clin pathological mimicsof chronic inflammatory bowel disease · non-specific chronic inflammation,...

J Clin Pathol 1991;44:726-733

Pathological mimics of chronic inflammatorybowel disease

N A Shepherd

IntroductionUlcerative colitis, Crohn's disease, and theirintermediate forms collectively referred to asidiopathic chronic inflammatory bowel disease,show a spectrum of pathological changes.Essentially their pathological diagnosis is oneof exclusion of the many causes of inflam-mation in the intestines. The ultimate diag-nosis may require corroboration with clinical,microbiological, and radiological data. Withthe advent of fibreoptic endoscopy, patho-logists are more often confronted withmucosal biopsy specimens of the gastro-intestinal tract, and the success of newer sur-gical techniques has meant that pathologistsare increasingly pressured to make un-equivocal diagnoses of ulcerative colitis andCrohn's disease. For instance, many wouldnow consider restorative proctocolectomywith ileal reservoir to be the operation ofchoice in ulcerative colitis when medicationhas failed to control the disease,' and yetCrohn's disease is an absolute contraindica-tion for this operation.2 It is essential,therefore, that the correct diagnosis is madebefore surgery is contemplated.The importance of macroscopic examina-

tion of surgical specimens to differentiateulcerative colitis and Crohn's disease has beenemphasised, but the pathologist more oftenhas to make important diagnostic decisions,which will directly affect patient management,on biopsy specimens. Although both ulcer-ative colitis and Crohn's disease show charac-teristic histological features in their classicform, none of the morphological changes seenin either condition is entirely specific. Patho-logists generally rely on a combination ofmorphological features to make the appro-priate diagnosis. This is especially so inCrohn's disease in which pathological featuresvary greatly from case to case. The two mostcharacteristic features, fissuring ulceration andgranulomas, can be seen in many other condi-tions. Transmural inflammation is a little morespecific but this feature is also seen in severalother conditions. In about 10% of cases, par-ticularly in acute fulminant colitis, it may beimpossible to differentiate the conditions andthe term "indeterminate colitis" is used.3 Forthe purposes of this review it is not intendedto concentrate on the pathological differenti-ation of ulcerative colitis and Crohn's diseaseas this has been the subject of standard textsand reviews.' This review is primarily con-cerned with recently described conditions andnewly recognised situations where the micro-scopic features may inappropriately suggesta diagnosis of chronic inflammatory boweldisease.

Iatrogenic inflammatory bowel diseaseVarious therapeutic manoeuvres, both sur-gical and medical, may directly cause inflam-matory pathology in the intestines. Certainsurgical procedures can result in mor-phological changes which simulate chronicinflammatory bowel disease: these changes arelargely independent of the original indicationfor surgery and represent a tissue response toan altered environment. Drugs can causeinflammation of the intestinal mucosa. In thesmall bowel enteric coated preparations andnon-steroidal anti-inflammatory drugs arecommon causes of inflammation, while in thelarge intestine mucosal inflammatory changesare seen particularly with anti-neoplasticagents and after enemas and suppositories.

DIVERSION COLITIS AND THE DEFUNCTIONEDRECTUMWhen part of the colon or rectum is excludedfrom the faecal stream for any reason, thecolorectal mucosa may become inflamed. Thepathogenesis of this diversion colitis probablyrelates to the lack of essential short-chain fattyacids, particularly butyrates, normallyproduced by anaerobic bacteria, which main-tain the healthy colonic mucosa.78 Such diver-sion colitis or proctitis occurs in the defunc-tioned large intestine in patients with func-tional disorders, diverticular disease, andcolorectal cancer.910 Macroscopically, the con-dition is said to resemble ulcerative colitis." 12Histologically, diversion colitis shows diffusechronic inflammation of the lamina propriaand crypt abscesses may be present: in generalthe lack of gross crypt architectural distortionand goblet cell depletion militates against adiagnosis of ulcerative colitis (fig 1)." Insevere cases, however, diversion colitis mayclosely simulate ulcerative colitis micros-copically.9"' Furthermore, the presence ofmucosal granulomas, together with the inflam-matory pathology, may create a histologicalappearance reminiscent of Crohn's disease.'4 15The picture is further complicated by the factthat ulcerative colitis and Crohn's disease areboth common indications for faecal streamdiversion. In ulcerative colitis the rectum isoften defunctioned as a mucus fistula aftertotal colectomy, and severe colonic Crohn'sdisease may be ameliorated by faecal streamdiversion.'6 17 The defunctioned rectum inpatients with ulcerative colitis may showtransmural inflammation, fissures, and gran-ulomas: an erroneous diagnosis of Crohn'sdisease may result.'8 It should be emphasisedthat the pathological examination of a defunc-tioned segment of bowel may be very mislead-ing: the diagnosis of chronic inflammatory

Department ofHistopathology,Gloucestershire RoyalHospital, GreatWestern Road,Gloucester GL1 3NNCorrespondence to:Dr N A ShepherdAccepted for publication20 February 1991

726

on August 21, 2019 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.44.9.726 on 1 Septem

ber 1991. Dow

nloaded from

http://jcp.bmj.com/

Pathological mimics of chronic inflammatory bowel disease 727

ar bowel disease should be restricted to thef,,<:', ' .' *", <@.<t4,S,b, examination of colon excised before faecal~,. t^.*.*;streamdiversion.'8

THE ILEAL RESERVOIRRestorative proctocolectomy with ileal

alofreservoir is now one of the more favoured'- surgical alteratives among both surgeons and

patients, for patients with ulcerative colitis

-.'.'r^^s.<F.- s chrequiring total colectomy, and for those withx4j;*,'t';,"familial adenomatous polyposis.n10 The_ creation of an ileal reservoir, both pelvic as in

restorative proctocolectomy, or abdominal as

ofheecumolowigveiccolcfstua ompictin dvericlardieas. herii Koorck's mcosatinen ileostomye coiiscompicAted

by inflammatory changes with varying degreesof villous atrophy in the ileal mucosa.an 2 The

.J' cvillous atrophy and crypt hyperplasia producea morphological appearance of colonic meta-

_ ~ plasia. In those cases with extensive acute andS.chronic inflammation, particularly those

patients with the chronic relapsing inflam-(>t3 *> t t i ' matory condition known as pouchitis, the

Figure 1 Diversion proctitis This rectal biopsy specimen was taken after defunctioning mucosa shows a close resemblance to theof the rectum following a vesicocolicflstula complicating diverticular disease. There is colorectal mucosa in ulcerative colitis (fig 2A).diffuse chronic inflammation of the lamina propria in the absence ofgross crypt distortion Both clinically and pathologically there areActive inflammation with crypt abscesses was also present (haematoxylin and eosin). close links between pouchitis and ulcerative

colitis; it may be that pouchitis and ulcerative'. '--colitis share immunopathogenetic mechan-'F.''isms" or that pouchitis represents a form of

z *.. ulcerative colitis in metaplastic ileal mucosa.yo

4.-W.It has been suggested that Crohn's disease---~ may develop in the reservoir after proc-tocolectomy for indisputable ulcerativecolitis and even that pouchitis may be amanifestation of Crohn's disease. The patho-logical hallmarks of Crohn's disease, however,

-.-~~~~~~ ~ ~~.~namely granulomas, transmural inflammation

.-:.~~~ ~ ~ ~ ~~~ ~and fissures, may all be seen as a consequenceof surgical manipulation and reservoir con-struction.26 As in many other sites in the gut,

~~)~~~ ~ '~~i~ granulomas are a major source of diagnostic-~E~~~ - ~:- -, confusion. They are occasionally seen, par--~~~~~~ticularly in lymphoid follicles, in the reservoir

mucosa of patients with an indisputable diag-~~. - ~. nosis of ulcerative colitis (fig 2B).26 Such gran-

,~~~.~~*.~~~~ ~ ~ ulomas probably represent a reaction of theW.. ~ $~; ileal mucosa, particularly the lymphoid tissue,:'~ ~ to an altered intraluminal environment or to

*141f.aAS~ ~ ~extraneous material. Grohn's disease remainsf~~~ ~ ~ ~ ~ ~~' an absolute contraindication for pelvic

~~~21~~~~~~reservoir surgery2: pathological changes~~I~~IJ-J .~~~~~* /~~within the reservoir may closely resemble

those seen in Crohn's disease but such a'I;114 ~~~~~~~~~diagnosis should not be made solely on the

- ~~~~~~~pathological changes seen in the reservoir.

DRUGS

.<~~~~~~~~~~~~~~~Small bowel ulceration may be caused by anynumber of conditions, although drugs seem tobe a- common cause.27 Enteric coated potas-

-'1W~~~~ - / ~~~~ sium supplements cause small bowel ulcera-,- s-~ ... tion, probably by an ischaemic effect due to

C';Ad ~~~~~~~~~~~~~~vascularconstriction induced by high concen-

-4;~trations of potassium ions: the histological-~~~-.- ~~~ ~ ~ ~ ~~ ~ ~~ changes affect only the mucosa and submucosa

Figure 2 Mucosal biopsy specimens from pelvic ileal reservoirs. (A) There is subtotal and are unlikely to be confused with activevillous atrophy, diffuse chronic in.jammation, andgross crypt distortion. (B) A well Crohn's disease.' Now accepted as one of thecircumscribed epithelioidgranuloma in the characteristic position within a lymphoid commonest causes of clinically important ilealfollicle in a rather distorted mucosal biopsy specimen. In the absence of other stigmata ofCrohn's disease such granulomas probably represent a tissue reaction to persorbed ulceration are non-steroidal anti-inflammatoryintraluminal material (haematoxylin and eosin). drugs (NSAIDs).29 These also cause mucosal



j.com/

J Clin P


ber 1991. Dow

nloaded from

http://jcp.bmj.com/

Shepherd

Figure 3 Sigmoid colonic biopsy specimenfrom crescentic or diverticular colitis. Thereis diffuse chronic inflammation with pronounced crypt distortion and crypt abscesses arepresent (right). There was no clinical or radiological evidence of ulcerative colitis andthe rectal mucosa was histologically normal (haematoxylin and eosin).

inflammation and ulceration in the colorectalmucosa.30 The picture is complicated by thefact that NSAIDs may exacerbate chronicinflammatory bowel disease.3' 32 Attention hasrecently focused on the late stage pathology ofsmall intestinal disease induced by NSAIDs.29So-called diaphragm disease has highlycharacteristic macroscopic and microscopicappearances, which, once recognised, arereadily differentiated from Crohn's disease.29NSAIDs, methyldopa, gold, antineoplastic

agents and penicillins are the most widelyrecognised drugs known to cause activeinflammation in the large intestinal mucosa.Methyldopa causes an acute colitis in a smallproportion of cases33 and gold causes a charac-teristic eosinophilic infiltrate.34 5'fluoro-uracilhas been the most widely studied of thechemotherapeutic agents to cause acute colitis.In the acute phase epithelial necrosis is thepredominant feature while in the resolvingphase crypt regeneration and distortionreminiscent of healed ulcerative colitis are

observed.35

ENEMAS AND SUPPOSITORIES

Most enemas and suppositories cause littlemucosal pathology but hypertonic salineenemas and bisacodyl, in particular, producecrypt epithelial proliferation and degenerationwith inflammatory change.637 In general thepathological changes due to enemas more

closely resemble ischaemia or infective colitisand lack the chronic inflammatory componentof chronic inflammatory bowel disease. Sup-positories, particularly those containingNSAIDs, may also cause mucosal damage andresult in rectal bleeding.' The resulting histo-logical changes are generally those of mild,non-specific chronic inflammation, butoccasionally active inflammation is seen.38

Diverticular diseaseAlthough diverticular disease is essentially a

functional disorder of the colon, a wide varietyof inflammatory pathology may be associatedwith the condition. For instance, mucosal

redundancy is a characteristic feature of thecondition, and inflammation, with histologicalfeatures ofmucosal prolapse, is sometimes seenin the sigmoid colon."'3 This syndrome wasoriginally described as segmental colitis, al-though it has become clear that inflammatorypathology restricted to the segment affected bydiverticular disease shows a wide spectrum ofmicroscopic disease, from non-specific inflam-matory changes through to florid active inflam-matory change associated with crypt architec-tural distortion simulating ulcerative colitis (fig3).41 42 In a small proportion of cases it has beensuggested that such segmental or crescenticcolitis, the latter named for the characteristicinvolvement of the mucosal folds in the sig-moid, may precede the development of distalulcerative colitis.4' This form of colitis, beingpredominantly an inflammation of the luminalmucosa, is quite distinct from diverticulitis, acondition in which stasis, infection, and inflam-mation occur primarily in the diverticula them-selves.43The clinical and histological diagnosis of

chronic inflammatory bowel disease restrictedto the sigmoid colon can be very difficult whendiverticular disease is present. While segmentalor crescentic colitis, associated with diver-ticular disease, can produce histological chan-ges that closely mimic ulcerative colitis, thethree pathological hallmarks of Crohn's disease-granulomas, transmural inflammation, andfissuring ulceration-may all be seen as a resultof diverticular disease of the sigmoid colon.4'An association between Crohn's disease anddiverticular disease has been postulated, but isprobably no more common than would beexpected by the natural prevalence of theconditions." 45 If all the pathological features ofulcerative colitis and Crohn's disease can beseen as a result of the inflammatory complica-tions ofdiverticular disease, can one make a firmdiagnosis of chronic inflammatory bowel dis-ease in the sigmoid colon with diverticulosis?Caution is needed, and other areas of theintestines should be investigated and biopsied,particularly the rectum if ulcerative colitis issuspected. Radiological studies of the sigmoidcolon may be helpful in establishing a doublediagnosis.46 Isolated involvement of the sig-moid colon in Crohn's disease is relativelyunusual and doubt should be cast on any suchdiagnosis in the presence of diverticular diseaseand the absence of any other stigmata ofCrohn's disease.

Infective enterocolitisInfectious (acute, self-limiting) colitis in gen-eral has rather characteristic histologicalfeatures which are unlikely to be confused withulcerative colitis. The predominance of acuteover chronic inflammatory cells, the lack ofcrypt architectural abnormalities, and thepresence of oedema and neutrophils within thecrypt epithelium rather than in the crypt lumenare all helpful features in favour of infectivecolitis.4748 Occasionally, ulcerative colitis-likechanges may be seen in more chronic formsof infectious colitis, particularly chronic

728



j.com/

J Clin P


ber 1991. Dow

nloaded from

http://jcp.bmj.com/

729Pathological mimics of chronic inflammatory bowel disease

Figure 4Microgranulomas(arrowheads) in resolvingphase of culture confirmedsalmonella enterocolitis(haematoxylin and eosin).

shigellosis and amoebiasis.49 Acute chlamydialproctitis may resemble active ulcerativecolitis.50 These ulcerative colitis-like changesoccur more often with the bacteria that cause

inflammation by direct invasion of the mucousmembrane as opposed to those which elaboratea toxin such as Salmonella and Yersinia.5' Itmay be difficult to differentiate infectious colitis,particularly in a resolving phase, from Crohn'sdisease in colonic biopsy specimens.6 Poorlycircumscribed microgranulomas are a featureofsome infective colitides, particularly salmon-ella52 (fig 4) and Campylobacter.53 Microbiologyis not always helpful-in only about 40% ofcases is an infectious agent identified.' Most

Figure S CMV colitis. The inflammatory changes with crypt architecturalabnormalities resemble ulcerative colitis. Inset: CMV inclusions are present(haematoxylin and eosin).

helpful histological variables are evidence ofinflammatory and architectural chronicity andthe presence of well formed epithelioid granu-lomas, these changes favouring a diagnosis ofCrohn's disease. Pseudomembranous entero-colitis, due to Clostridium difficile toxin, is mostlikely to be mistaken for acute ischaemia ratherthan chronic inflammatory bowel disease, al-though early lesions show non-specific patchyinflammation with surface epithelial degenera-tion.54Well formed granulomas are a feature of

some infective colitides, chlamydial infection,55yersiniosis,56 57 and tuberculosis being the mostcharacteristic. Yersiniosis is perhaps the mostlikely infectious enterocolitis to produce-pathological confusion with Crohn's disease.The most helpful differentiating features arecentral necrosis within granulomas and therelative lack of transmural inflammation inyersiniosis.57 Examination of local lymphnodes, if available, is also helpful.58 If there isany doubt yersinia serology should be perfor-med. In cases of Crohn's disease associatedwith a florid sarcoid-like granulomatous re-sponse it may be impossible to rule out tuber-culosis on histological grounds alone. Tuber-culosis is favoured if there is florid coalescentgranulomatous inflammation, extensive caseousnecrosis (some central necrosis is seen inCrohn's granulomas, particularly in the analregion), and nodal granulomas in the absence ofintramural granulomas.59 6' Acid fast bacilli areonly demonstrable in about 50% of intestinaltuberculosis cases but clinical data, includingchest radiology and Mantoux testing, may be ofvalue.61 Other infective granulomatous condi-tions, such as schistosomiasis, deep mycoses,and larval infestations are not often confusedwith Crohn's disease as the infecting organismsare usually readily identifiable.

Viral infection, in particular by cyto-megalovirus (CMV) and herpes simplex virus(HSV), may result in acute enterocolitis,especially in the immunosuppressed. Bothproduce characteristic histological appearancesreminiscent ofchronic inflammatory bowel dis-ease (fig 5): the clinical history and the presenceof inclusion bodies and multinucleate giantcells help to substantiate the diagnosis of viralenterocolitis. CMV colitis is a recognised com-plication of ulcerative colitis and may induceacute fulminant colitis with toxic dilatation.62CMV, HSV, and cryptosporidiosis may pro-duce a florid active enterocolitis, masqueradingas active ulcerative colitis, in patients withAIDS. The correct diagnosis relies on thedemonstration of the organisms or their cyto-pathic effects. The enterocolitis of atypicalmyobacteriosis in AIDS is not usually con-fused with Crohn's disease as the histiocyticinfiltrate diffusely involves the lamina propriaand Ziehl-Neelsen stains show overwhelmingnumbers of bacteria.

Non-specific (microscopic) colitis andlymphocytic colitisMicroscopic colitis denotes a patient groupwith chronic watery diarrhoea and normal



j.com/

J Clin P


ber 1991. Dow

nloaded from

http://jcp.bmj.com/

Shepherd

radiological and sigmoidoscopic findings butmicroscopically abnormal colorectal biopsyspecimens.6364 The term is ambiguous as otherconditions, including both ulcerative colitisand Crohn's disease, may show histopath-ological abnormalities in the face of normalmacroscopic appearances.65"' A characteristicpathological feature of the disease is surfaceepithelial degeneration with a noticeablyincreasedintraepithelial lymphocyte infiltrate,67and it has been proposed that the condition berenamed lymphocytic colitis.6768 His-tologically, lymphocytic colitis shares some ofthe features of collagenous colitis, but thesecolitides can be differentiated on examination ofthe biopsy specimen.67 The chronic inflam-matory changes that occur in lymphocytic(microscopic) colitis may suggest a diagnosis ofchronic inflammatory bowel disease: the lack ofcrypt architectural distortion and active cryp-

;

Zjw -s*J;-4P....... = J Stw;;'*4 iSXv W=gg '

Figure 6 Primary malignant lymphoma of the intestines. (A) deep destructive fissuringulceration in a small intestinal lymphoma. At this magnification the lymphomatousinfiltrate is not obvious and the pathology could easily be mistaken for Crohn's disease.(B) diffuse intramucosal lymphoma of the colon mimicking chronic inflammatory boweldisease (haematoxylin and eosin).

titis, together with the characteristic changes oflymphocytic colitis, enables the pathologist todistinguish the condition from chronic inflam-matory bowel disease.67

Ischaemic enterocolitis and Beh9et'ssyndromeAcute forms of ischaemic enterocolitis presentlittle differential diagnostic dilemma. Chronicstrictures in ischaemic enterocolitis may sug-gest a diagnosis of Crohn's disease. Histo-logically, the reparative and chronic phases ofischaemia disclose microscopic fissures, cryptepithelial regeneration and distortion, andchronic inflammation, features suggestive ofchronic inflammatory bowel disease. Helpfulmicroscopic features for establishing a diag-nosis of ischaemia are the presence of hae-mosiderin laden macrophages in the laminapropria, fibrosis ofthe lamina propria, a relativepaucity ofchronic inflammatory cells and selec-tive damage to the more superficial epitheliumof the crypt.69 7The multisystem disorder Behcet's syn-

drome may affect the intestines: colonic diseaseis sometimes a prominent feature of the syn-drome.7' Ulceration is the most characteristicfeature, either localised to the ileo-caecal regionor more diffusely affecting the colon.72 Colitis inBehcet's syndrome shows typical aphthoidulcers and mucosal cobblestoning; granulomasmay be present.7273 Vasculitis is often seen inBehcet's colitis, affecting small veins and ven-ules.74 Controversy still abounds about theenterocolitis of Behqet's syndrome: it has beenintimated that it is not a specific entity butrather a variant of chronic inflammatory boweldisease.6 7172 Although many ofthe pathologicalfeatures are similar to those in Crohn's colitis,Behcet's syndrome is said to lack theaggregated lymphocytic transmural inflamma-tion of Crohn's disease.75 Current opinion isthat Behcet's enterocolitis should be regardedas a distinct entity until more information onthe pathogenesis of the disease is forth-coming.75 The diagnosis of Behcet'senterocolitis clearly relies on corroboration ofpathological findings with clinical data.

Malignant lymphomaThere are two situations in which primarymalignant lymphoma of the gut may mimicinflammatory bowel disease. Deep, destructivefissuring ulceration is a highly characteristicfeature of high grade small and large intestinallymphomas of both B and T cell phenotype76(fig 6A) and this may mimic Crohn's disease,particularly in those tumours with relativelyfew neoplastic cells and innumerable eosino-phils.77 In the large bowel diffuse lymphoma isrelatively unusual but may be seen in bothprimary and secondary disease.78 Occasionallythe disease is predominantly mucosal andshows similar macroscopic and histologicalchanges to those of acute ulcerative colitis (fig6B).7`1 In this situation immunohisto-chemistry may help to differentiate the condi-tions." Very occasionally ulcerative colitis and

730



j.com/

J Clin P


ber 1991. Dow

nloaded from

http://jcp.bmj.com/

Pathological mimics of chronic inflammatory bowel disease

Figure 7 Pneumatosis cystoides intestinalis. The mucosa shows gross crypt atrophy anddistortion and the muscularis mucosae is thickened. Close inspection of the lower edge ofthe biopsy specimen shows the histiocytic lining of a submucosal gas cyst. Intramucosalgranulomas were also present (haematoxylin and eosin).

malignant lymphoma may coexist: primarymalignant lymphoma of the large bowel is arare but well recognised complication ofchronic ulcerative colitis.82

MiscellaneousIMMUNE DEFICIENCY SYNDROMESIt is the acquired immunodeficiency syndromesthat are most often confused histologically withchronic inflammatory bowel disease. Graft-versus-host disease (GvHD) in the acute phaseshows crypt distortion and degeneration withcrypt abscesses83: as the disease progressesthere is gross crypt atrophy.84 The absence of apredominant inflammatory component and thepresence of the "exploding crypt lesion", inwhich individual cell necrosis can be shownhistologically and ultrastructurally in the cryptepithelium, are helpful in distinguishingGvHD from chronic ulcerative colitis.85Although clinically important intestinal path-ology is caused by specific infection in patientswith AIDS, AIDS is also complicated by a

non-specific enterocolitis.86 This is the result ofimmunologically mediated damage to the intes-tinal epithelium similar to that seen in GvHD,with similar histological appearances.87 In theabsence ofother histological clues the diagnosismay be reached by confirming the presence ofHIV in rectal crypt epithelium by in situhybridisation.88The autosomal recessive disorder chronic

granulomatous disease of childhood (CGD) iscomplicated by a colitis with similarities toCrohn's disease.89 The conditions may be dif-ferentiated by the presence of a histiocyticinfiltrate containing lipid vacuoles and pigmenthistochemically similar to lipofuscin in CGD90and by the normal leucocyte bactericidalactivity in Crohn's disease.9' Inherited immunedeficiency syndromes cause malabsorption,probably as a result ofchronic enteric infection,

731

particularly giardiasis, with histologicalfeatures like those of coeliac disease,92 andoccasionally they cause nodular lymphoidhyperplasia particularly in the ileum93: in-flammatory pathology similar to chronicinflammatory bowel disease is generally notencountered.

NEUROMUSCULAR AND VASCULAR CHANGES IN THEINTESTINESThe lesions of Crohn's disease may undergospontaneous healing.94 In these cases a confus-ing histological picture is produced withhyalinised granulomas and little or no inflam-mation. The most striking findings are oftenseen in the connective tissues. Neuronal hyper-plasia, both of ganglion cells95 and nervetrunks,96 may mimic diffuse neurofibroma orintestinal ganglioneuromatosis,97 while the vas-cular degenerative changes may suggest aprimary vasculitis. Another striking feature oflate stage Crohn's disease is the muscularisa-tion of the fibrotic submucosa, a feature alsoseen in ischaemic and radiation enterocolitis.98The late stage of any localised intestinal ulcercould be misinterpreted as Crohn's disease. Inthese cases the presence of fissuring ulceration,granulomas, or active inflammation elsewhereare features which can be used toward a diag-nosis of Crohn's disease.

EOSINOPHILIC INFILTRATES OF THE GUTSeveral heterogeneous conditions may cause atissue eosinophilia in the intestines99: eosino-phils may be a prominent component of theinflammatory infiltrate in both ulcerative colitisand Crohn's disease.'"' Eosinophilic gas-troenteritisl0l 102 iS characterised by a floridtissue and peripheral blood eosinophilia, withsome histopathological similarities to chronicinflammatory bowel disease. Nevertheless,eosinophilic gastroenteritis has characteristicclinical associations: the condition lacks thecharacteristic histopathological changes ofCrohn's disease and the crypt architecturalabnormalities of ulcerative colitis are notpresent.'

ISOLATED MUCOSAL GRANULOMASSmall isolated granulomas are not an unusualfeature in rectal mucosal biopsy specimens. Inthe absence of collateral evidence to support adiagnosis of Crohn's disease a cause may not befound. Refractile crystals are always demon-strable in barium granulomas'03 and it is mostunusual to see evidence of sarcoidosis in rectalmucosa.1'0 It is likely that most of theseincidental granulomas represent a tissue res-ponse to mucin, possibly as a result of cryptobstruction, inflammation, and disruption.

PNEUMATOSIS CYSTOIDES INTESTINALISPneumatosis is a rare condition characterisedby gas filled cysts which particularly affect thecolonic submucosa.'05 Clinically, the conditionis more likely to simulate polyposis syn-dromes,'06 but histologically the overlyingmucosa shows features which may suggestchronic inflammatory bowel disease if biopsiesare superficial."0 These include chronic inflam-

I

il . '.'

I



j.com/

J Clin P


ber 1991. Dow

nloaded from

http://jcp.bmj.com/

Shepherd

mation, crypt distortion as a result of theunderlying submucosal gas cysts, and intra-mucosal granulomas (fig 7). Only when biopsyspecimens that include the submucosa areexamined does the true nature of the diseasebecome apparent.'07

SummaryWhen all of the macroscopic and microscopicfeatures of Crohn's disease and ulcerativecolitis are present, the correct diagnosis isusually made without difficulty. When some ofthe changes are absent, the accuracy of diag-nosis is reduced. This review has outlined thosediseases which feature some of these patho-logical changes and may masquerade asidiopathic chronic inflammatory bowel disease.Some of the pathological mimics are iatrogenicwhile other common diseases, such as bacterialinfection, ischaemia, and diverticulosis mayproduce confusing histological appearances.The picture is complicated by the fact thatmany of these pathological imitators maythemselves cause or predispose to chronicinflammatory bowel disease, or may complicatechronic inflammatory bowel disease. Forexample, drugs and infectious agents are recog-nisable causes of relapse in ulcerative colitis;Crohn's disease may cause diverticulitis inpatients with diverticulosis; and lymphomamay complicate ulcerative colitis. It behoves allpractising histopathologists to recognise thesemimics ofulcerative colitis and Crohn's diseaseto ensure appropriate management for patientswith inflammatory pathology of the intestines.

I am most grateful to Ms Jill Maybee for photographic assistanceand to Dr Bryan Warren for helpful discussions.

1 Mortensen N. Progress with the pouch-restorative procto-colectomy for ulcerative colitis. Gut 1988;29:561-5.

2 Beart RW. Proctocolectomy and ileoanal anastamosis.World J Surg 1988;12:160-3.

3 Price AB. Overlaps in the spectrum of non-specific inflam-matory bowel disease-"colitis indeterminate". J ClinPathol 1978;31:567-77.

4 Jass JR. The large intestine. In Morson BC, ed. Systemicpathology. 3rd ed. Vol 3: Alimentary tract. Edinburgh:Churchill Livingstone, 1987:334-46.

5 Morson BC, Dawson IMP, Day DW, Jass JR, Price AB,Williams GT. Inflammatory disorders. In: Morson &Dawson's Gastrointestinal Pathology. 3rd ed. Oxford:Blackwell Scientific, 1990:514-20.

6 Tanaka M, Riddell RH. The pathological diagnosisand differential diagnosis of Crohn's disease. Hepato-gastroenterology 1990;37:18-31.

7 Lusk LB, Reichen J, Levine JS. Aphthous ulceration indiversion colitis. Gastroenterology 1984;87:1 171-3.

8 Harig JM, Soergel KH, Komorowski RA, Wood CM.Treatment of diversion colitis with short chain fatty acidirrigation. N Engl J Med 1989;320:23-38.

9 Komorowski RA. Histologic spectrum of diversion colitis.Am J Surg Pathol 1990;14:548-54.

10 Roe AM, Warren BF, Brodribb AJM, Brown C. Diversioncolitis and involution of the defunctioned anorectum. Gut1990;31:Al 174.

11 Glotzer DJ, Glick ME, Goldman H. Proctitis and colitisfollowing diversion of the faecal stream. Gastroenterology1981;80:438-41.

12 Korelitz BI, Cheskin LJ, Sohn N, Sommers SC. Proctitisafter fecal diversion in Crohn's disease and its eliminationwith reanastamosis: implications for surgical man-agement. Gastroenterology 1984;87:710-13.

13 Murray FE, O'Brien MJ, Birkett DH, Kennedy SM,LaMont JT. Diversion colitis: pathological findings in aresected sigmoid colon and rectum. Gastroenterology1987;93:1404-8.

14 Ma CK, Gottlieb C, Haas PA. Diversion colitis: a clinico-pathological study of 21 cases. Hum Pathol 1990;21:429-36.

15 Anonymous. Diversion colitis. [Editorial]. Lancet1989;i:764.

16 Truelove SC, Ellis H, Webster CU. Place of double-barrelled ileostomy in ulcerative colitis and Crohn'sdisease of the colon. Br MedJ 1965;i:150-3.

17 Burman JH, Thompson H, Cooke W, Alexander-WilliamsJ. The effects of diversion of intestinal contents on theprogress of Crohn's disease of the large bowel. Gut1971;12:1 1-15.

18 Warren BF, Shepherd NA, Bartolo DCC, Bradfield JWB.Histological changes of the defunctioned rectum inulcerative colitis. J Pathol 1991;63:69A.

19 Pemberton JH, Kelly KA, Beart RW, Dozois RR, WolffBG, Ilstrup DM. Ileal pouch-anal anastamosis for chroniculcerative colitis. Long-term results. Ann Surg1987;206:504-13.

20 Nicholls RJ, Pezim ME. Restorative proctocolectomy withileal reservoir for ulcerative colitis and familial aden-omatous polyposis: a comparison of three reservoirdesigns. Br J Surg 1985;72:470-4.

21 Hulten L, Svaninger G. Facts about the Kock continentileostomy. Dis Colon Rectum 1984;27:553-7.

22 Shepherd NA, Jass JR, Duval I, Moskowitz RL, NichollsRJ, Morson BC. Restorative proctocolectomy with ilealreservoir-pathological and histochemical study ofmucosal biopsy specimens. J Clin Pathol 1987;40:601-7.

23 Tytgat GNJ, Deventer SJH. Pouchitis. Int J Colorectal Dis1988;3:226-8.

24 Shepherd NA. The pelvic ileal reservoir: apocalypse later?Br Med J 1990;301:886-7.

25 Keighley MRB. Discussiorl in Pouchitis Workshop. Int JColorectal Dis 1989;4:213.

26 Shepherd NA. The pelvic ileal reservoir: pathology andpouchitis. Neth J Med 1990;37:S57-S64.

27 Davies DR, Brightmore T. Idiopathic and drug-inducedulceration ofthe small intestine. BrJSurg 1970;57:134-9.

28 Leyonmarck CE, Raf L. Ulceration of the small intestinedue to slow-release potassium tablets. Acta Chir Scand1985;151:273-8.

29 Lang J, Price AB, Levi AJ, Burke M, Gumpel JM,Bjarnason I. Diaphragm disease: pathology of the smallintestine induced by non-steroidal anti-inflammatorydrugs. J Clin Pathol 1988;41:516-26.

30 Sheers R, Williams WR. NSAIDs and gut damage. Lancet1989;ii:1 154.

31 Hall RI, Pelty AH, Cobden I, Lendrum R. Enteritis andcolitis associated with mefanamic acid. Br Med J1983;287:1 182.

32 Anonymous. NSAIDs and gut damage. [Editorial]. Lancet1989;ii:600.

33 Graham CF, Gallagher K, Jones JK. Acute colitis withmethyldopa. N Engl JMed 1981;304:1044-5.

34 Martin DM, Goldman JA, Gilliam J, Nasrallah SM. Gold-induced eosinophilic enterocolitis: response to oralchromolyn sodium. Gastroenterology 1981;80:1567-70.

35 FlochMH, Hellman L. The effectof5-fluorouracil on rectalmucosa. Gastroenterology 1965;48:430-7.

36 Saunders DR, Sillery J, Rachmilewitz D, Rubin CE,Tytgat GNJ. Effect of bisacodyl on the structure andfunction of rodent and human intestine. Gastroenterology1977;72:849-56.

37 Leriche M, Devroede G, Sanchez G, Rossano J. Changes inthe rectal mucosa induced by hypertonic enemas. DisColon Rectum 1978;21:227-36.

38 Levy N, Gaspar E. Rectal bleeding and indomethacinsuppositories. Lancet 1975;i:577.

39 Cawthome SJ, Gibbs NM, Marks CG. Segmental colitis: anew complication of diverticular disease. Gut1983;24:A500.

40 Mathus-Vliegen EMH, Tytgat GNJ. Polyp-simulatingmucosal prolapse syndrome in (pre)-diverticular syn-drome. Endoscopy 1986;18:84-6.

41 Gore S, Shepherd NA, Wilkinson SP. Crescentic colitis:the clinical and histopathological spectrum ofa distinctiveendoscopic appearance in the sigmoid colon. J Pathol1991 (in press).

42 Sladen GE, Filipe MI. Is segmental colitis a complicationofdiverticular disease? Dis Colon Rectum 1984;27:513-14.

43 du Boulay CEH. Diverticular disease of the colon andsolitary ulcer syndrome. In Whitehead R, ed. Gastrointes-tinal and oesophageal pathology. Edinburgh: ChurchillLivingstone, 1989:355-60.

44 Schmidt GT, Lennard-Jones JE, Morson BC, Young AC.Crohn's disease of the colon and its distinction fromdiverticulitis. Gut 1968;9:7-16.

45 McCue J, Copper MJ, Rasbridge SA, Lock MR. Coexis-tent Crohn's disease and sigmoid diverticulosis. PostgradMed J 1989;65:636-9.

46 Meyers AA, Alonso DR, Morson BC, Bartram CI. Patho-genesis of diverticulitis complicating granulomatouscolitis. Gastroenterology 1978;74:24-31.

47 Surawicz CM, Belic L. Rectal biopsy helps to distinguishacute self-limited colitis from idiopathic inflammatorybowel disease. Gastroenterology 1984;86:104-13.

48 Nostrant TT, Kumar NB, Appelman HD.Histopathologydifferentiates acute self-limited colitis from ulcerativecolitis. Gastroenterology 1987;92:318-28.

49 Anand BS, Malhotra V, Bhattacharya SK, et al. Rectalhistology in acute bacillary dysentery. Gastroenterology1986;90:654-60.

50 Levine JS, Smith PD, Brugge WR. Chronic proctitis inmale homosexuals due to lymphogranuloma venereum.Gastroenterology 1980;79:563-5.

51 Jass JR. The large intestine. In: Morson BC, ed. Systemicpathology. 3rd ed. Vol 3: Alimentary tract. Edinburgh:Churchill Livingstone, 1987:325-9.

52 Day DW, Mandal BK, Morson BC. The rectal biopsyappearances of salmonella colitis. Histopathology1978;2:1 17-31.

53 Price AB, Jewkes J, Sanderson PJ. Acute diarrhoea:

732



j.com/

J Clin P


ber 1991. Dow

nloaded from

http://jcp.bmj.com/

Pathological mimics of chronic inflammatory bowel disease

campylobacter colitis and the role of rectal biopsy. J ClinPathol 1979;32:990-7.

54 Price AB, Davies DR. Pseudomembranous colitis. J ClinPathol 1977;30:1-12.

55 Quinn TC, Goodell SE, Mkrtichian E. Chlamydiatrachomatis proctitis. N EnglJ Med 1981;305:195-200.

56 Vantrappen G, Agg HO, Ponette E, Geboes K, Bertrand P.Yersinia enteritis and enterocolitis: gastroenterologicalaspects. Gastroenterology 1977;72:220-7.

57 Haggitt RC. The differential diagnosis ofidiopathic inflam-matory bowel disease. In: Norris HT, ed. Pathology of thecolon, small intestine and anus. New York: ChurchillLivingstone, 1983:21-59.

58 El-Meraghi NRH, Mair NS. The histopathology ofentericinfection with yersinia pseudotuberculosis. Am J ClinPathol 1979;71:631-9.

59 Jass JR, Shepherd NA, Maybee JD. Inflammatory dis-orders. In: Atlas of surgical pathology of the colon, rectumand anus. Edinburgh: Churchill Livingstone. 1989:188-9.

60 Morson BC, Dawson IMP, Day DW, Jass JR, Price AB,Williams GT. Inflammatory disorders. In: Morson &Dawson's Gastrointestinal Pathology. 3rd ed. Oxford:Blackwell Scientic, 1990:275-6.

61 Haggitt RC. The differential diagnosis ofidiopathic inflam-matory bowel disease. In: Norris HT, ed. Pathology of thecolon, small intestine and anus. New York: ChurchillLivingstone, 1983:21-59.

62 Cooper HS, Raffensperger ED, Jonas L, Fitts WT.Cytomegalovirus inclusions in patients with ulcerativecolitis and toxic dilatation requiring colonic resection.Gastroenterology 1977;72:1253-6.

63 Read NW, Kreis GJ, Read MG, Santa Ana CA, MorawskiSG, Fordtran JS. Chronic diarrhoea of unknown origin.Gastroenterology 1980;78:264-71.

64 Kingham JGC, Levison DA, Ball JA, Dawson AM.Microscopic colitis-a cause of chronic watery diarrhoea.Br JMed 1982;285:1601-4.

65 Elliott PR, Williams CB, Lennard-Jones JE, et al. Colono-scopic diagnosis of minimal change colitis in patients witha normal sigmoidoscopy and normal air-contrast bariumenema. Lancet 1982;i:650-1.

66 Sanderson IR, Boyle S, Williams CB, Walker-Smith JA.Histological abnormalities in biopsies from macro-scopically normal colonoscopies. Arch Dis Child1986;61:271-4.

67 Lazenby A, Yardley J, Giardiello F, Jessurun J, Bayless T.Lymphocytic ("microscopic") colitis. A comparativehistopathologic study with particular reference tocollagenous colitis. Hum Pathol 1989;20:18-28.

68 Giardiello FM, Lazenby AJ, Bayless TM, et al. Lympho-cytic (microscopical) colitis. Clinicopathologic study of 18patients and comparison to collagenous colitis. Dig Dis Sci1989;34: 1730-8.

69 Whitehead R. The pathology of intestinal ischaemia. ClinGastroenterol 1972;1:613-37.

70 Morson BC, Dawson IMP, Day DW, Jass JR, Price AB,Williams GT. Vascular disorders. In: Morson & Dawson'sgastrointestinal pathology. 3rd ed. Oxford: BlackwellScientific, 1990:551-9.

71 Smith GE, Kime LR, Pitcher JL. The colitis of Behcet'sdisease: a separate entity? AmJ Dig Dis 1973;18:987-99.

72 O'Connell DJ, Courtney JV, Riddell RH. Colitis ofBehcet's syndrome: radiologic and pathologic features.Gastrointest Radiol 1980;5:173-9.

73 Baba S, Maruta M, Ando K, Teramoto T, Endo I.Intestinal Behcet's disease; report of five cases. Dis ColonRectum 1976;9:428-40.

74 Lee RG. The colitis of Behcet's syndrome. Am J SurgPathol 1986;10:888-93.

75 Morson BC, Dawson IMP, Day DW, Jass JR, Price AB,Williams GT. Inflammatory disorders. In: Morson &Dawson's gastrointestinal pathology. 3rd ed. Oxford:Blackwell Scientific, 1990:290-1.

76 Morson BC, Dawson IMP, Day DW, Jass JR, Price AB,Williams GT. Non-epithelial tumours. In: Morson &Dawson'sgastrointestinalpathology. 3rd ed. Oxford: Black-well Scientific, 1990:376-7.

77 Shepherd NA, Blackshaw AJ, Hall PA, et al. Malignantlymphoma with eosinophilia of the gastrointestinal tract.Histopathology 1987;11:1 15-30.

78 Richards MA. Lymphoma of the colon and rectum. Post-grad MedJ 1986;62:615-20.

79 Friedman HB, Silver GM, Brown CH. Lymphoma of thecolon simulating ulcerative colitis. Am J Dig Dis 1968;13:910-17.

80 Myerson P, Myerson D, Miller D, DeLuca VA, LawsonJP. Lymphosarcoma of the bowel masquerading asulcerative colitis. Dis Colon Rectum 1974;17:710-5.

81 Sagar S, Selby P, Sloane J, McElwain TJ. Colorectallymphoma simulating inflammatory colitis and diagnosedby immunohistochemistry. Postgrad Med J 1986;62:51-3.

82 Shepherd NA, Hall PA, Williams GT, et al. Primarymalignant lymphoma of the large intestine complicatingchronic inflammatory bowel disease. Histopathology 1989;15:325-37.

83 Epstein RJ, McDonald GB, Sale GE, Shulman HM,Thomas ED. The diagnostic accuracy of the rectal biopsyin acute graft-versus-host disease: a prospective study ofthirteen patients. Gastroenterology 1980;78:764-71.

84 Thorning D, Howard JD. Epithelial denudation in thegastrointestinal tract of two bone marrow transplantrecipients. Hum Pathol 1986;17:560-6.

85 Sale GE, McDonald GB, Shulman HM, Thomas ED.Gastrointestinal graft-versus-host disease in man. Aclinicopathological study of rectal biopsy. Am J SurgPathol 1979;3:291-9.

86 Kotler DP, Gaetz HP, Lange M, Klein EB, Holt PR.Enteropathy associated with the acquired immuno-deficiency syndrome. Ann Intern Med 1984;101:421-8.

87 Kotler DP, Weaver SC, Terzakis JA. Ultrastructuralfeatures of epithelial cell destruction in rectal crypts ofpatients with AIDS. Am J Surg Pathol 1986;1O:531-8.

88 Nelson JA, Wiley CA, Reynolds-Kohler C, Reese CA,MargarettenW, Levy JA. Human immunodeficiency virusdetected in bowel epithelium from patients with gastro-intestinal symptoms. Lancet 1988;i:259-62.

89 Werlin SL, Chusid MJ, Caya J, Oechler HW. Colitis inchronic granulomatous disease. Gastroenterology 1982;82:328-31.

90 Ament ME, Ochs HD. Gastrointestinal manifestations ofchronic granulomatous disease. N Engl J Med1973;288:382-7.

91 Yogman MW, Touloukian RJ, Gallagher R. Intestinalgranulomatosis in chronic granulomatous disease and inCrohn's disease. N Engl J Med 1974;290:228.

92 Ament ME, Rubin CE. Relation of giardiasis to abnormalintestinal structure and function in gastrointestinalimmunodeficiency syndromes. Gastroenterology 1972;62:216-26.

93 Hermans PE, Huizenga kA, Hoffman HN, Brown AL,Markowitz H. Dysgammaglobulinaemia associated withnodular hyperplasia of the small intestine. Am J Med1966;40:78-89.

94 Gray BK, Lockhart-Mummery HE, Morson BC. Crohn'sdisease of the anal region. Gut 1965;6:515-24.

95 Davis DR, Dockerty MB, Mayo CW. The myentericplexus in regional enteritis: a study of ganglion cells in theileum in 24 cases. Surg Gynecol Obstet 1955;101:208-16.

96 Whitehead R. Pathology of Crohn's disease. In: Kirsner JB,Shorter RG, eds. Inflammatory bowel disease. 2nd ed.Philadelphia: Lea & Febiger, 1980:296-307.

97 Carney JA, Go VLW, Sizemore GW, Hayle AB. Alimen-tary tract ganglioneuromatosis. N Engl J Med 1976;295:1287-91.

98 Shepherd NA, Jass JR. Neuromuscular and vascularhamartoma: is it Crohn's disease? Gut 1987;28:1663-8.

99 Blackshaw AJ, Levison DA. Eosinophilic infiltrates in thegastrointestinal tract. J Clin Pathol 1986;39:1-7.

100 Heatley RV, James PD. Eosinophils in the rectal mucosa. Asimple method of predicting the outcome of ulcerativecolitis. Gut 1979;20:787-91.

101 Klein NC, Hargrove RL, Sleisinger MH, Jeffries GH.Eosinophilic gastroenteritis. Medicine 1970;49:299-319.

102 Johnstone JM, Morson BC. Eosinophilic gastroenteritis.Histopathology 1978;2:335-48.

103 Levison DA, Crocker PR, Smith A, Blackshaw AJ, BartramCI. Varied light and scanning electron microscopicappearances of barium sulphate in smears and histologicalsections. J Clin Pathol 1984;37:481-7.

104 Tobi M, Kobrin I, Ariel I. Rectal involvement in sarco-idosis. Dis Colon Rectum 1982;25:491-3.

105 Galandiuk S, Fazio VW. Pneumatosis cystoides intestin-alis. A review of the literature. Dis Colon Rectum 1986;29:358-63.

106 Calne RY. Gas cysts of the large bowel simulating multiplepolyposis. BrJ Surg 1959;47:212-15.

107 Pieterse AS, Leong ASY, Rowland R. The mucosalchanges and pathogenesis of pneumatosis cystoides intes-tinalis. Hum Pathol 1986;16:683-8.

733



j.com/

J Clin P


ber 1991. Dow

nloaded from

http://jcp.bmj.com/

clin pathological mimicsof chronic inflammatory bowel disease · non-specific chronic inflammation,...

Documents