Ultrasound Shear Wave Elastography and Contrast Enhancement ...

ORIGINAL ARTICLE

Ultrasound Shear Wave Elastography and Contrast Enhancement: Effective Biomarkers in Crohn’s Disease Strictures Cathy Lu, MD, MSc,* Xianyong Gui, MD,† Wenqian Chen, MD,† Tak Fung, PhD,‡ Kerri Novak, MD, MSc,* and Stephanie R. Wilson, MD* ,§

Background: Inflammation, fibrosis, and muscular hypertrophy lead to thickened bowel in Crohn’s disease forming strictures. Ultrasound shear wave elastography (SWE) measures bowel wall stiffness. Contrast-enhanced ultrasound (CEUS) uniquely detects bowel wall inflammation. We aim to correlate SWE of ileal Crohn’s disease in vivo to CEUS peak enhancement and pathology grades of inflammation, fibrosis, and muscular hypertrophy.

Methods: In a prospective institutional review board–approved study, 105 consecutive ileal patients with Crohn’s disease received ultrasound. At maximal bowel wall thickness (.4 mm), SWE and CEUS were performed. Fifteen patients had ileal resection within a mean time interval of 71.0 6 66.9 days. Pathology scores for inflammation, fibrosis, and muscular hypertrophy were compared with SWE and CEUS measurements. Results: Mean in vivo SWE velocity for patients with and without surgery was 2.8 6 0.7 and 2.2 6 0.8 m/s (P , 0.01), respectively. In all ileal

specimens, chronic exceeded active inflammatory change (P , 0.001). There was an inverse relationship between CEUS peak enhancement and both fibrosis, r ¼ 20.59, P ¼ 0.02, and SWE velocity measurements, r ¼ 20.61, P ¼ 0.03. Strictured bowel specimens had more smooth muscle hypertrophy than fibrosis, P , 0.001. There was moderate correlation between SWE and muscular hypertrophy, r ¼ 0.59, P ¼ 0.02 and no significant relationship between SWE and fibrosis scores (P . 0.05).

Conclusions: Stiffer bowel from smooth muscle hypertrophy increases SWE measurements. We report a novel relation of high SWE with muscle hypertrophy, and inverse relationship with CEUS peak enhancement; providing differentiation between active and chronic bowel wall inflammation to improve selection between medical therapy and surgery. (Inflamm Bowel Dis 2017;23:421–430) Key Words: Crohn’s disease, stricture, ultrasonography, elastography

C

rohn’s Disease (CD) is a chronic inflammatory bowel disorder mainly affecting the terminal ileum. A stricturing phenotype is characterized by inflammation, fibrosis, and muscular hypertrophy, although bowel fibrosis has been the major focus of previous imaging studies and animal models of CD.1–3 Strictured bowel in CD occurs unpredictably, found in 20% of patients at diagnosis and occurring in 60% within 10 years of disease onset,4 half inevitably requiring surgery.5 Fibrotic bowel is thought to be generally more unresponsive to drug therapies and often requires surgical intervention.6 Thus, differentiating between patients who Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.ibdjournal.org). Received for publication November 21, 2016; Accepted December 3, 2016. From the *Division of Gastroenterology, Department of Medicine, University of Calgary, Alberta, Canada; †Department of Pathology and Laboratory Medicine, University of Calgary, Alberta, Canada; ‡Department of Mathematics and Statistics, Information Technologies, University of Calgary, Alberta, Canada; and §Department of Diagnostic Imaging, University of Calgary, Alberta, Canada. The authors have no conflict of interest to disclose. Address correspondence to: Stephanie R. Wilson, MD, Department of Diagnostic Imaging, Foothills Medical Centre, University of Calgary, Alberta T2N2T9, Canada (e-mail: [email protected]). Copyright © 2017 Crohn’s & Colitis Foundation of America, Inc. DOI 10.1097/MIB.0000000000001020 Published online 26 January 2017.

Inflamm Bowel Dis Volume 23, Number 3, March 2017

have a predominantly inflammatory from a mainly fibrotic component of their stricture is imperative to improve selection between medical therapy and surgery. Radiological imaging with computed tomography (CT), magnetic resonance imaging, and bowel ultrasound (US) may be important in evaluating such disease behavior. Bowel US has gained recognition as an accurate, costeffective, and easily repeatable technique for diagnosing and assessing CD.7,8 US has been shown in reviews and metaanalyses to be equally sensitive and specific to CT and magnetic resonance imaging in diagnosing and monitoring CD.9–11 With all 3 of these modalities, both increased bowel wall thickness (BWT) and increased blood flow detected as wall enhancement on CT or magnetic resonance imaging and mural vascularization on US with color Doppler imaging (CDI) are important components of inflammation.12–14 Both BWT and grade of CDI have been shown to be significantly correlated with severity grade at endoscopy.15 In addition to using BWT and CDI to analyze disease activity, the quantity of inflammatory fat16 and wall layer preservation are also indicators of activity on US.14 These gray scale parameters allow us to generate an overall assessment of disease activity (Ultrasound Global Assessment [USGA] of Disease Activity, Appendix 1, Supplemental Digital Content, http://links.lww.com/IBD/B450).17 Easily detected on US, strictures of bowel with CD appear as thickened segments with a fixed narrowed lumen often with www.ibdjournal.org |

Copyright © 2017 Crohn’s & Colitis Foundation of America, Inc. Unauthorized reproduction of this article is prohibited.

421


Lu et al

prestenotic dilatation and dysfunctional peristalsis. Characterization of strictures with imaging requires quantification of the inflammatory and fibrotic components. Identifying fibrosis on CT enterography (CTE) in correlation with pathological specimens has been described by Chiorean et al (2007).18 However, Adler et al has cautioned against the use of CTE in predicting the presence of small bowel fibrosis that is lacking concurrent inflammation. In other words, strictures considered “inactive” with low or absent inflammation and presumed to have chronic changes such as fibrosis were surprisingly associated with less histological fibrosis.19 Furthermore, magnetic resonance enterography (MRE) has also been used to correlate inflammation and fibrosis to resected bowel specimens based on factors such as the presence of stenosis, intramural T2 signal hyperintensity, and mucosal enhancement.1 The percentage of gadolinium enhancement gain measured over a period of elapsed time is described by Rimola et al1 to successfully identify areas with a high amount of fibrosis, irrespective of the amount of concurrent inflammation. Although CTE and MRE are thought to be similar in their accuracies for detecting active inflammation, MRE has been described to be superior to CTE in illustrating fibrosis.20 In addition, the potential role of US for characterizing bowel strictures is of increasing interest and focuses on assessment of stiffness and inflammation, essentially performed on animal models to date. Shear wave elastography (SWE) uses acoustic radiation force impulse technology to assess elastic properties of tissue through an US force that propagates a shear wave through tissue. Both SWE and strain elastography have been used in rodent models with considerable success.21–23 Furthermore, SWE has been used to assess fibrosis in resected CD segments.21 Inflammatory assessments on US are traditionally based on subjective assessment of CDI recognizing that neoangiogenesis is an important component of CD bowel wall inflammation confirmed on histology.24,25 Microcirculation of the bowel wall is a valuable parameter to evaluate and has been well correlated with clinical26 and endoscopic severity.15,26 Quantification of bowel wall microvascular and macrovascular perfusion can be uniquely detected by contrast-enhanced ultrasound (CEUS).15,27,28 Contrast enhancement provides both a qualitative and objective quantitative measurement of inflammatory activity.29 This prospective study aims to demonstrate the positive contribution of SWE and CEUS to conventional US of fibrostenotic ileal CD in vivo. To our knowledge, no studies have analyzed CEUS combined with SWE in CD, and only 2 studies have analyzed fibrosis using elastography in vivo.30,31 We predict that patients with chronic stricture will have a higher SWE score with greater bowel stiffness attributed to more fibrosis, and less inflammation than those with inflammatory bowel wall thickening. More specifically, this study aims to correlate SWE to CEUS peak enhancement (PE) as a measurement of inflammation, and to pathology grades of inflammation, fibrosis, and muscular hypertrophy of resected strictured small bowel.

422

MATERIALS AND METHODS This study followed a prospective design. Between March and November 2014, 105 consecutive adult patients assessed by gastroenterologists in the University of Calgary IBD centre who were referred for a routine bowel US with CEUS were enrolled in the study to have additionally SWE. Inclusion criteria included ileal or ileocolonic CD, an ileal BWT of greater than 4 mm as thinner bowel is unsuitable for both SWE and CEUS, and previous microscopically confirmed CD of the terminal ileum. Disease behavior was characterized according to the Montreal Classification for CD where B1 is nonstricturing and nonpenetrating, B2 is a stricturing disease, and B3 is a penetrating phenotype.32 Patients were prospectively followed for a minimum of 8 months and up to one year from time of their sonography to identify those with medical management and those with surgical resection. The patient’s medication and surgical history were obtained at the time of US appointment. The clinical characteristics of the patients are displayed in Table 1. Exclusion criteria included patients #18 years old, those with inadequate gray scale examination, failed CEUS, or interquartile range (IQR) /median .0.30 for SWE (n ¼ 10). CEUS was not performed if any of the following contraindications to use of Definity (Lantheus Medical Imaging, Billerica, MA) contrast were present: pregnancy, known or suspected cardiac shunts, acute or chronic heart failure, hypersensitivity to perflutren or Definity, and chronic obstructive pulmonary disease.

Sonography As part of our routine protocol, all patients fasted for 6 hours before US to reduce luminal content and bowel blood flow to minimize false readings. Four experienced sonographers and a radiologist (S.W.) with more than 35 years of experience with IBD US were aware of CD diagnosis and surgical history at the time of US. US was performed using a multifrequency convex C9-2 (2–9 MHz) transducer with a Philips Epiq 5 machine, C9-4 (4–9 MHz) transducer with a Philips IU-22 (Philips Healthcare, Bothell, WA) machine, or a C8-3 (3–8 MHz) transducer with Acuson S3000, (Siemens Medical Solutions, Malvern, PA) where all bowel segments were assessed in long and short axes. Abnormal bowel loops on gray scale US were identified and captured as images to show BWT, length of affected bowel, the presence of inflammatory fat, and the amount of color signal on Doppler imaging. Grading of these gray scale parameters for CD activity used an USGA (Appendix 1, Supplemental Digital Content, http://links.lww.com/IBD/B450). The retention of bowel wall layering or loss of bowel stratification was also documented. Stricture was defined as a thickened segment with a fixed narrowed lumen with prestenotic dilatation with or without dysfunctional peristalsis.

Point SWE Point SWE measures the velocity of a generated shear wave under direct visualization. SWE readings were collected in the supine position using acoustic radiation force impulse technology

| www.ibdjournal.org



Bowel Ultrasound Elastography and Contrast

TABLE 1. Clinical Characteristics of CD Patients with and Without Surgery Postsonography Characteristic

Surgical Resection, n ¼ 15

No Surgery, n ¼ 80

Men (%) Age at study (mean 6 SD), yr Age at diagnosis (mean 6 SD), yr Disease duration (mean 6 SD), yr

5 (33.3) 41.0 6 14.4 29.5 6 14.0 12.7 6 13.9

40 (50.0) 46.1 6 16.5 32.3 6 16.7 14.8 6 13.7

t

Degrees of Freedom

P

— 21.1 20.6 20.6

— 93 93 93

0.3a 0.3b 0.4b 0.8b Fisher’s Exact Test

Medication at the time of US, n (%) 5-aminosalicylic acid Immunomodulatorc Biological agent Biological agent and immunomodulator Corticosteroid Disease behavior at sonography, n (%) Inflammatory Stricturing disease Penetrating disease Indication for surgery, n (%) IMBO Complete obstruction Fistulizing disease

1 (6.7) 5 (33.3) 7 (46.7) 1 (6.7) 5 (33.3)

5 17 31 8 12

(6.3) (21.3) (38.8) (10) (15)

1.0 0.2 0.6 0.6 0.1

0 12 (80) 3 (20)

12 (15) 57 (71.3) 11 (13.8)

0.3

10 5 2d

— — —

—

a

Fisher’s Exact Test. Independent t test. c Includes 6-mercaptopurine, azathioprine, or methotrexate. d Also had evidence of obstruction. b

with Virtual Touch Quantification (VTQ); Acuson S3000, Siemens Medical Solutions USA, Inc. or ElastPQ; Philips Epiq 5 (Philips Healthcare). An average of 10 SWE readings were obtained at the point of maximal BWT in the ileum and/or at the

point of maximal stricture, also known as region of interest (ROI) (Fig. 1). SWE cannot be reliably performed on normal bowel or bowel with a thickness that is insufficiently large enough (,4 mm). The ROI is 10 mm by 5 mm in size and was placed on

FIGURE 1. Forty-seven-year-old woman with fibrostenotic CD and symptoms of bloating and nausea. A, Longitudinal view of very strictured terminal ileum with luminal apposition and a thin white horizontal line representing air in the bowel lumen on the left side of image, and resulting prestenotic dilatation on the right side of photograph. B, Virtual Touch Quantification SWE measurement of bowel stricture in long axis. ROI box is placed on bowel wall with a high shear wave speed of 3.73 m/s, suggesting disease chronicity. www.ibdjournal.org |


423

Lu et al

the bowel segment in axial or longitudinal view, with exclusion of surrounding tissues and luminal content. In the axial view, the ROI is placed at the 3 o’clock or 9 o’clock position to incorporate as much bowel wall as possible for analysis. In the long axis view, the positioning of the ROI to incorporate only bowel wall is easier and less stringent. This location of interest was photographically documented and recorded as distance from the ileocecal valve or anastomosis, in those with previous surgery, for comparison to pathology grades. After administration of the acoustic radiation force impulse technology pulse, the shear wave speed was reported in meter per second. The IQR was used to assess quality of the SWE data. The IQR measures the statistical dispersion between the upper and lower quartiles. An IQR/median ,0.30 suggests that a data set is good.33 Therefore, an additional exclusion criterion is an IQR/median $0.30. Of the 95 patients, 80 SWE measurements were collected using VTQ and 15 SWE were obtained from ElastPQ. Of the 15 patients who went for surgery, 14 had VTQ measurements. For our population, the average IQR/median was 0.23. Of our consecutive 105 patients, 6 patients had very reliable IQR ratios (#0.10), 89 had reliable (0.10 , IQR/median # 0.30), and 10 were excluded (IQR/median .0.30); 1 of whom had subsequent surgery.34

CEUS Quantification of bowel wall vascular perfusion with microbubble contrast agents is useful for activity determination on US. All patients who received IV contrast had PE measurements made on the Philips Epiq 5 or Philips IU-22 with consistent acquisition settings. CEUS was performed as closely as possible to the location where the SWE measurements were obtained, which was at an area of maximum ileal BWT and/or maximal stricture to assess for hypervascularization of the bowel wall. CEUS was performed using Definity, a perflutren lipid microsphere injectable suspension (Lantheus Medical Imaging). Definity is used off-label for evaluation of IBD activity with a letter of no-objection from Health Canada for research in this area. Hyoscine butylbromide was used in 5 cases where the bowel loop showed excess peristalsis. On identification of the point of interest, 0.4 mL of Definity was injected intravenously through an 18-gauge catheter in the antecubital vein followed by a 5 mL normal saline flush. The completion of the normal saline flush signified time zero and the enhancement subsequently monitored in real time without any motion adjustment to the transducer and scan plane, and stored as a 2-minute video file. All CEUS imaging was saved as video and DICOM (Digital Imaging and Communications in Medicine) format, the latter analyzed using Q-Lab quantification software version 9.0 (Philips Healthcare). Time-enhancement intensity linear and logarithmic curves, maximum PE values measured in decibels (dB), time to PE (seconds), and area under the curve values in arbitrary intensity units (AIUs) were obtained.

Pathology Assessment A most representative full-thickness cross section of the strictured bowel was retrieved for histopathological analysis. Two

424


gastrointestinal pathologists (X.G., W.C.) blinded to US findings, scored all specimens for active and chronic inflammation, fibrosis, and muscular hypertrophy. After grading of all of the slides, Masson’s trichrome stain (Artisan Link; DAKO, Carpinteria, CA) was used in selected cases to confirm correlation of grading of fibrosis and collagen deposition with the original hematoxylin and eosin interpretation (Fig. 2). Histological grading scheme is available in Appendix 2 (Supplemental Digital Content, http://links. lww.com/IBD/B450).

Statistical Analyses Statistical analyses were performed using IBM SPSS (The Statistical Package for the Social Sciences) for Windows version 22 software (IBM, Armonk, NY). Descriptive statistics were used to determine the distribution of demographic and disease (Montreal Classification) characteristics among the patient populations. Between groups, unpaired t tests for continuous variables and the Fisher’s exact test for categorical variables were performed. With our small sample size, nonparametric tests such as Mann–Whitney U tests, and Wilcoxon signed-rank tests were also conducted for comparison with results of parametric tests for verification purposes. Analysis of variance was used to compare patients with high and low PE, and differences of mean SWE with histological grades. Correlations of elastography, PE, inflammation, fibrosis, and muscle hypertrophy were determined by either Pearson correlation for continuous variables or Spearman rank correlation for ordinal variables. A 2-sided significance level of a ¼ 0.05 was used. Simultaneous and stepwise logistic regression was performed to identify which variables could be used in our model to significantly predict surgery.

ETHICAL CONSIDERATIONS This study was reviewed and approved by the Conjoint Research Ethics Board at the University of Calgary. Each subject provided signed informed consent.

RESULTS All 105 patients had SWE and CEUS; 10 patients were excluded (IQR/median .0.30). Of the 95 patients meeting inclusion criteria, 80 subsequently had medical management and 15 patients had resection allowing for histologic analysis. The mean time interval between US and surgical resection was 71.0 days (SD 66.9 d, range 1–230 d). The mean age of patients without surgery was 46.1 years (SD 16.5), whereas it was 41.0 years (SD 14.4), P ¼ 0.3 in those with surgery. In the surgical group, 3 of the 15 (20%) patients had previous surgery. Of the nonsurgical group, 36 of the 80 patients (45%) had previous surgery. There was no significant difference in age at diagnosis, sex, disease duration, behavior, and medications at the time of sonography between the surgical and nonsurgical groups (Table 1). Stricturing phenotype was dominant for the entire population of 95 patients. Fifteen patients received ileal resection for stricture with incomplete mechanical bowel obstruction (IMBO)





FIGURE 2. Resected terminal ileum with fibrostenosis (·20 magnification). A, Hematoxylin and eosin (H&E) stain shows significant smooth muscular hypertrophy with volume expansion of the circular and longitudinal layers of the MP (lower aspect of the micrographs), which shows the predominant change. Within the SM, fibrosis (loose and haphazardly arranged collagen bundles and fibroblastic proliferation) and adipose tissue hyperplasia are prominent. The fibrous tissue also partly extends into the MP. B, On Masson’s trichrome stain, the fibrosis component is highlighted (in blue), and its distribution and extent are identical to that seen on H&E stain (left). MP, muscularis propria; SM, submucosa.

with or without fistulizing disease. Of these patients receiving surgery, 8 (53.3%) procedures were elective, whereas 7 procedures (46.7%) were urgent. Five of the 15 patients were on corticosteroids at the time of sonography and surgery. Two patients who had not been on previous corticosteroids received the intravenous form during their hospital admission before surgery.

and IMBO were present in 100% (15/15) of patients, whereas 52.5% (42/80) and 47.5% (38/80) of the nonsurgical patients had strictures and IMBO, respectively. CTE or MRE was also performed in 12 of the 15 patients. Of these 12 scans, 10 of the patients also had evidence of IMBO. In all 12 CTE or MRE scans when compared with US, concordant areas of identified bowel wall abnormalities consistent with stricture were present.

Gray Scale US Findings and CEUS

Shear Wave Enhancement and Histological Grading

The terminal ileum was identified in all patients. The mean values BWT for the surgical and nonsurgical groups were 8.8 6 2.6 and 7.8 6 1.9 mm (P . 0.05), respectively, both consistent with moderate to severe disease. Inflammatory fat, color Doppler signal, wall layer preservation did not differ between both groups (Table 2). However, of the 15 surgical patients, 5 (33.3%) had complete loss of wall stratification appearing completely black on US (Fig. 3). Furthermore, when assessing PE, there was no statistical significant difference between patients with and without surgery after sonography. However, for the nonsurgical group, there was a fair correlation between the severity of PE and color Doppler signal, r ¼ 0.3, P ¼ 0.01, and a good correlation in those who had surgery postsonography, r ¼ 0.7, P ¼ 0.1. Regarding CD complications as predicted on sonography, the only significant difference between the surgery and nonsurgery patients was that the latter group of patients had fewer strictures and IMBO, P , 0.001. In the surgery group, strictures

Mean in vivo SWE measurements for all patients who had and did not have surgery were 2.8 6 0.7 m/s (range 1.5–3.9 m/s) and 2.2 6 0.8 m/s (range 0.64–4.1 m/s) (P , 0.01), respectively (Table 3). With histological assessment, there was a moderate correlation between SWE and muscular hypertrophy, r ¼ 0.59, P ¼ 0.02. All other correlations between inflammation, fibrosis, and SWE were not significant.

CEUS PE and Histological Grading Mean CEUS measurements PE for all nonsurgical and surgical patients were 22.4 6 4.6 and 22.9 6 3.0 dB (P . 0.05) (Table 3), respectively, suggesting moderate inflammatory activity.4 PE was negatively correlated with fibrosis (r ¼ 20.59, P ¼ 0.02), and there was a trend toward an inverse relationship with muscular hypertrophy (r ¼ 20.49, P ¼ 0.06). PE had a fair correlation with chronic inflammation histological scores (r ¼ 0.6, P ¼ 0.03), and not with active inflammation scores, P . 0.05. www.ibdjournal.org |


425


Lu et al

TABLE 2. Comparison of Gray Scale US Parameters and Complications Surgical Resection, n ¼ 15 BWT, millimetres (mean 6 SD) Inflammatory fat score, No. of patients (%) None Mild Moderate Severe Color Doppler score, No. of patients None Mild Moderate Severe Wall layer preservation present Strictures present IMBO Fistulas present

No Surgery, n ¼ 80

Pa

8.8 6 2.6

7.8 6 1.9

0.6b

0 0 6 (40) 9 (60)

2 (2.5) 25 (31.3) 35 (43.8) 17 (21.3)

0.5 ,0.01c 0.8 0.003

1 (6.7) 5 (33.3) 4 (26.7) 5 (33.3) 10 (66.7) 15 (100) 15 (100) 2 (13.3)

7 (8.8) 29 (36.3) 20 (25.0) 18 (22.5) 72 (90) 42 (52.5) 38 (47.5) 7 (8.8)

0.8 0.8 0.9 0.4 0.6 ,0.01c ,0.01c 0.7

Histological Grading of Surgical Specimens Fifteen ileal specimens were graded histologically (Table 4). In all cases, inflammation was present. Overall, the inflammation of the strictured bowel showed a higher degree of chronic than active inflammatory change, t(14) ¼ 211.5, P , 0.001. However, active and chronic inflammation were moderately correlated, r ¼ 0.67, P , 0.01. Although fibrosis and muscular hypertrophy were significant observations in all cases and were moderately correlated (r ¼ 0.58, P ¼ 0.02), the quantity of muscular hypertrophy was significantly greater than the amount of fibrosis in resected bowel specimens, t(14) ¼ 217.2, P , 0.001. Space volume expansion of the muscularis propria layer associated with muscular hypertrophy was significantly greater than the expansion for both the mucosa and the submucosa layers (P , 0.001) (Fig. 4). All other correlations between inflammation with fibrosis and muscle hypertrophy were poor (Table 5).

Simultaneous and Stepwise Logistic Regression Analysis Using a simultaneous and stepwise regression analysis, no variables that were chosen in the model (sex, age at study, age at diagnosis, disease duration, previous resection status, medication at the time of US, disease behavior, gray scale US variables [BWT, CDI, and inflammatory fat], SWE, and histological scores) significantly predicted surgery.

a

Fisher’s exact test. Independent samples t test. c Significant at 0.05 level. b

Shear Wave and CEUS PE Of the patients with surgical resection, SWE was moderately inversely correlated with PE, r ¼ 20.61, P ¼ 0.02, whereas there was no significant correlation for those without resection, r ¼ 20.07, P . 0.05.

DISCUSSION The process of inflammation and fibrosis in intestinal strictures in CD is closely intertwined. Chronic inflammation mediates multiple factors to alter tissue architecture leading to cell proliferation and stricture formation in CD.35 The relative composition of damaged segments of bowel is central to decisions for management. US is gaining increasing acceptance36 and in particular, CEUS and SWE are innovative and objective biomarkers

FIGURE 3. Fifty-year-old man with fibrostenotic CD for 30 years. A, Axial gray scale image of thickened terminal ileum measuring 1.2 cm with loss of all wall layer stratification, narrowed lumen with mucosal apposition, and with surrounding inflammatory fat. B, Longitudinal gray scale image of strictured terminal ileum with prestenotic dilation on the right side of photograph.

426





TABLE 3. SWE and PE Values for Patient Population Characteristic SWE (m/s 6 SD) PE

Surgical Resection, n ¼ 15

No Surgery, n ¼ 80

P

2.8 6 0.7 22.9 6 3.0

2.2 6 0.8 22.4 6 4.6

,0.01a 0.7

a

Significant at 0.05 level.

that can be successfully used to impact selection of medical and surgical therapies for CD. Our pretest belief was that active inflammatory change and fibrosis are major components of strictures in CD, whereas our study suggests that although inflammatory change is present in all strictures, chronic inflammation, reflective of long-standing disease, dominates active changes in those requiring resection (P , 0.001). However, the presence of both active and chronic inflammation occurred concurrently, r ¼ 0.67, P , 0.01. To our surprise, high CEUS values are associated with histologic chronic inflammation rather than acute inflammatory change in this population. Furthermore, fibrosis, although present in surgical specimens, is overshadowed by marked muscular hypertrophy as the dominant explanation for bowel thickening and stricture formation. In this population, PE had a statistically significant correlation with (1) histological chronic and not active

inflammatory changes (r ¼ 0.6, P ¼ 0.03), and with (2) CDI for both the surgical (r ¼ 0.7, P ¼ 0.01) and nonsurgical groups (r ¼ 0.3, P ¼ 0.01). The latter finding is expected as contrast enhancement is a quantitative measurement of uptake of blood vessels.28 Currently, there is no imaging modality other than US that is capable of capturing both inflammation and the stiffness of bowel. CEUS accurately and reliably quantifies inflammation in bowel,29,37 and SWE provides numeric quantification of its stiffness.21 CEUS has successfully been used in differentiating inflammatory from fibrotic ileal CD strictures in a small study of 28 patients, where inflammatory ileal strictures had a significantly higher maximal contrast enhancement than fibrotic strictures.37 A limitation of this study was that fibrosis was identified as abnormal collagen deposits in the edges of mucosal ulceration from endoscopic biopsies, rather than resected gross specimens. Nonetheless, the value of small bowel CEUS to easily follow-up and differentiate active CD from mural fibrosis was emphasized.37 Moreover, our results show fibrosis was not a major component of bowel wall thickening. Rather interestingly, muscular hypertrophy was the primary component of strictured bowel specimens in our study. This was a surprising observation as to date, the imaging literature on CT and magnetic resonance has focused on fibrosis as a major factor in stricture formation with much less emphasis on muscular hypertrophy.1,12,19 However, in our study, fibrosis did moderately correlate with muscular hypertrophy in the resected bowel specimens (r ¼ 0.58, P ¼

TABLE 4. Details of Patients with CD and Histology Scores

Patient 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Age at Diagnosis

Age at Surgery

21 38 18 37 20 20 19 28 14 62 43 21 18 34 50

29 38 46 59 50 20 62 43 16 62 43 41 25 35 50

CD Sex Duration F F F M M F F M F M F M F F F

8 0 28 22 30 0 43 15 2 0 0 20 7 1 0

CD Locationa

CD Behaviorb

Active Inflammation Score (%)c

Chronic Inflammation Score (%)c

Fibrosis Score (%)c

Muscular Hypertrophy Score (%)c

1 1 1 1 3 1 3 1 1 1 1 1 1 1 1

2 2 2 2 2 3 2 2 2 3 2 3 2 2 2

27.2 12.1 21.2 15.2 21.2 36.4 12.1 39.4 3.0 21.2 9.1 27.3 33.3 15.2 30.3

53.3 50.0 36.7 43.3 63.3 73.3 36.7 66.7 20 63.3 43.3 50.0 50.0 70.0 63.3

25.0 16.7 25.0 25.0 33.3 16.7 16.7 33.3 33.3 25.0 25.0 25.0 16.7 16.7 16.7

83.2 88.9 83.3 100.0 94.4 83.3 61.1 100 77.8 77.8 66.7 77.8 83.3 83.3 88.9

a

Montreal Classification location: 1, ileal; 2, colonic; 3, ileocolonic. Montreal Classification behavior: 1, nonstricturing, nonpenetrating; 2, stricturing; 3, penetrating. c Mild score is ,33%. Moderate score is from 33% to 66%. Severe score is .66%. b

www.ibdjournal.org |


427


Lu et al

FIGURE 4. A, Normal terminal ileum with hematoxylin and eosin (H&E) stain (·50 magnification). B, Resected terminal ileum with fibrostenotic behavior; H&E stain of ileal mucosa with profuse smooth muscular hypertrophy of MP (severe grade of volume expansion), and smooth muscle proliferation with replacement of normal loose connective tissue in submucosa. Minimal fibrosis present (·20 magnification). MP, muscularis propria; SM, submucosa.

0.02). In addition, there was no significant relationship between fibrosis and SWE. This is likely due to our pathology grading scheme separating fibrosis from muscular hypertrophy, whereas other studies that have reported a relationship between elastography or disease activity and fibrosis used pathology classifications that do not include muscular hypertrophy21,22 or incorporate muscular hypertrophy into their overall fibrosis score.1 Regarding PE and fibrosis, histological analysis of the 15 ileal specimens showed that those with lower contrast PEs had significantly more fibrosis (r ¼ 20.59, P ¼ 0.02). Conversely and as expected, patients who had higher PEs had less fibrosis. Although PE was significantly negatively correlated with fibrosis, there was also a trend toward an inverse relationship with muscular hypertrophy (r ¼ 20.49, P ¼ 0.06).

Investigations originally using elastography to study bowel stiffness have used rat models of colitis23 and ex vivo human CD bowel specimens.21 Stidham et al23 had concluded that US elasticity imaging can distinguish inflammatory from fibrotic intestine in rat models of colitis, and between fibrotic and unaffected bowel in humans with CD. Baumgart et al30 used strain elastography to assess strictured bowel with CD and concluded that fibrosis includes changes related to an increased width of muscularis propria and muscularis mucosa in strictured CD segments when compared with unaffected portions. There are limitations to strain elastography, which measures the strain of the tissue from external compression as compared to the surrounding reference tissue. Our experience suggests that this technique has limitations to practical application of bowel. By

TABLE 5. Pearson Correlation of SWE Measurements with Inflammation, Fibrosis, and Muscle Hypertrophy of Surgically Resected Bowel Specimens

SWE Active inflammation (P) Chronic inflammation (P) Fibrosis (P)

Active Inflammation (P)

Chronic Inflammation (P)

Fibrosis (P)

20.02 (0.96) — — —

0.13 (0.66) 0.67 (,0.01)a — —

0.38 (0.16) 0.01 (0.97) 20.13 (0.65) —

Muscle Hypertrophy (P) 0.59 0.20 20.06 0.58

(0.02)a (0.48) (0.85) (0.02)a

a

Significant at 0.05 level.

428




comparison, SWE is a reproducible, objective, quantitative technique for measuring organ stiffness and does not have the same inherent limitations as strain elastography.38,39 From our knowledge, the evaluation of SWE of in vivo bowel is in the preliminary phases of investigation. Our study has a few limitations. Consecutive patients referred for bowel US who met inclusion criteria were included in our study. The University of Calgary currently has the largest integrated IBD and US unit in North America, performing over one thousand scans each year. However, as our center is a quaternary referral center, there is a possibility of applicability to the general IBD population and referral bias although all IBD physicians at our center refer patients to the US unit. Second, one-third of patients who had surgery were on steroids at the time of sonography, and an additional 2 patients received IV corticosteroids before surgery. Although these medications may alter active inflammation, these patients were not excluded as it was felt that their histological changes from long-standing disease would not be greatly altered. In addition, we only followed patients for up to a period of one year from sonographic evaluation. This may decrease the number of patients who went for surgery, limiting comparisons with the nonsurgical group. In addition, differences between the surgical and nonsurgical groups may be confounded as patients of similar disease severity may exist in the latter group, as surgical resection is declined. Thus, this could explain an overlap in the range of SWE values between both groups. In our study, the time from sonography to surgery ranged from 1 to 230 days. However, it was felt that chronic changes in long-standing CD will not necessarily change over a several-month interval. Furthermore, although we measured SWE and contrast parameters at the point of maximal BWT and assessed this corresponding point for histological changes, an exact overlap of the point of interest cannot be possible after formalin fixation. The histologic evaluation of the bowel specimens was from the expert opinion of 2 gastrointestinal pathology specialists whose method is not yet validated.40 At the location of SWE measurement, the ROI box (10 by 5 mm) is not adjustable with the US systems we used and ideally should only incorporate the bowel wall and not lumen. We appreciate that this is a limitation if the bowel wall is less than 1 cm thick and could incorporate content other than bowel wall. Another limitation of our study was that SWE was measured on 2 different US systems, ElastPQ and Virtual Touch Quantification. Different systems may have dissimilar SWE estimates. However, we do not expect that this would have a large impact on our assessments as SWE measurements using VTQ were collected in 84.2% (80/95) of patients, and in all patients who went for resection except for one. Inter- and intraobserver variability is a limitation, and future studies will need to address the reproducibility of SWE and CEUS measurements between US operators. However, the reproducibility of SWE for assessing liver fibrosis has been demonstrated to be high with intraobserver intraclass correlation coefficients of 0.95 and 0.93 for an expert and a novice sonographer, respectively, with an interobserver


agreement of 0.88.41 Likewise, the reproducibility of CEUS has been described as calculations of PE are performed by software analysis.42 The reproducibility of US in diagnosing and following CD has also been well shown.43 However, our elastography parameters have not been validated which would also affect the generalizability of our results across centers. There is currently a scarcity of published data with CEUS and SWE parameters and we need to validate these US measurement indices. Last, a further limitation includes a small sample size. However, our sample size exceeds or is comparable with previously published studies.21,30 In conclusion, identifying the combination of inflammation and muscle hypertrophy within strictures in CD using CEUS and SWE has yet to be described. SWE measurements of in vivo small bowel CD increase when there is stiffer bowel, which is attributed to fibrosis and muscular hypertrophy. In this study, the dominant cause of increased BWT in strictures was muscular hypertrophy. Furthermore, SWE is a novel surrogate for assessing inflammation where there is a compelling relation of higher SWE with bowel stiffness and muscular hypertrophy, and an inverse relationship with CEUS PE. Chronic inflammation dominated active changes in those requiring resection. Inflammation, fibrosis, and muscular hypertrophy are not mutually exclusive of each other in the bowel wall of CD; their relative contribution impacts selection of medical or surgical therapy. Results from this study are promising and prompt us to perform further investigations to assess whether SWE and CEUS parameters improve over time with medical therapy or conversely, whether lack of improvement is predictive of those requiring surgery. Second, we would like to determine an SWE threshold above which patients would more likely benefit from surgical intervention. We anticipate that CEUS and SWE parameters of more stiff bowel will not respond to therapy as readily as bowel with less chronic changes. Overall, bowel US and more specifically, CEUS and SWE are safe, noninvasive, objective biomarkers that we predict will play an increasing role in management of strictures in the future.

ACKNOWLEDGMENTS We acknowledge Siemens AG for ultrasound equipment support.

REFERENCES 1. Rimola J, Planell N, Rodriguez S, et al. Characterization of inflammation and fibrosis in Crohn’s disease lesions by magnetic resonance imaging. Am J Gastroenterol. 2015;110:432–440. 2. Makanyanga J, Punwani S, Taylor SA. Assessment of wall inflammation and fibrosis in Crohn’s disease: value of T1-weighted gadoliniumenhanced MR imaging. Abdom Imaging. 2012;37:933–943. 3. Li C, Iness A, Yoon J, et al. Noncanonical STAT3 activation regulates excess TGF-beta1 and collagen I expression in muscle of stricturing Crohn’s disease. J Immunol. 2015;194:3422–3431. 4. Lovasz BD, Lakatos L, Horvath A, et al. Evolution of disease phenotype in adult and pediatric onset Crohn’s disease in a population-based cohort. World J Gastroenterol. 2013;19:2217–2226. 5. Peyrin-Biroulet L, Loftus EV, Colombel JF, et al. The natural history of adult Crohn’s disease in population-based cohorts. Am J Gastroenterol. 2009;105:289–297. www.ibdjournal.org |


429

Lu et al

6. Spinelli A, Correale C, Szabo H, et al. Intestinal fibrosis in Crohn’s disease: medical treatment or surgery? Curr Drug Targets. 2010;11:242–248. 7. Haber HP, Busch A, Ziebach R, et al. Ultrasonographic findings correspond to clinical, endoscopic, and histologic findings in inflammatory bowel disease and other enterocolitides. J Ultrasound Med. 2002;21:375–382. 8. Maconi G, Radice E, Greco S, Bianchi Porro G. Bowel ultrasound in Crohn’s disease. Best Pract Res Clin Gastroenterol. 2006;20:93–112. 9. Allocca M, Fiorino G, Danese S. Cross-sectional imaging modalities in Crohn’s disease. Dig Dis. 2013;31:199–201. 10. Horsthuis K, Bipat S, Bennink RJ, et al. Inflammatory bowel disease diagnosed with US, MR, scintigraphy, and CT: meta-analysis of prospective studies. Radiology. 2008;247:64–79. 11. Panés J, Bouzas R, Chaparro M, et al. Systematic review: the use of ultrasonography, computed tomography and magnetic resonance imaging for the diagnosis, assessment of activity and abdominal complications of Crohn’s disease. Aliment Pharmacol Ther. 2011;34:125–145. 12. Wu YW, Tang YH, Hao NX, et al. Crohn’s disease: CT enterography manifestations before and after treatment. Eur J Radiol. 2012;81:52–59. 13. Gücer FI, Senturk S, Özkanli S, et al. Evaluation of Crohn’s disease activity by MR enterography: derivation and histopathological comparison of an MR-based activity index. Eur J Radiol. 2015;84: 1829–1834. 14. Novak KL, Wilson SR. The role of ultrasound in the evaluation of inflammatory bowel disease. Semin Roentgenol. 2013;48:224–233. 15. Ripollés T, Martínez MJ, Paredes JM, et al. Crohn disease: correlation of findings at contrast-enhanced US with severity at endoscopy. Radiology 2009;253:241–248. 16. Maconi G, Greco S, Duca P, et al. Prevalence and clinical significance of sonographic evidence of mesenteric fat alterations in Crohn’s disease. Inflamm Bowel Dis. 2008;14:1555–1561. 17. Medellin-Kowalewski A, Wilkens R, Wilson A, et al. Quantitative contrast-enhanced ultrasound parameters in Crohn disease: their role in disease activity determination with ultrasound. AJR Am J Roentgenol. 2016;206:64–73. 18. Chiorean MV, Sandrasegaran K, Saxena R, et al. Correlation of CT enteroclysis with surgical pathology in Crohn’s disease. Am J Gastroenterol. 2007;102:2541–2550. 19. Adler J, Punglia DR, Dillman JR, et al. Computed tomography enterography findings correlate with tissue inflammation, not fibrosis in resected small bowel Crohn’s disease. Inflamm Bowel Dis. 2012;18:849–856. 20. Migaleddu V, Scanu AM, Quaia E, et al. Contrast-enhanced ultrasonographic evaluation of inflammatory activity in Crohn’s disease. Gastroenterology. 2009;137:43–52. 21. Dillman JR, Stidham RW, Higgins PD, et al. Ultrasound shear wave elastography helps discriminate low-grade from high-grade bowel wall fibrosis in ex vivo human intestinal specimens. J Ultrasound Med. 2014;33:2115–2123. 22. Dillman JR, Stidham RW, Higgins PD, et al. US elastography-derived shear wave velocity helps distinguish acutely inflamed from fibrotic bowel in a Crohn disease animal model. Radiology. 2013;267: 757–766. 23. Stidham RW, Xu J, Johnson LA, et al. Ultrasound elasticity imaging for detecting intestinal fibrosis and inflammation in rats and humans with Crohn’s disease. Gastroenterology. 2011;141:819–826.e1. 24. Taylor SA, Punwani S, Rodriguez-Justo M, et al. Mural Crohn disease: correlation of dynamic contrast-enhanced MR imaging findings with angiogenesis and inflammation at histologic examination–pilot study. Radiology. 2009;251:369–379.

430


25. Danese S, Fiorino G, Angelucci E, et al. Narrow-band imaging endoscopy to assess mucosal angiogenesis in inflammatory bowel disease: a pilot study. World J Gastroenterol. 2010;16:2396–2400. 26. De Franco A, Di Veronica A, Armuzzi A, et al. Ileal Crohn disease: mural microvascularity quantified with contrast-enhanced US correlates with disease activity. Radiology. 2012;262:680–688. 27. Paredes JM, Ripollés T, Cortés X, et al. Contrast-enhanced ultrasonography: usefulness in the assessment of postoperative recurrence of Crohn’s disease. J Crohns Colitis. 2013;7:192–201. 28. Romanini L, Passamonti M, Navarria M, et al. Quantitative analysis of contrast-enhanced ultrasonography of the bowel wall can predict disease activity in inflammatory bowel disease. Eur J Radiol. 2014;83:1317–1323. 29. Ripolles T, Rausell N, Paredes JM, et al. Effectiveness of contrastenhanced ultrasound for characterisation of intestinal inflammation in Crohn’s disease: a comparison with surgical histopathology analysis. J Crohns Colitis. 2013;7:120–128. 30. Baumgart DC, Muller HP, Grittner U, et al. US-based real-time elastography for the detection of fibrotic gut tissue in patients with stricturing Crohn disease. Radiology. 2015;275:889–899. 31. Fraquelli M, Branchi F, Cribiù FM, et al. The role of ultrasound elasticity imaging in predicting ileal fibrosis in Crohn’s disease patients. Inflamm Bowel Dis. 2015;21:2605–2612. 32. Satsangi J, Silverberg MS, Vermeire S, et al. The montreal classification of inflammatory bowel disease: controversies, consensus, and implications. Gut. 2006;55:749–753. 33. Barr RG, Ferraioli G, Palmeri ML, et al. Elastography assessment of liver fibrosis: society of radiologists in ultrasound consensus conference statement. Radiology. 2015;32:92–104. 34. Boursier J, Zarski JP, de Ledinghen V, et al. Determination of reliability criteria for liver stiffness evaluation by transient elastography. Hepatology. 2013;57:1182–1191. 35. Rieder F, Zimmermann EM, Remzi FH, et al. Crohn’s disease complicated by strictures: a systematic review. Gut. 2013;62:1072–1084. 36. Novak K, Tanyingoh D, Petersen F, et al. Clinic-based point of care transabdominal ultrasound for monitoring Crohn’s disease: impact on clinical decision making. J Crohns Colitis. 2015;9:795–801. 37. Quaia E, De Paoli L, Stocca T, et al. The value of small bowel wall contrast enhancement after sulfur hexafluoride-filled microbubble injection to differentiate inflammatory from fibrotic strictures in patients with Crohn’s disease. Ultrasound Med Biol. 2012;38:1324–1332. 38. Ferraioli G, Tinelli C, Lissandrin R, et al. Point shear wave elastography method for assessing liver stiffness. World J Gastroenterol. 2014;20: 4787–4796. 39. Cantisani V, Grazhdani H, Drakonaki E, et al. Strain US elastography for the characterization of thyroid nodules: advantages and limitation. Int J Endocrinol. 2015;2015:908575. 40. Chen W, Lu C, Hirota C, et al. Smooth muscle hyperplasia/hypertrophy is the most prominent histological change in Crohn’s fibrostenosing bowel strictures: a semiquantitative analysis by using a novel histological grading scheme. J Crohns Colitis. 2016;11:92–104. 41. Ferraioli G, Tinelli C, Dal Bello B, et al. Accuracy of real-time shear wave elastography for assessing liver fibrosis in chronic hepatitis C: a pilot study. Hepatology. 2012;56:2125–2133. 42. Girlich C, Jung EM, Iesalnieks I, et al. Quantitative assessment of bowel wall vascularisation in Crohn’s disease with contrast-enhanced ultrasound and perfusion analysis. Clin Hemorheol Microcirc. 2009;43:141–148. 43. Fraquelli M, Sarno A, Girelli C, et al. Reproducibility of bowel ultrasonography in the evaluation of Crohn’s disease. Dig Liver Dis. 2008;40:860–866.

