Quantified small bowel motility assessment on MR enterography in paediatric inflammatory bowel disease – does it reflect clinical response?

doi:10.21203/rs.3.rs-4712819/v1

Download PDF

Research Article

Quantified small bowel motility assessment on MR enterography in paediatric inflammatory bowel disease – does it reflect clinical response?

https://doi.org/10.21203/rs.3.rs-4712819/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Quantified small bowel motility assessment using cine magnetic resonance enterography (MRE) has shown promise as a biomarker in adult inflammatory bowel disease. Whether quantified motility corresponds to treatment response in paediatric inflammatory bowel disease (PIBD) is unknown. Our objective to test whether changes in motility reflect response.

Materials and Methods

Local ethics approval was granted for this single institution, retrospective study. All children < 18 years with confirmed PIBD, who had > 1 MRE between Jan 2011-Jan 2022 were included. sMaRIA and motility index (quantified motility) at all terminal ileum (TI) and diseased non-TI segments were independently assessed by two radiologists each with ≥ 9 years’ experience. Change in (Δ) motility index was compared to clinical (gastroenterologist physician’s global assessment) and consensus radiological reference standard (response = decrease in sMaRIA ≥ 2) in responders versus non-responders using Mann-Whitney test. Sensitivity and specificity of Δmotility index ≥ 0 was compared to decrease in sMaRIA ≥ 2 for identifying clinical response.

Results

Of 64 children aged 5–16; 21 out of 64 (33%) were responders, 37 out of 64 (58%) were non-responders and 6 out of 64 (9%) had inactive disease according to clinical reference standard. Δmotility index by both radiologists was higher in responders (+ 16, + 39) than non-responders (-43, -44), p = 0.04, p = 0.009 each radiologist, respectively. Motility index was more sensitive (57% versus 24%), but less specific (67% versus 93%) than sMaRIA in identifying clinical response.

Conclusion

Motility index on cine MRE corresponds to clinical response, and is more sensitive at detecting response compared to sMaRIA in PIBD.

Crohn Disease

Gastrointestinal motility

Inflammatory bowel diseases

Magnetic Resonance Imaging (Cine)

Paediatrics

MR Enterography (MRE) is integral to the management of paediatric inflammatory bowel disease (PIBD)[1] [2] and assessment of treatment response is a key part of imaging follow-up which influences clinical decisions. It is therefore important to find reliable, quantitative MRI techniques to measure disease activity and response. MRE disease activity scores, such as the Magnetic Resonance Index of Activity (MaRIA) [3], simplified MaRIA (sMaRIA)[4] and Crohn’s Disease MRI Index (CDMI) [5], enable a semi-objective and reproducible assessment of disease activity in response to treatment [6, 7] and correlate to clinical markers in paediatric Crohn’s disease [8]. Although sensitivity and specificity of sMaRIA for evaluating treatment response has been reported as 89% and 87% in adults, respectively [4], scores can be time-consuming to calculate and rely on the measurement of subjective indices which introduce variability [9, 10, 11]. They are therefore not easily nor routinely used in normal clinical practice [6].

Cine motility MRE has been recognised by 2019 guidance from the European Crohn’s and Colitis Organisation and the European Society of Gastrointestinal and Abdominal Radiology as an important emerging tool in Crohn’s disease assessment [6]. Quantified small bowel motility (motility index) measured on cine MRE has shown promise as a biomarker in adult studies, where reduced motility index correlates with histopathological disease activity, symptoms and elevated faecal calprotectin, [12, 13, 14, 15, 16, 17]. Reduced motility index is also seen in bowel segments which appear subjectively normal on MRE [18], inferring a more diffuse impact on bowel motility, even when only the terminal ileum (TI) is actively inflamed. Motility index in adults outperforms subjective radiologist assessment in terms of correlation to clinical disease activity [19] and is more sensitive than MaRIA at detecting histopathological disease activity [17]. There is a lack of published data in paediatric patients. One feasibility study mirrored results seen in adult studies [20], and a small prospective trial showed increased motility index at the TI after 6-weeks of biological therapy [21], similar to adults [22]. There have been no trials to date testing motility index changes at non-TI sites in adult or PIBD. The limited ability of endoscopy to reach non-TI sites, and poor sensitivity of scoring systems in detecting disease at the proximal small bowel [23, 24], mean an additional biomarker for non-TI disease would be clinically beneficial.

Whilst motility index could provide an objective and sensitive measure of disease activity over time, it is currently unknown to what extent it could be incorporated into every day clinical practice, outside of controlled research intervals and with variable disease sites, severity and treatment regimens. Our primary hypothesis is that the motility index at all affected bowel segments will increase in a clinical cohort of PIBD patients who have responded to treatment. Further objectives are to (i) compare the sensitivity and specificity of change in (Δ) motility index to sMaRIA in response assessment; and (ii) to measure inter-reader agreement in measuring motility index.

This study received local hospital and university research ethics committee approval and a multi-institution data sharing agreement was in place (REC#10-H0-720-91, local R&D#18BB10). Informed consent was waived. All data was fully anonymised at source.

Study Participants

Participants were identified via single centre, retrospective search of all consecutive children aged under 18 who had two or more MRE examinations from January 2011 to January 2022 using the electronic patient record (Epic 2022, SlicerDicer, Wisconsin, United States). Those who did not have a final diagnosis of PIBD or did not have a cine sequence as part of their MRE were excluded. Demographic patient data (age and sex) and final diagnoses were recorded, as documented by PIBD specialist clinic letters and multi-disciplinary team meetings. All patients were included regardless of treatment regimen; we did not seek to assess response to a particular treatment as this study aimed to identify bowel motility changes in response to any treatment. There was no requirement for contemporaneous colonoscopy or histopathology for patients to be included in the study as these are not obtained as standard in the follow up assessment of PIBD. The first two MREs for each patient were labelled timepoint 1 and timepoint 2. Any subsequent MREs were not included in the analysis.

Clinical Reference Standard

Clinic letters, laboratory results, multidisciplinary discussions and discharge notes from the electronic record system were used to retrospectively assign a Paediatric Crohn’s Disease Activity Index (PCDAI) score [25]. Those with ulcerative colitis were instead assigned a Paediatric Ulcerative Colitis Activity Index (PUCAI) score [26]. Medical treatment and relevant blood and faecal markers of disease activity (haemoglobin, haematocrit, albumin, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR) and faecal calprotectin) were recorded. Values closest in time to the MRE timepoints 1 and 2 were included. Values outside of a 3-month window either side of the two timepoints were excluded and note was made of missing data so that this could be considered by the evaluating clinician. A consultant gastroenterologist (4 years’ paediatric inflammatory bowel disease experience at consultant level) was asked to use PCDAI validated scoring (< 10 = inactive; moderate or severe > 30; clinical response = decrease by > 12.5 points) [27] and trends in biochemical markers to provide a physician’s global assessment (PGA) blinded to the MRE report. This PGA was used to classify each patient into “inactive” (no active disease at either timepoint), “response” or “non-response” groups.

Imaging Protocol

All MREs were performed on 1.5-T Siemens Avanto or 3-T Prisma according to institution protocol (Supplementary Material 1), including coronal free breathing balanced steady-state free precession cine images with a temporal resolution of 30 frames/8 seconds/slice position (slice thickness 5mm; TR 271.41ms, TE 1.64ms; flip angle 60°), coronal balanced steady-state free precession (TruFISP), axial and coronal T2-weighted single-shot fast spin echo (HASTE) with and without fat saturation, axial diffusion-weighted imaging with corresponding apparent diffusion coefficient map, and coronal gradient-echo T1-weighted breath-hold (VIBE) with and without gadolinium (0.2ml/kg).The department protocol includes the use of buscopan to limit bowel motility, given after the cine sequence has been obtained and a oral contrast regimen to provide luminal distension.

MRE Analysis

Radiological analysis was carried out on the cloud based DICOM viewer, Entrolytics™ (Motilent, London, United Kingdom). Two paediatric radiology consultants (TW, 10 years and JB, 9 years subspeciality experience in gastrointestinal paediatric radiology), with no access to clinical information or original report, used sMaRIA [4] to independently score disease activity at the TI and worst additional site of small bowel disease (if non-TI) based on visual assessment of any sequence, with measurement of bowel wall thickness on axial/coronal T2 HASTE, for all patients at both timepoints. The TI was scored in all cases by both radiologists regardless of whether it visually appeared diseased or normal: it was scored on the MRE slice that captured the worst disease, or if normal, the slice that captured the longest segment of TI. The sMaRIA score was chosen as it has demonstrated reasonable inter reader agreement and has been shown to decrease over time in paediatric Crohn’s disease in response to therapy [7]. Each TI segment was classified radiologically as “inactive” if there was no active disease at either timepoint (sMaRIA of 0 at both timepoints), “response” (decrease of ≥ 2 points in sMaRIA), or “non-response” (increase, no change, or decrease of 1 point in sMaRIA) [4]. Each non-TI segment was classified radiologically as “response” or “non-response” using the same criteria, with non-TI segments not assessed if they did not have active disease at either time point. Where the non-TI segment identified at timepoint 1 was not easily identifiable at timepoint 2 due to resolution of the inflammation, the readers identified the segment most likely to represent the same segment based on location and course. Discordant selection of most severe segment and response outcomes were resolved in consensus between the two readers to obtain the radiological reference standard as per sMaRIA classification.

Motility Index Assessment

GIQuant (Motilent, London, United Kingdom) is a commercially available FDA approved, CE marked quantified small bowel motility tool for assessment of small bowel Crohn’s disease activity [17]. Timepoint 1 and 2 MREs were deidentified and uploaded to Entrolytics™ (Motilent, London, United Kingdom). Coronal cine images were processed with GIQuant® postprocessing software, based on non-rigid optic flow image registration technique, whereby deformation fields were summarised as the pixel-wise standard deviation of the Jacobian [28]. The results were displayed on a parametric “map” and motility index as a “GIQuant Score” in arbitrary units (a.u.). Following training and 5 test cases with a GIQuant expert, two radiology readers placed regions of interest (ROIs) on the postprocessed coronal images at the TI, and worst site of small bowel disease (if non-TI) to measure the motility index at the two timepoints, corresponding to bowel segments scored using sMaRIA at the same occasion. ROIs were placed to include bowel wall and lumen on the single slice that captured the longest segment of interest, and did not have to be the same for both readers.

Statistical Analysis

Study data were collected and managed using REDCap electronic data capture tools, hosted at University College London [29, 30] and R version 4.2.2 was used to analyse data [31]. The primary outcome was the magnitude of Δ in motility index and secondary outcomes included: (i) the direction of Δ motility index – where stable or increased motility (Δ ≥ 0) was classified as “response” vs reduced motility (Δ < 0) was “non-response” [21]; (ii) sensitivity and specificity of this threshold compared to sMaRIA; and (iii) inter-reader agreement. These outcomes were compared using the non-parametric Mann-Whitney U test between patients classified as “response” versus “non-response”, based on the clinical response assessment and radiological response assessment. Results were visually demonstrated using boxplots with individual readers and TI and non-TI segments analysed separately. Normally distributed outcomes were reported as mean ± standard deviation (S.D), median and interquartile range (IQR) were reported. Student’s T-test was applied to test for changes in clinical parameters in clinical response versus non-response.

Sensitivity, specificity and % agreement with clinical response of motility index Δ ≥ 0 [22] and sMaRIA Δ ≤-2 were calculated using the epiR package [32]. Inter-reader agreement in Δ motility index and Δ sMaRIA was assessed using proportional agreement in direction of change (+/-) using the Agree package [33]. Reliability of Δ motility index was quantified as Bland Altman bias and limits of agreement using the blandr [34]. A sample size calculation was not deemed necessary as all potential participants had been included in the selection. P values of < 0.05 were considered significant.

Study Participants

The search yielded 64 eligible individuals after exclusions, Fig. 1, aged 5–16 years (normally distributed, mean 11 years), 33/64 (52%) male, Table 1. According to clinical reference standard, 21/64 (33%) were classified as response, 37/64 (58%) non-response, 6/64 (9%) inactive. There were 8/64 (12%) classified as response, 42/64 (66%) non-response and 13/64 (20%) inactive according to sMaRIA radiological reference standard. One case was excluded in consensus due to being a post-surgical case with complications and no TI for assessment. The median time interval between MRE was 18 months (range 3–69 months). All patients were on medical treatment under gastroenterology for PIBD, including aminosalicylates, azathioprine, methotrexate and anti-tumour necrosis factor agents.

Δ Motility Index in Clinical Response and Non-response

A visual example of the post-processed image and motility map are displayed, Fig. 2 and motility index in TI and non-TI segments is summarised in Supplementary Material 2. A decrease in the motility index between the two timepoints (indicated by a negative value) represents a decrease in motion, whereas an increase (indicated by a positive value) represents an increase in motion. Patients who were classified as “response” according to the clinical assessment had a median (IQR) Δ motility index of + 16 (-81 to + 105) for reader 1, and + 39 (-56 to + 102) for reader 2. Conversely, patients clinically classified as “non-response” had a median (IQR) Δ motility index of -43 (-148 to + 25) for reader 1, and − 44 (-151 to + 10) for reader 2. The difference in Δ motility between the “response” and “non-response” groups was statistically significant for both reader 1 (p = 0.04) and reader 2 (p = 0.009), Fig. 3(a).

Table 1: Demographics, diagnosis, disease severity at timepoint 1 and 2 in those that had a clinical response to treatment according to expert gastroenterologist assessment, versus those that did not respond, or had inactive disease.

Characteristic

Response

N=21

Non-response

N=37

Inactive

N=6

P-value (response Vs non response)

Demographics

Mean age (y)

Sex (male)

12 (5-16)

9 (43%)

11 (5-15)

21 (57%)

12 (9-15)

3 (50%)

Diagnosis

Crohn’s

Ulcerative colitis

Very early onset IBD

IBD-unspecified

19 (90%)

2 (10%)

33 (89%)

2 (6%)

2 (6 %)

5 (83%)

1 (17%)

Disease severity at timepoint 1

Mean PCDAI/PUCAI

Mean FCP

Mean CRP (mg/L)

Mean ESR

1843

1261

369

Disease severity at timepoint 2 (change compared to timepoint 1)

Mean PCDAI/PUCAI

Mean FCP

Mean CRP (mg/L)

Mean ESR

9 (¯9)

735 (¯1108)

8 (¯3)

14 (¯14)

15 (4)

2032 (771)

15 («)

32 (10)

3 (¯5)

20 (¯349)

4 (¯5)

6 (¯17)

0.02

0.009

0.006

0.08

Cases with missing data at either timepoint (% of total)

Activity score parameters

FCP

CRP

ESR

6 (29%)

9 (43%)

3 (14%)

5 (14%)

18 (49%)

6 (16%)

7 (19%)

1 (17%)

4 (67%)

CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; FCP, faecal calprotectin; IBD, inflammatory bowel disease; PCDAI, paediatric Crohn’s disease activity index; PUCAI, paediatric ulcerative colitis activity index.

Δ Motility Index in Radiological Response and Non-response

Using the consensus radiological (sMaRIA) assessment, patients who were classified “response” had a median (IQR) Δ motility index of + 50 (-26 to + 97) for reader 1, and + 21 (-42 to + 93) for reader 2. In comparison, those who were radiologically classified as “non-response” had a median (IQR) Δ motility index of -25 (-105 to + 45) for reader 1, and − 16 (-141 to + 49) for reader 2. Although the median Δ motility index values were numerically higher for radiologically classified response, this did not reach significance, Fig. 3(b). When assessed separately, the motility index for the TI (n = 63) and non-TI (n = 26) segments were significantly higher in radiological response versus non-response for both readers, except for the TI segments for reader 2, which neared significance (p = 0.056), Supplementary Material 3. All Δ motility index values by both readers are presented, Table 2.

Table 2

Reader 1 and 2 Δmotility index values for cases classified according to clinical and radiological response.
Response Assessment	Reader 1 Median ΔMotility Index (a.u.) (IQR)			Reader 2 Median Δ Motility Index (a.u.) (IQR)
	Most severe	TI	Non-TI	Most severe	TI	Non-TI
Clinical Response Non-response Inactive	+ 16 (-81; +105) -43 (-148; +25) -49 (-101; -0)	-8 (-80; +63) -56 (-126; -4) -49 (-102; 0)	+ 91 (+ 16; +195) -1 (-71; +40) NA	+ 39 (-56; +102) -44 (-151; +10) -98 (-126; -22)	+ 3 (-56; +51) -40 (-135; +25) -98 (-120; +2)	+ 100 (-16; +182) -47 (-146; +8) NA
Radiological (overall) Response Non-response Inactive	+ 50 (-26; +97) -25 (-105; +38) -45 (-118; -1)	+ 37 (-26; +61) -47 (-111; +5) -45 (-118; -1)	+ 133 (-90; +279) + 22 (-21; +85) NA	+ 21 (-42; +93) -16 (-141; +49) -75 (-102; +2)	+ 21 (-42; +78) -31 (-120; +49) -75 (-102; +2)	+ 201 (-16; +224) -17 (-141; +11) NA
Radiological (segmental) Response Non-response Inactive	+ 85 (+ 19; +115) -40 (-123; +29) -45 (-118; -1)	+ 27 (-44; +61) -53 (-116; +4) -45 (-118; -1)	+ 115 (+ 85; +200) + 15 (-71; +40) NA	+ 51 (-18; +129) -38 (-144; +33) -75 (-102; +2)	-2 (-57; +55) -37 (-130; +49) -75 (-102; +2)	+ 175 (+ 83; +201) -47 (-143; +9) NA
TI, Terminal Ileum.

Sensitivity and Specificity of Response Detection

A confusion matrix of sMaRIA and Δ motility index versus gastroenterologist defined clinical response is displayed, Table 3. Consensus radiological assessment of response by sMaRIA had low sensitivity of 24% (95% CI: 8–47%) and high specificity 93% (81–99%). Sensitivity was higher when analysing the sMaRIA radiological assessment of the readers separately (29% and 43% for each reader), Table 4. When using Δ motility index of ≥ 0 as an identifier of “response”, Δ motility index had sensitivity 57% (34–78%) and specificity of 67% (50–80%) against the gold standard of clinical assessment.

Table 3

Confusion matrix consensus sMaRIA and mean Δ motility index response assessment against clinical response as the reference standard. Motility index Δ ≥ 0 and consensus sMaRIA Δ ≤-2 were considered ‘Response’ (N = 63).
	Clinical response + (n = 21)	Clinical response – (n = 42)
sMaRIA response+ (n = 8)	5	3
sMaRIA response- (n = 55)	16	39
ΔMotility index response+ (n = 26)	12	14
ΔMotility index response- (n = 37)	9	28

Table 4

Sensitivity and specificity of MRI assessed response compared to clinical response as the reference standard. Motility index Δ ≥ 0 [22] and sMaRIA Δ ≤-2 were considered ‘Response’ (N = 63).
	Proportional agreement (95% CI)	Sensitivity (95% CI)	Specificity (95% CI)
Expert A
Δ Motility Index	63% (50–75%)	57% (34–78%)	67% (50–80%)
Δ sMaRIA	73% (60–75%)	29% (11–52%)	95% (84–99%)
Expert B
Δ Motility Index	67% (54–78%)	67% (43–85%)	67% (50–80%)
Δ sMaRIA	75% (62–85%)	43% (22–66%)	90% (77–97%)
Combined
Mean Δ Motility Index	63% (50–75%)	57% (34–78%)	67% (50–80%)
Consensus Δ sMaRIA	70% (57–81%)	24% (8–47%)	93% (81–99%)

Inter-reader Agreement & Reliability

Readers agreed in the direction of Δ motility index in 78% (68–86%) at the most severe segment of disease overall, Supplementary Material 4. Inclusion of non-TI segments did increase the bias numerically, but there was no evidence of systematic bias at either TI or non-TI segments. Bland Altman limits of agreement demonstrated similar spread when the readers measured either motility at either TI or non-TI segments, Supplementary Material 2 and 5. Compared to the magnitude of the changes in motility index observed in each group following treatment ( < ± 50 a.u.) the Bland-Altman limits of agreement were relatively large ( > ± 165). A similar spread was observed when the readers measured motility at either TI or non-TI segmentsalthough wider differences were evident when selecting most severely affected segments in this cohort.

Readers agreed in the direction of Δ sMaRIA score in 86% (78–92%) of cases at the most severe segment of disease overall: 57% (29–79%) of cases where there was an increase in score and 91% (85–96%) of cases where there was a decrease in score.

Children with inflammatory bowel disease from a clinical, real-life cohort who responded to treatment had an overall significant increase in their motility index at small bowel sites, compared to those who clinically did not respond to treatment. This was particularly true for non-TI segments, where substantial increases in motility index were seen in patients who responded clinically and radiologically, compared to those who did not. The sensitivity of changes in motility index was higher compared to validated scoring system, sMaRIA, (57% versus 24%), although these sensitivities are both relatively low and neither should be used as sole response assessment tools. The specificity of change in motility index was lower than sMaRIA (67% versus 93%). There was 78% agreement in the direction of change in motility index between two readers. Motility index is a feasible supplementary tool for response assessment and could add to radiologist and gastroenterologist armamentarium.

The significant difference in the Δ motility index in patients who responded according to clinical markers are in keeping with the findings of Dillman et al[21] and a prior adult study [22], both of which reported significantly increased motility index over time in patients with inflammatory bowel disease treated with biologics. Compared to Dillman et al,[21] our study cohort had a wide variety of MRE interval times and were included regardless of treatment regimen, more reflective of clinical practice. This work provides further evidence that motility index can act as a biomarker of disease response in clinical settings, including at non-TI segments; an encouraging early result that has not been reported previously in adult or paediatric studies. Non-TI small bowel disease is difficult to assess as it cannot be reached endoscopically, and video capsule endoscopy only provides an overview of the mucosal surface, not transmural or mesenteric disease. In the authors’ opinion, MRE assessment of the proximal small bowel can be limited by poor distension as the oral contrast protocol is often optimised for assessment of the distal small bowel, which can cause a relative increase in diffusion signal abnormality and contrast enhancement. A tool that could non-invasively and accurately provide an assessment of non-TI small bowel disease would therefore be beneficial.

Interestingly, we did not find that classification of response using sMaRIA corresponded well to classification clinically and the differences in motility index between sMaRIA responders and non-responders did not reach significance. Although sMaRIA is a validated technique, it has been developed for mainly the TI and colon and only tested in a single paediatric small bowel study [7] so may be less well suited to PIBD. The definition of response using sMaRIA decreasing by ≥ 2 points is based on ulcer healing in adult studies [4], and anecdotally, paediatric disease (particularly in younger children) tends to be inflammatory initially, with a lower prevalence of penetrating disease. This threshold of change in sMaRIA may be too high, limiting sensitivity of response detection. In addition, given that bowel wall thickening is a predominant feature in PIBD, the poor correlation between clinical score and sMaRIA may reflect variability of bowel wall thickness measurement between the two readers, and a high number of false negatives seen in our cohort. In addition, sMaRIA does not take into account diffusion restriction abnormalities – a useful adjunct in PIBD [35]. Paediatric specific scoring systems such as the Pediatric Inflammatory Crohn’s MRE Index (PICMI)[36] does include a measure of diffusion abnormalities, but had not been validated as extensively as sMaRIA at the time of writing. In addition, ultrasound is a useful modality in PIBD particularly for younger children who are unable to tolerate MR [37], and a novel scoring system awaits external validation [38].

This study does have a number of limitations. Firstly, all MREs were clinically indicated, and this will have caused a selection bias towards the worst cases and those that were not clinically responding. This means that the results of our cohort may not be reflected in other institutions, where scanning habits may vary. In addition, the use of a clinical, paediatric cohort meant that there were suboptimal studies with movement artefact and inadequate bowel distension. These were not excluded as we felt they reflected real life clinical practice. We used a clinical, retrospective cohort with missing data and therefore could not rely on one single biomarker that was consistently available for every case. Histopathological correlation was also not available as follow up endoscopy is not routinely performed. We aimed to counteract this by using a global assessment by a single expert clinician, which we believe gave a fair evaluation of the patient as a whole, given the available information. More than one clinician providing an assessment may have improved response classification, had this been available.

Motility index increases with clinical response in PIBD in a real-life cohort. Agreement was good at 78% between two radiologists in terms of direction of change. Using the tool in initial and subsequent MREs in the same patient provides the radiologist and gastroenterologist with one additional biomarker to determine response in real world clinical practice– particularly helpful in non-TI regions that are difficult to assess otherwise on MR or via endoscopy. Further work should investigate whether the tool could be reliably used by non-expert readers. In addition, a threshold value for change in motility index that could be applied to all paediatric or adult patients, which reliably corresponds to clinical disease response, has not yet been established and would be beneficial.

Competing Interests

The authors of this manuscript declare relationships with the following companies: Motilent Ltd as follows: HEF – previous internship SAT – Holds share options

Author Contribution

T.W. is the guarantor for the article. R.M., H.F., J.B., K.J., and T.W. planned and conducted the study. R.M. and H.F collected and interpreted the data. R.M. drafted the manuscript. All authors edited and revised the initial drafts. All authors have approved the final submitted draft.

Acknowledgement

Thanks are extended to Motilent CEO, Alex Menys, for use of the Entrolytics platform and GIQuant®. Research support was provided by Judith Holmes, Research Assistant, and Tom Parry, Medical Statistician, at University College London Hospital Biomedical Research Centre.

Fidler J, Guimaraes L, Einstein D (2009) MR imaging of the small bowel. Radiographics 29:1181–1125. 10.1148/rg.296095507
Sinha R, Murphy P, Hawker P, Sanders S, Rajesh A, Verma R (2009) Role of MRI in Crohn's disease. Clin Radiol 64:341–352. 10.1016/j.crad.2008.08.006
Rimola J, Ordás I, Rodriguez S et al (2011) Magnetic resonance imaging for evaluation of Crohn's disease: validation of parameters of severity and quantitative index of activity. Inflamm Bowel Dis 17:1759–1768. 10.1002/ibd.21551
Capozzi N, Ordás I, Fernandez-Clotet A et al (2020) Validation of the Simplified Magnetic Resonance Index of Activity [sMARIA] Without Gadolinium-enhanced Sequences for Crohn's Disease. J Crohns Colitis 14:1074–1081. 10.1093/ecco-jcc/jjaa030
Steward M, Punwani S, Proctor I et al (2012) Non-perforating small bowel Crohn’s disease assessed by MRI enterography: deri- vation and histopathological validation of an MR-based activity index. Eur J Radiol 81:2080–2088. 10.1016/j.ejrad.2011.07.013
European Crohn’s and Colitis Organisation [ECCO] and the European Society of Gastrointestinal and Abdominal Radiology [ESGAR] (2019) ECCO-ESGAR Guideline for Diagnostic Assessment in IBD Part 2: IBD scores and general principles and technical aspects. J Crohns Colitis 13:273–284. 10.1093/ecco-jcc/jjy114
Dillman J, Ata N, Towbin A et al (2023) The Simplified MR Index of Activity Score in Pediatric Small-Bowel Crohn Disease: An Interreader Agreement and Responsiveness Study. AJR Am J Roentgenol 220:126–133. 10.2214/AJR.22.28123
Pomerri F, Al Bunni F, Zuliani M et al (2017) Assessing pediatric ileocolonic Crohn's disease activity based on global MR enterography scores. Eur Radiol 27:1044–1051. 10.1007/s00330-016-4443-y
D'Amico F, Chateau T, Laurent V, Danese S, Peyrin-Biroulet L (2020) Which MRI Score and Technique Should Be Used for Assessing Crohn's Disease Activity? J Clin Med 9:1691. 10.3390/jcm9061691
Tielbeek J, Makanyanga J, Bipat S et al (2013) Grading Crohn disease activity with MRI: interobserver variability of MRI features, MRI scoring of severity, and correlation with Crohn disease endoscopic index of severity. AJR Am J Roentgenol 201:1220–1228. 10.2214/AJR.12.10341
Rozendorn N, Amitai M, Eliakim R, Kopylov U, Klang E (2018) A review of magnetic resonance enterography-based indices for quantification of Crohn's disease inflammation. Th Adv Gastroenterol. 10.1177/1756284818765956
Menys A, Helbren E, Makanyanga J et al (2013) Small bowel strictures in Crohn's disease: a quantitative investigation of intestinal motility using MR enterography. Neurogastroenterol Motil 25:967–e775. 10.1111/nmo.12229
Menys A, Atkinson D, Odille F et al (2012) Quantified terminal ileal motility during MR enterography as a potential biomarker of Crohn's disease activity: a preliminary study. Eur Radiol 22:2494–2501. 10.1007/s00330-012-2514-2
Hahnemann M, Nensa F, Kinner S (2015) Improved detection of inflammatory bowel disease by additional automated motility analysis in magnetic resonance imaging. Invest Radiol 50:67–72. 10.1097/RLI.0000000000000097
Hahnemann M, Nensa F, Kinner S et al (2015) Quantitative assessment of small bowel motility in patients with Crohn's disease using dynamic MRI. Neurogastroenterol Motil 27:841–848. 10.1111/nmo.12558
Gollifer R, Menys A, Makanyanga J et al (2018) Relationship between MRI quantified small bowel motility and abdominal symptoms in Crohn's disease patients-a validation study. Br J Radiol 91:20170914. 10.1259/bjr.20170914
Menys A, Puylaert C, Tutein Nolthenius C et al (2018) Quantified Terminal Ileal Motility during MR Enterography as a Biomarker of Crohn Disease Activity: Prospective Multi-Institution Study. Radiology 289:428–435. 10.1148/radiol.2018180100
Menys A, Makanyanga J, Plumb A et al (2016) Aberrant Motility in Unaffected Small Bowel is Linked to Inflammatory Burden and Patient Symptoms in Crohn's Disease. Inflamm Bowel Dis 22:424–432. 10.1097/MIB.0000000000000601
Gollifer R, Menys A, Plumb A et al (2019) Automated versus subjective assessment of spatial and temporal MRI small bowel motility in Crohn's disease. Clin Radiol 74:814e9–814e19. 10.1016/j.crad.2019.06.016
Cococcioni L, Fitzke H, Menys A et al (2021) Quantitative assessment of terminal ileum motility on MR enterography in Crohn disease: a feasibility study in children. Eur Radiol 31:775–784. 10.1007/s00330-020-07084-1
Dillman J, Tkach J, Imbus R, Towbin A, Denson L (2022) MRI-Based Characterization of Intestinal Motility in Children and Young Adults With Newly Diagnosed Ileal Crohn Disease Treated by Biologic Therapy: A Controlled Prospective Study. AJR Am J Roentgenol 219:655–664. 10.2214/AJR.22.27792
Plumb A, Menys A, Russo E et al (2015) Magnetic resonance imaging-quantified small bowel motility is a sensitive marker of response to medical therapy in Crohn's disease. Aliment Pharmacol Ther 42:343–355. 10.1111/apt.13275
González-Suárez B, Rodriguez S, Ricart E et al (2018) Comparison of Capsule Endoscopy and Magnetic Resonance Enterography for the Assessment of Small Bowel Lesions in Crohn's Disease. Inflamm Bowel Dis 24:775–780. 10.1093/ibd/izx107
Prichard D, Hamilton Z, Savage T et al (2019) Capsule Endoscopy Complements Magnetic Resonance Enterography and Endoscopy in Evaluating Small Bowel Crohn's Disease. J Can Assoc Gastroenterol 3:279–287. 10.1093/jcag/gwz028
Hyams J, Ferry G, Mandel F et al (1991) Development and validation of a pediatric Crohn's disease activity index. J Pediatr Gastroenterol Nutr 12:439–447
Turner D, Otley A, Mack D et al (2007) Development, validation, and evaluation of a pediatric ulcerative colitis activity index: a prospective multicenter study. Gastroenterology 133:423–432. 10.1053/j.gastro.2007.05.029
Hyams J, Markowitz J, Otley A, Pediatric Inflammatory Bowel Disease Collaborative Research Group (2005) Evaluation of the pediatric crohn disease activity index: a prospective multicenter experience. Pediatr Gastroenterol Nutr 41:416–421. 10.1097/01.mpg.0000183350.46795.42
Odille F, Menys A, Ahmed A, Punwani S, Taylor S, Atkinson D (2012) Quantitative assessment of small bowel motility by nonrigid registration of dynamic MR images. Magn Reson Med 68:783–793. 10.1002/mrm.23298
Harris P, Taylor R, Thielke R, Payne J, Gonzalez N, Conde J (2009) Research electronic data capture (REDCap) – A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inf 42:377–381. 10.1016/j.jbi.2008.08.010
Harris P, Taylor R, Minor B (2019) REDCap Consortium, The REDCap consortium: Building an international community of software partners. J Biomed Inf 95:103208. 10.1016/j.jbi.2019.103208
R Core Team (2024) A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria, URL https://www.R-project.org/. Accessed 9th
Stevenson M, Sergeant E (2023) April (5th epiR Version 2.0.60. https://mvs.unimelb.edu.au/research/groups/veterinary-epidemiology-melbourne. Accessed 9th July 2024.
de Vet H, Mullender M, Eekhout I (2018) Specific agreement on ordinal and multiple nominal outcomes can be calculated for more than two raters. J Clin Epidemiol 96:47–53. 10.1016/j.jclinepi.2017.11.024
Datta D (2017) blandr: A Bland-Altman Method Comparison package for R. https://github.com/deepankardatta/blandr. Accessed 9th July 2024.
Morani A, Smith E, Ganeshan D, Dillman J (2015) Diffusion-weighted MRI in pediatric inflammatory bowel disease. AJR Am J Roentgenol 204:1269–1277. 10.2214/AJR.14.13359
Church P, Greer M, Cytter-Kuint R et al (2017) Magnetic resonance enterography has good inter-rater agreement and diagnostic accuracy for detecting inflammation in pediatric Crohn disease. Pediatr Radiol 47:565–575. 10.1007/s00247-017-3790-4
van Wassenaer E, Benninga M, van Limbergen J, D'Haens G, Griffiths A, Koot B (2022) Intestinal Ultrasound in Pediatric Inflammatory Bowel Disease: Promising, but Work in Progress. Inflamm Bowel Dis 28:783–787. 10.1093/ibd/izab125
van Wassenaer E, van Rijn R, de Voogd F et al (2023) Assessing Disease Activity in Pediatric Crohn's Disease Using Ultrasound: The Pediatric Crohn Disease Intestinal Ultrasound Score. J Pediatr Gastroenterol Nutr 76:582–589. 10.1097/MPG.0000000000003727

Competing interest reported. The authors of this manuscript declare relationships with the following companies: Motilent Ltd as follows:
HEF – previous internship SAT – Holds share options

Download PDF

Reviewers agreed at journal
04 Sep, 2024
Reviewers agreed at journal
30 Aug, 2024
Reviews received at journal
28 Aug, 2024
Reviewers agreed at journal
26 Aug, 2024
Reviews received at journal
24 Aug, 2024
Reviewers agreed at journal
19 Aug, 2024
Reviewers invited by journal
15 Jul, 2024
Editor assigned by journal
10 Jul, 2024
Submission checks completed at journal
10 Jul, 2024
First submitted to journal
09 Jul, 2024

You are reading this latest preprint version

Quantified small bowel motility assessment on MR enterography in paediatric inflammatory bowel disease – does it reflect clinical response?

Status:

Version 1

Abstract

Background

Materials and Methods

Results

Conclusion

Figures

INTRODUCTION

MATERIALS AND METHODS

Study Participants

Clinical Reference Standard

Imaging Protocol

MRE Analysis

Motility Index Assessment

Statistical Analysis

RESULTS

Study Participants

Δ Motility Index in Clinical Response and Non-response

Δ Motility Index in Radiological Response and Non-response

Sensitivity and Specificity of Response Detection

Inter-reader Agreement & Reliability

DISCUSSION

CONCLUSION

Declarations

Author Contribution

Acknowledgement

References

Additional Declarations

Supplementary Files

Status:

Version 1