11
718 AJR:186, March 2006 AJR 2006; 186:718–728 0361–803X/06/1863–718 © American Roentgen Ray Society M E D I C A L I M A G I N G A C E N T U R Y O F Pickhardt and Taylor Extracolonic Findings at CT Colonograp hy Gastrointestinal Imaging Pictorial Essay Extracolonic Findings Identified in Asymptomatic Adults at Screening CT Colonography Perry J. Pickhardt 1,2 Andrew J. Taylor 1 Pickhardt PJ, Taylor AJ Keywords: abdominal imaging, colonography, CT, CT screening DOI:10.2214/AJR.04.1748 Received November 11, 2004; accepted after revision February 7, 2005. The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Navy or the Department of Defense. 1 Department of Radiology, University of Wisconsin Medical School, E3/311 Clinical Science Center, 600 Highland Ave., Madison, WI 53792-3252. Address correspondence to P. J. Pickhardt. 2 Department of Radiology, Uniformed Services University of the Health Sciences, Bethesda, MD 20814. OBJECTIVE. The purpose of this article is to demonstrate the wide variety of extracolonic findings that may be encountered at screening CT colonography (CTC) in asymptomatic adults as well as to discuss the pertinent issues regarding the detection of potential abnormalities in a healthy population. CONCLUSION. Regardless of whether extracolonic evaluation resulting from CTC screening is viewed as a net benefit or liability, it is an unavoidable responsibility that must be handled with care by the interpreting radiologist. Although many potential abnormalities may be questioned, the pretest probability of clinically relevant disease is quite low in average-risk asymptomatic adults, which may influence subsequent management decisions. he primary indication for CT colonography (CTC), also known as virtual colonoscopy, is the de- tection of colorectal polyps and masses. When state-of-the-art technique is applied, CTC represents an effective screen- ing tool that is comparable to optical colonoscopy [1]. Because it is believed that most colorectal cancers can be prevented through effective screening, including asymptomatic adults at average risk, CT colonography is quite distinct from self- referred whole-body CT screening, for which there is currently insufficient scien- tific data to support routine use [2]. The reality for CTC, however, is that the extracolonic abdomen and pelvis are un- avoidably screened in a limited fashion with low-dose, unenhanced CT. Therefore, it is important for radiologists involved in CTC screening to appreciate the unique aspects that surround CT evaluation of healthy adults, where the likelihood of a clinically significant extracolonic finding is very low. A CTC classification system to codify and track extracolonic findings was recently de- veloped and published by the Working Group on Virtual Colonoscopy [3]. Extracolonic evaluation at CTC repre- sents a double-edged sword: the potential benefits include personal reassurance for most adults for whom nothing ominous is found and, in a small minority, discovery of an unsuspected but clinically significant process at an early, presymptomatic stage; the potential limitations include undue anx- iety and added costs stemming from addi- tional workup for findings that eventually prove to be of no consequence. Most studies to date on extracolonic findings at CTC have reported on symptomatic or high-risk indi- viduals [4–7]. In contrast, this pictorial es- say will focus on extracolonic findings gath- ered from more than 3,000 CTC studies of asymptomatic adults. Emphasis will be placed on findings that could potentially af- fect the patient’s health and therefore may require further workup or intervention or cases that can be adequately diagnosed from CTC alone. It is not our intent to pro- vide scientific, evidence-based recommen- dations but rather to broach an important subplot of CTC screening. Technique and Handling of Extracolonic Evaluation at CTC MDCT imaging for CTC at our institution generally entails 1.25-mm collimation, 1- mm reconstruction interval, 120 kVp, and 50–75 mAs. Automatic reconstruction of the supine series to 5-mm contiguous images is performed in all cases to facilitate extraco- lonic evaluation. Advantages of this ap- proach include fewer images to review (< 100), decreased image noise, and easier archiving and future retrieval since the im- age-rich original CTC series are stored as a separate source file. Although these 5-mm T

Extracolonic Findings Identified in Asymptomatic Adults at

  • Upload
    others

  • View
    0

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Extracolonic Findings Identified in Asymptomatic Adults at

718 AJR:186, March 2006

AJR 2006; 186:718–728

0361–803X/06/1863–718

© American Roentgen Ray Society

M E D I C A L I M A G I N G

A C E N T U R Y O F

Pickhardt and TaylorExtracolonic Findings at CT Colonography

G a s t ro i n t e s t i n a l I m ag i n g • P i c t o r i a l E s s ay

Extracolonic Findings Identified in Asymptomatic Adults at Screening CT Colonography

Perry J. Pickhardt1,2

Andrew J. Taylor1

Pickhardt PJ, Taylor AJ

Keywords: abdominal imaging, colonography, CT, CT screening

DOI:10.2214/AJR.04.1748

Received November 11, 2004; accepted after revision February 7, 2005.

The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Navy or the Department of Defense.

1Department of Radiology, University of Wisconsin Medical School, E3/311 Clinical Science Center, 600 Highland Ave., Madison, WI 53792-3252. Address correspondence to P. J. Pickhardt.

2Department of Radiology, Uniformed Services University of the Health Sciences, Bethesda, MD 20814.

OBJECTIVE. The purpose of this article is to demonstrate the wide variety of extracolonicfindings that may be encountered at screening CT colonography (CTC) in asymptomatic adultsas well as to discuss the pertinent issues regarding the detection of potential abnormalities in ahealthy population.

CONCLUSION. Regardless of whether extracolonic evaluation resulting from CTCscreening is viewed as a net benefit or liability, it is an unavoidable responsibility that must behandled with care by the interpreting radiologist. Although many potential abnormalities maybe questioned, the pretest probability of clinically relevant disease is quite low in average-riskasymptomatic adults, which may influence subsequent management decisions.

he primary indication for CTcolonography (CTC), also knownas virtual colonoscopy, is the de-tection of colorectal polyps and

masses. When state-of-the-art technique isapplied, CTC represents an effective screen-ing tool that is comparable to opticalcolonoscopy [1]. Because it is believed thatmost colorectal cancers can be preventedthrough effective screening, includingasymptomatic adults at average risk, CTcolonography is quite distinct from self-referred whole-body CT screening, forwhich there is currently insufficient scien-tific data to support routine use [2].

The reality for CTC, however, is that theextracolonic abdomen and pelvis are un-avoidably screened in a limited fashion withlow-dose, unenhanced CT. Therefore, it isimportant for radiologists involved in CTCscreening to appreciate the unique aspectsthat surround CT evaluation of healthyadults, where the likelihood of a clinicallysignificant extracolonic finding is very low.A CTC classification system to codify andtrack extracolonic findings was recently de-veloped and published by the WorkingGroup on Virtual Colonoscopy [3].

Extracolonic evaluation at CTC repre-sents a double-edged sword: the potentialbenefits include personal reassurance formost adults for whom nothing ominous isfound and, in a small minority, discovery ofan unsuspected but clinically significant

process at an early, presymptomatic stage;the potential limitations include undue anx-iety and added costs stemming from addi-tional workup for findings that eventuallyprove to be of no consequence. Most studiesto date on extracolonic findings at CTC havereported on symptomatic or high-risk indi-viduals [4–7]. In contrast, this pictorial es-say will focus on extracolonic findings gath-ered from more than 3,000 CTC studies ofasymptomatic adults. Emphasis will beplaced on findings that could potentially af-fect the patient’s health and therefore mayrequire further workup or intervention orcases that can be adequately diagnosedfrom CTC alone. It is not our intent to pro-vide scientific, evidence-based recommen-dations but rather to broach an importantsubplot of CTC screening.

Technique and Handling of Extracolonic Evaluation at CTC

MDCT imaging for CTC at our institutiongenerally entails 1.25-mm collimation, 1-mm reconstruction interval, 120 kVp, and50–75 mAs. Automatic reconstruction of thesupine series to 5-mm contiguous images isperformed in all cases to facilitate extraco-lonic evaluation. Advantages of this ap-proach include fewer images to review(< 100), decreased image noise, and easierarchiving and future retrieval since the im-age-rich original CTC series are stored as aseparate source file. Although these 5-mm

T

Page 2: Extracolonic Findings Identified in Asymptomatic Adults at

Extracolonic Findings at CT Colonography

AJR:186, March 2006 719

unenhanced images resemble CT examina-tions obtained for urolithiasis evaluation,there is a fundamental difference: CTCscreening patients are asymptomatic. Theprobability of finding a clinically relevant al-ternative diagnosis is much greater in thesymptomatic “rule out calculus” group, re-portedly in the range of 10–30% [8].

For CTC screening of asymptomaticadults, IV contrast material is generally notindicated, in part because its addition wouldprobably not significantly increase polyp de-tection, particularly when oral contrast tag-ging is used. Furthermore, it is unlikely thatany incremental benefit of IV contrast mate-

rial would offset the added risks, expense, andtime. We specifically mention in our dictatedreports that the lack of IV contrast materialand low-dose technique limit the evaluationof CT findings outside of the colon.

Although we directly communicate the co-lonic findings to patients immediately afterCTC interpretation, we generally do not relayextracolonic findings directly to patients.This enables the referring physician, who hasbuilt a rapport with the patient and is ulti-mately responsible for arranging furtherworkup, to maintain appropriate control. Wedo, however, keep a careful log of potentiallyimportant extracolonic findings, which we

periodically check to confirm resolution. Wedo not accept self-referred patients for CTCscreening but instead require physician refer-ral. This also helps to ensure appropriate fol-low-up of extracolonic findings, thus elimi-nating an area of potential weakness from ourscreening program.

Common Extracolonic CT Findings of Little or No Clinical Significance

A wide variety of minor incidental CTfindings, such as uncomplicated renal or he-patic cysts, arterial vascular calcification,calcified granulomata, hernias (particularlyhiatal and inguinal), fatty liver, benign skel-

A

Fig. 1—Biliary calculi in asymptomatic adults undergoing routine colorectal screening.A, Unenhanced transverse CT image in 58-year-old man shows cholelithiasis with two large gallstones showing rim calcification (arrowheads). Note also gas within one calculus (arrow).B and C, Unenhanced transverse CT images in 68-year-old woman show cholelithiasis (arrowhead, B) and choledocholithiasis (arrow, C). There was mild biliary ductal dilatation and there were additional common duct stones (not shown) but no clinical symptoms or elevated bilirubin.

B C

Page 3: Extracolonic Findings Identified in Asymptomatic Adults at

Pickhardt and Taylor

720 AJR:186, March 2006

etal findings (e.g., enostosis, hemangioma,degenerative changes), and pelvic phlebo-liths, are encountered on virtually a dailybasis. Except for extreme cases, these find-ings almost never require further evaluation.

The reported frequency of these findingshas varied from 1% to 65% [4–7], perhapsreflecting that many radiologists reasonablychoose not to include many of these findingsin their reports. For asymptomatic adults

undergoing routine CTC screening, benign-appearing low-attenuation renal or hepaticlesions do not require further workup (suchas sonography) unless unequivocal com-plexity is present.

Fig. 2—Unenhanced transverse CT image in asymptomatic 50-year-old man undergoing CT colonography screening shows a 5-mm left lower pole renal calculus (arrow).

Fig. 3—Unenhanced transverse CT image in asymptomatic 51-year-old man undergoing CT colonography screening shows multiple subcentimeter noncalcified pulmonary nodules. Patient has no history of malignancy and is currently undergoing CT surveillance to assess stability of these lesions.

Fig. 4—Unenhanced transverse CT image in asymptomatic 58-year-old man undergoing routine colorectal screening shows unsuspected 5.4-cm abdominal aortic aneurysm (arrowhead) with intimal calcification and subtle crescentic mural thrombus. Patient subsequently underwent successful surgical repair of aneurysm.

Page 4: Extracolonic Findings Identified in Asymptomatic Adults at

Extracolonic Findings at CT Colonography

AJR:186, March 2006 721

Extracolonic CT Findings of Variable Clinical Significance

Published studies have tended to reportthe frequency of extracolonic findings interms of “moderate importance” and “high

importance” (with “low importance” gener-ally assumed to represent a clinically insig-nificant finding) [4–7]. This practice greatlyoverstates the frequency of truly significantextracolonic findings because even most

findings reported as highly important ulti-mately prove to be of no consequence (e.g.,a hepatic hemangioma) (Kang PS et al., pre-sented at the 2003 Radiological Society ofNorth America meeting). Therefore, we

A B

Fig. 5—Benign cystic adnexal lesions in asymptomatic women undergoing routine colorectal screening.A, Unenhanced transverse CT image in 59-year-old woman shows large unilocular cyst (C) in right adnexal region and adjacent solid lesion (F), which represents pedunculated broad ligament fibroid. U = uterus.B, Unenhanced transverse CT image in 68-year-old woman shows left adnexal cystic lesion (arrow) that was complex at subsequent pelvic sonography (not shown) and proved to be benign fibroadenoma after surgical resection.

Fig. 6—Unenhanced transverse CT image in asymptomatic 68-year-old man undergoing CT colonography screening shows minimally complicated left renal cyst with thin focal rim calcification (arrow). Lesion was unchanged from CT performed more than 5 years earlier for prostate cancer staging (not shown). Note also cholelithiasis (arrowhead).

Fig. 7—Unenhanced transverse CT image in asymptomatic 57-year-old woman undergoing CT colonography screening shows subtle 5-cm hepatic lesion (arrowheads) not compatible with simple cyst. Lesion was confirmed to represent cavernous hemangioma on dynamic IV contrast-enhanced CT (not shown).

Page 5: Extracolonic Findings Identified in Asymptomatic Adults at

Pickhardt and Taylor

722 AJR:186, March 2006

A B

C D

Fig. 8—Unsuspected extracolonic malignancy in asymptomatic adults undergoing routine colorectal screening.A, Unenhanced transverse CT image in 56-year-old woman shows complex solid and cystic left adnexal mass that proved to be papillary serous adenocarcinoma of ovary. Note mural soft-tissue nodule (arrowhead).B, Unenhanced coronal CT image in 52-year-old man shows solid exophytic mass extending off upper pole of left kidney (arrowheads), which proved to be renal cell carcinoma. This case reinforces utility of multiplanar evaluation because this lesion may be difficult to detect on transverse images alone.C, Unenhanced transverse CT image in 51-year-old woman shows confluent retroperitoneal lymphadenopathy (asterisk), which was subsequently diagnosed as non-Hodgkin’s lymphoma by CT-guided biopsy.D, Unenhanced coronal CT image in 63-year-old man shows spiculated left lower lobe pulmonary nodule (arrow), which was subsequently diagnosed as non–small cell lung carcinoma by CT-guided biopsy. Patient underwent successful surgical excision of this T1 lesion.

Page 6: Extracolonic Findings Identified in Asymptomatic Adults at

Extracolonic Findings at CT Colonography

AJR:186, March 2006 723

A B

C D

Fig. 9—Congenital variants in asymptomatic adults undergoing routine colorectal screening.A, Unenhanced coronal CT image in 42-year-old man with family history of colon cancer shows malrotation (nonrotation) with air-filled colon predominately occupying left abdomen and small bowel predominately on right. Absence of duodenal sweep and reversal of normal superior mesenteric artery–superior mesenteric vein relationship were evident on other images (not shown). C = cecum.B, Unenhanced transverse CT image in 51-year-old man shows multiple small spleens (short arrows), abrupt shortening of pancreas (long arrow), and preduodenal portal vein (arrowhead), all compatible with heterotaxy (polysplenia). ICV interruption was not present but borderline cardiomegaly was suggested on CT scout (not shown); cardiac evaluation has not yet been pursued.C, Unenhanced transverse CT image in 55-year-old man shows unsuspected horseshoe kidney (arrows). Small calculus was present in left upper pole moiety (not shown).D, Unenhanced transverse CT image in 54-year-old man shows inferior vena cava duplication (arrows).

Page 7: Extracolonic Findings Identified in Asymptomatic Adults at

Pickhardt and Taylor

724 AJR:186, March 2006

report such findings at CTC to be of “poten-tial” importance to underscore both the needfor further evaluation and the reasonablelikelihood for a good outcome [1].

Cholelithiasis (Figs. 1A and 1B) and neph-rolithiasis (Fig. 2) are relatively common find-ings of potentially moderate clinical impor-tance, with each seen in approximately 5–10%of patients undergoing CTC [1, 4, 5]. Unsus-pected gallstones are generally seen within an

otherwise normal-appearing gallbladder. Ofnote, we have also encountered asymptomaticcholedocholithiasis on several occasions(Fig. 1C). Unsuspected renal calculi are typi-cally 5 mm or smaller and without associatedhydronephrosis. Indeterminate pulmonary nod-ules detected in asymptomatic adults are likelybenign but may require additional follow-up toconfirm stability (Fig. 3). We generally followthe recently published guidelines from the

Fleischner Society [9]. The likelihood of de-tecting an unsuspected abdominal aortic aneu-rysm is largely related to patient age, and thesignificance is primarily determined by the sizeof the aneurysm (Fig. 4).

Because most women undergoing CTC arepostmenopausal, prominent adnexal lesionsoften necessitate sonographic follow-up. Find-ings range from simple-appearing unilocularcysts that are almost certainly benign and prob-

A B

C D

Fig. 10—Fat- or lipid-containing lesions in asymptomatic adults that allow specific diagnosis on CT colonography.A, Unenhanced transverse CT image in 59-year-old woman shows an exophytic lesion extending off left lower pole kidney (arrow), diagnostic of angiomyolipoma.B, Unenhanced transverse CT image in 74-year-old man shows right adrenal lesion containing macroscopic fat (arrow), diagnostic of myelolipoma.C, Unenhanced transverse CT image in 54-year-old man shows bilateral low-attenuation adrenal lesions (arrowheads). Attenuation measured less than 10 H for both lesions, diagnostic of nonhyperfunctioning adenomas.D, Unenhanced transverse CT image in 54-year-old woman shows ovoid lipoma (arrow) within proximal jejunum. Note also subtle cholelithiasis within distended gallbladder. Subsequently, 5-cm jejunal lipoma was resected via enterotomy during open cholecystectomy.(Fig. 10 continues on next page)

Page 8: Extracolonic Findings Identified in Asymptomatic Adults at

Extracolonic Findings at CT Colonography

AJR:186, March 2006 725

ably nonneoplastic to more complex solid andcystic masses that invariably require surgicalevaluation (Fig. 5). Most uniform solid le-sions, however, are likely to represent pedun-culated fibroids extending into the broad liga-ment (Fig. 5A). In our experience, the majorityof complex cystic renal lesions encountered atCTC are also benign (Fig. 6), although unsus-pected renal cell carcinomas will be identifiedon occasion. Most large indeterminant hepaticlesions identified at asymptomatic screeningCTC are subsequently diagnosed as cavernoushemangiomas of essentially no clinical impor-tance on IV contrast-enhanced studies (Fig. 7).Overall, the frequency of extracolonic findingsof potentially high importance is much loweramong average-risk cohorts (4–6%) [1] com-pared with higher-risk populations (10–23%)[4, 5, 7]. Fortunately, the majority of these willultimately prove to be of little or no clinicalsignificance (Kang PS et al., 2003 RSNAmeeting).

Unsuspected extracolonic malignancy isrelatively uncommon in asymptomatic adultsundergoing CTC, with approximately one caseper 200 patients screened in our cumulative ex-perience. However, it should be noted thatCTC paradoxically uncovers more extraco-lonic malignancies than colon cancers in this

group [1] since the more common target lesionfound in an asymptomatic screening popula-tion is the potentially precancerous advancedadenoma, not colon cancer itself. To date, wehave encountered at least two cases each of un-suspected ovarian cancer, renal cell carcinoma,non-Hodgkin’s lymphoma, and lung cancer(Fig. 8). Detection of malignancy during thepresymptomatic phase was probably of realbenefit in at least some of these patients.

We have encountered a wide array of inci-dental congenital variants, only a few ofwhich may impart some clinical significance.Notably, incidental malrotation in adultsshould not be assumed to automaticallyrepresent an insignificant finding since de-layed complications can rarely occur [10](Fig. 9A). Similarly, patients with polysple-nia may evade detection into adulthood if sig-nificant congenital cardiac defects are notpresent [10] (Fig. 9B). More commonly, mi-nor congenital variants of little or no conse-quence are identified (Figs. 9C and 9D).

An imaging-specific diagnosis is possibleon unenhanced CT for a variety of fat-containing lesions in the abdomen and pel-vis. Solitary renal angiomyolipomas aremost often seen in middle-aged women(Fig. 10A). Adrenal myelolipomas are oc-

casionally identified (Fig. 10B) and, as withangiomyolipomas, are at low risk for spon-taneous hemorrhage unless they are large.Nonhyperfunctioning adrenal adenomastend not to contain macroscopic fat, butrather most contain sufficient amounts ofcytoplasmic lipid, allowing confident diag-nosis on unenhanced CT (Fig. 10C). Lipo-mas arising from the gastrointestinal tract(Fig. 10D) or a variety of other abdomi-nopelvic locations can be seen. Fat-contain-ing gynecologic entities include the rela-tively common benign ovarian teratoma(Fig. 10E) and the rare uterine lipoleiomy-oma (Fig. 10F).

In addition to the intended evaluation ofthe colon and rectum, a variety of incidentalfocal gastrointestinal lesions may be identi-fied in the distal esophagus, stomach, smallbowel, and appendix. Gastrointestinal lipo-mas have already been discussed (Fig. 10D).We have encountered ileal carcinoid tumors,all of which have been relatively smalland without extension beyond the bowelwall (Figs. 11A–11C). Incidental tumors inthe more proximal small bowel are rare(Figs. 11D and 11E). Appendiceal findingshave included appendicoliths and mucoceles(Fig. 11F). Benign lesions seen in the esoph-

E F

Fig. 10 (continued)—Fat- or lipid-containing lesions in asymptomatic adults that allow specific diagnosis on CT colonography.E, Unenhanced transverse CT image in 42-year-old woman with family history of colon cancer shows pelvic mass (arrows) containing fat, soft tissue, and calcifications, diagnostic of ovarian teratoma.F, Unenhanced transverse CT image in 57-year-old woman shows large fat-containing mass (asterisk) centrally within uterus, which was confirmed to represent lipoleiomyoma after surgical resection.

Page 9: Extracolonic Findings Identified in Asymptomatic Adults at

Pickhardt and Taylor

726 AJR:186, March 2006

agus and stomach have included a duplica-tion cyst (Fig. 11G) and a densely calcifiedleiomyoma (Fig. 11H).

As mentioned above, skeletal findings suchas bone islands, degenerative changes, and ver-tebral hemangiomata are of little or no clinical

concern. We have, however, seen several casesof bilateral pars defects (spondylolysis) withvarying degrees of spondylolisthesis (Fig. 12)

A

Fig. 11—Noncolorectal gastrointestinal lesions seen in asymptomatic adults undergoing routine screening.A–C, Unenhanced transverse CT image (A) in 65-year-old woman shows small, subcentimeter soft-tissue lesion involving distal ileum (arrow). Volume-rendered 3D endoluminal image from CT colonography (B) and digital photograph from optical colonoscopy (C) show same lesion, which proved to be carcinoid tumor. We do not routinely perform 3D fly-through of distal ileum at CT colonography.(Fig. 11 continues on next page)

B C

Page 10: Extracolonic Findings Identified in Asymptomatic Adults at

Extracolonic Findings at CT Colonography

AJR:186, March 2006 727

D E

F G

Fig. 11 (continued)—Noncolorectal gastrointestinal lesions seen in asymptomatic adults undergoing routine screening.D and E, Unenhanced transverse CT image with polyp window setting (D) and volume-rendered 3D endoluminal image (E) in 55-year-old woman show incidental polypoid mass (arrowhead, D) in distal jejunum. Lesion proved to be jejunal hamartoma after surgical recision.F, Unenhanced curved reformatted sagittal CT image in 63-year-old man shows grossly dilated appendix (asterisk) with subtle mural calcification (arrowheads), consistent with mucocele. Proximal appendix near base appears normal (arrow). Gas-filled sigmoid colon (S) abuts cecum (C). Mucinous adenoma of appendix was confirmed after surgical recision.G, Unenhanced transverse CT image in 67-year-old woman shows cystic lesion adjacent to distal esophagus (arrow). Comparison with chest CT performed 2 years earlier showed lesion was stable and is believed to most likely represent foregut duplication cyst.(Fig. 11 continues on next page)

Page 11: Extracolonic Findings Identified in Asymptomatic Adults at

Pickhardt and Taylor

728 AJR:186, March 2006

that may be of clinical relevance. Unexplainedmultifocal lytic or blastic lesions have rarelyled to further evaluation. An example is unsus-pected osseous metastases in one patient witha remote history of breast cancer.

ConclusionRegardless of whether extracolonic evalu-

ation resulting from CTC screening is viewedas a net benefit or liability, it is an unavoidableresponsibility that must be handled with careby the interpreting radiologist. Althoughmany abnormalities will inevitably be uncov-ered, the pretest probability of clinically rele-vant disease is quite low among these aver-age-risk asymptomatic adults. In terms ofreceiver operating characteristic (ROC) anal-ysis, one should perhaps “slide down” theROC curve somewhat to decrease the false-positive fraction and avoid overcalling extra-

colonic findings in this cohort, which couldhave a negative impact on both cost-effective-ness and overall patient care.

References1. Pickhardt PJ, Choi JR, Hwang I, et al. CT virtual

colonoscopy to screen for colorectal neoplasia in

asymptomatic adults. N Engl J Med 2003;

349:2189–2198

2. Stanley RJ. Inherent dangers in radiologic screen-

ing. AJR 2001; 177:989–992

3. Zalis ME, Barish MA, Choi JR, et al. (Working

Group on Virtual Colonoscopy). CT colonoscopy

reporting and data system: a consensus proposal.

Radiology 2005; 236:3–9

4. Hara AK, Johnson CD, MacCarty RL, Welch TJ. In-

cidental extracolonic findings at CT colonography.

Radiology 2000; 215:353–357

5. Gluecker TM, Johnson CD, Wilson LA, et al. Ex-

H

Fig. 11 (continued)—Noncolorectal gastrointestinal lesions seen in asymptomatic adults undergoing routine screening.H, Unenhanced transverse CT image in 58-year-old woman shows densely calcified gastric mass (arrow), which proved to be leiomyoma after surgical wedge resection.

tracolonic findings at CT colonography: evaluation

of prevalence and cost in a screening population.

Gastroenterology 2003; 124:911–916

6. Edwards JT, Wood CJ, Mendelson RM, Forbes GM.

Extracolonic findings at virtual colonoscopy: impli-

cations for screening programs. Am J Gastroenterol

2001; 96:3009–3012

7. Hellström M, Svensson MH, Lasson A. Extraco-

lonic and incidental findings at CT colonography

(virtual colonoscopy). AJR 2004; 182:631–638

8. Rucker CM, Menias CO, Bhalla S. Mimics of renal

colic: alternative diagnoses at unenhanced helical

CT. RadioGraphics 2004; 24:S11–S33

9. MacMahon H, Austin JHM, Gamsu G, et al. Guide-

lines for management of small pulmonary nodules

detected on CT scans: a statement from the

Fleischner Society. Radiology 2005; 237:395–400

10. Pickhardt PJ, Bhalla S. Intestinal malrotation in ad-

olescents and adults: spectrum of clinical and im-

aging features. AJR 2002; 179:1429–1435

Fig. 12—Unenhanced sagittal CT image in asymptomatic 56-year-old woman undergoing CT colonography screening shows unsuspected spondylolisthesis and degenerative changes at the L5–S1 level (arrow), resulting from bilateral L5 pars defects (spondylolysis). These defects can be seen better on other images.