Printer Friendly

Performing Colonic Mast Cell Counts in Patients With Chronic Diarrhea of Unknown Etiology Has Limited Diagnostic Use.

The American Gastroenterology Association defines chronic diarrhea as the production of liquid or loose stools for longer than 4 weeks, with or without increased stool frequency. (1) Common causes include inflammatory bowel disease, malabsorption syndromes, microscopic colitis, drugs, and infections. However, in many cases, the etiology cannot be ascertained, and the patient is given the diagnosis of diarrhea-predominant irritable bowel syndrome, according to the Rome III criteria (2) for the classification of functional gastrointestinal disorders. Diarrhea-predominant irritable bowel syndrome remains a diagnosis of exclusion, and the lack of diagnostic biomarkers limits the ability of clinicians to study and recognize diarrhea-predominant irritable bowel syndrome in a systematic manner.

Recently, Jakate et al (3) found increased mast cells (MCs) in duodenal and colonic mucosal biopsies from patients with chronic diarrhea, a subset (45%, 21 of 47) of which were clinically diagnosed with possible diarrhea-predominant irritable bowel syndrome. The authors proposed the novel term mastocytic enterocolitis to describe this population of patients with chronic diarrhea, normal mucosal histology on routine hematoxylin-eosin sections, increased mucosal MCs demonstrated by immunohistochemistry, and effective treatment with MC stabilizers ([H.sub.1] and [H.sub.2] receptor antagonists, with or without a MC mediator release inhibitor). The cutoff value of more than 20 MCs per high-power field (hpf) was used to define patients with increased MCs because that represented 2 standard deviation (SD) beyond the normal mean values for the colon and duodenum (13.6 and 13.2 MCs/hpf, respectively). However, a major limitation of that study was a lack of a comparison control group. Several years later, another study, (4) which described a group of 24 patients with clinically demonstrated gastrointestinal dysmotility, nocturnal awakening attributed to uncontrolled asthma, elevated histamine levels, and a history of food or environmental allergy, was published. Those 24 patients had increased mucosal MCs at levels more than twice (mean [SD]; colon, 37 [17] MCs/hpf; small intestine, 53 [18] MCs/hpf) those reported by Jakate et al. (3) The higher MC counts and notable atopic clinical presentations prompted the authors to suggest another entity, allergic mastocytic gastroenteritis.

These articles and others, (5-7) which suggest a role for MCs in patients with chronic diarrhea of unknown etiology (CDUE), have prompted some clinicians to request that MC stains be performed on colon biopsies with normal histology to investigate the possibility of MC-related chronic diarrhea. In addition, some authors (5-7) have advocated for colonoscopic biopsies with special MC stains as part of the routine diagnostic workup in patients with chronic, unexplained diarrhea. Those recommendations are based on intriguing but minimal and often poorly controlled evidence, and a universally accepted cutoff for significant MC counts remains unknown. Thus, the diagnostic utility of routine MC stains in colon biopsies remains unclear and requires further investigation.

The aims of this study were to (1) determine whether colonic MCs are increased in patients with CDUE, who otherwise have normal histologic findings; (2) investigate the microscopic distribution of colonic MCs to determine whether biopsy location is an important factor to consider when interpreting MC counts; and (3) gain a better understanding of how MC counts might be interpreted based on our data and recent literature that examines the subject. To our knowledge, this is the largest study to date to examine the diagnostic utility of colonic MCs in patients with CDUE.

MATERIALS AND METHODS

The Yale New Haven Hospital (New Haven, Connecticut) pathology database was searched for a 1-year period (July 2010 to July 2011) for all patients who had presented with a clinical history of chronic diarrhea. The pathology reports were reviewed, and 404 patients noted to have normal colon biopsy results were selected for possible inclusion in the study group. Patients were excluded if concurrent duodenal or gastric biopsies were reported to have malabsorption features or moderate to severe gastritis. In addition, patients with concurrent esophageal diagnoses were also reviewed, and patients with eosinophilic esophagitis were excluded from the study group (slides not reviewed).

Of the 404, 153 patients (37.9%) identified were included in the potential study group. Those patients had undergone the investigational protocol recommended by the American Gastroenterology Association for the workup of chronic diarrhea, as determined by retrospective review of the medical charts. The components of the recommended workup included a past medical history review, a physical exam, a medication history, a travel history, diet, and comprehensive infectious disease testing. Seventy-seven additional patients from the 153 (50.3%) were excluded from the study group because a possible cause of chronic diarrhea was identified during chart review. A group of 76 patients was, therefore, identified who met our criteria for CDUE with normal colon biopsy results.

From the same period, 89 control patients were selected, consisting of consecutive, asymptomatic patients who presented for screening colonoscopy and who had biopsies taken of healthy colonic mucosa that was perceived as a polyp by the endoscopist.

Routinely processed, formalin-fixed, paraffin-embedded, hematoxylin-eosin-stained tissue sections obtained from all study group and control patients were reviewed to confirm the diagnosis of normal colonic mucosa. Immunohistochemical analysis with the c-Kit (CD117) antibody (catalog no. A4502, 1:200 dilution, Dako, Carpinteria, California) was performed on all biopsies to stain the MCs. Antigen retrieval was performed using citrate buffer at pH 6.0. One observer (J.A.H.) performed blinded counts of mucosal MCs. The MCs were counted in a single X400 hpf (0.55 mm diameter) in the area of highest density.

The MC counts were compared between the patients in the study group and those in the control group, independent of biopsy location. Comparisons were also made between the left and right side of the colon. A 2-sample t test was used for all comparisons. Using logistic regression, receiver operating characteristic (ROC) curves were generated to examine putative sensitivity and specificity cutoff values. A 1-way analysis of variance examined the relationship between MC counts and the severity of diarrhea. In addition, the sex and age of patients and controls were examined to determine how these influenced MC counts. Overall MC counts were analyzed on a log scale because the distribution of counts was skewed to the right. Patient groups had some younger individuals than control groups for both men/women, so analyses were adjusted by age. Patients were compared with controls and males to females in a 2-way analysis of variance with age as an adjusting covariate (ie, with analysis of covariance).

The study patients were stratified according to the severity of their diarrhea as follows: mild, 2 to 3 loose stools per day (n = 21; 28%), moderate, 3 to 5 loose stools per day (n = 43; 57%), and severe, more than 5 loose stools per day (n = 12; 16%).

RESULTS

Seventy-six patients with CDUE were identified. The mean age of patients in the study group was 50 years (range, 18-81 years) (Table 1). There were 48 women (63%) and 28 men (37%). Forty-seven patients (62%) had random biopsies only. Eighteen patients (24%) had designated right-and left-sided biopsies. Two patients (3%) had biopsies from the right side of the colon but no designated biopsies from the left side. Nine patients (12%) had biopsies from the left side of the colon but no designated biopsies from the right side. Review of the patients' medical charts did not reveal a cause of diarrhea.

The mean age of the patients in the control group (n = 89) was 57 years (range, 36-78 years) (Table 1). There were 41 women (46%) and 48 men (54%). Four control patients (4%) had random biopsies only. Six control patients (7%) had both right- and left-sided biopsies. Forty-one control patients (46%) had biopsies from the right side of the colon but no designated biopsies from the left side. Thirty-eight control patients (43%) had biopsies from the left side of the colon but no designated biopsies from the right side.

For all colon biopsies, independent of site, the patients in the control group had a mean (SD) 24 (9) MCs/hpf versus 31 (11) MCs/hpf for patients in the study group (P < .001) (Table 2; Figures 1, A through D, and 2, A through D, respectively). In many biopsies, MCs were noted to be concentrated in the deeper part of the lamina propria (Figures 2, D, and 3, A and B), although that observation was not systematically studied because tangential sections were often difficult to interpret. An ROC curve demonstrated that, when MC counts for all biopsy sites were considered, the counts did not yield discriminatory cutoff values (area under the curve, 0.68) (Figure 4).

When biopsies were stratified by location, MCs were significantly increased among study patients in the left, but not right, side of the colon, when compared with the control group. Left colon MCs increased from a mean (SD) of 22 (9) MCs/hpf in the control patients to 31 (16) MCs/hpf in the study patients (P = .01) (Table 3). Right colon MC counts, on the other hand, had a mean (SD) of 28 (11) MCs/hpf in the study patients and 25 (9) MCs/hpf in the control patients (P = 0.32) (Table 3). When only left-sided MC counts were considered, the ROC curve was slightly more discriminatory compared with the curve generated for all biopsy sites (area under the curve, 0.74) (Figure 5). As done by previous authors, we established an arbitrary cutoff value of 2 SD above the left-sided control patient mean as "abnormal" and, thus, indicative of increased MCs. (3) Doing so yielded a value of 39 MCs/hpf. That value was then used to calculate sensitivity, specificity, positive predictive value, and negative predictive value in our study group patients with the momentary assumption that counts greater than 39 MCs/ hpf were diagnostic of MC-related diarrhea. The results were as follows: sensitivity, 19%; specificity, 98%; positive predictive value, 83%; and negative predictive value, 66%.

In study group patients who had biopsies on both the right and left sides of the colon (n = 18; 24%), the mean difference between right- and left-sided MCs counts (right minus left) was not significant (P = .68) (Table 4). In control patients, the mean difference (also right counts minus left) was significant (P = .02), despite a low number of patients (n = 6; 8%) (Table 4). The difference in right and left MC counts generated the most discriminatory ROC curve (area under the curve, 0.81) (Figure 6). We established an arbitrary cutoff value of 2 SD below the control mean difference as "abnormal" and, thus, indicative of increased MCs and arrived at a value of -5 MCs/hpf. That value was then used to calculate sensitivity, specificity, positive predictive value, and negative predictive value in our study group patients, again, with the momentary assumption that the MC count differences between left- and right-sided colon of more than 5 MCs/hpf was diagnostic of MC-related diarrhea. The results were as follows: sensitivity, 33%; specificity, 100%; positive predictive value, 100%; negative predictive value, 33%.

Overall and left-sided MC counts did not significantly differ with diarrhea severity, as stratified by the severity score previously described (overall counts, F = 0.99, P = .37; left counts, F = 0.86, P = .43). In the analysis of covariance, overall MC counts showed no significant 2-way interactions, and age was not significant despite the study group having some younger subjects than the control group. Higher MC counts were seen in women compared with men (P = .04). Table 5 provides a classification of the groups by sex. There was no statistical interaction between sex and group, meaning that MC counts increased proportionally in the study group for both men and women, but that increase was not influenced by sex (Figure 7).

COMMENT

Recent studies (3,4,8) have suggested a possible role for MCs in some cases of chronic diarrhea. Mast cells are bone marrow-derived cells that are located proximate to nerves and blood vessels in the gastrointestinal tract. (5,9,10) In addition to their well-known role in immunity and allergic reactions, they function as key regulators of intestinal sensory and motor function. (11) Mast cell activation results in degranulation and the subsequent release of lipid-derived substances and allergy/inflammatory mediators, including histamine, serotonin, and various cytokines and chemokines. (11) Motor neurons located near MCs are stimulated and are believed to induce colonic motility and propulsion-causing diarrhea. (5) Additionally, vagal and spinal sensory afferent neurons have receptors for MC mediators, suggesting a link to abdominal pain. (2) Several studies (10,12) have shown an increase in colonic MC mediators in patients with irritable bowel syndrome, and their presence correlates with abdominal pain in both diarrhea- and constipation-predominant irritable bowel syndrome.

Although it seems intuitive that activated MCs can cause diarrhea and other abdominal symptoms, as exemplified by the common gastrointestinal complaints in patients with systemic mastocytosis, the clinical significance of subtle MC infiltration in various gastrointestinal disorders remains unknown. Mast cells have been found to be increased, decreased, and within reference range in inflammatory bowel disease. (13-16) In addition, investigators have found that they are often increased relative to control populations in patients diagnosed with diarrhea- or constipation-predominant irritable bowel syndrome, (10,12) but the distribution is variable. (17) Other studies, (18,19) however, found no differences in the degree of MC infiltration in patients with irritable bowel syndrome. The reason for such variability among studies is unclear, although MC detection methods may have a role because a uniform stain has not been used in all prior studies, and some studies lacked the number of subjects needed to determine a statistically significant difference.

Some of the most intriguing evidence for increased mucosal MCs having a causative role in patients with CDUE was reported by Jakate et al (3) in 2006. The authors demonstrated that 67% (22 of 33) of their patients with CDUE and increased MCs were effectively treated using drugs targeting MC-mediated function and release. However, those drugs were not administered to a control group of patients (patients with CDUE and normal MC counts), so it remains unknown whether the drugs treated MC-mediated dysfunction or improved symptoms caused by other mechanisms or because of the placebo effect.

Despite the limitations of that study, it suggested that MCs may have a role in some cases of CDUE, which led to an evaluation of MC counts in patients with CDUE who had histologically normal colon biopsy results. However, practices among different laboratories and clinicians are varied and CDUE is a diverse group of diseases with many causes for which increased mast cell activity may not be expected to contribute to all or even a majority of these cases. Thus, pathologists must, therefore, understand the usefulness and limitations of performing MC counts in this heterogeneous group of patients.

Our study is, to our knowledge, the largest to date to examine the diagnostic utility of colonic MC counts and, on the surface, supports assertions by prior investigators that MCs are increased in patients with CDUE. Our data demonstrate that when biopsy site is not considered, patients with CDUE show a statistically significant increase in colonic MCs, although the absolute differences were small, especially when considering the range of counts for patients in both the study group (range, 13-67 MCs/hpf) and the control group (range, 8-47 MCs/hpf) (Table 2). In addition, MC counts were different depending on the biopsy location. Mast cell counts were only significantly increased in the left side of the colon in our study group population, and as found with total counts, that increase was small given the range of left-sided counts for patients in both the study and control groups (Table 3).

What do these data tell us? First, biopsy location matters. Mast cells appear to be preferentially increased in the left side of the colon in patients with CDUE. Therefore, if we agree, for the moment, that MCs are a cause of some cases of CDUE, we can reasonably suggest that MC counts are uninterpretable on random colon biopsies and may not identify an increase if biopsies are performed only on the right side of the colon. Second, where MC counts are increased in the left side of the colon, there is significant interindividual variation, which causes a range of counts that can be problematic if we want to use those counts as a sensitive and specific means of testing in a group as heterogeneous as the patients with CDUE. Not surprisingly, with such a range, the ROC curves examining absolute counts are nondiscriminatory (Figures 4 and 5). The most discriminatory ROC curve was generated by examining the difference in right- and left-sided MC counts in single patients, despite the few subjects (Figure 6), probably because that type of intraindividual analysis partially addressed the issue of interindividual variation and the left-sided increase.

There are several possible explanations for these results. A methodological approach shows that our study differs from prior studies in the method of counting MCs. (3,4) We chose 1 hpf (X400, 0.55 mm field diameter) with the highest density of MCs because of its easier implementation in routine clinical practice. Many biopsies consist of only small tissue fragments that are not large enough for counting cells in 10 hpfs. Counting MCs on different levels does not compensate for that limitation because one would end up counting many of the same cells. A similar counting method is recommended to enumerate eosinophils in esophageal biopsies and has proven effective and easy to apply in routine practice. (20-22) However, it is possible that counting MCs in selected fields could contribute to the wide range seen in our study and control groups. Although we did not specifically study the distribution of MCs in the lamina propria, it appears that, in some cases, the increase was due to a higher density of MCs in the deeper part of the lamina propria (Figures 2, D, and 3, A and B), which would imply that tangential sections could lead to skewed MCs counts.

The inherent etiologic heterogeneity of any group of patients with CDUE will always be problematic, making this a difficult group to study. If MC-related diarrhea exists, a cohort of patients with CDUE would almost certainly comprise some patients with MC-related diarrhea, which may represent the high end of the range of MC counts, and many other patients with a plethora of non-MC-related causes of diarrhea, which may represent the low end of the range. The way to identify a subset of patients with CDUE and true MC-related diarrhea would be to divide those patients into a group with increased MCs and a group with normal MC counts and then randomize each group into treatment (MC-targeted medications) and placebo groups. Such a process might identify a subset of patients with MC-related diarrhea, but it would not necessarily imply that MCs will be increased enough in that subset to provide for a sensitive and specific means of testing. Such a study would be difficult because it would require many compliant patients, would need careful clinical exclusion of known causes of chronic diarrhea, and would, ideally, be multi-institutional because of the possibility that MC counts might vary significantly among different geographic regions, as has been demonstrated with eosinophils in the colon.

With this potential degree of variability in mind, an additional confounding factor in interpreting MC counts is sex. Our data illustrate that women have higher MC counts than men do (Table 5), and our box plots of overall MC counts separated by control and study groupings demonstrated no significant interaction (Figure 7). Thus, although the difference in MC counts we observed between patients in the study group and those in the control group did not depend on sex, it appears as if a reference range (and subsequent potential increase from that range in patients with CDUE) differs by sex, adding one more variable to the proper interpretation of MC counts. To date, reference ranges for men and women have not been established, and our data suggest that would be an important step in fully understanding what a particular count in an individual patient means.

Thus, these variables raise an important, practical question: How should a pathologist interpret MC counts in a patient with CDUE and a normal colon biopsy? Despite an increase in MCs in patients with CDUE, a discriminatory cutoff value for designating patients with MC-related diarrhea remains elusive, and MC counts become especially problematic if the biopsy site is not designated for the pathologist (an unfortunately common practice because random biopsies are often submitted when an endoscopist wishes to rule out microscopic colitis). Further studies, ideally with the aid of morphometric quantitative analysis, are needed to better define MC counts in healthy and diseased states, and such an analysis would likely need to be performed throughout a range of patient demographics. In all likelihood, each laboratory would have to define its own reference range.

Therefore, we conclude that, with our current knowledge, individual MC counts in a patient with CDUE are uninterpretable. In fact, we could argue that such counts are counterproductive to patient care. If MC counts were used to diagnose MC-related diarrhea and subsequently to select patients for treatment with MC stabilizers, the poor sensitivity of possible cutoff values, as illustrated in our ROC curves and sensitivity and specificity calculations, would mean many patients would go "undiagnosed," and potentially beneficial medications might be withheld from patients with mildly to moderately elevated counts. Conceivably, patients with chronic diarrhea and MCs counts within reference range might also benefit from MC-targeted medications because it may not be the absolute number of MCs but the degree of MC activation that leads to gastrointestinal dysfunction. If that were the case, counting MCs becomes unnecessary because many patients could simply be treated empirically because MC-targeted drugs, such as [H.sub.1] and [H.sub.2] blockers and MC mediator release inhibitors, such as cromolyn sodium, are readily available with few side effects.

In conclusion, although the absolute differences are small, MC counts were increased in patients with CDUE, primarily in the left side of the colon. However, whether the MCs are physiologic bystanders or a cause of the diarrhea is still unclear and requires further investigation. Our analysis demonstrates nondiscriminatory ROC curves because of the wide range of MCs seen in both healthy patients and patients with CDUE. Therefore, from a testing standpoint, performing MC stains on normal colon biopsies in patients with chronic diarrhea yields little useful information at this point. More important, it remains unknown whether medications targeting MCs are beneficial in all patients with CDUE, irrespective of MC counts, or in only a subset of patients with increased MCs. We conclude that it is, therefore, premature to recommend MC counts as part of a routine diagnostic workup for CDUE.

Please Note: Illustration(s) are not available due to copyright restrictions.

References

(1.) Clinical Practice Committee of the American Gastroenterological Association (AGA); Practice Economics Committee of the AGA. American Gastroenterological Association medical position statement: guidelines for the evaluation and management of chronic diarrhea. Gastroenterology. 1999; 116(6):1461-1463.

(2.) Sperber AD, Shvartzman P, Friger M, Fich A. A comparative reappraisal of the Rome II and Rome III diagnostic criteria: are we getting closer to the 'true' prevalence of irritable bowel syndrome? Eur J Gastroenterol Hepatol. 2007; 19(6): 441-447.

(3.) Jakate S, Demeo M, John R, Tobin M, Keshavarzian A. Mastocytic enterocolitis: increased mucosal mast cells in chronic intractable diarrhea. Arch Pathol Lab Med. 2006; 130(3):362-367.

(4.) Akhavein MA, Patel NR, Muniyappa PK, Glover SC. Allergic mastocytic gastroenteritis and colitis: an unexplained etiology in chronic abdominal pain and gastrointestinal dysmotility. Gastroenterol Res Pract. 2012; 2012:950582.doi: 10.1155/2012/950582.

(5.) Ramsay DB, Stephen S, Borum M, Voltaggio L, Doman DB. Mast cells in gastrointestinal disease. Gastroenterol Hepatol (N Y). 2010; 6(12):772-777.

(6.) Thonhofer R, Siegel C, Trummer M, Langner C. Mastocytic enterocolitis as a rare cause of chronic diarrhea in a patient with rheumatoid arthritis. Wien Klin Wochenschr. 2011; 123(9-10):297-298.

(7.) Bradesi S, Eutamene H, Theodorou V, Fioramonti J, Bueno L. Effect of ovarian hormones on intestinal mast cell reactivity to substance P. Life Sci. 2001; 68(9):1047-1056.

(8.) Collins SM, Bercik P. The relationship between intestinal microbiota and the central nervous system in normal gastrointestinal function and disease. Gastroenterology. 2009; 136(6):2003-2014.

(9.) Park CH, Joo YE, Choi SK, Rew JS, Kim SJ, Lee MC. Activated mast cells infiltrate in close proximity to enteric nerves in diarrhea-predominant irritable bowel syndrome. J Korean Med Sci. 2003; 18(2):204-210.

(10.) Metcalfe DD, Baram D, Mekori YA. Mast cells. Physiol Rev. 1997; 77(4): 1033-1079.

(11.) Barbara G, Stanghellini V, De Giorgio R, et al. Activated mast cells in proximity to colonic nerves correlate with abdominal pain in irritable bowel syndrome. Gastroenterology. 2004; 126(3):693-702.

(12.) O'Sullivan M, Clayton N, Breslin NP, et al. Increased mast cells in the irritable bowel syndrome. Neurogastroenterol Motil. 2000; 12(5):449-457.

(13.) Stoyanova II, Gulubova MV. Mast cells and inflammatory mediators in chronic ulcerative colitis. Acta Histochem. 2002; 104(2):185-192.

(14.) He SH. Key role of mast cells and their major secretory products in inflammatory bowel disease. World J Gastroenterol. 2004; 10(3):309-318.

(15.) Minocha A, Thomas C, Omar R. Lack of crucial role of mast cells in pathogenesis of experimental colitis in mice. Dig Dis Sci. 1995; 40(8):1757-1762.

(16.) Bischoff SC, Wedemeyer J, Herrmann A, et al. Quantitative assessment of intestinal eosinophils and mast cells in inflammatory bowel disease. Histopathology. 1996; 28(1):1-13.

(17.) Weston AP, Biddle WL, Bhatia PS, Miner PB Jr. Terminal ileal mucosal mast cells in irritable bowel syndrome. Dig Dis Sci. 1993; 38(9):1590-1595.

(18.) Talley NJ, Butterfield JH. Mast cell infiltration and degranulation in colonic mucosa in the irritable bowel syndrome. Am J Gastroenterol. 1996; 91(8):1675-1676.

(19.) Irvine E, Gaebel K, Driman D, Riddell R, Collins S. Mucosal mast cells (MMC) numbers are normal in biopsies of patients with irritable bowel syndrome (IBS). Gastroenterology. 1995; 108(4)(suppl 2):A620. AGA Abstracts. American Gastroenterology Association Annual Meeting. doi:10.1016/ 0016-5085(95)26793-x.

(20.) Chang F, Anderson S. Clinical and pathological features of eosinophilic oesophagitis: a review. Pathology. 2008; 40(1):3-8.

(21.) Parfitt JR, Gregor JC, Suskin NG, Jawa HA, Driman DK. Eosinophilic esophagitis in adults: Distinguishing features from gastroesophageal reflux disease: a study of 41 patients. Mod Pathol. 2006; 19(1):90-96.

(22.) Rodrigo S, Abboud G, Oh D, et al. High intraepithelial eosinophil counts in esophageal squamous epithelium are not specific for eosinophilic esophagitis in adults. Am J Gastroenterol. 2008; 103(2):435-442.

Aisha Sethi, MD; Dhanpat Jain, MD; Bani Chander Roland, MD; Jason Kinzel, MD; Joanna Gibson, MD; Ronald Schrader, PhD; Joshua Anspach Hanson, MD

Accepted for publication February 26, 2014.

From the Department of Pathology (Drs Sethi and Hanson) and the Clinical and Translational Science Center (Dr Schrader), University of New Mexico School of Medicine, Albuquerque; the Departments of Pathology (Drs Jain and Gibson) and Internal Medicine, Section of Digestive Diseases (Dr Kinzel), Yale University School of Medicine, New Haven, Connecticut; and the Department of Internal Medicine, Division of Gastroenterology and Hepatology, Johns Hopkins School of Medicine, Baltimore, Maryland (Dr Roland).

The authors have no relevant financial interest in the products or companies described in this article.

Presented in part at the 102nd Annual Meeting of the United States and Canadian Academy of Pathology; March 2-8, 2013; Baltimore, Maryland.

Reprints: Aisha Sethi, MD, Department of Pathology, University of New Mexico, MSC08 4640, BMSB, Room 335, 1 University of New Mexico, Albuquerque, NM 87131 (e-mail: Asethi@salud.unm.edu).

Caption: Figure 1. Examples of colon biopsy results from patients in the control group. A and B, Low mast cell counts. C and D, High mast cell counts (hematoxylin-eosin and c-Kit immunohistochemical stains, original magnifications X20 [A through D]).

Caption: Figure 2. Examples of colon biopsy results from patients in the study group. A and B, Low mast cell counts. C and D, High mast cell counts; also demonstrating the tendency of mast cells to cluster in the deep mucosa, a feature seen both in study and control group patients (hematoxylin-eosin and c-Kit immunohistochemical stains, original magnifications X20 [A through D]).

Caption: Figure 3. A, Poorly oriented, tangentially sectioned biopsy demonstrating the variability of the mast cell distribution in the lamina propria with increased mast cells in the deeper mucosa. B, Higher magnification of the slide shown in A, with increased mast cells in the basal mucosa (c-Kit immunohistochemical stain, original magnifications X20 [A] and X40 [B]).

Caption: Figure 4. Mast cell counts for all biopsy sites: receiver operating characteristic curve (area under the curve, 0.68).

Caption: Figure 5. Left-sided mast cell counts: receiver operating characteristic curve (area under the curve, 0.74).

Caption: Figure 6. Difference in mast cell counts (right minus left): receiver operating characteristic curve (area under the curve, 0.81).

Caption: Figure 7. Sex differences in mast cell counts: box plots of overall mast cell counts separated by control and study group patient and by sex, with each observation also plotted along with the summary box. Lines join the median of each sex within control and study groups. The plots are on a log scale because mast cell count distribution was skewed right. Mast cell counts are significantly higher in the study group than in the control group (P < .001) and counts are significantly higher for women than they are for men (P = .04). However, there is no significant interaction, so the difference between study and control group patients does not depend on sex, and the difference between men/women does not depend on whether the subject was in the study or control group. Abbreviation: hpf, high-power field.
Table 1. Patient Demographics and Biopsy Distribution

                                        Colon Biopsy Location

Patient   Patients,   Mean     Sex,
Group        No.      Age y    F (M)    Random,   Right and
                                        No. (%)     Left,
                                                   No. (%)

Control      89        57     41 (48)    4 (4)      6 (7)
Study        76        50     48 (28)   47 (62)    18 (24)

          Colon Biopsy Location

Patient
Group     Right,     Left,
          No. (%)   No. (%)

Control   41 (46)   38 (43)
Study      2 (3)    9 (12)

Table 2. Mast Cell (MC) Counts for All Biopsy Sites

Patient   Patients,    Mean,      SD,       SE,     Minimum,
Group        No.      MCs/hpf   MCs/hpf   MCs/hpf   MCs/hpf

Control      89        24.1       8.7      0.92        8
Study        76        30.7      10.5       1.2        13

Patient   Maximum,   P Value
Group     MCs/hpf

Control      47       <.001
Study        67

Abbreviation: hpf, high-power field.

Table 3. Mast Cell (MC) Counts From the Right Versus Left Side of
the Colon

Side of Colon/      Site Designated
Patient Group          Biopsies,       Mean,      SD,       SE,
                      No. (%) (a)     MCs/hpf   MCs/hpf   MCs/hpf

Right side
  Control, n = 89       47 (53)        25.4       9.0       1.3
  Study, n = 76         20 (26)        28.2      11.0       2.5
Left side
  Control, n = 89       44 (49)        22.2       8.6       1.3
  Study, n = 76         27 (36)        31.0      15.9       3.1

Side of Colon/      Minimum,   Maximum,
Patient Group       MCs/hpf    MCs/hpf    P Value

Right side
  Control, n = 89      10         47        .32
  Study, n = 76        12         50
Left side
  Control, n = 89      8          40        .01
  Study, n = 76        14         86

Abbreviation: hpf, high-power field.

Table 4. Differences in Mast Cell (MC) Counts (Right -Left)

Patient Group     Patients,    Mean,      SD,       SE,     Minimum,
                   No. (%)    MCs/hpf   MCs/hpf   MCs/hpf   MCs/hpf

Study, n = 76      18 (24)      -1       10.3       2.4      -17.5
Control, n = 89     6 (7)      10.1       7.4       3.0        1

Patient Group     Maximum,   P Value
                  MCs/hpf

Study, n = 76        25        .68
Control, n = 89      20        .02

Abbreviation: hpf, high-power field.

Table 5. Mast Cell (MC) Counts in Men and Women

Patient           Patients,   Mean, MCs/hpf
Group/Sex          No. (%)     (95% CI *)

Control, n = 89
  Women            41 (46)     24 (22-27)
  Men              48 (54)     21 (19-24)
Study, n = 76
  Women            48 (63)     30 (27-34)
  Men              28 (37)     27 (24-31)

Abbreviations, CI, confidence interval; hpf, high-power field.
COPYRIGHT 2015 College of American Pathologists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Sethi, Aisha; Jain, Dhanpat; Roland, Bani Chander; Kinzel, Jason; Gibson, Joanna; Schrader, Ronald;
Publication:Archives of Pathology & Laboratory Medicine
Date:Feb 1, 2015
Words:5316
Previous Article:Documenting the Cancer Journey.
Next Article:Clinical Validation of Targeted Next-Generation Sequencing for Inherited Disorders.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters