This is an author version of the contribution published on:
Questa è la versione dell’autore dell’opera:
[Scand J Gastroenterol., 49(12), 2014, 10.3109/00365521.2014.934913.]
ovvero [Caviglia GP, Pantaleoni S, Touscoz GA, Adriani A, Rosso C, Smedile
A, Pellicano R, Astegiano M, Bresso F, 49, Taylor & Francis, 2014,
pagg.1419-1424]
The definitive version is available at:
La versione definitiva è disponibile alla URL:
Fecal calprotectin is an effective diagnostic tool that differentiates inflammatory from functional intestinal disorders
GIAN PAOLO CAVIGLIA1, STEFANO PANTALEONI2, GIOVANNI ANTONIO TOUSCOZ1, ALESSANDRO ADRIANI2, CHIARA ROSSO1, ANTONINA SMEDILE1, RINALDO PELLICANO2, MARCO ASTEGIANO2 & FRANCESCA BRESSO3
1Laboratory of Liver Physiopathology, Department of Medical Science, University of
Turin, Turin, Italy, 2Department of Gastroenterology and Hepatology, Città della
Salute e della Scienza Hospital, Turin, Italy, and 3Gastrocentrum Medicin, Karolinska
University Hospital, Stockholm, Sweden
Gian Paolo Caviglia: Laboratory of Liver Physiopathology, Department of Medical Science, University of Turin, Città della Salute e della Scienza Hospital – SGAS, Via Cavour, n° 31, 10123 Turin, Italy; tel (0039)0116333532, fax (0039)0116333976
caviglia.giampi@libero.it
Stefano Pantaleoni: Department of Gastroenterology and Hepatology, Città della Salute e della Scienza Hospital - SGAS, Via Cavour, n° 31, 10123 Turin, Italy; tel (0039)0116333565, fax (0039)0116333623,
stefano.pantaleoni@tiscali.it
Giovani Antonio Touscoz: Laboratory of Liver Physiopathology, Department of Medical Science, University of Turin, Città della Salute e della Scienza Hospital –
SGAS, Via Cavour, n° 31, 10123 Turin, Italy; (0039)0116333532, fax (0039)0116333976
gtouscoz@cittadellasalute.to.it
Alessandro Adriani: Department of Gastroenterology and Hepatology, Città della Salute e della Scienza Hospital - SGAS, Via Cavour, n° 31, 10123 Turin, Italy; tel (0039)0116333565, fax (0039)0116333623,
alessandro_adriani@alice.it
Chiara Rosso: Laboratory of Liver Physiopathology, Department of Medical Science, University of Turin, Città della Salute e della Scienza Hospital – SGAS, Via Cavour, n° 31, 10123 Turin, Italy; tel (0039)0116333532, fax (0039)0116333976
chiara.rosso84@tiscali.it
Antonina Smedile: Laboratory of Liver Physiopathology, Department of Medical Science, University of Turin, Città della Salute e della Scienza Hospital – SGAS, Via Cavour, n° 31, 10123 Turin, Italy; tel (0039)0116333532, fax (0039)0116333976
asmedile@cittadellasalute.to.it
Rinaldo Pellicano: Department of Gastroenterology and Hepatology, Città della Salute e della Scienza Hospital - SGAS, Via Cavour, n° 31, 10123 Turin, Italy; tel (0039)0116333565, fax (0039)0116333623,
rinaldo_pellican@hotmail.com
Corresponding author: Marco Astegiano: Department of Gastroenterology and
Hepatology, Città della Salute e della Scienza Hospital - SGAS, Via Cavour, n° 31, 10123 Turin, Italy; tel (0039)3356348383, fax (0039)0116333623,
mastegiano@cittadellasalute.to.it
7-9, 14183 Stockholm, Sweden; tel +46 8 6089143, fax: +46 8 7745538, Francesca.Bresso@ki.se
This work was not supported by grants.
Abstract
The clinical pictures of functional gastrointestinal disorders and inflammatory diseases can be quite similar leading to inappropriate and expensive investigations.Objective. To
investigate fecal calprotectin (FC) diagnostic performance in different gastrointestinal conditions. Material and Methods. Stool specimens of 66 outpatients referred for colonoscopy were collected for further FC determination. Diagnostic accuracy was assessed by the area under the receiver operating characteristic curve (AUC). Sensitivity (Se), specificity (Sp), positive (PPV) and negative predictive values (NPV) were calculated according to the presence of inflammation and the main final diagnosis.
Results. Histological inflammation was found in 45 (68%) patients: 24 had a diagnosis
of inflammatory bowel disease (IBD) while 21 reported miscellaneous conditions (5 microscopic colitis, 2 eosinophilic colitis and 14 nonspecific chronic colitis). The diagnosis in the 21 (32%) patients without inflammation was irritable bowel syndrome (IBS). Median FC values were 268 µg/g (95% CI, 151-343) and 49 µg/g (95% CI, 23-101) in patients with and without inflammation, respectively (p=0.0001). AUC value of FC was 0.811 (Se=68.9%, Sp=71.4%, PPV=83.8% and NPV=56.3% with a cut-off value of 100 µg/g) for discriminating between patients with and without inflammation and 0.931 (Se=87.5%, Sp=90.5%, PPV=91.3% and NPV=86.4% with a cut-off value of 150 µg/g) for discriminating between patients with IBS and IBD. Using the cut-off value recommended by the manufacturer (50 µg/g), we found Se=100%, Sp=52.4%, PPV=70.6% and NPV=100% values for the diagnosis of IBD. Conclusions. FC appears to be a reliable non-invasive biomarker of intestinal inflammation useful to improve the appropriateness of colonoscopy requests.
Introduction
Discrimination of patients with functional intestinal disorders such as irritable bowel syndrome (IBS) from those with organic intestinal diseases, in particular inflammatory bowel disease (IBD), cannot be based exclusively on symptoms or on routine biochemical investigations. In fact, both conditions share many overlapping features including abdominal pain, excessive flatus, bloating and altered bowel habit. Available blood biomarkers like C-reactive protein (CRP) and eytrhocyte sedimentation rate (ESR) are not sufficiently specific and sensitive. Currently, colonoscopy with multiple biopsies is considered the gold standard to establish IBD diagnosis but it is an invasive and expensive procedure. An ideal diagnostic approach should be sensitive enough to reliably detect intestinal inflammation with a reasonable specificity to avoid inappropriate investigations.
Calprotectin is a calcium binding protein which is found mainly in human neutrophil granulocytes [1] but also in monocytes and macrophages [2]. The amount of calprotectin in feces is proportional to the neutrophil migration through the gut wall and increases with the severity of inflammation. An increase in fecal calprotectin (FC) levels has been associated with IBD [3] but also with various gastrointestinal diseases such as microscopic colitis [4], infectious diarrhea [5], non-steroidal anti-inflammatory enteropathy [6] and neoplastic conditions [7].
FC is resistant to colonic bacterial degradation, it is stable in stool for up one week at room temperature [8,9] and no changes over time have been found by storing feces at -20°C [10].
diagnostic accuracy in patients with and without bowel inflammation.
Methods
A total of 82 outpatients from the Division of Gastroenterology and Hepatology of San Giovanni Battista University Hospital of Turin were consecutively approached between January 2010 and December 2011. All patients referred to our outpatient clinic for the first time and reported abdominal pain and/or altered bowel habit lasting at least 4 weeks. Patients taking regularly non-steroidal anti-inflammatory drugs, patients with known colorectal cancer or polyps and patients with known acute gastrointestinal infection diseases were considered ineligible. A full medical history was obtained from all patients enrolled and a physical examination was performed. All patients underwent first-step haematology and biochemistry tests, including white blood cell count, ESR, CRP and thyroid hormones. In addition, a serological screening for celiac disease was performed. Stool samples were collected for bacteria and parasite examination and for further FC determination. All patients underwent colonoscopy with biopsies for histological examination to evaluate the presence of macro and micro inflammation, respectively. Both endoscopists and pathologists were blinded to FC levels. IBS was defined according to the Rome III Diagnostic Criteria [11]. The study protocol was conformed to the guidelines of the 1975 Declaration of Helsinki. All patients gave their written informed consent prior to recruitment.
Stool specimens for FC determination were collected within 1 week of the first visit and stored at -20°C until the time of measurement. Before analysis frozen stool samples were thawned at room temperature. Laboratory personnel performing the analyses was blinded to clinical diagnosis and details of the patients' clinical histories. FC was assayed by ELISA (Calprest Eurospital, Trieste, Italy) using polyclonal antibody against calprotectin, as previously described [12]. Calprotectin concentrations were given in µg/g of feces.
Statistical analysis
FC values are expressed in medians and ranges. Kruskal-Wallis test was used to evaluate differences in calprotectin levels among groups, while Fisher's exact test was performed to analyze categorical data. Diagnostic performance, sensitivity (Se), specificity (Sp), positive predictive value (PPV) and negative predictive value (NPV) of FC were assessed by using receiver operating characteristic (ROC) curves. A p-value <0.05 was considered statistically significant.
All statistical analyses were performed using Med Calc software, version 9.2.1.0.
Results
A total of 66 (80.1%) out of 82 subjects were enrolled in the study. The reasons for exclusion were: 9 (10.7%) diverticular disease, 1 (1.2%) small intestinal bacterial overgrowth syndrome, 4 (4.8%) adenoma, 1 (1.2%) adenocarcinoma and 1 (1.2%) amyloidosis (Figure 1). The 66 enrolled patients included 45 (68%) patients with macro
and/or micro inflammation and 21 (32%) with no inflammation.
Figure 1
Median FC values were 268 µg/g (95% CI, 151-343) and 49 µg/g (95% CI, 23-101) in patients with and without inflammation, respectively (p=0.0001).
According to the presence of inflammation, the main demographic and clinical characteristics of the studied patients are reported in Table I.
Table I
No difference in gender and age distribution was found among the groups (p=0.4365 and p=0.0902, respectively). Abdominal pain (p=0.0128) and bloating (p=0.0027), CRP (p=0.0126) and FC (p=0.0001) values significantly differed according to the presence of inflammation. However, based on a multivariate regression analysis, FC levels remained the only factor significantly and independently associated with inflammation (p=0.0211).
The ROC curve was constructed to predict the presence of inflammation according to FC and the area under the curve (AUC) for FC, Se, Sp, PPV and NPV at different cut-off levels are reported in Table II.
Table II
n=5; Crohn’s disease, n=12; indeterminate colitis, n=7), 5 patients with microscopic colitis, 2 patients with eosinophilic colitis and 14 patients with nonspecific chronic colitis. A diagnosis of nonspecific chronic colitis was set in patients presenting no macroscopic or microscopic evidence of IBD but a low-grade microscopic inflammation that does not fulfill diagnostic criteria of lymphocytic/collagenous/eosinophilic colitis [13]. For further analysis, patients with miscellaneous conditions including microscopic colitis, eosinophilic colitis and nonspecific chronic colitis were clustered in the subgroup of ‘other colitis’. The final diagnosis among the 21 patients without inflammation was IBS in all cases. In detail, 4 patients had constipation-predominant IBS, 10 diarrhea-predominant IBS, 5 alternating bowel habit IBS, and 2 post-infectious IBS. Due to the low number of patients, a comparison between the subgroups was not possible.
According to final diagnosis, median FC values were 349 µg/g (95%CI 286-500 µg/g), 92 µg/g (95%CI 39-256 µg/g) and 49 µg/g (95%CI 23-101 µg/g) in patients with IBD, other colitis and IBS respectively (p<0.0001) (Figure 2).
Figure 2
The ROC curves showed FC has high diagnostic accuracy in distinguish IBS from IBD (AUC=0.931) but a moderate diagnostic accuracy in discriminate other colitis from IBS and IBD (AUC=0.673 and AUC=0.793, respectively). Se, Sp, PPV and NPV at different cut-off levels are reported in Table III.
Using the cut-off value recommended by the manufacturer (50 µg/g), 24 out of 24 (100%) patients with IBD, 3 out of 5 (60%) patients with microscopic colitis, 1 out of 2 (50%) patients with eosinophilic colitis, 10 out of 14 (71.4%) patients with nonspecific colitis and 10 out of 21 (47.6%) patients with IBS tested positive.
Discussion
The studied population is young (mean age 42 years), consisting mainly of female patients (69.7%). In particular, the number of women with IBS diagnosis (16 women out of 21 patients) reflects the available epidemiological data for which IBS affects more frequently women, especially under 45 years old [14].
This study shows that FC levels in patients with macro and/or micro inflammation are significantly higher than those in patients without inflammation (268 µg/g vs 49 µg/g; p=0.0001). In agreement to a recent study [15], ROC curve analysis showed that a FC concentration of 150 µg/g was the best cut-off for discriminating between patients with and without inflammation. At this cut-off, FC had Se=66.7% and Sp=90.5%, while using the cut-off value recommended by the manufacturer (50 µg/g), Se increased to 84.4% whereas Sp decreased to 52.4%. A higher sensitivity allows to exclude functional diseases and identify the majority of inflammatory cases despite the risk to perform colonoscopy and/or further examination on patients who had false-positive results.
In fact, about 48% of IBS patients had FC levels higher than 50 µg/g. Four patients had FC values between 50 and 100 µg/g representing only a weakly positive
result, while concentrations > 100 µg/g could be explained by a possible daily variation of FC concentration in patients without neoplasm or inflammation [16]. Moreover, we found high FC concentration (225 µg/g and 416 µg/g) in 2 IBS patients with a family history of IBD; this might be explained by an increased intestinal permeability associated with an increased migration of granulocytes into the gut lumen, leading to IBD development [17]. Furthermore, several novel mechanism of IBS pathophysiology have recently challenged the classical concept of IBS as an absolute idiopathic functional disease. Recent evidence supports the role of post-infectious low-grade inflammation, immunologic factors such as imbalanced IL-10 or TNFα levels, altered intestinal microbiota, dietary factors and abnormalities in neuroendocrine peptides derived from enteroendocrine cells in the pathogenesis of IBS [11]. In fact, several studies reported an increased number of activated immunocompetent cells, including mast cells, T-lymphocytes, plasma cells and neutrophils in the gastrointestinal mucosa of a subset of IBS patients [18-21]. These findings could explain higher FC levels found in patients with a final diagnosis of IBS.
Based on the above considerations, an intermediate cut-off of 100 µg/g might be a good compromise that would result in clinical practice in a more accurate and restrictive selection of patients for colonoscopy. However, the risk of missing patients with mild inflammation could be compensated with a thorough clinical follow-up in association with FC monitoring over time.
Considering the final diagnosis, median FC values were 49 µg/g in IBS patients, 92 µg/g in patients with miscellaneous conditions and 349 µg/g in patients with IBD (p<0.0001). The results of the present study support the use of FC as a routine test for discriminating patients with and without bowel inflammation (AUC=0.811), especially
IBS from IBD (AUC=0.931), confirming what was previously reported in other studies [22]. In fact, FC values above 150 µg/g predicts IBD with a specificity of 90.5% and a PPV of 91.3% while FC values below 50 µg/g are strongly indicative of IBS (Se=100% and NPV=100%). However, FC diagnostic accuracy in distinguish IBS from other colitis (AUC=0.673) and other colitis from IBD (AUC=0.793) is moderate and not enough to establish a diagnosis, but could be useful to the clinician in patients management.
The present study is limited by the rather small number of patients included. A bigger sample size would be ideal to investigate FC diagnostic accuracy, specially in order to evaluate FC performance within each subgroup of patients. Despite the small size of the study cohort, our results are in agreement with previous published data regarding FC diagnostic accuracy in discriminating functional from inflammatory bowel disorders.
In conclusion, FC appears to be a reliable marker of intestinal inflammation and may be useful to improve the appropriateness of colonoscopy requests avoiding unnecessary investigations in patients with abdominal complaints. However, FC monitoring over time in association with the clinical follow-up should be suggested before endoscopy indication, also considering that some organic intestinal diseases requiring endoscopic approach cannot be determined only by FC levels.
References
[1] Berntzen HB, Fagerhol MK. L1, a major granulocyte protein: antigenic properties of its subunits. Scand J Clin Lab Invest, 1988; 48:647-652.
[2] Foell D, Wittkowski H, Roth J. Monitoring disease activity by stool analyses: from occult blood to molecular markers of intestinal inflammation and damage. Gut, 2009; 58:859-868.
[3] Burri E, Beglinger C. Faecal calprotectin in the diagnosis of inflammatory bowel disease. Biochem Med (Zagreb), 2011; 21:245-253.
[4] Wildt S, Nordgaard Lassen I, Bendtsen F, Rumessen JJ. Metabolic and inflammatory faecal markers in collagenous colitis. Eur J Gastroenterol Hepatol, 2007; 19:567-574.
[5] Shastri YM, Bergis D, Povse N, Schafer V, Shastri S, Weindel M et al. Prospective multicenter study evaluating fecal calprotectin in adult acute bacterial diarrhea. Am J Med, 2008; 121:1099-1106.
[6] Maiden L, Thjodleifsson B, Theodors A, Gonzalez J, Bjamason I. A quantitative analysis of NSAID-induced small bowel pathology by capsule enteroscopy. Gastroenteroly, 2005; 128:1172-1178.
[7] Roseth AG, Kristinsson J, Fagerhol MK, Schjonsby H, Aadland E, Nygaard K et al. Faecal calprotectin: a novel test for the diagnosis of colorectal cancer? Scand J Gastroenterol, 1993; 28:1073-1076.
[8] Johne B, Fagerhol MK, Lyberg T, Prydz H, Brandtzaeg P, Naess Andresen CF et al. Functional and clinical aspects of the myelomonocyte protein calprotectin. J Clin Pathol: Mol Pathol, 1997; 50:113-123.
[9] Schoepfer AM, Trummler M, Seeholzer P, Seibold Schmid B, Seibold F. Discriminating IBD from IBS: comparison of test performance of fecal markers, blood leukocytes, CRP, and IBD antibodies. Inflamm Bowel Dis, 2008; 14:32-39. [10] Ton H, Brandsnes, Dale S, Hotlund J, Skuibina F, Schjonsby H et al. Improved
assay for fecal calprotectin. Clin Chim Acta, 2000; 292:41-54.
[11] Lee YJ, Park KS. Irritable bowel syndrome: Emerging paradigm in pathophysiology. World J Gastroenterol, 2014; 20:2456-2469
[12] D'Incà R, Dal Pont E, Di Leo V, Benazzato L, Martinato M, Lamboglia F et al. Can Calprotectin predict relapse risk in inflammatory bowel disease? Am J Gastroenterol, 2008; 103:2007-2014.
[13] Fiehn AM, Bjørnbak C, Warnecke M, Engel PJ, Munck LK. Observer variability in the histopathologic diagnosis of microscopic colitis and subgroups. Hum Pathol, 2013;44:2461-6.
[14] Talley NJ. Irritable bowel syndrome: definition, diagnosis, and epidemiology. Bailleres Best Pract Res Clin Gastroenterol, 1999; 13:371-384.
[15] Licata A, Randazzo C, Cappello M, Calvaruso V, Butera G, Florena AM et al. Fecal calprotectin in clinical practice. A noninvasive screening tool for patients with chronic diarrhea. J Clin Gastroenterol, 2012; 46:504-508.
[16] Costa F, Mumolo MG, Bellini M, Romano MR, Ceccarelli L, Arpe P et al. Role of faecal calprotectin as non-invasive marker of intestinal inflammation. Dig Liver Dis, 2003; 35:642-647.
[17] Berstad A, Arslan G, Folvik G. Relationship between intestinal permeability and calprotectin concentration in gut lavage fluid. Scand J Gastroenterol, 2000; 35:64-69.
[18] Chadwick Vs, Chen W, Shu D, Paulus B, Bethwaite P et al. Activation of the mucosal immune system in irritable bowel syndrome. Gastroenterology, 2002; 122:1778-1783.
[19] Cremon C, Gargano L, Morselli-Labate AM, Santini D, De Giorgio R, Stenghellini V et al. Mucosal immune activation in irritable bowel syndrome: gender-dependence and association with digestive symptoms. Am J Gastroenterol, 2009; 104:392-400
[20] Akbar A, Yiangou Y, Facer P, Walters JR, Anand P, Ghosh S. Increased capsaicin receptor TRPV1-expressing sensory fibres in irritable bowel syndrome and their correlation with abdominal pain. Gut, 2008; 57:923-929.
[21] Ohman L, Isaksson S, Lindmark AC, Posserud I, Stotzer PO, Strid H et al. T-cell activation in patients with irritable bowel syndrome. Am J Gastroenterol, 2009; 104:1205-1212.
[22] Van Rheenen PF, Van de Vijer E, Fidler V. Faecal calprotectin for screening of patients with suspected inflammatory bowel disease: diagnostic mata-analysis. BMJ 2010;341:c3369.
Table I. Demographic and clinical characteristics of patients enrolled according to the presence of macro and/or micro inflammation
Overall No
inflammation Inflammation p* p** Number of patients, n (%) 66 (100%) 21 (32%) 45 (68%)
Gender ,M/F 20/46 5/16 15/30 0.4365
-Age, mean (range) 42 (18-78) 37 (20-62) 45 (18-78) 0.0902 -Symptoms, % ● Abdominal pain ● Diarrhea ● Bloating 69.7% 63.6% 22.7% 90.0% 55.0% 47.6% 60.9% 67.4% 13.3% 0.0128 0.4572 0.0027 0.1771 -0.0820 ESR > 15 mm/h, % 19.6% 6.3% 25.7% 0.0839 -CRP > 3 mg/L, % 32.7% 11% 40% 0.0126 0.1268 FC, median (95% CI) 139 (71-264) 49 (23-101) 268 (151-343) 0.0001 0.0211
p*= univariate analysis; p**= multivariate analysis
M, male; F, female; ESR, erythrocyte sedimentation rate; CRP, C-reactive protein; FC, fecal calprotectin.
Table II. Diagnostic accuracy for FC at different cut-off levels according to the presence of macro and/or micro inflammation
Inflammation vs no inflammation AUC cut-off Se (%) Sp (%) PPV (%) NPV (%) 0.811 50 µg/g 84.4 52.4 79.2 61.1
100 µg/g 68.9 71.4 83.8 56.3 150 µg/g 66.7 90.5 93.9 54.3 AUC, area under the curve; Se, sensitivity; Sp, specificity; PPV, positive predictive
Table III. Diagnostic accuracy for FC at different cut-off levels according to the final diagnosis
AUC cut-off Se (%) Sp (%) PPV (%) NPV (%) IBS vs Other colitis 0.673 50 µg/g 66.7 52.4 58.3 61.1
100 µg/g 47.6 71.4 62.5 57.7
IBS vs IBD 0.931 50 µg/g 100 52.4 70.6 100
100 µg/g 87.5 71.4 77.8 83.3 150 µg/g 87.5 90.5 91.3 86.4 IBD vs Other colitis 0.793 200 µg/g 83.3 61.9 71.4 76.5 250 µg/g 79.2 73.9 76.0 77.3 IBS, irritable bowel syndrome; IBD, inflammatory bowel disease; AUC, area under the
curve; Se, sensitivity; Sp, specificity; PPV, positive predictive value; NPV, negative predictive value.
