Pediatric Inflammatory Bowel Disease Type Unclassified: A Nationwide Cohort Study in Scotland With up to 20 Years Follow-up Shows Reclassification in the Majority and Mild Course in Those Whose Diagnosis Is Unchanged

Abstract

Background

Given the paucity of long-term longitudinal data for inflammatory bowel disease type unclassified (IBDU), we aimed to clarify IBDU disease course and reclassification rate by presenting nationwide data with up to 20 years of follow-up.

Methods

We analyzed a prospectively identified 11-year cohort of pediatric patients diagnosed with IBDU between January 1, 2003 and December 31, 2013 at all Scottish pediatric IBD centers and followed up into adult services until December 31, 2022. Data were obtained from electronic medical records at fixed timepoints (5 and 10 years post-diagnosis) and at the final follow-up.

Results

Overall, 102 patients were included in the analysis (57/102 [56%] male, median [interquartile range {IQR}] age at diagnosis: 11.5 [9.1-13.2] years) with a median (IQR) follow-up length of 10.5 (8.6-14.0) years. A change of diagnosis was made in 61 of 102 patients (60%); of these, 30 patients (29%) were reclassified to Crohn’s disease (CD) and 31 patients (30%) to ulcerative colitis (UC). Patients who remained with IBDU had higher 1- to 5-year remission rates (IBDU 30/39 [77%] vs reclassified 16/57 [28%], P < .05), with lower rates of moderate-to-severe disease (IBDU 3/39 [8%] vs reclassified 31/57 [54%], P < .05) and less need for biologics across all timepoints (IBDU vs reclassified: first timepoint 1/39 [3%] vs 17/57 [30%], second timepoint 1/33 [3%] vs 26/56 [46%], third timepoint 0/18 [0%] vs 16/33 [49%]; all P < .05). Higher rates of surgical resections were observed in reclassified patients (reclassified 11/61 [18%] vs IBDU 1/41 [2%], P = .02).

Conclusions

In our nationwide pediatric IBDU cohort, 60% of patients were reclassified to either UC or CD over 10.5 years of median follow-up; those who remained with IBDU had a milder disease course.

Introduction

Inflammatory bowel disease (IBD) encompasses Crohn’s disease (CD), ulcerative colitis (UC), and inflammatory bowel disease type unclassified (IBDU).¹ IBDU is the least common subtype and accounts for 7.1% of incident cases in children according to the most recent meta-analysis.² It represents a distinct phenotype of colonic IBD where there are insufficient clinical, endoscopic, histological, and radiological findings to confidently diagnose either UC or CD.³ In a Canadian study of 7143 participants, Benchimol et al⁴ revealed an age-associated decline in frequency. Those younger than 6 years constituted 11% of new diagnoses, those aged 6-9 years constituted 10%, and those older than 10 years constituted 6.5%.⁴ Unsurprisingly, IBDU is more common in pediatric practice compared to adults, where around 6% of prevalent patients have a diagnosis of IBDU.⁵ Patients with pediatric-onset IBD have more extensive disease and early progression than those with adult-onset IBD within the same nationwide or regional population.⁶ IBDU as a distinct IBD phenotype is supported by clinical, genetic, and immunological data^3,7–9; however, many studies have a follow-up period below 5 years; therefore, the long-term reclassification rate and natural disease course are not fully characterized; follow-up is often terminated at transition to adult IBD services. Longitudinal studies assessing the clinical course, disease behavior, response to therapy, and associated complications are vital for optimizing diagnostic approaches, treatment strategies, and prognostic assessments.

Efforts have been made over recent years to standardize the diagnostic criteria for IBDU with the introduction of the Montreal classification,¹ followed by the Paris classification,¹⁰ and subsequently the Paediatric Inflammatory Bowel Disease (PIBD) classes,³ which remain the current standard to differentiate PIBD subtypes. While machine learning utilizing clinical, endoscopic, and histological inputs has been demonstrated to have moderate predictive value,¹¹ it is far from being validated and used in routine practice. With lower heterogeneity in disease definitions and improved proportions of children being fully investigated¹² in parallel with the emergence of more sensitive small bowel imaging techniques such as magnetic resonance enterography (MRE) and wireless capsule endoscopy, we would perhaps expect a decrease in the incidence. However, the body of research has yielded heterogeneous results, with some studies demonstrating escalating incidence figures and others a reduction.^3,13,14 Further population-based IBDU cohorts with complete accrual are required to accurately characterize the trends of incidence and prevalence.

This retrospective study aimed to examine the clinical characteristics and outcomes of pediatric patients with IBDU by longitudinally analyzing a well-characterized nationwide cohort of children diagnosed with IBDU with up to 20 years of follow-up. This research intended to contribute to the existing body of knowledge surrounding IBDU, ultimately providing additional insights into the natural disease course and presenting features of this IBD subtype.

Materials and Methods

Scotland had a population of 5 327 700 in 2013¹⁵ at the end of patient identification, and the point prevalence of IBD is one of the highest demonstrated worldwide.¹⁶ Pediatric IBD care then and now is provided through 3 tertiary-managed clinical networks (centered in Aberdeen, Edinburgh, and Glasgow) including all district general hospitals. PIBD disease registries are prospectively maintained, allowing for complete case accrual of patients with IBDU over the study period. In this nationwide study, which is part of the PIES (Paediatric IBD Epidemiology in Scotland) nationwide collaborative research program, we analyzed a prospectively identified cohort of pediatric patients (<17 years) diagnosed with IBDU in Scotland between January 1, 2003 and December 31, 2013, and followed up until December 31, 2022, including those who transitioned to adult services. The diagnosis of IBDU was established and re-evaluated in 2014 from clinical, endoscopic, radiological, and histological data based on the revised Porto criteria.¹⁰ Historic cases were reviewed for diagnostic accuracy with questionable cases checked and confirmed with agreement between 2 PIBD experts (D.C.W. and R.K.R.). Data were retrospectively collected in 2023 from electronic medical records (demographics, endoscopic and radiological assessments, longitudinal disease severity based on defined physician’s global assessment, medical treatment, surgical outcomes, and eventual IBD reclassification) at fixed timepoints (closest clinical assessment to 0, 5, and 10 years post-diagnosis) and at the final follow-up. For analysis purposes, timepoints in follow-up were grouped into 1-5 years (first timepoint), 6-10 years (second timepoint), and >10 years (third timepoint) of follow-up.

Disease course between timepoints was defined as either remission, mild, moderate, or severe. Remission was defined as disease in continuous clinical and biochemical (normal fecal calprotectin and C-reactive protein) remission between the considered timepoints. Mild disease was defined as mild disease flares occurring less than twice per year with no requirement for treatment escalation. Moderate disease was defined as more than 2 disease flares per year or a disease flare requiring escalation of therapy. Severe disease was defined as more than 2 severe disease flares requiring treatment escalation, including anti-TNF-α therapy or resectional surgery. Treatment escalation was defined as any type of therapeutic change including additional medical therapy (immunomodulators, steroids, exclusive enteral nutrition (EEN) or advanced therapies), surgical treatment, or optimization of advanced therapies. Patients requiring multiple biologics were defined as requiring more than 1 biologic class during the follow-up period. Surgery was defined as any elective or emergency resectional surgery, abscess drainage, or perianal surgical intervention. In patients with multiple changes of diagnosis, the patient was considered reclassified according to the latest and final diagnosis.

The study was reported as per the STROBE statement for observational cohort studies.¹⁷ Categorical variables were summarized as counts and frequencies (%). Non-normally distributed variables were described as median and interquartile range (IQR), while normally distributed variables were described as mean and standard deviation (SD). Differences in continuous variables were analyzed using the Mann-Whitney U test or 2-sample t test, whereas differences in categorical data were explored using the chi-squared test. Comparisons were made between patients who retained an IBDU diagnosis at the end of follow-up and those who were reclassified into either CD or UC. Statistical significance was set at a P-value of <.05. Statistics were performed using GraphPad Prism (version 10.1.1, GraphPad Software, 2023).

In line with local protocol, ethical approval was not required as this was a review of service delivery and clinical practice.¹⁸

Results

Population Characteristics

A total of 116 patients with IBDU were identified across the 3 managed clinical networks; 14 patients were excluded from further analysis as they emigrated out of the area before the first follow-up timepoint and consequently had incomplete notes (Supplementary Figure 1). Therefore, 102 patients were included in the analysis (57/102 [56%] male, median [IQR] age at diagnosis: 11.5 [9.1-13.2] years) with a median (IQR) length of follow-up of 10.5 (8.6-14.0) years. The clinical characteristics of these patients including age at disease diagnosis, presenting symptoms, most proximal colonoscopic extent, and disease distribution at diagnosis are provided in Table 1. All patients presented with recognized symptoms of IBD, most commonly colitic symptoms. At diagnosis, all patients underwent upper and lower gastrointestinal (GI) endoscopies, with only 9 of 102 patients (9%) missing further small bowel imaging (MRE, wireless capsule endoscopy, or barium follow-through). All patients who did not undergo a complete endoscopy (to the terminal ileum) had small bowel imaging. Considering initial endoscopic disease distribution, the majority of patients (80/102; 78%) had a pancolitis (E4 phenotype) at diagnosis, with 58/102 (57%) having macroscopic involvement of the upper GI tract. Accurate IBD prevalence data are available between January 1, 2003 and December 31, 2008 (the first 5 years)¹⁹; out of the total 436 incident cases of IBD, 43 patients in our cohort during this timeframe were diagnosed with IBDU, giving an IBDU prevalence rate of 9.8%.

Disease Reclassification

A change of diagnosis was made in 61 of 102 patients (60%), with a median (IQR) length of time to change of diagnosis of 4.9 (1.9-7.7) years (Figure 1). The diagnosis was changed to CD in 30 of 102 patients (29%) and to UC in 31 of 102 patients (30%), with a median (IQR) time to change of diagnosis of 5.0 (2.0-8.2) years and 2.0 (2.0-3.0) additional endoscopies for CD, and 4.7 (1.8-7.7) years and 2.0 (1.0-2.3) additional endoscopies for UC. IBDU remained the diagnosis in 41 of 102 patients (40%) after a median (IQR) follow-up of 9.7 (7.0-13.9) years and 1.0 (1.0-2.0) additional endoscopy. One patient experienced multiple changes in diagnosis: initially to UC at 1.4 years and later to CD at 10.3 years post-diagnosis. The median age at diagnosis was not significantly different between those who remained with IBDU and those who were reclassified into either UC or CD (IBDU: 11.7, CD: 10.4, and UC: 11.2 years, P > .05) (Table 1). The majority of disease reclassifications were made in the first follow-up period with 39 of 61 patients (64%) being reclassified in the first 5 years, 17 of 61 (28%) between 5 and 10 years, and the remaining 5 of 61 (8%) after 10 years of follow-up. There was no difference in the type of reclassification, either UC or CD, at the different timepoints: at the first timepoint, 18 of 30 patients (60%) were reclassified to CD and 21 of 31 (68%) to UC; at the second timepoint, 9 of 30 patients (30%) were reclassified to CD and 8 of 31 (26%) to UC; and at the third timepoint, 3 of 30 patients (10%) were reclassified to CD and 2 of 31 (6%) to UC (all P > .05) (Figure 1). Of the patients who did not have small bowel imaging performed at the initial diagnostic work-up, 2 patients were reclassified to CD following endoscopic reassessment for persistently active disease with histology favoring CD; 3 patients were reclassified to UC based on repeat histology; and 4 patients retained an IBDU diagnosis: of these, 3 did not require further endoscopic assessment due to quiescent disease maintained on aminosalicylates (ASA) and 1 underwent subtotal colectomy for acute severe colitis (ASC). Patients changed their diagnosis to CD, UC, or retained an IBDU diagnosis after a mean follow-up of 2.5, 5.2, and 8.4 years, respectively.

Disease Course

Data were available for 96 of 102 patients (94%) at the first timepoint, 89 of 102 (87%) at the second timepoint, and 51 of 102 (50%) at the third timepoint (Table 2). Patients who remained with a diagnosis of IBDU had a milder disease course within the first 5 years compared to those who were reclassified (first timepoint remission rates: 30/39 [77%] in IBDU vs 16/57 [28%] in CD/UC, P < .05), while equivalent remission rates were observed at the other timepoints (IBDU vs CD/UC, second timepoint: P = .63 and third timepoint: P = .92) (Table 2 and Supplementary Table 1). In parallel, those patients reclassified to UC or CD were more likely to have a moderate-to-severe disease course in the first 5 years but not after that (3/39 [8%] in IBDU vs 31/57 [54%] in CD/UC at first timepoint, P < .05; IBDU vs CD/UC at second and third timepoints were P = .47 and P = .54, respectively) (Table 2 and Supplementary Table 1). When comparing individuals reclassified to CD and UC over the first 5 years, a higher proportion of patients with active disease of any severity was observed in those reclassified to CD (24/28, 86%) compared to those reclassified to UC (17/29, 59%) (P = .04) (Table 2 and Supplementary Table 2). Absolute figures for each timepoint and diagnosis are given in Table 2.

Medical Treatment

A decreasing requirement for treatment escalation was observed in the entire cohort over time with 45 of 96 patients (47%) requiring an escalation during the first timepoint, 22 of 89 (25%) during the second timepoint, and 15 of 51 (29%) at the third timepoint (Table 3). Patients whose disease was reclassified to CD or UC were more likely to require treatment escalation at all timepoints (IBDU vs CD/UC: first timepoint: 8/39 [21%] vs 37/57 [65%]; second timepoint: 4/33 [12%] vs 18/56 [32%]; and third timepoint: 1/18 [6%] vs 14/33 [42%]; all P < .05) and also had a higher biologic utilization (IBDU vs CD/UC: first timepoint: 1/39 [3%] vs 17/57 [30%]; second timepoint: 1/33 [3%] vs 26/56 [46%]; and third timepoint 0/18 [0%] vs 16/33 [49%]; all P < .05) (Figures 2 and 3; Table 3; Supplementary Table 3). Moreover, patients reclassified to CD were more likely to be on a biologic than patients reclassified to UC at all the timepoints (CD vs UC: first timepoint: 12/28 [43%] vs 5/29 [17%]; second timepoint: 19/29 [66%] vs 7/27 [26%]; and third timepoint: 13/19 [68%] vs 3/14 [21%]; all P < .05); however, they did not have more frequent treatment escalations (CD vs UC: first timepoint: 20/28 [71%] vs 17/29 [59%]; second timepoint: 12/29 [41%] vs 6/27 [22%]; and third timepoint: 9/19 [47%] vs 5/14 [36%]; all P > .05) (Table 3 and Supplementary Table 4). Patients reclassified to UC or CD had higher use of immunomodulators (azathioprine, mercaptopurine, and methotrexate) in the first 5 years but not at subsequent timepoints (IBDU vs CD/UC at first timepoint: 11/39 [28%] vs 36/57 [63%], P < .05; second timepoint: 5/33 [15%] vs 18/56 [32%], P = .07; and third timepoint: 1/18 [6%] vs 6/33 [18%], P = .21) (Table 3 and Supplementary Table 3). Patients who retained an IBDU diagnosis more frequently had their disease controlled on ASA or no treatment throughout follow-up (IBDU vs CD/UC: first timepoint: 31/39 [80%] vs 11/57 [19%]; second timepoint: 25/33 [76%] vs 21/56 [38%]; third timepoint: 13/18 [72%] vs 12/33 [36%]; all P < .05) (Figure 3; Table 3; Supplementary Table 3). Across all timepoints, 12 of 41 patients (29%) who retained an IBDU diagnosis did not require any treatment escalation, while only 5 of 61 patients (8%) who were reclassified were not therapeutically escalated (P < .05). Absolute figures for each timepoint and diagnosis are given in Table 3.

Surgical Treatment

Surgery was required in 3 of 41 patients (7%) who remained with an IBDU diagnosis compared to 6 of 30 patients (20%) reclassified to CD and 6 of 31 patients (19%) reclassified to UC across all timepoints; this difference was not statistically significant (IBDU 3/41 vs CD/UC 12/61, P = .08). There was no difference in the median (IQR) time to surgery between IBDU (4.7 [2.4-9.1] years), CD (7.0 [4.5-8.7] years), or UC (3.1 [1.4-10.6] years, all P > .05). When examining differences in IBD-related surgical resections, a higher proportion of patients with reclassified CD and UC required surgery compared to patients with IBDU (IBDU 1/41 vs CD 5/30 [P = .03], IBDU 1/41 vs UC 6/31 [P = .02]). The colectomy rate for patients with UC in the cohort was 6 of 31 (19%), with 3 undergoing elective surgery for medically refractory disease and 3 for emergency management of ASC due to unresponsive first- and second-line therapies. The surgery of patients with CD included 3 subtotal colectomies for medically refractory disease, 1 spontaneous perforation in active CD requiring a minor resection, 1 stricturoplasty, and 1 perianal fistulotomy. The surgery in patients with IBDU included 1 colectomy for ASC, 1 resection following volvulus, and 1 abdominal abscess drainage; none resulting in a change in diagnosis.

Discussion

This study has the longest published follow-up periods of pediatric patients with IBDU in the literature to current date and aimed to provide population-based data on the reclassification rate and disease course of this important but relatively understudied IBD subtype. Up to 60% of patients in our cohort were reclassified to either CD or UC, with patients who remained diagnosed with IBDU having a milder disease course and less need for treatment escalation. While this is a historic cohort, the IBDU prevalence rate (2003-2008) is comparable to the literature,^{2,4,7,14,20–22} and the latest interpretation of diagnostic criteria has been applied allowing for distinctions in historic investigation utilization and greater generalizability to modern cohorts and practice.

Reclassification

There are a wide range of published reclassification rates of IBDU evident within the literature; however, there is a trend to studies with longer follow-up having higher rates of IBDU reclassification. When examining studies with under 5 years of follow-up, including prospective multicenter studies and large disease registry studies, reclassification rates range from 21% to 44% after a median follow-up period of 1.9-4.3 years. These studies also have significant heterogeneity in the IBD reclassification subtype ranging from 22% to 70% in CD and 30% to 78% in UC.^{7,12,14,20,23–28} The relatively short follow-up in these studies has the potential to underreport the lifetime reclassification rate. Our reclassification rate, during 1-5 years of follow-up, is comparable at 38%. When we examine studies with longer follow-up, this rate rises. In a single-center non-population-based Israeli study, Rinawi et al²⁹ followed a cohort of 53 patients with IBDU for a median of 6.8 (±6.7) years and showed a reclassification rate of 54% (83% to CD and 17% to UC). While the majority of reclassifications occurred within the first 5 years, a significant proportion required longer follow-up to display differentiating features which is consistent with our findings.²⁹ Reclassification rates are lower in adult cohorts, potentially due to the lower baseline IBDU diagnosis rate. In the adult Epi-IBD cohort including 29 centers across 22 countries, 1289 patients were diagnosed with IBD, of whom 112 (9%) were classified as IBDU at diagnosis with 25% reclassified over 5 years of follow-up (71% to UC and 29% to CD) after a median (IQR) of 6 (4-12) months.²⁴

In our study, 9% of patients were diagnosed with IBDU without initial small bowel imaging. We agree with the modern practice that a label of IBDU should not be applied without small bowel imaging, but note that our rate of incomplete small bowel assessment is low compared to studies from a similar time period (2003-2013). The multicenter Eurokids cohort was recruited between 2005 and 2013 and showed 38.1% of patients diagnosed with IBDU did not have diagnostic small bowel imaging (vs 27.5% in CD and 64.1% in UC).¹² Better access to newer imaging techniques has likely increased the number of patients receiving complete diagnostic work-up; however, there is currently no clear evidence that this has resulted in lower IBDU incidence rates.^3,12–14

In our data, with a median of 10.5 years of follow-up, our reclassification rate continued to increase up to 60%. This is supported by a recent large systematic review and meta-analysis of 7 studies containing 397 patients with IBDU, showing that 50% of patients were reclassified as either UC (32.7%) or CD (17.3%).² While a significant number of patients will be reclassified, it still leaves a distinct cohort of patients with a different disease course and potentially different management needs.

Disease Course

There is a relative paucity of data on long-term disease course in IBDU; however, there is a trend to less severe disease and a lower medication burden.^24,29 In our cohort, patients who retained a diagnosis of IBDU were more likely to be maintained in remission with ASA or no medication, have a lower need for disease reassessments, have lower rates of biological use, and have lower rates of resectional surgery compared to those reclassified to either CD or UC. Moreover, patients reclassified to CD showed higher rates of active disease of any severity and a higher biologic utilization compared to those reclassified to UC. This is mirrored in the literature. Rinawi et al showed that an increased need for surgical interventions (33% vs 7%, P = .031), systemic steroids (88% vs 35%, P < .001), and biological treatments (66.7% vs 34.5%, P = .02) were associated with reclassification to CD compared to IBDU in a pediatric cohort.²⁹ In the adult Epi-IBD cohort, patients reclassified to UC had a higher burden of IBD-related surgery compared to those who remained diagnosed with IBDU (30% vs 2%), as well as a higher need for biologics and steroids (25% vs 6% and 75% vs 42%, respectively).²⁴

Our data showed IBDU is potentially more benign than both UC and CD once established as IBDU and without migration to another disease type; however, a direct comparison between the 3 different IBD subtypes was not feasible in our exclusively IBDU population. When comparing disease course between different subtypes from diagnosis, we observed a similar trend compared to those with reclassified CD and UC. Aloi et al²³ in a large (IBDU n = 260) retrospective cohort from 23 centers showed that patients with IBDU had a less severe disease course with more patients with IBDU either in remission or with mild disease by the end of follow-up compared to CD (69% IBDU vs 46% CD, P = .0001), lower corticosteroid use compared to UC (59% IBDU vs 71% UC, P = .004), and lower surgery rates compared to both CD and UC (IBDU 2% vs CD 8% [P = .008] and UC 8% [P = .009]).²³ In contrast, patients with CD had a higher biologic utilization than those with IBDU or UC (CD 34% vs IBDU 12% and UC 17%, P < .0001). An analysis of the RISK study supports this with patients with CD more likely to receive anti-TNF therapy compared to IBDU or UC (CD 50% vs IBDU 24% and UC 28%, P < .0001).⁷ Again, in the adult Epi-IBD cohort, patients with CD had a higher risk of both IBD-related surgery and hospitalization compared to those with IBDU (CD vs IBDU: odds ratio [OR] 3.5; 95% confidence interval [CI], 1.6-7.6 for surgery and OR 2.8; 95% CI, 1.6-4.9 for hospitalization); these were not significantly increased in UC.²⁴

Disease reassessments and additional investigation results from endoscopy, radiology, and surgical specimens are essential to gain sufficient information to change the diagnosis from IBDU. Likely as a result of quiescent disease, controlled on lower levels of therapy, our patients who remained with a diagnosis of IBDU had fewer endoscopies and disease reassessments than the reclassified groups. This was also reported in the Eurokids registry which showed that those patients who were reclassified to either CD or UC had a greater number of disease reassessments (82% and 82% of reclassified patients received a repeat ileoscopy and colonoscopy, respectively, compared to 28% and 34% of the IBDU group).¹² Although the current paradigm considers IBDU a distinct entity, whether a proportion of patients with IBDU represent genuine indeterminate cases with distinct management requirements or merely constitute milder forms of CD or UC that have undergone less intensive reassessment or those whose disease is in evolution, remains an open question. It is possible that with more frequent reassessments and longer follow-ups, the prevalence of IBDU will decrease. With such a high proportion of patients reclassified within 5 years of diagnosis, there should be a balance between accurately characterizing a patient’s phenotype and limiting invasive investigations.

Limitations

There are limitations in this study, in terms of its retrospective nature that did not enable complete data collection of all the variables considered, particularly as this is a historic cohort assembled before the 2014 publication of the revised Porto criteria and the 2017 publication of “PIBD classes.” We have presented the proportion of data missing at each timepoint; those with missing data in the first 5-year period who later had follow-up data are those whose historic, non-computerized notes were insufficient to complete the dataset but who later have clear documentation. A drawback of such long follow-up is that many patients were originally diagnosed using different criteria and less sensitive testing. We aimed to mitigate this by applying modern diagnostic criteria within the limits of data availability and having cases reviewed by IBD experts. The clear limitation of receiving an IBDU label without appropriate small bowel imaging is acknowledged and discussed above, including historical context. We accepted the potential limitations of our data as the price paid to provide the long-term data in terms of reclassification rates and course—which can then be repeated in future years in better phenotyped contemporary cohorts of IBDU.

Conclusions

In this population-based nationwide study of IBDU reclassification and disease course with complete case accrual over 10 years and the longest published median follow-up in the literature to date, reclassification rate continued to increase with longer follow-up, and patients who remained with a diagnosis of IBDU had a milder disease course. While further study is required to delineate the optimal management strategies to achieve the best outcomes in this group of patients, this work will help inform clinicians when counseling patients with a new diagnosis of IBDU and their families regarding long-term disease course, reinforcing the importance of a lower threshold than in confirmed CD or UC for complete disease reassessment during IBDU exacerbations for diagnosis confirmation and treatment tailoring.

Supplementary Data

Supplementary data are available at Inflammatory Bowel Diseases online.

Conflicts of Interest

D.I.F.W. has been supported by an Edinburgh Children’s Hospital Charity (ECHC) research fellowship. D.I.F.W. also reports receiving an honorarium from Nutricia-Danone. L.G. fellowship has been supported by the University of Milan. F.C. was supported by a Crohn’s in Childhood Research Association (CICRA) Dave Casson Paediatric Gastroenterology Research Fellowship February 1, 2013 to January 31, 2015. R.K.R. reports receiving speaker fees, travel support, and participation on advisory boards from Nestlé, AbbVie, Celltrion, and Pharmacosmos. D.C.W. reports receiving consultancy fees, speaker fees, and/or travel support from Celltrion and AbbVie. R.H. has no conflicts of interest to declare.

Ethical Considerations

In line with local protocol, ethics approval was not required.

References

Author notes