Ultrasonography of the Colon in Pediatric Ulcerative Colitis: A Prospective,Blind, Comparative Study with Colonoscopy

Fortunata Civitelli, MD1, Giovanni Di Nardo, MD, PhD1, Salvatore Oliva, MD1, Federica Nuti, MD, PhD1, Federica Ferrari, MD1,

Anna Dilillo, MD1, Franca Viola, MD1, Nadia Pallotta, MD2, Salvatore Cucchiara, MD, PhD1, and Marina Aloi, MD, PhD1

Objectives To evaluate the usefulness of colonic ultrasonography (US) in assessing the extent and activity ofdisease in pediatric ulcerative colitis (UC) and to compare US findings with clinical and endoscopic features.Study design Consecutive pediatric patients (n = 60) with a diagnosis of UC and suspected disease flare-up wereprospectively enrolled; of these, 50 patients were eligible for the study. All underwent clinical evaluation, bowel USwith color Doppler examination and colonoscopy. Blind USwas performed the day before endoscopy in all patients.The US assessed variables were bowel wall thickness >3 mm, bowel wall stratification, vascularity, presence ofhaustra coli, and enlarged mesenteric lymph nodes.Results The endoscopic extent of disease was independently confirmed in 47 patients by US that yielded a 90%concordance with endoscopy (95%CI 0.82-0.96). Multiple regression analysis showed that USmeasurements withan independent predictive value of severity at endoscopy were increased bowel wall thickness (P < .0008),increased vascularity (P < .002), loss of haustra (P = .031), and loss of stratification of the bowel wall (P = .021).Each variable was assigned a value of 1 if present. The US score strongly correlated with clinical (r = 0.94) and endo-scopic activity (r = 0.90) of disease (P < .0001).Conclusions Colonic US is a useful first line noninvasive tool to assess the extent and activity of disease in chil-dren with UC and to estimate the severity of a flare-up, prior to further invasive tests. (J Pediatr 2014;165:78-84).

Ulcerative colitis (UC) andCrohn’s disease (CD) are chronic disorders characterized by a relapsing-remitting clinical course,belonging to the group of inflammatory bowel disease (IBD). In patients with UC, an exhaustive evaluation of diseaseextent and activity is crucial for planning medical and surgical decisions.1,2 Unfortunately, clinical and endoscopic scores

of activity are either indirect or invasive, such as in acute patients in whom a complete colonoscopy should not be performedbecause of the risk of procedure-related complications.3Moreover, during follow-up, repeated investigations are not well acceptedby children and adolescents. Therefore, a direct, noninvasive method for assessing disease location and severity would represent anoteworthy progress in the clinicalmanagement of pediatric UC, particularlywhen a severe disease is suspected. Bowel ultrasonog-raphy (US) has become a well-established tool in the diagnostic work-up and in the follow-up of IBD, especially CD.4-7 US has theadvantages of being noninvasive, radiation-free, inexpensive, widely available, and generally well-tolerated by patients, so that itmay repeatedly be performed overtime. In adults, the diagnostic performance of US in detecting IBD colonic involvement largelydependson the site of disease; sensitivity is higher (97%) for sigmoid/descending colon and lower for those areasmore difficult tobeassessed, such as the rectum.8 In some studies, bowel US has also been shown to be useful in evaluating IBDactivity andmonitoringresponse tomedical treatment.9-14 A few studies have specifically investigated the role of bowel US in adult patients withUC,13,15,16

whereas pediatric studies included a relatively small numberof patients and consideredboth IBDentities,CDandUC.17,18 Thus,weconducted a prospective double-blind study aimed to evaluate the usefulness of colonic US in assessing the extent and activity ofdisease in a population of children with UC and to compare US findings with clinical and endoscopic features.

BW Bowel wall

BWT Bowel wall thickness

CD Crohn’s disease

CMV Cytomegalovirus

IBD Inflammatory bowel disease

MRE Magnetic resonance enterog

PUCAI Pediatric Ulcerative Colitis A

UC Ulcerative colitis

US Ultrasonography

78

Methods

Consecutive pediatric patients with an established diagnosis of UCwho required endoscopy for suspected flare-up of disease wereprospectively studied at a single tertiary center for pediatric gastroenterology and hepatology betweenMay 2012 andMay 2013. All

raphy

ctivity Index

patients underwent a complete clinical eval-uation, laboratory investigations, bowel US

From the 1Pediatric Gastroenterology and Liver Unit,Department of Pediatrics, and the 2Department ofInternal Medicine and Medical Specialties, SapienzaUniversity of Rome, Rome, Italy

The authors declare no conflicts of interest.

0022-3476/$ - see front matter. Copyright ª 2014 Elsevier Inc.

All rights reserved.

http://dx.doi.org/10.1016/j.jpeds.2014.02.055

Vol. 165, No. 1 � July 2014

with color Doppler examination, and ileocolonoscopy; USwasperformed the day before endoscopy in all patients. Treatmentwas kept stable in the interval between the 2 procedures. Thestudy was approved by the ethics committee of the UmbertoI University Hospital, and written informed consent was ob-tained from all enrolled children and their parents.

US and colonoscopy were performed by different opera-tors blinded to the patient’s clinical data but aware of thediagnosis of UC. All investigators reported their results ona predefined data collection form immediately after finishingthe examination. Endoscopy was considered the referencestandard for the evaluation of UC extent and activity. Theextent of disease was defined according to the Paris Classifi-cation for pediatric IBD19: ulcerative proctitis (E1) wasdefined as a disease limited to the rectum; left-sided UC(E2) as an involvement limited to the portion of the colorec-tum distal to the splenic flexure; extensive UC (E3) as a dis-ease proximal to the splenic flexure but distal to the hepaticflexure; and pancolitis (E4) included a colitis extended prox-imally to the hepatic flexure. Based on prior published datareporting a low US accuracy for disease confined to therectum, because of technical difficulties in exploring thedeep pelvis,8 patients with proctitis alone at endoscopywere excluded from the study. Patients with histologicallyproven cytomegalovirus (CMV) infection were excludedfrom the statistical analysis to avoid potential confoundingfactors that could alter the actual disease activity and extent.

Endoscopic activity was evaluated according to the Mayoendoscopic sub-score20: 0 = normal or inactive disease,1 = mild (erythema, decreased vascular pattern, mild fria-bility), 2 = moderate (marked erythema, absent vascularpattern, friability, erosions), and 3 = severe (spontaneousbleeding, ulceration). The clinical activity of disease was as-sessed by the Pediatric Ulcerative Colitis Activity Index(PUCAI): remission: 0-10, mild disease: 10-34, moderate:35-64, and severe: >65.21 Flare-ups were determined througha detailed clinical history and defined when the PUCAIwas >10.

EndoscopyIleocolonoscopy was performed in all patients under generalanesthesia. Colonic cleansing was obtained following thestandard protocol in our unit.22 In patients with severe dis-ease, endoscopy was carried out with limited insufflationsof air to minimize the risk of acute traumatic dilatation orperforation. Patients in which colonoscopy was notcompleted because of the severity of inflammation werenot included in the study. In all patients, biopsies to ruleout CMV infection were performed. To allow comparisonwith US, the colon was divided into 3 segments: right colon(cecum-ascending colon), transverse colon, and left colon(descending-sigmoid colon). Mayo endoscopic subscorewas calculated globally and per segment. All endoscopic pro-cedures were carried out by 2 experienced operators withmore than 7 (G.dN.) and 4 (S.O.) years of training in IBDendoscopy, using a pediatric colonoscope (PCF-Q180I;Olympus, Tokyo, Japan).

Colonic USAll US examinations (Aplio XG; Toshiba Equipment, Tokyo,Japan) were performed the day before colonoscopy by thesame pediatric gastroenterologist (F.C.), who had 5 years ofexperience in the setting of bowel US, and was unaware ofthe patient’s clinical data but was informed of the diagnosisof UC. The US was carried out after 6 hours fasting, withno specific bowel preparation, first by means of a3.5-5 MHz convex probe for a panoramic view of theabdomen and then with a high frequency (7.5-12 MHz)linear-probe, for a better evaluation of the bowel wall(BW).23 The right iliac fossa was first examined and the ter-minal ileum identified; the entire colon was then reviewedsystematically, starting in the left lower quadrant with visual-ization of the sigmoid colon and using left iliac vessels aslandmarks, proceeding backwards to the right lower quad-rant to identify all colonic segments. For the US analysis,the same division into 3 segments (right colon, transverse co-lon, and left colon) as for endoscopy was established. The USassessed variables were selected based on a literature searchand our own expertise in bowel US and included: BW thick-ness (BWT), BW stratification, vascularity, presence of haus-tra coli, and enlarged mesenteric lymph nodes.13-16,23,24 Themaximum BWT (obtained as a mean of three measures of themost thickened tract wall) was recorded, in a longitudinaland transverse section, from the central hyperechoic line ofthe lumen to the outer hyperechoic margin of the wall; aBWT greater than 3 mm was considered as abnormal.11,18,23

Extension of disease at US was thus defined by the identifica-tion of all continuous colonic segments displaying a wallthickness greater than 3 mm.16 Vascularity was studiedwith the color Doppler examination using the lowest pulserepetition frequency available (6.5/s): presence of color sig-nals within the BW was considered as pathologic.15,23 Lossof haustra coli, absence of the normal 5-layer stratificationof the BW23 as well as the presence of enlarged mesentericlymph nodes23,25 were other measurements of abnormality.These variables were assessed in each colonic segment. Inaddition, based on all US features, the overall segmental dis-ease activity was categorized as inactive, mild, moderate, andsevere.

Statistical AnalysesQuantitative variables are represented as median and rangewhereas frequencies and percentages are used for categoricaldata. Sensitivity, specificity, positive predictive value, andnegative predictive value of US for detecting UC lesionswere calculated in each colonic segment considering endos-copy as the gold standard. Comparison between the BWTand endoscopic severity and between the values of BWT inthe different colonic segments was performed usingKruskal-Wallis Test (nonparametric ANOVA) followed byDunn Multiple Comparisons Test. Multiple regression anal-ysis considering the Mayo endoscopic score as dependentvariable was used to determine the US findings (BWT, vascu-larity, stratification, presence of haustra, and enlargedmesenteric lymph nodes) with independent predictive value

79

Figure 2. Concordance betweenUS and endoscopy for loca-tion of disease.

Table II. SE, SP, PPV, and NPV of US in detection ofUC colonic lesions

US

Right colon Transverse colon Left colon

% 95% CI % 95% CI % 95% CI

SE 75 42-93 86 60-97 96 80-100SP 100 74-100 100 70-100 100 62-97PPV 100 62-100 100 71-100 100 83-100NPV 83 57-98 85 56-97 80 30-100

NPV, negative predictive value; PPV, positive predictive value; SE, sensitivity; SP, specificity.

THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. 165, No. 1

for quantification of endoscopic severity on a segment basis.All variables with P value of .05 or less were retained in themodel. An US score was, thus, obtained by adding 1 pointfor each of the US predictors of disease activity when present.A receiver operating characteristic curve analysis was used todetermine the accuracy of the US score in detecting severedisease at endoscopy. Correlation between US index, Mayoscore, and PUCAI were calculated using Spearman correla-tion coefficient. Concordance between US and endoscopyin assessing disease extent and in classifying disease activityas inactive, mild, moderate, and severe was calculated usingK statistics. The quantitative significance of k was calculatedaccording to Landis and Koch 0.01-0.20, slight agreement;0.21-0.40, fair agreement; 0.41-0.60, moderate agreement;0.61-0.80, substantial agreement, and 0.81-1.00, almost per-fect agreement.26 Statistical analysis was performed usingGraph Pad software for windows (GraphPad Software, Inc,San Diego, California).

Results

Consecutive patients with UCwith suspected disease flare-up(n = 60) were prospectively enrolled (Figure 1; available atwww.jpeds.com). Blinded US was performed in all patients.Five patients did not have a complete colonoscopy becauseof the severity of inflammation and were excluded from thestudy; 2 patients with histologically confirmed CMVinfection and 3 with isolated proctitis were not included inthe statistical analysis. A total of 50 children were eligiblefor the study, thus, providing data for comparison from150 colonic segments proximal to the rectum. Patientbaseline characteristics are shown in Table I (available atwww.jpeds.com). Extent of disease was E2 in 22 (44%)patients, E3 in 6 (12%), and E4 in 22 (44%). This extentwas independently confirmed in 47/50 patients by US, thatyielded a 90% concordance (k) with endoscopy for diseaselocation (95% CI 0.82-0.96) (Figure 2). In the 5 patients inwhich endoscopy was incomplete, US was able todocument an extensive colitis (E3) in 2 and a pancolitis(E4) in 3. Two of these patients finally underwentcolectomy, and the US extent of the disease (pancolitis)was confirmed in both by operative findings.

In the 3 patients with proctitis alone at endoscopy, UScorrectly assessed rectal involvement in 1, whereas it missedlesions in 2. There were no false positive US results in theother colonic segments. In the 2 patients with CMV infection,US documented extensive colitis in 1 and pancolitis in theother. This extent was confirmed on endoscopy that showeda moderately activity of disease in both patients.

The values of sensitivity, specificity, positive predictivevalue, and negative predictive value of US for detection ofUC lesions in the different colonic segments are shown inTable II.

A total of 150 colonic segments were considered for thestatistical analysis: 35 (24%) segments were normal at endos-copy, 29 (19%) showed mild disease activity, 40 (26%) mod-erate, and 46 (31%) severe disease. Themean BWT (mm) was

80

2 � 0.3 in the normal colonic segments, and it was 3 � 0.6,4.5� 0.7, and 5.5� 0.8 mm in the segments with mild, mod-erate, and severe disease, respectively. There was no statisti-cally significant difference between the BWT of normal andmildly inflamed segments nor between moderately andseverely inflamed segments, whereas the BWT was signifi-cantly different between moderately/severely inflamed andnormal (P < .001) and mildly (P < .01) inflamed segments(Figure 3, A). The mean value of BWT in affected colonicsegments significantly increased from the ascending colon(3 � 1.6 mm) to the left colon (5 � 1.8 mm) (P = .001)(Figure 3, B). The US features in the colonic segments withdifferent degree of endoscopic activity are reported inTable III (available at www.jpeds.com).Multiple regression analysis considering the Mayo endo-

scopic subscore as dependent variable showed that variableswith independent predictive value of endoscopic severitywere BWT (P = .0008), increased vascularity (P = .002),loss of the normal stratification of the BW (P = .021), andabsence of haustra coli (P = .031). The presence of enlargedmesenteric lymph nodes did not reach a statistical signifi-cance, but it was a frequent finding in moderate-severe dis-ease (Figure 4; available at www.jpeds.com). Based onthese results, each variable independently predicting theendoscopic severity was assigned a value of 1 (present) or0 (absent), thus obtaining a final US score ranged from0 to 4. The presence of more than 2 variables (US score

Civitelli et al

Figure 3. Dot plot of BWT and endoscopic disease severity inall bowel segments examined (n = 150). The mean value ofBWT was significantly different between normal segmentsand moderately and severely inflamed segments whereasthere was no statistically significant difference betweennormal and mildly inflamed segments nor between segmentswith moderate and severe disease at endoscopy.

Figure 5. Concordance between US and endoscopy for dis-ease activity.

July 2014 ORIGINAL ARTICLES

>2) had a sensitivity of 100% and a specificity of 93% (areaunder curve of 0.98) for detection of severe disease atendoscopy. The US score showed a strong and significantcorrelation with both endoscopic and clinical activity ofdisease, r = 0.94 and r = 0.90 (P < .0001), respectively.Thirty-nine (26%) segments were considered normal byUS, whereas 25 (17%), 42 (28%), and 44 (29%) wereclassified as mildly, moderately, and severely inflamed,respectively. The overall concordance between US andendoscopy in classifying segmental disease activity as

Ultrasonography of the Colon in Pediatric Ulcerative Colitis: A Pro

inactive, mild, moderate, and severe was K = 0.94 (95% CI0.88-1) (Figure 5).

Discussion

The management of patients with UC requires evaluationwith objective tools, both at the diagnosis and throughoutthe variable course of the disease.1,2 Pediatric-onset UC isdistinguished from adult-onset disease by a greater preva-lence of extensive inflammation.27,28 A disease extent beyondthe splenic flexure occurs in up to 80% of children and ado-lescents, increasing the likelihood for acute severe exacerba-tions.27-30 Therefore, the majority of young patients withUC undergo repeated endoscopies at the follow-up to deter-mine the extension, activity, and severity of inflammatory le-sions, as well as the occurrence of complications and tomonitor the response to treatment. Endoscopy is currentlyconsidered the standard tool to accomplish all these evalua-tions.1,2 Nevertheless, conventional colonoscopy may not al-ways be feasible in the setting of severe disease because of therisk of complications or disease activity deterioration.3More-over, patients are usually reluctant to undergo repeated colo-noscopies at every flare-up, because of procedure-relateddiscomfort and the inconvenience of the extensive bowelcleansing required for an optimal examination.31 Althoughclinical scores of activity have shown a fine correlation withendoscopic features,21,32 they are indirect indices, mainlyrelying on symptoms referred by patients.33 Moreover,some clinical symptoms in UC are not directly related tomucosal inflammation and endoscopic healing tends to lagbehind symptom improvement.2 Therefore, a noninvasivetool to assess the extent and activity of UC, especially whendisease extension or exacerbation is suspected, would bedesirable. Compared with endoscopy, US has the advantageof being noninvasive, less costly, well-tolerated, and easilyrepeatable, thus, representing a valuable tool in the manage-ment of pediatric patients with UC.In the last decade, bowel US has become a well-established

technique in the diagnostic work-up and in the follow-up of

spective, Blind, Comparative Study with Colonoscopy 81

THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. 165, No. 1

patients with IBD, particularly in CD.4-7 The usefulness of USfor detecting small bowel CD has been largely evaluated inprevious studies, whereas its application to assess colonic le-sions is less extensively investigated.13,15,16 The accuracy ofUS in evaluating IBD colonic involvement in adults dependson the anatomic distribution of the disease, being very high(>90%) for those areas easily accessible at US (ie, the left co-lon), and fairly low [14%] for the tracts more difficult to beassessed, such as the rectum.4,8,34 In pediatric patients, the re-ported overall sensitivity and specificity of US for colonic in-flammatory lesions is 88% and 93%, respectively.17 Faureet al evaluated the role of bowel US in detecting terminalileum and colonic lesions in 38 children who underwent co-lonoscopy for suspected or already known IBD; they reporteda sensitivity of 93% in the left-sigmoid colon and of 77% inthe right one, with a specificity of 100% and 91%, respec-tively.17 However, this study, as well as previous reports inadults, examined a population including both IBD entities,CD and UC. In our study, we have investigated in a prospec-tive, blinded, comparative fashion the role of US vs colonos-copy in defining the site and disease activity in a pediatricpopulation of UC only. The sensitivity of US was variable ac-cording to the different colonic segments, being higher forthe left (96%) rather than for the right colon (75%). Weshowed that US can accurately detect the anatomic locationof the disease, with a high level of agreement with endoscopy(k = 0.094).Moreover, in the 5 patients in which full colonos-copy was not done because of the severity of inflammation,US documented an extensive disease, suggesting that bowelUS might represent a useful noninvasive tool for establishingthe location of UC when endoscopy might be risky.

The potential use of US in assessing disease activity andmonitoring the response to medical therapy in IBD is still un-settled.9-13,15,35 The most studied variables to evaluate diseaseactivity at US are the BWT and the presence of blood flowwithin the BW that have shown a fine reproducibility.36

However, there are not yet formally validated scores to assessUC activity by US. Maconi et al reported a strong correlationbetween the degree of BWT at US and clinical and endo-scopic activity of disease, both before and after treatmentwith corticosteroids. Moreover, the value of BWT was signif-icantly different between patients with moderate/severeendoscopic activity and those with mild/absent activity,both before and after treatment (P < .05).13 In a populationof 44 children with IBD, Bremner et al showed that a BWT>3 mm has a good predictive value for moderate/severe dis-ease in the colon proximal to the rectum.18 Parente et al pro-spectively evaluated 83 adults with moderate-to-severe UCundergoing colonoscopy and US at 3, 9, and 15 months afterrecruitment. They analyzed the maximum thickness ofcolonic wall, the echo-stratification, and the presence ofincreased vascularity within the BW, and they reported ahigh and consistent concordance between endoscopic andUS activity of disease at the different study times (k between0.76 and 0.90).13

In our study, the mean value of BWT was significantlydifferent between normal and moderately/severely inflamed

82

segments (P = .001). However, as previously reported,13 thevalue of BWT did not allow to distinguish between normaland mildly inflamed segments. Indeed, superficial lesions(ie, erythema and decreased vascular pattern) may be missedby US.23 Moreover, the value of BWT in the different colonicsegments significantly increased from the right to the left co-lon (P = .001). This might partially explain the difference insensitivity of US for lesions located in the ascending colon(75%) in respect to the left one (96%).We defined disease activity at US by considering conven-

tional variables (ie, BWT and increased vascularity of theBW)13,15,23-25 as well as other variables such as loss of haustracoli, altered stratification of the BW, and presence of enlargedmesenteric lymph nodes.23-25 The latter was very frequent inpatients withmoderate-to-severe disease, whereas at multipleregression analysis, it was not an independent predictor ofendoscopic disease severity. If the occurrence of extra-colonic lesions is a well-established feature of CD,37 this isa somewhat unexpected finding in UC, which probably re-flects the more severe course and the more systemic involve-ment of UC in children.In our study, the US features with an independent predic-

tive value of disease severity at endoscopy were BWT,increased vascularity, loss of normal stratification, and lossof haustra. According to this result, we also attempted toconstruct a US score to quantify the disease severity at US.The alteration of more than 2 of those variables had a highaccuracy (sensitivity 100%, specificity 93.3%) for detectingsevere disease at endoscopy; moreover, the US score stronglycorrelated with both endoscopic (r = 0.94) and clinical(r = 0.90) activity of disease. The strong relationship wefound between US and clinical and endoscopic data suggeststhat bowel US may represent a useful first line, noninvasivetool for evaluating disease extent and activity in childrenwith UC, particularly those with suspected severe disease inwhich endoscopy might not always be feasible. US may behelpful to determine in a relatively rapid manner whether asignificant disease flare-up is present and to guide in thework up of the majority of clinical circumstances, thus, al-lowing to delay or even avoid more invasive tests such as co-lonoscopy. In this context, bowel US may therefore bepreferred in clinical practice for monitoring disease courseand for assessing short-term treatment response. Moreover,colonic US does not require a specific bowel preparation, isradiation-free, and is generally tolerated well. The patientfriendliness of this technique is of importance in a chronic in-flammatory disorder such us UC, requiring frequent reassess-ment of the disease.There has been an interest for noninvasive tools in the

management of IBD. This has implications both for healthcare costs and for reducing risks due to invasive procedures.Although stool subclinical markers (ie, calprotectin and lac-toferrin) have gained wide acceptance as ancillary test for dis-ease activity and inflammation,38 imaging tools such as USand magnetic resonance enterography (MRE) can revealboth mucosal and structural bowel damage.7 This is of inter-est because therapeutic goals in IBD are evolving from

Civitelli et al

July 2014 ORIGINAL ARTICLES

control of clinical symptoms to achieving and maintainingmucosal healing as well as preventing progression of boweldamage. MRE seems to be a promising tool in adults withUC, however, it requires a bowel preparation with a consider-able volume of orally ingested polyethylene glycol solutionand usually needs a rectal enema for optimal colonic disten-sion.31 Finally, MRE is an expensive technique and other usesare often prioritized ahead of colonography.

This study has several strengths. It was prospectively con-ducted in a large homogenous population of pediatric pa-tients with UC, all undergoing colonoscopy, which is thegold standard tool in UC, with each investigator blinded tothe results of the other examinations. The main limitationof the study, however, is that US was performed at a singlecenter by the same operator with a sound expertise in IBDUS. Indeed, the latter requires peculiar skillfulness, not al-ways available in routine US procedures. In addition, super-ficial mucosal lesions may be missed at US in mildinflammation, so that the usefulness of US seems to be higherin patients with moderate-severe disease. Thus, further largermulticenter studies are warranted to confirm our observa-tions and to test the reproducibility of the technique as wellas the effect of the operator expertise on diagnosticperformance.

In conclusion, the high diagnostic accuracy in compari-son to endoscopy, noninvasiveness, low cost, and wide-spread availability make colonic US a useful tool forassessing the extent of the disease and activity in pediatricpatients with UC, especially in the setting of moderate-to-severe disease. n

Submitted for publication Oct 8, 2013; last revision received Jan 29, 2014;

accepted Feb 25, 2014.

Reprint requests: Marina Aloi, MD, PhD, Pediatric Gastroenterology and Liver

Unit, Department of Pediatrics, Inflammatory Bowel Disease Program,

Sapienza University of Rome, Viale Regina Elena, 324-00161 Rome, Italy.

E-mail: marina.aloi@uniroma1.it

References

1. Stange EF, Travis SPL, Vermeire S, Reinisch W, Geboes K,

Barakauskiene A, et al. European evidence-based consensus on the diag-

nosis and management of ulcerative colitis: definitions and diagnosis.

J Crohn’s Colitis 2008;2:1-23.

2. Turner D, Levine A, Escher JC, Griffiths AM, Russell RK, Dignass A, et al.

Management of pediatric ulcerative colitis: joint ECCO and ESPGHAN

evidence-based consensus guidelines. J Pediatr Gastroenterol Nutr 2012;

55:340-61.

3. Navaneethan U, Kochhar G, Phull H, Venkatesh PG, Remzi FH,

Kiran RP, et al. Severe disease on endoscopy and steroid use increase

the risk for bowel perforation during colonoscopy in inflammatory

bowel disease patients. J Crohns Colitis 2012;6:470-5.

4. Horsthuis K, Bipat S, Bennink RJ, Stoker J. Inflammatory bowel disease

diagnosed with US,MR, scintigraphy, and CT: meta-analysis of prospec-

tive studies. Radiology 2008;247:64-79.

5. Panes J, Bouhnik Y, ReinischW, Stoker J, Taylor SA, Baumgart DC, et al.

Imaging techniques for assessment of inflammatory bowel disease: joint

ECCO and ESGAR evidence-based consensus guidelines. J Crohns Coli-

tis 2013;7:556-85.

6. Fraquelli M, Colli A, Casazza G, Paggi S, Colucci A, Massironi S, et al.

Role of US in detection of Crohn disease: meta-analysis. Radiology

2005;236:95-101.

Ultrasonography of the Colon in Pediatric Ulcerative Colitis: A Pro

7. Di Nardo G, Aloi M, Oliva S, Civitelli F, Casciani E, Cucchiara S. Inves-

tigation of small bowel in pediatric Crohn’s disease. Inflamm Bowel Dis

2012;18:1760-76.

8. Parente F, Greco S, Molteni M, Cucino C, Maconi G, Sampietro GM,

et al. Role of early ultrasound in detecting inflammatory intestinal disor-

ders and identifying their anatomical location within the bowel. Aliment

Pharmacol Ther 2003;18:1009-16.

9. Drews BH, Barth TF, Hanle MM, Akinli AS, Mason RA, Muche R, et al.

Comparison of sonographically measured bowel wall vascularity, histol-

ogy, and disease activity in Crohn’s disease. Eur Radiol 2009;19:1379-86.

10. Haber HP, Busch A, Ziebach R, Dette S, Ruck P, Stern M. Ultrasono-

graphic findings correspond to clinical, endoscopic, and histologic find-

ings in inflammatory bowel disease and other enterocolitides.

J Ultrasound Med 2002;21:375-82.

11. Epifanio M, Baldisserotto M, Spolidoro JV, Gaiger A. Grey-scale and co-

lor Doppler sonography in the evaluation of children with suspected

bowel inflammation: correlation with colonoscopy and histological find-

ings. Clin Radiol 2008;63:968-78.

12. Ruess L, Blask AR, Bulas DI, Mohan P, Bader A, Latimer JS, et al. Inflam-

matory bowel disease in children and young adults: correlation of sono-

graphic and clinical parameters during treatment. Am J Roentgenol

2000;175:79-84.

13. Maconi G, Ardizzone S, Parente F, Bianchi Porro G. Ultrasonography in

the evaluation of extension, activity, and follow-up of ulcerative colitis.

Scand J Gastroenterol 1999;34:1103-7.

14. Antonelli E, Giuliano V, Casella G, Villanacci V, Baldini V, Baldoni M.

Ultrasonographic assessment of colonic wall in moderate-severe ulcera-

tive colitis: comparison with endoscopic findings. Dig Liver Dis 2011;43:

703-6.

15. Parente F, Molteni M, Marino B, Colli A, Ardizzone S, Greco S, et al. Are

colonoscopy and bowel ultrasound useful for assessing response to

short-term therapy and predicting disease outcome of moderate-

to-severe forms of ulcerative colitis?: a prospective study. Am J Gastro-

enterol 2010;105:1150-7.

16. Haber HP, Busch A, Ziebach R, Stern M. Bowel wall thickness measured

by ultrasound as a marker of Crohn’s disease activity in children. Lancet

2000;355:1239-40.

17. Faure C, Belarbi N, Mougenot JF, Besnard M, Hugot JP, C�ezard JP, et al.

Ultrasonographic assessment of inflammatory bowel disease in children:

comparison with ileocolonoscopy. J Pediatr 1997;130:147-51.

18. Bremner AR, Griffiths M, Argent JD, Fairhurst JJ, Beattie RM. Sono-

graphic evaluation of inflammatory bowel disease: a prospective,

blinded, comparative study. Pediatr Radiol 2006;36:947-53.

19. Levine A, Griffiths A, Markowitz J, Wilson DC, Turner D, Russell RK,

et al. Pediatric modification of the Montreal classification for inflamma-

tory bowel disease: the Paris classification. Inflamm Bowel Dis 2011;17:

1314-21.

20. Schroeder KW, Tremaine WJ, Ilstrup DM. Coated oral 5-aminosalicylic

acid therapy for mildly to moderately active ulcerative colitis. A random-

ized study. N Engl J Med 1987;317:1625-9.

21. Turner D, Otley AR, Mack D, Hyams J, de Bruijne J, Uusoue K, et al.

Development, validation, and evaluation of a pediatric ulcerative colitis

activity index: a prospective multicenter study. Gastroenterology 2007;

133:423-32.

22. Di Nardo G, Oliva S, Passariello M, Pallotta N, Civitelli F, Frediani S,

et al. Intralesional steroid injection after endoscopic balloon dilation

in pediatric Crohn’s disease with stricture: a prospective, randomized,

double-blind, controlled trial. Gastrointest Endosc 2010;72:1201-8.

23. Darge K, Anupindi S, Keener H, Rompel O. Ultrasound of the bowel in

children: how we do it. Pediatr Radiol 2010;40:528-36.

24. Alison M, Kheniche A, Azoulay R, Roche S, Sebag G, Belarbi N. Ultraso-

nography of Crohn’s disease in children. Pediatr Radiol 2007;37:1071-82.

25. Ord�as I, Rimola J, Garc�ıa-Bosch O, Rodr�ıguez S, Gallego M,

Etchevers MJ, et al. Diagnostic accuracy of magnetic resonance colonog-

raphy for the evaluation of disease activity and severity in ulcerative co-

litis: a prospective study. Gut 2013;11:1566-72.

26. Kramer SM, Feinstein AR. Clinical biostatistics LIV. The biostatistics of

concordance. Clin Pharmacol Ther 1981;29:111-23.

spective, Blind, Comparative Study with Colonoscopy 83

THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. 165, No. 1

27. Van Limbergen J, Russell RK, Drummond HE, Aldhous MC,

Round NK, Nimmo ER, et al. Definition of phenotypic characteristics

of childhood-onset inflammatory bowel disease. Gastroenterology

2008;135:1114-22.

28. Levine A, de Bie CI, Turner D, Cucchiara S, Sladek M, Murphy MS,

et al. Atypical disease phenotypes in pediatric ulcerative colitis: 5-

year analyses of the EUROKIDS Registry. Inflamm Bowel Dis 2013;

19:370-7.

29. Gower-RousseauC,Dauchet L, Vernier-Massouille G, Tilloy E, Brazier F,

Merle V, et al. The natural history of pediatric ulcerative colitis: a

population-based cohort study. Am J Gastroenterol 2009;104:2080-8.

30. Torres J, Billioud V, Sachar DB, Peyrin-Biroulet L, Colombel JF. Ulcer-

ative colitis as a progressive disease: the forgotten evidence. Inflamm

Bowel Dis 2012;18:1356-63.

31. Ordas I, Rimola J, Rodriguez S, Gallego M, Ricart E, Panes J. Imaging of

the colon in inflammatory bowel disease: ready for prime time? Curr

Drug Targets 2012;13:1252-60.

32. Turner D, Hyams J, Markowitz J, Lerer T, Mack DR, Evans J, et al.

Appraisal of the pediatric ulcerative colitis activity index (PUCAI). In-

flamm Bowel Dis 2009;15:1218-23.

84

33. Turner D, Griffiths AM, Mack D, Otley AR, Seow CH, Steinhart AH,

et al. Assessing disease activity in ulcerative colitis: patients or their phy-

sicians? Inflamm Bowel Dis 2010;16:651-6.

34. Horsthuis K, Stokkers PC, Stoker J. Detection of inflammatory bowel

disease: diagnostic performance of cross-sectional imaging modalities.

Abdom Imaging 2008;33:407-16.

35. PascuM, Roznowski AB,M€uller HP, Adler A,Wiedenmann B, Dignass AU,

et al. Clinical relevance of transabdominal ultrasonography and magnetic

resonance imaging in patientswith inflammatory bowel disease of the termi-

nal ileum and large bowel. Inflamm Bowel Dis 2004;10:373-82.

36. Fraquelli M, Sarno A, Girelli C, Laudi C, Buscarini E, Villa C, et al.

Reproducibility of bowel ultrasonography in the evaluation of Crohn’s

disease. Dig Liver Dis 2008;40:860-6.

37. Maconi G, Di Sabatino A, Ardizzone S, Greco S, Colombo E, Russo A,

et al. Prevalence and clinical significance of sonographic detection of

enlarged regional lymph nodes in Crohn’s disease. Scand J Gastroenterol

2005;40:1328-33.

38. D’Haens G, Ferrante M, Vermeire S, Baert F, Noman M, Moortgat L,

et al. Fecal calprotectin is a surrogate marker for endoscopic lesions in

inflammatory bowel disease. Inflamm Bowel Dis 2012;18:2218-24.

Civitelli et al

Table I. Patients baseline characteristics (N = 50)

Age (y), median; range 13 (2.6-18)Sex (males), n (%) 23 (46)Age (y) at diagnosis; median (range) 11 (2-16)Weight z-score (mean � SD) 0.33 � 0.91BMI z-score (mean � SD) 0.43 � 0.91Disease duration (y); median (range) 3 (0.5-5)Disease extensionE2, n (%) 22 (44)E3, n (%) 6 (12)E4, n (%) 22 (44)

PUCAI, mean � SD 40.5 � 24.5Clinical activity (PUCAI)Mild, n (%) 13 (26)Moderate, n (%) 20 (40)Severe, n (%) 17 (34)

Mayo score, median (range) 2 (1-3)Endoscopic activity (Mayo score)Mild, n (%) 11 (22)Moderate, n (%) 11 (22)Severe, n (%) 28 (56)

BMI, body mass index.

Table III. US findings in the colonic segments withdifferent endoscopic activity

US variablesNormal(n = 35)

Mild(n = 29)

Moderate(n = 40)

Severe(n = 46)

BWT (mm; mean � SD) 2 � 0.6 3 � 0.6 4.5 � 0.7 5.5 � 0.8BWT >3 mm (n, %) 1 (2.85) 25 (86.2) 40 (100) 46 (100)Vascularity (n, %) 0 3 (7%) 26 (65%) 40 (87%)Loss of haustra coli (n, %) 0 0 22 (55%) 40 (87%)Enlarged lymph nodes(n, %)

4 (11%) 7 (24%) 15 (37.5%) 27 (57%)

Loss of stratification(n, %)

0 0 8 (22%) 40 (87%)

Figure 1. Flow chart illustrating the number of recruited sub-jects (n = 60) and the number of eligible patients (n = 50) whounderwent a complete diagnostic work-up and were finallyincluded in the study.

July 2014 ORIGINAL ARTICLES

Ultrasonography of the Colon in Pediatric Ulcerative Colitis: A Prospective, Blind, Comparative Study with Colonoscopy 84.e1

Figure 4. US and endoscopic appearance of the A, left and B, transverse colon in a patient with active disease. A, Longitudinaland transverse scan of the descending colon. Thickened BW with presence of color signals and loss of haustra coli. The BWstratification is preserved. B, Transversal US scan of the transverse colon showing thickened BW, increased vascularity, andloss of the normal stratification. Note the presence of enlarged mesenteric lymph nodes (arrow).

THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. 165, No. 1

84.e2 Civitelli et al