- Email: [email protected]
Clinical disease activity and endoscopic severity correlate poorly in children newly diagnosed with Crohn’s disease
ORIGINAL ARTICLE
Clinical disease activity and endoscopic severity correlate poorly in children newly diagnosed with Crohn’s disease Nicholas Carman, MBBS,1 Diane Tomalty, MSc,2 Peter C. Church, MD,2 David R. Mack, MD, PhD,1 Eric I. Benchimol, MD, PhD,1 Anthony R. Otley, MD,3 Kevan Jacobson, MD,4 Hien Q. Huynh, MD,5 Jennifer C. DeBruyn, MD,6 Wael El-Matary, MD,7 Mary Sherlock, MD,8 Johan Van Limbergen, MD, PhD,3 Anne M. Griffiths, MD,2 Thomas D. Walters, MD2; on behalf of the Canadian Children Inflammatory Bowel Disease Network: A Joint Partnership of Canadian Institutes of Health Research and the Children with Intestinal and Liver Disorders Foundation Ottawa, Toronto, Halifax, Vancouver, Edmonton, Calgary, Winnipeg, Hamilton, Canada
Background and Aims: Treatment goals in Crohn’s disease (CD) have evolved to target mucosal healing. There is now a drive to determine if noninvasive measures can adequately identify the attainment and persistence of this goal. Currently, data describing the relationship between clinical indices and endoscopic appearance in pediatric CD are sparse. Our aim was to compare endoscopic severity with the weighted Pediatric Crohn’s Disease Activity Index (wPCDAI) in children with newly diagnosed CD. Methods: All children aged 17 years newly diagnosed with CD enrolled in an inception cohort at sites of the Canadian Children Inflammatory Bowel Disease Network were eligible. Clinical disease activity at presentation was evaluated by the wPCDAI and conventional biochemical parameters. Severity of disease at ileocolonoscopy was assessed by the simple endoscopic score for CD (SES-CD), with segmental subscores noted. We evaluated the association of SES-CD and disease activity markers using the Pearson test of correlation, the Spearman rank coefficient, and linear regression models. Results: Two hundred eighty patients from 11 centers were included in the analysis. The median wPCDAI score was 60 (interquartile range, 40-80; 53% severe). Median SES-CD was 16 (interquartile range 10-22; 51% severe). The wPCDAI correlated weakly with SES-CD (r Z .39, P < .001). Examination of the individual components that contribute to the wPCDAI demonstrated weak correlation with the SES-CD for all items apart from stooling (moderate correlation, r Z .50, P < .001). Routine blood tests did not correlate well with the SES-CD. In regression models, variation in clinical symptoms accounted for most of the variation in both the wPCDAI and SES-CD, with no additional benefit from routine blood tests. Conclusions: In children with newly diagnosed CD, wPCDAI correlates poorly with endoscopic disease activity. As treatment paradigms evolve to target mucosal healing, clinical markers should not be used in isolation to determine disease activity. (Gastrointest Endosc 2018;-:1-9.)
Abbreviations: CD, Crohn’s disease; CDAI, Crohn’s Disease Activity Index; CIDsCaNN, Canadian Children Inflammatory Bowel Disease Network; CRP, C-reactive protein; IBD, inflammatory bowel disease; PCDAI, Pediatric Crohn’s Disease Activity Index; SES-CD, simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index. DISCLOSURE: During fellowship training at The Hospital for Sick Children, funding support for N. Carman was provided by Abbvie Canada. All authors disclosed no financial relationships relevant to this publication. Copyright ª 2018 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 https://doi.org/10.1016/j.gie.2018.09.025 Received May 14, 2018. Accepted September 20, 2018.
www.giejournal.org
Current affiliations: Children’s Hospital of Eastern Ontario, Department of Pediatrics, University of Ottawa, Ottawa, Canada (1), The Hospital for Sick Children, Department of Pediatrics, University of Toronto, Toronto, Canada (2), IWK Health Centre, Department of Pediatrics, Dalhousie University, Halifax, Canada (3), BC Children’s Hospital, Department of Pediatrics, University of British Columbia, Vancouver, Canada (4), Stollery Children’s Hospital, Department of Pediatrics, University of Alberta, Edmonton, Canada (5), Alberta Children’s Hospital, Department of Pediatrics, University of Calgary, Calgary, Canada (6), Children’s Hospital Research Institute of Manitoba, Department of Pediatrics, University of Manitoba, Winnipeg, Canada (7), McMaster Children’s Hospital, Department of Pediatrics, McMaster University, Hamilton, Canada (8). Reprint requests: Nicholas Carman, MBBS, FRACP, Division of Gastroenterology, Hepatology and Nutrition, Children’s Hospital of Eastern Ontario, 401 Smyth Road, Ottawa, ON, K1H 8L1, Canada.
Volume
-,
No.
-
: 2018 GASTROINTESTINAL ENDOSCOPY 1
Endoscopic severity evaluation in children with newly diagnosed CD
Treatment goals for Crohn’s disease (CD) have advanced beyond control of symptoms and facilitation of normal growth; the aspiration now is intestinal healing. It is hoped that the attainment of such “deep remission” will prevent disease-related adverse events. Studies among adults with CD demonstrate the lack of correlation between disease activity, as measured by the Crohn’s Disease Activity Index (CDAI), and the endoscopic appearance of the ileocolonic mucosa.1 Accordingly, clinical trials evaluating new therapies in CD require objective assessment of inflammatory activity in the intestine.2 The simple endoscopic score for CD (SES-CD) is the multi-item measure most often used.3 CD develops during childhood in up to 20% of patients.4 Achievement of intestinal healing is particularly important in young patients with CD, given their long life ahead during which disease progression and repeated need for intestinal resection may occur despite achieving clinical remission based on symptomatology.5 In contrast to adult CD, however, studies assessing endoscopic severity of disease and correlation with clinical assessment of disease activity have been lacking. The Pediatric CD Activity Index (PCDAI) is the most commonly used pediatric multi-item measure of disease activity.6 The weighted PCDAI (wPCDAI) is its validated, mathematically weighted, short form.7 In comparison with the adult CDAI, both the PCDAI and wPCDAI rely less on subjectively reported symptoms and more on objective findings and laboratory investigations; they might therefore be expected to better reflect endoscopic inflammation. In the context of a pan-Canadian multicenter prospective inception cohort study, wherein the ileocolon is systematically evaluated, we aimed to explore the relationship before treatment between endoscopic activity and clinical disease severity using the SES-CD and the wPCDAI.
METHODS The network The Canadian Children Inflammatory Bowel Disease (IBD) Network: a Joint Partnership of the Canadian Institutes of Health Research and Children with Intestinal and Liver Disorders Foundation (CIDsCaNN), was formed in 2013 and now has participation from 12 academic pediatric IBD centers across Canada. All children and adolescents younger than 17 years presenting with new-onset CD or ulcerative colitis are eligible for enrollment in an inception cohort study. Baseline evaluation of suspected IBD is undertaken based on Porto diagnostic criteria8 and includes ileocolonoscopy and upper endoscopy with mucosal biopsy specimens for histologic assessment and small bowel imaging, most commonly using magnetic resonance enterography. Pretreatment demographic and phenotypic data are prospectively recorded, including for CD patients: PCDAI/ wPCDAI, physician global assessment of disease activity and SES-CD. All data, including routine laboratory parameters (complete blood count, C-reactive protein [CRP], serum al2 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2018
Carman et al
bumin, erythrocyte sedimentation rate) are captured using standardized case report forms. The diagnosis of CD is made at individual sites based on conventional criteria but confirmed centrally via central review of key clinical, endoscopic, and histologic reports. Macroscopic localization of CD is based on the Paris classification of pediatric IBD.9
Study patients Patients enrolled in the inception cohort with a diagnosis of CD involving the colon and/or terminal ileum (Paris classification L1 [ileal] or L2 [colonic] or L3 [ileocolonic] L4 [upper GI]) who had undergone complete ileocolonoscopy with visualization of the terminal ileum were included in the analysis. Patients with CD limited to the upper GI tract and/or small intestine proximal to the terminal ileum (Paris classification L4a and/or L4b) were excluded, as were patients with incomplete ileocolonoscopic examination or incomplete data. Additionally, patients were excluded if the baseline endoscopic and clinical assessments were more than 4 weeks apart or if treatment was started before the completion of both assessments.
Clinical and endoscopic activity evaluation: PCDAI and SES-CD Individual elements encompassing the clinical and endoscopic disease activity indices were collected after ileocolonoscopy or the clinic visit by the local site IBD physician on a standardized case report form and uploaded onto a REDCap database. These were subsequently downloaded by a single central physician (N.C.), who then calculated total scores for both the clinical and endoscopic tool in an attempt to avoid biasing the local physician’s scoring. This method of comparing clinical and endoscopic scores using local site readers, without central reads, is consistent with existing adult and pediatric literature on this topic.1,10,11 The PCDAI is a multi-item measure of inflammatory disease activity.6 Items include abdominal pain, number of stools per day, general well-being, hematocrit, erythrocyte sedimentation rate, albumin, weight, height velocity, abdominal tenderness, perirectal disease, and extraintestinal manifestations of disease. The total PCDAI score ranges from 0 to 100. The wPCDAI was derived by reweighting the PCDAI mathematically using data from 4 large prospective studies of children with CD, in which physician global assessment of inflammatory disease activity served as the dependent variable. Hematocrit and height velocity contributed little to the reweighting, and so these items were excluded. The wPCDAI is composed of 3 domains (clinical symptoms, physical examination, and laboratory variables), with individual items mathematically weighted to produce an overall score that classifies patients into 4 disease activity categories: none, <12.5; mild, 12.5 to 40; moderate, >40 to 57.5; and severe, >57.57 (Fig. 1). For the SES-CD, the intestine is divided into 5 anatomic segments: rectum, left side of the colon segment, transverse colon, right side of the colon segment, and ileum. Items of www.giejournal.org
Carman et al
Endoscopic severity evaluation in children with newly diagnosed CD
Clinical History – Recall, 1 week Abdominal Pain 0 = None
Score
10 = Mild: Brief, does not interfere with activities
20 = Moderate/Severe: Daily, longer lasting, affects activities, nocturnal
Patient Functioning, General Well Being 10 = Occasional difficulty in 0 = No limitation of maintaining age appropriate activities, well activities, below par Stools (per day) 0 = 0-1 liquid stools, no 7.5 = Up to 2 semi-formed blood with small blood, or 2-5 liquid
Score 20 = Frequent limitation of activity, very poor Score 15 = Gross bleeding, or >/= 6 liquid, or nocturnal diarrhea
Laboratory Erythrocyte Sedimentation Rate (ESR) 0 = <20 mm/hr 7.5 = 20-50 mm/hr
15 = >50 mm/hr
Albumin 0 = >/= 3.5 g/dl
20 =
Score
Score 10 = 3.1-3.4 g/dl
Examination Weight 0 = Weight gain or voluntary weight stable/loss Perirectal Disease 0 = None, asymptomatic tags
Score 5 = Involuntary weight stable, weight loss 1-9%
10 = Weight loss >/= 10%
Score
15 = Active fistula, drainage, 7.5 = 1-2 indolent fistula, tenderness, or abscess scant drainage, no tenderness Extra-intestinal Manifestations (fever >/= 38.5 for 3 days over the past week, definite arthritis, uveitis, E.nodosum, P. gangrenosum 0 = None 10 = One or more
Score
Total Score (0 – 125) Scoring 0 – <12.5 12.5 – 40 >40 – 57.5 >57.5
None/Remission Mild disease Moderate Disease Severe Disease Figure 1. Weighted Pediatric Crohn’s Disease Activity Index.
interest include ulcer size, estimates of the ulcerated and affected surface, and the presence of luminal narrowing. Categorical descriptions are recorded by the endoscopist for each of these items within each segment (Fig. 2). The SESCD score is calculated by combining scores of individual segments; hence, the maximum score possible in L1 disease is 12, 45 in L2, and 56 in L3. Overall SES-CD scores are classified into 4 disease categories: none, 0 to 2; mild, 2 to 6; moderate, 7 to 15; and severe, 16.3 CIDsCaNN uses a Central Phenotyping Committee to ensure the consistency of phenotypic characteristics of included patients. This committee is made up of experienced pediatric IBD physicians (A.M.G., J.C., H.Q.H.). Before www.giejournal.org
commencing patient recruitment to the Network, the site principle investigators reviewed and practiced using the endoscopy scoring tool at a face-to-face meeting. This exercise was repeated annually at subsequent face-to-face meetings.
Statistical analysis Laboratory values, indices, and their component subscores are presented as medians with interquartile range (IQR). Differences in the distributions when stratified by disease location were tested using analysis of variance and Mann-Whitney U tests as appropriate. A correlation matrix consisting of each index, its component subscores, and the Volume
-,
No.
-
: 2018 GASTROINTESTINAL ENDOSCOPY 3
Endoscopic severity evaluation in children with newly diagnosed CD
Carman et al
SES-CD Score 2 0.5 – 2cm
3 >2cm
<10%
10 – 30%
>30%
Unaffected Segment
<50%
50 – 75%
>75%
None
Single, can be passed
Multiple, can be Cannot be passed passed
Variable Size of Ulcers
0 None
1 0.1 – 0.5cm
Ulcerated Surface
None
Affected Surface
Presence of Narrowings
Total
Total score for this segment
Total Score = sum of score for each segment (Ileum + Right Colon + Transverse Colon + Left Colon + Rectum) Figure 2. The simple endoscopic score for Crohn’s disease (SES-CD).
various laboratory values was examined using the Pearson method or the Spearman rank coefficient as appropriate.12 The kappa statistic was used to determine the level of agreement between categorical subclassifications of each index.13 The Bowker test was used to determine if patterns of classification discordance were random. A series of linear regression models was used to explore the relative contribution of the various clinical and endoscopic subscores to the scoring of each activity index. The relative model fit and estimation of a variable’s relative contribution to any demonstrated association was assessed using the r2 statistic. A significance level of .05 was used throughout. All statistical analyses were performed using IBM SPSS Statistics version 22.0 for Windows software (IBM Corp, Armonk, NY).
Ethics This study protocol was reviewed and approved by the Research Ethics Board at each participating institution.
patients the time between clinical and endoscopic assessment was longer than 4 weeks. As such, 280 patients had complete baseline assessments within the designated time frame and were eligible for analysis. The demographic and phenotypic characteristics are reported in Table 1. The distributions of the wPCDAI and SES-CD scores within each Paris classification of the disease location are presented in Figure 3. The distributions of wPCDAI scores within each of the location subgroups were relatively similar, with 52% of patients having “severely active” disease at diagnosis (L1: median, 50; IQR, 32.5-67.5; L2: median, 55; IQR, 39-85; L3: median, 62.5; IQR, 42-80). Reflecting differences in the maximal SES-CD scores possible within each of L1, L2, and L3 disease, the median SES-CD for patients with isolated ileal (L1) disease (7; IQR, 4-9), was substantially lower than that for those with either isolated colonic (L2) disease (17; IQR, 1224) or ileocolonic (L3) disease (17; IQR, 13-23) (L1 vs L2, P < .001; L1 vs L3, P < .001; L2 vs L3, P Z .39).
Relationship between SES-CD and wPCDAI RESULTS Patients Seven hundred children and adolescents with CD were enrolled in the inception cohort study at the time of data extraction. Of these, 187 had an incomplete baseline wPCDAI (reflective of the unavailability of erythrocyte sedimentation rate in some centers), 74 had an incomplete endoscopic evaluation before treatment (eg, did not reach terminal ileum before initiation of treatment), and in 159 4 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2018
Despite relatively similar distributions, there was only a weak correlation between the SES-CD and wPCDAI score (r Z .39; P < .001) (Fig. 4). To ascertain if there was a differential relationship between the SES-CD and wPCDAI that was dependent on clinical disease severity, we subsequently categorized the subjects according to disease activity (none/mild, wPCDAI 0-40; moderate/severe, wPCDAI >40) and repeated the correlation analysis. This demonstrated a weaker relationship between clinical disease activity and endoscopic appearance in patients with clinically www.giejournal.org
Carman et al
Endoscopic severity evaluation in children with newly diagnosed CD
TABLE 1. Baseline patient characteristics stratified by disease location Paris Classification Baseline characteristics Male gender
All patients (n [ 280)
Ileocolonic (L3) (n [ 163)
Colonic (L2) (n [ 74)
Ileal (L1) (n [ 43)
153 (55)
84 (52)
42 (57)
27 (63)
13.4 (11.2-15)
13.4 (11.3-15)
12.8 (9.7-14.9)
14.2 (12.4-15.7)
3 (0-11)
2 (0-11)
3 (1-13)
4 (0-12)
SES-CD
16 (10-22)
17 (13-23)
17 (12-24)
7 (4-9)
wPCDAI
60 (40-80)
62.5 (42-80)
55 (39-85)
50 (32.5-67.5)
ESR
41 (22-61)
42 (25-61)
40 (18-63)
42 (23.7-59)
Age, y Days between endoscopic and clinical/biochemical assessments
Albumin
34 (29-38)
33 (29-37)
36 (30-40)
35 (31-39)
hsCRP
33 (10.1-70.3)
44.3 (16.7-79)
12.5 (2-43.5)
30 (12.7-82)
CRP
24.7 (12.1-59)
47/163: 30 (14.2-61)
25/74: 17.8 (7.6-63)
22/43: 19.3 (12-39)
Mild
63 (23)
33 (20)
15 (20)
15 (35)
Moderate
108 (38)
57 (35)
32 (43)
18 (42)
Severe
110 (39)
73 (45)
27 (37)
10 (23)
PGA
Values are n (%) or median (interquartile range). SES-CD, Simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index; ESR, erythrocyte sedimentation rate; hsCRP, high-sensitivity C-reactive protein; CRP, C-reactive protein; PGA, physician’s global assessment.
less-active disease (none/mild: r Z .08, P Z .52; moderate/ severe: r Z .33, P < .001). The wPCDAI was originally formatted to be an ordinal classifier of disease activity (none, mild, moderate, severe disease activity), and so we examined its relationship with the SES-CD using this structure. As illustrated in Figure 5, although a modest linear relationship exists (analysis of variance: F Z 15.4, P < .001), minimal variance is explained (R2 Z .05). The SES-CD also has previously validated cut-points that describe severity of mucosal disease (none, mild, moderate, severe mucosal disease). Overall agreement between the ordinal classifiers for each tool was poor (122/280 agreed, k Z .14, P Z .003). Notably, the pattern of discordance was not random (Bowker test P < .001) with mucosal severity classifications in the discordant cases systematically more severe than clinical activity classifications.
Correlation of the SES-CD with wPCDAI according to disease location In patients with isolated colonic (L2) disease, there was a better relationship between clinical and endoscopic activity, demonstrating moderate correlation between the 2 scores (r Z .49, P < .001). A weaker relationship was seen in patients with ileocolonic (L3) disease (r Z .36, P < .001) and no relationship in isolated ileal (L1) disease (r Z .05, P Z .97).
Correlation of laboratory measures with endoscopic and clinical scores There was a weak relationship overall between SES-CD and each of the laboratory measures (Table 2). When stratified by www.giejournal.org
disease location, there was a strong relationship between high-sensitivity CRP and SES-CD in colonic disease (r Z .67, P < .001) but not with erythrocyte sedimentation rate or albumin (r Z .37, P < .001 and r Z –.37, P < .001, respectively) (Table 2). In contrast, the wPCDAI showed a better relationship to all the laboratory measures (erythrocyte sedimentation rate: r Z .47, P < .001; albumin: r Z –.53, P < .001; high-sensitivity CRP: r Z .56, P < .001) (Table 3). This relationship persisted regardless of disease location.
Analysis of subscores Examination of each individual subscore that contributed to the wPCDAI are summarized in Table 2. Overall, all subscores only weakly correlated with the SES-CD score apart from stooling (r Z .41, P < .001). Notably, this relationship was only present in patients with colonic involvement (L2/L3), especially in isolated colonic disease where there was moderate correlation (r Z .50, P < .001). When examining the SESCD subscores (Table 3) and their relationship to the clinical score, ulcer size showed the best relationship overall with a moderate correlation (r Z .41, P < .001). Again, this relationship was strongest in isolated colonic disease (r Z .60, P < .001).
Regression models of SES-CD and wPCDAI We examined the wPCDAI and its subdomains to determine the overall contribution of each variable in the multiitem index to the overall score. The clinical components of the wPCDAI (stooling, pain, and functioning) accounted for 71% of the variance within the wPCDAI (R Z .84, P < .001). When the clinical components were examined Volume
-,
No.
-
: 2018 GASTROINTESTINAL ENDOSCOPY 5
Endoscopic severity evaluation in children with newly diagnosed CD
Carman et al
120
100 30 80
60
20
wPCDAI
Simple Endoscopic Score (SES-CD)
40
40 10 20
0
0
Ileal (L1)
Colonic (L2) Paris Classification
SES-CD
Ileocolonic (L3) wPCDAI
Figure 3. wPCDAI and SES-CD score stratified by disease location (n Z 280). SES-CD, Simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index.
in isolation, 80% of the variance within this subgroup could be attributed to pain and functioning. Indeed, when examined further it was determined that variation in these 2 items accounted for approximately 65% of the variance in the total wPCDAI scores (R Z .8, P < .001). We modeled the SES-CD score and explored its relative relationship to the various domains of the wPCDAI. When all 8 subcomponents that make up the wPCDAI were included as individual variables in a linear regression model of the SES-CD (3 symptom-based items, 3 examinationbased items, and 2 laboratory tests) 24% of the variance was explained (P < .001). This was somewhat improved from the relationship seen between the SES-CD and the composite wPCDAI score (R2 Z .15, P < .001). The addition of CRP to the model marginally improved the fit (R2 Z .24 vs R2 Z .32). The wPCDAI elements that contributed significantly to the model were the symptom-based items (stooling: b Z .5 [95% confidence interval, .34-.64], P < .001; function: b Z .14 [95% confidence interval, .06-.30], P Z .04). Abdominal pain was noncontributory. There was no appreciable benefit from the physical examination and laboratory items.
DISCUSSION Our data concerning the spectrum of severity of endoscopic disease in an inception cohort of children with 6 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2018
new-onset CD are unique. Endoscopic severity of disease at presentation, specifically the presence of deep ulcers, has been suggested, albeit in limited data accrued among adults, to be a risk factor for progressive disease and an indication of need for aggressive treatment.14 If pediatric gastroenterologists are to attempt to risk stratify and thereby individualize treatment regimens, greater attention to consistent description of endoscopic disease using validated measurement tools is required. To date, there are scarce data on the relationship between mucosal appearance at endoscopy and clinical disease activity scores in pediatric patients with CD. In this study, we demonstrate a weak correlation between the wPCDAI and endoscopic appearance at diagnosis as described by the SES-CD (r Z .39; P < .001). A small pediatric study (n Z 24) previously demonstrated a similarly poor correlation between the PCDAI and SES-CD at diagnosis (r Z .33, P > .05).10 A more recent study of 100 children with CD of varying duration that examined the relationship between SES-CD and clinical indices revealed a slightly better association (wPCDAI r Z .43, P < .001; PCDAI r Z .45, P < .001).11 However, 19 of these patients (19%) had longstanding mucosal healing; in patients with well-established intestinal healing one might anticipate a better relationship between the 2 indices. Our results are similar to data from adult CD studies.3,15,16 When examining the distributions of scoring across disease locations, clinical index scores were relatively consistent www.giejournal.org
Carman et al
Endoscopic severity evaluation in children with newly diagnosed CD
Simple Endoscopic Score (SES-CD)
120 100
wPCDAI
80 60 40 20 0
0
10
20
30
40
40
30
20
10
0 None/Mild
Simple Endoscopic Score (SES-CD)
Figure 4. Correlation between the SES-CD and wPCDAI (n Z 280) (r Z .39, P <.001). SES-CD, Simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index.
across isolated ileal, colonic, and ileocolonic disease. However, the SES-CD showed a distinct difference between ileal and colonic disease. The SES-CD is calculated by the sum of scores across involved segments. In cases of severe isolated ileal disease, the maximal score possible is 12, which only corresponds to moderate endoscopic severity. Therefore, defining disease severity in isolated ileal disease is difficult when using the SES-CD, raising the question of whether the same scoring system should continue to be used for both isolated ileal disease and more extensive disease. The 2 location subgroups often present with different symptomatology, have a different natural history, and may even require a different treatment paradigm. We examined the components of the wPCDAI and SESCD to explore the contributions of individual variables to overall score using a series of linear regression models. When examining the wPCDAI against its subscores, our regression model demonstrated that the variation in scoring is determined largely by symptoms and their effect, with reports of abdominal pain, stooling, and functional impairment together accounting for 70% of the variation in total score. These symptoms account for a significant amount of patient discomfort and disease experience and have a considerable impact on disease-related quality of life for the patient.17,18 The more objective variables in the wPCDAI (physical examination, laboratory tests) contributed minimally to the overall composite score. When we examined the relationship between wPCDAI variables and SES-CD, stooling showed the best relationship in patients with colonic involvement (r Z .50, P < .001). Correlation between laboratory variables and SES-CD was weak, but it was reasonable with the clinical activity score. In addition, CRP marginally improved the model fit when added to wPCDAI subscores in our linear regression analysis, but the overall relationship was still poor. This suggests that biochemical measures of inflammation www.giejournal.org
Moderate wPCDAI Categories
Severe
Figure 5. SES-CD scores stratified by wPCDAI (n Z 280). Distribution of SES-CD scores across wPCDAI categories assessed by analysis of variance (analysis of variance: F Z 15.0, P < .001, R2 Z .05). SES-CD, Simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index.
cannot necessarily be used in place of endoscopic assessment and are in fact a third axis of disease experience that needs to be considered along with endoscopic assessment and clinical symptoms. A number of adult studies have demonstrated that improvements in laboratory markers are predictive of better clinical outcomes.19,20 The discrepancy between clinical symptoms and endoscopic appearance demonstrated here in pediatric CD is in contrast to the relationship seen in pediatric ulcerative colitis, where there is good correlation between the Pediatric Ulcerative Colitis Activity Index and endoscopic assessment (r Z .77, P < .001).21,22 This discrepancy is likely multifactorial but is potentially in part because of variability in disease location and extent in pediatric CD. Whereas pediatric ulcerative colitis is continuous and generally uniform throughout the colon extensively, pediatric CD patients have much more varied segmental disease. Colonic mucosal inflammation correlated more strongly with clinical disease score in our study, with the overall SES-CD and wPCDAI relationship showing the best correlation coefficient, (r Z .49, P < .001) and stooling being the variable driving the relationship (r Z .50, P < .001). Bloody diarrhea is a consistent symptom in pediatric ulcerative colitis, and symptoms related to stooling make up a large proportion of the Pediatric Ulcerative Colitis Activity Index. In addition, ulcer size (r Z .60, P < .001) was the endoscopic variable most contributory to the relationship with clinical score. Therefore, it appears that stooling frequency/bleeding is most reflective of mucosal ulceration in colonic inflammation. Given the importance of achieving mucosal remission and assessing endoscopic improvement, valid and responsive endoscopic tools are essential for disease activity assessment. However, no endoscopic tools have been validated in children. The SES-CD is the most frequently used Volume
-,
No.
-
: 2018 GASTROINTESTINAL ENDOSCOPY 7
Endoscopic severity evaluation in children with newly diagnosed CD
Carman et al
TABLE 2. Correlation between SES-CD and wPCDAI stratified by disease location SES-CD (Pearson’s coefficient [r]) All patients (n [ 280)
Ileocolonic (L3) (n [ 163)
Colonic (L2) (n [ 74)
Ileal (L1) (n [ 43)
.39*
.36*
.49*
.05
Pain
.16*
.15*
.20*
.07
Stooling
.41*
.33*
.50*
–.24
Function
.28
.21*
.38*
–.03
.16*
.18y
.25y
–.02
Outcome measure wPCDAI wPCDAI clinical
wPCDAI exam Weight Perianal disease
.05
.05
.08
.09
Extraintestinal manifestations
.18*
.18y
.24y
.12
wPCDAI labs ESR
.22*
.23*
.37*
.01
Albumin
–.24*
–.17y
–.37*
–.14
hsCRP
.31*
.46*
.67*
.21
CRP
.08*
–.03
.19
.15
SES-CD, Simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index; ESR, erythrocyte sedimentation rate; hsCRP, high-sensitivity C-reactive protein; CRP, C-reactive protein. *P < .01. yP < .05.
TABLE 3. Correlation among wPCDAI and SES-CD, endoscopic subscores, and biochemical parameters stratified by disease location wPCDAI (Pearson’s coefficient [r]) Outcome measure SES-CD
All patients (n [ 280)
Ileocolonic (L3) (n [ 163)
Colonic (L2) (n [ 74)
Ileal (L1) (n [ 43)
.39*
.36*
.49*
.05
SES-CD subscores and components Ulcerated surface
.38*
.38*
.45*
–.08
Ulcer size
.41*
.35*
.60*
–.21
Affected surface
.32*
.32*
.31*
.12
Narrowing
.04
–.04
.15
.28
Laboratory investigations ESR
.47*
.43*
.53*
.51*
Albumin
–.53*
–.55*
–.51*
–.55y
hsCRP
.56*
.60*
.63*
.42y
CRP
.37*
.28*
.53*
.35
SES-CD, Simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index; ESR, erythrocyte sedimentation rate; hsCRP, high-sensitivity Creactive protein; CRP, C-reactive protein. *P < .01. yP < .05.
endoscopic tool in pediatric CD, yet its utility in pediatrics has not been completely evaluated. The SES-CD was originally developed and validated by Daperno et al in 20043 using the Crohn’s Disease Endoscopic Index of Severity as the reference standard. As such, the SES-CD was never validated against a global assessment of disease severity. The patients enrolled in the development study were consecutive adult patients needing colonoscopy for CD, with no stratification of disease severity before ensuring a 8 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2018
spectrum of disease severity (ie, mild, moderate, severe). Total score is influenced heavily by number of segments affected, because the total score is calculated by the sum of the individual scores for each segment. Despite good reliability data for adult patients in the hands of IBD experts,3,15 reliability assessments have not been performed by pediatric gastroenterologists. The strengths of our study include its prospective, multicenter design with rigorous and standardized clinical workup www.giejournal.org
Carman et al
Endoscopic severity evaluation in children with newly diagnosed CD
and data collection across the network. Because most children with IBD in Canada are treated at the specialized pediatric centers comprising the CIDsCaNN network (ie, very few community-based pediatric gastroenterology practices exist that treat pediatric IBD patients), this study includes a diverse patient population, making our conclusions generalizable. In addition, patients in our study were treatment naïve, and as such the endoscopic appearance was not influenced by the use of medical therapies. However, our study has some potential limitations. There is limited information on intra- and interrater reliability of SES-CD scoring, especially between sites. To attempt to address this limitation, patients were only enrolled if the endoscopy was performed/scored by a pediatric IBD expert. Adult studies have shown agreement between IBD expert readers to be excellent.15 This increases confidence in the reliability of our endoscopic scores. Subsequent validation studies are underway examining the interobserver reliability of the SES-CD in CIDsCaNN patients. Previous studies looking at the relationship between scoring tools have been heterogeneous and included both treatment naïve and treated patients. A strength of our study is that the relationship is examined solely on treatment naïve patients. This does, however, lead to an inherent weakness in that we cannot comment on the relationship between the tools after treatment exposure or in the context of mucosal healing. The goals of clinical care in IBD should bring together factors most valuable to patients (improvement of clinical symptoms) with intestinal healing. This model of care is the cornerstone of the treat-to-target strategy and emphasized by the STRIDE program.2,23 To achieve this end, accurate assessment of overall disease activity is required. This study demonstrates that currently available composite clinical disease activity scores and commonly used blood tests correlate poorly with endoscopic assessment and therefore should not be used in isolation to assess global disease activity. Clinical assessment remains an essential aspect of routine care in IBD, representing the patient experience of their disease, so the use of appropriate clinical indices remains important. This should, however, be combined with objective mucosal assessment to monitor treatment outcomes, both in the clinic and in the context of clinical trials.
d’Etudes Therapeutiques des Affections Inflammatoires Digestives. Gut 1994;35:231-5. Levesque BG, Sandborn WJ, Ruel J, et al. Converging goals of treatment of inflammatory bowel disease from clinical trials and practice. Gastroenterology 2015;148:37-51. Daperno M, D'Haens G, Van Assche G, et al. Development and validation of a new, simplified endoscopic activity score for Crohn’s disease: the SES-CD. Gastrointest Endosc 2004;60:505-12. Benchimol EI, Manuel DG, Guttmann A, et al. Changing age demographics of inflammatory bowel disease in Ontario, Canada: a population-based cohort study of epidemiology trends. Inflamm Bowel Dis 2014;20:1761-9. Pigneur B, Seksik P, Viola S, et al. Natural history of Crohn’s disease: comparison between childhood- and adult-onset disease. Inflamm Bowel Dis 2010;16:953-61. Hyams JS, Ferry GD, Mandel FS, et al. Development and validation of a pediatric Crohn’s disease activity index. J Pediatr Gastroenterol Nutr 1991;12:439-47. Turner D, Griffiths AM, Walters TD, et al. Mathematical weighting of the pediatric Crohn’s disease activity index (PCDAI) and comparison with its other short versions. Inflamm Bowel Dis 2012;18:55-62. Levine A, Koletzko S, Turner D, 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. Levine A, Griffiths A, Markowitz J, et al. Pediatric modification of the Montreal classification for inflammatory bowel disease: the Paris classification. Inflamm Bowel Dis 2011;17:1314-21. Grover Z, Lewindon P. Predicting endoscopic Crohn’s disease activity before and after induction therapy in children: a comprehensive assessment of PCDAI, CRP, and fecal calprotectin. Inflamm Bowel Dis 2015;21:1386-91. Turner D, Levine A, Walters TD, et al. Which PCDAI version best reflects intestinal inflammation in pediatric Crohn disease? J Pediatr Gastroenterol Nutr 2017;64:254-60. Swinscow TD, Campbell MJ. Statistics at square one. Ninth Ed. London: BMJ Publishing group; 1997. Fleiss JL. Statistical methods for rates and proportions. 2nd ed. New York, NY: John Wiley and Sons; 1981. Allez M, Lemann M, Bonnet J, et al. Long term outcome of patients with active Crohn’s disease exhibiting extensive and deep ulcerations at colonoscopy. Am J Gastroenterol 2002;97:947-53. Khanna R, Zou G, D'Haens G, et al. Reliability among central readers in the evaluation of endoscopic findings from patients with Crohn’s disease. Gut 2016;65:1119-25. Sipponen T, Nuutinen H, Turunen U, et al. Endoscopic evaluation of Crohn’s disease activity: comparison of the CDEIS and the SES-CD. Inflamm Bowel Dis 2010;16:2131-6. Richardson G, Griffiths AM, Miller V, et al. Quality of life in inflammatory bowel disease: a cross-cultural comparison of English and Canadian children. J Pediatr Gastroenterol Nutr 2001;32:573-8. De Boer M, Grootenhuis M, Derkx B, et al. Health-related quality of life and psychosocial functioning of adolescents with inflammatory bowel disease. Inflamm Bowel Dis 2005;11:400-6. Schnitzler F, Fidder H, Ferrante M, et al. Long-term outcome of treatment with infliximab in 614 patients with Crohn’s disease: results from a single-centre cohort. Gut 2009;58:492-500. Kiss LS, Szamosi T, Molnar T, et al. Early clinical remission and normalisation of CRP are the strongest predictors of efficacy, mucosal healing and dose escalation during the first year of adalimumab therapy in Crohn’s disease. Aliment Pharmacol Ther 2011;34:911-22. Turner D, Otley AR, Mack D, et al. Development, validation, and evaluation of a pediatric ulcerative colitis activity index: a prospective multicenter study. Gastroenterology 2007;133:423-32. Turner D, Hyams J, Markowitz J, et al. Appraisal of the pediatric ulcerative colitis activity index (PUCAI). Inflamm Bowel Dis 2009;15:1218-23. Peyrin-Biroulet L, Sandborn W, Sands BE, et al. Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE): determining therapeutic goals for treat-to-target. Am J Gastroenterol 2015;110:1324-38.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12. 13. 14.
15.
16.
17.
18.
19.
ACKNOWLEDGMENT The Canadian Children Inflammatory Bowel Disease Network is a national collaboration funded by the generous support of the Ch.I.L.D. Foundation, Canada.
20.
21.
REFERENCES 1. Cellier C, Sahmoud T, Froguel E, et al. Correlations between clinical activity, endoscopic severity, and biological parameters in colonic or ileocolonic Crohn’s disease. A prospective multicentre study of 121 cases. The Groupe
www.giejournal.org
22. 23.
Volume
-,
No.
-
: 2018 GASTROINTESTINAL ENDOSCOPY 9
Clinical disease activity and endoscopic severity correlate poorly in children newly diagnosed with Crohn’s disease Nicholas Carman, MBBS,1 Diane Tomalty, MSc,2 Peter C. Church, MD,2 David R. Mack, MD, PhD,1 Eric I. Benchimol, MD, PhD,1 Anthony R. Otley, MD,3 Kevan Jacobson, MD,4 Hien Q. Huynh, MD,5 Jennifer C. DeBruyn, MD,6 Wael El-Matary, MD,7 Mary Sherlock, MD,8 Johan Van Limbergen, MD, PhD,3 Anne M. Griffiths, MD,2 Thomas D. Walters, MD2; on behalf of the Canadian Children Inflammatory Bowel Disease Network: A Joint Partnership of Canadian Institutes of Health Research and the Children with Intestinal and Liver Disorders Foundation Ottawa, Toronto, Halifax, Vancouver, Edmonton, Calgary, Winnipeg, Hamilton, Canada
Background and Aims: Treatment goals in Crohn’s disease (CD) have evolved to target mucosal healing. There is now a drive to determine if noninvasive measures can adequately identify the attainment and persistence of this goal. Currently, data describing the relationship between clinical indices and endoscopic appearance in pediatric CD are sparse. Our aim was to compare endoscopic severity with the weighted Pediatric Crohn’s Disease Activity Index (wPCDAI) in children with newly diagnosed CD. Methods: All children aged 17 years newly diagnosed with CD enrolled in an inception cohort at sites of the Canadian Children Inflammatory Bowel Disease Network were eligible. Clinical disease activity at presentation was evaluated by the wPCDAI and conventional biochemical parameters. Severity of disease at ileocolonoscopy was assessed by the simple endoscopic score for CD (SES-CD), with segmental subscores noted. We evaluated the association of SES-CD and disease activity markers using the Pearson test of correlation, the Spearman rank coefficient, and linear regression models. Results: Two hundred eighty patients from 11 centers were included in the analysis. The median wPCDAI score was 60 (interquartile range, 40-80; 53% severe). Median SES-CD was 16 (interquartile range 10-22; 51% severe). The wPCDAI correlated weakly with SES-CD (r Z .39, P < .001). Examination of the individual components that contribute to the wPCDAI demonstrated weak correlation with the SES-CD for all items apart from stooling (moderate correlation, r Z .50, P < .001). Routine blood tests did not correlate well with the SES-CD. In regression models, variation in clinical symptoms accounted for most of the variation in both the wPCDAI and SES-CD, with no additional benefit from routine blood tests. Conclusions: In children with newly diagnosed CD, wPCDAI correlates poorly with endoscopic disease activity. As treatment paradigms evolve to target mucosal healing, clinical markers should not be used in isolation to determine disease activity. (Gastrointest Endosc 2018;-:1-9.)
Abbreviations: CD, Crohn’s disease; CDAI, Crohn’s Disease Activity Index; CIDsCaNN, Canadian Children Inflammatory Bowel Disease Network; CRP, C-reactive protein; IBD, inflammatory bowel disease; PCDAI, Pediatric Crohn’s Disease Activity Index; SES-CD, simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index. DISCLOSURE: During fellowship training at The Hospital for Sick Children, funding support for N. Carman was provided by Abbvie Canada. All authors disclosed no financial relationships relevant to this publication. Copyright ª 2018 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 https://doi.org/10.1016/j.gie.2018.09.025 Received May 14, 2018. Accepted September 20, 2018.
www.giejournal.org
Current affiliations: Children’s Hospital of Eastern Ontario, Department of Pediatrics, University of Ottawa, Ottawa, Canada (1), The Hospital for Sick Children, Department of Pediatrics, University of Toronto, Toronto, Canada (2), IWK Health Centre, Department of Pediatrics, Dalhousie University, Halifax, Canada (3), BC Children’s Hospital, Department of Pediatrics, University of British Columbia, Vancouver, Canada (4), Stollery Children’s Hospital, Department of Pediatrics, University of Alberta, Edmonton, Canada (5), Alberta Children’s Hospital, Department of Pediatrics, University of Calgary, Calgary, Canada (6), Children’s Hospital Research Institute of Manitoba, Department of Pediatrics, University of Manitoba, Winnipeg, Canada (7), McMaster Children’s Hospital, Department of Pediatrics, McMaster University, Hamilton, Canada (8). Reprint requests: Nicholas Carman, MBBS, FRACP, Division of Gastroenterology, Hepatology and Nutrition, Children’s Hospital of Eastern Ontario, 401 Smyth Road, Ottawa, ON, K1H 8L1, Canada.
Volume
-,
No.
-
: 2018 GASTROINTESTINAL ENDOSCOPY 1
Endoscopic severity evaluation in children with newly diagnosed CD
Treatment goals for Crohn’s disease (CD) have advanced beyond control of symptoms and facilitation of normal growth; the aspiration now is intestinal healing. It is hoped that the attainment of such “deep remission” will prevent disease-related adverse events. Studies among adults with CD demonstrate the lack of correlation between disease activity, as measured by the Crohn’s Disease Activity Index (CDAI), and the endoscopic appearance of the ileocolonic mucosa.1 Accordingly, clinical trials evaluating new therapies in CD require objective assessment of inflammatory activity in the intestine.2 The simple endoscopic score for CD (SES-CD) is the multi-item measure most often used.3 CD develops during childhood in up to 20% of patients.4 Achievement of intestinal healing is particularly important in young patients with CD, given their long life ahead during which disease progression and repeated need for intestinal resection may occur despite achieving clinical remission based on symptomatology.5 In contrast to adult CD, however, studies assessing endoscopic severity of disease and correlation with clinical assessment of disease activity have been lacking. The Pediatric CD Activity Index (PCDAI) is the most commonly used pediatric multi-item measure of disease activity.6 The weighted PCDAI (wPCDAI) is its validated, mathematically weighted, short form.7 In comparison with the adult CDAI, both the PCDAI and wPCDAI rely less on subjectively reported symptoms and more on objective findings and laboratory investigations; they might therefore be expected to better reflect endoscopic inflammation. In the context of a pan-Canadian multicenter prospective inception cohort study, wherein the ileocolon is systematically evaluated, we aimed to explore the relationship before treatment between endoscopic activity and clinical disease severity using the SES-CD and the wPCDAI.
METHODS The network The Canadian Children Inflammatory Bowel Disease (IBD) Network: a Joint Partnership of the Canadian Institutes of Health Research and Children with Intestinal and Liver Disorders Foundation (CIDsCaNN), was formed in 2013 and now has participation from 12 academic pediatric IBD centers across Canada. All children and adolescents younger than 17 years presenting with new-onset CD or ulcerative colitis are eligible for enrollment in an inception cohort study. Baseline evaluation of suspected IBD is undertaken based on Porto diagnostic criteria8 and includes ileocolonoscopy and upper endoscopy with mucosal biopsy specimens for histologic assessment and small bowel imaging, most commonly using magnetic resonance enterography. Pretreatment demographic and phenotypic data are prospectively recorded, including for CD patients: PCDAI/ wPCDAI, physician global assessment of disease activity and SES-CD. All data, including routine laboratory parameters (complete blood count, C-reactive protein [CRP], serum al2 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2018
Carman et al
bumin, erythrocyte sedimentation rate) are captured using standardized case report forms. The diagnosis of CD is made at individual sites based on conventional criteria but confirmed centrally via central review of key clinical, endoscopic, and histologic reports. Macroscopic localization of CD is based on the Paris classification of pediatric IBD.9
Study patients Patients enrolled in the inception cohort with a diagnosis of CD involving the colon and/or terminal ileum (Paris classification L1 [ileal] or L2 [colonic] or L3 [ileocolonic] L4 [upper GI]) who had undergone complete ileocolonoscopy with visualization of the terminal ileum were included in the analysis. Patients with CD limited to the upper GI tract and/or small intestine proximal to the terminal ileum (Paris classification L4a and/or L4b) were excluded, as were patients with incomplete ileocolonoscopic examination or incomplete data. Additionally, patients were excluded if the baseline endoscopic and clinical assessments were more than 4 weeks apart or if treatment was started before the completion of both assessments.
Clinical and endoscopic activity evaluation: PCDAI and SES-CD Individual elements encompassing the clinical and endoscopic disease activity indices were collected after ileocolonoscopy or the clinic visit by the local site IBD physician on a standardized case report form and uploaded onto a REDCap database. These were subsequently downloaded by a single central physician (N.C.), who then calculated total scores for both the clinical and endoscopic tool in an attempt to avoid biasing the local physician’s scoring. This method of comparing clinical and endoscopic scores using local site readers, without central reads, is consistent with existing adult and pediatric literature on this topic.1,10,11 The PCDAI is a multi-item measure of inflammatory disease activity.6 Items include abdominal pain, number of stools per day, general well-being, hematocrit, erythrocyte sedimentation rate, albumin, weight, height velocity, abdominal tenderness, perirectal disease, and extraintestinal manifestations of disease. The total PCDAI score ranges from 0 to 100. The wPCDAI was derived by reweighting the PCDAI mathematically using data from 4 large prospective studies of children with CD, in which physician global assessment of inflammatory disease activity served as the dependent variable. Hematocrit and height velocity contributed little to the reweighting, and so these items were excluded. The wPCDAI is composed of 3 domains (clinical symptoms, physical examination, and laboratory variables), with individual items mathematically weighted to produce an overall score that classifies patients into 4 disease activity categories: none, <12.5; mild, 12.5 to 40; moderate, >40 to 57.5; and severe, >57.57 (Fig. 1). For the SES-CD, the intestine is divided into 5 anatomic segments: rectum, left side of the colon segment, transverse colon, right side of the colon segment, and ileum. Items of www.giejournal.org
Carman et al
Endoscopic severity evaluation in children with newly diagnosed CD
Clinical History – Recall, 1 week Abdominal Pain 0 = None
Score
10 = Mild: Brief, does not interfere with activities
20 = Moderate/Severe: Daily, longer lasting, affects activities, nocturnal
Patient Functioning, General Well Being 10 = Occasional difficulty in 0 = No limitation of maintaining age appropriate activities, well activities, below par Stools (per day) 0 = 0-1 liquid stools, no 7.5 = Up to 2 semi-formed blood with small blood, or 2-5 liquid
Score 20 = Frequent limitation of activity, very poor Score 15 = Gross bleeding, or >/= 6 liquid, or nocturnal diarrhea
Laboratory Erythrocyte Sedimentation Rate (ESR) 0 = <20 mm/hr 7.5 = 20-50 mm/hr
15 = >50 mm/hr
Albumin 0 = >/= 3.5 g/dl
20 =
Score
Score 10 = 3.1-3.4 g/dl
Examination Weight 0 = Weight gain or voluntary weight stable/loss Perirectal Disease 0 = None, asymptomatic tags
Score 5 = Involuntary weight stable, weight loss 1-9%
10 = Weight loss >/= 10%
Score
15 = Active fistula, drainage, 7.5 = 1-2 indolent fistula, tenderness, or abscess scant drainage, no tenderness Extra-intestinal Manifestations (fever >/= 38.5 for 3 days over the past week, definite arthritis, uveitis, E.nodosum, P. gangrenosum 0 = None 10 = One or more
Score
Total Score (0 – 125) Scoring 0 – <12.5 12.5 – 40 >40 – 57.5 >57.5
None/Remission Mild disease Moderate Disease Severe Disease Figure 1. Weighted Pediatric Crohn’s Disease Activity Index.
interest include ulcer size, estimates of the ulcerated and affected surface, and the presence of luminal narrowing. Categorical descriptions are recorded by the endoscopist for each of these items within each segment (Fig. 2). The SESCD score is calculated by combining scores of individual segments; hence, the maximum score possible in L1 disease is 12, 45 in L2, and 56 in L3. Overall SES-CD scores are classified into 4 disease categories: none, 0 to 2; mild, 2 to 6; moderate, 7 to 15; and severe, 16.3 CIDsCaNN uses a Central Phenotyping Committee to ensure the consistency of phenotypic characteristics of included patients. This committee is made up of experienced pediatric IBD physicians (A.M.G., J.C., H.Q.H.). Before www.giejournal.org
commencing patient recruitment to the Network, the site principle investigators reviewed and practiced using the endoscopy scoring tool at a face-to-face meeting. This exercise was repeated annually at subsequent face-to-face meetings.
Statistical analysis Laboratory values, indices, and their component subscores are presented as medians with interquartile range (IQR). Differences in the distributions when stratified by disease location were tested using analysis of variance and Mann-Whitney U tests as appropriate. A correlation matrix consisting of each index, its component subscores, and the Volume
-,
No.
-
: 2018 GASTROINTESTINAL ENDOSCOPY 3
Endoscopic severity evaluation in children with newly diagnosed CD
Carman et al
SES-CD Score 2 0.5 – 2cm
3 >2cm
<10%
10 – 30%
>30%
Unaffected Segment
<50%
50 – 75%
>75%
None
Single, can be passed
Multiple, can be Cannot be passed passed
Variable Size of Ulcers
0 None
1 0.1 – 0.5cm
Ulcerated Surface
None
Affected Surface
Presence of Narrowings
Total
Total score for this segment
Total Score = sum of score for each segment (Ileum + Right Colon + Transverse Colon + Left Colon + Rectum) Figure 2. The simple endoscopic score for Crohn’s disease (SES-CD).
various laboratory values was examined using the Pearson method or the Spearman rank coefficient as appropriate.12 The kappa statistic was used to determine the level of agreement between categorical subclassifications of each index.13 The Bowker test was used to determine if patterns of classification discordance were random. A series of linear regression models was used to explore the relative contribution of the various clinical and endoscopic subscores to the scoring of each activity index. The relative model fit and estimation of a variable’s relative contribution to any demonstrated association was assessed using the r2 statistic. A significance level of .05 was used throughout. All statistical analyses were performed using IBM SPSS Statistics version 22.0 for Windows software (IBM Corp, Armonk, NY).
Ethics This study protocol was reviewed and approved by the Research Ethics Board at each participating institution.
patients the time between clinical and endoscopic assessment was longer than 4 weeks. As such, 280 patients had complete baseline assessments within the designated time frame and were eligible for analysis. The demographic and phenotypic characteristics are reported in Table 1. The distributions of the wPCDAI and SES-CD scores within each Paris classification of the disease location are presented in Figure 3. The distributions of wPCDAI scores within each of the location subgroups were relatively similar, with 52% of patients having “severely active” disease at diagnosis (L1: median, 50; IQR, 32.5-67.5; L2: median, 55; IQR, 39-85; L3: median, 62.5; IQR, 42-80). Reflecting differences in the maximal SES-CD scores possible within each of L1, L2, and L3 disease, the median SES-CD for patients with isolated ileal (L1) disease (7; IQR, 4-9), was substantially lower than that for those with either isolated colonic (L2) disease (17; IQR, 1224) or ileocolonic (L3) disease (17; IQR, 13-23) (L1 vs L2, P < .001; L1 vs L3, P < .001; L2 vs L3, P Z .39).
Relationship between SES-CD and wPCDAI RESULTS Patients Seven hundred children and adolescents with CD were enrolled in the inception cohort study at the time of data extraction. Of these, 187 had an incomplete baseline wPCDAI (reflective of the unavailability of erythrocyte sedimentation rate in some centers), 74 had an incomplete endoscopic evaluation before treatment (eg, did not reach terminal ileum before initiation of treatment), and in 159 4 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2018
Despite relatively similar distributions, there was only a weak correlation between the SES-CD and wPCDAI score (r Z .39; P < .001) (Fig. 4). To ascertain if there was a differential relationship between the SES-CD and wPCDAI that was dependent on clinical disease severity, we subsequently categorized the subjects according to disease activity (none/mild, wPCDAI 0-40; moderate/severe, wPCDAI >40) and repeated the correlation analysis. This demonstrated a weaker relationship between clinical disease activity and endoscopic appearance in patients with clinically www.giejournal.org
Carman et al
Endoscopic severity evaluation in children with newly diagnosed CD
TABLE 1. Baseline patient characteristics stratified by disease location Paris Classification Baseline characteristics Male gender
All patients (n [ 280)
Ileocolonic (L3) (n [ 163)
Colonic (L2) (n [ 74)
Ileal (L1) (n [ 43)
153 (55)
84 (52)
42 (57)
27 (63)
13.4 (11.2-15)
13.4 (11.3-15)
12.8 (9.7-14.9)
14.2 (12.4-15.7)
3 (0-11)
2 (0-11)
3 (1-13)
4 (0-12)
SES-CD
16 (10-22)
17 (13-23)
17 (12-24)
7 (4-9)
wPCDAI
60 (40-80)
62.5 (42-80)
55 (39-85)
50 (32.5-67.5)
ESR
41 (22-61)
42 (25-61)
40 (18-63)
42 (23.7-59)
Age, y Days between endoscopic and clinical/biochemical assessments
Albumin
34 (29-38)
33 (29-37)
36 (30-40)
35 (31-39)
hsCRP
33 (10.1-70.3)
44.3 (16.7-79)
12.5 (2-43.5)
30 (12.7-82)
CRP
24.7 (12.1-59)
47/163: 30 (14.2-61)
25/74: 17.8 (7.6-63)
22/43: 19.3 (12-39)
Mild
63 (23)
33 (20)
15 (20)
15 (35)
Moderate
108 (38)
57 (35)
32 (43)
18 (42)
Severe
110 (39)
73 (45)
27 (37)
10 (23)
PGA
Values are n (%) or median (interquartile range). SES-CD, Simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index; ESR, erythrocyte sedimentation rate; hsCRP, high-sensitivity C-reactive protein; CRP, C-reactive protein; PGA, physician’s global assessment.
less-active disease (none/mild: r Z .08, P Z .52; moderate/ severe: r Z .33, P < .001). The wPCDAI was originally formatted to be an ordinal classifier of disease activity (none, mild, moderate, severe disease activity), and so we examined its relationship with the SES-CD using this structure. As illustrated in Figure 5, although a modest linear relationship exists (analysis of variance: F Z 15.4, P < .001), minimal variance is explained (R2 Z .05). The SES-CD also has previously validated cut-points that describe severity of mucosal disease (none, mild, moderate, severe mucosal disease). Overall agreement between the ordinal classifiers for each tool was poor (122/280 agreed, k Z .14, P Z .003). Notably, the pattern of discordance was not random (Bowker test P < .001) with mucosal severity classifications in the discordant cases systematically more severe than clinical activity classifications.
Correlation of the SES-CD with wPCDAI according to disease location In patients with isolated colonic (L2) disease, there was a better relationship between clinical and endoscopic activity, demonstrating moderate correlation between the 2 scores (r Z .49, P < .001). A weaker relationship was seen in patients with ileocolonic (L3) disease (r Z .36, P < .001) and no relationship in isolated ileal (L1) disease (r Z .05, P Z .97).
Correlation of laboratory measures with endoscopic and clinical scores There was a weak relationship overall between SES-CD and each of the laboratory measures (Table 2). When stratified by www.giejournal.org
disease location, there was a strong relationship between high-sensitivity CRP and SES-CD in colonic disease (r Z .67, P < .001) but not with erythrocyte sedimentation rate or albumin (r Z .37, P < .001 and r Z –.37, P < .001, respectively) (Table 2). In contrast, the wPCDAI showed a better relationship to all the laboratory measures (erythrocyte sedimentation rate: r Z .47, P < .001; albumin: r Z –.53, P < .001; high-sensitivity CRP: r Z .56, P < .001) (Table 3). This relationship persisted regardless of disease location.
Analysis of subscores Examination of each individual subscore that contributed to the wPCDAI are summarized in Table 2. Overall, all subscores only weakly correlated with the SES-CD score apart from stooling (r Z .41, P < .001). Notably, this relationship was only present in patients with colonic involvement (L2/L3), especially in isolated colonic disease where there was moderate correlation (r Z .50, P < .001). When examining the SESCD subscores (Table 3) and their relationship to the clinical score, ulcer size showed the best relationship overall with a moderate correlation (r Z .41, P < .001). Again, this relationship was strongest in isolated colonic disease (r Z .60, P < .001).
Regression models of SES-CD and wPCDAI We examined the wPCDAI and its subdomains to determine the overall contribution of each variable in the multiitem index to the overall score. The clinical components of the wPCDAI (stooling, pain, and functioning) accounted for 71% of the variance within the wPCDAI (R Z .84, P < .001). When the clinical components were examined Volume
-,
No.
-
: 2018 GASTROINTESTINAL ENDOSCOPY 5
Endoscopic severity evaluation in children with newly diagnosed CD
Carman et al
120
100 30 80
60
20
wPCDAI
Simple Endoscopic Score (SES-CD)
40
40 10 20
0
0
Ileal (L1)
Colonic (L2) Paris Classification
SES-CD
Ileocolonic (L3) wPCDAI
Figure 3. wPCDAI and SES-CD score stratified by disease location (n Z 280). SES-CD, Simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index.
in isolation, 80% of the variance within this subgroup could be attributed to pain and functioning. Indeed, when examined further it was determined that variation in these 2 items accounted for approximately 65% of the variance in the total wPCDAI scores (R Z .8, P < .001). We modeled the SES-CD score and explored its relative relationship to the various domains of the wPCDAI. When all 8 subcomponents that make up the wPCDAI were included as individual variables in a linear regression model of the SES-CD (3 symptom-based items, 3 examinationbased items, and 2 laboratory tests) 24% of the variance was explained (P < .001). This was somewhat improved from the relationship seen between the SES-CD and the composite wPCDAI score (R2 Z .15, P < .001). The addition of CRP to the model marginally improved the fit (R2 Z .24 vs R2 Z .32). The wPCDAI elements that contributed significantly to the model were the symptom-based items (stooling: b Z .5 [95% confidence interval, .34-.64], P < .001; function: b Z .14 [95% confidence interval, .06-.30], P Z .04). Abdominal pain was noncontributory. There was no appreciable benefit from the physical examination and laboratory items.
DISCUSSION Our data concerning the spectrum of severity of endoscopic disease in an inception cohort of children with 6 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2018
new-onset CD are unique. Endoscopic severity of disease at presentation, specifically the presence of deep ulcers, has been suggested, albeit in limited data accrued among adults, to be a risk factor for progressive disease and an indication of need for aggressive treatment.14 If pediatric gastroenterologists are to attempt to risk stratify and thereby individualize treatment regimens, greater attention to consistent description of endoscopic disease using validated measurement tools is required. To date, there are scarce data on the relationship between mucosal appearance at endoscopy and clinical disease activity scores in pediatric patients with CD. In this study, we demonstrate a weak correlation between the wPCDAI and endoscopic appearance at diagnosis as described by the SES-CD (r Z .39; P < .001). A small pediatric study (n Z 24) previously demonstrated a similarly poor correlation between the PCDAI and SES-CD at diagnosis (r Z .33, P > .05).10 A more recent study of 100 children with CD of varying duration that examined the relationship between SES-CD and clinical indices revealed a slightly better association (wPCDAI r Z .43, P < .001; PCDAI r Z .45, P < .001).11 However, 19 of these patients (19%) had longstanding mucosal healing; in patients with well-established intestinal healing one might anticipate a better relationship between the 2 indices. Our results are similar to data from adult CD studies.3,15,16 When examining the distributions of scoring across disease locations, clinical index scores were relatively consistent www.giejournal.org
Carman et al
Endoscopic severity evaluation in children with newly diagnosed CD
Simple Endoscopic Score (SES-CD)
120 100
wPCDAI
80 60 40 20 0
0
10
20
30
40
40
30
20
10
0 None/Mild
Simple Endoscopic Score (SES-CD)
Figure 4. Correlation between the SES-CD and wPCDAI (n Z 280) (r Z .39, P <.001). SES-CD, Simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index.
across isolated ileal, colonic, and ileocolonic disease. However, the SES-CD showed a distinct difference between ileal and colonic disease. The SES-CD is calculated by the sum of scores across involved segments. In cases of severe isolated ileal disease, the maximal score possible is 12, which only corresponds to moderate endoscopic severity. Therefore, defining disease severity in isolated ileal disease is difficult when using the SES-CD, raising the question of whether the same scoring system should continue to be used for both isolated ileal disease and more extensive disease. The 2 location subgroups often present with different symptomatology, have a different natural history, and may even require a different treatment paradigm. We examined the components of the wPCDAI and SESCD to explore the contributions of individual variables to overall score using a series of linear regression models. When examining the wPCDAI against its subscores, our regression model demonstrated that the variation in scoring is determined largely by symptoms and their effect, with reports of abdominal pain, stooling, and functional impairment together accounting for 70% of the variation in total score. These symptoms account for a significant amount of patient discomfort and disease experience and have a considerable impact on disease-related quality of life for the patient.17,18 The more objective variables in the wPCDAI (physical examination, laboratory tests) contributed minimally to the overall composite score. When we examined the relationship between wPCDAI variables and SES-CD, stooling showed the best relationship in patients with colonic involvement (r Z .50, P < .001). Correlation between laboratory variables and SES-CD was weak, but it was reasonable with the clinical activity score. In addition, CRP marginally improved the model fit when added to wPCDAI subscores in our linear regression analysis, but the overall relationship was still poor. This suggests that biochemical measures of inflammation www.giejournal.org
Moderate wPCDAI Categories
Severe
Figure 5. SES-CD scores stratified by wPCDAI (n Z 280). Distribution of SES-CD scores across wPCDAI categories assessed by analysis of variance (analysis of variance: F Z 15.0, P < .001, R2 Z .05). SES-CD, Simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index.
cannot necessarily be used in place of endoscopic assessment and are in fact a third axis of disease experience that needs to be considered along with endoscopic assessment and clinical symptoms. A number of adult studies have demonstrated that improvements in laboratory markers are predictive of better clinical outcomes.19,20 The discrepancy between clinical symptoms and endoscopic appearance demonstrated here in pediatric CD is in contrast to the relationship seen in pediatric ulcerative colitis, where there is good correlation between the Pediatric Ulcerative Colitis Activity Index and endoscopic assessment (r Z .77, P < .001).21,22 This discrepancy is likely multifactorial but is potentially in part because of variability in disease location and extent in pediatric CD. Whereas pediatric ulcerative colitis is continuous and generally uniform throughout the colon extensively, pediatric CD patients have much more varied segmental disease. Colonic mucosal inflammation correlated more strongly with clinical disease score in our study, with the overall SES-CD and wPCDAI relationship showing the best correlation coefficient, (r Z .49, P < .001) and stooling being the variable driving the relationship (r Z .50, P < .001). Bloody diarrhea is a consistent symptom in pediatric ulcerative colitis, and symptoms related to stooling make up a large proportion of the Pediatric Ulcerative Colitis Activity Index. In addition, ulcer size (r Z .60, P < .001) was the endoscopic variable most contributory to the relationship with clinical score. Therefore, it appears that stooling frequency/bleeding is most reflective of mucosal ulceration in colonic inflammation. Given the importance of achieving mucosal remission and assessing endoscopic improvement, valid and responsive endoscopic tools are essential for disease activity assessment. However, no endoscopic tools have been validated in children. The SES-CD is the most frequently used Volume
-,
No.
-
: 2018 GASTROINTESTINAL ENDOSCOPY 7
Endoscopic severity evaluation in children with newly diagnosed CD
Carman et al
TABLE 2. Correlation between SES-CD and wPCDAI stratified by disease location SES-CD (Pearson’s coefficient [r]) All patients (n [ 280)
Ileocolonic (L3) (n [ 163)
Colonic (L2) (n [ 74)
Ileal (L1) (n [ 43)
.39*
.36*
.49*
.05
Pain
.16*
.15*
.20*
.07
Stooling
.41*
.33*
.50*
–.24
Function
.28
.21*
.38*
–.03
.16*
.18y
.25y
–.02
Outcome measure wPCDAI wPCDAI clinical
wPCDAI exam Weight Perianal disease
.05
.05
.08
.09
Extraintestinal manifestations
.18*
.18y
.24y
.12
wPCDAI labs ESR
.22*
.23*
.37*
.01
Albumin
–.24*
–.17y
–.37*
–.14
hsCRP
.31*
.46*
.67*
.21
CRP
.08*
–.03
.19
.15
SES-CD, Simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index; ESR, erythrocyte sedimentation rate; hsCRP, high-sensitivity C-reactive protein; CRP, C-reactive protein. *P < .01. yP < .05.
TABLE 3. Correlation among wPCDAI and SES-CD, endoscopic subscores, and biochemical parameters stratified by disease location wPCDAI (Pearson’s coefficient [r]) Outcome measure SES-CD
All patients (n [ 280)
Ileocolonic (L3) (n [ 163)
Colonic (L2) (n [ 74)
Ileal (L1) (n [ 43)
.39*
.36*
.49*
.05
SES-CD subscores and components Ulcerated surface
.38*
.38*
.45*
–.08
Ulcer size
.41*
.35*
.60*
–.21
Affected surface
.32*
.32*
.31*
.12
Narrowing
.04
–.04
.15
.28
Laboratory investigations ESR
.47*
.43*
.53*
.51*
Albumin
–.53*
–.55*
–.51*
–.55y
hsCRP
.56*
.60*
.63*
.42y
CRP
.37*
.28*
.53*
.35
SES-CD, Simple endoscopic score for Crohn’s disease; wPCDAI, weighted Pediatric Crohn’s Disease Activity Index; ESR, erythrocyte sedimentation rate; hsCRP, high-sensitivity Creactive protein; CRP, C-reactive protein. *P < .01. yP < .05.
endoscopic tool in pediatric CD, yet its utility in pediatrics has not been completely evaluated. The SES-CD was originally developed and validated by Daperno et al in 20043 using the Crohn’s Disease Endoscopic Index of Severity as the reference standard. As such, the SES-CD was never validated against a global assessment of disease severity. The patients enrolled in the development study were consecutive adult patients needing colonoscopy for CD, with no stratification of disease severity before ensuring a 8 GASTROINTESTINAL ENDOSCOPY Volume
-,
No.
-
: 2018
spectrum of disease severity (ie, mild, moderate, severe). Total score is influenced heavily by number of segments affected, because the total score is calculated by the sum of the individual scores for each segment. Despite good reliability data for adult patients in the hands of IBD experts,3,15 reliability assessments have not been performed by pediatric gastroenterologists. The strengths of our study include its prospective, multicenter design with rigorous and standardized clinical workup www.giejournal.org
Carman et al
Endoscopic severity evaluation in children with newly diagnosed CD
and data collection across the network. Because most children with IBD in Canada are treated at the specialized pediatric centers comprising the CIDsCaNN network (ie, very few community-based pediatric gastroenterology practices exist that treat pediatric IBD patients), this study includes a diverse patient population, making our conclusions generalizable. In addition, patients in our study were treatment naïve, and as such the endoscopic appearance was not influenced by the use of medical therapies. However, our study has some potential limitations. There is limited information on intra- and interrater reliability of SES-CD scoring, especially between sites. To attempt to address this limitation, patients were only enrolled if the endoscopy was performed/scored by a pediatric IBD expert. Adult studies have shown agreement between IBD expert readers to be excellent.15 This increases confidence in the reliability of our endoscopic scores. Subsequent validation studies are underway examining the interobserver reliability of the SES-CD in CIDsCaNN patients. Previous studies looking at the relationship between scoring tools have been heterogeneous and included both treatment naïve and treated patients. A strength of our study is that the relationship is examined solely on treatment naïve patients. This does, however, lead to an inherent weakness in that we cannot comment on the relationship between the tools after treatment exposure or in the context of mucosal healing. The goals of clinical care in IBD should bring together factors most valuable to patients (improvement of clinical symptoms) with intestinal healing. This model of care is the cornerstone of the treat-to-target strategy and emphasized by the STRIDE program.2,23 To achieve this end, accurate assessment of overall disease activity is required. This study demonstrates that currently available composite clinical disease activity scores and commonly used blood tests correlate poorly with endoscopic assessment and therefore should not be used in isolation to assess global disease activity. Clinical assessment remains an essential aspect of routine care in IBD, representing the patient experience of their disease, so the use of appropriate clinical indices remains important. This should, however, be combined with objective mucosal assessment to monitor treatment outcomes, both in the clinic and in the context of clinical trials.
d’Etudes Therapeutiques des Affections Inflammatoires Digestives. Gut 1994;35:231-5. Levesque BG, Sandborn WJ, Ruel J, et al. Converging goals of treatment of inflammatory bowel disease from clinical trials and practice. Gastroenterology 2015;148:37-51. Daperno M, D'Haens G, Van Assche G, et al. Development and validation of a new, simplified endoscopic activity score for Crohn’s disease: the SES-CD. Gastrointest Endosc 2004;60:505-12. Benchimol EI, Manuel DG, Guttmann A, et al. Changing age demographics of inflammatory bowel disease in Ontario, Canada: a population-based cohort study of epidemiology trends. Inflamm Bowel Dis 2014;20:1761-9. Pigneur B, Seksik P, Viola S, et al. Natural history of Crohn’s disease: comparison between childhood- and adult-onset disease. Inflamm Bowel Dis 2010;16:953-61. Hyams JS, Ferry GD, Mandel FS, et al. Development and validation of a pediatric Crohn’s disease activity index. J Pediatr Gastroenterol Nutr 1991;12:439-47. Turner D, Griffiths AM, Walters TD, et al. Mathematical weighting of the pediatric Crohn’s disease activity index (PCDAI) and comparison with its other short versions. Inflamm Bowel Dis 2012;18:55-62. Levine A, Koletzko S, Turner D, 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. Levine A, Griffiths A, Markowitz J, et al. Pediatric modification of the Montreal classification for inflammatory bowel disease: the Paris classification. Inflamm Bowel Dis 2011;17:1314-21. Grover Z, Lewindon P. Predicting endoscopic Crohn’s disease activity before and after induction therapy in children: a comprehensive assessment of PCDAI, CRP, and fecal calprotectin. Inflamm Bowel Dis 2015;21:1386-91. Turner D, Levine A, Walters TD, et al. Which PCDAI version best reflects intestinal inflammation in pediatric Crohn disease? J Pediatr Gastroenterol Nutr 2017;64:254-60. Swinscow TD, Campbell MJ. Statistics at square one. Ninth Ed. London: BMJ Publishing group; 1997. Fleiss JL. Statistical methods for rates and proportions. 2nd ed. New York, NY: John Wiley and Sons; 1981. Allez M, Lemann M, Bonnet J, et al. Long term outcome of patients with active Crohn’s disease exhibiting extensive and deep ulcerations at colonoscopy. Am J Gastroenterol 2002;97:947-53. Khanna R, Zou G, D'Haens G, et al. Reliability among central readers in the evaluation of endoscopic findings from patients with Crohn’s disease. Gut 2016;65:1119-25. Sipponen T, Nuutinen H, Turunen U, et al. Endoscopic evaluation of Crohn’s disease activity: comparison of the CDEIS and the SES-CD. Inflamm Bowel Dis 2010;16:2131-6. Richardson G, Griffiths AM, Miller V, et al. Quality of life in inflammatory bowel disease: a cross-cultural comparison of English and Canadian children. J Pediatr Gastroenterol Nutr 2001;32:573-8. De Boer M, Grootenhuis M, Derkx B, et al. Health-related quality of life and psychosocial functioning of adolescents with inflammatory bowel disease. Inflamm Bowel Dis 2005;11:400-6. Schnitzler F, Fidder H, Ferrante M, et al. Long-term outcome of treatment with infliximab in 614 patients with Crohn’s disease: results from a single-centre cohort. Gut 2009;58:492-500. Kiss LS, Szamosi T, Molnar T, et al. Early clinical remission and normalisation of CRP are the strongest predictors of efficacy, mucosal healing and dose escalation during the first year of adalimumab therapy in Crohn’s disease. Aliment Pharmacol Ther 2011;34:911-22. Turner D, Otley AR, Mack D, et al. Development, validation, and evaluation of a pediatric ulcerative colitis activity index: a prospective multicenter study. Gastroenterology 2007;133:423-32. Turner D, Hyams J, Markowitz J, et al. Appraisal of the pediatric ulcerative colitis activity index (PUCAI). Inflamm Bowel Dis 2009;15:1218-23. Peyrin-Biroulet L, Sandborn W, Sands BE, et al. Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE): determining therapeutic goals for treat-to-target. Am J Gastroenterol 2015;110:1324-38.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12. 13. 14.
15.
16.
17.
18.
19.
ACKNOWLEDGMENT The Canadian Children Inflammatory Bowel Disease Network is a national collaboration funded by the generous support of the Ch.I.L.D. Foundation, Canada.
20.
21.
REFERENCES 1. Cellier C, Sahmoud T, Froguel E, et al. Correlations between clinical activity, endoscopic severity, and biological parameters in colonic or ileocolonic Crohn’s disease. A prospective multicentre study of 121 cases. The Groupe
www.giejournal.org
22. 23.
Volume
-,
No.
-
: 2018 GASTROINTESTINAL ENDOSCOPY 9