Mooiweer, Erik MD; Fidder, Herma H. MD, PhD; Siersema, Peter D. MD, PhD; Laheij, Robert J. F. MD, PhD; Oldenburg, Bas MD, PhD
Longstanding colonic inflammation in patients with Crohn's disease (CD) or ulcerative colitis (UC) increases the risk of colorectal cancer (CRC).1,2 For this reason, patients with a disease duration of more than 8 years are advised to undergo regular endoscopic surveillance aimed at the detection of dysplasia and early stage asymptomatic CRC.3–5 Detection of dysplasia in patients with inflammatory bowel disease (IBD) is challenging; however, because dysplastic lesions are often subtle, flat, or not endoscopically identifiable altogether. Chromoendoscopy has been advocated in recent guidelines because this technique improves visualization of subtle flat lesions and thereby increases the dysplasia yield.6–8
If inflammation is present during surveillance colonoscopy, identification of lesions containing dysplasia is severely hampered and distinction between inflammation-associated atypia and dysplasia is challenging for pathologists.9 This is reflected by the low interobserver agreement between pathologists for the diagnosis of indefinite and low-grade dysplasia (LGD), especially if active inflammation is present.10,11 Therefore, it is of paramount importance to perform CRC surveillance only in patients in complete endoscopic remission and to repeat the procedure if severe extensive inflammation is present.
However, mucosal disease activity can also be present when patients with IBD have no or only mild symptoms.12,13 One previous study from the Netherlands showed that as much as 34% of patients with IBD undergoing surveillance had endoscopic and histological signs of active inflammation, despite being in clinical remission.14 Several blood and stool-based markers have been evaluated to predict the presence of intestinal inflammation, of which fecal calprotectin seems to be the most promising. This marker was found to correlate well with endoscopic disease activity both in CD and UC patients.15–17
We aimed to investigate whether fecal calprotectin testing before surveillance colonoscopy can reliably identify patients with active inflammation, thereby preventing ineffective procedures.
All consecutive patients with UC and CD scheduled for surveillance colonoscopy between July 2011 and August 2012 were invited to participate in this study. Patients collected a sample of feces before the start of bowel cleansing and stored the sample at room temperature for a maximum of 48 hours. Patients also completed a questionnaire to collect data for the simple clinical colitis activity index (patients with UC) or the Harvey–Bradshaw index without the parameter abdominal mass (patients with CD) before the colonoscopy.
Bowel preparation was performed with 4 L of polyethylene glycol according to the prevailing protocol in our center. Three experienced endoscopists performed the surveillance colonoscopies using chromoendoscopy with methylene blue (0.1%). All lesions suspected of neoplasia were removed; and additionally, 2 random biopsies were taken in 4 colonic segments (i.e., 8 biopsies in total) to assess histological disease activity. Endoscopic disease severity and extent of inflammation were scored in all 4 colonic segments, i.e., the ascending, transverse, descending, and the rectosigmoid colon. Disease severity in each segment was scored as no inflammation, mild (erythema, decreased vascular pattern, mild friability), moderate (marked erythema, absent vascular pattern), and severe inflammation (spontaneous bleeding, ulceration) in accordance with the Mayo endoscopic score. Because patients with CD could only be included if colitis was present, we decided to use the same score for UC and CD patients to increase comparability between groups. Endoscopists were unaware of the calprotectin levels when performing the colonoscopies.
Patients in whom the cecum could not be reached or with insufficient bowel preparation were excluded from further analysis. If moderate or severe inflammation was present in at least 1 colonic segment, the procedure was categorized as ineffective surveillance. The remaining procedures in which no inflammation or mild inflammation was found were categorized as effective surveillance. The pathology reports were reviewed to assess whether any of the targeted or random biopsies contained neoplasia.
All stool samples were stored at −80°C directly after they were received from the patients. Calprotectin was measured with a quantitative enzyme-linked immunoassay according to the manufacturer's instructions (Ridascreen; R-Biopharm, Darmstadt, Germany). Calprotectin levels were also analyzed with the commonly available Buhlmann assay (Buhlmann laboratories AG, Basel, Switzerland), which showed an excellent correlation with the Ridascreen results (correlation coefficient, r = 0.93, results not shown).
Baseline characteristics were analyzed using standard descriptive statistics. Calprotectin levels were compared between patients with effective and ineffective surveillance using the Mann–Whitney U test. The accuracy and appropriate cutoff values of calprotectin for identifying patients with ineffective surveillance were analyzed using receiver operator characteristic (ROC) statistics. The method described by Hanley and McNeil18 was used to assess whether there was a significant difference between the area under the ROC curves of CD and UC patients. The percentage of patients diagnosed with neoplasia was compared between patients with no inflammation, mild, moderate, or severe inflammation using chi-square testing. A 2-sided P value of <0.05 was considered statistically significant. Data were analyzed using SPSS version 20 for windows.
This study was approved by the medical ethical committee of our institution. All patients gave written informed consent.
In total, 176 surveillance colonoscopies were performed in 164 patients. The recruitment of the study population is summarized in Figure 1. Seventy-four patients were diagnosed with UC (45%), 83 had Crohn's colitis (51%), and 7 had IBD-unclassified (IBD-U) (4%). Baseline characteristics of the study population are shown in Table 1.
Complete endoscopic remission was observed in 111 colonoscopies (63%), whereas mild inflammation was present in 40 (23%), moderate inflammation in 15 (8%), and severe inflammation in 10 colonoscopies (6%) (Table 2). Therefore, 151 procedures (86%) were categorized as effective surveillance and 25 procedures (14%) as ineffective surveillance.
The median simple clinical colitis activity index and Harvey Bradshaw score was 0 (range, 0–15). Increased Harvey Bradshaw (>4) and simple clinical colitis activity index (>3) scores were observed in 12 patients with CD and 12 patients with UC or IBD-U.18,19 Of the 24 patients with increased Harvey Bradshaw or simple clinical colitis activity index scores, only 7 (29%) were classified as ineffective surveillance during endoscopy.
Median calprotectin levels for the effective and ineffective surveillance group were 84 mg/kg (range, 20–4609) and 1605 mg/kg (range, 66–26,336), respectively (P < 0.01; Fig. 2).
ROC curve analysis showed that a cutoff level of 539 mg/kg optimally distinguished patients with effective surveillance from patients with ineffective surveillance with 84% sensitivity, 89% specificity, a positive predictive value of 55%, and a negative predictive value of 97% (Fig. 3). Overall, the diagnostic accuracy of calprotectin as reflected by the area under the ROC curve was 0.89. These results were similar for patients with UC as compared with CD (area under the ROC curves 0.86 versus 0.92, respectively, P = 0.42).
If the calprotectin cutoff level of 539 mg/kg would be applied to the study population, 38 patients (22%) had levels above the cutoff of which 21 (55%) were classified as ineffective surveillance and 17 (45%) as effective surveillance. Of the 138 patients with levels below the cutoff, only 4 patients (3%) were classified as ineffective surveillance. Therefore, calprotectin testing before a scheduled surveillance colonoscopy could potentially reduce the incidence of ineffective surveillance because of inflammation from 25 procedures to 4 procedures (14%–3%, P < 0.01) with a number needed to test to identify 1 patient with ineffective surveillance of 8.4.
During a total of 176 surveillance colonoscopies, 30 dysplastic lesions were detected in 23 colonoscopies (13%). Two patients were diagnosed with LGD detected in a random biopsy. The remaining 28 dysplastic lesions were diagnosed in endoscopically visible lesions, which were classified by the endoscopists as adenoma-like in 24 lesions and as non-adenoma-like in the remaining 4 lesions (Table 2). The adenoma-like lesions were histologically classified as tubular adenomas (n = 21) serrated adenomas (n = 2), and tubulovillous adenoma (n = 1) all containing LGD. The 4 non-adenoma-like lesions contained LGD in all 4 cases.
Among the 111 surveillance colonoscopies with no inflammation, the neoplasia yield was 24 dysplastic lesions in 18 colonoscopies (16%). In 4 of 40 colonoscopies with mild inflammation, 5 dysplastic lesions were detected (10%). Only 1 dysplastic lesion in 15 colonoscopies with moderate inflammation (7%) and no dysplastic lesions in 10 colonoscopies with severe inflammation (0%) was found (P = 0.34; Fig. 4). The dysplasia yield among the 151 patients with effective surveillance was 15% compared with 4% among the 25 patients with ineffective surveillance (P = 0.15).
Endoscopic surveillance in patients with longstanding colonic IBD is challenging because IBD-associated dysplasia is more difficult to detect than sporadic adenomas in patients without IBD. Good bowel preparation, optimal visualization with chromoendoscopy, and an absence of mucosal disease activity are therefore of utmost importance to optimize the effectiveness of this procedure and increase dysplasia detection rates. This study shows that fecal calprotectin testing before a scheduled surveillance colonoscopy can significantly reduce the number of ineffective procedures, defined as colonoscopies showing persistent moderate or severe inflammation from 14% to 3%.
The large proportion of active inflammation encountered among patients undergoing surveillance in our study (37% including patients with mild inflammation) is in line with previous studies, which reported endoscopic inflammation in 34% to 50% of cases, despite the fact that these patients were in clinical remission.14,20 This is most likely because clinical symptoms do not accurately predict mucosal inflammation in both CD and UC patients.12,13
Although our results show that a cutoff of 539 mg/kg of calprotectin can predict ineffective surveillance, we did not investigate what the best management is when calprotectin is above this threshold. Baars et al21 showed that a short course of corticosteroids before a surveillance colonoscopy might decrease histological disease severity, although the decrease in endoscopic disease severity was not statistically significant. Postponing the procedure and initiating a step-up in maintenance therapy to achieve mucosal healing seems to be a logical approach, although this was not the scope of this study.
We also reported a lower dysplasia yield as the severity of inflammation found during surveillance increased. Although not statistically significant, the dysplasia yield decreased from 16% in patients with complete mucosal healing to 0% in patients with severe inflammation. This seems in contrast to the commonly accepted concept that persistent inflammation in IBD is a risk factor for the development of colitis-associated carcinoma.22–24 We believe that this can be attributed to the difficulty of detecting dysplasia in inflamed mucosa. Missed lesions and the carcinogenic effect of ongoing inflammation most likely explain the increased risk of neoplasia in these cases over time as reported by previous studies.
Analysis of 1 calprotectin sample costs around 30 euros (40 dollars) as compared with 459 euros (623 dollars) for colonoscopy. Based on our study cohort, calprotectin testing would have cost 5280 euros (7161 dollars) and saved 9639 euros (13,074 dollars) in colonoscopy costs. Therefore, the costs of calprotectin testing before a surveillance colonoscopy seem to be balanced out by the cost savings of reducing the number of ineffective surveillance procedures.
Our study has several limitations that need to be addressed. The low positive predictive value of 55% implies that the surveillance procedure would be incorrectly postponed in 45% of cases with calprotectin levels above the cutoff of 539 mg/kg. Therefore, postponing the procedure and initiating a step-up in therapy cannot be based solely on an elevated calprotectin level, which limits its usefulness in clinical practice.
Because of the low incidence of dysplasia, we were only able to identify trends regarding differences in dysplasia yield between patients with effective and ineffective surveillance. Although the strategy of fecal calprotectin testing was able to significantly reduce the number of ineffective surveillance procedures, the strategy of postponing all surveillance colonoscopies when calprotectin is above the cutoff level would have resulted in a nonsignificant increase in dysplasia yield from 13% to 15% in our study population (P = 0.64).
Therefore, we cannot definitively conclude that calprotectin testing before a surveillance colonoscopy increases the dysplasia yield and therefore the effectiveness of the procedure.
There is no well-defined cutoff to determine when a surveillance procedure is ineffective because of inflammation. Current surveillance guidelines simply state that surveillance colonoscopies, whenever possible, should be performed when the disease is in remission.4 We chose to classify patients with moderate and severe inflammation as ineffective procedures and mild or no inflammation as effective. The gradually decreasing neoplasia yield as the severity of inflammation increases would suggest that the neoplasia yield might be even better when patients with mild inflammation were also classified as ineffective surveillance. If patients with mild inflammation had classified as ineffective surveillance as well, 65 procedures (37%) would have been ineffective. Using a lower cutoff of 140-mg/kg calprotectin would have identified 56 of these patients (86%). Regarding the dysplasia yield, postponing all these procedures would increase the dysplasia yield only marginally to 16% as compared with 15% when only patients with moderate or severe inflammation were classified as ineffective. Because 40 patients were diagnosed with mild inflammation leading to a considerable rise in the number of procedures that need to be postponed, we feel that a cutoff at moderate or severe inflammation offers the best compromise between the number of procedures that need to be postponed and the increase in neoplasia yield.
The decision to abort the surveillance colonoscopy and repeat the procedure after induction therapy was left at the discretion of the endoscopists in this study population, rather than the aforementioned cutoff of patients with moderate or severe inflammation. Therefore, the procedure was repeated in only 7 patients of the 25 classified as ineffective surveillance. Because the endoscopists knew that the procedure would be repeated, their effort to detect neoplasia was probably lower, which could explain the lower yield in neoplasia in patients with inflammation. If these 7 patients were excluded, neoplasia yield changed from 8% to 7% for the patients with moderate inflammation and remained 0% for the patients with severe inflammation.
In conclusion, low fecal calprotectin can accurately identify patients with IBD without active inflammation in which CRC surveillance is most effective. Future studies should focus on whether the routine use of this tool and postponing surveillance procedures based on the calprotectin level is cost effective.
1. Herrinton LJ, Liu L, Levin TR, et al.. Incidence and mortality of colorectal adenocarcinoma in persons with inflammatory bowel disease from 1998 to 2010. Gastroenterology. 2012;143:382–389.
2. Lutgens MW, van Oijen MG, van der Heijden GJ, et al.. Declining risk of colorectal cancer in inflammatory bowel disease: an updated meta-analysis of population-based cohort studies. Inflamm Bowel Dis. 2013;19:789–799.
3. Farraye FA, Odze RD, Eaden J, et al.. AGA medical position statement on the diagnosis and management of colorectal neoplasia in inflammatory bowel disease. Gastroenterology. 2010;138:738–745.
4. Cairns SR, Scholefield JH, Steele RJ, et al.. Guidelines for colorectal cancer screening and surveillance in moderate and high risk groups (update from 2002). Gut. 2010;59:666–689.
5. Annese V, Daperno M, Rutter MD, et al.. European evidence concensus for endoscopy in inflammatory bowel disease. J Crohns Colitis. 2013;7:982–1018.
6. Kiesslich R, Fritsch J, Holtmann M, et al.. Methylene blue-aided chromoendoscopy for the detection of intraepithelial neoplasia and colon cancer in ulcerative colitis. Gastroenterology. 2003;124:880–888.
7. Rutter MD, Saunders BP, Schofield G, et al.. Pancolonic indigo carmine dye spraying for the detection of dysplasia in ulcerative colitis. Gut. 2004;53:256–260.
8. Marion JF, Waye JD, Present DH, et al.. Chromoendoscopy-targeted biopsies are superior to standard colonoscopic surveillance for detecting dysplasia in inflammatory bowel disease patients: a prospective endoscopic trial. Am J Gastroenterol. 2008;103:2342–2349.
9. Riddell RH, Goldman H, Ransohoff DF, et al.. Dysplasia in inflammatory bowel disease: standardized classification with provisional clinical applications. Hum Pathol. 1983;14:931–968.
10. van Schaik FD, ten Kate FJ, Offerhaus GJ, et al.. Misclassification of dysplasia in patients with inflammatory bowel disease: consequences for progression rates to advanced neoplasia. Inflamm Bowel Dis. 2011;17:1108–1116.
11. Odze RD, Goldblum J, Noffsinger A, et al.. Interobserver variability in the diagnosis of ulcerative colitis-associated dysplasia by telepathology. Mod Pathol. 2002;15:379–386.
12. 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 d'Etudes Therapeutiques des Affections Inflammatoires Digestives. Gut. 1994;35:231–235.
13. Osada T, Ohkusa T, Okayasu I, et al.. Correlations among total colonoscopic findings, clinical symptoms, and laboratory markers in ulcerative colitis. J Gastroenterol Hepatol. 2008;23(suppl 2):S262–S267.
14. Baars JE, Nuij VJ, Oldenburg B, et al.. Majority of patients with inflammatory bowel disease in clinical remission have mucosal inflammation. Inflamm Bowel Dis. 2012;18:1634–1640.
15. D'Haens G, Ferrante M, Vermeire S, et al.. Fecal calprotectin is a surrogate marker for endoscopic lesions in inflammatory bowel disease. Inflamm Bowel Dis. 2012;18:2218–2224.
16. Schoepfer AM, Beglinger C, Straumann A, et al.. Fecal calprotectin more accurately reflects endoscopic activity of ulcerative colitis than the Lichtiger Index, C-reactive protein, platelets, hemoglobin, and blood leukocytes. Inflamm Bowel Dis. 2013;19:332–341.
17. Schoepfer AM, Beglinger C, Straumann A, et al.. Fecal calprotectin correlates more closely with the Simple Endoscopic Score for Crohn's disease (SES-CD) than CRP, blood leukocytes, and the CDAI. Am J Gastroenterol. 2010;105:162–169.
18. Hanley JA, McNeil BJ. A method of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology. 1983;148:839–843.
19. Higgins PD, Schwartz M, Mapili J, et al.. Patient defined dichotomous end points for remission and clinical improvement in ulcerative colitis. Gut. 2005;54:782–788.
20. Rosenberg L, Nanda KS, Zenlea T, et al.. Histologic markers of inflammation in patients with ulcerative colitis in clinical remission. Clin Gastroenterol Hepatol. 2013;11:991–996.
21. Baars JE, Vogelaar L, Wolfhagen FH, et al.. A short course of corticosteroids prior to surveillance colonoscopy to decrease mucosal inflammation in inflammatory bowel disease patients: results from a randomized controlled trial. J Crohns Colitis. 2010;4:661–668.
22. Rutter M, Saunders B, Wilkinson K, et al.. Severity of inflammation is a risk factor for colorectal neoplasia in ulcerative colitis. Gastroenterology. 2004;126:451–459.
23. Gupta RB, Harpaz N, Itzkowitz S, et al.. Histologic inflammation is a risk factor for progression to colorectal neoplasia in ulcerative colitis: a cohort study. Gastroenterology. 2007;133:1099–1105; quiz 1340–1341.
24. Ullman TA, Itzkowitz SH. Intestinal inflammation and cancer. Gastroenterology. 2011;140:1807–1816.
© Crohn's & Colitis Foundation of America, Inc.