, and 5-mm image collimation reconstructed at 2-mm intervals), CTC demonstrated a 91% (20 of 22 polyps) sensitivity at the 10-mm threshold, which decreased to 82% (33 of 40) for 6- to 9-mm polyps. All cancers in the series (3 of 3) were seen with CTC.

The positive results seen in these small, single-institution series then paved the way for large multicenter trials in low-polyp prevalence or screening cohorts. These studies assessed whether CTC performance could be maintained in the more challenging and pertinent clinical situation in which most of the examinations would be negative. Here, there were discrepant results, leading to controversy regarding the true capabilities of this modality. First, Pickhardt and colleagues reported the results from the multicenter Department of Defense (DoD) CTC trial in 2003. In a large cohort ( N = 1233 screening individuals, three institutions), the per-patient sensitivity for adenomatous polyps was 94% at the 10-mm threshold, decreasing to 89% at the 6-mm threshold. In comparison, optical colonoscopy sensitivity was 88% at the 10-mm level and 92% at the 6-mm level. The lower than traditionally reported sensitivity levels for colonoscopy were likely related to trial design, in which an enhanced polyp validation system of segmental unblinding was used. In contrast to using colonoscopy in isolation as the gold standard, in which colonoscopy misses would be unaccounted for, segmental unblinding created an enhanced reference standard in which the CTC results were revealed segmentally at colonoscopy. Thus, if no polyp was initially seen at colonoscopy but noted with CTC, the colonoscopist reexamined the segment for the missed polyp. Such a standard led to more accurate sensitivity and specificity measures, correcting the situation in which a false-negative colonoscopy result would be inaccurately counted as a CTC false-positive result.

Following the DoD results, however, two smaller multicenter trials demonstrated poorer results. ^, Rockey and colleagues ( N = 614) and Cotton and colleagues ( N = 615) reported sensitivities ranging from 55% to 59% for large polyps (≥10 mm) and decreasing to even lower levels for subcentimeter polyps. At the time, this led to great debate about the adequacy of CTC performance. However, with the results of the subsequent larger validation trials published several years later, it is now evident that the trials by Rockey and colleagues and Cotton and colleagues were flawed, accounting for the observed trial results. In addition to using older CTC technology, one major criticism concerned the lack of reader training. For example, the trial by Cotton and colleagues required no prior reader experience or training for the CTC readers. It is now evident that there is a learning curve for CTC interpretation that requires acquisition of CTC-specific skills in addition to existing CT cross-sectional skills. Liedenbaum and colleagues demonstrated such a learning curve in which substantial improvement among novice readers was seen through a study set of 200 CTC cases. Sensitivity increased from an initial value of 76% at the 6-mm threshold for the first set of 50 CTC cases to 91% sensitivity for the fourth set of 50 cases. Furthermore, studies suggested that training and experience may not be enough and that competency testing may ultimately be required.

During this evaluation period of CTC performance, it is important to be aware that CTC was a dynamic technology with continuing changes in the underlying CTC software, hardware, and examination protocol. Technical components were refined, leading to more optimally prepared and tagged bowel, improved and consistent colonic distention, and increased spatial resolution with decreased artifact. ^, ^, The CTC protocol evolved, such as standardizing the use of both supine and prone imaging for the typical examination, with additional decubitus series for problem solving. ^, ^, Underlying improvements in computer hardware allowed easy manipulation of even larger volumetric datasets. The current CTC examination, as noted earlier, now reflects the tremendous advancements over the years.

In 2008, the results of the ACRIN 6664 protocol or National CT colonography trial were published, answering the controversy about the discrepant initial screening results. This landmark trial validated the prior excellent results seen by Pickhardt and colleagues. It was a large multicenter trial ( N = 2531 participants, 15 institutions) performed in a low-prevalence or screening cohort. It used state-of-the-art techniques and required a high level of experience and competence testing by the readers. In this trial, a per-patient sensitivity of 90% and specificity of 86% was seen for large polyps (≥10 mm). The performance decreased to a sensitivity of 78% at the 6-mm threshold. In addition to the ACRIN trial, other series in Europe corroborated a high level of CTC performance. The Italian IMPACT trial ( N = 937, 12 institutions) showed a 90.8% sensitivity (84.5% specificity) at the 10-mm threshold, and the Munich Colorectal Cancer screening trial also showed similar results (92% sensitivity, 98% specificity). The United Kingdom SIGGAR trial demonstrated equivalent detection rates for cancers and large polyps between CTC and colonoscopy screening programs.

Outside of trial data, observational outcomes from large-scale clinical practices suggested that CTC-based screening could be effective. A large Wisconsin series by Kim and colleagues reported results from parallel operating CTC ( N = 3120) and colonoscopy ( N = 3163) CRC screening programs, in which each drew from the same regional population. The CTC program reported advanced neoplastic yields equivalent to the colonoscopy program ( N = 123 CTC; N = 121 colonoscopy) but with a fourfold decrease in the number of polypectomies (561 vs. 2434) and subsequent marked decrease in the number of complications (0 vs. 7 perforations).

In addition, data regarding longer-term outcomes for CTC is now available, reinforcing the clinical practices of setting the detection threshold at 6 mm to ignore diminutive polyps, a 5-year routine follow-up for negative screening CTC exams ^, as well as the safety of 3-year imaging surveillance for one or two isolated subcentimeter polyps. CTC has allowed the confirmation of the natural history of small colorectal polyps as low-risk, indolent lesions with only a minority transforming to cancer over many years. In 2016, the CTC experience with serrated polyps has been reported and appears to detect these lesions despite their flat, subtle nature.

Computed Tomography Colonography Exam

TECHNICAL COMPONENTS

CTC is a multicomponent examination that is undertaken over several days. The major technical components include bowel preparation, colonic distention, and image acquisition. It is key to optimize these components to allow high-quality interpretation.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here