Standards for Gastroenterologists for Performing and Interpreting Diagnostic Computed Tomographic Colonography

Prominent among a number of new techniques with which to image the colon, computed tomographic (CT) colonography is extremely attractive because it is noninvasive and also relatively simple for patients to undergo. As the technology evolves, it is important that gastroenterologists not only understand the multiple issues surrounding CT colonography but also that they be able to interpret this examination.¹ The American Gastroenterological Association (AGA) Institute's Governing Board convened the CT Colonography Task Force to develop training standards for gastroenterologists for CT colonography. These standards are intended to outline the basic requirements that board-certified gastroenterologists should meet to be involved in and/or perform CT colonography. All recommendations are based on the literature available at the time this manuscript was developed. A wide range of sensitivities have been reported for CT colonography; therefore, the current use of CT colonography in clinical practice is controversial. Several studies have evaluated the use of CT colonography after failed colonoscopy; its sensitivity for detecting important lesions is comparable with or better than results with air contrast barium enema (ACBE). CT colonography appears to also be useful for evaluation of the colon proximal to an obstructing lesion. Minimal data are available regarding the use of CT colonography as a screening test in patients with contraindications to colonoscopy or who refuse other screening options. The results of studies using CT colonography as a colorectal cancer (CRC) screening test suggest that this is an area requiring further study. The use of CT colonography for CRC screening is currently controversial and this test has not yet been endorsed as a primary CRC screening tool in asymptomatic, normal-risk adults by any multidisciplinary group involved in CRC screening guideline development. CT colonography has few contraindications; however, it should not be performed in patients in whom perforation is a risk and should probably not be performed immediately after failed colonoscopy in patients who had polyps removed or large biopsy specimens taken during colonoscopy because of the risk of perforation from colonic insufflation. Specific clinical circumstances may also exist in which endoscopic examination is preferred to CT colonography (such as patients with known inflammatory bowel disease, high-risk symptoms, and others). Overall, the Task Force finds that CT colonography is appropriate in certain circumstances and has developed the following recommendations to guide gastroenterologists who are interested in performing CT colonography. CT scanning should be performed by American Registry of Radiologic Technologists-certified radiologic technologists. The extent of training for gastroenterologists to read accurately CT colonography has not been fully defined. However, research shows that response to training is unpredictable, and the "learning curve" for CT colonography interpretation will vary widely among observers. Available literature suggests that review of at least 75 endoscopically confirmed cases is appropriate as a requirement for minimal competence in detecting and characterizing colorectal neoplasia detected by CT colonography. Subsequently, interpretation under the supervised guidance of a qualified physician mentor is required. To maintain clinical expertise in CT colonography after formalized training, physicians should supervise and interpret a minimum number of cases per year, in addition to participating in continuing medical education activities, and update them relating to advances in the field. Most bowel preparative regimens employ a cathartic agent, the selection of which will depend on patient factors as well as physician preferences. Fecal and fluid tagging may permit identification of submerged polyps and reduce false-positive examinations. CT colonography performed without a bowel purge is an area of great promise but cannot currently be recommended because no large clinical studies have verified its performance in a large cohort. Colonic insufflation with automated insufflators results in improved colonic distention compared with manual insufflation. High-resolution CT is performed in the supine and prone positions following review of an initial CT scout. CT colonography evaluation involves the following 2 steps: first, a primary search for suspicious colonic lesions and, second, lesion characterization. The primary search can be achieved using either a primary 2-dimensional (2D) search or a primary 3-dimensional (3D) search; optimal performance likely involves both search methods. Lesion characterization includes determination of lesion density and lesion mobility. Reading: All intracolonic findings should be examined, and any segment not adequately evaluated should be documented. All large masses and lesions that compromise luminal caliber should be communicated. The size and location of colorectal lesions should be reported. Extracolonic findings are common, but the majority of these lesions are not clinically significant and do not require follow-up. Characterizing these extracolonic lesions requires expertise in recognizing abnormalities of the lungs, the solid organs, the retroperitoneum, and the extracolonic gastrointestinal tract. A radiologist should review the extracolonic portion of the study. Reporting: A standardized CT colonography report should encompass elements of preprocedure documentation, patient demographics, indications, technical description, findings, clinical assessment, and recommendations (plan) for follow-up. Reporting by polyp size is controversial. General agreement exists that all polyps ≥10 mm should be reported. However, full consensus relating to the reporting or management of subcentimeter polyps discovered at CT colonography has not been reached. The referral of patients to endoscopy for diminutive lesions (when CT colonography specificity is low) could lead to inappropriate referrals to colonoscopy. Moreover, current CT colonography acquisition parameters are tailored to the detection of polyps 6 to 10 mm in diameter. Based on these considerations, it is recommended that all polyps 6 mm or larger should be reported. Controversy exists for small lesions; these should be reported when reader confidence is very high. Extracolonic findings should be reported. A comprehensive technical and professional quality control program is necessary. Technical quality control should encompass both the CT scanner and the CT colonography workstation. Professional quality assessment monitors outcomes within a practice for internal quality assessment purposes. Such measures will alert physicians that changes may need to be made in patient educational materials, patient preparation regimens, or interpretation techniques. Retrospective, sporadic review of CT colonography parameters and reports can also ensure that appropriate technique and practice patterns are being followed. Standardized practices followed by all physicians and allied health personnel within a practice can also improve patient safety. Regulatory Issues: Federal anti-kickback laws and Stark statutes influence who can perform CT colonography as well as the subject of split interpretation (a situation in which one physician interprets intracolonic images and another performs the extracolonic images). Both performing and interpreting CT colonography constitute "designated health services" and are therefore subject to Stark statutory requirements regarding referrals and billing for split interpretation. Compensation arrangements in a situation in which there is dual interpretation are potentially complicated but should not exclude any group from reading CT colonography. A personal services and management agreement ("safe harbor") is a potentially applicable compensation arrangement between the gastroenterologist and the radiologist in a split interpretation scheme. Key Executive Summary Recommendations: The key Task Force recommendations related to the basic requirements that board-certified gastroenterologists should meet to be involved in and/or perform CT colonography are summarized below. A complete list of recommendations is included in the full Task Force report.•CT colonography is effective for evaluation of the colon proximal to an obstructing lesion.•CT colonography is indicated for adults with failed colonoscopy in whom evaluation of the colon is deemed necessary.•Minimal data are available regarding the use of CT colonography as a CRC screening test in patients with contraindications to colonoscopy or those who refuse other screening options. CT colonography may be considered in patients unwilling to undergo other primary screening modalities.•Based on currently available data, CT colonography is not endorsed as a primary screening modality for CRC in asymptomatic adults.•Training for CT colonography interpretation should include review and interpretation of at least 75 cases with endoscopic correlation.•Subsequent to formal training, the gastroenterologist should participate in a mentored CT colonography preceptorship lasting 4 to 6 weeks, occurring within 6 months of the initial training, with the candidate physically present and involved in the interpretation of at least 25-50 additional cases.•It is expected that those performing CT colonography will undertake ongoing training and self-assessment including attending formal continuing medical education accredited courses in CT colonography.•Gastroenterologists should work collaboratively with board-certified radiologists to review the extracolonic portion of the CT colonography examination.•Any polyp ≥6 mm in size (ie, widest diameter) should be reported and the patient referred for consideration of endoscopic polypectomy...