| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Clinical Radiology, St. James's University Hospital, Leeds, West
Yorkshire LS9 7TF, United Kingdom.
2 Department of Clinical Radiology, Derriford Hospital, Plymouth, Devon, United
Kingdom.
Received February 7, 2007;
accepted after revision May 20, 2007.
Address correspondence to D. J. M. Tolan
(djmtolan{at}doctors.net.uk).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We performed a retrospective analysis of 400 consecutive patients older than 70 years undergoing CTC over a 14-month period. All patients presented with weight loss, alteration of bowel habits, rectal blood loss, abdominal pain, or anemia; these symptoms led to clinical suspicion of lower gastrointestinal abnormalities.
RESULTS. Five hundred five separate extracolonic abnormalities were detected in 268 of 400 patients (67%). One hundred thirty-nine pathologic processes were deemed significant in 116 patients. Of these, 110 lesions (79%) were previously unknown in 96 of the 400 patients (24.0%). Forty-nine of the 400 patients (12.3%) had at least one malignancy, including 23 extracolonic malignancies and 29 colorectal malignancies. Thirteen patients had early cancers (T1N0M0 or T2N0M0). Twenty of the colon cancer patients had significant previously unknown extracolonic abnormalities, half of which were related to the primary tumor and half of which were unrelated extracolonic abnormalities.
CONCLUSION. In patients older than 70 years being examined because of lower gastrointestinal symptoms, CTC findings yield a high number of new significant extracolonic abnormalities. This finding makes a compelling case for targeting this group of patients for a CTC service.
Keywords: barium enema • colorectal cancer • colorectal cancer screening • CT colonography • elderly patients
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
10 mm) of at least 90%, and its sensitivity for
colorectal cancer approaches 100%
[1–6]. CTC is well tolerated in comparison with colonoscopy [7] and barium enema [8] and is a safe technique; a recent United Kingdom survey by Burling and colleagues [9] recorded adverse events in 0.08% of symptomatic patients undergoing CTC, with no deaths, and symptomatic perforations in 0.03%. CTC requires less patient movement than barium enema, which is a particular advantage in elderly patients or infirm patients.
In many hospitals, however, because of equipment availability and staff experience issues, barium enema remains the primary investigation for patients referred for lower gastrointestinal symptoms. At our institution, after decommissioning a fluoroscopy room and using that space to house a new MDCT scanner, all patients older than 70 years referred for imaging because of lower gastrointestinal symptoms had CTC performed if fit for bowel cleansing. We describe our experience with CTC examinations in this age group.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All CTC examinations were performed on either a 4-MDCT scanner or a 16-MDCT scanner. Patients received full bowel preparation either with sodium picosulphate plus magnesium citrate (Picolax, Nordic) or, for those with impaired renal function or mild congestive cardiac failure, with polyethylene glycol (Kleen-Prep, Norgine).
For each CTC examination, a thin rectal tube was inserted, and the small retention balloon was inflated for the first series only. Colonic insufflation was achieved with carbon dioxide using an automated device (PROTOCO2L, E-Z-EM). Unless there was a contraindication to contrast material, patients received 20 mg of hyoscine-N-butylbromide (Buscopan, Boehringer Ingelheim) IV and 100 mL of iodinated contrast material (iohexol [Omnipaque 300, GE Healthcare]) for those with normal renal function or of isoosmolar contrast material (iodixanol [Visipaque 300, GE Healthcare]) for those with impaired renal function. If a patient's creatinine level was greater than 200 mg/mL, no IV contrast material was administered (five patients). Patients who had diabetes and were being treated with metformin stopped treatment with metformin for 2 days after receiving IV contrast material.
Scans were obtained with the patient in the supine position followed by either the prone position or the lateral decubitus position if the prone position was not tolerated. The rectal balloon was deflated for prone imaging to allow adequate visualization of the lower rectum. Data sets were acquired with a 2.5-mm (4-MDCT) or 1-mm (16-MDCT) reconstructed slice thickness, respectively, at 120 kV and a minimum of 100 mAs with automatic dose modulation. All supine data sets were acquired 45 seconds after the administration of IV contrast material, followed by a prone series. If a cancerous tumor, either colonic or extracolonic, was seen while a patient was being scanned, chest imaging for staging was included as part of the examination.
Data sets were reported by one of four gastrointestinal radiologists using a workstation (Leonardo, Siemens Medical Solutions) with CTC software. At the time of the study, all the reviewers had attended a European Society of Gastrointestinal and Abdominal Radiologists CTC training course and had a combined experience of performing and reporting more than 1,000 CTC examinations. A primary 3D review was followed by 2D analysis. Polyp size was assessed by manual 2D measurement. The whole data set from the supine IV contrast–enhanced imaging data set was reviewed on axial with coronal multiplanar reformatted images to exclude extracolonic abnormalities.
Radiology reports and laboratory studies, including pathology results, case notes, and records from the hospital database, were analyzed after a mean interval of 12.5 months (range, 7.0–20.5 months) after imaging to determine the frequency and significance of extracolonic abnormalities and outcome. The extracolonic lesions that we categorized as significant (Table 1) were radiologic abnormalities normally associated with significant mortality or morbidity or were possibly benign conditions requiring confirmation with other tests, similar to the categorization used by Ng et al. [10].
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Colonic Abnormalities
Colorectal cancer—The true incidence of all primary
colorectal malignancy in the study cohort was 7.5% (30/400)
(Fig. 1 and
Table 2).
|
Thirty-six lesions of malignant appearance were diagnosed in 34 patients (two patients having synchronous lesions). In five patients, histology results were not obtained: four had unequivocal extensive metastatic disease with a clear colonic primary tumor and the fifth patient was too ill to undergo further investigation. Of the 29 patients with colonoscopic or surgical histologic correlation, colorectal adenocarcinoma was confirmed in 24, cecal carcinoid tumor in one, prostate carcinoma directly invading the rectum in one, and large severely dysplastic tubulovillous adenomas in two.
Because of the relatively short follow-up period of this study, the true-negative detection rate for colorectal cancer cannot be determined, and a specificity calculation would be meaningless. Two patients, one of whom had proven synchronous cecal carcinoma, had a sigmoid stricture that was thought at CTC to be a malignant tumor but was found to be diverticulitis at histology.
A false-negative case, the only one in our series, arose in a patient with a sigmoid diverticular stricture perforation and pelvic abscess in whom a low, flat rectal tumor was missed because of persistent nondistention of the rectum. The tumor was detected at subsequent sigmoidoscopy, which was performed as recommended by the reporting radiologist.
If the cases of prostate cancer, carcinoid, and large tubulovillous adenoma are counted as false-positives and if the patient diagnosed as too sick for follow-up is excluded, the sensitivity per patient for colorectal adenocarcinoma in our group is 84.8% (28/33). This sensitivity rises to 97% (32/33) when all significant malignancies of or involving the colon and large, severely dysplastic tubulovillous adenomas are included.
Significant polyps—Only large polyps ( 10 mm) were
deemed significant in this symptomatic group of patients older than 70 years.
Sixteen large polyps (mean, 15 mm; range, 10–26 mm) were detected at CTC
in 14 patients (14/400, 3.5%); one patient had three large polyps, and another
had a large polyp and a synchronous rectal cancer. A third patient had a flat
cecal polyp measuring 20 mm that was missed at CTC and was discovered when
colonoscopy was performed to remove a correctly identified 26-mm sessile polyp
in the ascending colon. Nine patients had endoscopic correlation. Nine polyps
were found with eight true-positive, one false-positive, and one
false-negative case (sensitivity = 88.9% per polyp). All of these large polyps
were either villous or tubulovillous adenomas with dysplasia.
Diverticular disease—One hundred twenty-five patients (31.3%) had uncomplicated diverticular disease of the colon alone. Complications of diverticular disease were present in another 15 patients (3.8%): seven had evidence of active diverticulitis, five had diverticular abscesses, two had diverticular strictures, and one had a diverticular colovesical fistula. In eight patients, diverticular disease complications were the only abnormality on CTC.
Extracolonic Abnormalities
Five hundred five separate extracolonic abnormalities were detected in 268
patients (67%) (Table 3). If
all the findings were attributable to a single pathologic process (e.g., lung
cancer with mediastinal lymphadenopathy and adrenal and bone metastases),
these findings were considered one entity. One hundred thirty-nine separate
significant pathologic processes were detected in 116 of the 400 patients
(29%), including 10 in patients with metastatic colorectal malignancy.
|
One hundred ten of these 139 processes (79%) were previously unknown, present in 96 of the 400 patients (24.0%). Twenty-one patients had two or more significant extracolonic abnormalities (5.3%).
Patients with significant extracolonic abnormalities were older (80.7 vs 79.2 years, p = 0.019) and were significantly more likely to be male (65/250 [26%] female vs 51/150 [34%] male, p = 0.0001) than those with insignificant abnormalities or no extracolonic abnormalities. Insignificant abnormalities were more frequent in female patients (106/250 [42.4%] females vs 46/150 [30.7%] males, p =0.016).
Malignancies—Primary colorectal malignancy diagnosed at CTC was confirmed in 29 patients, which included five early cancers (T1N0M0 or T2N0M0). There were also two large severely dysplastic tubulovillous adenomas. Twenty-two previously unknown extracolonic malignancies were detected (pathologic or unequivocal radiologic diagnosis), of which eight were early cancers (T1N0M0 or T2N0M0). In all, there were 52 cancers in 49 patients (three patients had two cancers) (Fig. 2); 26.5% (13/49) of these patients had early-stage tumors.
|
Extracolonic abnormalities in patients with colorectal cancer found at CTC— Twenty-nine patients were diagnosed with primary colorectal malignancy either by histology or the presence of extensive metastatic disease with a primary colonic tumor at presentation. Twenty-six (89.7%) of these 29 patients had extracolonic abnormalities, 20 (69.0%) of whom had significant previously unknown abnormalities.
Of the 20 patients, 10 (50%) had findings directly related to the primary colonic tumor or its management, such as metastasis, ureteric or small-bowel obstruction, or indeterminate liver lesions requiring further characterization. The other 10 patients (50%) had significant new extracolonic abnormalities unrelated to the primary tumor, including two patients with lung cancer and one with a large intramuscular lipoma suspicious for liposarcoma; the latter patient died of colon cancer before other investigations to confirm the diagnosis could be performed.
Follow-up investigations for newly detected extracolonic
abnormalities—Fifty-one patients (53%) with significant previously
unknown findings did not undergo additional examinations either because the
diagnosis was definitive—for example, disseminated metastatic
disease—or because the patient was too infirm to justify treatment of
any of the differential diagnoses. The costs of further investigations were
calculated from the 2006–2007 National Health Service (NHS) local
interprovider tariff, the NHS directory price, or the batch cost to the NHS
for a local private health care provider. Forty-five (46.9%) of the 96
patients underwent 68 further investigations at a total cost of
£13,511.64 (
20,537.69, $26,617.93). The examinations included 13
MRI, nine CT, 17 sonography, and eight endoscopic examinations and seven blood
tests, along with a smaller number of a range of other tests. The average cost
was £300.26 (456.40, $591.90) for each of the 45 patients (range,
£4–889.34 [6.08–351.80, $7.88–1,771.70]). This
equates to an added cost of £33.78 (51.27, $66.59) for each CTC
performed (exchange rates for 2007).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
...as we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns—the ones we don't know we don't know. (U.S. Secretary of Defense Donald H. Rumsfeld [11]).
The symptoms that lead patients to undergo lower gastrointestinal investigations to exclude colorectal malignancy are varied and frequently are nonspecific [12]. These include weight loss, altered bowel habits, rectal bleeding, abdominal pain, and anemia. Hamilton et al. [13] in a population-based case–control study found that the positive predictive values (95% CI) for those with rectal bleeding was 2.4% (CI, 1.9–3.2); however, for weight loss, it was 1.2% (0.91–1.6); abdominal pain, 1.1% (0.86–1.3); diarrhea, 0.94% (0.73–1.1); and constipation, 0.42% (0.34–0.52). Except for rectal bleeding [14], the association between gastrointestinal symptoms and colorectal cancer is less strong in the older-than-70-years age group [15] because of a general increase in lower gastrointestinal symptoms in the "healthy" old. Therefore, in elderly patients in whom serious organic disease is more prevalent and lower gastrointestinal symptoms are more frequent, it is of particular value to be able to diagnose or exclude significant extracolonic abnormalities.
CTC has several advantages over barium enema for the investigation of symptomatic patients. In addition to its obvious advantage in revealing extracolonic abnormalities, a CTC examination is less physically demanding on patients than a barium enema, and CTC has a higher sensitivity and specificity for colonic polyps and cancer than barium enema, particularly on the right side where barium examinations are prone to problems associated with inadequate bowel preparation and poor mucosal coating [16]. In addition, in patients who have difficulty maintaining continence with a rectal tube in situ, the loss of carbon dioxide is less problematic than the loss of barium.
The disadvantages of CTC compared with barium enema are that the radiation dose with CTC is increased—in the region of 6.4 mSv for CTC with tube-current modulation [17] compared with 3.5 mSv for barium enema [18]—and that fine mucosal details are better seen with barium enema. These issues are both generally of less importance in the older symptomatic population.
In implementing the transition from a barium enema service to a CTC service, we were limited by CT access, radiologist and radiographer experience, relatively high barium enema capacity, and concern about radiation dose. We therefore elected to target the group with the least concern about radiation, the greatest limitation in mobility, and the likely highest incidence of extracolonic abnormalities to implement the transition. We chose 70 years as a lower age limit. On the basis of the number of barium enema referrals in preceding years, we expected that the number of examinations would match our CTC capacity. All patients referred for radiologic investigation over the study period with the appropriate symptoms and who were older than 70 years and deemed suitable to undergo bowel cleansing were included in our study cohort. No screening examinations were included.
Minimal preparation CT has been advocated by some as the investigation of choice in the elderly population [19]. Although minimal preparation CT of the colon with oral and IV contrast material allows definition of significant extracolonic abnormalities and visualization of some colonic tumors and the negative predictive value of minimal preparation CT is good (96–100%) [20], the positive predictive value for a diagnosis of colon cancer, at approximately 50%, is disappointing [20]. This implies that 50% of patients may unnecessarily require further investigation with full bowel preparation to exclude malignancy even when a "confident" minimal preparation CT diagnosis has been made. The patient consumes a large volume (1.2–1.5 L) of dilute meglumine diatrizoate (Gastrografin, Schering) before the examination over a 48-hour period [19, 21] that, in itself, may elicit a cathartic effect, but this has never been formally quantified in relation to full bowel preparation.
Nevertheless, minimal preparation CT has an important role to play in the investigation of patients who are unfit to undergo bowel cleansing. In articles advocating minimal preparation CT, elderly patients are usually included on the grounds that they are not fit for colonoscopy or barium enema. What is probably more pertinent is whether the patient, irrespective of age, can tolerate a full bowel preparation. Furthermore, reported sensitivities and specificities for large colon polyp and colorectal cancer detection at CTC are comparable to colonoscopy and better than barium enema or minimal preparation CT [1, 2, 5, 16, 19, 21–25].
Our finding of primary colon malignancy in 7.3% (29/400) of patients is
similar to that of another large series of older patients
[19]. The incidence of large
polyps ( 10 mm), colorectal cancer, or both in our study (10.5%) is
similar to a recent large colonoscopic study of patients older than 70 years
(13.4%) [26], although those
investigators identified more polyps than cancers (10.2% vs 3.2%,
respectively). We also identified a small number of patients with
complications of diverticular disease (3.8%).
CTC and minimal preparation CT have clear advantages over colonoscopy and barium enema for the diagnosis and exclusion of extracolonic intraabdominal abnormalities. Studies published to date about the use of CT for the investigation of likely colonic abnormalities are broadly split into either CTC with full bowel cleansing, generally for younger (< 70 years old) and often asymptomatic patients for screening low- to high-risk groups, or minimal preparation CT for the frail and elderly symptomatic patient. In our study, 29.0% (116/400) of all patients had significant extracolonic abnormalities and 24.0% (96/400) of these had previously unknown significant lesions. This is more than twice as high as a pooled analysis of previous studies by Xiong and colleagues [27] (10.5%). This discrepancy may be explained by the inclusion of asymptomatic populations in their meta-analysis, that most studies included did not use IV contrast material, and that the average patient age was substantially lower than that of our group.
If the 29 patients with primary colorectal cancer, whose lesions would arguably have been detected as part of a standard workup with CT of the thorax and abdomen after colonoscopy or barium enema, are excluded, 20.5% (76/371) of the remaining patients had significant previously unknown abnormalities and 5.4% (20/371) had probable or definite malignancy.
|
We routinely administer IV contrast material to allow maximal characterization of extracolonic lesions, which again minimizes additional investigation of extracolonic abnormalities; contrast-enhanced CT is reported to be the most requested follow-up investigation after CTC performed without IV contrast material [27].
There are several drawbacks to our study. Data were collected retrospectively, and we have not been able to determine accurately the precise clinical impact of the colonic and extracolonic abnormalities reported in all cases by case note review. All patients did not undergo full characterization of their colonic and extracolonic findings. Although this lack of follow-up information is a frustration, it reflects real clinical practice in that a clinician supervising a patient's care must judge the appropriateness of further investigation in the context of the patient's current health and other factors. We are unable to measure the specificity of CTC for colorectal cancer or polyps because the period of follow-up is insufficient to clinically detect all missed colorectal cancer and because colonoscopy was not performed in all patients. However, we recorded a high sensitivity for both polyps and colorectal cancer.
A consensus is developing about the format for reporting both colonic [6] and extracolonic abnormalities at CTC [28]. Whether one should report nonsignificant extracolonic abnormalities, such as renal cysts or gallbladder calculi, is a matter of debate, but lengthening a report by including all of these details could distract the referring doctor from a significant abnormality when it is present. It is also important that doctors requesting CTC examinations understand the value of CTC for excluding extracolonic disease. This knowledge will reduce the number of requests for additional unnecessary investigations that may have been requested after traditional colon-only investigations (for example, abdominal sonography). The Virtual Colonoscopy Working Group [28] has proposed the CT Colonography Reporting System (C-RADS), analogous to that used for breast disease, in which colonic and extracolonic findings are given a clinical significance weighting by the reporting radiologist, thereby guiding the referring clinician to the next appropriate management step. This guidance can be particularly helpful when dealing with nonspecialist colleagues.
In the United States and United Kingdom, large numbers of barium enema examinations are still performed (E-Z-EM, written communication, April 2007). In the United Kingdom, barium enema continues to be used because of limited colonoscopy capacity and because adequate radiology provisions, resulting from radiographers performing the barium enema examinations, are available for barium enema. However, the number of barium enema examinations is slowly falling by 5–8% per year in the United Kingdom, which is, in part, explained by increasing CTC capacity, albeit from a low baseline.
The thrust of recent research publications has been related to computer-aided detection of colon polyps and to CTC in screening populations, which may have inadvertently discouraged some gastrointestinal radiologists from performing this technique for investigating symptomatic patients. Although we recognize the primary method of colonic investigation will vary among health care organizations, the main purpose of our article is to raise awareness about the importance of detecting extracolonic abnormalities in older patients presenting with lower gastrointestinal symptoms and to report the power of CTC to depict significant colonic and extracolonic abnormalities using a single examination in the older-than-70-years age group.
We are eagerly awaiting the results of the Special Interest Group in Gastrointestinal and Abdominal Radiology 1 Trial (SIGGAR1 Trial). Based in the United Kingdom, the trial is powered to detect a difference in the sensitivity of CTC versus barium enema versus colonoscopy in 4,500 symptomatic patients older than 55 years; the results are due in 2008. We hope that the results of the SIGGAR1 Trial will clearly define the role of CTC in the investigation of symptomatic patients and provide further impetus for increased implementation of CTC among the wider radiology community.
In conclusion, in a growing elderly population, who have a higher incidence of significant extracolonic abnormalities, those who will tolerate bowel preparation are probably best served by undergoing an examination with the highest positive and negative predictive values both for significant colonic and extracolonic abnormalities. By performing CTC in symptomatic patients older than 70 years, previously unknown significant extracolonic abnormalities were detected in 24.0% of patients, 7.5% of whom had colorectal cancer and 5.8% of whom had primary extracolonic malignancies; the overall yield for malignancy of all types was 12.3%. We recommend the use of IV contrast material for lesion detection and characterization in all symptomatic patients without a contraindication to contrast material.
If access to CT is a limiting factor, we suggest that the combination of good patient tolerance, high sensitivity and specificity for colorectal cancer and large polyps, and additional yield of significant new extracolonic abnormalities makes a compelling case for choosing to target elderly symptomatic patients for implementing a CTC service.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. J. Pickhardt, M. E. Hanson, D. J. Vanness, J. Y. Lo, D. H. Kim, A. J. Taylor, T. C. Winter, and J. L. Hinshaw Unsuspected Extracolonic Findings at Screening CT Colonography: Clinical and Economic Impact Radiology, October 1, 2008; 249(1): 151 - 159. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A SLATER Alteration in bowel habit Imaging, December 1, 2006; 18(4): 208 - 217. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
