CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2012;10:886–892
Diagnostic Accuracy of Computed Tomography Using Lower Doses ofRadiation for Patients With Crohn’s Disease
ORLA CRAIG, SIOBHAN O’NEILL, FIONA O’NEILL, PATRICK McLAUGHLIN, ANNEMARIE McGARRIGLE,SEBASTIAN McWILLIAMS, OWEN O’CONNOR, ALAN DESMOND, ELIZABETH KENNY WALSH, MAX RYAN,MICHAEL MAHER, and FERGUS SHANAHAN
Departments of Medicine and Radiology, Cork University Hospital and Alimentary Pharmabiotic Centre, University College Cork, National University of Ireland, Cork, Ireland
This article has an accompanying continuing medical education activity on page e67. Learning Objectives—At the end of thisactivity, the successful learner will be able to select the most appropriate radiological imaging technique for assessment of
acute Crohn’s disease and will be able to rank different imaging techniques according to the level of radiation they impart.
apktasbtwta
BACKGROUND & AIMS: Magnetic resonance and ultra-onography have increasing roles in the initial diagnosis ofrohn’s disease, but computed tomography (CT) with posi-
ive oral contrast agents is most frequently used to identifyhose with acute extramural complications. However, CTnvolves exposure of patients to radiation. We prospectivelyompared the diagnostic accuracy of low-dose CT (at a doseomparable to that used to obtain an abdominal radiograph)ith conventional-dose CT in patients with active Crohn’sisease. METHODS: Low and conventional dose CT of thebdomen and pelvis were acquired from 50 patients with Crohn’sisease, referred from an inflammatory bowel disease service (20ale; median age, 34 years). Acute complications of Crohn’s dis-
ase were suspected. Iterative reconstruction was performed on allT datasets to facilitate dose reduction. Three radiologists re-iewed the low-dose CT images before the conventional-dose CTmages. RESULTS: The median effective dose (interquartileange) of radiation for the low-dose CT was reduced by 72% fromhat of conventional CT: from 3.5 mSv (3–5.08 mSv) to 0.98 mSv0.77–1.42 mSv) (P � .001). As expected, the quality indexes of theow-dose images were inferior to those of the conventional-dosemages, but no clinically significant diagnostic findings were
issed with low-dose imaging. Follow-up CT examinations wereecommended for 5 patients; 1 had a cervical tumor, 1 had aancreatic lesion, and 3 had intra-abdominal abscess. In each case,he image obtained by low-dose CT was considered sufficient foriagnosis. CONCLUSIONS: Although low-dose CT images
are of lower quality than images obtained with conven-tional doses of radiation, no clinically significant diagnos-tic findings were missed from low-dose CT images of pa-tients with Crohn’s disease. The low-dose CT was obtainedat a median effective dose equivalent to 1.4 abdominalradiographs.
Watch this article’s video abstract and others at http://tiny.cc/bz9jv.
Scan the quick response (QR) code to the left withyour mobile device to watch this article’s video ab-stract and others. Don’t have a QR code reader? Getone at mobiletag.com/en/download.php.
In contrast to magnetic resonance and computed tomogra-phy (CT) enterography (with negative oral contrast), which
re useful for the initial diagnosis of Crohn’s disease, CT withositive oral contrast remains the investigation of choice innown Crohn’s patients with suspected extramural complica-ions such as abscess or perforation. Positive oral contrastbdominal CT has been recommended in this setting by theecond European evidence-based consensus on Crohn’s diseaseecause of better detection of acute extramural complicationshan with CT or magnetic resonance (MR) enterography.1 Theidespread availability of CT and its relatively rapid acquisition
ime confer added advantage in this setting but ionizing radi-tion is an important disadvantage.2,3 We and others showed
that some patients with Crohn’s disease are exposed to a cu-mulative dose of radiation that may be associated with anincreased risk of malignancy, primarily through repeated CTscanning.4 –9
Imaging modalities with no radiation are evolving inCrohn’s disease.10 –12 However, inadequate availability at manycenters, relatively longer acquisition times, greater patient dis-comfort and cost of magnetic resonance imaging (MRI) as wellas the operator-dependent nature of ultrasound means that CTis likely to retain its role for adults with Crohn’s disease,particularly, in the acute setting. Therefore, strategies for reduc-ing diagnostic radiation dose are important.
Increased image noise and reduced image quality representcompromises that accompany methods for reducing the effec-tive dose of a CT scan using the current method of imagereconstruction on modern CT systems, namely, filtered backprojection (FBP). Iterative reconstruction (IR) reduces radiationfrom CT scanning and enables reconstruction of CT images toselectively identify and subtract noise from the image, therebyallowing lower radiation doses and minimal compromise ofimage quality.13 Significant reduction in radiation dose with
Abbreviations used in this paper: ASIR, Adaptive Statistical IterativeReconstruction; BMI, body mass index; CT, computed tomography;IQR, interquartile range; IR, iterative reconstruction; MR, magneticresonance; SD, standard deviation.
August 2012 DIAGNOSTIC ACCURACY OF LOW-DOSE CT RADIATION 887
preservation of interpretability has been achieved with coronaryCT angiography.14 IR also permits reductions in radiation dose
ssociated with abdominal CT while preserving diagnostic in-ormation,15,16 and recent studies suggest that a low-radiationose CT protocol with IR may be applied to the Crohn’sopulation.17,18
We prospectively evaluated a novel low-dose CT protocolreducing radiation to levels close to that of a plain abdominalradiograph and compared the images with conventional-doseCT, performed contemporaneously on the same scanner, andreconstructed with an IR algorithm in patients with activeCrohn’s disease: (1) to assess the potential reduction in radia-tion dose and image quality; and (2) to compare diagnosticaccuracy in addressing the posed clinical question.
MethodsPatient PopulationThe study was approved by the university ethics com-
mittee and registered with http://ClinicalTrials.gov (IdentifierNCT01244386). Fifty patients with Crohn’s disease referred fora clinically indicated abdominopelvic CT scan in the acutesetting were consecutively recruited from an inflammatorybowel disease speciality clinic at Cork University Hospital be-tween April 2010 and March 2011. Each gave written consentfor an additional low-dose CT immediately prior to the clini-cally indicated conventional CT. The protocol for both CTscans was designed such that the combined radiation exposurefrom both scans would not exceed or minimally exceed that ofa single conventional abdominopelvic CT. Disease severity wasassessed using the Harvey–Bradshaw index.19 Body mass index
able 1. Patient Demographics and Clinical Characteristics
Sex (n)Male 19Female 31
Age (y), median (IQR) 34 (26–46)Disease duration (y), mean (SD) 12.5 (9.6)Tobacco use (n)
Smokers 8Former smokers 5Nonsmokers 33
BMI, mean (SD) 24.6 (4.8)Disease severity (Harvey–Bradshaw Index)
MedicationPurine analogues 11Biological agents (anti-TNFs) 11Purine analogues and biologics 2Steroids
Prednisolone 2Budesonide 3
No active treatment 21C-reactive protein (mg/L), median (IQR) 11.5 (2.5–27.4)Clinical question asked (n)
Disease complication 37Disease extent and distribution 36Other 3
(BMI) was determined. Laboratory tests included C-reactive
protein (CRP) as standard of care. The patient characteristicsare summarized in Table 1. For each patient, the indication forthe imaging was recorded as a specific clinical question fromthe referring clinician to be addressed by the CT examination.Categories of question/indication included: (1) confirmation orexclusion of a complication of Crohn’s disease such as perfo-ration, abscess, new fistula or stricture; (2) disease activity anddistribution; and (3) suspected non-Crohn’s related pathology.There were no exclusions and no patients refused.
Study Population SizeBecause this was a comparative assessment of 2 diag-
nostic strategies performed on the same day with the same CTscanner, whereby each patient acted as his/her own control,rather than an interventional study with a primary end point,the usual parameters for calculation of sample size did notapply. The clinical intent at the outset was to determine iflow-dose scanning could be used without missing any clinicallysignificant lesions in the acute setting, and for this, it wasarbitrarily decided that a consecutive series of 50 subjects wouldbe required.
CT TechniqueSingle-phase images through the abdomen and pelvis
were acquired during the portal-venous phase using a 64-slice,multidetector CT scanner (LightSpeed XTE; GE Healthcare,Milwaukee, WI). All patients underwent 90 minutes of oralpreparation (Gastrografin, Bracco Diagnostics, Inc, Princeton,NJ, in H20 solution to give 2% dilution) before CT scan. Patients
ere scanned cranio-caudally in the supine, feet-first positionith arms raised over their head. Mediolateral and anteropos-
erior scout views were obtained and used to plan subsequentanges. Each patient then received 100 mL nonionic intrave-ous contrast media (iohexol, Omnipaque 300, GE Healthcare,ississauga, ON, Canada) at a rate of 2.5 mL per second
njected via a power injector (Stellant; Medrad, Warrendale, PA).utomatic bolus-tracking software (SmartPrep; GE Healthcare)as used to monitor and identify peak arterial vascular en-ancement, and the scan was manually triggered when a thresh-ld of 100 Hounsfield Units was reached within the region of
Table 2. Radiological Crohn’s Disease Activity
Small bowel disease Normal small bowel 0Wall thickening 1Stricture(s) without obstruction 2Stricture(s) with obstruction 3
Large bowel disease Normal large bowel 0Wall thickening 1Stricture(s) without obstruction 2Stricture(s) with obstruction 3
888 CRAIG ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 8
interest in the abdominal aorta. Portal-venous phase scanningof the abdomen and pelvis on arrested inspiration commencedautomatically 45 seconds after the manual trigger. A secondrange was acquired using the conventional-dose protocol 6.2seconds after completion of the low-dose acquisition.
The scanning parameters for the low-dose protocol weretube voltage, 100 kV, gantry rotation time 0.5 seconds andtolerated noise index 70%. For the conventional-dose protocolthe scanning parameters were: tube voltage, 120 kV; gantryrotation time, 0.8 seconds; and tolerated noise index, 38%. Forboth protocols, automatic tube current modulation resulted ina variable tube current dependent upon the composition of thearea being imaged with a minimum threshold of 20 mA and amaximum threshold of 350 mA.
CT Image ReconstructionImages acquired with both protocols were recon-
structed with a 40% Adaptive Statistical Iterative Reconstruc-tion (ASIR) algorithm (GE Healthcare). Reconstructions wereretrospectively performed from the CT raw data using inbuiltiterative reconstruction software on the CT scanner (ASIR; GEHealthcare).
Figure 1. (A) Diagnostic acceptability of conventional- and low-dosergans, large bowel, small bowel, pericolonic fat, perienteric fat, and oveorta hepatis, renal hilum, iliac crest, and acetabulum. Streak artifact
mages.
Estimation of Radiation DoseFor each protocol, the volume CT dose index (CTDIvol)
and dose length product (DLP) values calculated by the scannerwere recorded. These values and the original scanning param-eters for each protocol, corrected for the anatomical area, wereused to estimate the effective dose with The Imaging Perfor-mance and Assessment in CT (ImPACT) patient dosimetrycalculator (version 0.99x, London, England). This approach isvalid for an adult of 70 kg and is used by most practitioners forthe estimation of effective dose.20 While this method may un-derestimate the effective dose for those weighing less than 70 kgand overestimate it for those weighing greater than 70 kg,21 aswe were primarily interested in the percentage dose reduction,we postulated that when dose was over or underestimated, bothprotocols for that individual would be affected in equal pro-portion and the percentage dose reduction would therefore beaccurate.
Image AnalysisThree radiologists reviewed the CT images on a picture-
archiving and communication system (Impax, 6.3.1, AGFAHealthcare, Morstel, Belgium) in a Digital Imaging and Com-
es subjectively determined at anatomical structures such as the solid) Distribution and severity of streak artifact at the level of the diaphragm,bscure important anatomic and pathologic findings on low-dose CT
imagrall. (Bcan o
essa
(ipr
p5
August 2012 DIAGNOSTIC ACCURACY OF LOW-DOSE CT RADIATION 889
munications in Medicine (DICOM) format and on a modalityworkstation (Advantage Workstation version 4.4, GE Health-care). Images obtained with the low-dose protocol were re-viewed first, and the conventional-dose images were reviewed 2weeks later. Reviewers were blinded to the identity of the pa-tient and all clinical details other than the diagnosis of Crohn’sdisease. A consensus was reached between the reviewers onclinically relevant abnormalities, radiologic Crohn’s disease ac-tivity, and image quality for each of the conventional- andlow-dose scans. Crohn’s disease activity was graded and scoredradiologically according to the presence and severity of mor-phologic changes in both the large and small bowel, changessuggestive of active inflammation and penetrating disease andthe presence or absence of acute complications such as acuteobstruction, ileus, or visceral perforation (Table 2). This scoringsystem for radiological Crohn’s disease activity has recentlybeen validated by our group.22
Image quality was assessed in terms of diagnostic acceptabil-ity and the presence of streak artifact. Diagnostic acceptabilitywas scored on a 5-point visual scale, (1 � unacceptable, 2 �barely acceptable, 3 � adequate, 4 � more than adequate, 5 �xcellent) in 5 anatomical structures: solid organs, large bowel,mall bowel, pericolonic fat, and perienteric fat. The mediancore from all 5 levels was taken as an overall score of diagnosticcceptability. Streak artifact was scored on a 3-point visual scale
Table 3. Agreement Between Low- and Conventional-DoseCT for Complications of Crohn’s Disease
NOTE. Number of complications of Crohn’s disease identified onconventional-dose CT on horizontal axis and low-dose CT on verticalaxis. The number of these Crohn’s associated complications identi-fied on both sets of corresponding scans is shown on the diagonal.
Table 4. Agreement Between Low- and Conventional-Dose CT
NOTE. Number of incidental findings identified on conventional-dose C
incidental findings identified on both sets of corresponding scans is show
0 � absent, 1 � present and not interfering with imagenterpretation, 2 � present and interfering with image inter-retation) at the level of the diaphragm, porta hepatis, rightenal hilum, iliac crest, and acetabulum.
CT noise was measured as the standard deviation (SD) of theixel values from a spherical region of interest (ROI); volume19 mm3 drawn in 5 individual anatomical regions: liver pa-
renchyma at the level of the diaphragm, liver parenchyma at thelevel of the porta hepatis, erector spinae at the level of the rightrenal hilum, psoas at the level of the iliac crest, and gluteusmaximus at the level of the acetabulum.
Statistical AnalysisData compilation and statistical analyses were per-
formed using Microsoft Excel (Microsoft Corporation, Red-mond, WA) and GraphPad Prism version 5.0 (GraphPadSoftware Incorporated, San Diego, CA). The effective dose,noise, and radiological Crohn’s disease activity were testedfor normality using the Kolmogorov–Smirnov test. Wheredata were normally distributed, mean values were comparedusing paired t test and where nonnormally distributed, me-dian values were compared using Wilcoxon matched-pairssigned rank test. Clinical and radiological Crohn’s diseaseactivity scores were correlated using the Spearman’s rank-order correlation coefficient.
ResultsEffective Diagnostic Radiation Dose andImage QualityThe median (interquartile range [IQR]) volume CT dose
index was significantly lower for low-dose vs conventional-doseCT (1.55 [1.21–2.26] mGy vs 4.85 [3.96 – 6.97] mGy). Similarlymedian (IQR) dose length product was significantly lower forlow-dose compared with conventional-dose CT (68.13 [52.08 –94.79] mGy vs 229 [181.1–323.6] mGy). The effective radiationdose for low-dose CT ranged from 0.46 to 4.7 mSv with amedian (IQR) of 0.98 (0.76 –1.4) mSv. The effective radiationdose for conventional-dose CT ranged from 2 to 17 mSv with amedian (IQR) of 3.5 (3–5.08) mSv. The difference in effectivedose between low-dose CT scan and conventional-dose CT scancorresponded to a radiation dose reduction of 72% (P � .001).
Incidental Findings
Conventional dose
asis)
Hepatic steatosis(n � 2)
Diverticulardisease(n � 2)
Pancreaticlesion
(n � 1)Cervical tumor
(n � 1)
22
11
horizontal axis and low-dose CT on vertical axis. The number of these
for
rolithi� 1
1
T on
n on the diagonal.
890 CRAIG ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 8
With regard to diagnostic acceptability, the most frequentoverall rating for conventional-dose CT was 5 (excellent) in43/50 scans. The most frequent overall rating for low-dose CTwas 4 (more than adequate) in 24/50 scans. For conventional-dose CT, no scan was rated below 4. For low-dose, 16/50 and7/50 scans were rated as 3 (adequate) and 2 (barely acceptable),respectively. None were rated as 1 (unacceptable) (Figure 1A).
Streak artifact was worse on the low-dose compared with theconventional-dose images particularly at the level of the acetab-ulum where 6 of the low-dose and 1 of the conventional-doseexaminations were rated as 2, ie, artifact present and interferingwith image interpretation. Image noise (defined as SD ofHounsfield value) was significantly higher in the low-dose vsconventional-dose protocol at all levels. Further data on thestreak artifact and image noise scores can be obtained in thesupplemental material section.
Radiologic Crohn’s Disease ActivityThe mean (SD) radiological Crohn’s disease activity
score was lower in the low-dose protocol compared to the
conventional-dose protocol (3.5 [2.7] vs 4.1 [2.5]; P � .0086).
Clinical Crohn’s disease activity (Harvey–Bradshaw index) cor-related poorly with the radiological Crohn’s disease activityscore regardless of protocol used, although this was not statis-tically significant (low dose, � � 0.13 [P � .4]; conventionaldose, � � 0.09 [P � .6]).
Diagnostic AccuracyWith regard to Crohn’s-associated complications, an
intra-abdominal abscess was identified in 3 patients on theconventional-dose scan; each was identified on the correspond-ing low-dose scan. A fistula was identified in 4 patients on theconventional-dose scan. Each was also identified on the lowdose scan. A stricture was identified in 16 patients on theconventional-dose scan with associated partial obstruction in 4patients. Thirteen of these strictures were identified on thelow-dose scan, including all 4 of those with partial obstruction.A nonobstructing stricture was identified in another 3 patientson the low-dose scan that was not seen on the conventional-dose scan (Table 3). Using a large sample normal approxima-tion method (nQuery Advisor version 7; Statistical Solutions,
Figure 2. (A) Axial CT imagesdemonstrating an enterocolicfistula in a 31-year-old man witha BMI of 26.9 shown on low-dose scan with an effective radi-ation dose of 1.6 mSv (i) andconventional-dose scan with aneffective dose of 6.5 mSv (ii) rep-resenting a 75% dose reduction.(B) Sagittal CT images demon-strating a pericolonic abscess ina 46-year-old man with a BMI of26.8 shown on low-dose scanwith an effective dose of 1.5 mSv(i) and conventional-dose scanwith an effective dose of 5 mSv(ii) representing a 70% dose re-duction.
Saugus, MA), with an observed absolute difference in the de-
cfi3
ogyt
August 2012 DIAGNOSTIC ACCURACY OF LOW-DOSE CT RADIATION 891
tection of any Crohn’s-related complication (with the use of lowdose as compared with conventional CT) of �6% and with asample size of 50, the 2-sided 95% confidence interval for thisdifference would extend from 0.02% to �12.6%. Of note, thesensitivity of low dose CT within this sample for a majorcomplication that would have altered clinical management was100%.
Regarding incidental abnormalities, there was 1 case of cho-lelithiasis, 1 porcelain gallbladder, 1 case of nephrolithiasis, 2hepatic steatosis, 2 cases of diverticular disease, 1 indeterminatepancreatic lesion, and 1 tumor of the cervix detected on theconventional-dose scan. Each of these was also identified on thelow-dose scan (Table 4). Follow-up with an additional CT wasrecommended in 5 patients (cervical tumor [n � 1], pancreaticlesion [n � 1], and intra-abdominal abscess [n � 3]). A low-doseCT was subjectively deemed sufficient for follow– up therebyreducing the cumulative exposure. In all cases, the referringgastroenterologist considered that the consensus report ade-quately answered the clinical question (indication) posed, re-gardless of the imaging protocol used. Representative imagesare shown in Figures 2 and 3.
DiscussionThe value of CT in this cohort is shown by finding 3
abscesses, an incidental tumor of the cervix and a pancreaticlesion. While the number of such lesions was modest, theiraccurate detection by low-dose CT changed management; fur-thermore, the accurate exclusion of such lesions in the remain-ing patients facilitated safe consideration of immunosuppres-sive therapy. Although subjective and objective measures ofimage quality were reduced with the low-dose protocol, imageswere diagnostically acceptable. Previous studies have demon-strated dose reductions of 33% and 50% in unselected patientsundergoing abdominopelvic CT and CT colonography, respec-tively.16,23 Our median dose reduction was 72% and now ap-proaches that of the plain abdominal radiograph.
Software to simulate noise in retrospectively collected ab-dominopelvic CT scans of patients with Crohn’s disease hassuggested that noisier images do not compromise the diagno-sis.24 In addition, when low-dose abdominal CT with ASIR wasompared with previously performed standard dose CT withltered back projection in Crohn’s disease, a dose reduction of4% was achieved without loss of image quality.17 Ours was a
Figure 3. Tumor of the cervix ina 73-year-old woman with a BMIof 31.4 shown on low-dose scanwith an effective dose of 4.4 mSv(A) and conventional-dose scanwith an effective dose of 17 mSv(B) representing a 74% dose re-duction.
prospective trial of low-dose CT vs conventional CT, both t
reconstructed with ASIR performed contemporaneously, on thesame scanner in patients with active Crohn’s disease.
It is noteworthy that a clinical trial necessarily introducesartificiality. To overcome the impossibility of complete blindingbecause of the obvious difference in appearance of low vsconventional images and to avoid recall bias, the low-dose scanswere analyzed first and the conventional-dose scans 2 weekslater. Furthermore, the scans were interpreted in isolation with-out bias from previous reports or imaging studies. This createda disadvantage for the radiologists reading both types of imageswho would normally have access to all previous radiologicaldata and participation in a multidisciplinary team meeting.This may have influenced the high rate of reporting of nonob-structing strictures.
Another limitation in the application of our study results isthat we used standard multidetector CT with positive oralcontrast rather than CT enterography with neutral or negativeoral contrast. CT with positive oral contrast is suboptimal inthe initial diagnosis of Crohn’s disease where CT and MRenterography are superior. However, consensus guidelines onthe diagnosis of Crohn’s disease indicate the continuing role ofmultidetector CT with positive oral contrast for the detectionof acute extramural complications in known Crohn’s disease.1
Most authors agree that extra luminal complications are bettervisualized with positive oral contrast.25 Furthermore, routineabdominal CT with positive oral contrast in this setting hascompared well with MR enterography in the assessment ofinflammatory changes in the bowel wall.25 With CT enterogra-phy, the ingestion of large amounts of neutral or negative oralcontrast achieves greater luminal distension and improved mu-cosal enhancement, allowing better detection of mucosal ab-normalities than standard CT.26 However, the ingestion of alarge volume of contrast and the fact that some neutral contrastagents cause diarrhea means that it is less well tolerated in theacute setting. Therefore, our results may not apply beyond thepractice of CT with positive oral contrast for the detection ofacute extramural complications in known Crohn’s disease, al-though Lee et al have reported radiation reduction with IR andCT enterography in patients with Crohn’s disease.18 In addition,
ur results should not be loosely extrapolated to other areas ofastroenterology or to those in whom Crohn’s disease has notet been diagnosed. Furthermore, current iterative reconstruc-ion algorithms are suboptimal for adequate image quality of
he solid organs such as the liver—at the very low radiation
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
892 CRAIG ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 8
doses at which the low-dose CT scans in the current study wereacquired. This is particularly evident in our study (Figure 1A).
We propose that where CT is clinically indicated in theassessment of known active Crohn’s disease, low-dose CT withIR can be used to substantially reduce radiation dose whileaccurately addressing the clinical query.
Supplementary MaterialNote: To access the supplementary material accompa-
nying this article, visit the online version of Clinical Gastroenter-ology and Hepatology at www.cghjournal.org, and at http://dx.doi.org/10.1016/j.cgh.2012.03.014.
References
1. Van Assche G, Dignass A, Panes J, et al. The second Europeanevidence-based consensus on the diagnosis and management ofCrohn’s disease: definitions and diagnosis. J Crohns Colitis2010;4:7–27.
2. Brenner DJ, Hall EJ. Computed tomography—an increasingsource of radiation exposure. N Engl J Med 2007;357:2277–2284.
3. Smith-Bindman R, Lipson J, Marcus R, et al. Radiation doseassociated with common computed tomography examinationsand the associated lifetime attributable risk of cancer. Arch InternMed 2009;169:2078–2086.
4. Desmond AN, O’Regan K, Curran C, et al. Crohn’s disease:factors associated with exposure to high levels of diagnosticradiation. Gut 2008;57:1524–1529.
5. Peloquin JM, Pardi DS, Sandborn WJ, et al. Diagnostic ionizingradiation exposure in a population-based cohort of patients withinflammatory bowel disease. Am J Gastroenterol 2008;103:2015–2022.
6. Newnham E, Hawkes E, Surender A, et al. Quantifying exposure todiagnostic medical radiation in patients with inflammatory boweldisease: are we contributing to malignancy? Aliment PharmacolTher 2007;26:1019–1024.
7. Kroeker KI, Lam S, Birchall I, et al. Patients with IBD are exposedto high levels of ionizing radiation through CT scan diagnosticimaging: a five-year study. J Clin Gastroenterol 2011;45:34–39.
8. Palmer L, Herfarth H, Porter CQ, et al. Diagnostic ionizing radia-tion exposure in a population-based sample of children withinflammatory bowel diseases. Am J Gastroenterol 2009;104:2816–2823.
9. Jaffe TA, Gaca AM, Delaney S, et al. Radiation doses from small-bowel follow-through and abdominopelvic MDCT in Crohn’s dis-ease. AJR Am J Roentgenol 2007;189:1015–1022.
0. Siddiki H, Fidler J. MR imaging of the small bowel in Crohn’sdisease. Eur J Radiol 2009;69:409–417.
1. Sinha R, Murphy P, Hawker P, et al. Role of MRI in Crohn’sdisease. Clin Radiol 2009;64:341–352.
2. Dietrich CF. Significance of abdominal ultrasound in inflammatorybowel disease. Dig Dis 2009;27:482–493.
3. Hara AK, Paden RG, Silva AC, et al. Iterative reconstructiontechnique for reducing body radiation dose at CT: feasibilitystudy. AJR Am J Roentgenol 2009;193:764–771.
4. Leipsic J, Labounty TM, Heilbron B, et al. Estimated radiationdose reduction using adaptive statistical iterative reconstructionin coronary CT angiography: the ERASIR study. AJR Am J Roent-genol 2010;195:655–660.
5. Singh S, Kalra MK, Hsieh J, et al. Abdominal CT: comparison of
adaptive statistical iterative and filtered back projectionreconstruction techniques. Radiology 2010;257:373–383.
6. Sagara Y, Hara AK, Pavlicek W, et al. Abdominal CT: comparisonof low-dose CT with adaptive statistical iterative reconstructionand routine-dose CT with filtered back projection in 53 patients.AJR Am J Roentgenol 2010;195:713–719.
7. Kambadakone AR, Chaudhary NA, Desai GS, et al. Low-doseMDCT and CT enterography of patients with Crohn disease: fea-sibility of adaptive statistical iterative reconstruction. AJR Am JRoentgenol 2011;196:W743–W752.
8. Lee SJ, Park SH, Kim AY, et al. A prospective comparison ofstandard-dose CT enterography and 50% reduced-dose CT en-terography with and without noise reduction for evaluating Crohndisease. AJR Am J Roentgenol 2011;197:50–57.
9. Harvey RF, Bradshaw JM. A simple index of Crohn’s-diseaseactivity. Lancet 1980;1:514.
0. Huda W, Ogden KM, Khorasani MR. Converting dose-length prod-uct to effective dose at CT. Radiology 2008;248:995–1003.
1. DeMarco JJ, Cagnon CH, Cody DD, et al. Estimating radiationdoses from multidetector CT using Monte Carlo simulations:effects of different size voxelized patient models on magnitudesof organ and effective dose. Phys Med Biol 2007;52:2583–2597.
2. Desmond AN, O’Regan K, McWilliams SR, et al. Selection ofsymptomatic patients with Crohn’s disease for abdomino-pelvicCT: role of serum C-reactive protein. Can Assoc Radiol J 2012Mar 14 [Epub ahead of print].
3. Flicek KT, Hara AK, Silva AC, et al. Reducing the radiation dosefor CT colonography using adaptive statistical iterative recon-struction: a pilot study. AJR Am J Roentgenol 2010;195:126–131.
4. Kambadakone AR, Prakash P, Hahn PF, et al. Low-dose CT ex-aminations in Crohn’s disease: impact on image quality, diag-nostic performance, and radiation dose. AJR Am J Roentgenol2010;195:78–88.
5. Schreyer AG, Hoffstetter P, Daneschnejad M, et al. Comparisonof conventional abdominal CT with MR-enterography in patientswith active Crohn’s disease and acute abdominal pain. AcadRadiol 2010;17:352–357.
6. Paulsen SR, Huprich JE, Fletcher JG, et al. CT enterography as adiagnostic tool in evaluating small bowel disorders: review ofclinical experience with over 700 cases. RadioGraphics 2006;26:641–657; Discussion: 57–62.
Reprint requestsAddress requests for reprints to: Fergus Shanahan, MD, Department
of Medicine, Alimentary Pharmabiotic Centre, University College Cork,National University of Ireland, Cork, Ireland. e-mail: [email protected]; fax: 353 (0)21-4345300.
AcknowledgmentsThe help and advice of Jackie Bye, GE Healthcare, UK, is gratefully
acknowledged. The authors are also grateful for the assistance andgoodwill of the patients and clinical staff, without whom this workwould not have been possible.
Conflicts of interestThe authors disclose no conflicts.
Funding
The authors are supported in part by Science Foundation Ireland.
