Lymphocytic Follicles and Aggregates Are a Determinant of Mucosal Damage and Duration of Diarrhea Nilesh Shah, MD; Bhavin Thakkar, MD; Emily Shen, FRACP; Marie Loh, MSc; Pei Yi Chong, BSc; Wee Hoe Gan, MRCP; Tian Ming Tu, MRCP; Liang Shen, PhD; Richie Soong, PhD; Manuel Salto-Tellez, MD, FRCPath

Context.—Nonspecific changes (nonspecific chronic Ninflammation) in patients with chronic diarrhea represent the commonest diagnosis in colorectal biopsy interpretation, but these changes are of little clinical significance. Objective.—To find, within this group, histologic and immunohistologic diagnostic criteria to predict the duration and resolution of diarrhea. Design.—Detailed clinical features and histologic findings were analyzed in a cohort of 47 patients with chronic diarrhea, with near-normal histology and no clear-cut known etiologic agent. Immunohistochemistry to mast cells (CD117) and Treg cells (FOXP3) was also assessed in 39 patients. Results.—Increased number of lymphoid follicles and aggregates, increased number of mast cells, and paucity of Treg were the statistically significant key findings (P = .003, P = .008, and P = .04, respectively). The duration of diarrhea was correlated with the number of large lymphoid istologic changes in colonic biopsy materials from H patients suffering of diarrhea are often reported by pathologists as ‘‘near-normal,’’ ‘‘nonspecific changes,’’ ‘‘within normal limits’’ or ‘‘mild, nonspecific inflammation.’’ 1 This group represents, arguably, the commonest diagnostic label in endoscopic colorectal pathology. Despite the inability of the pathologist to characterize these cases further, there are important clinical implications. Although diarrhea in some of these patients is ‘‘selflimiting and resolving,’’ 2 in other cases it is ‘‘prolonged and unresolved.’’ 3 Many times patients with prolonged and unresolved diarrhea and nonspecific histologic changes are misdiagnosed as having irritable bowel syndrome1,3 or ‘‘minimal colitis.’’ 4 Clinically, the distincAccepted for publication March 12, 2012. From the Cancer Science Institute of Singapore (Drs Shah, Thakkar, Soong, and Salto-Tellez, Ms Loh, and Ms Chong), the Department of Pathology, National University Hospital, Yong Loo Lin Medical School (Drs Soong and Salto-Tellez), and the Biostatistics Unit, Yong Loo Lin Medical School (Dr L. Shen), National University of Singapore, Singapore; the Endoscopy Centre (Gastroenterology), Tan Tock Seng Hospital, Singapore (Dr E. Shen, Dr Gan, and Dr Tu); and the Centre for Cancer Research and Cell Biology, Queen’s University, Belfast, United Kingdom (Dr Salto-Tellez). Presented orally in part at Digestive Disease Week (American Gastroenterological Association); May 21, 2008; San Diego, California. The authors have no relevant financial interest in the products or companies described in this article. Reprints: Manuel Salto-Tellez, MD, FRCPath, 97 Lisburn Road, Belfast BT9 7BL, United Kingdom (e-mail: [email protected]). Arch Pathol Lab Med—Vol 137, January 2013

follicles and aggregates (P = .001, r = .48), number of total lymphoid follicles and aggregates (P = .003, r = .43), density of lymphoid follicles and aggregates (P = .009, r = .38), and total lymphoid follicles and aggregates per biopsy (P = .004, r = .42) and the number of mast cells (P = .001, r = .52). The number of mast cells and Treg cells showed significant difference between resolved and unresolved cases (P = .001 and P = .01 respectively). Conclusions.—Lymphocytic follicles and aggregates colitis, previously regarded as of negligible diagnostic significance, allows the prediction of the behavior of chronic diarrhea in a subset of patients with nonspecific changes on colonic biopsy. The increased number of mast cells and paucity of Treg cells further helps to identify such unresolved cases. (Arch Pathol Lab Med. 2013;137:83–89; doi: 10.5858/ arpa.2011-0430-OA) tion between these 2 groups of patients has implications for their counseling and clinical management. On purely pathologic grounds, the characterization of those specific histomorphologic and immunohistologic features associated with chronicity would increase the biological understanding and the diagnostic accuracy of the disease. The near-normal spectrum of changes evokes microscopic colitis (MC) and other related entities. Patients with MC have persistent, watery, bloodless diarrhea, a normal colonoscopy, and chronic inflammation with other specific histologic changes.4,5 Microscopic colitis comprises 2 entities, namely collagenous colitis (CC) and lymphocytic colitis (LC), which are indistinguishable clinically but are separated by histopathologic characteristics.4–6 Collagenous colitis is characterized by the presence of a subepithelial collagen band (10 mm or more) adjacent to basal membrane,4–6 whereas LC biopsies show more than 20 surface epithelial lymphocytes per 100 epithelial cells.4–6 In addition, many atypical histopathologic forms of MC and LC have been described,7,8 such as MC not otherwise specified,9,10 MC with giant cells,9 and paucicellular,10 borderline,11 and cryptal LC,12 all of which show more subtle histomorphologic changes than classical MC or LC. The aims of this study are (1) to characterize the clinicopathologic features of chronic diarrhea with no clear etiology and nonspecific changes on colonic biopsy and (2) to search for specific histologic and immunohistochemical characteristics that may indicate (retrospective) Lymphocytic Follicles and Aggregates—Shah et al

83

and/or predict diarrhea.

(prospective)

long-standing

chronic

MATERIALS AND METHODS Study Group The original study group consisted of 75 patients (Alexandra and National University Hospitals, Singapore) who had undergone colonoscopy for chronic diarrhea (defined as 3 or more loose motions a day for more than 1-month duration13) between August 2002 and March 2005. Clinical and endoscopic appraisal of these patients was done by an experienced gastroenterologist; the biopsy materials were consensually reviewed by 2 gastrointestinal pathologists without the knowledge of the patients’ clinical or endoscopic findings. Of the initial 75 cases, 47 were finally entered in the study once all were reviewed in detail and there was unequivocal evidence of sufficient clinicopathologic information, near-normal histologic appearance, and no clear-cut known etiologic agent. Immunohistochemistry was performed on 39 cases (because paraffin blocks of 8 of the 47 cases were exhausted) and evaluated by an experienced pathologist. Institutional Review Board approval was granted from the collaborating institutions. Follow-up information was available for 39 patients.

Control Group Control group colonic biopsies were obtained from segments of absolute normal-looking mucosa of colorectal cancer colectomy resections in patients with no diarrhea or colitis at their clinical presentation. The samples were taken from the superficial portions of the mucosa, and were of a size akin to that of conventional colonic biopsies, with a mean of 6.7 biopsies per control case (range, 5–12).

Clinical Data The clinical information collected included duration of diarrhea, presence/absence of rectal bleeding, other associated illness, colonoscopic findings, other investigations (particularly stool culture and thyroid function test), treatment, and clinical outcome. The mean follow-up was 4.74 6 5.46 months (range, 1– 22 months). The clinical course was categorized as resolved (cessation of diarrhea with or without empirical treatment in the form of probiotics, fiber supplement, Lomotil, and loperamide), unresolved (diarrhea persisted with or without partial improvement), or lost to follow-up.

Histology Analysis A total of 358 single colonic biopsy tissues were obtained from 47 patients (mean, 7.3 per patient; range, 5–12). The following histologic parameters were examined according to previous studies14–23: 1. The number of surface intraepithelial lymphocytes per 100 epithelial cells, counted in areas of high lymphocytic density in a sequential manner and avoiding epithelium overlying lymphoid follicles.15 2. Increased chronic inflammation, characterized as reversal of the normal mucosal plasma cell gradient and/or a continuous layer of lymphoplasmacytic inflammatory infiltrate from epithelium to muscularis mucosae.16 3. The number of lymphoid follicles and aggregates in each biopsy.17,18 Lymphoid follicles were defined as aggregates of lymphocytes with a germinal center. A lymphoid aggregate was defined as accumulation of lymphocytes and plasma cells without a germinal center. The lymphoid follicles and aggregates were categorized in 3 groups: (1) sub-epithelial small, present just below the epithelium and occupying half or less of the lamina propria; (2) basal small, present just above the muscularis mucosae and occupying half or less of 84

Arch Pathol Lab Med—Vol 137, January 2013

the lamina propria; and (3) large, occupying more than half the thickness of the lamina propria. The density of lymphoid follicles and aggregates was calculated by counting the number of biopsy specimens with lymphoid follicles and aggregates versus the number of biopsy specimens examined. The total number of lymphoid follicles and aggregates per biopsy was also analyzed. 4. Eosinophils of the surface epithelium and lamina propria were graded separately on a scale from 0 to 3 (0, no eosinophils; 1, rare eosinophils; 2, moderate number of eosinophils; 3, marked increase in eosinophils).19 5. Neutrophils of the surface epithelium and lamina propria were graded in the same way as eosinophils.19 6. Surface epithelial injury was graded on a scale from 0 to 3 (0, no visible injury; 1, mucin-depleted epithelial cells with columnar configuration; 2, mucin-depleted epithelial cells with cuboidal configuration; 3, mucin-depleted flattened epithelial cells).20 7. Crypt distortion was categorized as 1, crypt atrophy (presence of inflammatory infiltrate between muscularis mucosae and crypt); 2, pointed crypts (triangular diamond-shaped, or quadrangular crypts); and 3, other architectural irregularities (2 or more than 2 branched crypts and/or cystic crypts).19,20 8.Muscularis mucosae duplication, defined as apparent thickening of muscularis mucosae with flaying of the fascicles.21 9. Lamina propria fibrosis, described as presence of distinct bands of collagen.22 A subepithelial collagen table was searched for. 10. Presence or absence of mucosal edema, surface ulceration, pseudolipomatosis, giant cells,9 and muciphages.23 The pattern of distribution for features such as inflammation, surface epithelial injury, crypt injury, lamina propria fibrosis, and mucosal edema was categorized as diffuse (present in the whole area of a biopsy and in all biopsies) or focal (present in some areas of a biopsy and/or in few of the biopsies).19 Crypt distortion was considered diffuse if more than 50% of crypts were affected and was considered focal if less than 50% of crypts were affected.17,20

Immuohistochemistry Additional sections were analyzed with FOXP3 (forkhead transcription factor; dilution 1:500; clone 236/E7, Abcam, Cambridge, UK) and CD117 (c-Kit; dilution 1:400; clone A4502, Dako, Glostrup, Denmark) antibodies, targeting regulatory T (Treg) cells and mast cells respectively. Briefly, formalin-fixed, paraffin-embedded 4-mm sections were mounted on precoated slides, dewaxed, subjected to automated antigen retrieval with EDTA buffer (pH 9.0), and immunostained in the Bond-MAX automated immunostainer (Leica, Wetzlar, Germany) according to manufacturer’s recommendations.

Cell Counting Treg+ lymphocytes and c-Kit+ mast cells were counted in 4 well-oriented, nonoverlapping high-power fields by a Leica DM 1000 microscope (field area 5 0.52 mm2). Mast cells showed cytoplasmic staining and Treg cells showed nuclear staining. The Treg and c-Kit densities represented the average of positive cells per high-power field, as described previously.24

Statistics The statistical analysis was performed using SPSS v.14.0 (SPSS Inc, Chicago, Illinois). The associations between the outcome and the clinicopathologic and immunohistochemistry parameters were evaluated by the x2 test or the Mann-Whitney U test. The correlations between duration of diarrhea and the categorical Lymphocytic Follicles and Aggregates—Shah et al

Table 1.

Demographic, Clinical, and Colonoscopic Characteristics of the Patients Features

Results

M/F, No. (ratio) Age, y, median (range) Race Chinese, No./Total (%) Indian, No./Total (%) Malay, No./Total (%)

26/21 (1.2:1) 44 (22–77) 40/47 (85) 4/47 (9) 3/47 (6)

Symptoms and signs Duration of diarrhea before colonoscopy, mo, median (range) Watery diarrhea, No./Total (%) Rectal bleeding, No./Total (%) Weight loss, No./Total (%)

38/47 (81) 9/47 (19) 8/47 (17)

Laboratory and endoscopic investigations Stool culture, No./total (%) Positive stool culture, No. Negative stool culture, No./total

24/47 (51) 0 24/24

Thyroid function test Hypothyroid, No./total Hyperthyroid, No./total Abnormal endoscopy, No./total (%) Rectal pseudopolyps, No./total Erythema of colon and rectum, No./total Diverticular disease of colon, No./total Cecal polyp, 3-mm size, No./total Slightly nodular ileum, No./total Treatment received, No./total (%) Autoimmune disease Hashimoto thyroiditis, No./total

4 (1–432)

1/32 1/32 8/47 (17) 1/8 3/8 2/8 1/8 1/8 26/47 (55) 1/47

Outcome Diarrhea resolved, No./total (%) 26/39 (67) Diarrhea unresolved, No./total (%) 13/39 (33) Lost to follow-up, No./total (%) 8/47 (17) Duration of follow-up, mo, mean (range) 4.74 6 5.46 (1–22)

variables were assessed by the Mann-Whitney U test, and the correlations between duration of diarrhea and the numerical variables were analyzed by the Spearman correlation coefficient. The correlation between study and control cases was analyzed by the Mann-Whitney U test.

RESULTS Clinical and Endoscopic Results Demographic, clinical, and colonoscopic data are summarized in Table 1. The study cohort predominantly consisted of Chinese patients (85%), and a slight majority (55%) were male. The median age of the patients was 44 years (range, 22–77). Twenty-four patients had stool cultures done; all were negative for bacterial growth and parasites. None of these patients had a history of inflammatory bowel disease or irritable bowel syndrome or developed them during the course of the follow-up. Twenty-six patients received empirical treatment in the form of probiotics, fiber supplement, Lomotil, and loperamide. In addition, 2 received antibiotics and 1 received prednisolone. None of the patients received nonsteroidal anti-inflammatory drugs or proton pump inhibitors (lansoprazole). No association with persistence of diarrhea was found between the treated and the untreated group. Eight of the 47 patients were lost to follow-up. Out of 39 patients, diarrhea was resolved in 67% and unresolved in 33%. Arch Pathol Lab Med—Vol 137, January 2013

Histology Results The frequency and distribution of the histology findings are described in Table 2. Lymphoid follicles and aggregates were present in the majority of cases (96%) with median count of 4 (range, 0–10) and mean density of 0.49 6 0.22. Large was the commonest subtype of lymphoid follicles and aggregates, with median count of 3 (range, 0– 10); the values were significantly higher in the study group than in the control group (total lymphoid follicles and aggregates range, 0–4, P 5 .003; mean density of lymphoid follicles and aggregates, 0.23 6 0.14, P 5 .001; large lymphoid follicles and aggregates range, 0–3, P , .001). Basal small lymphoid follicles and aggregates were present in 25% of cases. The mean total lymphoid follicles and aggregates per biopsy was 0.59 6 0.32 for study cases and 0.31 6 0.21 for controls (P 5 .005) (Table 3). The different subtypes of lymphoid follicles and aggregates are illustrated in Figure 1. Chronic inflammation was present in all of the study cases. This is not surprising in that all the cases of our study had ‘‘nonspecific changes’’ to start with, which, in most instances, included a mild, patchy, nonspecific increase in chronic inflammation as perceived by the reporting pathologist. The pattern of chronic inflammation was diffuse in 12 cases (26%) and focal or patchy in 35 cases (74%). The mean number of surface epithelial lymphocytes per 100 epithelial cells was 12, which is significantly higher than in idiopathic inflammatory bowel disease, acute colitis, and normal colonic mucosa epithelium,11,15 but lower than the diagnostic criterion for LC.4,5 All cases had less than 20 surface epithelial lymphocytes per 100 epithelial cells, except for 3 cases having counts of 20, 21, and 23; overall, these counts were lower than the mean counts reported in classic LC in other studies (29.3 by Goldstein et al10 and 24.6 by Lazenby et al11). A subepithelial collagen band was not present in any of the cases. Muscularis mucosae thickening was present in 50% of the 44 assessable cases; in 5 cases muscularis mucosae duplication was not assessable because of partial thickness of the muscularis mucosae in the section or improper orientation of the colon mucosa. Crypt distortion was noted in the form of pointed crypts in all 47 cases, diffuse in 11 cases (23%) and focal in 36 cases (77%). Six patients had a moderate number of eosinophils, predominantly in the lamina propria, but none exhibited marked increased numbers. The presence of neutrophils in the mucosa was detected in 9 cases (minimal in 6, moderate in 3). Comparison of Histology Features With Clinical Findings The duration of diarrhea was correlated to the following parameters: number of large lymphoid follicles and aggregates (r 5 .48, P 5 .001), number of total lymphoid follicles and aggregates (r 5 .43, P 5 .003), density of lymphoid follicles and aggregates (r 5 .38, P 5 .009). The median number of total lymphoid follicles and aggregates for patients with resolved diarrhea was less than that for patients with unresolved diarrhea (4 versus 6). However, in 6 (28.2%) patients with unresolved diarrhea, fewer than 6 lymphoid follicles and aggregates were observed. The median density of lymphoid follicles and aggregates for patients with resolved diarrhea was less than that for patients with unresolved diarrhea (50% versus 60%), but in 8 (37.6%) patients with unresolved diarrhea, lymphoid Lymphocytic Follicles and Aggregates—Shah et al

85

Table 2.

Summary of Frequency of Occurrence and Distribution of Various Histologic Features in Study Population Pattern of Distribution Histologic Feature

,20 surface epithelial lymphocytes/100 epithelial cells $20 surface epithelial lymphocytes/100 epithelial cells Surface epithelial injury Chronic inflammation Neutrophilic infiltration Mild (6/9) Moderate (3/9) Severe (0/9) Eosinophilic infiltration Mild (0/6) Moderate (6/6) Severe (0/6) Lymphoid follicles and aggregates More than 3 large lymphoid follicles and aggregates or more than 4 total lymphoid follicles and aggregates Edema Apoptotic bodies Pseudoliposis Lamina propria fibrosis Minimal crypt distortion Muscularis mucosae duplication Muscularis mucosae duplication, not assessable

follicle and aggregate density of less than 60% was observed. There was no relation between the age of the patient and the number of lymphoid follicles and aggregates identified. Comparison of Immunohistochemistry and Clinical Findings The median number of mast cells was higher in unresolved cases than in resolved cases, and also higher in the study group than in the control group (P 5 .001 and P 5 .008, respectively). Similarly, a significant difference was seen in the number of Treg cells between resolved and unresolved cases (P 5 .01), and also between the study group and the control group (P 5 .04). In addition, the duration of diarrhea was significantly related to the number of mast cells (r 5 .52, P 5 .001), but not to Treg cells (r 5 .05, P 5 .75). Expression of CD117 and FOXP3 is illustrated in Figure 2, and the comparisons of CD117 and FOXP3 scores for control, resolved, and unresolved groups are shown in Figures 3 and 4 respectively. COMMENT Nonspecific changes/inflammation amounts to a significant percentage of the gastrointestinal pathology diagnostic practice that, in many cases, represents chronic inflammatory cells in lamina propria, with or without subtle cytologic or architectural alteration.25 The results of this study suggest that lymphoid follicles and aggregates in colonic biopsies are evidence of gut mucosa damage and may help to predict the resolution and/or duration of diarrhea. Lymphoid follicles and aggregates have been postulated as a normal component in colonic biopsies. Whenever they have been perceived as being increased in numbers, it was always when accompanying other specific features that would allow a diagnosis of ulcerative colitis, Crohn disease, MC, or lymphoma.14 In our study, the total number of lymphoid follicles and aggregates and the 86

Arch Pathol Lab Med—Vol 137, January 2013

No. of Patients (%)

44/47 3/47 26/47 47/47 9/47

(94) (6) (55) (100) (18)

Diffuse

Focal

6 12

20 35

2

22

5 19

24 28

6/47 (12)

45/47 (96) 26/47 (55) 24/47 45/47 4/47 29/47 47/47 22/44 5/47

(51) (96) (8) (62) (100) (50) (11)

number of large lymphoid follicles and aggregates (the commonest subtype) are found to be significantly increased compared to the control group. As with the above diseases, an increased number of lymphoid follicles and aggregates in our study is a marker of gut mucosal damage. Furthermore, these features showed a significant correlation with the duration of diarrhea, as well as with its resolved or unresolved status (the latter being a trend, with no statistical significance, presumably because of the relatively small sample size of the study). Thus, we propose a more specific diagnostic relevance of the presence of lymphoid follicles and aggregates in the clinicopathologic context of chronic diarrhea and histologic features of subtle, general mucosal injury. In this framework, a biopsy specimen with more than 3 large lymphoid follicles and aggregates or more than 4 total (small or large) lymphoid follicles and aggregates in a biopsy sample of 5 or more tissue portions representative of the full colonic length is an indicator of mucosal damage of unknown etiology and a predictor of the length of the symptoms. To our knowledge, this is the first study reporting the link of lymphoid follicles and aggregates and unresolved diarrhea in the background of nonspecific histologic changes. In addition, the statistically significant differential expression of established immune-related markers (paucity of Treg cells and increased number of mast cells), both in the study group and in the unresolved group, when lymphoid follicles and aggregates are present, may represent a unique form of colonic immune response. The isolated lymphoid follicles, along with Payer patches and immune lymphocytes, are normal components of the mucosal immune system.26 In animal models, the presence and numbers of isolated lymphoid follicles has been related to the response to intraluminal antigen.27 The role of mast cells is well known in allergic and parasitic inflammation, and increasingly so in the regulation of epithelial barrier, mucosal immune function, Lymphocytic Follicles and Aggregates—Shah et al

Figure 1. Representation of lymphoid follicles and aggregates. A, Large lymphoid aggregates. B, Basal lymphoid aggregates. C, Subepithelial lymphoid aggregates. D, Lymphoid follicle. E, Low-power appearance of multiple lymphoid aggregates in multiple biopsies in same sample. F, A lymphoid follicle and a lymphoid aggregate coexisting in same biopsy (hematoxylin-eosin, original magnifications 3200 [A through D], 325 [E], and 350 [F]).

motility, and gut visceral sensitivity.28,29 The role of Treg cells has also been described in diabetes, inflammatory bowel disease, autoimmune gastritis, and thyroiditis.30,31 Their role of maintaining tolerance in the gut has been described in an animal model. Thus, one could argue that lymphoid follicles and aggregates are the most prominent Table 3.

histologic feature of an overall abnormal immune response. The proposed morphoimmunologic spectrum is, in some ways, unique. It differs from that of infective colitis, as none of these patients had marked neutrophilic infiltration, crypt abscess, or positive stool culture.2,3 It

Comparison of Lymphoid Follicles and Aggregates (LFA) of Study Cases and Control Group Study group (n = 47)

No. of large lymphoid follicles and aggregates, median (range) Total No. of lymphoid follicles and aggregates, median (range) Mean density of lymphoid follicles and aggregates, No. specimens with LFA/No. specimens examined Mean total lymphoid follicles and aggregates per biopsy fragment Arch Pathol Lab Med—Vol 137, January 2013

Control group (n = 11)

P

3 (0–10)

1 (0–3)

,.001

4 (0–10)

2 (0–4)

.003

0.49 6 0.22

0.23 6 0.14

.001

0.59 6 0.32

0.31 6 0.21

.005

Lymphocytic Follicles and Aggregates—Shah et al

87

Figure 2. Expression of CD117 and FOXP3 in lamina propria. A, Few CD117-positive mast cells (shown by arrow) in resolved diarrhea; inset, CD117-positive mast cell. B, Increased number of CD117-positive mast cells (shown by arrow) in unresolved diarrhea; inset, CD117-positive mast cells. C, Occasional FOXP3-positive lymphocytes (shown by arrow) in unresolved diarrhea; inset, FOXP3-positive lymphocyte. D, Increased number of FOXP3-positive lymphocytes in (shown by arrow) resolved diarrhea; inset, FOXP3-positive lymphocytes (original magnifications 3400 [A through D] and 31000 [insets, A through D]).

lacks clinical and histodiagnostic features of inflammatory bowel disease.3 None of these patients carried the Rome criteria, which are required for diagnosis of irritable bowel syndrome.3,24 Although MC sensu stricto embraces CC and LC,4–6 the results of our study suggest a specific mucosal response, perhaps in predisposed individuals, to various noxious luminal agents, leading to mucosal injury and concomitant immune dysfunction not unlike that of CC and/or LC. Indeed, the role for lymphoid follicles and aggregates has occasionally been suggested in established CC or LC. Ayata et al,17 in a review of 79 LC and 71 CC cases, found no evidence of basal lymphoid aggregates, but others suggest that lymphoid follicles and aggregates are an early event in the course of CC and LC.18 Although it would be interesting to hypothesize that our observation could be better placed in the MC spectrum, our patients differ clinically by (1) slightly male predominance; (2) a younger mean age; (3) and lack of association with 88

Arch Pathol Lab Med—Vol 137, January 2013

autoimmune disease and nonsteroidal anti-inflammatory drug intake. In any case, this should be placed in the limiting context of the relative small sample size of this study. Lymphoid nodular hyperplasia,32,33 characterized by the presence of abundant numbers of large lymphoid follicles throughout the colon, is a different entity altogether, in that (1) it tends to occur in children, (2) it has a clear tendency to represent an allergic reaction, or represent part of a non-Hodgkin lymphoma (then, it does occur in adults); and (3) it is so aggressive, even in the nonneoplastic cases, that it may result in physical obstruction and the need for surgery. In summary, the recognition of lymphoid follicles and aggregates may help in the future characterization of patients with diarrhea showing an immunologic pattern. Their recognition may help to increase our clinical, pathologic, and etiologic understanding of nonspecific Lymphocytic Follicles and Aggregates—Shah et al

Figure 3. Box plots for CD117 score in control, resolved, and unresolved groups (P 5 .008, control versus resolved and unresolved diarrhea cases; P 5 .001, resolved versus unresolved group). Abbreviation: hpf, high power field. Figure 4. Box plots for FOXP3 score in control, resolved, and unresolved groups (P 5 .04, control versus resolved and unresolved diarrhea cases; P 5 .01, resolved versus unresolved group). Abbreviation: hpf, high-power field.

chronic diarrhea and open the way for more targeted studies on the etiology of, arguably, the commonest clinicopathologic picture in the routine colonoscopic practice. References 1. McKenna BJ. Is it really colitis?: dealing with the nearly normal colonic biopsy and variations of microscopic colitis. Pathol Case Rev. 2004;9(2):106– 114. 2. Spiller RC, Jenkins D, Thornley JP, et al. Increased rectal mucosal enteroendocrine cells, T lymphocytes, and increased gut permeability following acute Campylobacter enteritis and in post-dysenteric irritable bowel syndrome. Gut. 2000;47(6):804–811. 3. Jenkins D, Goodall A. Simple objective criteria for diagnosis of causes of acute diarrhea on rectal biopsy. J Clin Pathol. 1997;50(7):580–585. 4. Fraser AG, Warren BF, Chandrapala R, Jewell DP. Microscopic colitis: a clinical and pathological review. Scand J Gastroenterol. 2002;37(11):1241– 1245.

Arch Pathol Lab Med—Vol 137, January 2013

5. Nyhlin N, Bohr J, Eriksson S, Tysk C. Microscopic colitis: a common and an easily overlooked cause of chronic diarrhoea. Eur J Intern Med. 2008;19:181– 186. 6. Tangri V, Chande N. Microscopic colitis: an update. J Clin Gastroenterol. 2009;43(4):293–296. 7. Chang F, Deere H, Vu C. Atypical forms of microscopic colitis: morphological features and review of the literature. Adv Anat Pathol. 2005; 12(4):203–211 8. Jawhari A, Sheaf M, Forbes A, Kamm MA, Talbot IC. Microscopic colitis: widening the definition. Gastroenterology. 1995;108:843A. 9. Brown IS, Lambie DL. Microscopic colitis with giant cells: clinicopathological review of 11 cases and comparison with microscopic colitis without giant cells. Pathology. 2008;40(7):671–675. 10. Goldstein NS, Bhanot P. Paucicellular and asymptomatic lymphocytic colitis: expanding the clinicopathologic spectrum of lymphocytic colitis. Am J Clin Pathol. 2004;122(3):405–411. 11. Lazenby AJ, Yardley JH, Giardiello FM, Jessurun J, Bayless TM. Lymphocytic (‘‘microscopic’’) colitis: a comparative histopathologic study with particular reference to collagenous colitis. Hum Pathol. 1989;20(1):18–28. 12. Rubio CA, Lindholm J. Cryptal lymphocytic coloproctitis: a new phenotype of lymphocytic colitis? J Clin Pathol. 2002;55(2):138–140. 13. Schiller LR. Diarrhea. Med Clin North Am. 2000;84(5):1259–1274. 14. Talbot I, Price AB, Salto-Tellez M. Assessment of abnormalities: diagnostic signposts. In: Talbot I, Price A, Salto-Tellez M, eds. Biopsy Pathology in Colorectal Disease. 1st ed. London, England: Hodder Arnold; 2006:14–53. 15. Levine DS, Haggitt RC. Normal histology of the colon. Am J Surg Pathol. 1989;13(11):966–984. 16. Warren BF, Edwards CM, Travis SP. ‘‘Microscopic colitis’’: classification and terminology. Histopathology. 2002;40(4):374–376. 17. Ayata G, Ithamukkala S, Sapp H, et al. Prevalence and significance of inflammatory bowel disease-like morphologic features in collagenous and lymphocytic colitis. Am J Surg Pathol. 2002;26(11):1414–1423. 18. Shaz BH, Reddy SI, Ayata G, et al. Sequential clinical and histopathological changes in collagenous and lymphocytic colitis over time. Mod Pathol. 2004; 17(4):395–401. 19. Wang N, Dumot JA, Achkar E, Easley KA, Petras RE, Goldblum JR. Colonic epithelial lymphocytosis without a thickened subepithelial collagen table: a clinicopathologic study of 40 cases supporting a heterogeneous entity. Am J Surg Pathol. 1999;23(9):1068–1074. 20. Warren BF, Dankwa EK, Davies JD. ‘‘Diamond-shaped’’ crypts and mucosal elastin: helpful diagnostic features in biopsies of rectal prolapse. Histopathology. 1990;17(2):129–134. 21. Ridell RH, Montgomery E. Traps in the diagnosis of ulcerative colitis. Pathol Case Rev. 2004;9(3):87–92. 22. Lazenby AJ, Yardley JH, Giardiello FM, Bayless TM. Pitfalls in the diagnosis of collagenous colitis: experience with 75 cases from a registry of collagenous colitis at the Johns Hopkins Hospital. Hum Pathol.1990;21(9):905– 910. 23. Salto-Tellez M, Price AB. What is the significance of muciphages in colorectal biopsies?: muciphages and other mucosal accumulations in the colorectal mucosa. Histopathology. 2000;36(6):556–559. 24. Dunlop SP, Jenkins D, Spiller RC. Distinctive clinical, psychological and histological features of postinfective irritable bowel syndrome. Am J Gastroenterol. 2003;98(7):1578–1583. 25. Haboubi NY, Kamal F. Non-specific colitis, is it a justifiable diagnosis? Colorectal Dis. 2001;3(4):263–265. 26. Lorenz RG, Chaplin DD, McDonald KG, et al. Isolated lymphoid follicle formation is inducible and dependent upon lymphotoxin-sufficient B lymphocytes, lymphotoxin beta receptor, and TNF receptor I function. J Immunol. 2003; 170(11):5475–5482. 27. Little MC, Bell LV, Cliffe LJ, Else KJ. The characterization of intraepithelial lymphocytes, lamina propria leukocytes, and isolated lymphoid follicles in the large intestine of mice infected with the intestinal nematode parasite Trichuris muris. J Immunol. 2005;175(10):6713–6722. 28. Guilarte M, Santos J, Torres I, et al. Diarrhea-predominant IBS patients show mast cell activation and hyperplasia in the jejunum. Gut. 2007;56(2):203–209. 29. Santos J, Yang PC, Soderholm JD, Benjamin M, Perdue M. Role of mast cells in chronic stress induced colonic epithelial barrier dysfunction in the rat. Gut. 2001;48(5):630–636. 30. Buckner JH. Mechanisms of impaired regulation by CD4+CD25+FOXP3+ regulatory T cells in human autoimmune diseases. Nat Rev Immunol. 2010; 10(12):849–859. 31. Ochs HD, Gambineri E, Torgerson TR. IPEX, FOXP3 and regulatory Tcells: a model for autoimmunity. Immunol Res. 2007;38(1–3):112–121. 32. Iacono G, Ravelli A, Di Prima L, et al. Colonic lymphoid nodular hyperplasia in children: relationship to food hypersensitivity. Clin Gastroenterol Hepatol. 2007;5(3):361–366. 33. Shuhaiber J, Jennings L, Berger R. Nodular lymphoid hyperplasia: a cause for obscure massive gastrointestinal bleeding. J Pediatr Surg. 2005;40(4):E17– E19.

Lymphocytic Follicles and Aggregates—Shah et al

89