Imaging Patients with Acute Abdominal Pain › ... › abdomen-agudo-1.pdf · Imaging Patients with Acute Abdominal Pain1 JaapStoker,MD AdriennevanRanden,MD WytzeLame´ris,MSc MarjaA.Boermeester,MD

Imaging Patients with AcuteAbdominal Pain1

Jaap Stoker, MDAdrienne van Randen, MDWytze Lameris, MScMarja A. Boermeester, MD

Acute abdominal pain may be caused by a myriad of diag-noses, including acute appendicitis, diverticulitis, and cho-lecystitis. Imaging plays an important role in the treatmentmanagement of patients because clinical evaluation resultscan be inaccurate. Performing computed tomography (CT)is most important because it facilitates an accurate andreproducible diagnosis in urgent conditions. Also, CT find-ings have been demonstrated to have a marked effect onthe management of acute abdominal pain. The cost-effec-tiveness of CT in the setting of acute appendicitis wasstudied, and CT proved to be cost-effective. CT can there-fore be considered the primary technique for the diagnosisof acute abdominal pain, except in patients clinically sus-pected of having acute cholecystitis. In these patients,ultrasonography (US) is the primary imaging technique ofchoice. When costs and ionizing radiation exposure areprimary concerns, a possible strategy is to perform US asthe initial technique in all patients with acute abdominalpain, with CT performed in all cases of nondiagnostic US.The use of conventional radiography has been surpassed;this examination has only a possible role in the setting ofbowel obstruction. However, CT is more accurate andmore informative in this setting as well. In cases of bowelperforation, CT is the most sensitive technique for depict-ing free intraperitoneal air and is valuable for determiningthe cause of the perforation. Imaging is less useful in casesof bowel ischemia, although some CT signs are highlyspecific. Magnetic resonance (MR) imaging is a promisingalternative to CT in the evaluation of acute abdominal painand does not involve the use of ionizing radiation expo-sure. However, data on the use of MR imaging for thisindication are still sparse.

� RSNA, 2009

Supplemental material: http://radiology.rsna.org/content/253/1/31/suppl/DC1

1 From the Departments of Radiology (J.S., A.v.R., W.L.)and Surgery (A.v.R., W.L., M.A.B.), Academic MedicalCenter, University of Amsterdam, Meibergdreef 9, 1105AZ Amsterdam, the Netherlands. Received February 17,2009; revision requested March 26; revision receivedApril 27; accepted May 13; final version accepted May19. Address correspondence to J.S. (e-mail: [email protected] ).

� RSNA, 2009

REVIEWS

ANDCOM

MENTARY

�STATE

OFTHE

ART

Radiology: Volume 253: Number 1—October 2009 ▪ radiology.rsna.org 31

Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.

Acute abdominal pain is a commonchief complaint in patients exam-ined in the emergency department

(ED) and can be related to a myriad ofdiagnoses. Of all patients who present tothe ED, 4%–5% have acute abdominalpain (1). Obtaining a careful medical his-tory and performing a physical examina-tion are the initial diagnostic steps forthese patients. On the basis of the resultsof this clinical evaluation and laboratoryinvestigations, the clinician will considerimaging examinations to help establishthe correct diagnosis.

Acute abdomen is a term frequentlyused to describe the acute abdominal painin a subgroup of patients who are seri-ously ill and have abdominal tendernessand rigidity. Before the advent of wide-spread use of imaging, these individualswere candidates for surgery. However,with the present role of imaging, somepatients with acute abdomen will not un-dergo surgery. Other patients with acuteabdominal pain that does not meet thecriteria to be defined as acute abdomen—for example, many patients suspected ofhaving acute appendicitis—will need sur-gery. In this article, we use the term acuteabdominal pain to refer to the completespectrum of acute abdominal pain in pa-tients who are treated in the ED and re-quire imaging.

A considerable number of articles onthe accuracy of imaging in determiningspecific diagnoses that may cause acuteabdominal pain, such as acute appendici-tis and diverticulitis, have been published.The accuracy of imaging techniques per-formed in carefully selected patients sus-pected of having a specific diagnosis inresearch studies cannot always be gener-alized to routine clinical practice in non-selected patients with acute abdominalpain because the pretest probabilities dif-fer per disease in different settings. Thespectrum of disease in this group of pa-tients is broad and varies according toreferral and demographic patterns (TableE1 [online]). The added value of imagingafter clinical evaluation—particularly itseffect on diagnostic accuracy and cer-tainty and patient treatment—is impor-tant.

In this review, we discuss the role ofimaging in adults who present with acuteabdominal pain to the ED. Our focus isacute abdominal pain in general, but wealso discuss a number of frequently en-countered urgent diagnoses in patientswith acute abdominal pain: appendicitis,diverticulitis, cholecystitis, and bowel ob-struction. Although perforated viscus andmesenteric ischemia are less frequentlyencountered, these are also addressedbecause imaging is of paramount impor-tance for the timely diagnosis of theseabnormalities. Other conditions such asacute pancreatitis are not described. Im-aging is important in the clinical manage-ment but is not pivotal for the diagnosis ofconditions such as acute pancreatitis.

Conventional radiography, ultra-sonography (US), and computed tomog-raphy (CT) are frequently used in the di-agnostic work-up of patients with acuteabdominal pain. Magnetic resonance(MR) imaging and diagnostic laparoscopyare also available, but they are used farless frequently for initial diagnostic work-up. In the literature, the accuracy of im-aging in patients with acute abdominalpain usually is not expressed in terms ofwell-known parameters such as sensitiv-ity, specificity, and predictive values be-cause of the lack of adequate referencestandards in many reports. Furthermore,acute abdominal pain may be due to anumber of discreet diagnoses, so it is im-

possible to generate a straightforward2 � 2 contingency table. Therefore, thediagnostic value of imaging modalities isoften expressed in terms of the change indiagnoses, the change in clinical manage-ment, and/or the extent to which thetreating physician in the ED found thegiven imaging examination helpful or di-agnostic. If a given imaging examination ishelpful (eg, leading to a higher level ofdiagnostic confidence) or diagnostic ac-cording to the treating physician, it is con-sidered to have yielded positive findingsor results at diagnostic work-up (2,3).

Acute Abdominal Pain

The causes of acute abdominal pain rangefrom life-threatening to benign self-limiting disorders. Acute appendicitis, di-verticulitis, cholecystitis, and bowel ob-struction are common causes of acute ab-dominal pain. Other important but lessfrequent conditions that may cause acuteabdominal pain include perforated viscusand bowel ischemia.

A confident and accurate diagnosiscan be made solely on the basis of medicalhistory, physical examination, and labo-ratory test findings in only a small propor-tion of patients. The clinical manifesta-tions of the various causes of acuteabdominal pain usually are notstraightforward. For proper treat-ment, a diagnostic work-up that en-ables the clinician to differentiate be-tween the various causes of acute abdom-inal pain is important, and imaging playsan important role in this process. Manypatients are referred without a clear pre-test diagnosis, and imaging is warrantedto determine the diagnosis and guidetreatment in these patients. According toAmerican College of Radiology (ACR) ap-propriateness criteria (4), contrast mate-

Published online10.1148/radiol.2531090302

Radiology 2009; 253:31–46

Abbreviations:ACR � American College of RadiologyED � emergency departmentSBO � small-bowel obstruction

Authors stated no financial relationship to disclose.

Essentials

� CT is the technique of choice forthe diagnosis of acute abdominalpain, except in cases of acute cho-lecystitis.

� CT findings have a substantial ef-fect on the treatment manage-ment of patients with acute ab-dominal pain.

� Despite its limitations, CT is thepreferred imaging technique forthe diagnosis of bowel ischemia.

� Conventional radiography mayhave a role in the assessmentand/or diagnosis of bowel ob-struction, especially if CT is notavailable.

� US is a well-established alterna-tive—and MR imaging is a poten-tial alternative—to CT in the set-ting of acute abdominal pain.

STATE OF THE ART: Imaging Patients with Acute Abdominal Pain Stoker et al

32 radiology.rsna.org ▪ Radiology: Volume 253: Number 1—October 2009

rial–enhanced CT of the abdomen andpelvis is considered the most appropriateexamination for patients with fever, non-localized abdominal pain, and no recentsurgery. Nonenhanced CT, US, and con-ventional radiography are considered lessappropriate initial imaging examinationsfor these patients.

Conventional Chest and AbdominalRadiographyConventional radiography is commonlythe initial imaging examination per-formed in the diagnostic work-up of pa-tients who present with acute abdominalpain to the ED. This examination is widelyavailable, can be easily performed in ad-mitted patients, and is used to excludemajor illness such as bowel obstructionand perforated viscus. Conventional radi-ography includes supine and upright con-ventional abdominal radiography and up-right chest radiography.

The accuracy values for conventionalradiography in the diagnostic work-up ofpatients with acute abdominal pain arenot convincing. Some study investigatorshave reported an accuracy of 53% (5). Inone study, treatment managementchanges after review of the radiographswere reported for only 4% of patients (6).In the majority of patients, further imag-ing is warranted after conventional radi-ography. US and CT (5,7), as comparedwith conventional radiography, yieldmarkedly higher accuracy values. Theoverall sensitivity of CT is reportedly 96%compared with 30% for conventional ra-diography (5). Despite a lack of evidenceto justify the extensive use of conventionalradiography, it is often the initial diagnos-tic imaging examination performed in pa-tients with acute abdominal pain at manyinstitutions. Exact data on the number ofindividuals who present with acute ab-dominal pain to the ED and undergo con-ventional radiography are not available.Fifty to seventy-five percent of patientssuspected of having acute appendicitis un-dergo conventional radiography (8,9), de-spite evidence in the literature that con-ventional radiography has no diagnosticvalue in these patients (7,9). In selectcases, such as those of patients suspectedof having bowel obstruction, perforatedviscus, urinary tract calculi, or foreign

bodies (7,10,11), conventional radiogra-phy has been reported to have good ac-curacy. Conventional radiography mightbe useful in these patients. In a recentstudy, however, only the sensitivity forthe diagnosis of bowel obstruction wassignificantly higher after conventional ra-diograph evaluation (12). Thus, theuse of conventional radiography mightjustifiably be limited to these patientsonly, especially if CT is not available.The radiation dose delivered at con-ventional radiography is relatively lim-ited (approximately 0.1–1.0 mSv)compared with that delivered at CT(approximately 10 mSv) (13).

US ExaminationUS is another imaging modality com-monly used in the diagnostic work-up ofpatients with acute abdominal pain. WithUS, the abdominal organs and the alimen-tary tract can be visualized. US is widelyavailable and is easily accessible in theED. It is important that US is a real-timedynamic examination that can reveal thepresence or absence of peristalsis and de-pict blood flow. Furthermore, it is possi-ble to correlate US findings with the pointof maximal tenderness. Wide availabilityin the ED, lower costs, and absence ofradiation exposure are advantages of US,as compared with CT. When radiologistsperform US in patients, relevant addi-tional information can be obtained duringthe examination. For example, US find-ings may suggest a previously unexpecteddiagnosis, in which case additional clinicalhistory information becomes important.

The most common US technique usedto examine patients with acute abdominalpain is the graded-compression procedure(14). With this technique, interposing fatand bowel can be displaced or compressedby means of gradual compression to showunderlying structures. Furthermore, if thebowel cannot be compressed, the noncom-pressibility itself is an indication of inflam-mation. Curved (3.5–5.0-MHz) and linear(5.0–12.0-MHz) transducers are usedmostcommonly, with frequencies depending onthe application and the patient’s stature(15,16).

Although values for the accuracy ofUS performed in patients with acute ab-dominal pain are not available, in one

study (17), US reportedly provided usefulinformation for 56% of patients withacute abdominal pain, and in anotherstudy (18), it either yielded unique diag-nostic information or confirmed one ofthe differential diagnoses in 65% of pa-tients. Among 496 patients who pre-sented with acute abdominal pain to anED, the proportion of patients with a cor-rect diagnosis after clinical evaluation in-creased from 70% to 83% after evalua-tion with US (19). In two published stud-ies (17,20), US findings led to analteration in treatment management for22% of patients; however, the patient co-horts in these two investigations were notthe same: Walsh et al (17) excluded pa-tients who were strongly clinically sus-pected of having perforated viscus, bowelobstruction, or appendicitis, whereas Dhil-lon et al (20) included all patientswith acuteabdominal pain for whom US was re-quested by the treating physician in the ED.

CT ExaminationUse of CT in the evaluation of acute ab-dominal pain has increased to a large ex-tent. For example, in the United States,the number of CT examinations per-formed for this indication increased141% between 1996 and 2005 (21). Thisincrease was related to the high accuracyof CT in the diagnosis of specific diseases(eg, appendicitis and diverticulitis[15,16]) and the rapid patient throughputthat can be achieved with use of multide-tector CT scanners.

The CT technique used to examinepatients with acute abdominal pain gen-erally involves scanning of the entire ab-domen after intravenous administrationof an iodinated contrast medium. Al-though abdominal CT can be performedwithout contrast medium (5), the intrave-nous administration of contrast materialfacilitates good accuracy—with a positivepredictive value of 95% reported for thediagnosis of appendicitis (22)—and a highlevel of diagnostic confidence, especiallyin rendering diagnoses in thin patients, inwhom fat interfaces may be almost ab-sent (Fig 1). Although rectal or oral con-trast material may be helpful in differen-tiating fluid-filled bowel loops from ab-scesses in some cases, the use of oralcontrast material can markedly increase



the time these patients spend in the ED(23). The lack of enteral contrast mediumdoes not seem to hamper the accuratereading of CT images obtained in patientswith acute abdominal pain as it does inpostoperative patients. For example, in aseries of 1021 consecutive patients withacute abdominal pain in whom only intra-venous contrast medium was adminis-tered, there were no inconclusive CTscans due to the lack of enteral contrastmedium (24). Multiplanar reformation is

beneficial, especially in cases of equivocalCT scans, and it increases the radiolo-gist’s level of confidence in the diagnosis.

Studies to evaluate the accuracy ofabdominal CT performed in patients withacute abdominal pain generally arescarce. In the cohort study of 1021 con-secutive patients with acute abdominalpain, US and CT were compared for thedetermination of urgent diagnoses (24).CT was significantly more sensitive thanUS (89% vs 70%, P � .001). The highestsensitivity (only 6% missed urgent cases)was obtained with a diagnostic strategyinvolving the use of initial US, followed byCT, only in negative or inconclusive UScases. Use of this approach also led to areduction in radiation exposure becauseCT was needed for only 49% of the pa-tients. Alternative strategies based on thebody mass index or age of the patient oron the location of the pain resulted in aloss of sensitivity. In the literature, thereare two randomized controlled trials inwhich standard practice was comparedwith early CT—in one study, early CTwas performed within 1 hour of presen-tation, and in the other study, it was per-formed within 24 hours—in patients whopresented with acute abdominal pain(25,26). In these two studies, standardpractice involved conventional abdominaland chest radiography and, if necessary,additional CT. CT was requested in halfthe patients in the standard practicegroup. In the first trial (25), patients inthe early CT group had shorter hospitalstays, but this finding was not reproducedin the second randomized trial, whichwas powered to evaluate reductions inhospital stay (26). The percentages ofcorrect diagnoses 24 hours after admis-sion did not differ significantly betweenthe two patient groups (76% for early CT,75% for standard practice). However, asignificantly greater percentage of seriousdiagnoses were missed in the standardpractice group (21% vs 4%, P � .001).

Prospective studies involving the ex-amination of patients for whom the clini-cian ordered CT scanning have shownthat CT findings have a significant effecton diagnoses. In one study, the accuracyof the clinical diagnosis made before CTwas performed improved from 71% to93% after CT was performed (27). The

accompanying change in treatment man-agement was 46%. Another study re-vealed a significant increase in the level ofconfidence of the diagnosis made withCT: The treatment management for 60%of patients was changed (28). AbdominalCT reportedly yields good overall interob-server agreement and very good interob-server agreement for the determinationof specific urgent diagnoses, with re-ported � values of 0.84, 0.90, and 0.81 foragreement regarding the diagnoses of ap-pendicitis, diverticulitis, and bowel ob-struction, respectively (29).

Exposure to ionizing radiation is a dis-advantage of CT. The effective radiationdose for abdominal CT is approximately10 mSv. In comparison, the annual back-ground radiation dose in the UnitedStates is approximately 3 mSv. A 10-mSvCT examination performed in a 25-year-old person is associated with an esti-mated risk of induced cancer of one in900 individuals and a risk of induced fatalcancer of about one in 1800 individuals(30). For older individuals, these risks areconsiderably lower. These risks should beweighed against the direct diagnostic ben-efit and related to the lifetime cancer risk:One in three people will develop cancerwithin their lifetime. It is important tonote that the effective radiation dose atabdominal CT may be reduced to someextent. In studies of appendicitis and di-verticulitis, standard-radiation-dose(100- and 120-mAs) nonenhanced CTwas compared with 30-mAs nonen-hanced CT (31,32). There were no signif-icant differences in accuracy between thelow- and standard-dose CT examinations.In a series of 58 patients suspected ofhaving appendicitis, low-dose CT withoral contrast medium had accuracy com-parable to that of standard-dose CT withintravenous contrast medium (33). Ingeneral, the effective CT dose is influ-enced by the current dose modulationmethods, which balance image qualityand dose. The use of intravenous contrastmedium is a drawback in patients withimminent renal insufficiency.

MR ImagingMR imaging is not yet widely used in thediagnostic work-up of patients whopresent with acute abdominal pain to the

Figure 1

Figure 1: Axial CT images in 26-year-old slen-der woman clinically suspected of having acuteappendicitis. Differentiation between pelvic in-flammatory disease and appendicitis on US im-ages was difficult; therefore, CT was performed.(a) Nonenhanced CT scan findings were also in-conclusive because of absence of delineating fat.(b) CT scan obtained after intravenous and rectalcontrast material administration shows appendici-tis: a distended appendix with thickened wall (ar-row) and surrounding infiltration. B � bladder,C � cecum. Appendicitis was confirmed at sur-gery and histopathologic analysis. (Images cour-tesy of Saffire S.K.S. Phoa, MD, AcademischMedisch Centrum, Universiteit van Amsterdam,Amsterdam, the Netherlands.)



ED. The major advantage of MR imagingis the lack of ionizing radiation exposure.The high intrinsic contrast resolution ren-dered with MR imaging is another advan-tage, as intravenous contrast mediummay not be required. The high intrinsiccontrast resolution has the potential to beparticularly valuable for assessment anddiagnosis of pelvic disease in female pa-tients, but this has not been substanti-ated. In the past, MR imaging requiredlong examination times. Currently, withrecently introduced high-speed tech-niques, MR imaging protocols for patientswith acute abdominal pain involve exam-ination times shorter than 15 minutes.However, the lack of around-the-clockavailability of MR imaging is still a logisticproblem at many hospitals.

MR imaging has demonstrated prom-ising accuracy for the assessment and di-agnosis of appendicitis, albeit in a rela-tively small series of patients, who oftenwere pregnant (34) (Fig 2). MR imagingis also accurate in the diagnosis of diver-ticulitis (35). MR imaging is more accu-rate than CT for the diagnosis of acutecholecystitis and the detection of commonbile duct stones (36). However, the bodyof scientific research on the use of MRimaging in patients with acute abdominalpain is relatively limited. Therefore, theavailability of and expertise with this ex-amination are limited, and the cost-effectiveness has not been studied. Fur-ther research should be directed towardbetter defining the role of MR imaging inthe setting of acute abdominal pain, espe-cially as compared with US and CT. Atthis time, MR imaging is used in only se-lect cases at many institutions, primarilyafter US yields nondiagnostic findings inpregnant women. Current evidence indi-cates that MR imaging already could beused for a broader range of indications(37). MR imaging has contraindications,including claustrophobia, which may pre-vent MR imaging from being performed.

Acute Appendicitis

The prevalence of appendicitis in patientswho present with abdominal pain to theED is approximately 14% (Table E1 [on-line]). The starting symptom is generallynondescriptive visceral pain in the peri-

umbilical region, followed by nausea andvomiting. When the disease progresses,the pain typically migrates to the rightlower quadrant because of more localizedperitoneal inflammation. Owing to thisfrequent cause of acute abdominal pain,approximately 250 000 appendectomiesare performed annually in the UnitedStates (38). After mortality and morbid-ity, the important quality indicators ofcare in patients suspected of having ap-pendicitis are negative appendectomyrate and percentage of perforated appen-dicitis.

Making an accurate and timely diag-nosis of appendicitis is challenging for cli-nicians. A false-positive diagnosis maylead to unnecessary surgical exploration,which is associated with increased mor-tality risk, prolonged hospital stay, andincreased infection-related complicationrisk (39). A false-negative (missed) diag-nosis can lead to prolonged time to treat-ment and increased risk of perforation.Several nonmedical factors (ie, prehospi-tal time, availability of operating room foremergency surgery, time of presentation)have been shown to be significantly asso-ciated with perforated appendicitis (40).Compared with uncomplicated appendi-citis, perforated appendicitis is associatedwith a two- to tenfold increase in mortal-ity (Fig 3) (41).

Traditionally, acute appendicitis hasbeen diagnosed on the basis of clinicalfindings. There has been a low thresholdto perform appendectomy on the basis ofthe assumption that missed appendici-tis—and thus the chance of perforation—has more hazardous consequences thandoes appendectomy that reveals negativefindings. As a result, negative-finding ap-pendectomy rates of 12%–40% havebeen reported (42). Despite having highsensitivity (up to 100%), clinical evalua-tion has relatively low specificity (73%)(43). This means that surgeons are likelyto overestimate the presence of appendi-citis in patients who present to the ED.Some clinicians hold the view that imag-ing should be performed only in patientswho have equivocal clinical findings atpresentation. Direct appendectomy canbe performed in patients with classicsigns and symptoms at presentation, par-ticularly young men (44), whereas mak-

ing a clinical diagnosis is more difficult inwomen. A combination of clinical fea-tures, including pain migration, abdomi-nal rigidity, and elevated inflammatory

Figure 2

Figure 2: Axial fat-saturated half-Fourier ac-quisition with single-shot turbo spin-echo MRimage (1900/72 [repetition time msec/echo timemsec]) obtained in 28-year-old woman who was at18 weeks gestation, was clinically suspected ofhaving appendicitis, and had nondiagnostic USfindings. Image shows thickened retrocecal ap-pendix (arrow) with increased signal intensity andminimal infiltration of surrounding fat. Fundusuteri is seen directly anterior to the aorta. The diag-nosis of appendicitis was confirmed at surgery.

Figure 3

Figure 3: Coronal contrast-enhanced reformat-ted CT image in 28-year-old man clinically suspectedof having appendicitis shows inflamed appendix(straight arrow). Inflammation is more pronounced atthe appendiceal tip, and discontinuity of the appen-diceal wall is suggested. Small amounts of fluid(curved arrow) adjacent to the appendix are present.Adjacent fat inflammation (arrowheads) is morepronounced at the appendiceal tip. Surgery and his-topathologic analysis results proved the presence ofperforated retrocecal appendicitis.



parameters (45), has a high predictivevalue for appendicitis but is present inonly a small proportion of patients sus-pected of having appendicitis. This makesclinically determining the diagnosis diffi-cult in the majority of patients and em-phasizes the added value of imaging inpatients suspected of having appendicitis.

The exact role of imaging in the set-ting of suspected appendicitis is still amatter of debate. In general, radiography

does not play a role in the work-up (9);US and CT have important roles, al-though CT has better accuracy (16) (Figs4, 5). CT is the preferred imaging tech-

nique for the diagnosis and assessment ofappendicitis in the United States (46) andhas been shown to reduce the negative-finding appendectomy rate from 24% to3%, with a simultaneous increase in CTuse, from 20% to 85%, over a period of10 years (47). Two randomized trials re-vealed negative-finding appendectomyrates of 5% and 2% in patients who un-derwent routine CT, compared with ratesof 14% and 19% in patients in whom CTwas performed selectively on the basis ofclinical judgment (48,49). Furthermore,it has been shown that the routine use ofCT is cost-effective because it facilitates areduction in in-hospital costs of $447 perpatient by preventing unnecessary hospi-talizations and surgical explorations (50).

There are several individual CT find-ings that suggest a diagnosis of appendi-citis; an enlarged (�6 mm) appendix hasa high positive predictive value (51,52).Likewise, the sensitivity of adjacent fatinfiltration is high for the diagnosis of ap-pendicitis (52,53). However, the visual-ization of appendicoliths has been shownto have a low positive predictive value forthe diagnosis of appendicitis becausethese may also be present in individualswho do not have appendicitis (Fig 6) (51).CT has limitations in the detection of ap-pendiceal perforation. For the detectionof perforated appendicitis, extraluminalgas, abscess, focal appendiceal wall de-fect, and small-bowel obstruction (SBO)have high specificity at CT; however,these findings are not very sensitive(54,55). If appendicitis can be ruled out,the most common alternative imaging-based diagnoses are gynecologic diseases(Fig 7), diverticulitis, and colitis (31,56).Other alternative conditions, which re-quire conservative treatment, includeright-sided diverticulitis and epiploic ap-pendagitis (Fig 8).

Despite the high diagnostic perfor-mance of CT, this modality has the afore-mentioned drawbacks. Therefore, alter-native strategies for the diagnosis of acuteappendicitis that involve less use of CThave been proposed—for example, USperformed as the initial diagnostic test,with CT performed only secondarily, af-ter US has yielded nondiagnostic findings.However, US can be limited by gas-filledbowel, which may obscure the underlying

Figure 4

Figure 4: US image in 43-year-old man clini-cally suspected of having acute appendicitisshows noncompressible thickened (10-mm) ap-pendix surrounded by inflamed mesenteric fat(arrows) and some fluid (arrowhead). Calipersmark the appendix. A � iliac artery, V � iliac vein.

Figure 5

Figure 5: Axial CT image obtained after intra-venous contrast medium administration in 47-year-old man with 2-day history of right lowerquadrant pain and clinically suspected of havingacute appendicitis shows thickened appendix(arrow) with maximal diameter of 14 mm and adja-cent fat infiltration (arrowheads). C � cecum.

Figure 6

Figure 6: Axial abdominal CT image in 62-year-old woman with known factor V Leiden trom-bophilia and a fever without an apparent cause formore than 2 weeks, obtained after the administra-tion of oral and intravenous contrast media toexclude lymphoma shows an appendicolith (ar-row) in a noninflamed appendix as an incidentalfinding.

Figure 7

Figure 7: Axial CT scan obtained after intrave-nous contrast medium administration in 24-year-old woman with right lower quadrant pain, a clini-cal differential diagnosis of gynecologic disorder(tubo-ovarian abscess, pelvic inflammatory dis-ease, ovarian torsion) or appendicitis, and incon-clusive US findings shows a normal appendix(straight arrow) and an enlarged right ovary (ar-rowheads), which most likely is due to tubo-ovar-ian abscess or ovarian torsion. Free fluid andsome thickening of the peritoneum (curved ar-rows) are also visible. Laparoscopy revealed ovar-ian torsion. U � uterus.



abnormality and thus necessitate second-ary CT in many individuals (Fig 9). MRimaging might be an alternative to CT inthe future.

Acute Diverticulitis

Acute colonic diverticulitis is the secondmost common cause of acute abdominalpain and leads to 130 000 hospitalizationsin the United States annually (57). Theprevalence—and thus the incidence—ofdiverticulosis increases with age. Ten per-cent of the general population youngerthan 40 years and more than 60% of peo-ple older than 80 years are affected bydiverticulosis (58). Ten percent to 20% ofthe affected people will develop divertic-ulitis, which is localized on the left side ofthe colon in 90% of cases (59). Seventy-two percent of patients admitted to thehospital for diverticulitis have uncompli-cated diverticulitis. A sensitivity of 64%for the clinical diagnosis of acute divertic-ulitis in the ED has been reported—thatis, one-third of the cases are missed clin-ically (60). These patients are most oftensuspected of having acute appendicitis.Because the treatments for acute appendi-citis (appendectomy) and acute diverticuli-tis (mainly conservative treatment) are dif-ferent, the differentiation of these two diag-noses is important. The reported positivepredictive value of 53% for the presence ofdiverticulitis after clinical evaluation indi-cates that approximately 40% of primaryclinical diagnoses are false-positive (60).

The disease stage in patients with di-verticulitis is often determined by usingthe modified Hinchey classification sys-tem (61), in which imaging and/or surgi-cal findings are incorporated (Table).Most patients with uncomplicated di-verticulitis can follow a conservativetreatment regimen of antibiotics anddiet modification. In mildly ill patientswith a presentation clearly suggestive ofuncomplicated diverticulitis (Hincheystage 0 or 1a), the treatment decision isnot based on the imaging results butrather on the patient’s clinical status. Inpatients who have Hinchey stage 1b di-verticulitis with a small (�2 cm) abscess,treatment can be conservative as well.Patients with larger abscesses are treatedwith percutaneous drainage. Diverticuli-

tis-associated abscesses are found at CTin approximately 15% of patients (62)(Fig 10). The majority of these collec-tions, approximately 36%–59%, are me-socolic abscesses, which can be treatedwith percutaneous drainage. The divertic-ulitis recurrence rate is the highest (40%)in this group (63,64).

If patients do not respond to or dete-riorate while undergoing conservativetreatment, they will undergo surgery(61,65). Approximately 13% of patients

who are treated conservatively will have arecurrence of diverticulitis, and only 4%of patients will have a third episode (66).On the basis of the low recurrence rates,Broderick-Villa et al (66) proposed thatelective surgery is not indicated in thesepatients. In another study, 10% of thepatients underwent surgery after initialconservative treatment for diverticulitis(62). CT plays a role in confirming thediagnosis and staging suspected compli-cated disease. CT assists in therapeuticdecisions and in the detection of alterna-tive diseases, according to guidelines ofthe American Society of Colorectal Sur-geons (59,65). In a recent meta-analysis,the accuracies of US and CT in the assess-ment and diagnosis of diverticulitis werenot significantly different (15). Overallsensitivities were 92% for US and 94%for CT (P � .65), and overall specificitieswere 90% for US and 99% for CT (P �.07). The sensitivity of CT for the diagno-sis of alternative diseases was higher andranged between 50% and 100%.

Two frequently present CT findingsthat have high sensitivity for the diagnosis

Figure 8

Figure 8: Axial CT images in 25-year-oldwoman suspected of having appendicitis. At US,the appendix was not well visualized; therefore, CTwas performed after intravenous contrast mediumadministration. (a) Image shows right-sided diver-ticulitis, indicated by right-sided colon diverticula(arrow) and fecalith with thickened wall, wall en-hancement, and adjacent fat infiltration (arrow-heads). (b) Image shows some secondary wallthickening of the adjacent appendix (arrow), withair in the lumen. Only some secondary changes—and no appendicitis—are seen. C � cecum.

Figure 9

Figure 9: Axial CT image obtained after intra-venous administration of contrast medium in 39-year-old woman with classic clinical manifesta-tions of appendicitis shows retrocecal inflamedappendix (arrow) with thickened wall and somesurrounding infiltration. The appendix could notbe visualized at US because of overlying (bowel)gas. Appendicitis was confirmed at surgery andhistopathologic analysis. C � cecum.



of diverticulitis are wall thickening (95%sensitivity) and fat stranding (91% sensi-tivity). Although fascial thickening and in-flamed diverticulum are less frequentfindings, they have reported specificitiesof 97% and 91%, respectively (67). CT isused not only to make diagnoses but alsoto stage disease in patients with divertic-ulitis. CT can also be used to differentiatecolorectal cancer from diverticulitis. Fea-tures associated with the diagnosis of co-lon carcinoma are pericolonic lymph nodesand luminal mass, whereas pericolonic in-flammation and segment involvementlarger than 10 cm are more commonly as-sociated with diverticulitis (Figs 10, 11).However, these signs arenot very accurate,and cancer can be missed (Fig 10). There-fore, endoscopy and biopsy are often re-quired to make this differentiation afterthe clinical symptoms have resolved—often after 6 weeks.

Acute Cholecystitis

Cholecystolithiasis is the main cause ofacute cholecystitis, for which an esti-mated 120 000 cholecystectomies areperformed annually in the United States(68). The prevalence of acute cholecysti-tis is approximately 5% in patients whopresent with acute abdominal pain to theED. Traditionally, the diagnosis has beenbased on the clinical triad of right upperquadrant tenderness, elevated body tem-perature, and elevated white blood cellcount. In a prospective series of patientswith acute cholecystitis (69), however,this triad was present in only 8% of pa-tients. Relatively recently published To-kyo guidelines introduced diagnostic andseverity assessment criteria (70). The di-

agnostic criteria for acute cholecystitisare one local sign of inflammation (Mur-phy sign; mass, pain, and/or tendernessin right upper quadrant), one systemicsign of inflammation (fever, elevated C-reactive protein level, elevated whiteblood cell count), and confirmatory imag-ing findings. Cholecystitis severity is clas-sified as mild, moderate, or severe(stages I, II, and III, respectively). Mildcholecystitis is defined as cholecystitis in apatient who has mild inflammatorychanges adjacent to the gallbladder with-out organ dysfunction. Findings of mod-erate cholecystitis are elevated whiteblood cell count, palpable tender mass inthe right upper quadrant, duration ofcomplaints longer than 72 hours, andmarked local inflammation. Severe chole-cystitis is defined as cholecystitis com-bined with multiple organ dysfunctionsyndrome.

Radiologic findings have an importantinfluence on treatment management inpatients with cholecystitis and organ fail-ure due to sepsis. Percutaneous drainageof the inflamed gallbladder with delayedcholecystectomy can be a safe option. Inall other cases that do not involve severeinflammation or surrounding infiltrationat imaging, laparoscopic cholecystectomyshould be performed within 96 hours af-ter the start of the complaints. Imagingfindings are therefore essential in makingdecisions regarding treatment for chole-cystitis.

Several imaging techniques are avail-able for the evaluation of suspected acutecholecystitis. US is the most frequentlyperformed modality for right upper quad-rant pain and yields a sensitivity of 88%and a specificity of 80% in the diagnosis of

acute cholecystitis (71). Features of cho-lecystitis include gallbladder wall thicken-ing; enlarged tender, noncompressiblegallbladder; and adjacent infiltration orfluid collections (Figs 12, 13). Accordingto ACR appropriateness criteria, US isconsidered the most appropriate imagingmodality for patients suspected of havingacute calculous cholecystitis (72). In ahighly select study sample, CT alsoshowed good accuracy, with a sensitivityof 92% and a specificity of 99% (73). Inpatients with acute abdominal pain, CThas demonstrated accuracy comparableto that of US in the diagnosis of acutecholecystitis (69). US should be consid-ered the primary imaging technique forpatients clinically suspected of havingacute cholecystitis (72).

Bowel Obstruction

Bowel obstruction is a relatively frequentcause of acute abdominal pain (Table E1[online]). The majority of patients foundto have bowel obstruction after theypresent to the ED have an SBO.

SBO DiseaseSBO is primarily caused by postoperativeadhesions. The combination of vomiting,distended abdomen, and increased bowelsounds is suggestive of SBO and has apositive predictive value of 64% (74).Other patient characteristics and risk fac-tors associated with bowel obstructionare previous abdominal surgery, ageolder than 50 years, and history of consti-pation (74). In patients with SBO compli-cated by ischemia (strangulated hernia),immediate surgery is warranted, whereasmany other patients with low-grade ob-struction can be treated conservativelywith a nasogastric tube and bowel rest.Seventy-three percent of all patients whoare treated conservatively will not be re-admitted. In one series, however, 19% ofthe patients were readmitted for recur-rent obstruction—one-third of these sub-jects underwent surgery—and 8% of thepatients died (75). Approximately onequarter of patients who are initially foundto have SBO in the ED will undergo sur-gery (75). For adequate treatment, it isimportant to identify the cause (eg, adhe-sion, neoplasm, or hernia) and severity of

Modified Hinchey Classification of Disease Stage in Patients with Diverticulitis

Stage Characteristic

0 Mild clinical diverticulitis1a Confined pericolic inflammation, no abscess1b Confined pericolic inflammation with local abscess2 Pelvic, retroperitoneal, or distant intraperitoneal abscess3 Generalized purulent peritonitis, no communication with bowel lumen4 Feculent peritonitis, open communication with bowel lumen

Source.—reference 61.



the obstruction. An obstruction can bepartial or complete and complicated byischemia, especially in the case of closed-loop obstruction (strangulation). Unlikeadhesions, internal hernia is an uncom-mon cause of bowel obstruction in theWestern world. Accurate diagnosis ismandatory because of the risk of strangu-lation. Because clinical evaluation has lim-itations in the diagnosis of bowel obstruc-tion (74), imaging is routinely performedto identify the site, cause, and severity(high- vs low-grade) of the obstruction.All of these are important parameters tohelp guide patient treatment.

Radiography has long been the pri-mary imaging modality of choice for pa-tients suspected of having bowel obstruc-tion. Radiography has been reported tohave 69% sensitivity and 57% specificityin the diagnosis of bowel obstruction (76)(Fig 14).

CT has the best reported accuracyfor the diagnosis of SBO, with a sensitiv-ity of 94% and a specificity of 96% (77).In one study, the cause of the obstruc-tion was correctly identified at CT in

85% of patients with a clinically equivo-cal diagnosis of bowel obstruction (77).The clinical management was correctlyaltered for 23% of patients—mainlyfrom conservative to surgical treat-ment. Evaluating the location and differ-entiating the common causes of bowelobstruction are more difficult on ab-dominal radiographs (76). CT can alsoenable the differentiation between high-and low-grade obstruction (79). How-ever, the sensitivity of CT is markedlylower for the diagnosis of low-grade(64%) SBO than for the diagnosis ofhigh-grade SBO (80).

An important CT finding that may

suggest SBO is a clear change in boweldiameter. With SBO, loops proximal tothe transition point are distended,whereas loops distal to the transition

Figure 10

Figure 10: Axial CT image obtained after intra-venous contrast medium administration in 54-year-old man with a history of colitis that was diag-nosed at age 15 and a several-month-long historyof abdominal pain and weight loss, who presentedto the ED with progressive abdominal pain of 1week duration. Image shows a thickened sigmoidcolon with some surrounding infiltration (ar-row), a contained perforation (arrowhead), andmultiple abscesses (A). Histopathologic analy-sis revealed extensive perforated diverticulitisand adenocarcinoma. (Image courtesy of C.Yung Nio, MD, Academisch Medisch Centrum,Universiteit van Amsterdam, Amsterdam, theNetherlands.)

Figure 11

Figure 11: (a, b) Axial CT images obtainedafter intravenous, oral, and rectal contrast materialadministration in 46-year-old man with 2-yearhistory of abdominal pain and recent progressiveacute abdominal pain. He had not defecated for thepast 2 days and had experienced weight loss of 12kg during the past year. Acute diverticulitis wasclinically suspected, with colorectal cancer as adifferential diagnosis. Images show (a) apple-corestenosis (arrow) of the sigmoid colon caused bycolorectal cancer and (b) proximal prestenoticdilatation of descending colon and cecum (arrow).

Figure 12

Figure 12: US image obtained in 79-year-oldman with 4-day history of right upper quadrantpain, nausea, and vomiting shows a thickenedgallbladder wall (arrowheads) and an obstructinggallstone (arrow), which was position indepen-dent. The patient was initially treated with percuta-neous gallbladder drainage.

Figure 13

Figure 13: Axial CT image obtained after ad-ministration of oral and intravenous contrast me-dia in 73-year-old obese woman with abdominalpain, fever, elevated C-reactive protein level (400mg/L), and a normal white blood cell count showscholecystitis with wall thickening, radio-opaquegallstones (arrow), and some adjacent fat infiltra-tion. The broad clinical differential diagnosis inthis patient included cholecystitis, diverticulitis,and appendicitis. The acute cholecystitis wastreated with percutaneous drainage because of thispatient’s critical condition.



point are collapsed. A helpful sign foridentifying the point of obstruction is thesmall-bowel feces sign—that is, feces-likematerial in the distended small bowel(81) (Fig 15). The transition point shouldbe scrutinized for the cause of the ob-struction (adhesion, neoplasm, hernia, orinflammatory disorder).

Because SBO is most often due to ad-hesions, which are usually not visible atCT, in most patients, no cause will beidentified at CT. In this setting, the diag-nosis of SBO due to adhesions is made bymeans of exclusion. In some patients withadhesions, an adhesive band may be sug-gested if extraluminal compression andkinking (acute angle) of the bowel arepresent (82).

Internal hernias can be difficult toidentify. The sensitivity and specificity ofCT in the diagnosis of internal hernias aremoderate, 63% and 73%, respectively(83). A cluster of small-bowel loops is aCT finding strongly associated with thediagnosis of internal hernia (odds ratio,7.9) (84).

The ACR proposes that abdominopel-vic CT with intravenous contrast mediumis the most appropriate imaging examina-tion when complete or high-grade SBO issuspected (85). If low-grade or intermit-tent SBO is suspected, several CT tech-niques (barium or water as the contrastagent, enteroclysis) or small-bowel exam-inations (follow-through, enteroclysis)are more or less equally appropriate, ac-cording to the ACR.

Other imaging modalities such as USand MR imaging are not widely used, al-though US reportedly has good accuracy(81%) in the diagnosis of bowel obstruc-tion (84). Fluid-filled loops are easily visu-alized at US, and one can easily differen-tiate between mechanical obstruction andparalytic ileus by visualizing peristalticmovement (84). US has important limita-tions: Gas-filled loops may obscure theunderlying abnormality, which has im-portant treatment management–relatedimplications, and the obstruction is diffi-cult to stage accurately. The ACR statesthat US is the least appropriate imagingmodality when high- or low-grade SBO issuspected (85). Evidence supporting theaccuracy of MR imaging in the diagnosisof bowel obstruction is limited, but results

are promising. A sensitivity of 95% and aspecificity of 100% were reported in asmall study by Beall et al (86). Overall, CTcan be considered the primary imagingtechnique for patients suspected of havingSBO.

Large-Bowel ObstructionLarge-bowel obstruction (LBO) is mostcommonly caused by colorectal cancer(60% of cases) (Fig 11), sigmoid volvulus(10%–15% of cases), and diverticulitis(10% of cases). The clinical features ofLBO—abdominal pain, constipation, andabdominal distention—are not very spe-cific. Therefore, the clinical diagnosis isoften incorrect. Traditionally, conven-tional radiography has been the initialimaging examination performed. Atpresent, CT is more often used to identifythe cause of the obstruction, the level ofobstruction, and the presence of a com-plicated obstruction (eg, strangulation).LBO is diagnosed at CT if the colon isdilated (colon diameter � 5.5 cm, cecumdiameter � 10 cm) and filled with feces,gas, and fluid proximal to an abrupt tran-sition point, after which the colon is col-lapsed distally. The accuracy of CT in thediagnosis of LBO has been reported inonly one study—that by Frager et al (87).In that study, CT had a sensitivity of 96%and a specificity of 93%. LBO can be di-agnosed with barium enema examina-tion. Although obstruction can often beadequately detected with barium enemaexamination, unlike CT, barium enemaexamination generally does not enablethe visualization of mural changes and ex-tracolonic abnormalities. CT is the imag-ing modality of choice in the diagnosis ofLBO in patients. For short luminal seg-mental obstruction caused by colorectalcancer, a colonic stent can be placed aseither a palliative treatment or a “bridge”to elective surgery (88). Volvulus of thelarge bowel can be reduced endoscopi-cally.

Perforated Viscus

Acute abdominal pain as a result of gas-trointestinal tract perforation is mostcommonly caused by a perforated gas-troduodenal peptic ulcer or diverticulitis.Less frequent causes include carcinoma

Figure 14

Figure 14: Upright conventional abdominalradiograph obtained in 59-year-old man who hadabdominal pain and a distended abdomen at clini-cal evaluation, as well as a history of SBO 3 yearsago, for which he was treated conservatively,shows distended small-bowel loops and air-fluidlevels (arrowheads), consistent with SBO. Theprevious obstruction was most likely caused byadhesions because the patient had previouslyundergone appendectomy. This patient was againtreated conservatively.

Figure 15

Figure 15: Sagittal reconstructed CT image in47-year-old woman who had a history of lysis ofadhesions and presented with cramping pain of 2days duration, nausea, and vomiting shows thetransition point (arrow) and the small-bowel fecessign (arrowheads) proximal to the transition point.No mass is visible, and the diagnosis is obstruc-tion by adhesions. The patient was treated conser-vatively with a successful outcome.



and bowel ischemia. Perforated viscus isa generally recognized diagnosis, al-though the incidence of this abnormalitywith free perforation is low (1%) in theED (Table E1 [online]). Perforation of apeptic ulcer is now less frequent because ofthe availability of adequate medical therapyfor peptic ulcer disease. Among patientswho are evaluated for possible acute diver-ticulitis, only 1%–2% have free perforation(57). Most perforated diverticula are con-tained perforations (Fig 16).

Because the clinical symptoms offree perforation are associated with theunderlying cause of the perforation, theclinical presentations of patients withperforated viscus are quite variable. Be-sides the variable symptoms of the un-derlying mechanism, a rigid abdomenusually is present. Recognizing a perfo-ration and establishing the cause andsite of the perforation can yield crucialinformation for the surgeon. Formerly,suspected free intraperitoneal air wasalways an indication to perform sur-gery. Currently, with the increased useof CT, contained perforations are morecommonly diagnosed, and the initialtreatment for these may be conserva-tive. For example, contained perforatedpeptic ulcers and Hinchey type 2 diver-ticulitis with peridiverticular air bubbles(Fig 16) are often treated with conser-vative management.

Upright posteroanterior chest radiog-raphy traditionally has been used for theinitial examination of patients suspected

of having pneumoperitoneum. Pneumo-peritoneum is visualized as a translucentcrescent or area below the diaphragm.Left lateral decubitus radiography is analternative in patients who are not able tostand upright. CT is currently replacingconventional radiography for this indica-tion. This reflects the fact that multisec-tion CT is more sensitive for the detectionof smaller amounts of free intraperitonealair. Conventional radiography is insensi-tive for the detection of air pocketssmaller than 1 mm and only 33% sensi-tive for the detection of 1–13-mm pockets(89). The major advantage of CT, as com-

pared with radiography and US, is that itcan correctly depict the actual site of per-foration in 86% of cases (90). A concen-tration of extraluminal air bubbles, a focaldefect of the bowel wall, and segmentalbowel wall thickening are CT findingssubstantially associated with correct iden-tification of the location of a perforation(Fig 17). The location of the free air is auseful indicator of the site of the perfora-tion. If free air is located around the liverand stomach, this most likely indicates agastroduodenal perforation. Free air de-tected predominantly in the pelvis andsupramesocolic and inframesocolic re-

Figure 16

Figure 16: Axial CT images obtained in (a) abdominal and (b) lung window settings after intravenous contrast medium administration in 71-year-old woman who hada 2-day history of left lower quadrant pain and was suspected of having diverticulitis show diverticulitis of the sigmoid colon with a contained perforation (arrow) andinfiltration of pericolic fat. The patient was treated conservatively with antibiotics.

Figure 17

Figure 17: Axial CT images obtained after intravenous administration of contrast medium in 54-year-oldwoman who presented to the ED with acute periumbilical abdominal pain that radiated to the back. The ab-dominal pain started after the woman ingested a nonsteroidal antiinflammatory drug. (a) Image obtained inlung window setting shows free intraperitoneal air (arrow). (b) Image shows wall thickening at the duodenalbulb and evidence of perforation (arrow), with adjacent soft-tissue infiltration and air bubbles (arrowhead). Adiagnosis of perforated duodenal ulcer was made and confirmed at surgery.



gions makes perforation of the colon orappendix more likely (90). Multiplanarreformations at CT are helpful for identi-fying perforations (91). A perforation canbe diagnosed at US when echogenic linesor spots with comet-tail reverberation ar-tifacts representing free intraperitonealair are seen adjacent to the abdominalwall in a supine patient. A sensitivity of92% and a specificity of 53% have beenreported for the detection of perforationwith US and constitute an overall accu-racy of 88% (92). It is important to notethat establishing the cause and location ofthe perforation is difficult with US.

Bowel Ischemia

Although bowel ischemia is a potentiallylife-threatening cause of acute abdominalpain, it is present in only about 1% ofpatients who present with acute abdom-inal pain (Table E1 [online]) (92,93).Gastrointestinal blood flow normallycomprises 20% of cardiac output. Thiscan increase to 35% postprandially and

decrease to 10% in critical situations suchas hypovolemia (93). If the blood supplyto the bowel decreases any further, mes-enteric ischemia will develop. Acutebowel ischemia can be caused by occlu-sion of the arteries (60%–70% of cases)or veins (5%–10% of cases) or by nonoc-clusive diminished vascular perfusion(20%–30% of cases) (94,95).

Patients with bowel ischemia oftenhave a short clinical history of prominentabdominal pain, while other possiblesymptoms such as nausea, vomiting, diar-rhea, and distended abdomen are sub-stantially less prominent. All of thesesymptoms are nonspecific. A diagnosis ofbowel ischemia is often made after themore frequently occurring diagnoses withsimilar associated symptoms are ex-cluded. Bowel ischemia should be consid-ered especially in elderly patients withknown cardiovascular disease (eg, atrial

fibrillation) and in younger patientsknown to have diseases that may causeinadequate mesenteric blood flow, suchas vasculitis, hereditary or familial coagu-lation disorders such as antiphospholipidsyndrome, and protein C or S deficiency.Laboratory findings such as elevated lac-tate level, elevated amylase level, and leu-kocytosis are nonspecific nonearly signsof ischemia (93).

In many patients with mesenteric isch-emia, the differential diagnosis is broadand includes peptic ulcer disease, bowelobstruction, pancreatitis, inflammatorybowel disease, appendicitis, and divertic-ulitis. Diagnostic imaging is always war-ranted to establish the diagnosis in atimely manner and differentiate betweenarterial and venous occlusive bowel isch-emia. Acute arterial mesenteric ischemiais treated surgically, with percutaneousthrombolytic treatment as an alternative

Figure 18

Figure 18: Portal venous phase CT imageobtained after intravenous administration of con-trast medium in 58-year-old woman who pre-sented to the ED with abdominal pain, anal bloodloss of several hours duration, and an abdomenthat had been distended for the past 2 days. Shehad a history of breast cancer and hypertension.Bowel ischemia was clinically suspected. Imageshows superior mesenteric vein occlusion(straight arrow); the superior mesenteric artery isopen. A transition point (arrowheads) is alsoclearly visible, with a normal small bowel proxi-mally and a thickened bowel wall with decreasedenhancement distally. Free peritoneal fluid(curved arrow) is also seen. This patient under-went surgery, during which a large part of the smallbowel was resected, and recovered uneventfully.

Figure 19

Figure 19: Multiplanar reformatted abdominal CT images obtained in (a) soft-tissue and (b) lung windowsafter intravenous administration of contrast material in 59-year-old woman with nausea and vomiting who hadundergone sigmoid colon resection for a gastrointestinal stromal tumor 5 days earlier. A distended abdomenidentified at physical examination and an increasing C-reactive protein level were noted. Images show por-tovenous gas (straight arrows) in the periphery of the liver and pneumatosis (curved arrows). The bowel wall(arrowheads) is thickened and enhanced. On the basis of these CT findings, bowel ischemia was considered.However, the clinical findings were more suggestive of bacterial translocation. The patient responded well totreatment with antibiotics. (Images courtesy of Ludo F.M. Beenen, MD, Academisch Medisch Centrum, Uni-versiteit van Amsterdam, Amsterdam, the Netherlands.)



(96). In some patients with low-gradeischemia, vascular reconstructive surgerycan be performed to preserve the mesen-teric blood supply. In these patients, CTinformation can be of vital importance.Venous mesenteric ischemia is usually nottransmural and can often be treated con-servatively with anticoagulative therapy(97). Venous mesenteric ischemia inclosed-loop obstruction requires treat-ment of the mechanical obstruction.

Formerly, the diagnosis of bowel is-chemia was made with angiography. CTis currently used to identify the primarycause, severity, location, and extent ofthe bowel ischemia. Angiography can beused to confirm the diagnosis of bowelischemia and treat occlusive bowel isch-emia. Angiography is less accurate incases of nonocclusive mesenteric isch-emia than in cases of occlusive mesen-teric ischemia.

In patients suspected of having mes-enteric ischemia, biphasic CT performedduring the arterial and venous phases isparticularly useful. Volume rendering ormultiplanar reformation facilitates evalu-ation of the vessels. Arterial phase CT isvery helpful for evaluating the celiac trunkand the mesenteric arteries. A venousphase CT scan also can show occlusionsof mesenteric arteries, but it predomi-nantly enables evaluation of the mesen-teric veins, bowel wall, and other causesof acute abdominal pain (Fig 18).

Although several CT signs are associ-ated with bowel ischemia, these signs arenot very frequent or specific (Fig 19). Vi-sualized occluded mesenteric arteries orvenous thrombus is a clear sign of mesen-teric ischemia (Fig 18). The bowel wallmay be thickened (�3 mm) because ofmural edema, hemorrhage, congestion,or superinfection. Thickening owing toedema, congestion, or hemorrhage is afrequent finding of venous obstruction.Bowel wall hypoattenuation (edema),bowel wall hyperattenuation (hemor-rhage), abnormal bowel wall enhance-ment (target sign), and absence of bowelwall enhancement are features of bowelischemia. The absence of bowel wall en-hancement is highly specific but is oftenmissed. The bowel wall may become pa-per thin, and this may indicate impendingperforation. Luminal dilatation and fluid

levels (fluid exudation of the ischemicbowel segments) are common in irrevers-ible bowel ischemia, and mesentericstranding and ascites are nonspecific CTfindings of bowel ischemia.

Pneumatosis cystoides intestinaliscan be present and manifest as a singlegas bubble or a broad rim of air divid-ing the bowel wall into two layers.Pneumatosis was formerly thought tobe highly associated with a diagnosisof bowel ischemia. However, pneuma-tosis can also be caused by infectious,inflammatory, or neoplastic disordersand is therefore a nonspecific sign forthe diagnosis of bowel ischemia.When pneumatosis cystoides intesti-nalis is seen in combination with por-tal venous gas, especially in the liverperiphery, it is definitely associatedwith bowel ischemia but is not a pa-thognomonic finding. Portal venousgas is an ominous sign that is generallyseen in patients with a poor prognosis.

The reported accuracy of CT in thediagnosis of bowel ischemia is compara-ble to the accuracy of angiography. Sen-sitivities of 93% for CT and 96% for an-giography (98) and specificities of 79%for CT and 99% for angiography (94)have been reported. In contrast, a morerecent study showed CT to have sensitiv-ity as low as 14% and a specificity of 94%(99). The disappointing results of thatstudy suggest that radiologists may be un-aware of the signs and symptoms ofbowel ischemia. This diagnosis should bemore commonly considered in patientswith acute abdominal pain, especiallyolder patients with known cardiovasculardisease. The use of contrast-enhanced USfor the diagnosis of bowel ischemia hasbeen evaluated. Contrast-enhancedDoppler US reportedly has a sensitivityof 63% when the color signals are di-minished and 80% when the color sig-nals are absent (100). CT is currentlythe preferred imaging modality for theassessment and diagnosis of bowel is-chemia, despite a number of indeter-minate signs.

Conclusions

The clinical findings–based diagnosis ren-dered in patients with acute abdominal

pain is often inaccurate. Therefore, imag-ing plays an important role in the treat-ment of patients with acute abdominalpain. Because US and CT are widely avail-able, radiography is rarely indicated forthe examination of patients with acute ab-dominal pain, with the exception of selectpatients groups—for example, patientswith bowel obstruction. CT is an effectiveexamination with results that have a pos-itive effect on the treatment of many pa-tients with acute abdominal pain. Atpresent, CT can be considered the pri-mary imaging technique for patients withacute abdominal pain, with the exceptionof patients suspected of having acute cho-lecystitis. US is preferable in these pa-tients, but CT is an acceptable alterna-tive.

The widespread use of CT raises im-aging costs. To our knowledge, the cost-effectiveness of increased CT use hasbeen studied—with increased CT useproved to be cost-effective—only in pa-tients suspected of having acute appendi-citis. This issue should be further evalu-ated for patients with acute abdominalpain who present to the ED. Radiationexposure is a drawback of CT; therefore,US may serve as an initial diagnostic test.CT may then be reserved for patientswith nondiagnostic US results (24). MRimaging has the potential to advance as avaluable alternative to CT, but supportivedata are still scarce.

Acknowledgment: Johan S. Lameris, MD, isacknowledged for his comments regarding thismanuscript.

References

1. Powers JH. Acute appendicitis during thelater decades of life: some remarks on theincidence of the disease in a rural area. AnnSurg 1943;117:221–233.

2. Fryback DG, Thornbury JR. The efficacy ofdiagnostic imaging. Med Decis Making1991;11:88–94.

3. Mackenzie R, Dixon AK. Measuring the ef-fects of imaging: an evaluative framework.Clin Radiol 1995;50:513–518.

4. ACR appropriateness criteria, 2006. Ameri-can College of Radiology Web site. http://www.acr.org/SecondaryMainMenuCategories/quality_safety/app_criteria/pdf/ExpertPanelonGastrointestinalImaging/AcuteAbdominal



PainandFeverorSuspectedAbdominalAbscessDoc1.aspx. Accessed October 15, 2008.

5. MacKersie AB, Lane MJ, Gerhardt RT, etal. Nontraumatic acute abdominal pain: un-enhanced helical CT compared with three-view acute abdominal series. Radiology2005;237:114–122.

6. Kellow ZS, MacInnes M, Kurzencwyg D, etal. The role of abdominal radiography in theevaluation of the nontrauma emergency pa-tient. Radiology 2008;248:887–893.

7. Ahn SH, Mayo-Smith WW, Murphy BL,Reinert SE, Cronan JJ. Acute nontraumaticabdominal pain in adult patients: abdomi-nal radiography compared with CT evalua-tion. Radiology 2002;225:159–164.

8. Otero HJ, Ondategui-Parra S, Erturk SM,Ochoa RE, Gonzalez-Beicos A, Ros PR. Im-aging utilization in the management of ap-pendicitis and its impact on hospitalcharges. Emerg Radiol 2008;15:23–28.

9. Rao PM, Rhea JT, Rao JA, Conn AK. Plainabdominal radiography in clinically sus-pected appendicitis: diagnostic yield, re-source use, and comparison with CT. Am JEmerg Med 1999;17:325–328.

10. Anyanwu AC, Moalypour SM. Are abdomi-nal radiographs still overutilised in the as-sessment of acute abdominal pain? a dis-trict general hospital audit. J R Coll SurgEdinb 1998;43:267–270.

11. Gupta K, Bhandari RK, Chander R. Com-parative study of plain film abdomen andultrasound in non-traumatic acute abdo-men. Ind J Radiol Imaging 2005;15:109–115.

12. van Randen A, Lameris W, Bossuyt PM,Boermeester MA, Stoker J. Comparison ofaccuracy of ultrasonography and computedtomography in patients with acute abdomi-nal pain at the emergency department [ab-str]. In: Radiological Society of NorthAmerica scientific assembly and annualmeeting program. Oak Brook, Ill: Radiolog-ical Society of North America, 2009; 519.

13. Brenner DJ, Hall EJ. Computed tomography:an increasing source of radiation exposure.N Engl J Med 2007;357:2277–2284.

14. Puylaert JB, Rutgers PH, Lalisang RI, et al.A prospective study of ultrasonography inthe diagnosis of appendicitis. N Engl J Med1987;317:666–669.

15. Lameris W, van Randen A, Bipat S, BossuytPM, Boermeester MA, Stoker J. Gradedcompression ultrasonography and com-puted tomography in acute colonicdiverticulitis: meta-analysis of test accu-racy. Eur Radiol 2008;18:2498–2511.

16. van Randen A, Bipat S, Zwinderman AH,

Ubbink DT, Stoker J, Boermeester MA.Acute appendicitis: meta-analysis of diag-nostic performance of CT and graded com-pression US related to prevalence of dis-ease. Radiology 2008;249:97–106.

17. Walsh PF, Crawford D, Crossling FT,Sutherland GR, Negrette JJ, Shand J. Thevalue of immediate ultrasound in acuteabdominal conditions: a critical appraisal.Clin Radiol 1990;42:47–49.

18. McGrath FP, Keeling F. The role of earlysonography in the management of the acuteabdomen. Clin Radiol 1991;44:172–174.

19. Allemann F, Cassina P, Rothlin M, Largia-der F. Ultrasound scans done by surgeonsfor patients with acute abdominal pain: aprospective study. Eur J Surg 1999;165:966–970.

20. Dhillon S, Halligan S, Goh V, Matravers P,Chambers A, Remedios D. The therapeuticimpact of abdominal ultrasound in patientswith acute abdominal symptoms. Clin Ra-diol 2002;57:268–271.

21. Levin DC, Rao VM, Parker L, Frangos AJ,Sunshine JH. Ownership or leasing of CTscanners by nonradiologist physicians: arapidly growing trend that raises concernabout self-referral. J Am Coll Radiol 2008;5:1206–1209.

22. Mun S, Ernst RD, Chen K, Oto A, Shah S,Mileski WJ. Rapid CT diagnosis of acuteappendicitis with IV contrast material.Emerg Radiol 2006;12:99–102.

23. Huynh LN, Coughlin BF, Wolfe J, Blank F,Lee SY, Smithline HA. Patient encountertime intervals in the evaluation of emer-gency department patients requiring ab-dominopelvic CT: oral contrast versus nocontrast. Emerg Radiol 2004;10:310–313.

24. Lameris W, van Randen A, van Es HW, etal. Imaging strategies for detection of ur-gent conditions in patients with acute ab-dominal pain: diagnostic accuracy study.BMJ 2009;338:b2431.

25. Ng CS, Watson CJ, Palmer CR, et al. Eval-uation of early abdominopelvic computedtomography in patients with acute abdomi-nal pain of unknown cause: prospectiverandomised study. BMJ 2002;325:1387.

26. Sala E, Watson CJ, Beadsmoore C, et al. Arandomized, controlled trial of routineearly abdominal computed tomography inpatients presenting with non-specific acuteabdominal pain. Clin Radiol 2007;62:961–969.

27. Tsushima Y, Yamada S, Aoki J, MotojimaT, Endo K. Effect of contrast-enhancedcomputed tomography on diagnosis and

management of acute abdomen in adults.Clin Radiol 2002;57:507–513.

28. Rosen MP, Sands DZ, Longmaid HE 3rd,Reynolds KF, Wagner M, Raptopoulos V.Impact of abdominal CT on the manage-ment of patients presenting to the emer-gency department with acute abdominalpain. AJR Am J Roentgenol 2000;174:1391–1396.

29. van Randen A, Lameris W, Nio CY, et al.Inter-observer agreement for abdominalCT in unselected patients with acute ab-dominal pain. Eur Radiol 2009;19:1394–1407.

30. The 2007 recommendations of the Interna-tional Commission on Radiological Protec-tion. ICRP publication 103. Ann ICRP 2007;37:1–332.

31. Keyzer C, Zalcman M, De Maertelaer V,Coppens E, Bali MA, Gevenois PA, VanGansbeke D. Acute appendicitis: compar-ison of low-dose and standard-dose unen-hanced multi–detector row CT. Radiology2004;232:164–172.

32. Tack D, Bohy P, Perlot I, et al. Suspectedacute colon diverticulitis: imaging with low-dose unenhanced multi-detector row CT.Radiology 2005;237:189–196.

33. Platon A, Jlassi H, Rutschmann OT, et al.Evaluation of a low-dose CT protocol withoral contrast for assessment of acute ap-pendicitis. Eur Radiol 2009;19:446–454.

34. Oto A. MR imaging evaluation of acute ab-dominal pain during pregnancy. Magn Re-son Imaging Clin N Am 2006;14:489–501.

35. Oh KY, Gilfeather M, Kennedy A, et al.Limited abdominal MRI in the evaluation ofacute right upper quadrant pain. AbdomImaging 2003;28:643–651.

36. Aube C, Delorme B, Yzet T, et al. MR cholan-giopancreatography versus endoscopic sono-graphy in suspected common bile ductlithiasis: a prospective, comparative study.AJR Am J Roentgenol 2005;184:55–62.

37. Stoker J. Magnetic resonance imaging andthe acute abdomen. Br J Surg 2008;95:1193–1194.

38. Owings MF, Lawrence L. Detailed diag-noses and procedures: national hospitaldischarge survey, 1997. Vital Health Stat 131999;145:1–157.

39. Flum DR, Koepsell T. The clinical andeconomic correlates of misdiagnosed ap-pendicitis: nationwide analysis. Arch Surg2002;137:799–804.

40. Sicard N, Tousignant P, Pineault R, Dube S.Non-patient factors related to rates of rup-



tured appendicitis. Br J Surg 2007;94:214–221.

41. Andersson RE. The natural history andtraditional management of appendicitisrevisited: spontaneous resolution and pre-dominance of prehospital perforations im-ply that a correct diagnosis is more impor-tant than an early diagnosis. World J Surg2007;31:86–92.

42. Andersson RE. Meta-analysis of the clinicaland laboratory diagnosis of appendicitis.Br J Surg 2004;91:28–37.

43. Hong JJ, Cohn SM, Ekeh AP, Newman M,Salama M, Leblang SD. A prospective ran-domized study of clinical assessment versuscomputed tomography for the diagnosis ofacute appendicitis. Surg Infect (Larchmt)2003;4:231–239.

44. Paulson EK, Kalady MF, Pappas TN. Sus-pected appendicitis. N Engl J Med 2003;348:236–242.

45. Wagner JM, McKinney WP, Carpenter JL.Does this patient have appendicitis? JAMA1996;276:1589–1594.

46. Ralls PW, Balfe DM, Bree RL, et al. Eval-uation of acute right lower quadrant pain:American College of Radiology—ACR ap-propriateness criteria. Radiology 2000;215(suppl):159–166.

47. Raman SS, Osuagwu FC, Kadell B, Cryer H,Sayre J, Lu DS. Effect of CT on false posi-tive diagnosis of appendicitis and perfora-tion. N Engl J Med 2008;358:972–973.

48. Lee CC, Golub R, Singer AJ, Cantu R Jr,Levinson H. Routine versus selective ab-dominal computed tomography scan in theevaluation of right lower quadrant pain: arandomized controlled trial. Acad EmergMed 2007;14:117–122.

49. Walker S, Haun W, Clark J, McMillin K,Zeren F, Gilliland T. The value of limitedcomputed tomography with rectal contrastin the diagnosis of acute appendicitis. Am JSurg 2000;180:450–454.

50. Rao PM, Rhea JT, Novelline RA, MostafaviAA, McCabe CJ. Effect of computed tomog-raphy of the appendix on treatment of pa-tients and use of hospital resources. N EnglJ Med 1998;338:141–146.

51. Daly CP, Cohan RH, Francis IR, Caoili EM,Ellis JH, Nan B. Incidence of acute appendi-citis in patients with equivocal CT findings.AJR Am J Roentgenol 2005;184:1813–1820.

52. Rao PM. Cecal apical changes with appen-dicitis: diagnosing appendicitis when theappendix is borderline abnormal or notseen. J Comput Assist Tomogr 1999;23:55–59.

53. Pereira JM, Sirlin CB, Pinto PS, Jeffrey RB,Stella DL, Casola G. Disproportionate fatstranding: a helpful CT sign in patients withacute abdominal pain. RadioGraphics2004;24:703–715.

54. Bixby SD, Lucey BC, Soto JA, TheysohnJM, Ozonoff A, Varghese JC. Perforatedversus nonperforated acute appendicitis:accuracy of multidetector CT detection. Ra-diology 2006;241:780–786.

55. Horrow MM, White DS, Horrow JC. Dif-ferentiation of perforated from nonperfo-rated appendicitis at CT. Radiology 2003;227:46–51.

56. Wise SW, Labuski MR, Kasales CJ, et al.Comparative assessment of CT and sono-graphic techniques for appendiceal imag-ing. AJR Am J Roentgenol 2001;176:933–941.

57. Jacobs DO. Clinical practice: diverticulitis.N Engl J Med 2007;357:2057–2066.

58. Ferzoco LB, Raptopoulos V, Silen W. Acutediverticulitis. N Engl J Med 1998;338:1521–1526.

59. Stollman N, Raskin JB. Diverticular diseaseof the colon. Lancet 2004;363:631–639.

60. Laurell H, Hansson LE, Gunnarsson U.Acute diverticulitis: clinical presentationand differential diagnostics. Colorectal Dis2007;9:496–501.

61. Kaiser AM, Jiang JK, Lake JP, et al. Themanagement of complicated diverticulitisand the role of computed tomography.Am J Gastroenterol 2005;100:910–917.

62. Ambrosetti P, Becker C, Terrier F. Colonicdiverticulitis: impact of imaging on surgicalmanagement—a prospective study of 542patients. Eur Radiol 2002;12:1145–1149.

63. Kumar RR, Kim JT, Haukoos JS, et al. Fac-tors affecting the successful management ofintra-abdominal abscesses with antibioticsand the need for percutaneous drainage.Dis Colon Rectum 2006;49:183–189.

64. Siewert B, Tye G, Kruskal J, et al. Impact ofCT-guided drainage in the treatment of di-verticular abscesses: size matters. AJRAm J Roentgenol 2006;186:680–686.

65. Rafferty J, Shellito P, Hyman NH, BuieWD. Practice parameters for sigmoid di-verticulitis. Dis Colon Rectum 2006;49:939–944.

66. Broderick-Villa G, Burchette RJ, Collins JC,Abbas MA, Haigh PI. Hospitalization foracute diverticulitis does not mandate rou-tine elective colectomy. Arch Surg 2005;140:576–581.

67. Kircher MF, Rhea JT, Kihiczak D, NovellineRA. Frequency, sensitivity, and specificity of

individual signs of diverticulitis on thin-section helical CT with colonic contrast ma-terial: experience with 312 cases. AJR Am JRoentgenol 2002;178:1313–1318.

68. Strasberg SM. Clinical practice: acute cal-culous cholecystitis. N Engl J Med 2008;358:2804–2811.

69. Lameris W, van Randen A, Ten Hove W,Bossuyt PM, Boermeester MA, Stoker J.The clinical diagnosis of acute cholecystitisis unreliable [abstr]. In: Radiological Soci-ety of North America Scientific Assemblyand Annual Meeting Program. Oak Brook,Ill: Radiological Society of North America,2008;110.

70. Hirota M, Takada T, Kawarada Y, et al.Diagnostic criteria and severity assess-ment of acute cholecystitis: Tokyo guide-lines. J Hepatobiliary Pancreat Surg 2007;14:78–82.

71. Shea JA, Berlin JA, Escarce JJ, et al. Re-vised estimates of diagnostic test sensitivityand specificity in suspected biliary tract dis-ease. Arch Intern Med 1994;154:2573–2581.

72. Bree RL, Ralls PW, Balfe DM, et al. Evalua-tion of patients with acute right upperquadrant pain: American College of Radiol-ogy—ACR appropriateness criteria. Radiol-ogy 2000;215(suppl):153–157.

73. Bennett GL, Rusinek H, Lisi V, et al. CTfindings in acute gangrenous cholecystitis.AJR Am J Roentgenol 2002;178:275–281.

74. Bohner H, Yang Q, Franke C, Verreet PR,Ohmann C. Simple data from history andphysical examination help to exclude bowelobstruction and to avoid radiographic stud-ies in patients with acute abdominal pain.Eur J Surg 1998;164:777–784.

75. Foster NM, McGory ML, Zingmond DS, KoCY. Small bowel obstruction: a population-based appraisal. J Am Coll Surg 2006;203:170–176.

76. Maglinte DD, Balthazar EJ, Kelvin FM,Megibow AJ. The role of radiology in thediagnosis of small-bowel obstruction. AJRAm J Roentgenol 1997;168:1171–1180.

77. Megibow AJ. Bowel obstruction: evaluationwith CT. Radiol Clin North Am 1994;32:861–870.

78. Taourel PG, Fabre JM, Pradel JA, Sene-terre EJ, Megibow AJ, Bruel JM. Value ofCT in the diagnosis and management ofpatients with suspected acute small-bowelobstruction. AJR Am J Roentgenol 1995;165:1187–1192.

79. Yaghmai V, Nikolaidis P, Hammond NA,Petrovic B, Gore RM, Miller FH. Multide-tector-row computed tomography diagno-



sis of small bowel obstruction: can coronalreformations replace axial images? EmergRadiol 2006;13:69–72.

80. Maglinte DD, Gage SN, Harmon BH, et al.Obstruction of the small intestine: accuracyand role of CT in diagnosis. Radiology 1993;188:61–64.

81. Lazarus DE, Slywotsky C, Bennett GL,Megibow AJ, Macari M. Frequency andrelevance of the “small-bowel feces” signon CT in patients with small-bowel ob-struction. AJR Am J Roentgenol 2004;183:1361–1366.

82. Petrovic B, Nikolaidis P, Hammond NA,Grant TH, Miller FH. Identification of adhe-sions on CT in small-bowel obstruction.Emerg Radiol 2006;12:88–93.

83. Blachar A, Federle MP, Brancatelli G,Peterson MS, Oliver JH 3rd, Li W. Radi-ologist performance in the diagnosis ofinternal hernia by using specific CT find-ings with emphasis on transmesenterichernia. Radiology 2001;221:422–428.

84. Schmutz GR, Benko A, Fournier L, PeronJM, Morel E, Chiche L. Small bowelobstruction: role and contribution of sonog-raphy. Eur Radiol 1997;7:1054–1058.

85. Ros PR, Huprich JE. ACR appropriatenesscriteria on suspected small-bowel obstruc-tion. J Am Coll Radiol 2006;3:838–841.

86. Beall DP, Fortman BJ, Lawler BC, Regan F.

Imaging bowel obstruction: a comparisonbetween fast magnetic resonance imagingand helical computed tomography. Clin Ra-diol 2002;57:719–724.

87. Frager D, Rovno HD, Baer JW, Bashist B,Friedman M. Prospective evaluation of co-lonic obstruction with computed tomogra-phy. Abdom Imaging 1998;23:141–146.

88. Dionigi G, Villa F, Rovera F, et al. Colonicstenting for malignant disease: review ofliterature. Surg Oncol 2007;16(suppl 1):S153–S155.

89. Stapakis JC, Thickman D. Diagnosis ofpneumoperitoneum: abdominal CT vs. up-right chest film. J Comput Assist Tomogr1992;16:713–716.

90. Hainaux B, Agneessens E, Bertinotti R, etal. Accuracy of MDCT in predicting site ofgastrointestinal tract perforation. AJRAm J Roentgenol 2006;187:1179–1183.

91. Ghekiere O, Lesnik A, Millet I, Hoa D,Guillon F, Taourel P. Direct visualization ofperforation sites in patients with a non-trau-matic free pneumoperitoneum: added diag-nostic value of thin transverse slices andcoronal and sagittal reformations for multi-detector CT. Eur Radiol 2007;17:2302–2309.

92. Chen SC, Wang HP, Chen WJ, et al. Selec-tive use of ultrasonography for the detec-tion of pneumoperitoneum. Acad EmergMed 2002;9:643–645.

93. Levine JS, Jacobson ED. Intestinal ischemicdisorders. Dig Dis 1995;13:3–24.

94. Wiesner W, Hauser A, Steinbrich W. Ac-curacy of multidetector row computed to-mography for the diagnosis of acute bowelischemia in a non-selected study popula-tion. Eur Radiol 2004;14:2347–2356.

95. Herbert GS, Steele SR. Acute and chronicmesenteric ischemia. Surg Clin North Am2007;87:1115–1134.

96. Schoots IG, Levi MM, Reekers JA, LamerisJS, van Gulik TM. Thrombolytic therapy foracute superior mesenteric artery occlusion.J Vasc Interv Radiol 2005;16:317–329.

97. Kumar S, Sarr MG, Kamath PS. Mesentericvenous thrombosis. N Engl J Med 2001;345:1683–1688.

98. Zalcman M, Sy M, Donckier V, Closset J,Gansbeke DV. Helical CT signs in the diag-nosis of intestinal ischemia in small-bowelobstruction. AJR Am J Roentgenol 2000;175:1601–1607.

99. Sheedy SP, Earnest F, Fletcher JG, FidlerJL, Hoskin TL. CT of small-bowel ischemiaassociated with obstruction in emergencydepartment patients: diagnostic perfor-mance evaluation. Radiology 2006;241:729–736.

100. Hata J, Kamada T, Haruma K, Kusunoki H.Evaluation of bowel ischemia with contrast-enhanced US: initial experience. Radiology2005;236:712–715.



Documents

Imaging Patients with Acute Abdominal Pain › ... › abdomen-agudo-1.pdf · Imaging Patients with Acute Abdominal Pain1 JaapStoker,MD AdriennevanRanden,MD WytzeLame´ris,MSc MarjaA.Boermeester,MD