The DCBE, sometimes referred to as air-contrast barium enema, evaluates the colon in its entirety by coating the mucosal surface with high-density barium and distending the colon with air introduced through a flexible catheter that is inserted into the rectum. Multiple radiographs are acquired while varying the patient position during direct fluoroscopic evaluation and, subsequently, with conventional radiographic equipment. Colonic preparation, usually a 24-hour dietary and laxative regimen, is essential for an optimal examination. Sedation is not utilized, and the duration of the procedure averages about 20 to 40 minutes. Patients may experience mild to moderate discomfort during and after the procedure, but a prompt return to normal activity is typical.
DCBE was contemporaneously adopted as a CRC screening option by the Multi-Society Gastroenterology Consortium and the ACS in 1997 and has continued to be included among the recommended screening options in periodic updates of those guidelines,12,17,24,132 as well as those of the US Preventive Services Task Force.21 It is also considered appropriate for screening by the ACR.133 CRC screening of the average-risk population with DCBE also has been a designated Medicare benefit since 1997.134
DCBE—Efficacy and Test Performance. There have been no randomized controlled trials evaluating the efficacy of DCBE as a primary screening modality to reduce incidence or mortality from CRC in average-risk adults, and there also are no case-control studies evaluating the performance of DCBE. Further, the existing literature describing the test performance of DCBE also is limited by study designs that are retrospective and commonly do not report findings from an asymptomatic or average-risk population.135,136 In some reports, asymptomatic individuals were selected for investigation during neoplasm surveillance or after a prior screening test (eg, FSIG or FOBT). Finally, similar to the literature related to other CRC screening technologies, the DCBE literature varies considerably in terms of measurement and outcome metrics (ie, polyps, cancers, all neoplasms, adenomas, size categorizations, etc.), and these measurements may be estimated by lesion or by population.
Most studies evaluating the cancer-detection capability of DCBE utilized a methodology in which all patients in an institution- or population-based database that had been diagnosed with CRC were assessed for a history of a prior DCBE within a defined time frame, the length of which was not consistent between studies but usually ranged from 2 to 5 years. The assumption was that missed cancers on DCBE would subsequently be clinically detected. The majority of these studies showed sensitivity for cancer of 85% to 97%.137, , , , , , , , , , , , –150
Review of the literature concerning the performance of DCBE for polyps is more difficult due to the described biases and heterogeneity of study design; in particular, the target lesion and thresholds considered clinically significant often varied based upon size and/or morphology. Two studies involving truly asymptomatic individuals were performed in surveillance groups with a history of prior adenoma removal.151,152 These demonstrated sensitivities of 48% (N = 23) for adenomas >1 cm and 73% (N = 56) for adenomas >7 mm, respectively. It should be noted that in the former study, the DCBE detected 75% (6 of 8) with advanced histology.153
DCBE—Benefits, Limitations, and Harms. The potential benefits derived from the DCBE are that it evaluates the entire colon in almost all cases and can detect most cancers and the majority of significant polyps. DCBE also provides an opportunity for a full structural examination for individuals for whom colonoscopy has either failed or is contraindicated.
DCBE has several limitations. The acceptability of DCBE may be limited by the requirement for extensive colonic preparation, and some patients experience discomfort during and after the procedure. Suboptimal preparation can reduce both sensitivity and specificity. Further, there is no opportunity for biopsy or polypectomy, and any individual with findings of polyps >6 mm on DCBE should undergo colonoscopy. The lower sensitivity for significant adenomas when compared with colonoscopy may result in less favorable outcomes regarding morbidity and mortality from CRC. DCBE is also limited by the operator dependence of the radiologist or technologist performing the examination, as well as by the radiologist interpreting the examination. DCBE is a relatively safe procedure with a lower perforation rate when compared with colonoscopy (1 of 25,000 versus 1 of 1,000 to 2,000).154
Quality Assurance. The DCBE is a full structural examination of the entire colon that can be performed by radiologists or radiology residents and trained technicians under the supervision of a radiologist. Factors that can affect the quality of the DCBE examination include (1) ability to fully evaluate the entire colon due to lack of retained barium or collapse of segments of the colon; (2) adequacy of the bowel preparation; (3) patient's ability to stand and be imaged in prone and supine positions; and (4) reader's experience in interpretation. Caution is advised when performing a DCBE on the same day after polypectomy to avoid a perforation. The ACR has published guidelines that detail the basic requisites for a high-quality examination,155 as well as a quality-assurance manual for the DCBE.155 Interaction with referring physicians to correlate radiologic findings with endoscopic and/or surgical outcomes may also be an effective ongoing quality assurance in clinical practice.
DCBE—Other Issues. It is likely that the decline in the use of DCBE for CRC screening in average-risk adults will continue.156, –158 This decline in the utilization of DCBE has had an impact on training programs, as radiology residents have had less opportunity to develop the necessary skills to perform the procedure properly. Moreover, although there likely are sufficient numbers of radiologists in clinical practice who are available currently to perform DCBE studies, there has been a decline in radiologists' enthusiasm for the DCBE due to its labor-intensive nature, the low reimbursement rate, and greater interest in newer and more complex technologies such as computed tomography (CT) and magnetic resonance imaging (MRI). Based on these trends, it is likely that in the next 5 years, that there will be even fewer radiologists adequately trained to perform this procedure due to the low volume of DCBE studies currently being requested, as well as low professional interest. At present, the DCBE remains an option for direct imaging of the entire colon and may be of particular value where colonoscopy resources are limited or colonoscopy is contraindicated or less likely to be successful (eg, prior incomplete colonoscopy, prior pelvic surgery, etc.).
DCBE—Conclusions and Recommendations. DCBE every 5 years is an acceptable option for CRC screening in average-risk adults aged 50 years and older. Discussions with patients should include a description of the test characteristics, the importance of adherence to a thorough colon cleansing, test accuracy, the likelihood of a positive test, and the need for subsequent colonoscopy if the test is abnormal. The choice of DCBE for screening can be made on an individual basis, depending on factors such as personal preference, cost, and the local availability of trained radiologists able to offer a high-quality examination.
CTC, also referred to as virtual colonoscopy, is a minimally invasive imaging examination of the entire colon and rectum. CTC uses CT to acquire images and advanced 2-dimensional (2D)- and 3-dimensional (3D)-image display techniques for interpretation. Since its introduction in the mid-1990s, there have been rapid advancements in CTC technology. Multidetector CT now permits image acquisition of thin 1 to 2 mm slices of the entire large intestine well within breath-hold imaging times. Computer imaging graphics allow for visualization of 3D endoscopic flight paths through the inside of the colon, which are simultaneously viewed with interactive 2D images. The integrated use of the 3D and 2D techniques allows for ease of polyp detection, as well as characterization of lesion density and location. The 2D images also allow for limited evaluation of the extracolonic structures.
Adequate bowel preparation and gaseous distention of the colon are essential to ensure a successful examination. Patients typically undergo full cathartic preparation along with a clear liquid diet the day before the study, similar to the requirements for colonoscopy. Tagging of residual solid stool and fluid with barium and/or iodine oral contrast agents is being increasingly used and validated in large trials. At CT, a small-caliber rectal catheter is inserted into the rectum, followed by automated or manual insufflation of room air or carbon dioxide. Intravenous contrast generally is not given to patients undergoing screening but can be helpful in some patients with more advanced symptoms. Typically, the entire procedure on the CT table takes approximately 10 minutes, with no sedation or recovery time needed. Research into noncathartic approaches to minimize the bowel preparation is underway, but this technique has not yet been validated in a multicenter screening trial.159, –161 However, under conditions where same-day or next-day referral for colonoscopy would be possible, one drawback of noncathartic CTC is that a cathartic bowel preparation would still be required prior to removal of polyps.
CTC—Efficacy and Test Performance. No prospective, randomized, controlled clinical trial has been initiated (nor is one planned) to directly demonstrate the efficacy of CTC in reducing mortality from CRC. Given the cumulative body of evidence in support of CRC screening for reducing mortality and the value of polypectomy in reducing incidence, studies of CTC have focused on the detection of advanced neoplasia.
The test performance characteristics of CTC for polyp detection are derived by using optical colonoscopy (OC) as the reference standard. Early single-center CTC clinical trials involving small, polyp-rich cohorts162, –164 provided encouraging initial results and served as proof of concept that paved the way for larger multicenter screening trials. Two early trials by Cotton et al165 and Rocky et al166 included approximately 600 subjects each and observed per-patient sensitivity for large polyps of 55% and 59%, respectively. However, these 2 studies did not evaluate screening in an asymptomatic population, nor did they apply the latest CTC techniques. A more recently initiated multi-institutional screening trial using more advanced CTC techniques demonstrated more favorable performance. Pickhardt et al studied 1,233 asymptomatic adults and introduced the techniques of stool tagging and primary 3D polyp detection, neither of which were used in the 2 earlier multi-institutional trials.167 This trial reported a 94% sensitivity for large adenomas, with a per-patient sensitivity for adenomas >6 mm of 89%.
In 2005, 2 meta-analyses reviewed the cumulative published CTC performance data, including both high-risk and screening cohorts, with one analysis representing 33 studies on 6,393 patients.168,169 On a per-patient basis, pooled CTC sensitivity and specificity for large (>10 mm) polyps was found to be 85% to 93% and 97%, respectively. Pooled sensitivity and specificity for detection of small polyps (6 to 9 mm) was 70% to 86% and 86% to 93%, respectively. Of note, the pooled CTC sensitivity for invasive CRC was 96%,168 comparable with the reported sensitivity for OC.119,121
There also are a number of CTC trials currently in progress within the United States and Europe. Initial results from smaller screening trials utilizing 3D polyp detection by Cash et al170 and Graser et al171 have shown CTC performance characteristics similar to that of Pickhardt et al, providing at least a measure of independent validation for this screening technique. Also of particular interest is the recently completed ACRIN Study 6664: National CT Colonography Trial, which is sponsored and funded by the National Cancer Institute. The primary aim of this trial was to assess CTC performance for large adenomas and advanced neoplasia in a large screening cohort of 2,500 patients across 15 institutions. State-of-the-art techniques included oral contrast tagging, colonic distention with automated carbon dioxide delivery, multidetector row CT (>16 slice) with thin collimation, and both 2D and 3D polyp detection on dedicated CTC software systems. Specialized training and achievement of a high level of expertise were required of the radiologists prior to participation in the study. Preliminary findings announced at the 2007 annual meeting of ACRIN on September 28, 2007, were consistent with other recent studies using state-of-the-art techniques.
Beyond validation, a recent study demonstrated the efficacy of CTC to select patients who would benefit from therapeutic polypectomy. Kim et al recently reported comparative results from primary CTC (with selective recommendation for therapeutic colonoscopy) and primary OC screening arms among 3,120 and 3,163 mostly asymptomatic adults, respectively.172 Although this study did not randomize participants to CTC versus OC, apart from a slightly higher proportion of individuals with a family history in the OC group, the 2 groups were similar. Similar rates of advanced neoplasia were found in each group, with 3.2% in the CTC group and 3.4% in the OC group.172
CTC—Benefits, Limitations, and Harms. CTC provides a time-efficient procedure with good accuracy and minimal invasiveness. No sedation or recovery time is required, nor is a chaperone needed to provide transportation after the procedure. Time permitting, patients can return to work on the same day. However, some limitations to CTC exist, ranging from access issues to potential harms from the examination. Because CTC is relatively early in its utilization, there are fewer data relative to other CRC screening tests for evaluating benefits, limitations, and harms. Thus, continued development of best practice standards is a high priority, as is monitoring the performance of CTC as access and utilization increases. At this time, reimbursement for screening CTC is very limited, although 47 states now offer Medicare reimbursement for diagnostic CTC where the clinical indication is limited to incomplete OC.173 However, because reimbursement for screening still is uncommon, the current professional capacity to deliver CTC also is limited, although capacity is expected to increase when third-party payers begin providing reimbursement for screening.
CTC requires the same full cathartic bowel preparation and restricted diet as colonoscopy, which may decrease patient adherence. As an “imaging-only,” nontherapeutic evaluation of the colon, patients with polyps of significant size will require therapeutic colonoscopy for subsequent polypectomy. Thus, it is possible to offer same-day polypectomy to patients for whom colonoscopy is recommended without the need for additional bowel preparation, although this convenience for patients requires coordination between radiology and gastroenterology departments.174 Where such coordination does not exist, patients will need to undergo an additional bowel preparation. While older oral tagging protocols would have precluded same-day colonoscopy, revised, more efficient tagging protocols have successfully allowed therapeutic colonoscopy on the same day.
CTC is similar to endoscopy and DCBE with respect to the quality of interpretation being highly operator dependent, and thus initiatives towards training and certification are important. Detection of flat lesions has been variable, ranging from sensitivities of 13% to 65% in early CTC studies175 to 80% when using multidetector CT and combined 3D-2D polyp detection.176 However, debate continues over the prevalence and significance of flat colorectal lesions.177, –179
The accuracy of CTC is influenced by lesion size, and the sensitivity and specificity of CTC improves with polyp size. The accuracy of CTC in measuring polyp size is of particular importance since accurate size estimation is critical for appropriate patient management and for minimizing the false-positive rate. While earlier studies using rudimentary software applied to wide-slice thicknesses and 2D images showed poor concordance with prefixation polyp size,180 modern CT technology producing 3D images results in more accurate size estimates.181, –183 The ability to ensure consistent polyp size measurements during examinations is a high priority for quality-assurance initiatives since it will influence referrals for polypectomy. Pickhardt et al showed that specificity (when polyps were matched for size) was 97.4% for lesions >1 cm but declines to 84.5% for all lesions to all lesions >6mm.167 The incremental increase in the false-positive rate associated with polyps between 6 to 8 mm could add significantly to the cost of screening, and thus it will be important to monitor sensitivity and specificity in the clinical setting and identify strategies to improve specificity without diminishing sensitivity.
There is controversy over the long-term potential harms associated with radiation dose effects from CT examinations. One aspect of this controversy relates to risk-estimation models, and the other pertains to the long-term risk of cancer from single and repeated medical imaging exposures.184,185 While current estimates of the potential cancer risk related to low-dose radiation exposures during medical procedures derive from linear nonthreshold models based on long-term outcomes in survivors of acute radiation doses from atomic weapons, there is disagreement over whether this model truly is applicable to periodic exposures from medical imaging.186 In a recent position statement issued by the Health Physics Society, the health effects of low-dose radiation exposure (defined as below 50 to 100 mSv—a threshold many times higher than typical CTC levels) were considered to be “either too small to be observed or are nonexistent.”187 Nevertheless, although this risk may be theoretical, there is a growing concern that more individuals are receiving multiple diagnostic evaluations with ionizing radiation over a lifetime and that for some individuals the doses over a lifetime can reach levels that are sufficiently high to be of concern. It is important to put these issues into context with respect to screening with CTC.
Using the linear, no-threshold radiation-risk estimate, a CTC examination in a 50-year-old individual with an estimated organ dose to the colon of 7 to 13 mSv (65 mAs) is estimated to add an additional 0.044% to the lifetime risk of colon cancer.188 Because organ radiosensitivity declines with increasing age, this organ dose is halved for the same examination taking place at age 70 years. In this same evaluation, the additional lifetime risk of cancer in any site associated with a single CTC examination at age 50 years was 0.14%, although the authors stated with optimized techniques this risk could be reduced by a factor of 5- to10-fold. More efficient dose protocols using 50 mAs on 4DCT, similar to the ACR-defined protocols, have demonstrated decreased estimated organ dose ranges of 5 to 8 mSv.189 While acknowledging there is uncertainty about potential harms from single or multiple CTC screening examinations, current ACR quality metrics for CTC define low-dose parameters as a best practice for minimizing risk to patients.190
Since CTC is a minimally invasive test, the risk for colonic perforation during screening is extremely low. In the collective experience of the International Working Group on Virtual Colonoscopy, there were no cases of perforation in over 11,000 screening CTC examinations, and out of nearly 22,000 total CTC examinations (screening and diagnostic), there was only one symptomatic perforation, corresponding to a symptomatic perforation rate of 0.005%.191 Some studies of symptomatic patients, however, have reported higher perforation rates, ranging from 0.03% (1 in 3,400 patients) to 0.06% (1 in 1,700 patients).192,193 Colonic distention with low-pressure carbon dioxide delivery may be safer than insufflation of room air.191 Rates of perforation are part of the quality metrics being collected by the ACR.
Because CTC produces an image not only of the colon but also the upper and lower abdomen, there is a chance that incidental extracolonic findings will be observed. Although the overall rates of extracolonic findings have been reported to range from 15% to 69%, the incidence of clinically significant extracolonic findings at CTC has ranged from 4.5% to 11% in various patient cohorts.194, , –197 In an asymptomatic screening population, the incidence of unsuspected but potentially important extracolonic findings is approximately 4.5%, but findings of minimal or moderate potential clinical significance, such as cholelithiasis (6%) and nephrolithiasis (8%), are more common.197 While there are potential benefits from serendipitous findings, there also are associated risks and costs that need to be considered when these findings are false positives. These include further radiologic imaging and, thus, added organ dose, potential for adverse outcomes associated with tissue sampling for abnormalities that are not resolved with additional imaging, as well as the direct and indirect costs to the patient. The implementation of structured reporting of extracolonic findings and monitoring trends in subsequent diagnostic workups and adherence with quality metrics are being evaluated through the National Radiology Data Registry (NRDR), the ACR's national data warehouse.
Quality Assurance. Similar to the call to action for measuring quality of colonoscopy,198 the implementation of CTC will require quality metrics to be defined and implemented in clinical practice. Quality of CTC examinations will depend on (1) proper bowel preparation; (2) adequate insufflation of the colon and appropriate use of CTC technique parameters at image acquisition; (3) adequate training of the interpreting physician in the use of 2D- and 3D-image display techniques; and (4) documentation of clinically significant colonic and extracolonic lesions to referring physicians. In 2005, the ACR Practice Guidelines for the Performance of Computed Tomography (CT) Colonography in Adults was published, encompassing the techniques, quality control, clinical uses, training, and communication of results for CTC.190 An update of these guidelines is planned following publication of the results of the ACRIN CTC screening trial. In 2006, the ACR Colon Cancer Committee outlined practice-based quality metrics for CTC, encompassing process measures of CTC technique and image quality; patient preparation; and outcomes measures such as rates of true positives, colonic perforation, and incidence of extracolonic findings. These quality metrics are to begin a pilot phase in late 2007, with data entry in the NRDR database. The ACR has begun construction of an interactive, hands-on training facility for CTC and will begin training courses in early 2008. A process for individual certification and proficiency is being evaluated.
CTC—Other Issues. Standardization of the evolving technology and consensus related to the reporting of findings will be essential for effective implementation of CTC screening. A consensus statement of a standardized reporting structure for CTC findings was recently published, modeled after the Breast Imaging Reporting and Data System's (BI-RADS) reporting of mammography.199 This reporting structure, termed the “CT Colonography Reporting and Data System (C-RADS),” describes how to report lesion size, morphology, and location, with a summary category score per patient.
The management of CTC findings is an important part of a CTC screening program. At this time, there is consensus that all patients with one or more polyps >10 mm or 3 or more polyps >6 mm should be referred for colonoscopy.200 The management of patients with fewer polyps (<3) in which the largest polyp is 6 to 9 mm remains controversial. Such polyps are routinely removed if found at OC because of the opportunity and the risk, albeit low, of advanced neoplasia. However, in studies that have been limited to screening cohorts, among individuals whose largest polyp is 6 to 9 mm in size, the prevalence of advanced features tends to be low (3.4% to 6.6%).201,202 At this time, there is ongoing research using CTC surveillance to evaluate the natural history of polyps in this size range. Based on expert consensus and until further evidence is available to provide additional guidance, a reasonable approach at this time for patients with 6 to 9 mm polyps identified on CTC is to recommend therapeutic colonoscopy. Patients who decline referral to colonoscopy or who are not good candidates for colonoscopy should be offered surveillance with CTC.
Optimal management of patients whose largest polyp is <6 mm detected on CTC is uncertain. Experts from the American Gastroenterological Association, the American College of Gastroenterology, and the ACR have reported a range of policies on how to handle these lesions.190,203,204 There is general agreement that the risk of advanced features in patients whose largest polyp is ≤5 mm is very low. In a recent study that is able to provide this estimate in a screening cohort, the prevalence of advanced neoplasia in patients whose largest polyp was ≤5 mm was 1.7% (D.A.L, personal communication, December 14, 2007).202 At this time, there is a pressing need for multidisciplinary consensus on the reporting and clinical management of patients whose largest polyp is <6 mm.
CTC—Conclusions and Recommendations. In terms of detection of colon cancer and advanced neoplasia, which is the primary goal of screening for CRC and adenomatous polyps, recent data suggest CTC is comparable to OC for the detection of cancer and polyps of significant size when state-of-the-art techniques are applied. In previous assessments of the performance of CTC, the ACS concluded that data were insufficient to recommend screening with CTC for average-risk individuals.19 Based on the accumulation of evidence since that time, the expert panel concludes that there are sufficient data to include CTC as an acceptable option for CRC screening.
Screening of average-risk adults with CTC should commence at age 50 years. The interval for repeat exams after a negative CTC has not been studied, and is uncertain. However, if current studies confirm the previously reported high sensitivity for detection of cancer and of polyps ≥6 mm, it would be reasonable to repeat exams every 5 years if the initial CTC is negative for significant polyps until further studies are completed and are able to provide additional guidance. Until there is more research on the safety of observation, colonoscopy should be offered to patients whose largest polyp is 6 mm or greater. CTC surveillance could be offered to those patients who would benefit from screening but either decline colonoscopy or who are not good candidates for colonoscopy for one or more reasons. However, if colonoscopy is contraindicated because the patient is not likely to benefit from screening due to life-limiting comorbidity, then neither CTC nor any other CRC screening test would be appropriate.