Faecal calprotectin as reliable non-invasive marker to assess the severity of mucosal inflammation in children with inflammatory bowel disease

Results Faecal calprotectin showed a high correlation ( r = 0.655) with the histologic grade of mucosal inflammation, similar to that observed for endoscopy ( r = 0.699), and it resulted the most accurate tool (sensitivity 94%, specificity 64%, positive predictive value 81%, negative predictive value 87%) to detect the presence of active mucosal inflammation when compared to clinical scores and common serum markers. In patients with apparent clinical and laboratory remission the accuracy of faecal calprotectin resulted further improved (sensitivity 100%, specificity 80%, positive predictive value 67%, negative predictive value 100%). Conclusions A more accurate assessment of the severity of mucosal inflammation can be achieved by the determination of faecal calprotectin levels compared to other common clinical and laboratory indices. This non-invasive and objective method could be particular useful in patients with apparent clinical and laboratory remission. Keywords Crohn's disease Disease activity Endoscopy Microscopic inflammation Ulcerative colitis 1 Introduction The two major forms of inflammatory bowel disease (IBD), i.e. Crohn's disease (CD) and ulcerative colitis (UC), are idiopathic, lifelong, chronic intestinal inflammatory conditions characterized by periods of remission and recurrent relapses [1] . In the past decades, the increasing paediatric incidence and prevalence of IBD has determined a substantial growth for health care and social costs related to these conditions [2,3] . Many authors have recently argued that the optimal therapeutic target to modify the disease course in patients affected by IBD should not only to be the clinical remission, but also the full mucosa healing [4,5] . In this light, an accurate evaluation of the severity of mucosal inflammation is important to optimize the treatment on an individual patient basis [4] . The gold standard to asses intestinal inflammation is the endoscopy with biopsy [6] . Unfortunately, endoscopy is invasive, uncomfortable, expensive and requires frequently general anaesthesia in children. Non-invasive objective tools have been proposed in the years, from faecal excretion of technetium-99-labeled ( 99 Tc) white cell scanning to the 18 F-fluorodeoxy glucose (FDG) positron emission tomography (PET) with computerized tomography (CT) [7,8] , but these tests are expensive, not free from risks and cannot be repeated frequently in children. For these reasons, in paediatric gastroenterology practice, clinical scores and common serum markers of inflammation are largely used. The erythrocyte sedimentation rate (ESR) and serum concentration of C-reactive protein (CRP) are the most used laboratory surrogate markers for disease activity in IBD [9,10] , but their accuracy to identify active mucosal inflammation is still debated [11,12] . Recently, we have demonstrated that faecal calprotectin (FC) is a sensitive and non-invasive marker to detect active inflammation in the gastrointestinal tract of paediatric patients [13] , and that the determination of FC values is useful, combined with other non-invasive tools, in the initial diagnostic work-up of children with suspected IBD [14] . Calprotectin is a cytosolic protein of granulocytes with immunomodulatory, antimicrobial and anti-proliferative activities [13–16] . A positive correlation between FC values and excretion of 111 In-labeled neutrophils has been demonstrated by Roseth and co-workers [17] . This group also demonstrated a correlation between FC values and severity of inflammation assessed by colonoscopy in adult IBD patients [18,19] . Concerning the paediatric field, consistent with the findings of others [20,21] , we have demonstrated that children with active IBD, defined by clinical, endoscopic and histological parameters, showed significantly higher FC values compared with those obtained in healthy children or in subjects with IBD in remission [14] . Data on the possible correlation between FC values and the histologic and endoscopic grading of intestinal mucosa inflammation in paediatric age are sparse. Recently, Fageberger et al. [22] have reported a correlation between FC values and the microscopic extent and severity of colonic inflammation in children affected by IBD, but a comprehensive study comparing FC with other tools is still needed for paediatric age. In this study, we aimed to comparatively evaluate the accuracy of FC, clinical scores, serum markers and endoscopy in the assessment of intestinal mucosa inflammation in children affected by IBD. 2 Patients and methods All children (1–16-year-old) with a previously established diagnosis of IBD, consecutively referred at our Paediatric Gastroenterology Tertiary Center because of a planned endoscopy, during a 12-month period starting from January 2005, were considered eligible for the study. The diagnosis of IBD was confirmed in each case according to the presence of accepted clinical, radiographic, endoscopic and histopathological criteria [14] . Exclusion criteria were: infections (within 1 month), malignancy (current), primary immunodeficiency, cystic fibrosis, underlying chronic disease, menstrual or nasal bleeding (within 3 weeks), trauma or surgery (within 1 month), and incomplete diagnostic work-up. All medications for IBD taken within the 8 weeks before enrolment were recorded. In the 24 h preceding bowel cleaning for endoscopy, all the enrolled patients underwent a full clinical and laboratory assessment. Clinical disease activity was measured using the paediatric Crohn's disease activity index (PCDAI) in children with CD [23] , and with the Rachmilewitz index in patients affected by UC [24] . A blood sample was taken from each patient for estimation of ESR, and of CRP. At the same time, a stool sample was collected and stored at −20 °C for FC levels determination. The FC levels were measured by a previously described highly sensitive enzyme-linked immunosorbent assay (ELISA, Calprest ® , Eurospital, Trieste, Italy) [13,14,25] , and the results were expressed as microgram per gram of faeces. All children underwent endoscopy with histological examination of intestinal biopsy specimens. The macroscopic grading of gut inflammation was assessed by two experienced observers who were blind to clinical and laboratory data, including FC levels. During endoscopy at least two biopsy specimens were taken from each segment for the histological examination. The biopsies were performed at five standard sites: “rectum and sigmoid”, “descending colon”, “transverse colon”, “ascending colon and caecum”, and “terminal ileum”. The extension of inflammation was scored from 1 to 5 considering the sum of the number of intestinal standard sites involved according to histological examination. The biopsies were evaluated by one experienced pathologist who was unaware of any relevant clinical information concerning the patients. The severity of mucosal inflammation was defined selecting the most inflamed segment according to previously validated endoscopic and histologic scores reported in Tables 1 and 2 [26] . According to the microscopic evaluation of bioptic specimens the patients were considered affected by active disease when the histologic score was >1 ( Table 2 ). The study protocol was approved by the Ethics Committee of our Institution. Written informed consent was obtained from the parents of all subjects. The research was not sponsored by any company. 2.1 Statistical analysis For categorical variables the Pearson chi-square test was performed, unless the exact method was required for frequency tables when more than 20% of the expected values were less than 5. Continuous variables were expressed as mean and 95% confidence interval (CI). Standard cut off values for PCDAI (>10), Rachmilewitz score (>3), ESR (>20 mm), CRP (>1 mg/dL), and for endoscopic score (>1) were used. The cut off values for ESR and CRP were defined as outside the reference range at our institution's laboratory. The FC optimal cut off value (i.e. the maximal sum of sensitivity and specificity) was achieved by a receiver operating characteristics (ROC) curve calculation. Spearman rank test was used to define coefficient of correlation ( r ). Linear regression analysis was used to study the possible influence of such variable on the FC concentrations. All tests of significance were two-sided. A p value of 0.05 or less was considered significant. The statistical analysis was performed using the SPSS software package for Windows (release 14.0.2, Chicago, IL, USA) and StatsDirect (release 2.5.6). 3 Results During the study period 64 children with IBD were considered eligible. Four patients were excluded because of incomplete diagnostic work-up, and two patients were excluded because the consent was withdrawn. Data of 58 patients with IBD were analysed. Main demographic and clinical characteristics of the study population are shown in Table 3 . According to the histologic grading of intestinal inflammation the patients were grouped in two categories: remission ( n = 22) and active disease ( n = 36). In Table 4 are reported the results of clinical, laboratory inflammatory index, endoscopic and histologic data obtained in the two groups. The ROC curve revealed an optimal cut-off FC level to discriminate between remission and active disease at 143 μg/g faeces ( Fig. 1 ). This value was used to comparatively evaluate the accuracy of FC, clinical scores, serum markers and endoscopy in the assessment of intestinal mucosa inflammation in children affected by IBD ( Table 5 ). When we considered only patients with apparent clinical and laboratory remission, the accuracy of FC to identify active disease resulted further improved (sensitivity 100%, 95% CI 40–100; specificity 80%, 95% CI 44–97; positive predictive value 67%, 95% CI 23–96; negative predictive value 100%, 95% CI 63–100; likelihood ratio for positive test 5.0, 95% CI 1.3–16.0). The correlations between clinical scores, ESR, CRP, FC, endoscopic score and the histologic grading of mucosal inflammation, for CD and UC patients separately and overall considered, are reported in Table 6 . A significant correlation between clinical score, ESR values and histologic grading was observed only in patients affected by CD. The FC significantly correlated with clinical score ( r = 0.330), ESR ( r = 0.388) and macroscopic severity score ( r = 0.460). The linear regression analysis showed that FC concentration was not affected by treatment, disease duration, site and extension of inflammation, and that it was influenced only by the severity of histologic grading of inflammation (Beta 0.655, p < 0.0001). 4 Discussion In children with IBD to distinguish between remission or active disease and to assess the severity of mucosal inflammation could be sometime difficult and may results in therapeutic errors and overuse of invasive diagnostic procedures [14] . The PCDAI and the Rachmilewitz index, and the more recent Paediatric Ulcerative Colitis Activity Index (PUCAI) [27] , are the most used scoring systems in clinical practice and in trial setting to assess disease activity and response to treatment. These indices are largely influenced by patient's report of symptoms and physical signs, and several studies suggested that these scores poorly correlate with the degree of mucosal inflammation [11,12] . Our results confirm that this is particular true for children with UC. On the contrary, we found a significant correlation between PCDAI and microscopic grading of mucosal inflammation that was very similar to that obtained for FC. We believe that this difference could be related to the inclusion of objective laboratory parameters in the PCDAI. In our study FC resulted the most accurate tool to assess the presence of active mucosal inflammation when compared to clinical scores and common serum markers, with a high correlation with the histologic grading of mucosal inflammation, similar to that observed for endoscopy. In patients with apparent clinical and laboratory remission the accuracy of FC resulted further improved. These findings could represent an important advantage in the clinical practice to detect the presence of active mucosal inflammation in an asymptomatic patient. In addition, FC also appeared to have a better accuracy in paediatric population when compared to results obtained in adult patients [28] . This potential diagnostic advantage in paediatric age necessitates further investigations. Our data support the concept that the use of common laboratory indicators of inflammation (ESR and CRP) could not be accurate predictors of active intestinal inflammation in children with IBD. Remarkable heterogeneity in the CRP and ESR accuracy to assess mucosal inflammation in IBD has been already reported, with some differences between CD and UC affected adult patients [29] . This was also observed in our paediatric population and could be explained, at least in part, by different cytokines production [9] . The ROC curve revealed an optimal cut-off FC level, to discriminate between remission and active disease, at 143 μg/g faeces. This cut-off value resulted higher when compared with the cut-off level defined in previous studies aimed to identify children with active organic intestinal diseases [13,22] . We believe that this apparent discrepancy could be related to the different patient populations and study designs. We have previously indicated an optimal cut-off value of 102.9 μg/g faeces to distinguish patients with active organic/inflammatory disorders from healthy subjects and from patients with intestinal functional disorders [13] . These findings suggest a possible different interpretation of FC values for the diagnosis and for the follow up of children with IBD. Our results suggest that the major factor influencing FC values is the severity instead the site or the extent of mucosal inflammation. This data are in apparent contrast with previous results from similar studies performed in paediatric patients [15,18,22] . These differences could be due to the small number of children evaluated in these studies and to the fact that macroscopic inflammation was defined by a combination of extent and severity scores. Although our investigation was not focused on value of FC in predict disease relapse as previously showed in adult patients, we believe that our results support the routinely use of FC in the clinical practice [30] . The FC could be measured at intervals during follow-up, which may allow to monitor the response to therapy and the early detection of relapses limiting the use of more invasive and expensive investigations. More recently, a rapid bedside test using monoclonal antibodies have been developed [31] . This test could open new perspectives to monitor disease activity in IBD children also at home. The FC is stable in the stools up to 7 days at room temperature [13–15] . Thus, faecal samples can be easily collected at home (only a small 1–2 g spot stool sample is required), mailed to the laboratory and analysed with a simple ELISA method for a quantitative evaluation, at a low cost [14] . Certainly, FC cannot replace invasive tests which will always be necessary to obtain tissue samples, to know disease distribution and complications. However, FC appears as an excellent alternative to assess inflammation severity. Recently, it has been proposed that a combination of the stool markers with serum markers and a disease-specific activity index in a categorical comprehensive activity index can increase the diagnostic accuracy with reference to the endoscopic inflammation in adult with UC [32] . Further study is necessary to investigate whether this test could be included in a revised version of clinical IBD activity index in adjunction to traditional markers in the management of children affected by CD and UC. Practice points • Monitoring mucosal inflammation is important for an effective management of patients with IBD. • Faecal calprotectin is an accurate, non-invasive and simple method to assess the severity of mucosal inflammation in children with IBD. • Faecal calprotectin is the most accurate tool to assess the presence of active mucosal inflammation when compared to clinical scores and common serum markers, with a high correlation with the histologic grading of mucosal inflammation, similar to that observed for endoscopy. • The measurement of faecal calprotectin levels could be useful in the follow up of children with IBD. Research agenda • To investigate whether faecal calprotectin is able to predict relapse in children with IBD. • To prospectively evaluate the effect of different therapeutic approaches on faecal calprotectin values in paediatric patients. • To determine the accuracy of new rapid bedside test to measure faecal calprotectin levels in children affected by IBD. Conflict of interest statement None declared. References [1] J.C. Escher J.A. Taminiau E.E.S. Nieuwenhuis H.A. Buller R.J. Grand Treatment of inflammatory bowel disease in childhood: best available evidence Inflamm Bowel Dis 9 2003 34 58 [2] H. Hildebrand Y. Finkel L. Grahnquist J. Lindholm A. Ekbom J. Askling Changing pattern of paediatric inflammatory bowel disease in northern Stockholm 1990–2001 Gut 52 2003 1432 1434 [3] S. Kugathasan R.H. Judd R.G. Hoffmann J. Heikenen G. Telega F. Khan Epidemiologic and clinical characteristics of children with newly diagnosed inflammatory bowel disease in Wisconsin: a statewide population-based study J Pediatr 143 2003 525 531 [4] I.D. Arnott D. Watts S. Ghosh Review article: is clinical remission the optimum therapeutic goal in the treatment of Crohn's disease? Aliment Pharmacol Ther 16 2002 857 867 [5] R. Berni Canani G. Terrin O. Borrelli M.T. Romano F. Manguso A. Coruzzo Short and long-term therapeutic efficacy of nutritional therapy and corticosteroids in paediatric Crohn's disease Dig Liver Dis 38 2006 381 387 [6] D.S. Fefferman R.J. Farrell Endoscopy in inflammatory bowel disease: indications, surveillance, and use in clinical practice Clin Gastroenterol Hepatol 3 2005 11 24 [7] M.A. Del Rosario J.F. Fitzgerald A.R. Siddiqui S.K. Chong J.M. Croffie S.K. Gupta Clinical applications of technetium Tc 99m hexamethyl propylene amine oxime leukocyte scan in children with inflammatory bowel disease J Pediatr Gastroenterol Nutr 28 1999 63 70 [8] E. Louis G. Ancion A. Colard V. Spote J. Belaiche R. Hustinx Noninvasive assessment of Crohn's disease intestinal lesion with 18 F-FDG PET/CT J Nucl Med 48 2007 1053 1059 [9] C.A. Solem E.V. Loftus Jr. W.J. Tremaine W.S. Harmsen A.R. Zinsmeister W.J. Sandborn Correlation of C-reactive protein with clinical, endoscopic, histologic, and radiographic activity in inflammatory bowel disease Inflamm Bowel Dis 11 2005 707 712 [10] S. Karoui S. Ouerdiane M. Serghini T. Jommi L. Kallel M. Fekih Correlation between levels of C-reactive protein and clinical activity in Crohn's disease Dig Liv Dis 39 2007 1006 1010 [11] L. Holmquist C. Ahrén S.P. Fällström Clinical disease activity and inflammatory activity in the rectum in relation to mucosal inflammation assessed by colonoscopy Acta Paediatr Scand 79 1990 527 534 [12] C. Cellier T. Sahmoud E. Froguel A. Adenis J. Belaiche J.F. Bretagne The Groupe d’Etudes Thérapeutiques des Affections Inflammatoires Digestives Correlations between clinical activity, endoscopic severity, and biological parameters in colonic or ileocolonic Crohn's disease. A prospective multicentre study of 121 cases Gut 35 1994 231 235 [13] R. Berni Canani L. Rapacciuolo M.T. Romano L. Tanturri de Horatio G. Terrin F. Manguso Diagnostic value of faecal calprotectin in pediatric gastroenterology clinical practice Dig Liver Dis 36 2004 467 470 [14] R. Berni Canani L. Tanturri de Horatio G. Terrin M.T. Romano E. Miele A. Staiano The combined use of non-invasive tests is useful in the initial diagnostic approach to a child with suspected inflammatory bowel disease J Pediatr Gastroenterol Nutr 42 2006 9 15 [15] S.K. Bunn W.M. Bisset M.J.C. Main E.S. Gray S. Olson B.E. Golden Fecal calprotectin: validation as a noninvasive measure of bowel inflammation in childhood inflammatory bowel disease J Pediatr Gastroenterol Nutr 33 2001 14 21 [16] A.C. von Roon L. Karamountzos S. Purkayastha G.E. Reese A.W. Darzi J.P. Teare Diagnostic precision of fecal calprotectin for inflammatory bowel disease and colorectal malignancy Am J Gastroenterol 202 2007 803 813 [17] A.G. Roseth P.N. Schmidt M.K. Fagerhol Correlation between faecal excretion of Indium-111 labelled granulocytes and calprotectin, a granulocyte marker protein, in patients with inflammatory bowel disease Scand J Gastroenterol 34 1999 50 54 [18] A.G. Roseth E. Aadland J. Jahnsen N. Raknerud Assessment of disease activity in ulcerative colitis by faecal calprotectin, a novel granulocyte marker protein Digestion 58 1997 176 180 [19] A.G. Roseth E. Aadland K. Grzyb Normalization of faecal calprotectin: a predictor of mucosal healing in patients with inflammatory bowel disease Scand J Gastroenterol 39 2004 1017 1020 [20] E. Olafsdottir L. Aksnes G. Fluge A. Berstad Faecal calprotectin levels in infants with infantile colic, healthy infants, children with inflammatory bowel disease, children with recurrent abdominal pain and healthy children Acta Pediatr 91 2002 45 50 [21] S.K. Bunn W.M. Bisset M.J.C. Main B.E. Golden Fecal calprotectin as a measure of disease activity in childhood inflammatory bowel disease J Pediatr Gastroenterol Nutr 32 2001 171 177 [22] U.L. Fagerberger L. Loof J. Lindholm L.O. Hansson Y. Finkel Fecal calprotectin: a quantitative marker of colonic inflammation in children with inflammatory bowel disease J Pediatr Gastroenterol Nutr 45 2007 414 420 [23] J.S. Hyams G.D. Ferry F.S. Mandel J.D. Gryboski P.M. Kibort B.S. Kirschner Development and validation of a pediatric Crohn's disease activity index J Pediatr Gastroenterol Nutr 12 1991 439 447 [24] D. Rachmilewitz Coated mesalazine (5-aminosalicylic acid) versus sulphasalazine in the treatment of active ulcerative colitis: a randomised trial Br Med J 298 1989 82 86 [25] F. Costa M.G. Mumolo M. Bellini M.R. Romano L. Ceccarelli P. Arpe Role of faecal calprotectin as non-invasive marker of intestinal inflammation Dig Liver Dis 35 2003 642 647 [26] S.H. Saverymuttu M. Camilleri H. Rees J.P. Lavender H.J.F. Hodgson V.S. Chadwick Indium 111-granulocyte scanning in the assessment of disease extent and disease activity in inflammatory bowel disease: I. A comparison with colonscopy, histology, and fecal Indium 111- granulocyte excretion Gastroenterology 90 1986 1121 1128 [27] D. Turner A.R. Otley D. Mack J. Hyams J. de Bruijne K. Uusoue Development, validation, and evaluation of a pediatric ulcerative colitis activity index: a prospective multicenter study Gastroenterology 133 2007 423 432 [28] M.R. Konikoff L.A. Denson Role of fecal calprotectin as biomarker of intestinal inflammation in inflammatory bowel disease Inflamm Bowel Dis 12 2006 524 534 [29] L. Zilberman N. Maharshak Y. Arbel O. Rogowski M. Rozenblat I. Shapira Correlated expression of high-sensitivity C-reactive protein in relation to disease activity in inflammatory bowel disease: lack of differences between Crohn's disease and ulcerative colitis Digestion 73 2006 205 209 [30] J.A. Tibble G. Sigthorsson S. Bridger M.K. Fagerhol I. Bjarnason Surrogate markers of intestinal inflammation are predictive of relapse in patients with inflammatory bowel disease Gastroenterology 119 2000 15 22 [31] Y. Shastri N. Povse J. Stein Comparative study of new bedside fecal calprotectin test with an established ELISA to assess intestinal inflammation in a prospective study Gastroenterology 130 Suppl. 2 2006 A200 [32] J. Langhorst S. Elsenbruch J. Koelzer A. Rueffer A. Michalsen G.J. Dobos Noninvasive markers in the assessment of intestinal inflammation in inflammatory bowel diseases: performance of fecal lactoferrin, calprotectin, and PMN-elastase, CRP, and clinical indices Am J Gastroenterol 102 2007 1 8