Lewis Score Correlates More Closely with Fecal Calprotectin Than Capsule Endoscopy Crohn s Disease Activity Index

Transcription

1 Dig Dis Sci (2012) 57: DOI /s ORIGINAL ARTICLE Lewis Score Correlates More Closely with Fecal Calprotectin Than Capsule Endoscopy Crohn s Disease Activity Index Anastasios Koulaouzidis Sarah Douglas John N. Plevris Received: 19 June 2011 / Accepted: 20 October 2011 / Published online: 6 November 2011 Ó Springer Science+Business Media, LLC 2011 Abstract Background Small-bowel capsule endoscopy (SBCE) is an invaluable imaging method for the small bowel. The Lewis score (LS) and the Capsule Endoscopy Crohn s Disease Activity Index (CECDAI) have been developed to standardize the reporting of small-bowel inflammation. Fecal calprotectin (FC) represents a highly reliable biomarker of intestinal inflammation. Aim To assess the performance of the two SBCE inflammation scoring systems by correlating them with FC. Furthermore, to define threshold levels for CECDAI. Methods Retrospective study; patients who underwent SBCE and had FC measurement shortly before or after SBCE. LS and CECDAI were calculated by a single reviewer and correlated [Spearman s (r s )] with the FC results. Linear regression analysis was used to identify threshold levels for CECDAI. Results Forty-nine patients; three subgroups A, B and C (based on FC levels \100, , and C200 lg/g, respectively). LS appears to correlate with FC (r s = 0.448, p = ), unlike CECDAI, which does not demonstrate significant correlation (r s = 0.245, p = 0.089). Strongly positive correlation between FC and LS was observed in subgroup A (r s = 0.68, p = ), while in subgroups B and C, neither LS nor CECDAI showed correlation with FC. Significant correlation between LS and CECDAI was demonstrated (r s = , p \ ). Linear regression analysis demonstrates that LS thresholds of 135 and A. Koulaouzidis (&) S. Douglas J. N. Plevris Centre for Liver and Digestive Disorders, Endoscopy Unit, The Royal Infirmary of Edinburgh, 51 Little France Crescent, Edinburgh EH16 4SA, Scotland, UK akoulaouzidis@hotmail.com 790 correspond with CECDAI levels of 3.8 and 5.8, respectively. Conclusions LS performs better than CECDAI in describing small-bowel inflammation, especially at FC levels of \100 lg/g. Furthermore, CECDAI levels of 3.8 and 5.8 seem to correspond to LS thresholds of 135 and 790, respectively. Keywords Capsule endoscopy Intestine, small Inflammation Scoring method Fecal calprotectin Biological markers Introduction Since its introduction into clinical practice more than a decade ago, small-bowel capsule endoscopy (SBCE) has been established as a state-of-the-art imaging method for the small bowel. Its unique ability to provide direct mucosal visualization, hence aiding detection of minute ulcers or mucosal breaks (often missed by other imaging modalities), has led to its incremental use in the evaluation of patients with known or clinically suspected Crohn s disease (CD) and non-steroidal anti-inflammatory drugs (NSAID)-induced enteropathy [1 3]. A recent pool-data analysis showed that SBCE has a minimal 1% miss rate for small-bowel ulcers [4]. Of note, although SBCE provides the highest yield of information regarding mucosal lesions, it does not allow assessment of small-bowel wall thickness or extraluminal abdominal findings [5]. Calprotectin is an abundant human protein, constituting 5% of total body protein and is a major part (50 60%) of the neutrophilic cytosol [6]. It is released from activated granulocytes and inflamed epithelia, as part of the initial innate immune response [7]. The amount of calprotectin in

2 988 Dig Dis Sci (2012) 57: feces is proportional to the granulocyte migration to the gastrointestinal mucosa. To date, several publications have demonstrated an excellent correlation of the amount of fecal calprotectin (FC) with the severity of mucosal inflammation from idiopathic, infective, and/or neoplastic enteropathies [8 10]. Therefore, FC represents a simple, reliable, highly specific and non-invasive gold standard biomarker of gastrointestinal inflammation [10, 11]. Until recently, there has been a lack of standardization in the description of the extent and severity of small-bowel mucosa inflammation, as severity scale of mucosal disease activity has not been universally followed [12]. To this end, two groups of investigators (Gralnek et al. and Gal et al.)each developed and validated their own SBCE mucosal disease activity index [13, 14]. Gal et al. [13] produced the Capsule Endoscopy Crohn s Disease Activity Index (CECDAI). The scoring system consists of three parameters/components: an inflammation score, a disease-extent score, and a luminal narrowing (stricture) score. The total score is the sum of the two segmental scores, ranging from 0 to 36 [12, 13]. Although the authors have not set specific thresholds, higher CECDAI levels indicate increasing severity of mucosal inflammation. Gralnek et al. [14] devised and validated the Lewis score (LS) index, based on three endoscopic parameters: villous edema, ulcer and stenosis/stricture. Using these parameters, the authors established a score range of 8 4,800 points [12] where: LS \ 135 reflects normal mucosal appearances, LS mild mucosal inflammatory change and an LS value C790 moderate to severe mucosal inflammatory changes [14]. An LS calculator has been incorporated into the RAPID Ò (Given Imaging Ltd., Yokneam, Israel) software (from version 6 onwards), allowing assessment of a range of quantitative and qualitative descriptors relating to the aforementioned endoscopic variables. In contrast to CECDAI, the number of lesions (mucosal breaks/ulcers) is taken into consideration when calculating the LS [12]. The second significant difference between the two scoring indices is that in CECDAI, villous edema, and ulcers are categorized as the two poles of a continuum (inflammation) rather than independent variables, as in the LS [12]. To date, there are only a limited number of studies looking into the correlation/usefulness of FC with SBCE [15, 16]. Moreover, there is a paucity of studies evaluating either CECDAI or LS in relation to stool biomarker levels, which should be considered superior (to both SBCE inflammation scoring systems) in reflecting the small-bowel inflammation level. We have recently reported that the FC level (cut-off 200 lg/g) is associated with high diagnostic yield of SBCE (65%) in patients with continuous clinical suspicion of IBD, despite prior negative bi-directional endoscopy [15]. Furthermore, there is accumulating evidence that FC \ 100 lg/ g is not associated with clinically significant intestinal inflammation [10, 15, 16]. Aim The primary objective of this retrospective study was to assess the performance of CECDAI and LS in the evaluation of small-bowel inflammation by correlating scoring system results with FC levels (considered the gold standard marker of inflammation). Moreover, the performance of the two SBCE inflammation scoring systems was checked in patient subgroups, as defined by the previously identified as significant FC threshold levels of 100 and 200 lg/g [15]. We also aimed to identify potential threshold values for CECDAI. Methods A combined computerized search of the SBCE and biochemistry database (March 2005 to December 2010) was performed in order to identify patients who had an FC measurement shortly before or after their SBCE. Time intervals of up to 30 days from each other (FC measurement and SBCE) were not considered problematic, as it is unlikely that significant changes in mucosal inflammation would occur within this period, as long as no therapy is administered [10]. If the FC level was measured prior to the SBCE, the minus symbol (-) was used to represent the number of days between the two tests; the plus symbol (?) was used if SBCE preceded FC measurement. In our center, FC is measured by a quantitative enzyme linked immunoassay (ELISA) and the laboratory reference range is 0 50 lg/g, (as advised by the kit manufacturer, level \20 lg/g equals undetectable FC). Further selection was performed by checking the electronic records for previous endoscopic work-up. Only those who were subjected to a colonoscopy (with either unremarkable findings or findings/pathology, e.g., polyps that were dealt before obtaining the stool sample for calprotectin) in the 12-month period prior to the SBCE were included in further analysis. This was done to eliminate any concerns that an abnormal FC result could be the effect of a gastrointestinal condition other than small-bowel inflammation [10, 16]. The meticulous search and case selection was performed by one of the authors (SD). SBCE was performed with the PillCam Ò SB1/SB2 (Given Imaging Ltd, Yokneam, Israel) and the MiroCam Ò (IntroMedic Co., Seoul, South Korea) capsule endoscopes using the predefined for our unit pre-procedure small-bowel preparation (2 lit polyethylene glycol) regimen as well as procedural protocol (simethicone ± prokinetic). A certified gastroenterologist (AK), with extensive experience in capsule interpretation ([700 SBCE sequence reviews) and blinded to the FC results, reviewed the SBCE sequences for CECDAI and LS calculation. The sequence review was

3 Dig Dis Sci (2012) 57: performed using the RAPID Ò 7 software and the MiroView TM (2nd version) software. SBCE video sequences were not de-identified. All sequences were reviewed at an automatic viewing speed of ten frames per second (fps), with the reviewer sitting at arm-length s distance from the screen in a room with dimmed light for maximal yield; only a desktop spotlight was used when illumination was required for data input in the data-collecting Excel spreadsheet. Roll-through mode was applied for improved delineation of mucosal lesions and more accurate CECDAI and LS calculation. CECDAI was calculated using a laminated printout of the scoring system [13]. LS was calculated (using the necessary quantitative and qualitative descriptors) with the help of the LS screen incorporated into the RAPID Ò 7software. Thereafter, the initial SBCE reports were accessed and any discrepancies were discussed and resolved by team consensus. The age, gender, date of FC measurement, FC value, indication and date of SBCE, clinical diagnosis, as well as small-bowel transit time (SBTT) and quality of bowel preparation were recorded. To date, there is no standardized or generally accepted bowel-cleansing score for capsule endoscopy, hence a four-point grading scale (poor, fair, good, and very good; from 0 to 3) was applied, depending on the proportion of visualized mucosa: grade 3 (very good visibility): [75% mucosa seen, grade 2 (good visibility): 50 75% mucosa visible, grade 1 (average visibility): 25 50% mucosa seen, grade 0 (poor visibility): \25% mucosa seen. Each small-bowel tertile was scored separately, with the final cleansing score being the average of the sum of the three tertile cleansing scores. Continuous data are presented as mean ± standard deviation (SD) and range. Statistical analyses were carried out with a statistical package program for Windows (StatsDirect Ò version Software, StatsDirect Ltd., Altrincham, Cheshire, UK). Spearman s rank correlation coefficient (r s ) was used to assess the correlation between FC levels and LS or CECDAI results. A probability (p) value \ 0.05 (two-tailed) was considered to be statistically significant. Receiver operating characteristics (ROC) curves were visualized and the corresponding areas under the curves were calculated. Thereafter, areas under the ROC curves (AUC ROC ) were compared to each other. Linear regression analysis (model Y = a? bx, 95% CI) was applied in order to identify CECDAI threshold levels that could match the 135 and 790 threshold points of LS. Chi-square calculation was used to analyze the relationship between the two systems and FC (at 200 lg/g level). This study was conducted in accordance with UK research ethics guidelines. After review by the local ethics committee, further specific ethical review and approval were not required, as the study was considered a retrospective audit work using data obtained as part of regular patient care. Results Cohort Demographics and Clinical Characteristics Six hundred and ninety-four SBCE were performed in our tertiary referral center (catchment area: 1 million) in the predefined period; out these, 74 were carried out in patients who had FC measured within a 30-day period prior to or after their SBCE. In five cases, the SBCE was not realized in the cecum by the end of the recording. Four of these were due to technical issues, i.e., premature termination of the recording period, hence they were excluded from further analysis. Of the remaining 70, a cohort of 49 patients (11 male/38 female; mean age: 41.9 ± 16.5 year, range: years) was further selected, as they had records of undergoing a colonoscopy prior to the SBCE with either unremarkable findings or pathology that was dealt with before obtaining the stool sample for FC. Some of these patients were included in a previous study of our group [15]. In this cohort, 23 examinations were performed with the PillCam Ò SB1/SB2 and the remainder (n = 26) with the MiroCam Ò. The mean SBTT for the cohort was 252 ± 38 min. The mean preparation cleansing quality score of the small bowel was good (score: 2.33 ± 0.52, range: ). The mean FC was ± 378 lg/g, range\20 1,740 lg/g. The mean LS was ± 1,337, range: 0 5,392; the mean CECDAI was 5.63 ± 5.63, range: Based on the FC results, the cohort (n = 49) was further divided into three groups. Group A (n = 16) constituted patients with FC \ 100 lg/g, group B (n = 12) patients with FC C 100 but \200 lg/g, and group C (n = 21) patients with FC C 200 lg/g. The mean FC levels for groups A, B, and C were 45.3 ± 38.1 lg/g, range: \20 95 lg/g; ± 26.5 lg/g, range: lg/g; and, 589 ± lg/g, range: 230 1,740 lg/g, respectively. The respective LS results for groups A, B, and C were ± 860.9, range 0 3,456; ± 239, range: 0 790; and, 1,258 ± 1,775.73, range: 0 5,392, respectively. For CECDAI similar values for the three groups were 5.25 ± 5.22, range: 0 17; 2.58 ± 2.5, range: 0 8; and, 7.66 ± 6.49, range: 0 22, respectively. Correlation of LS and CECDAI with FC Levels In the entire cohort (n = 49), LS correlated with FC levels (r s = 0.448, p = ), versus CECDAI, which demonstrated no correlation (r s = 0.245, p = 0.089) (Fig. 1a, b). In group A (n = 16, FC \ 100 lg/g), LS showed strong correlation with FC levels versus CECDAI, which again did not exhibit any statistically significant correlation (r s = 0.68, p = vs. r s = 0.446, p = ). In group B (n = 12, FC: lg/g), neither of the two SBCE inflammation scores demonstrated any

4 990 Dig Dis Sci (2012) 57: a 8000 Scatter plot of correlation between LS and FC a LS Sensitivity FC 0.25 LS at FC 50 LS at FC 100 LS at FC 200 b Scatter plot of correlation between CECDAI and FC b Specificity 1.00 CECDAI FC Fig. 1 a, b Scatter gram plots demonstrating the correlation between the Lewis score (LS) and the Capsule Endoscopy Crohn s Disease Activity Index (CECDAI) with the fecal calprotectin (FC) levels (Spearman s rank correlation coefficient (r s ) = 0.448, p = vs. r s = 0.245, p = 0.089) correlation with FC levels (LS; r s =-0.293, p = vs. CECDAI; r s =-0.291, p = ). Finally, in group C (n = 21, FC C 200 lg/g) neither LS nor CECDAI showed any correlation with FC levels (LS; r s = 0.132, p = vs. CECDAI; r s = 0.149, p = ). Corresponding AUC ROC values for LS at FC levels of 50, 100, and 200 lg/g were 0.87, 0.68, and 0.72, respectively. Corresponding AUC ROC values for CECDAI at FC levels of 50, 100, and 200 lg/g were 0.63, 0.51, and 0.66, respectively (Fig. 2a, b). SBCE inflammation score correlation with FC levels was also examined in the group of patients with CD (either newly diagnosed or re-assessment). In this group (n = 12), there was a correlation between LS and FC, but no correlation was demonstrated between CECDAI and FC levels (r s = , p = vs. r s = , p = ). Furthermore, there was a statistically significant correlation between LS and CECDAI (r s = , p \ ). Simple linear regression analysis showed that Sensitivity Specificity the LS thresholds of 135 and 790 correspond to CECDAI levels of 3.8 and 5.8, respectively (Fig. 3). Chi-square (v 2 ) analysis for FC levels (at 200 lg/g level, shown previously to represent significant mucosal inflammation) [15, 16] and LS threshold 135 was v 2 = , p = , Pearson s contingency Similar analysis for CECDAI score of 3.8 was v 2 = , p = , Pearson s contingency Discussion CECDAI at FC 50 CECDAI at FC 100 CECDAI at FC 200 Fig. 2 a Receiver operator characteristics (ROC) and area under the curve (AUC) for LS at FC 50 (black), 100 (green), and 200 (red) lg/g. AUC 0.87, 0.68, and 0.72, respectively. The optimum cut-off point for LS, 8, 112, and 168, respectively. b Receiver operator characteristics (ROC) and area under the curve (AUC) for CECDAI at FC 50 (black), 100 (green), and 200 (red) lg/g. AUC 0.63, 0.61, and 0.66, respectively. Optimum cut-off point for CECDAI, 4, 10 and 3, respectively The assessment of type and degree of small intestinal inflammation in an objective and reliable way remains a challenging task in gastroenterology [17]. As gut

5 Dig Dis Sci (2012) 57: CECDAI SE and 95% CI for regression estimate LS Fig. 3 Linear regression plot graph (green line) of Capsule Endoscopy Crohn s Disease Activity Index (CECDAI) and Lewis score (LS), 95% confidence interval (CI): black lines, Standard Error (SE) of slop: pink lines. Spearman s rank correlation coefficient (r s ) = 0.68, p \ LS 135 CECDAI 3.82; LS 790 CECDAI 5.86 inflammation is not directly observable by patients or their physicians, many methods have been developed to quantify the degree and severity of that inflammation [18]. Symptoms are an important indicator, but they are subjective, and are often influenced by physiological factors [19]. Clinical activity indices are calculated using some objective parameters, but on the other hand, they were never appealing outside the remits of clinical studies, as most of them are cumbersome and difficult to use in everyday clinical practice [18]. Therefore, stool biomarkers feature as ideal tools for assessment of gut inflammation. Fecal calprotectin (FC), alongside fecal lactoferrin, are the major markers in current use [20, 21]. FC is released from the cytoplasm of neutrophils during cell activation and death. It is stable for up to 1 week at room temperature and is distributed evenly throughout the stool. Therefore, it can be readily detected in random stool samples at levels[20 lg/g, using a commercially available and validated quantitative ELISA [7, 22]. Furthermore, spot samples of calprotectin in stool are as reliable as 24-h stool collections, with a proven excellent correlation [18]. In fact, a positive correlation has been shown between the FC levels and the fecal concentration of indium-111 labeled autologous leucocytes, a test historically considered as the gold standard for measuring intestinal inflammation [23]. FC could distinguish between IBD and irritable bowel syndrome (IBS) with a sensitivity of 100% and a specificity of 97% [24]. Schoepfer et al. [22] showed a close correlation of FC with the Simple Endoscopic index of Severity for CD (SES-CD). Moreover, FC correlates with abnormalities seen on small-bowel barium radiology. In 2004, Dolwani et al. [25] demonstrated that by using in their patients cohort an FC cut-off value of 60 lg/g, they were able to predict all cases with abnormal small-bowel barium follow-through (SBFT) as well as all cases with a normal SBFT, but with organic intestinal disease. More recently, Zippi et al. [17] showed that FC levels correlate well with the degree of mucosal inflammation recorded on small-bowel magnetic resonance imaging (MRI). Therefore, it is plausible to hypothesize that high levels of FC, in the documented absence of colonic pathology, reflect more accurately than any other serologic or endoscopic index the amount of small-bowel mucosal inflammation/disease [26]. SBCE LS was developed (initially in 2004 and thereafter presented and validated in 2008) by a group of gastroenterologists with wide experience in SBCE and CD. They assessed the following SBCE parameters: edema, erythema, nodularity, presence of denuded mucosa, villous appearance, ulcer, and stenosis [14]. The score in current use is based on three remaining SBCE variables, i.e., ulcer, stenosis and villous appearance. For LS calculation, the small-bowel transit time is divided into three equal parts, creating tertiles. A short segment is defined B10% of the tertile; a long segment is defined as 11 50% of the tertile; and whole segment is [50% of the tertile. A total score is created as follows: Maximum tertile score {[(Villous parameter 9 extent 9 descriptor)? (Ulcer parameter 9 extent 9 size)] for tertile 1 or [(Villous parameter 9 extent 9 descriptor)? (Ulcer parameter 9 extent 9 size)] for tertile 2 or [(Villous parameter 9 extent 9 descriptor)? (Ulcer parameter 9 extent 9 size)] for tertile 3 }? (Stenosis number 9 ulcerated 9 traversed) [14]. It is once more important to underline that LS is a measuring index of mucosal disease activity and not a diagnostic index of smallbowel CD. The CECDAI, on the other hand, has been devised aiming to mostly help clinicians to diagnose CD, identify damage due to NSAIDs, measure disease activity and drug response (for clinical trials), and guide medical management [13]. The system involves dividing the small bowel into proximal and distal segments according to transit time and then rating each segment on the basis of three parameters: inflammation (A), extent of disease (B), and presence of strictures (C). The segmental score is calculated by multiplying the inflammation sub-score by the disease-sub extent score and adding the stricture sub-score (A 9 B? C); the final score is calculated by adding the two segmental scores: CECDAI = (A1 9 B1? C1)? (A2 9 B2? C2) [13]. In a previous study [15], we showed that FC \ 100 lg/g is a good predictor of a negative SBCE (negative predictive value 1), while FC [ 200 lg/g was associated with higher SBCE yield (65%) and confirmed Crohn s disease in 50% of cases. We have also suggested that FC [ 200 lg/g should be used as a selection tool in requesting investigations of the

6 992 Dig Dis Sci (2012) 57: small bowel [15]. In the present study, we show that at FC levels \ 100 lg/g, LS correlation was greater than CEC- DAI. However, at FC levels[100 lg/g neither CECDAI nor LS showed correlation with FC. This may be explained partly by the differing parameter values used in each index, specifically in the scoring of ulceration and mucosal edema. LS eliminates the differentiation of ulcers from mucosal breaks, erosions, aphthae, and other similar lesions, places less emphasis on extent and size of edema/ulceration, and focuses primarily on the presence/absence of each mucosal disease indicator. This, arguably, less subjective approach appears to favor an enhanced pick-up rate for more subtle changes consistent with lower FC levels. In order to explain discordant results of the best-performing SBCE inflammation scoring system, i.e., LS in the subgroups with FC levels [100 lg/g, we undertook a detailed review of the individual LS scoring components/ parameters. In LS, the higher marks are scored by the luminal stenosis descriptors, i.e., presence and number of stenosis, presence of ulcerations in the stenotic lumen, and obstruction or lack thereof. Luminal stenosis, however, is likely to contribute significantly less to FC production, due to its possible fibrotic nature. In fact, no correlation between the stenosis parameter and FC levels was found (r s = , p = 0.065), despite the fact that all cases (n = 6) that scored marks in the stenosis variable were also found to have peri-stenotic ulcers. In the patient subgroup, the mean level of FC was 453 ± 371 lg/g, range: 93 1,170 lg/g; five out of six had FC [ 200 lg/g. The two remaining LS parameters, i.e., villi and ulcers, were examined both separately as well as in combination and correlation with FC levels was demonstrated: villi (the cumulative score of all three tertiles was used) r s = 0.415, p = ; ulcers (cumulative score) r s = 0.299, p = 0.036; villi? ulcers (cumulative score) r s = , p = Twenty-nine patients (n = 29), with no luminal stenosis on SBCE, scored marks in the villi and/or ulcer parameter of LS. In this group, the mean FC level was ± lg/g; range: \20 1,740 lg/g. Interestingly though, statistically significant correlation of the cumulative villi, ulcers and villi? ulcers score with FC levels, was seen again only in subgroup A (patients with FC [ 100 lg/g). Looking again at the entire cohort (n = 49); in the subgroup of patients with LS \ 135 (n = 19), i.e., defined accordingly as those with either negligible or clinically insignificant small-bowel inflammation, the mean FC value was ± 168 lg/g; range:\ lg/g. There was no correlation between LS and FC levels (r s = , p = 0.369). Furthermore, patients with LS of 0 (n = 14) had mean FC level was ± 194 lg/g; range:\ lg/g. On the other hand, patients with FC \ 100 lg/g had a median LS of 45. Therefore, it is reasonable to suggest that although qualitative parametric scoring extent and degree of villous edema and (to a lesser extent) presence of single/few mucosal breaks is rather subjective, background pathology is enough to give rise to high FC levels. We have previously demonstrated that FC C 200 lg/g is consistent with increased yield of pathology in SBCE [15]. Of the 21 cases in this subgroup (C), four patients had LS \ 135 (three LS 0, one LS 112). None of them was diagnosed with Crohn s disease or other sinister small-bowel pathology. The remainder (n = 17) presented a median LS of 458, while seven (33%) had diagnosis of CD (either newly diagnosed or known for reassessment) [15]. Study Limitations We herein attempted to correlate results of two SBCE inflammation scoring systems, i.e., LS and CECDAI with FC levels. Although a rigorous methodology was applied in case selection, due to its retrospective nature, our study is probably affected by some of the problems inherent in every database review. For the purpose of this study, we adopted the widely accepted view that a time frame of ±30 days (between the FC and SBCE) [10], should not be considered problematic. We also recognize that although in the majority of included cases (32/49, 65%) stool sample collection from FC was done shortly before SBCE [16], bowel cleansing administered for the purpose of SBCE might impacted on FC results (although the use of PEG-based laxatives is not considered damaging to the intestinal mucosa). Finally, although both LS and CECDAI calculation was performed by a single and experienced in SBCE reviewer, observation and reporting/interpretation bias cannot be excluded. Furthermore, the number of included cases is considered small. Conclusions LS performs better than CECDAI in describing smallbowel mucosal inflammation, especially at FC levels of \100 lg/g. Furthermore, CECDAI levels of 3.8 and 5.8 seem to correspond to LS threshold levels of 135 and 790, respectively. Conflict of interest The authors report no conflicts of interest in relevance to this work. The authors alone are responsible for the content and writing of the paper. References 1. Triester SL, Leighton JA, Leontiadis GI, et al. A meta-analysis of the yield of capsule endoscopy compared to other diagnostic

7 Dig Dis Sci (2012) 57: modalities in patients with nonstricturing small bowel Crohn s disease. Am J Gastroenterol. 2006;101: Kornbluth A, Colombel JF, Leighton JA, et al. ICCE. ICCE consensus for inflammatory bowel disease. Endoscopy. 2005; 37: Lewis BS. Expanding role of capsule endoscopy in inflammatory bowel disease. World J Gastroenterol. 2008;14: Lewis BS, Eisen GM, Friedman S. A pooled analysis to evaluate results of capsule endoscopy trials. Endoscopy. 2005;37: Doherty GA, Moss AC, Cheifetz AS. Capsule endoscopy in suspected Crohn s disease: yield does not equal diagnosis. Am J Gastroenterol. 2010;105: Røseth AG, Fagerhol MK, Aadland E, et al. Assessment of the neutrophil dominating protein calprotectin in feces. A methodologic study. Scand J Gastroenterol. 1992;27: Logan R. Faecal calprotectin for the diagnosis of inflammatory bowel disease. BMJ. 2010;341:c3636. doi: /bmj.c Langhorst J, Elsenbruch S, Koelzer J, et al. 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. 2008;103: Sipponen T, Björkesten CG, Färkkilä M, et al. Faecal calprotectin and lactoferrin are reliable surrogate markers of endoscopic response during Crohn s disease treatment. Scand J Gastroenterol. 2010;45: van Rheenen PF, Van de Vijver E, Fidler V. Faecal calprotectin for screening of patients with suspected inflammatory bowel disease: diagnostic meta-analysis. BMJ. 2010;341:c3369. doi: /bmj.c Summerton CB, Longlands MG, Wiener K, et al. Faecal calprotectin: a marker of inflammation throughout the intestinal tract. Eur J Gastroenterol Hepatol. 2002;14: Lucendo AJ, Guagnozzi D. Small bowel video capsule endoscopy in Crohn s disease: what have we learned in the last ten years? World J Gastrointest Endosc. 2011;3: Gal E, Geller A, Fraser G, et al. Assessment and validation of the new capsule endoscopy Crohn s disease activity index (CECDAI). Dig Dis Sci. 2008;53: Gralnek IM, Defranchis R, Seidman E, et al. Development of a capsule endoscopy scoring index for small bowel mucosal inflammatory change. Aliment Pharmacol Ther. 2008;27: Koulaouzidis A, Douglas S, Rogers MA, et al. Fecal calprotectin: a selection tool for small bowel capsule endoscopy in suspected IBD with prior negative bidirectional endoscopy. Scand J Gastroenterol. 2011;46: Jensen MD, Kjeldsen J, Nathan T. Fecal calprotectin is equally sensitive in Crohn s disease affecting the small bowel and colon. Scand J Gastroenterol. 2011;46: Zippi M, Al Ansari N, Siliquini F, et al. Correlation between faecal calprotectin and magnetic resonance imaging (MRI) in the evaluation of inflammatory pattern in Crohn s disease. Clin Ter. 2010;161:e53 e Konikoff MR, Denson LA. Role of fecal calprotectin as a biomarker of intestinal inflammation in inflammatory bowel disease. Inflamm Bowel Dis. 2006;12: Bourreille A, Ignjatovic A, Aabakken L, World Organisation of Digestive Endoscopy (OMED) and the European Crohn s and Colitis Organisation (ECCO), et al. Role of small-bowel endoscopy in the management of patients with inflammatory bowel disease: an international OMED-ECCO consensus. Endoscopy. 2009;41: Sidhu R, Sanders DS, Wilson P, et al. Faecal lactoferrin, capsule endoscopy and Crohn s disease. Is there a three way relationship? A pilot study. J Gastrointestin Liver Dis. 2010;19: van Langenberg DR, Gearry RB, Wong HL, et al. The potential value of faecal lactoferrin as a screening test in hospitalized patients with diarrhoea. Intern Med J. 2010;40: Schoepfer AM, Beglinger C, Straumann A, et al. Non-invasive 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. 2008;103: Røseth AG, Schmidt PN, Fagerhol MK. Correlation between faecal excretion of indium-111-labelled granulocytes and calprotectin, a granulocyte marker protein, in patients with inflammatory bowel disease. Scand J Gastroenterol. 1999;34: Tibble J, Teahon K, Thjodleifsson B, et al. A simple method for assessing intestinal inflammation in Crohn s disease. Gut. 2000; 47: Dolwani S, Metzner M, Wassell JJ, et al. Diagnostic accuracy of faecal calprotectin estimation in prediction of abnormal small bowel radiology. Aliment Pharmacol Ther. 2004;20: Sprakes MB, Hamlin PJ, Ford AC. Utility of fecal calprotectin in differentiating active inflammatory bowel disease from coexistent irritable bowel syndrome. Am J Gastroenterol. 2011; 106:166.