The objectives of this study were: (i) to document the frequency of IBDU within the total number of children diagnosed with IBD in a regional population; and (ii) to document any change to this diagnosis in the long-term.

Data were collected at endoscopy for all children aged 0-17 years diagnosed as PIBD using the Porto criteria over 7-years (2004-2011). All patients had small bowel imaging (MRE or barium meal follow-through), Ileocolonoscopy and UGIE. Biopsies (2 to 4 per site) were obtained from terminal ileum, cecum, ascending colon, transverse colon, descending colon, sigmoid colon, rectum, duodenum, pylorus, stomach and esophagus. Repeat endoscopic assessments were carried out as per clinical indication on patients with persistent symptoms despite treatment.

All histological specimens were reported by a single specialist pediatric histopathologist both at initial diagnosis and reassessment. Information collected included: age at diagnosis, gender, ethnicity, histological findings, perinuclear anti-neutrophil cytoplasmic antibodies (pANCA), inflammatory markers (C-reactive protein, ESR), albumin, liver function tests, full blood count, and urea and electrolytes. Children with a diagnosis of IBDU were managed as per the British Society of Paediatric Gastroenterology, Hepatology and Nutrition guidelines [3].

In 2016 (follow-up period of 4.5–11.5 years), clinical notes were examined and data on any change in diagnosis from IBDU to CD or UC, and the histological basis for this change were collected.

A total of 344 new PIBD patients were diagnosed during 2004-2011. Fig. 1 shows the subtypes of PIBD. The mean age at diagnosis was: 11.5 years for CD (n=199), 11.6 years for UC (n=119) and 10.1 years (n=26) for IBDU. The age range of IBDU patients was 1.4 years to 16.1 years; only one child was aged <2 years in whom cow’s milk protein allergy had been ruled out.

Fig. 1 Subtypes of pediatric inflammatory bowel disease cases over 7 years (n=344).

Data were available for 25 out of 26 patients with IBDU; 16 were males (64%) and 9 females (36%). After a minimum 4.5 years and maximum 11.5 years follow-up period, case notes of these patients were examined. Seventeen had been re-evaluated by endoscopic assessment. Eight were in clinical remission, and had not clinically warranted an endoscopic reevaluation. Histological features of the group (Group A) who had revision of diagnosis from IBDU to either CD or UC are highlighted in Web Table I. Group B consisted of cases without revision of diagnosis. There were no significant differences at initial diagnosis in the histology of the eight children who remained in clinical remission as compared to IBDU patients (n=7) who needed re-evaluation but had no change in diagnosis.

Table I lists the initial therapeutic interventions used in 25 IBDU patients. Prednisolone use was similar between the two groups (P=0.95). Use of aminosalicylates alone suggestive of milder disease appeared to be higher in Group B but this was not statistically significant (P=0.49).

TABLE I	Therapeutic Interventions Used at Initial Diagnosis	

The median follow-up to revision of diagnosis was 51 months (range 34-87 months). At follow-up (by 2016) diagnosis of IBDU had changed in 10/25 (40%) cases; 7 to CD and 3 to UC. Two of these 10 patients required hemicolectomy (1 CD, 1 UC). Preoperative assessment endoscopy was considered inappropriate in these two patients as they were too sick. Detailed histological results were not available for three patients (2 transferred to their local hospital’s adult gastroenterology services and 1 migrated to another region).

In this study from a single Pediatric gastroenterology center strictly following Porto diagnostic criteria, IBDU comprised only 7.5% (26/344) of total IBD patient at initial diagnosis. Children with IBDU were younger than those with CD or UC and there was male preponderance. After a median follow-up period of 52 months, the diagnosis of IBDU needed revision in 10 (40%) children; 7 to CD and 3 to UC.

The study had some limitations. It included review of clinical notes in 2016 (4.5-11.5 years after initial diagnosis of IBDU), and was not a continuous longitudinal study. Some data were missing from the clinical notes, and thus unavailable for final analysis. Some blood results were not available as patients referred from secondary care hospitals had initial investigations done locally. Testing for Anti-Saccharomyces cerevisiae antibodies　was not offered by our laboratory. For three patients who had moved to other centers, only summaries of change of diagnosis but not detailed histology following repeat endoscopic assessment were available

Previous studies have recorded much higher proportion (12.7-22%) of IBDU out of total PIBD patients [4,5]. The frequency of IBDU in our series was much lower (7.5%), and is in concordance with only one recent multicenter European study with 3641 children with IBD diagnosed using the Porto criteria (EUROKIDS Registry) where IBDU frequency at initial diagnosis was 7.7% [6]. These low percentages in our study and EUROKIDS study are likely to be a reflection of strict adherence to the Porto criteria for diagnosis of PIBD. In our study, over time, IBDU decreased from 7.5% to 4.3% which is similar to the EUROKIDS study where IBDU decreased to 5.6% during a median follow-up of 5.7 years [6]. A retrospective US study carried out pre-publication of the Porto criteria with 78 IBDU children, documented a lower age at diagnosis [9.2±4 years], and 23% IBDU were reclassified: 8 CD, 5 UC, and 5 non-IBD conditions [7]. A 6-center retrospective study with 210 PIBD patients, IBDU was reclassified in 8/20 (40%) patients, median time to revision was 18.5 months [8]. In our study, use of prednisolone and aminosalicylates at initial diagnosis was similar between those whose diagnosis changed (Group A) and those whose diagnosis remained unchanged (Group B). A retrospective multicenter study which pooled data on roughly equal numbers of patients with CD, UC and IBDU, 260/797 had IBDU [9]. These IBDU patients had milder disease course, with lower medication burden and need for surgery [9].

Our study suggests that it is essential to strictly follow the recommended Porto criteria for optimizing diagnosis of IBD and specially IBDU. Early repeat reassessment with endoscopy and small-bowel imaging in cases with persistent symptoms or where surgery may be a possibility should be considered. As treatment of CD and UC differs considerably, the treatment for changed correct diagnosis is essential.

Contributors: SPP: study design, data collection, analysis, manuscript preparation and revision; BKS: concept, supervision, manuscript editing and revision, and provided expert opinion.

Acknowledgements: Dr Christine Spray and Dr Dharamveer Basude, Consultant Pediatric Gastroenterologists, Bristol Royal Hospital for Children.

Funding: None. Competing interests: BKS was a founder member of ESPGHAN working group on IBD (The Porto Group) and served on the group until 2015.

1. Levine A, Koletzko S, Turner D, Escher JC, Cucchiara S, de Ridder L, et al. ESPGHAN revised porto criteria for the diagnosis of inflammatory bowel disease in children and adolescents. J Pediatr Gastroenterol Nutr. 2014;58:795-806.

2. IBD Working Group of the European Society for Paediatric Gastroenterology, Hepatology and Nutrition. Inflammatory bowel disease in children and adolescents: recommendations for diagnosis – the Porto criteria. J Pediatr Gastroenterol Nutr. 2005;41:1-7.

3. Sandhu BK, Fell JM, Beattie RM, Mitton SG, Wilson DC, Jenkins H; IBD Working Group of the British Society of Paediatric Gastroenterology, Hepatology, and Nutrition. Guidelines for the Management of Inflammatory Bowel Disease in Children in the United Kingdom. J Pediatr Gastroenterol Nutr. 2010;50:S1-13.

4. Prenzel F, Uhlig HH. Frequency of indeterminate colitis in children and adults with IBD - a metaanalysis. J Crohns Colitis. 2009;3:277-81.

5. Turunen P, Kolho KL, Auvinen A, Iltanen S, Huhtala H, Ashorn M. Incidence of inflammatory bowel disease in Finnish children, 1987-2003. Inflamm Bowel Dis. 2006;12:677-83.

6. Winter DA, Karolewska-Bochenek K, Lazowska-Przeorek I, Lionetti P, Mearin ML, Chong SK, et al. Pediatric IBD-unclassified is less common than previously reported; Results of an 8-year audit of the EUROKIDS Registry. Inflamm Bowel Dis. 2015;21:2145-53.

7. Malaty HM, Mehta S, Abraham B, Garnett EA, Ferry GD. The natural course of inflammatory bowel disease-indeterminate from childhood to adulthood: within a 25 year period. Clin Exp Gastroenterol. 2013; 23:115-21.

8. Newby EA, Croft NM, Green M, Hassan K, Heuschkel RB, Jenkins H, et al. Natural history of paediatric inflammatory bowel diseases over a 5-year follow-up: A retrospective review of data from the register of paediatric inflammatory bowel diseases. J Pediatr Gastroenterol Nutr. 2008;46: 539-45.

9. Aloi M, Birimberg-Schwartz L, Buderus S, Hojsak I, Fell JM, Bronsky J, et al. Treatment options and outcomes of pediatric IBDU compared with other IBD subtypes: A retrospective multicenter study from the IBD Porto Group of ESPGHAN. Inflamm Bowel Dis. 2016;22:1378-83.