Journal of Crohn's and Colitis (2014) 8, 1438 1443 Available online at www.sciencedirect.com ScienceDirect Endoscopic biopsy samples of naïve colitides patients: Role of basal plasmacytosis Vincenzo Villanacci a, Elisabetta Antonelli b, Gianpaolo Reboldi b, Marianna Salemme a, Giovanni Casella c, Gabrio Bassotti b, a Pathology Section, Department of Molecular and Translational Medicine, Spedali Civili and University of Brescia, Italy b Gastroenterology Section, Department of Medicine, University of Perugia School of Medicine, Italy c Department of Medicine, Desio Hospital, Monza and Brianza, Italy Received 22 February 2014; received in revised form 10 April 2014; accepted 19 May 2014 KEYWORDS Basal plasmacytosis; Colitides; Crohn's disease; Eosinophils; Ulcerative colitis Abstract Background: Although it is usually not difficult to diagnose inflammatory bowel disease (IBD) on surgical resection specimens, difficulties sometimes arise in differentiating these entities from other forms of colitis on endoscopic biopsy specimens. Basal plasmacytosis is considered as an early feature of IBD colitis, but it is rare in non-ibd colitides. Aims: We assessed the value of basal plasmacytosis as an individual variable in untreated patients with colitis. Patients and methods: Archival slides of patients with untreated colitis (66 IBD and 49 non-ibd) and 20 controls with complete (from the terminal ileum to the rectum) endoscopic biopsy sampling were evaluated blindly for the presence of basal plasmacytosis and a possible association with the presence of eosinophils in the same anatomical location. Results: Overall, basal plasmacytosis was present in at least one anatomical segment in 58% of cases, and it was always present in patients with IBD, whereas it was sparsely found (9%) in patients with other colitides and in controls. Basal plasmacytosis in three or more segments had more than 80% probability for a patient to be classified as IBD, with the segmental distribution being different between ulcerative colitis and Crohn's disease. Additionally, basal plasmacytosis was always accompanied by eosinophils intermingled with plasma cells in the same anatomical position. Source of support: none. Corresponding author at: Clinica di Gastroenterologia ed Epatologia, Ospedale Santa Maria della Misericordia, Piazzale Menghini, 1, 06156 San Sisto, Perugia, Italy. E-mail address: gabassot@tin.it (G. Bassotti). http://dx.doi.org/10.1016/j.crohns.2014.05.003 1873-9946/ 2014 European Crohn's and Colitis Organisation. Published by Elsevier B.V. All rights reserved.
Basal plasmacytosis in IBD and non-ibd colitides 1439 Conclusion: As an individual feature, basal plasmacytosis (accompanied by eosinophils) is a strong feature suggesting IBD, particularly when present in three or more colonic segments. This fact may be useful in the evaluation of endoscopic biopsies from patients with colitis. 2014 European Crohn's and Colitis Organisation. Published by Elsevier B.V. All rights reserved. 1. Introduction Ulcerative colitis (UC) and Crohn's disease (CD) are the two most common forms of chronic inflammatory bowel disease (IBD). 1 The pathogenesis of these disorders is currently unknown 2 and there are no specific clinical or laboratory tests that may help to establish the diagnosis. 3 Thus, a correct diagnostic approach needs a combination of clinical, radiological, endoscopic, and histological criteria. 4,5 Histopathology provides an important contribution to the diagnosis of IBD 6 8 and clinicians usually use mucosal biopsy specimens not only to identify patients suspected to be affected by IBD, but also to determine disease extent and severity, and to differentiate UC from CD. 9 11 Other common diseases, such as bacterial infections, ischemia, and diverticulosis may mimic the histological appearance of IBD, and these imitators may themselves complicate the course of IBD. 12 In addition, the picture is complicated by the fact that several other types of chronic colitis, such as collagenous colitis (CC), lymphocytic colitis (LC), and diversion colitis, may show UC- or CD-like features, causing diagnostic confusion. 13 Structural changes of crypt architecture are thought to have a good predictive value to differentiate IBD from non-ibd entities; the three main histological features useful to distinguish IBD from other non-ibd colitides are the presence of crypt branching, cryptitis, and an increase of lymphocytes and plasma cells in the lamina propria. 13 17 In particular, basal plasmacytosis (defined as the presence of plasma cells between the base of the crypts and the muscularis mucosae)is an early feature observed in biopsies of adult and pediatric UC patients, 18,19 it is frequently observed in CD, but it is considered rare in non-ibd colitides, 20,21 and it seems to be the strongest predictor of IBD. 22 A correct diagnosis between UC, CD and other colitides is usually made without difficulty in resected surgical specimens or when all macroscopic and microscopic features of CD and UC are present in endoscopically-obtained specimens, but if some of the typical pathological changes are absent, or the samples are too few, the accuracy of diagnosis is reduced. 23 Purpose of the present study was to investigate the presence of basal plasmacytosis and to establish its value as an individual variable in endoscopic biopsies from firstly diagnosed, non-treated colitis patients. 2. Material and methods Archival slides from patients which in the period January 2005 January 2012 underwent complete histologic ileo-colonic mapping (terminal ileum, cecum and ascending colon, transverse colon, descending colon, sigmoid colon, and rectum) for colitis, with at least two biopsies taken for each anatomical segment, were retrieved. To be taken into consideration for the analyses, data were only analyzed when endoscopic procedures were carried out within four weeks from the onset of symptoms (diarrhea with or without presence of blood), and patients were free from any treatment. 2.1. Controls As a control group, we retrieved slides from patients investigated for painless diarrhea, in whom complete histological mapping was carried out as above, and a final diagnosis of functional diarrhea or lactose malabsorption was made. 2.2. Data analysis Slides were coded and interspersed by a pathologist and reviewed blindly by another pathologist. According to a previously validated description, basal plasmacytosis was defined as the presence of at least three plasma cells around (deep 1/5th of the lamina propria) or below the crypts, alongside or penetrating the muscularis mucosae. 19 In addition, we looked for the possible association of plasma cells with eosinophils in the same anatomical location. 2.3. Statistical analysis Data management and analysis were performed using SAS/STAT 9.3 (SAS Institute, Cary, NC, USA). Patients were grouped into three main categories according to primary diagnosis (IBD, non-ibd, controls) and then further categorized between UC and CD. Basal plasmacytosis was dummy coded as present (1) vs absent (0) in each examined segment. We then defined a summary measure obtained by adding the number of positive segments independently of their anatomic location. Results are given as means or proportions, as appropriate. Continuous variables were compared by one-way ANOVA and categorical variables by the chi-square test. Categorical data were tabulated as counts and percentages with 95% CIs. Logistic regression was used to evaluate the strength of the association between the presence of basal plasmacytosis and diagnostic category. 24 To evaluate the diagnostic accuracy of basal plasmacytosis, receiver operating characteristic (ROC) curves were constructed and the area under the curve (AUC) was estimated by non-parametric methods. 25 All statistical tests were two-tailed and evaluated at the 5% significance level. 2.4. Ethical considerations Since this was a retrospective study, no individual patient identification was involved and no study-driven clinical intervention was performed; therefore no ethical approval was necessary.
1440 V. Villanacci et al. 3. Results Overall, specimens from 135 patients (66 IBD and 49 non-ibd colitides, and 20 controls) fulfilling inclusion criteria were analyzed. Table 1 shows the demographic variables of these groups. The non-ibd group (considered together for the subsequent analyses) included 6 patients with collagenous colitis, 4 patients with lymphocytic colitis, 10 patients with colitis associated with colonic diverticular disease, 5 patients with ischemic colitis, 17 patients with infectious colitis, 6 patients with eosinophilic colitis, and 1 patient with radiation colitis. Diagnoses in controls were functional diarrhea (15 cases) and lactose malabsorption (5 cases). In the whole group, basal plasmacytosis was present in at least one anatomical segment in 58% of the cases, and it was always present in patients with IBD. Conversely, it was sparsely found (9%) in patients with other types of colitis, mainly collagenous colitis, and in controls (Table 2). Fig. 1 shows the prevalence of basal plasmacytosis by segment and diagnosis in patients and in controls. When basal plasmacytosis was present in three or more anatomical segments, the probability of IBD was greater than 80% and the ROC-area was 0.94 ± 0.03 (p b 0.0001; Fig. 2). The distribution of basal plasmacytosis within the anatomical segments showed some peculiarities between IBD subgroups. Basal plasmacytosis was always present in ileal and ascending colonic segments of CD patients, in 17% of the cases obtained from the transverse colon, and in 60%, 34.3% and 26% of the cases in the descending, sigmoid and rectum, respectively. Concerning UC, basal plasmacytosis was always absent in the ileum, it was present in 87% of the cases in the ascending colon, it was always present in the transverse and descending colon, and it was present in 84% and 77.4% of the cases in the sigmoid and the rectum. The probability of CD decreased as a function of the number of positive segments (Fig. 3; panel A). With more than 4 segments the probability of CD was less than 20% and the ROC area was 0.84 ± 0.05 (p b 0.0001; Fig. 3 panel B). In the non-ibd group, basal plasmacytosis was never present in patients with eosinophilic colitis, infectious colitis, lymphocytic colitis, pseudomembranous colitis and radiation colitis (Table 2). In contrast, the overall number of segments harboring basal plasmacytosis in CC overlapped IBD colonic segments, being more similar to UC than to CD. Table 1 Demographic variables of IBD and non-ibd patients. N. M/F Age IBD 66 33/33 37 (34 40) UC 31 11/20 36 (31 40) CD 35 21/14 38 (32 43) Non-IBD 49 23/26 47 (43 52) Controls 20 5/15 49 (28 64) Table 2 groups. In controls, basal plasmacytosis was found in the sigmoid colon in only two subjects. Interestingly, an additional feature was observed when analyzing the slides, i.e. the fact that the presence of basal plasmacytosis was always accompanied by at least one three eosinophils intermingled with plasma cells in the same anatomical position (Fig. 4). This feature was not present in cases where basal plasmacytosis was absent. 4. Discussion Presence of basal plasmacytosis in the various Disease Basal plasmacytosis Absent Present Total Ulcerative colitis 0 31 31 Crohn's disease 0 35 35 Collagenous colitis 0 6 6 Eosinophilic colitis 6 0 6 Infectious colitis 16 0 16 Lymphocytic colitis 4 0 4 Pseudomembranous col 1 0 1 Radiation colitis 1 0 1 Diverticular colitis 7 3 10 Ischemic colitis 4 1 5 Controls 18 2 20 Total 57 78 135 The three major histological features useful to distinguish IBD from other non-ibd causes are an increase of lymphocytes and plasma cells in the lamina propria, the presence of crypt branching, and cryptitis. 14 17 In particular basal plasmacytosis, defined as the presence of plasma cells between the base of the crypts and the muscularis mucosae, seems to be the strongest predictor of IBD, 22,26 even for clinical relapses (at least for UC). 27 It has been demonstrated that the diagnostic yield increases when more sections are examined and it is essential to obtain segmental biopsy specimens to differentiate between the various subtypes of colitis. 28 For this reason, histology it is Figure 1 Prevalence of basal plasmacytosis by segment and diagnosis in IBD, non-ibd colitides, and controls. 1 = terminal ileum; 2 = cecum and ascending colon; 3 = transverse colon; 4 = descending colon; 5 = sigmoid colon; 6 = rectum.
Basal plasmacytosis in IBD and non-ibd colitides 1441 Figure 2 Panel A: probability (with 95% confidence band) of IBD vs non-ibd/controls as a function of the number of positive segments. Panel B: ROC analysis. crucial to establish a proper diagnosis before starting an appropriatetherapy,whichinturncaninducechangesthe histological morphology. Thus, histological examination of endoscopic biopsies remains a key step in the work-up of IBD patients. However, difficulties arise when the pathologist receives specimens with insufficient endoscopic sampling and when the patient had received some form of treatment; these variables may decrease the likelihood of finding all the elements useful for a correct diagnosis, making it difficult to distinguish IBD from non-ibd. 23 This study, somewhat unique in that it took into consideration endoscopic sampling from only naïve (untreated) patients, showed that the presence of basal plasmacytosis as a single histological feature, at least in one colonic segment, is highly suggestive of IBD in patients with newly diagnosed colitis and no previous treatment. In fact, the presence of basal plasmacytosis in at least one colonic segment was present in all IBD patients, and sparsely found in other forms of (mostly collagenous) colitis and in controls. This observation is strengthened by the fact that every patient and control underwent complete histological mapping (from the terminal ileum to the rectum). Moreover, we found that the probability of IBD diagnosis dramatically increases with the number of segments positive for the presence of basal plasmacytosis. In fact, if basal plasmacytosis is present in three or more segments of the colon the histological diagnosis of IBD can be made in 90% of the cases. Interestingly, the segmental distribution of basal plasmacytosis showed some differences between CD and UC; in fact, in CD patients it was always present in the ileum, cecum, and ascending colon, and variably present (17 60%) in the other segments, whereas in UC patients it was always present in the transverse and descending colon, and consistently present (77 87%) in the other segments. In the other colitides, the segmental distribution of basal plasmacytosis revealed its complete absence in the terminal ileum, and was more similar to UC in the few cases where it was found. In our study non-ibd population comprised not only infectious cases, but also other forms of colitis and our data demonstrated that, in the entire group, BP was present in less than 9% of the cases, and it was mostly present in subjects with CC. Another interesting feature was the observation of a strict association of basal plasma cells with one three eosinophils, a feature only present in patients with basal plasmacytosis. Eosinophils are one of the several cell types Figure 3 analysis. Panel A: probability (with 95% confidence band) of CD vs UC as a function of the number of positive segments. Panel B: ROC
1442 V. Villanacci et al. Marianna Salemme: helped in pathological analysis and reviewed critically the paper. Giovanni Casella: reviewed critically the paper. Gabrio Bassotti: planned the study, helped in analyzing the data, and wrote the first draft of the manuscript. All authors reviewed and approved the final version of the manuscript. Conflict of interest statement Figure 4 A and B: representative histological pictures of basal plasmacytosis in IBD patients, showing the strict association with eosinophils (arrows). H&E, original magnification 40. involved in the pathogenesis of IBD, and may play a role in both the inflammatory response and the repair processes. 29 Although their exact role has yet to be elucidated in this setting, we feel that their presence may not be serendipitous, since eosinophils have been shown to be key providers of plasma cell survival factors. 30 In conclusion, the finding of basal plasmacytosis (always associated with the presence of eosinophils) in colonic endoscopic biopsies of naïve patients with colitis seems to be a strong histological clue toward the presence of an IBD, since this variable is completely absent in most other colitides. The segmental distribution also enforces this point, and its presence in the terminal ileum is suggestive of CD. Of course, in ideal conditions the pathologist should receive adequate biopsy sampling, correctly oriented on acetate cellulose filters if possible, and sufficient clinical informations, 21,31 and would therefore rely on other clues, such as crypt branching and distortion and cryptitis 32 ; however, in real world it is not usual to receive only a few fragments, often from a single or from two colonic segments. Therefore, the presence and distribution of basal plasmacytosis (associated to eosinophils) may reveal an important issue in the differential diagnosis of the various forms of colitis. Authors' contribution Vincenzo Villanacci: planned the study and carried out pathological evaluation. Elisabetta Antonelli: helped in drafting the manuscript and reviewed critically the paper. Gianpaolo Reboldi: carried out statistical analysis and reviewed critically the paper. None declared. References 1. Walfish A, Sachar D. Phenotype classification in IBD: is there an impact on therapy? Inflamm Bowel Dis 2007;13:1573 5. 2. Gersemann M, Stange EF, Wehkamp J. From intestinal stem cells to inflammatory bowel diseases. World J Gastroenterol 2011;17:3198 203. 3. Jellema P, van Tulder MW, van der Horst HE, Florie J, Mulder CJ, van der Windt DA. Inflammatory bowel disease: a systematic review on the value of diagnostic testing in primary care. Colorectal Dis 2011;13:239 54. 4. Stange EF, Travis SP, Vermeire S, et al. European evidence based consensus on the diagnosis and management of Crohn's disease: definitions and diagnosis. Gut 2006;55(Suppl 1):i1 i15. 5. Stange EF, Travis SP, Vermeire S, et al. European evidence-based consensus on the diagnosis and management of ulcerative colitis: definitions and diagnosis. JCrohnsColitis2008;2:1 23. 6. Yantiss RK, Odze RD. Optimal approach to obtaining mucosal biopsies for assessment of inflammatory disorders of the gastrointestinal tract. Am J Gastroenterol 2009;104:774 83. 7. Peyrin-Biroulet L, Ferrante M, Magro F, Campbell S, Franchimont D, Fidder H, Strid H, et al. Results from the 2nd Scientific Workshop of the ECCO. I: impact of mucosal healing on the course of inflammatory bowel disease, J Crohns Colitis 2011;5:477 83. 8. Vilela EG, Torres HO, Martins FP, Ferrari Mde L, Andrade MM, Cunha AS. Evaluation of inflammatory activity in Crohn's disease and ulcerative colitis. World J Gastroenterol 2012;18:872 81. 9. Haboubi NY, Shaath NM, Safarani MN, Ghafar F. Improved diagnostic accuracy of inflammatory bowel disease: a clinicopathological collaborative approach. Tech Coloproctol 2004;8:117 21. 10. Cornaggia M, Leutner M, Mescoli C, et al. Chronic idiopathic inflammatory bowel diseases: the histology report. Dig Liver Dis 2011;43(Suppl 4):S293 303. 11. Baumgart DC. Endoscopic surveillance in Crohn's disease and ulcerative colitis: who needs what and when? Dig Dis 2011;29(Suppl 1):32 5. 12. Shepherd NA. Pathological mimics of chronic inflammatory bowel disease. J Clin Pathol 1991;44:726 33. 13. Yantiss RK, Odze RD. Diagnostic difficulties in inflammatory bowel disease pathology. Histopathology 2006;48:116 32. 14. Rubio CA, Johansson C, Kock Y. A quantitative method of estimating inflammation in the rectal mucosa. II. Normal limits in symptomatic patients. Scand J Gastroenterol 1982;17:1077 781. 15. Rubio CA, Johansson C, Kock Y. A quantitative method of estimating inflammation in the rectal mucosa. III. Chronic ulcerative colitis. Scand J Gastroenterol 1982;17: 1083 7. 16. Rubio CA, Johansson C, Uribe A, Kock Y. A quantitative method of estimating inflammation in the rectal mucosa. IV. Ulcerative colitis in remission. Scand J Gastroenterol 1984;19:525 30.
Basal plasmacytosis in IBD and non-ibd colitides 1443 17. Seldenrijk CA, Morson BC, Meuwissen SG, Schipper NW, Lindeman J, Meijer CJ. Histopathological evaluation of colonic mucosal biopsy specimens in chronic inflammatory bowel disease: diagnostic implications. Gut 1991;32:1514 20. 18. Lessells AM, Beck JS, Burnett RA, et al. Observer variability in the histopathological reporting of abnormal rectal biopsy specimens. J Clin Pathol 1994;47:48 52. 19. Tanaka M, Riddell RH, Saito H, Soma Y, Hidaka H, Kudo H. Morphologic criteria applicable to biopsy specimens for effective distinction of inflammatory bowel disease from other forms of colitis and of Crohn's disease from ulcerative colitis. Scand J Gastroenterol 1999;34:55 67. 20. Nostrant T, Kumar NB, Appelman HD. Histopathology differentiates acute self-limited colitis from ulcerative colitis. Gastroenterology 1987;92:318 28. 21. Villanacci V, Manenti S, Antonelli E, Chiudinelli M, Giuliano V, Bassotti G. Non-IBD colitides: clinically useful histopathological clues. Rev Esp Enferm Dig 2011;103:366 72. 22. Schumacher G, Kollberg B, Sandstedt B. A prospective study of first attacks of inflammatory bowel disease and infectious colitis. Histologic course during the 1st year after presentation. Scand J Gastroenterol 1994;29:318 32. 23. Yantiss RK, Odze RD. Pitfalls in the interpretation of nonneoplastic mucosal biopsies in inflammatory bowel disease. Am J Gastroenterol 2007;102:890 904. 24. Hosmer DW, Lemeshow S. Applied Logistic Regression. 2nd ed. Chichester, England: John Wiley & Sons Ltd.; 2000. ISBN: 9780471356325. 25. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 1982;143:29 36. 26. Magro F, Langner C, Driessen A, et al. European consensus on the histopathology of inflammatory bowel disease. J Crohns Colitis 2013;7:827 51. 27. Bessissow T, Lemmens B, Ferrante M, et al. Prognostic value of serologic and histologic markers on clinical relapse in ulcerative colitis patients with mucosal healing. Am J Gastroenterol 2012;107:1684 92. 28. Dejaco C, Oesterreicher C, Angelberger S, et al. Diagnosing colitis: a prospective study on essential parameters for reaching a diagnosis. Endoscopy 2003;35:1004 8. 29. Al-Haddad S, Riddell RH. The role of eosinophils in inflammatory bowel disease. Gut 2005;54:1674 5. 30. Chu VT, Berek C. The establishment of the plasma cell survival niche in the bone marrow. Immunol Rev 2013;251:177 88. 31. Canavese G, Bassotti G, Astegiano M, et al. Inflammatory bowel disease: a proposal to facilitate the achievement of an unequivocal diagnosis. World J Gastroenterol 2013;19:426 8. 32. Bressenot A, Geboes K, Vignaud JM, Guéant JL, Peyrin-Biroulet L. Microscopic features for initial diagnosis and disease activity evaluation in inflammatory bowel disease. Inflamm Bowel Dis 2013;19:1745 52.