Histopathology of Diaphragm Disease of the Small Intestine A ...

8
518 Am J Clin Pathol 2008;130:518-525 518 DOI: 10.1309/7DDT5TDVB5C6BNHV © American Society for Clinical Pathology Anatomic Pathology / DIAPHRAGM DISEASE OF THE SMALL INTESTINE CME Histopathology of Diaphragm Disease of the Small Intestine A Study of 10 Cases From a Single Institution Giovanni De Petris, MD, 1 and José I. López, MD 2 Key Words: Nonsteroidal anti-inflammatory drugs; Diaphragm disease; Small intestine DOI: 10.1309/7DDT5TDVB5C6BNHV Abstract Diaphragm disease (DD) is an uncommon gastrointestinal abnormality typical of nonsteroidal anti-inflammatory drug (NSAID)-induced injury. DD of the small intestine is well defined in its gross appearance (multiple circumferential stenosing lesions); however, its histologic features have been studied in few cases. To better define and update the histologic features of DD, we describe 10 cases in which patients underwent resection of portions of small intestine for DD. Selected gross features and 12 microscopic features were assessed. The typical gross appearance of DD was associated with focal chronic injury of the mucosa with mild to moderate inflammatory infiltrate. A wide array of additional abnormalities was found: eosinophilic enteritis (3 cases), inflammatory fibroid polyp (1 case), enteritis cystica profunda (1 case), villous atrophy (1 case), and neuromuscular and vascular hamartoma-like changes (9 cases). DD and, therefore, NSAID-related injury, should be considered in the differential diagnosis of several conditions affecting the small intestine. Diaphragm disease (DD) is a rare gastrointestinal con- dition considered characteristic of injury due to prolonged intake of nonsteroidal anti-inflammatory drugs (NSAIDs). 1 The disease is characterized by circumferential lesions of short length (<5 mm), located most commonly in the small intestine, causing multiple stenosis of the lumen. The importance of NSAID-related enteropathy has been empha- sized recently, 2 but the histologic features of DD of the small intestine have been reported in only a limited number of cases. 3-14 The initial thorough description of DD by Lang et al 15 was followed by several others 16-19 that have revealed few additional findings. These additional findings have included villous atrophy, 16 association or overlap with so- called neuromuscular and vascular hamartoma (NMVH), 17 rare association with pseudoneoplastic lesions such as inflammatory fibroid polyp (IFP), 18 and, recently, an entity called NSAID-associated submucosal fibrous nodules of the small intestine. 19 Cases in which NSAID intake could not be proven have also been published. 20 The need to better recognize the histologic features of DD has been increased by the appearance of novel clinical means of imaging and recognition of cases in the small intestine (eg, video capsule endoscopy and double-balloon endoscopy). The study has one aim: to describe the spectrum of histopathologic fea- tures of DD to provide knowledge useful to the diagnosis of this entity. Materials and Methods The Mayo Clinic Arizona Internal Review Board approved the study. Ten cases of patients who underwent resection of a portion of small intestine because of DD Upon completion of this activity you will be able to: • describe the gross appearance of diaphragm disease of the intestine. • define the clinical features, including etiology, clinical presentation, and endoscopic appearance, of diaphragm disease of the intestine. • list the histologic features of the intestine affected by diaphragm disease. The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this educational activity for a maximum of 1 AMA PRA Category 1 Credit ™ per article. The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose. Questions appear on p 642. Exam is located at www.ascp.org/ajcpcme. Downloaded from https://academic.oup.com/ajcp/article/130/4/518/1765278 by guest on 16 September 2022

Transcript of Histopathology of Diaphragm Disease of the Small Intestine A ...

518 Am J Clin Pathol 2008;130:518-525518 DOI: 10.1309/7DDT5TDVB5C6BNHV

© American Society for Clinical Pathology

Anatomic Pathology / Diaphragm Disease of the small intestine

CME

Histopathology of Diaphragm Disease of the Small Intestine

A Study of 10 Cases From a Single Institution

Giovanni De Petris, MD,1 and José I. López, MD2

Key Words: Nonsteroidal anti-inflammatory drugs; Diaphragm disease; Small intestine

DOI: 10.1309/7DDT5TDVB5C6BNHV

A b s t r a c t

Diaphragm disease (DD) is an uncommon gastrointestinal abnormality typical of nonsteroidal anti-inflammatory drug (NSAID)-induced injury. DD of the small intestine is well defined in its gross appearance (multiple circumferential stenosing lesions); however, its histologic features have been studied in few cases. To better define and update the histologic features of DD, we describe 10 cases in which patients underwent resection of portions of small intestine for DD. Selected gross features and 12 microscopic features were assessed. The typical gross appearance of DD was associated with focal chronic injury of the mucosa with mild to moderate inflammatory infiltrate. A wide array of additional abnormalities was found: eosinophilic enteritis (3 cases), inflammatory fibroid polyp (1 case), enteritis cystica profunda (1 case), villous atrophy (1 case), and neuromuscular and vascular hamartoma-like changes (9 cases). DD and, therefore, NSAID-related injury, should be considered in the differential diagnosis of several conditions affecting the small intestine.

Diaphragm disease (DD) is a rare gastrointestinal con-dition considered characteristic of injury due to prolonged intake of nonsteroidal anti-inflammatory drugs (NSAIDs).1 The disease is characterized by circumferential lesions of short length (<5 mm), located most commonly in the small intestine, causing multiple stenosis of the lumen. The importance of NSAID-related enteropathy has been empha-sized recently,2 but the histologic features of DD of the small intestine have been reported in only a limited number of cases.3-14 The initial thorough description of DD by Lang et al15 was followed by several others16-19 that have revealed few additional findings. These additional findings have included villous atrophy,16 association or overlap with so-called neuromuscular and vascular hamartoma (NMVH),17 rare association with pseudoneoplastic lesions such as inflammatory fibroid polyp (IFP),18 and, recently, an entity called NSAID-associated submucosal fibrous nodules of the small intestine.19 Cases in which NSAID intake could not be proven have also been published.20 The need to better recognize the histologic features of DD has been increased by the appearance of novel clinical means of imaging and recognition of cases in the small intestine (eg, video capsule endoscopy and double-balloon endoscopy). The study has one aim: to describe the spectrum of histopathologic fea-tures of DD to provide knowledge useful to the diagnosis of this entity.

Materials and Methods

The Mayo Clinic Arizona Internal Review Board approved the study. Ten cases of patients who underwent resection of a portion of small intestine because of DD

Upon completion of this activity you will be able to:•describethegrossappearanceofdiaphragmdiseaseoftheintestine.• definetheclinicalfeatures,includingetiology,clinicalpresentation,andendoscopicappearance,ofdiaphragmdiseaseoftheintestine.

• listthehistologicfeaturesoftheintestineaffectedbydiaphragmdisease.

TheASCPisaccreditedbytheAccreditationCouncilforContinuingMedicalEducationtoprovidecontinuingmedicaleducationforphysicians.TheASCPdesignatesthiseducationalactivityforamaximumof1AMAPRACategory1Credit™perarticle.

Theauthorsofthisarticleandtheplanningcommitteemembersandstaffhavenorelevantfinancialrelationshipswithcommercialintereststodisclose.

Questionsappearonp642.Examislocatedatwww.ascp.org/ajcpcme.

Dow

nloaded from https://academ

ic.oup.com/ajcp/article/130/4/518/1765278 by guest on 16 Septem

ber 2022

Am J Clin Pathol 2008;130:518-525 519519 DOI: 10.1309/7DDT5TDVB5C6BNHV 519

© American Society for Clinical Pathology

Anatomic Pathology / original article

were identified from the surgical pathology files of Mayo Clinic between 2002 and 2006. The presence or absence of the histologic features listed in zTable 1z was assessed. The list includes features of Crohn disease, collagenous enteritis, eosinophilic infiltrate, and pseudoneoplasms (IFP and submucosal fibrous nodule) possibly associated with DD. The term ulcer-associated cell lineage (UACL) was preferred to pyloric metaplasia to indicate metaplastic pale glandular cells because these have been found to recapitu-late the Brunner gland line of differentiation21 rather than pyloric gland.

The following gross features were also recorded: (1) diaphragms (number and location) defined as an intraluminal, circumferential, and protruding ridge of tissue with luminal stenosis; (2) length of intestine removed; (3) site of intestine involved by the diaphragms as described by the surgeon; (4) presence of circumferential ulceration; (5) ring-like depres-sion on the external surface of the intestine; and (6) adipose tissue wrapping.

H&E-stained slides were used. Immunohistochemical stains for fascin (clone 55K-2, DAKO, Carpinteria, CA), CD34 (clone QbEnd-10, DAKO), smooth muscle actin (SMA; clone 1A4, DAKO), and CD68 (clone kp-1, DAKO) were used for the diagnosis of IFP and were performed according to the standard protocol for an automated immunostainer (DAKO

Autostainer Universal, DAKO), following the manufacturer’s directions. The dilutions used were as follows: CD34, 1:400; CD68, 1:300; fascin, 1:150; and SMA, 1:100. Positive and negative (omitting primary antibody) control samples were used for each antibody.

Results

Clinical FindingsOf the 10 cases identified, 7 patients were women and

3 were men. The median patient age was 69 years (range, 50-79 years; SD, 9.21 years). The clinical findings are sum-marized in zTable 2z. Five patients had been taking NSAIDs for more than 1 year at the time of resection; the remain-ing 5 patients had taken NSAIDs, not otherwise specified, at high and unspecified doses for more than 1 year in the past. No patient had a history of radiation enteritis, celiac disease, or potassium chloride intake. Nine patients were receiving pain control medications that included opioid analgesics. The impression of the surgeons was that they were dealing with an unclear cause of obstruction or with DD, while Crohn disease was not considered likely. Crohn disease had not developed in any patient at follow-up.

zTable 1zMorphologic Features Analyzed in the Series of Cases of Diaphragm Disease of the Small Intestine

Gross features Diaphragms (length, number, and location) Ulceration (circumferential) Ring-like depression of the external surface Adipose tissue wrappingMicroscopic features Injury pattern (focal or diffuse) Increase of lymphocytes/plasma cells Mild (obvious increase and small lymphoid aggregates) Moderate (obvious increase, <4/LPF aggregates with germinal centers) Severe (>4/LPF, clustering of lymphoid aggregates with germinal centers) Fissuring-type ulceration Transmural lymphoid aggregates outside the ulceration Increased intraepithelial lymphocytes (>20/100 superficial cells) Thickened collagen plate with frayed edges and entrapped capillaries Neutrophilic infiltrate in lamina propria (>3/HPF) Cryptitis/crypt abscess Increased eosinophils in lamina propria (>20/HPF) Reactive enterocytes with or without mucosal erosion Granulomas Features of chronic injury Architectural distortion Ulcer-associated cell lineage (as defined by Pera et al21) Duplicated muscularis mucosae Crypt shortfall (separation of crypts from muscularis mucosae) Enteritis cystica profunda (entrapped glands in the submucosa or deeper) Neurovascular/muscular hamartoma-like changes (as defined by Fernando and Mcgovern22) Neoplasm/pseudoneoplasm of the small intestine Inflammatory fibroid polyp (as defined by Noffsinger et al23) Nonsteroidal anti-inflammatory drug–associated submucosal fibrous nodule (as defined by Scott19)

HPF, high-power field; LPF, low-power field.

Dow

nloaded from https://academ

ic.oup.com/ajcp/article/130/4/518/1765278 by guest on 16 Septem

ber 2022

520 Am J Clin Pathol 2008;130:518-525520 DOI: 10.1309/7DDT5TDVB5C6BNHV

© American Society for Clinical Pathology

De Petris and López / Diaphragm Disease of the small intestine

Pathologic FindingsThe gross features zImage 1z were collected in all cases.

The results are summarized in zTable 3z. The diaphragms varied in height from a tall, broad, ulcerated plica circularis (Image 1D), to an ulcerated hump, to a flat, nonspecific ulcer. The terminal 30 cm of ileum were not involved. Every case showed circumferential ulcerations at the tip of the dia-phragms, as well as stenosis of the lumen and short ring-like puckering of the serosa in the same areas (Image 1A). Fat wrapping was not present.

The number of slides examined for each case ranged from 5 to 31 (median, 8 slides). The histopathologic findings are summarized in zTable 4z.

The diaphragms differed from intestine because of scle-rosed eosinophilic connective tissue in the submucosa inter-mixed with smooth muscle, ganglion cells, nerve fibers, and vessels. The smooth muscle component was represented by a simple irregular thickening of the muscularis mucosae or, more commonly, by muscle bundle chaotically arranged. Round to oval, better circumscribed submucosal masses with smooth muscle bundles converging toward the overlying ulceration were also seen zImage 2z. The latter were invariably placed at the apex of the diaphragm. Submucosal NMVH-like changes were in continuity with the adjacent splayed and thickened muscularis mucosae. Fibrosis of the lamina propria was typi-cally present between crypts and muscularis mucosae, often tapering in thickness toward the base of the diaphragms. In 1 patient, the presence of loose stroma between crypts and mus-cularis mucosae, with numerous blood vessels and eosinophils, raised the possibility of IFP zImage 3z. No concentric perivas-cular arrangement of cells was seen. Immunostains for CD34, fascin, and SMA were performed in the case with eosinophilic infiltrate and in 3 additional cases with no particular features. They revealed an increase in fascin-positive cells at the ulcer base but no immunoreactivity for CD34 or SMA in the stromal cells of the fibrous tissue.

The fully developed IFP found in 1 case was sessile, approximately 1 cm in size, based in the submucosa, and associated with overlying ulceration. It showed the typical hypervascular, myxoid, loose stroma with several eosinophils and spindle cells staining with antibodies for CD34 (diffuse positivity), fascin (diffuse positivity), and CD68 (patchy positivity). No perivascular circumferential arrangement of spindle cells was noted.

Reactive enterocytes and crypt distortion were present at the immediate periphery of the ulcers. In every case, focal reactive enteropathy, with or without erosion, was identified, also away from the diaphragms zImage 4z.

Cryptitis and the rare crypt abscesses occurred within the perimeter of the ulcers. The adjacent mucosa appeared normal within a few millimeters with the exception of the presence of crypt shortfall. Lymphoplasmacytic infiltrate in the wall of the intestine was mild in intensity in 8 cases, usually at the periphery of the submucosal soft tissue changes, and moderate in the remaining 2 cases with nodular infiltrates of lympho-cytes directly below ulcerations. Transmural inflammation, fissuring ulcerations, or strings of round lymphoid aggregates along the muscularis propria were never identified.

Among the changes commonly associated with chronic injury, UACL was detected near the ulcers of the diaphragms in 6 cases zImage 5z, while the muscularis mucosae showed double tracking in 1 diaphragm in a single case.

A single case showed enteritis cystica profunda (ECP) with few reactive-appearing glands entrapped in the submu-cosal fibromuscular proliferation.

Eosinophilia limited to the mucosa was identified in 3 cases (average of eosinophils in 10 high-power fields [HPFs] 90/HPF, 129/HPF, and 138/HPF, respectively). The eosino-phils were scattered in the lamina propria diffusely with no evidence of cryptitis or abscesses. No peripheral eosinophilia was present. Patchy villous atrophy zImage 6z was seen in 1 case that also had eosinophilia. The villi/crypt ratio reached

zTable 2zClinical Characteristics in 10 Cases of Diaphragm Disease of the Small Intestine

Obstructive Iron Protein- Orthopedic Retained Case No./ Intestinal Overt Deficiency Losing Procedure/ Video Use of Sex/Age (y) Symptoms Bleeding Anemia Enteropathy Disease Capsule NSAIDs* Follow-up

1/F/72 + + + – + + COX-2 inhibitor 2 y2/F/76 – – + – + – COX-2 inhibitor, ibuprofen 18 mo3/F/69 + – + – + + Past NA4/F/68 – – + + + – Past 1 mo5/F/50 + – + – – + Past 10 d6/M/50 + – + – + – Past 2 y7/M/71 – + + – – – COX-2 inhibitor, aspirin 2 y8/F/57 + – + – + + Past 7 mo9/M/69 + – – – – + Aspirin 5 mo10/F/67 + – + – Ankylosing spondylitis + Diclofenac, aspirin 2 y

NA, not available; NSAIDs, nonsteroidal anti-inflammatory drugs; +, present; –, absent.* Past indicates that NSAIDs were used in the past.

Dow

nloaded from https://academ

ic.oup.com/ajcp/article/130/4/518/1765278 by guest on 16 Septem

ber 2022

Am J Clin Pathol 2008;130:518-525 521521 DOI: 10.1309/7DDT5TDVB5C6BNHV 521

© American Society for Clinical Pathology

Anatomic Pathology / original article

BA

D

*

zImage 1z Typical gross features of diaphragm disease. A, A segment of small intestine shows several serosal constrictions (arrows). B, A diaphragm, consisting of a circumferential tent-like projection in the lumen of the intestine, causes reduction of the lumen to a diameter of a few millimeters. Reproduced with permission from Arch Surg. 2005;140:1162-1166. Copyright © 2005 American Medical Association. C, Multiple diaphragms (arrowheads) create multiple compartments of the lumen of the small intestine. Serosal constrictions are also shown (arrows). D, A low-power view of a section of intestine shows the relative height of the diaphragm (asterisk) compared with the adjacent plicae circularis (H&E, ×10).

zTable 3zGross Features in 10 Cases of Diaphragm Disease of the Small Intestine

Case No.

1 2 3 4 5 6 7 8 9 10

No. of diaphragms 4 7 13 22 7 5 2 14 8 3Ulcerations on diaphragms + + + + + + + + + +Ring-like serosa retraction + + + + + + + + + +Stenosis of the lumen + + + + + + + + + +Location Mid Distal Distal Mid Distal Distal Proximal Mid Mid distal Distal jejunum ileum ileum jejunum ileum ileum ileum jejunum ileum ileumLength of 12 30 61 83 52 32 6 53 37 13 segmentectomy (cm) Fat wrapping – – – – – – – – – –

+ present; –, absent.

C

Dow

nloaded from https://academ

ic.oup.com/ajcp/article/130/4/518/1765278 by guest on 16 Septem

ber 2022

522 Am J Clin Pathol 2008;130:518-525522 DOI: 10.1309/7DDT5TDVB5C6BNHV

© American Society for Clinical Pathology

De Petris and López / Diaphragm Disease of the small intestine

1:1 in the most severely affected areas. Intraepithelial lym-phocytosis and basal plasmacytosis were not identified.

In 1 case, thick smooth muscle bundles encircled the vessels of the submucosa zImage 7z. No fibrous nodules similar to those reported by Scott19 were found. No diffuse enteritis, diffuse crypt distortion, granuloma, markedly increased

intraepithelial lymphocytes, or thickened subsurface col-lagen plate was identified.

DiscussionNSAIDs are known to cause a high incidence of toxic

effects in the small intestine.2 The deleterious effects of NSAIDs in the small bowel are the results of a myriad of mechanisms (including microvascular injury, cyclooxygenase [COX] 1 and/or 2 inhibition, products of inflammatory cell infiltration, and

zTable 4zHistopathologic Features in 10 Cases of Diaphragm Disease of the Small Intestine

Feature No. of Cases

Injury pattern (focal/diffuse) 10/0Increase of lymphocytes/plasma cells Mild 8 Moderate 2 Severe 0Fissuring-type ulceration 0Transmural lymphoid aggregates outside the ulceration 0Intraepithelial lymphocytes increased 0Thickened collagen plate 0Neutrophilic infiltrate in lamina propria 7Cryptitis/crypt abscess 7/2Increased eosinophils in lamina propria 3Reactive enterocytes with or without mucosal erosion 10Granulomas 0Features of chronic injury Architectural distortion 10 Ulcer-associated cell lineage 6 Duplicated muscularis mucosae 1 Crypt shortfall 9Enteritis cystica profunda 1Neurovascular/muscular hamartoma-like changes 9Neoplasm/pseudoneoplasm of the small intestine Inflammatory fibroid polyp 1 Nonsteroidal anti-inflammatory drug–associated submucosal fibrous nodule 0

zImage 2z A low-power view of the club-shaped tip of a diaphragm shows a nodular lesion in the submucosa with convergence of smooth muscle toward the base of an ulcer. The muscularis mucosae is entangled with the submucosal lesion. Note the relatively scant inflammation and the crypt shortfall on both sides of the ulcer (H&E, ×20).

zImage 3z Crypt shortfall due to loose fibrous tissue at the base of the mucosa with several vessels and eosinophils (H&E, ×100).

Dow

nloaded from https://academ

ic.oup.com/ajcp/article/130/4/518/1765278 by guest on 16 Septem

ber 2022

Am J Clin Pathol 2008;130:518-525 523523 DOI: 10.1309/7DDT5TDVB5C6BNHV 523

© American Society for Clinical Pathology

Anatomic Pathology / original article

expression of inducible nitric oxide synthase). Injuries are the result of local/direct contact and systemic effects. The current hypothesis24 is that NSAIDs cause initial reduction in villous microcirculation combined with inhibitory action on the cel-lular oxidative processes of the epithelial cells. These combine to induce increased permeability of the small intestinal mucosa. The injury is aggravated by prolonged exposure to the drug due to enterohepatic recirculation of at least some of the NSAIDs. Increased permeability allows luminal aggressors (indigenous

enteral bacteria, their products, and bile) to gain access to the mucosa and cause inflammation with expression of inducible nitric oxide synthase. Reactive oxygen species formation due to infiltration by neutrophils and produced by injured endothelial cells is likely to bring about tissue injury.

The NSAID COX-2 inhibitors have substantially reduced the propensity for injury of the small intestine, although they retain a component of side effects, including diaphragm formation.11 The reason for the reduced toxic

zImage 4z Reactive enterocytes, in this case with impending erosion, were found in every case of diaphragm disease (H&E, ×100).

zImage 5z Ulcer-associated cell lineage, also known as pyloric metaplasia, was seen at the base of the mucosa, usually near the ulcerated tips of the diaphragms (H&E, ×200).

zImage 6z Severe, patchy villous atrophy was present in a case showing also eosinophilia of the mucosa (H&E, ×200).

zImage 7z Peculiar perivascular smooth muscle arrangement was seen in the submucosa of a diaphragm in 1 case (H&E, ×100).

Dow

nloaded from https://academ

ic.oup.com/ajcp/article/130/4/518/1765278 by guest on 16 Septem

ber 2022

524 Am J Clin Pathol 2008;130:518-525524 DOI: 10.1309/7DDT5TDVB5C6BNHV

© American Society for Clinical Pathology

De Petris and López / Diaphragm Disease of the small intestine

IFP. To our knowledge, this is the second reported case of IFP with NSAID-induced enteropathy.18 IFP is considered a peculiar type of granulation tissue.37 It is possible that a nonspecific injury (in this case due to NSAIDs) can cause the abnormal stromal reaction of IFP.

The presence of histopathologic evidence of chronic injury in DD is not stated explicitly in the original report of Lang et al15 or in a recent authoritative review of gastrointes-tinal diseases.23 Mucosal architectural distortion, crypt short-fall, and UACL are common at the apex of the diaphragms. Minimal active inflammation can also be present. The result-ing picture is that of chronic enteritis with occasional mild activity. The histologic changes appear related primarily to the NSAID-associated damage rather than obstructive enteri-tis. This conclusion is suggested by the presence of inflam-matory changes at the site of obstruction, rather than proximal to it, and by the absence of the ischemic pattern of injury (hemorrhages and larger, irregular ulcers) and of subserosal neutrophilic infiltrates described in obstructive enteritis.

In 1 of our cases, few normal-sized glands with reactive atypia were entrapped in the submucosa, as in ECP. ECP is characterized by glands, often dilated, present in the wall of the small intestine, typically in large polyps of Peutz-Jeghers38 and in Crohn disease.39 ECP arises after repeated ulceration and trauma of the mucosa that allows outgrowth in the sub-mucosa. The repeated or persistent injury due to NSAIDs can, therefore, cause gland entrapment and potentially mimic a neoplastic lesion.

NMVH is a controversial entity most commonly report-ed in the small bowel. It was originally described as “fascicles of smooth muscle derived from the muscularis mucosae, bundles of unmyelinated nerve fibres with scat-tered ganglion cells, and hemangiomatous vessels, causing stenosis.”22 Cortina et al17 argued convincingly that DD and NMVH show clinical and histologic similitude on the basis of the study of 2 patients. An identical conclusion is reached in 9 cases from our series. It appears that NMVH and DD are indistinguishable in at least a subset of patients. The presence of a range of sizes for the NMVH-like lesions in our cases and the association with ulceration and/or reactive enteropathy and/or mucosal fibrosis suggests that they are reactive rather than congenital. Sheperd and Jass40 found NMVH in Crohn disease, additional evidence of the nonspe-cific reactive nature of the lesion.

Our study has revealed that DD has a distinctive gross appearance associated, however, with a wide array of histologic findings. NSAID-related injury should be considered in the presence of chronic enteritis with scant inflammation, mucosal eosinophilic enteritis, ECP, IFP, villous atrophy, and NMVH.

From the 1Department of Pathology, Mayo Clinic Arizona, Scottsdale; and 2Department of Anatomic Pathology, Hospital de Cruces, Basque Country University, Baracaldo, Spain.

effects in the small bowel is not completely clear, although there is evidence to suggest that COX-1 products have a predominant role in the maintenance of the epithelial barrier integrity. The relationship of the drugs with DD is complicated by our incomplete knowledge of the natural history of these lesions and of the type, duration, interactions, and dosage of NSAIDs taken by patients.

DD is typical of NSAID-related injury of the small intestine1 but remains an uncommon, although not vanish-ingly rare, occurrence: In 2% of patients taking NSAIDs and COX-2 inhibitors long-term, DD developed.25 Women are more commonly affected than are men, and the small intestine is more commonly affected than is the colon.26 The route of ingestion of NSAIDs does not preclude the formation of dia-phragms: DD has been reported in a bypassed bowel segment, suggesting systemic effect of NSAIDs.27

It is possible that the formation of diaphragms is a non-specific response to various insults to the intestine. At least 2 cases have been reported in which no NSAID intake could be proven.20,28

We identified 10 cases of DD of the small intestine in 4 years. The majority of the specimens were obtained because of retention of the video capsule, suggesting the difficult preoperative diagnosis of this condition. The gross and clini-cal features were typical and similar to those reported in the literature,3-15,26,29 with the exception of the diagnosis being reached by video capsule endoscopy in 6 of our cases.

We found that DD was characterized histopathologi-cally by focality, lack of severe inflammation, and the reac-tive appearance of the changes. We were surprised to find a remarkable array of abnormalities associated with DD (Table 3). Eosinophilia limited to the mucosal layer with no evidence of eosinophilic cryptitis or eosinophilic abscesses was present in 3 cases. In each of them, the clinical criteria (gastrointestinal symptoms and no evidence of parasitic or extraintestinal dis-ease) for the diagnosis of eosinophilic gastroenteritis were satis-fied.30 Eosinophilia of the small intestine in NSAID enteropa-thy was noted by Lang et al15 and is a known manifestation of NSAID colitis.31 A similar “allergic” drug reaction with small intestine eosinophilia has been reported with other drugs such as enalapril,32 trimethoprim-sulfonamide,33 and gemfibrozil.34

Eosinophils produce mediators that contribute to tissue damage (major basic protein, eosinophil peroxidase, platelet activating factor, leukotrienes, and eosinophil-derived neu-rotoxin). In 1 of our cases, eosinophilia was associated with villous atrophy. Patchy villous atrophy has been reported pre-viously in patients taking NSAIDs such as mefenamic acid35 or sulindac.36 In both reports, the abnormalities disappeared with cessation of the drug and reappeared at rechallenge with the drug. Intraepithelial lymphocytosis was not present.

Interesting was the presence of IFP in 1 case and of occasional changes reminiscent of incompletely developed

Dow

nloaded from https://academ

ic.oup.com/ajcp/article/130/4/518/1765278 by guest on 16 Septem

ber 2022

Am J Clin Pathol 2008;130:518-525 525525 DOI: 10.1309/7DDT5TDVB5C6BNHV 525

© American Society for Clinical Pathology

Anatomic Pathology / original article

18. Muñiz-Grijalvo O, Reina-Campos F, Borderas F. Could a fibroid polyp be a manifestation of enteropathy induced by nonsteroidal anti-inflammatory drugs? Am J Gastroenterol. 1997;92:170-171.

19. Scott N. NSAID-associated submucosal fibrous nodules of the small intestine. Histopathology. 2007;51:408-410.

20. Santolaria S, Cabezali R, Ortego J, et al. Diaphragm disease of the small bowel: a case without apparent nonsteroidal anti-inflammatory drug use. J Clin Gastroenterol. 2001;32:344-346.

21. Pera M, Heppell J, Poulsom R, et al. Ulcer associated cell lineage glands expressing trefoil peptide genes are induced by chronic ulceration in ileal pouch mucosa. Gut. 2001;48:792-796.

22. Fernando SS, Mcgovern VJ. Neuromuscular and vascular hamartoma of small bowel. Gut. 1982;23:1008-1012.

23. Noffsinger A, Fenoglio-Preiser C, Maru D, et al. Gastrointestinal Diseases. Washington, DC: ARP Press; 2007. Atlas of Nontumor Pathology; First series, Fascicle 5.

24. Whittle BJR. Mechanisms underlying intestinal injury induced by anti-inflammatory COX inhibitors. Eur J Pharmacol. 2004;500:427-439.

25. Maiden L, Thjodleifsson B, Seigal A, et al. Long-term effects of nonsteroidal anti-inflammatory drugs and cyclooxygenase-2 selective agents on the small bowel: a cross-sectional capsule enteroscopy study. Clin Gastroenterol Hepatol. 2007;5:1040-1045.

26. Kwo PY, Tremaine JM. Nonsteroidal anti-inflammatory drug–induced enteropathy: case discussion and review of the literature. Mayo Clin Proc. 1995;70:55-61.

27. Monihan JM, Hensley SD, Sobin LH. Nonsteroidal anti-inflammatory drug–induced diaphragm disease arising in a bypassed ileal segment. Am J Gastroenterol. 1994;89:610-612.

28. Freeman HJ. Nonceliac diaphragm disease of the duodenum. Can J Gastroenterol. 2000;14:453-455.

29. Onwundike M, Sundaresan M, Melville D, et al. Diaphragm disease of the small bowel: a case report and literature review. Dig Surg. 2002;19:410-413.

30. Bridges AJ, Marshall JB, Diaz-Arias AA. Acute eosinophilic colitis and hypersensitivity reaction associated with naproxen therapy. Am J Med. 1990;89:526-527.

31. Rothenberg ME. Eosinophilic gastrointestinal disorders (EGID). J Allergy Clin Immunol. 2004;113:11-28.

32. Barak N, Hart J, Sitrin MD. Enalapril-induced eosinophilic gastroenteritis. J Clin Gastroenterol. 2001;33:157-158.

33. Wienard B, Sanner B, Liersch M. Eosinophilic gastroenteritis as an allergic reaction to a trimethoprim-sulfonamide preparation [in German]. Dtsch Med Wochenschr. 1991;116:371-374.

34. Lee JY, Medellin MV, Tumpkin C. Allergic reaction to gemfibrozil manifesting as eosinophilic gastroenteritis. South Med J. 2000;93:807-808.

35. Isaacs PE, Sladen GE, Filipe I. Mefenamic acid enteropathy. J Clin Pathol. 1987;40:1221-1227.

36. Freeman HJ. Sulindac-associated small bowel lesion. J Clin Gastroenterol. 1986;8:569-571.

37. Assarian GS, Sundareson A. Inflammatory fibroid polyp of the ileum. Hum Pathol. 1985;16:311-312.

38. Kyriakos M, Condon SC. Enteritis cystica profunda. Am J Clin Pathol. 1978;69:77-85.

39. Alexis J, Lubin J, Wallack M, Enteritis cystica profunda in a patient with Crohn’s disease. Arch Pathol Lab Med. 1989;113:947-949.

40. Sheperd NA, Jass JR. Neuromuscular and vascular hamartoma of the small intestine: is it Crohn’s disease? Gut. 1987;28:1663-1668.

Address reprint requests to Dr De Petris: Dept of Pathology, Mayo Clinic Arizona, 13400 E Shea Blvd, Scottsdale 85259, AZ.

The clinical aspects of 5 of the cases included in this study were reported previously (Kelly ME, McMahon LE, Jaroszewski DE, et al. Small bowel diaphragm disease: seven surgical cases. Arch Surg. 2005;140:1162-1166; and Yousfi MM, De Petris G, Leighton JA, et al. Diaphragm disease after use of nonsteroidal anti-inflammatory agents: first report of diagnosis with capsule endoscopy. J Clin Gastroenterol. 2004;38:686-691).

References 1. Lee FD. Drug related pathological lesions of the intestinal

tract. Histopathology. 1994;23:303-308. 2. Adebayo D, Bjarnason I. Is non-steroidal anti-inflammatory

drug (NSAID) enteropathy clinically more important than NSAID gastropathy? Postgrad Med J. 2006;82:186-191.

3. Sturges HF, Krone CL. Ulceration and stricture of the jejunum in a patient on long-term indomethacin therapy. Am J Gastroenterol. 1973;59:162-169.

4. Saverymuttu SH, Thomas A, Grundy A, et al. Ileal stricturing after long-term indomethacin treatment. Postgrad Med J. 1986;62:967-968.

5. Levi S, de Lacey G, Price AB, et al. “Diaphragm-like” strictures of the small bowel in patients treated with non-steroidal anti-inflammatory drugs. Br J Radiol. 1990;63:186-189.

6. Gaede PH, Helmsøe-Zinck L, Brynskov J. Diaphragm-like strictures of the small intestine after treatment with non-steroidal anti-inflammatory drugs [in Danish]. Ugeskr Laeger. 1993;155:2409-2411.

7. Blinder GH, Hautekeete ML, Holvoet JP, et al. Duodenal diaphragm–like stricture induced by acetylsalicylic acid. Dig Dis Sci. 1994;39:1365-1369.

8. Speed CA, Bramble MG, Corbett WA, et al. Non-steroidal anti-inflammatory induced diaphragm disease of small intestine: complexities of diagnosis and management. J Rheumatol. 1994;33:778-780.

9. Abrahamian GA, Polhamus CD, Muskat P, et al. Diaphragm-like strictures of the ileum associated with NSAID: a rare complication. South Med J. 1998;91:395-397.

10. Shumaker DA, Bladen K, Katon RM. NSAID-induced small bowel diaphragms and strictures diagnosed with intraoperative endoscopy. Can J Gastroenterol. 2001;15:619-623.

11. Mir O, Dhote R, Scavennec R. Cyclooxygenase 2 selective inhibitor induced bowel stricture: a case report. Gut. 2004;53:154.

12. McGonigal A, Moffat DF, Lindop GB, et al. Nonsteroidal anti-inflammatory drug associated diaphragm disease. Postgrad Med J. 1997;73:505-506.

13. Hemmings CT. Neuromuscular and vascular hamartoma arising in a Meckel’s diverticulum [letter]. Pathology. 2006:38:173-174.

14. Zhao B, Sanati S, Eltorky M. Diaphragm disease: complete small bowel obstruction after long-term nonsteroidal anti-inflammatory drugs use. Ann Diagn Pathol. 2005;9:169-173.

15. Lang J, Price AB, Levi AJ, et al. Diaphragm disease: pathology of disease of the small intestine induced by non-steroidal anti-inflammatory drugs. J Clin Pathol. 1988;41:516-526.

16. Peacey SR, Walls WD. Steatorrhoea and sub-total villous atrophy complicating mefenamic acid therapy. Br J Clin Pract. 1992;46:211.

17. Cortina G, Wren S, Armstrong B, et al. Clinical and pathologic overlap in nonsteroidal anti-inflammatory drug–related small bowel diaphragm disease and the neuromuscular and vascular hamartoma of the small bowel. Am J Surg Pathol. 1999;23:1414-1417.

Dow

nloaded from https://academ

ic.oup.com/ajcp/article/130/4/518/1765278 by guest on 16 Septem

ber 2022