Clinical Gastroenterology and Hepatology 2015;13:228–236

PERSPECTIVES IN CLINICAL GASTROENTEROLOGYAND HEPATOLOGY

Microscopic Colitis: Clinical and Pathologic Perspectives

Andreas Münch* and Cord Langner‡

*Division of Gastroenterology and Hepatology, Department of Clinical and Experimental Medicine, Faculty of Health Science,Linköpings University, Linköping, Sweden; and ‡Institute of Pathology, Medical University of Graz, Graz, Austria

Abbreviations used in this paper: AZA, azathioprine; CC, collagenous co-litis; IBD, inflammatory bowel disease; IEL, intraepithelial lymphocyte; IFN,interferon; LC, lymphocytic colitis; MC, microscopic colitis; QoL, quality oflife; RCT, randomized controlled trial; TNF, tumor necrosis factor.

© 2015 by the AGA Institute1542-3565/$36.00

http://dx.doi.org/10.1016/j.cgh.2013.12.026

Microscopic colitis is a chronic inflammatory bowel dis-ease characterized by chronic nonbloody diarrhea andspecific histopathology features. Active disease, defined as3 or more stools or 1 or more watery stools per day,significantly reduces quality of life. Epidemiologic studieshave found the incidence and prevalence of microscopiccolitis to be comparable with those of Crohn’s disease andulcerative colitis. Nevertheless, microscopic colitis is stillunder-recognized in clinical practice—most health careworkers know little about its etiology and pathophysiology.Furthermore, there are many challenges to the diagnosisand treatment of patients. We review the epidemiologicand clinical features of this disorder and discuss its path-ogenesis. We also outline the criteria for histopathologicevaluation of microscopic colitis, recently published by theEuropean Consensus on Inflammatory Bowel Disease, anddiscuss a treatment algorithm created by the EuropeanMicroscopic Colitis Group. Treatment options for patientswith budesonide-refractory disease are discussed.

Keywords: Colon; EMCG; Immune Response; Therapy.

During the past decade, microscopic colitis (MC) hasemerged as a common cause of chronic nonbloody

diarrhea, especially in the elderly population.1 Up to 10%to 20% of patients with chronic diarrhea are diagnosedwith MC.2 A normal or nearly normal endoscopic picture istypically seen and only histology can confirm the diag-nosis, differentiating between its major subtypes: collag-enous colitis (CC) or lymphocytic colitis (LC). Affectedindividuals present with frequent loose or watery stools,leading to urgency and, ultimately, fecal incontinence.Abdominal pain and weight loss are common. Hence, pa-tients with active MC have a severely deteriorated qualityof life (QoL).3 The only drug that has been tested in mul-tiple randomized controlled trials (RCT) and that fulfillsthe criteria of evidence-based medicine is budesonide.4

This drug is highly effective and achieves clinical remissionin approximately 80% of patients. However, symptomrelapse occurs in 60% to 80% of patients after withdrawalof treatment,5 necessitating a discussion regarding mainte-nance therapy in patients with an active chronic course.

The cause of MC is unknown, but it is believed thata luminal agent triggers an uncontrolled immunologicresponse in the mucosa of genetically predisposed in-dividuals. On a scientific level, MC has not received the

same attention as other inflammatory bowel diseases(IBDs) (ie, ulcerative colitis and Crohn’s disease). Therefore,knowledge of MC among physicians as well as patholo-gists is limited.

Epidemiology

Epidemiologic studies have been performed mainly inEurope, North America, and Canada. However, reports oncases and smaller cohorts from Africa, Asia, Latin America,and Australia have indicated that MC is a worldwide dis-ease.1 The most comprehensive population-based studieshave been performed in Olmsted County, Minnesota, andin Örebro, Sweden. Since 1984, continuous epidemiologicfollow-up evaluation in both centers has shown a paralleltrend with an initial increase in incidence that has stabi-lized during the last decade (Figure 1). Overall, the annualincidence rates are between 2.6 and 10.8 per 105 in-habitants for CC and between 2.2 and 14 per 105 in-habitants for LC.1 The prevalence of MC was 219 of 105

cases in Olmsted County (in 2010)6 and 123 of 105 casesin Örebro (in 2008).7 These figures show that MC is nearlyas common as classic IBD (ie, Crohn’s disease and ulcer-ative colitis). Consistent in all studies is a strong femalepredominance that is less pronounced in LC than in CC.Typically, MC is a disease of the elderly, with an averageage at diagnosis of approximately 65 years. However,younger patients with chronic diarrhea also should beevaluated for the disease because 25% of MC patients areyounger than age 45.8 MC in childhood is a rare phe-nomenon, but some cases reports have been published.9

Clinical Presentation

Regarding symptoms and clinical presentation, LCand CC are not distinguishable from each other. The keyclinical feature is chronic nonbloody diarrhea, which is

Figure 1. Incidence of MC in Olmsted County (1985–2011).Figure courtesy of Darrell Pardi (Mayo Clinic).

February 2015 Microscopic Colitis 229

typically watery, leading to urgency (70% of patients)and, ultimately, fecal incontinence (40% of patients).10 Insevere cases, bowel movements can exceed 15 per dayand nocturnal diarrhea is common (50%).10 Despiteconsiderable fluid loss, serious dehydration, electrolytechanges, or other complications are rare. The naturalcourse of MC is not well investigated but more recentprospective studies have shown that the risk of relapse ishigh (60%–80%) after cessation of budesonide treat-ment, indicating that the course of many patients ischronic relapsing.5,11

Active disease leads to impaired health-related QoLand often results in severe social handicap.3 Studies haveshown that not the stool frequency as such but rather thestool consistency is the main determining factor forchanges in QoL.3,12 One watery stool per day alreadyimpairs QoL significantly and therefore disease activityin MC is defined according to Hjortswang et al13 as 3 ormore stools/day or 1 or more watery stools/day (meanof a 1-week registration). Thus, treatment decisionsshould focus primarily on bowel consistency and can bejustified in patients with fewer than 3 bowel movementsper day if the stools are watery. Several RCTs haveshown that budesonide improves QoL significantly, andthe earlier-mentioned criteria for disease activity havebeen proven useful in clinical trials.12

Abdominal pain is a common symptom in MC.8,10

Abdominal discomfort or cramps may occur in up to50%, and the differential diagnosis between MC andirritable bowel syndrome may be challenging in thesepatients. In a recent prospective cohort study, 43% ofMC patients fulfilled the ROME II criteria.14 Intermedi-ate or severe pain is observed in periods with activedisease and is ameliorated with budesonide treatment(unpublished data, 2013). Weight loss can occur inactive disease and is observed in nearly half of pa-tients.8,10 It is uncertain whether this is purely an effectof fluid loss or a consequence of eating habits becausepatients may eat less to decrease bowel frequency. Inpatients with significant weight loss concomitant celiacdisease should be ruled out.

MC commonly is associated with other autoimmunediseases. In a recent Swedish multicenter study, auto-immune disorders occurred in a third of patients, mostcommonly celiac disease (12%), autoimmune thyroiddisease (10.3%), but also Sjögren’s syndrome (3.4%),diabetes mellitus (1.7%), as well as skin and joint dis-eases (6.0%). In most cases, the diagnosis of associatedautoimmune disease preceded that of MC. Notably, pa-tients had an earlier onset of colitis and more severegastrointestinal symptoms.15

The relative risk of overall malignancy and overallmortality in CC is not different compared with the gen-eral population despite an increased risk of lung cancerin women, which may be related to smoking habits.16

Histopathology

MC remains a histologic diagnosis, but only in pa-tients with chronic diarrhea. This clinical informationalways should be provided to the pathologist. A recentlarge retrospective analysis showed that history ofdiarrhea per se does not identify patients at higher riskof abnormal histology, but those older than age 60 had amarkedly increased likelihood of a specific histologicabnormality, and MC was the most common diagnosis.17

Histology is necessary not only to make the diagnosis,differentiating between the 2 major subtypes (ie, LC andCC), but also to rule out other causes of chronic diarrhea.On endoscopic evaluation, the mucosa of the colon isalmost always normal, but occasionally may show subtlechanges, such as edema, erythema, altered vascularpattern, or even mucosal defects.18

In MC, the morphologic findings may be patchy andnot continuous. A systematic analysis of patients with CCfrom 2 large prospective multicenter trials showed that acollagenous band more than 10 mm in thickness wasmore common in the right colon (with the highest levelsin the cecum and ascending colon), and less frequent inthe sigmoid and rectum, whereas the mononuclearinflammation in the lamina propria was found to bedistributed evenly among the different segments of thelarge bowel.19 It is recommended to take multiple biopsyspecimens throughout the whole colon because biopsyspecimens obtained only from the rectum or from therectum and sigmoid colon may miss 41% or 21% ofcases, respectively. The biopsy specimens should besubmitted, preferably, in separate containers.1,19,20

Lymphocytic Colitis

The predominant histologic feature of LC is intra-epithelial lymphocytosis (ie, an increased number ofsurface intraepithelial lymphocytes [IELs] with little orno crypt architectural distortion) (Figure 2).21–23 Mostinvestigators refer to a cut-off value of 20 or more IELsper 100 surface epithelial cells (normal, <5), but someinvestigators refer to 15 or more IELs.24 Aiming at

Table 1. Histologic Key Features of Different Forms of MC

LC CCMCi orMCnos

IELs >20 IELs Normal to slightlyincreased

5–20 IELs

Subepithelialcollagen layer

Normal to slightlythickened

>10 mm 5–10 mm

Surface epitheliumdamage

þ þþ (þ)

Lamina propriainflammation

þþ þþ þ/þþ

MCi, microscopic colitis incomplete; MCnos, microscopic colitis not otherwisespecified.

Figure 2. (A) LC with a significantly increased number of IELs,increased mononuclear inflammation of the lamina propria,and preserved mucosal architecture (original magnification:100�). (B) High-power view of IELs. Note the degenerativechanges of the surface epithelium (original magnification:250�).

230 Munch and Langner Clinical Gastroenterology and Hepatology Vol. 13, No. 2

standardization of the histologic diagnosis, the cut-offvalue of 20 or more IELs recently was recommendedin the European Consensus on the Histopathology ofInflammatory Bowel Disease, which was published onbehalf of the European Society of Pathology and theEuropean Crohn’s and Colitis Organisation.20 The termi-nal ileum may be affected in LC (and also CC) with amean villous IEL count greater than 5.25

On sections stained with H&E, IELs are characterizedby mostly round, compact nuclei with a dense chromatinpattern, slightly irregular nuclear outline, and peri-nuclear halo. The surface epithelium may show degen-erative and/or regenerative changes, such asvacuolization, flattening, and loss of mucin.26 Comparedwith healthy individuals, the cellularity in the laminapropria is increased diffusely. The inflammatory infil-trate consists mainly of lymphocytes and plasma cells,but eosinophils and neutrophils also may be present,sometimes within the epithelium (Table 1). In general,H&E-stained slides are sufficient to make the diagnosis,

and immunohistochemistry to identify intraepithelial Tcells by their positivity for CD3 is not routinely needed.

Collagenous Colitis

The predominant histologic feature of CC is a thick-ened collagen band underneath the surface epithelium.The band is most evident between the crypts. It has anirregular jagged appearance at the deeper border. Itmay contain entrapped capillaries, red blood cells, andinflammatory cells.21–23 Damage to the surface epithe-lium usually is pronounced and is more common thanin LC, with sloughing of surface epithelial cells fromsubepithelial collagen being a characteristic finding(Figure 3). An increased number of IELs is seen, but notto the same amount as in LC. Similar to LC, the cellularityin the lamina propria is increased by a predominantlymononuclear inflammatory infiltrate, and IBD-like fea-tures, such as active crypt inflammation with occasionalcrypt abscess formation, may occur (Table 1).27 Ac-cording to the European Consensus on the Histopathol-ogy of Inflammatory Bowel Disease, the thickness of thecollagen band should exceed 10 mm (normal, <3 mm) onwell-oriented biopsy specimens (ie, biopsy specimens cutperpendicular to the mucosal surface).20 In most cases,the diagnosis can be established on the basis of H&E-stained slides without problems. In borderline cases,additional stains, such as Masson trichrome or immu-nohistochemistry with antibodies directed againsttenascin, which is not present in normal adult mucosaand is synthesized by subepithelial myofibroblasts(indicative of matrix remodeling), may be helpful.28

Variants and Differential Diagnosis

Different variant forms have been reported underseparate names for patients who have clinical features ofMC but fall short of fulfilling the morphologic criteria ofLC or CC. In these cases, histology shows an increasedinflammatory infiltrate in the lamina propria and anabnormal collagenous band less than 10 mm or anincreased number of IELs less than 20 per 100 epithelial

Figure 3. (A) CC with an amorphous eosinophil subepithelialcollagen band, approximately 30-mm thick. Note secondarydegenerative changes with typical sloughing of the surfaceepithelium (original magnification: 100�). (B) Masson tri-chrome stain (original magnification: 100�) and (C) immuno-staining with antibodies directed against tenascin (originalmagnification: 100�) show the characteristic irregular (jag-ged) appearance of the collagen band at the lower border.

February 2015 Microscopic Colitis 231

cells. The following terms, which reflect variations inhistology, but not clinical differences, have been sug-gested: borderline LC,29 MC incomplete,1,10 MC nototherwise specified,30 or paucicellular LC.31 The histology

of these atypical forms of MC cannot be discussed herein detail and we refer to the review by Chang et al32

for further information. These histologic categoriesare still somewhat controversial and not universallyaccepted. In addition, it may be difficult to separatethese milder forms from irritable bowel syndrome.

Pathophysiology

The etiology and pathophysiology of MC is not wellunderstood but is likely to be multifactorial, involvingmucosal immune responses to luminal factors in agenetically predisposed individual. This theory is bestsupported by observations made in patients undergoingsurgery with ileostomy. In these patients, fecal streamdiversion leads to regression of intestinal inflammationand mucosal barrier dysfunction and, in turn, recon-struction of bowel continuity leads to reappearance ofthe classic histologic features of MC.33,34 The followingsection is a selection of relevant results of mainly smallobservational studies describing different factors andmediators that might be involved in the pathogenesis ofMC. However, it still largely is unclear which of thesefindings have true pathogenic relevance.

Genetics

Familial occurrence of MC has been reported, but theexact role of genetic factors remains to be defined.35,36

HLA studies have shown an association of MC withHLA-DQ2 or DQ1/3 and a higher frequency of HLA-DR3DQ2 haplotype and TNF2 allele carriage comparedwith controls.37 Furthermore, allelic variation of thematrix metalloproteinase-9 gene does appear to beassociated with CC.38 In contrast to Crohn’s disease,functional polymorphism in the nucleotide-binding olig-omerization domain-containing protein 2 and caspaserecruitment domain-containing protein 15 (NOD2/CARD15) gene has not been detected.39

Epithelial Barrier Function

Another factor may be a defect in mucosal barrierfunction, leading to increased transmucosal permeabilityof antigens and bacteria, thereby promoting intestinalinflammation. In vitro experiments on colonic biopsyspecimens showed significant mucosal barrier dysfunc-tion in CC patients in clinical remission, which wasaggravated in active disease presenting with increasedtransmucosal uptake of nonpathogenic bacteria. Mucosalbarrier dysfunction persisted despite short-term, clini-cally effective treatment with budesonide.40 Tagkalidiset al41 observed reduced E-cadherin and zonula occlu-dens 1 expression (induced by interferon [IFN]g) in CC,which is indicative of altered epithelial barrier function.Burgel et al42 noted reduced expression of the tightjunction proteins occludin and claudin 4. Their findings

232 Munch and Langner Clinical Gastroenterology and Hepatology Vol. 13, No. 2

correlated with reduced epithelial resistance, indicatingincreased paracellular permeability.

Mechanism of Diarrhea

By using an Ussing chamber technique, Burgel et al42

described the mechanism of diarrhea in CC as a conse-quence of reduced Naþ and Cl� absorption accompaniedby active Cl� secretion. The clinical observation thatfasting can reduce diarrhea indicates an osmoticcomponent,43 making it likely that watery stools in MCare driven by a combination of osmotic and secretorycomponents.

Because MC can present with abdominal pain, ur-gency, and fecal incontinence, the question ariseswhether MC patients have normal anal-rectal function. Ina small study, patients with active CC had normal analfunction and rectal compliance. Furthermore, there wereno signs of visceral hypersensitivity.44

Immunology

CD8þ T lymphocytes predominate in both theepithelium and the lamina propria with increased pro-portions of Ki67þ and CD45ROþ CD8þ mucosal T cells.CD4þ T lymphocytes are comparably rare in the laminapropria.45

MC shows a Th1 mucosal cytokine profile with IFNgas the predominantly increased cytokine in CC, tumornecrosis factor a (TNF-a) in LC, and increased mucosalmessenger RNA levels of interleukin (IL)8 and IL15.41,46

More recent immunologic studies showed a mixed Th17/Tc17 and Th1/Tc1 mucosal cytokine profile. Mucosalmessenger RNA levels of IFNg, IL12, IL17A, IL21, andIL22 were up-regulated significantly compared withcontrols, and these correlated with higher clinical activ-ity. Significantly enhanced IL21 and TNF-a protein levelswere noted in both CC and LC.47

Risk Factors

Drugs

Drug intake has been suggested to act as an envi-ronmental risk factor in causing or triggering MC.48

Beaugerie and Pardi49 proposed a scoring system withvarying grades of certainty, and drugs with a high like-lihood to cause MC were acarbose, aspirin, cyclo3 fort,lansoprazole, nonsteroidal anti-inflammatory drugs, ra-nitidine, sertraline, and ticlopidine. In a recently pub-lished systematic review, nonsteroidal anti-inflammatorydrugs and proton pump inhibitors were identified as the2 drugs with the highest likelihood to cause MC, but theinvestigators stressed that confirmation of drug-inducedMC is still lacking.50 Most of the drugs that have beenassociated with MC are known to have diarrhea as a sideeffect.51 Hence, higher use of colonoscopy plus biopsies

in patients with drug-induced diarrhea may simply in-crease the frequency of MC diagnoses, rather than thesemedications being a causative or triggering factor in MCdevelopment. Therefore, considering a drug to cause MCshould include clinical improvement, but also disap-pearance of histologic findings after withdrawal andrecurrence after rechallenge.

Smoking

Studies on smoking habits in MC patients have indi-cated that smoking is an equal risk factor for MC for menand women and that smokers develop the disease earlierthan nonsmokers (by a median of 14 years).52–54 Thisrisk is most prominent in current smokers (adjustedodds ratio, 5.36; 95% confidence interval, 2.63–11.13 forCC, and adjusted odds ratio, 3.81; 95% confidence in-terval, 1.92–7.12 for LC).

Treatment

The primary aim of therapy is to achieve clinicalremission and to improve the patient’s QoL.13 Whetheror not histologic remission is an important goal iscurrently unknown. Before embarking on medical ther-apy, the possibility of drug-induced MC always should betaken into account, and the patient’s medication shouldbe reviewed carefully for drugs with high likelihood tocause MC.49 Discontinuation of such drugs, if possible,may lead to resolution of symptoms. Smoking cessationcan be considered but the evidence for this is still weak.It is advisable that patients register their bowel move-ments in a diary before initiation and during follow-uptreatment.

The strongest treatment evidence is currently avail-able for budesonide, a locally active corticosteroid withextensive first-pass metabolism in the liver and lowsystemic exposure. In a meta-analysis, budesonide waseffective, with response rates greater than 80%, and waswell tolerated for inducing and maintaining remissionwith a “number needed to treat” of 2 for CC and 3 forLC.4 More recently, 2 larger RCTs have confirmed thegood efficacy of budesonide induction therapy in activeCC with significant improvement of QoL.12,55 Two addi-tional RCTs of CC showed that clinical remission andhistologic response can be maintained with budesonide 6mg/d for 6 months.11,56 However, 6 mg/d might be toohigh in the long run and bears the risk of developing sideeffects such as a skin hematoma, cataract, or increasedserum blood sugar level. After withholding short-termbudesonide treatment or maintenance therapy (6 mg),relapse rates are high (60%–80%).5,11

Currently, there are no evidence-based alternatives tobudesonide. Antidiarrheals such as loperamide are usedfrequently in MC, although they have never been testedformally in RCTs. Clinical experience suggests im-provement of symptoms in some patients.8,57 However,

February 2015 Microscopic Colitis 233

sustained clinical remission rarely is achieved andimpact on colonic inflammation is unlikely. Thus, anti-diarrheals may be used alone or in conjunction withother therapies, depending on symptom severity.

Prednisolone has been investigated mainly in retro-spective studies.8,57 In a recent population-based study,budesonide had a higher rate of complete response thanprednisolone (82.5% vs 52.9%). Furthermore, MC pa-tients treated with budesonide were less likely to expe-rience a recurrence than those treated withprednisolone.58 Therefore, prednisolone appears to haveno value for patients who do not respond to budesonide.

Bismuth subsalicylate has been tested in a smallRCT that was never fully published.59 A clinical andhistologic response was noted in all 7 bismuth subsa-licylate MC patients compared with none of the 7 pla-cebo recipients. Larger randomized trials are necessaryto determine the value of bismuth subsalicylate in thetreatment of MC.

Mesalazine achieved results comparable with placeboin a recent RCT of CC patients, clearly showing thatmesalazine should not be used as induction therapy.55

Although evidence is limited, immunosuppressivetherapies may be considered in patients with severesymptoms who are refractory or intolerant to budeso-nide. According to results from RCTs, MC patients areexpected to achieve clinical remission within 4 weeks oninduction therapy with 9 mg budesonide or maintainclinical remission on 6 mg or less of budesonide.Approximately 10% to 20% of short-term budesonide-treated patients are nonresponders and may be candi-dates for immunosuppressive therapy.12,55 How manypatients develop intolerance or lose effect on budesonidemaintenance therapy has not been reported.

To date, only a few case reports have been publishedshowing the beneficial effects of anti-TNF agents(infliximab and adalimumab) in selected CC patients. Theregimen and dose applied were equal to those normallyused for classic IBD.60–62 Because these approaches areexperimental, solid recommendations cannot be givenbecause of a lack of data, and the following remarks arebased mainly on personal experience. Budesonide non-responders often have a long history of diverse treat-ment failures. The chronic active disease with highlyfrequent watery stools binds the patient to the toilet,causing severely impaired QoL. Initiation of biologicaltherapy always should be considered on an individualbasis, taking age and comorbidity into account. A thor-ough risk-benefit calculation is necessary to avoidserious side effects and regular follow-up evaluation isimportant. The advantage of anti-TNF agents comparedwith thiopurines may lie in faster relief of symptoms.However, experience is limited and data on long-termtreatment with biologicals are not available yet.

Azathioprine (AZA) or mercaptopurine have beentested in a small group of patients (N ¼ 9) with steroid-dependent or refractory CC, showing a response rateof 89% and a steroid-sparing effect.63 In a larger

retrospective multicenter case series, 28% of patientstreated with AZA achieved and maintained clinicalremission for up to 57 months, however, the majority ofpatients developed an intolerance to AZA, causingcessation of treatment. Thirteen of the 31 AZA-intolerantpatients were put on mercaptopurine, of whom 6 pa-tients (48%) regained clinical remission, resulting in anoverall response rate of 41%.64 These results indicatethat thiopurine might be considered in chronic active MCprimarily as maintenance therapy, but prospective RCTsare necessary and eagerly awaited.

In a retrospective case series oral methotrexate hasshown beneficial effect on symptoms of CC patients naiveto budesonide.65 However, in a recent case series of 9patients intolerant or not responding to budesonide,methotrexate was administered subcutaneously at adose of 15 to 25 mg/wk for a period of 12 weeks. Noneof the patients achieved clinical remission and 4 patientsexperienced reversible adverse events leading to cessa-tion of medication.66 The limited data are conflicting andfurther studies are required for methotrexate, especiallyin budesonide nonresponders.

Surgical intervention in MC should be regarded as thelast treatment option in patients refractory to any med-ical therapy. Diverting ileostomy, subtotal colectomy, orileal pouch–anal anastomosis have been performed suc-cessfully in individual cases.33,34

Based on the currently available evidence, an algo-rithm for the treatment of MC was proposed by the Eu-ropean Microscopic Colitis Group (Figure 4).1 Accordingto this algorithm, a treatment attempt with antidiar-rheals and/or cholestyramine may be justified in patientswith only mild symptoms. However, patients with activedisease should be treated primarily with short-termbudesonide (6–8 wk). In patients initially respondingto this treatment, budesonide can be re-administered incase of relapse, either as intermittent therapy or as low-dose continuous therapy, always aiming to identify thelowest dose that maintains clinical remission. In patientslacking response to budesonide and mild symptoms,cholestyramine, bismuth, or loperamide, even in combi-nation, may be given. In more severe cases and in pa-tients with low risk for side effects (age, comorbidity),anti–TNF-a antibodies may be justified to restoreremission and improve QoL. As maintenance treatment,immunomodulators such as AZA or mercaptopurine canbe considered. Surgical treatment is a therapeutic optionin patients refractory to all medical therapies.

Future Considerations

Epidemiologic data clearly show that MC has becomea common gastrointestinal disorder and physicians andpathologists should be aware of the disease. All patients,especially elderly patients with chronic diarrhea, shouldbe referred for colonoscopy and biopsy specimensshould be obtained to confirm or rule out MC. On a

Figure 4. Algorithm for thetreatment of MC proposedby the European Micro-scopic Colitis Group.1 6-MP, 6 mercaptopurine.

234 Munch and Langner Clinical Gastroenterology and Hepatology Vol. 13, No. 2

practical level, patients should be asked to use a diary tomonitor symptoms. The criteria for disease activitypresented by Hjortswang et al13 are recommended fortreatment decision making. Because MC mainly presentswith a chronic relapsing course, regular follow-up eval-uation is needed and the definition of disease activity canbe applied to improve patient empowerment.

On a scientific level, continuous epidemiologic studiesshould be encouraged not only to monitor incidence databut also to focus on environmental factors, such as drugs,infections, or food that might play a pathogenic role. Themost striking finding in epidemiology is the consistentfemale predominance. This notion causes speculation onthe pathogenetic role of specific hormones or may simplyreflect women as more prone to develop autoimmunediseases. Large-scale genetic studies are warranted toidentify genetic variants associated with MC and focus onsusceptibility genes that overlap with variants seen inother immune-related conditions. Basic science shouldconcentrate on the investigation of immunologic abnor-malities that occur in MC. Improved knowledge of theunderlying immune processes could be the key to abetter understanding of patients with treatment failureand could stimulate the development of future treatmentstrategies. Despite the fact that many patients respond tobudesonide, patients who become refractory or intolerantto the drug are a clinical challenge and physicians have tomake therapeutic decisions tailored to individual patientneeds because evidence-based data are unavailable. Toovercome this shortcoming, future clinical trials should

focus on the treatment of budesonide nonresponders andinvestigate the efficacy and safety of therapies on main-taining clinical remission. For this task, it would bedesirable if biological markers or objective parameterscould be developed to monitor disease activity.

References

1. Munch A, Aust D, Bohr J, et al. Microscopic colitis: current

status, present and future challenges: statements of the Euro-pean Microscopic Colitis Group. J Crohns Colitis 2012;6:932–945.

2. Olesen M, Eriksson S, Bohr J, et al. Microscopic colitis: acommon diarrhoeal disease. An epidemiological study in Ore-bro, Sweden, 1993-1998. Gut 2004;53:346–350.

3. Hjortswang H, Tysk C, Bohr J, et al. Health-related quality of lifeis impaired in active collagenous colitis. Dig Liver Dis 2011;43:102–109.

4. Chande N, MacDonald JK, McDonald JW. Interventions fortreating microscopic colitis: a Cochrane Inflammatory BowelDisease and Functional Bowel Disorders Review Group sys-tematic review of randomized trials. Am J Gastroenterol 2009;104:235–241; quiz 234, 242.

5. Miehlke S, Madisch A, Voss C, et al. Long-term follow-up ofcollagenous colitis after induction of clinical remission withbudesonide. Aliment Pharmacol Ther 2005;22:1115–1119.

6. Gentile NM, Khanna S, Loftus EV, et al. The epidemiology ofmicroscopic colitis in Olmsted County from 2002 to 2010: apopulation-based study. Clin Gastroenterol Hepatol 2013 Oct 9.Epub ahead of print.

7. Wickbom A, Bohr J, Eriksson S, et al. Stable incidence ofcollagenous colitis and lymphocytic colitis in Orebro, Sweden,

February 2015 Microscopic Colitis 235

1999-2008: a continuous epidemiologic study. Inflamm BowelDis 2013;19:2387–2393.

8. Bohr J, Tysk C, Eriksson S, et al. Collagenous colitis: a retro-spective study of clinical presentation and treatment in 163patients. Gut 1996;39:846–851.

9. El-Matary W, Girgis S, Huynh H, et al. Microscopic colitis inchildren. Dig Dis Sci 2010;55:1996–2001.

10. Bjornbak C, Engel PJ, Nielsen PL, et al. Microscopic colitis:clinical findings, topography and persistence of histopatholog-ical subgroups. Aliment Pharmacol Ther 2011;34:1225–1234.

11. Bonderup OK, Hansen JB, Teglbjaerg PS, et al. Long-termbudesonide treatment of collagenous colitis: a randomised,double-blind, placebo-controlled trial. Gut 2009;58:68–72.

12. Münch A, Benoni C, Olesen M, et al. Budesonide induces clin-ical remission and improves quality of life in active collagenouscolitis: results from the open-label induction phase of BUC-63/COC (abstr). J Crohns Colitis 2013;7(Suppl I):S238(P567).

13. Hjortswang H, Tysk C, Bohr J, et al. Defining clinical criteria forclinical remission and disease activity in collagenous colitis.Inflamm Bowel Dis 2009;15:1875–1881.

14. Abboud R, Pardi DS, Tremaine WJ, et al. Symptomatic overlapbetween microscopic colitis and irritable bowel syndrome: aprospective study. Inflamm Bowel Dis 2013;19:550–553.

15. Vigren L, Tysk C, Strom M, et al. Celiac disease and otherautoimmune diseases in patients with collagenous colitis.Scand J Gastroenterol 2013;48:944–950.

16. Chan JL, Tersmette AC, Offerhaus GJ, et al. Cancer risk incollagenous colitis. Inflamm Bowel Dis 1999;5:40–43.

17. O’Mahony OH, Burgoyne M, Going JJ. Specific histologicalabnormalities are more likely in biopsies of endoscopicallynormal large bowel after the age of 60 years. Histopathology2012;61:1209–1213.

18. Koulaouzidis A, Saeed AA. Distinct colonoscopy findings ofmicroscopic colitis: not so microscopic after all? World J Gas-troenterol 2011;17:4157–4165.

19. Aust D, Münch A, Olesen M, et al. Topographical distribution ofcollagenous colitis–a pooled histological analysis of 2 Europeanprospective multicenter trials (abstr). Gastroenterology 2013;144:S-421(Su1183).

20. Magro F, Langner C, Driessen A, et al. European consensus onthe histopathology of inflammatory bowel disease. J CrohnsColitis 2013;7:827–851.

21. Chetty R, Govender D. Lymphocytic and collagenous colitis: anoverview of so-called microscopic colitis. Nat Rev GastroenterolHepatol 2012;9:209–218.

22. Yen EF, Pardi DS. Review article: microscopic col-itis–lymphocytic, collagenous and ‘mast cell’ colitis. AlimentPharmacol Ther 2011;34:21–32.

23. Brown WR, Tayal S. Microscopic colitis. A review. J Dig Dis2013;14:277–281.

24. Geboes K. Lymphocytic, collagenous and other microscopiccolitides: pathology and the relationship with idiopathic inflam-matory bowel diseases. Gastroenterol Clin Biol 2008;32:689–694.

25. Sapp H, Ithamukkala S, Brien TP, et al. The terminal ileum isaffected in patients with lymphocytic or collagenous colitis. AmJ Surg Pathol 2002;26:1484–1492.

26. Mahajan D, Goldblum JR, Xiao SY, et al. Lymphocytic colitis andcollagenous colitis: a review of clinicopathologic features andimmunologic abnormalities. Adv Anat Pathol 2012;19:28–38.

27. Ayata G, Ithamukkala S, Sapp H, et al. Prevalence and signifi-cance of inflammatory bowel disease-like morphologic features

in collagenous and lymphocytic colitis. Am J Surg Pathol 2002;26:1414–1423.

28. Muller S, Neureiter D, Stolte M, et al. Tenascin: a sensitive andspecific diagnostic marker of minimal collagenous colitis.Virchows Arch 2001;438:435–441.

29. Lazenby AJ, Yardley JH, Giardiello FM, et al. Pitfalls in thediagnosis of collagenous colitis: experience with 75 cases froma registry of collagenous colitis at the Johns Hopkins Hospital.Hum Pathol 1990;21:905–910.

30. Fraser AG, Warren BF, Chandrapala R, et al. Microscopic colitis:a clinical and pathological review. Scand J Gastroenterol 2002;37:1241–1245.

31. Goldstein NS, Bhanot P. Paucicellular and asymptomatic lym-phocytic colitis: expanding the clinicopathologic spectrum oflymphocytic colitis. Am J Clin Pathol 2004;122:405–411.

32. Chang F, Deere H, Vu C. Atypical forms of microscopic colitis:morphological features and review of the literature. Adv AnatPathol 2005;12:203–211.

33. Jarnerot G, Tysk C, Bohr J, et al. Collagenous colitis and fecalstream diversion. Gastroenterology 1995;109:449–455.

34. Munch A, Soderholm JD, Wallon C, et al. Dynamics of mucosalpermeability and inflammation in collagenous colitis before,during, and after loop ileostomy. Gut 2005;54:1126–1128.

35. Abdo AA, Zetler PJ, Halparin LS. Familial microscopic colitis.Can J Gastroenterol 2001;15:341–343.

36. Jarnerot G, Hertervig E, Granno C, et al. Familial occurrence ofmicroscopic colitis: a report on five families. Scand J Gastro-enterol 2001;36:959–962.

37. Fine KD, Do K, Schulte K, et al. High prevalence of celiac sprue-like HLA-DQ genes and enteropathy in patients with the micro-scopic colitis syndrome. Am J Gastroenterol 2000;95:1974–1982.

38. Madisch A, Hellmig S, Schreiber S, et al. Allelic variation of thematrix metalloproteinase-9 gene is associated with collagenouscolitis. Inflamm Bowel Dis 2011;17:2295–2298.

39. Madisch A, Hellmig S, Schreiber S, et al. NOD2/CARD15 genepolymorphisms are not associated with collagenous colitis. Int JColorectal Dis 2007;22:425–428.

40. Munch A, Soderholm JD, Ost A, et al. Increased transmucosaluptake of E. coli K12 in collagenous colitis persists after bude-sonide treatment. Am J Gastroenterol 2009;104:679–685.

41. Tagkalidis PP, Gibson PR, Bhathal PS. Microscopic colitisdemonstrates a T helper cell type 1 mucosal cytokine profile.J Clin Pathol 2007;60:382–387.

42. Burgel N, Bojarski C, Mankertz J, et al. Mechanisms of diarrheain collagenous colitis. Gastroenterology 2002;123:433–443.

43. Bohr J, Jarnerot G, Tysk C, et al. Effect of fasting on diarrhoea incollagenous colitis. Digestion 2002;65:30–34.

44. Walter SA, Munch A, Ost A, et al. Anorectal function in patientswith collagenous colitis in active and clinically quiescent phase,in comparison with healthy controls. Neurogastroenterol Motil2010;22:534–538, e118.

45. Kumawat AK, Strid H, Elgbratt K, et al. Microscopic colitis pa-tients have increased proportions of Ki67(þ) proliferating andCD45RO(þ) active/memory CD8(þ) and CD4(þ)8(þ) mucosal Tcells. J Crohns Colitis 2013;7:694–705.

46. Dey I, Beck PL, Chadee K. Lymphocytic colitis is associatedwith increased pro-inflammatory cytokine profile and up regu-lation of prostaglandin receptor EP4. PLoS One 2013;8:e61891.

47. Kumawat AK, Strid H, Tysk C, et al. Microscopic colitis patientsdemonstrate a mixed Th17/Tc17 and Th1/Tc1 mucosal cytokineprofile. Mol Immunol 2013;55:355–364.

236 Munch and Langner Clinical Gastroenterology and Hepatology Vol. 13, No. 2

48. Fernandez-Banares F, de Sousa MR, Salas A, et al. Epidemio-logical risk factors in microscopic colitis: a prospective case-control study. Inflamm Bowel Dis 2013;19:411–417.

49. Beaugerie L, Pardi DS. Review article: drug-induced micro-scopic colitis–proposal for a scoring system and review of theliterature. Aliment Pharmacol Ther 2005;22:277–284.

50. Keszthelyi D, Penders J, Masclee AA, et al. Is microscopiccolitis a drug-induced disease? J Clin Gastroenterol 2012;46:811–822.

51. Chassany O, Michaux A, Bergmann JF. Drug-induced diarrhoea.Drug Saf 2000;22:53–72.

52. Vigren L, Sjoberg K, Benoni C, et al. Is smoking a riskfactor for collagenous colitis? Scand J Gastroenterol 2011;46:1334–1339.

53. Yen EF, Pokhrel B, Du H, et al. Current and past cigarettesmoking significantly increase risk for microscopic colitis.Inflamm Bowel Dis 2012;18:1835–1841.

54. Fernandez-Banares F, de Sousa MR, Salas A, et al. Impact ofcurrent smoking on the clinical course of microscopic colitis.Inflamm Bowel Dis 2013;19:1470–1476.

55. Miehlke S, Madisch A, Kupcinskas L, et al. Double-blind,double-dummy, randomised, placebo-controlled, multicentertrial of budesonide and mesalazine in collagenous colitis. Gut2012;61(Suppl 3):A102.

56. Miehlke S, Madisch A, Bethke B, et al. Oral budesonide formaintenance treatment of collagenous colitis: a randomized,double-blind, placebo-controlled trial. Gastroenterology 2008;135:1510–1516.

57. Olesen M, Eriksson S, Bohr J, et al. Lymphocytic colitis: aretrospective clinical study of 199 Swedish patients. Gut 2004;53:536–541.

58. Gentile NM, Abdalla AA, Khanna S, et al. Outcomes of patientswith microscopic colitis treated with corticosteroids: apopulation-based study. Am J Gastroenterol 2013;108:256–259.

59. Fine KOF, Lee E, Lafon G, et al. Randomized, double-blind,placebo-controlled trial of bismuth subsalicylate for micro-scopic colitis. Gastroenterology 1999;116:A880.

60. Esteve M, Mahadevan U, Sainz E, et al. Efficacy of anti-TNFtherapies in refractory severe microscopic colitis. J CrohnsColitis 2011;5:612–618.

61. Munch A, Ignatova S, Strom M. Adalimumab in budesonide andmethotrexate refractory collagenous colitis. Scand J Gastro-enterol 2012;47:59–63.

62. Pola S, Fahmy M, Evans E, et al. Successful use of infliximab inthe treatment of corticosteroid dependent collagenous colitis.Am J Gastroenterol 2013;108:857–858.

63. Pardi DS, Loftus EV Jr, Tremaine WJ, et al. Treatment of re-fractory microscopic colitis with azathioprine and 6-mercapto-purine. Gastroenterology 2001;120:1483–1484.

64. Munch A, Fernandez-Banares F, Munck LK. Azathioprine andmercaptopurine in the management of patients with chronic,active microscopic colitis. Aliment Pharmacol Ther 2013;37:795–798.

65. Riddell J, Hillman L, Chiragakis L, et al. Collagenous colitis: orallow-dose methotrexate for patients with difficult symptoms: long-term outcomes. J Gastroenterol Hepatol 2007;22:1589–1593.

66. Munch A, Bohr J, Vigren L, et al. Lack of effect of methotrexatein budesonide-refractory collagenous colitis. Clin Exp Gastro-enterol 2013;6:149–152.

Reprint requestsAddress requests for reprints to: Andreas Münch, MD, PhD, Division ofGastroenterology and Hepatology, Department of Clinical and ExperimentalMedicine, Faculty of Health Sciences, Linköping University, 58185 Linköping,Sweden. e-mail: andreas.munch@lio.se; fax: (0046) 101033896.

Conflicts of interestThis author discloses the following: Andreas Münch received a research grantfrom Abbvie and has received lecture fees/honoraria from Dr Falk Pharma. Theremaining author discloses no conflicts.

GASTROENTEROLOGY ARTICLE OF THE WEEK May 7, 2015

Munch A, Langner C. Microscopic colitis: Clinical and pathologic perspectives. Clin Gastroenterol Hepatol 2015:13:228-236 1. Clinical features of microscopic colitis includes: a. incontinence b. nocturnal diarrhea c. dehydration and electrolyte disturbance d. lack of urgency e. abdominal pain 2. Histologic features of lymphocytic colitis a. >20 lymphocytes/hpf invading the crypt lumen b. distortion of the colonic crypst c. >20 lymphocytes per 100 epithelial cells of the colon d. increased inflammatory infiltrate in the lamina propria True or False 3. The pathophysiology of diarrhea in microscopic colitis includes increased chloride secretion and reduced sodium and chloride absorption 4. Diverting ileostomy in microscopic colitis leads to resolution of microscopic changes in the colon epithelium 5. Microscopic findings of collagenous colitis may be more common in the right colon compared with the left colon 6. Diarrhea from microscopic colitis is not improved by fasting 7. Patients who do not respond to budesonide should be given prednisolone 8. Mesalamine can be used as an steroid-sparing agent in microscopic colitis therapy 9. Microscopic colitis is associated with an increased risk of colon cancer 10. Patients with less than 3 watery movements per day rarely warrant therapy with budesonide 11. NSAID’s and PPI’s are two drugs described as possibly having the highest likelihood of causing microscopic colitis 12. A maintenance dose of 6mg/d of budesonide is safe and free from most steroid side effects