Printer Friendly

The Histopathology of 103 Consecutive Colonoscopy Biopsies From 82 Symptomatic Patients With Acquired Immunodeficiency Syndrome: Original and Look-Back Diagnoses.

Gastrointestinal disease remains a major worldwide problem of the human immunodeficiency virus (HIV) epidemic.[1-3] Both the clinical and pathologic evaluation can be quite challenging.[4,5] The physician is faced with a broad range of infectious and noninfectious processes in the setting of an immunologically compromised host. Symptoms such as diarrhea and pain may be due to a range of pathogens, which can be multiple, either at the same or at different levels of the patient's intestinal tract. General surgical pathologists are more accustomed to a relatively narrow range of differential diagnoses for diarrhea, headed by inflammatory bowel disease (IBD). Diagnosing intestinal infections is more within the domain of the microbiology laboratory, where stool is analyzed, than in the purview of the surgical pathologist.[6-10] The HIV epidemic has dramatically altered the field of gastroenterology. Infections have become paramount and include parasites, fungi, bacteria, and viruses; some pathogens, such as microsporidia, have only recently been identified.[5,11-14] The endoscopic intestinal biopsy specimen has become critical in HIV disease because it allows the pathologist to identify infectious processes not routinely detected in the stool.

Additionally, HIV itself may also be a direct enteric Pathogen.[15-19] Considering the sheer size of the lymphoid component of the bowel, its abundance of activated CD4 lymphocyte and mononuclear phagocyte targets, and the likely recruitment of HIV-infected cells into areas of intestinal pathology, it is not surprising to find evidence of HIV-specific RNA or antigen expression in intestinal biopsies. Among the alimentary tract lesions for which HIV may play a direct pathogenic role are idiopathic ulcers of the esophagus and rectum.[20-22]

The diagnosis of IBD in an immunocompromised host presents a special challenge to the clinician and pathologist alike. It is not clear whether the symptoms, diagnostic histopathologic features, and response to therapy are the same for HIV-positive and HIV-negative individuals with IBD.[22-25] Finally, diarrhea is a side effect for some nucleoside reverse-transcriptase inhibitors and for virtually all protease inhibitors. To date, no histopathologic correlation has been described for antiretroviral drug-associated diarrhea.

This study of 103 consecutive colonoscopy specimens from 82 different symptomatic HIV-infected patients was undertaken to compare the primary diagnoses rendered by a general surgical pathologist with those assigned by a pathologist specializing in this particular diagnostic area and to identify which diagnoses were overlooked. These cases are from the pre-highly active antiretroviral therapy era, between 1991 and mid-1994.


Colonoscopy with biopsy was performed on 82 patients a total of 103 times for large bowel-associated symptoms, including irritable bowel, frequent small volume-type diarrhea, tenesmus, and/or bright red blood per rectum. Routine ova, parasite, and bacterial stool studies for these patients had been negative or the patient had failed to respond to treatment for an identified stool pathogen. The slides examined for this study were all recuts prepared from the original paraffin blocks; no original slides were used. In addition to hematoxylin-eosin (H&E) staining, immunohistochemistry (IHC) for cytomegalovirus (CMV) was performed (Dako Corporation, Carpinteria, Calif) on 18 of 22 cases originally diagnosed as having CMV infection. Two other cases that were suspicious for CMV in H&E-stained sections also underwent CMV IHC.

Cases suspicious for bacterial infection were also stained with Giemsa and Brown-Brenn Gram stains.


Original Histopathologic Diagnoses The diagnoses entered on the surgical pathology reports at the time of the initial screening are listed in the Table. There was a high proportion of negative or nonspecific diagnoses (n = 70, 68%), including no specific histologic abnormality; no pathologic diagnosis; and focal acute, chronic, or acute and chronic changes. The most common specific diagnosis was CMV (n = 22, 21%), 1 case of which was associated with Mycobacterium avium complex coinfection. The next most common diagnosis was cryptosporidiosis (n = 5, 5%). Of the 22 CMV cases, only 5 were considered to have a heavy viral burden, while the rest were scored as relatively mild infections. There were also 2 cases each of Kaposi sarcoma and adenomatous polyps, and individual cases of squamous cell carcinoma of the anorectum and spirochetosis. No other diagnoses were rendered (eg, adenovirus colitis, bacterial enteritis, or IBD).
Diagnoses for 103 Consecutive Biopsy Specimens
From 82 Patients(*)

 Initial Review
 Diagnoses Diagnoses

Negative or nonspecific 70 55
IBD-type pathology 0 9
CMV alone 21 11
 + MAC 1 0
 + Cryptosporidium 0 1
 + Adenovirus 0 2
Cryptosporidium 5 4
Adenovirus alone 0 7
 + CMV + MAC 0 1
 + AEEC 0 1
 + Spirochetosis 0 1
Spirochetosis 1 3
AEEC 0 2
Squamous cell carcinoma 1 1
Adenomatous polyp 2 2
Kaposi sarcoma 2 3

(*) IBD indicates inflammatory bowel disease; CMV, cytomegalovirus;
MAC, Mycobacterium avium complex; and AEEC, attaching and attaching
Escherichia coli.

Look-Back Histopathologic Diagnoses

Since many of the original H&E-stained slides from which the initial diagnoses were rendered were not available, the histopathologic review was carried out on single levels of H&E-stained recuts prepared from the original paraffin blocks (Table). Up to 10 pieces of tissue were present in a section. The number of cases without specific diagnoses was still very high (n = 64, 62%), although they were not always the same biopsies as in the original evaluation. Sixteen of these specimens showed prominent mucosa-associated lymphoid hyperplasia, often with secondary follicles.

Eleven (11%) lone CMV infections (instead of 21) were diagnosed in the recuts (Figure, A), only 9 of which had been originally diagnosed as CMV colitis. Two of these 9 cases were the only cases in which CMV inclusions were not detected in the H&E-stained sections; only single positive-staining cell nuclei were identified by CMV IHC in each section. The 2 newly diagnosed CMV-positive biopsies originally had been without specific diagnoses, and rare CMV-infected cells were found on the H&E-stained sections and confirmed in the IHC preparation. The cases that had slides that were originally read as positive, but that were negative on review, were all among the cases that were initially considered to have a low CMV burden, and thus could have been negative because of sampling. Two of the original CMV-positive cases were found on review to be coinfected with adenovirus. One CMV-infected specimen was also found to be coinfected with Cryptosporidium (Figure, B).


On review, a total of 12 specimens were found to at least be infected with adenovirus (Figure, C and D). Seven biopsies with only adenovirus had been originally given negative or nonspecific diagnoses. Two more adenovirus infections were found separately to have spirochetosis (Figure, E) and attaching and effacing bacterial infection. In addition to the 2 original CMV-positive cases that were found to also have adenovirus on review, an original Mycobacterium avium complex (Figure, F) plus CMV coinfection was also found to have adenovirus; this was the only case with 3 coinfections. Ten of the adenovirus specimens were colonic biopsies, while 2 were rectal biopsies.

The original case of spirochetosis could not be confirmed, but 4 other spirochete-positive biopsies were found, 1 with adenovirus coinfection (Figure, E). The review yielded 2 cases of enteropathogenic bacterial infection (Figure, G); 1 of these cases was also infected with adenovirus. There were 3 cases of Kaposi sarcoma, 1 of which was new. Diagnoses for the 2 cases of adenomatous polyps and 1 case of squamous cell carcinoma of the anorectum remained the same.

There were 5 cases diagnosed originally with colonic cryptosporidiosis, and 4 were found on review; 3 of the review diagnoses were the same as the originals. One new case had initially been given a negative diagnosis, and 1 case that was originally considered positive could not be confirmed.

Nine biopsies had no pathogen identified during either of the 2 examinations, but appeared to be different from all of the other specimens. These sections had features suggestive of IBD with ulcers, cryptitis, goblet cell hyperplasia, increased crypt mitoses, and deformed (branching, elongated, and dilated) crypts.

The 103 biopsies were from 82 different patients; 7 patients had multiple colonoscopies with biopsies: (1) the 1 patient who had squamous cell carcinoma of the anorectum also had 2 negative biopsies; (2) 1 patient had Cryptosporidium identified in the biopsies from only 1 of 3 specimens; (3) 1 patient had adenovirus in 2 specimens and only a single CMV inclusion identified by IHC in a third biopsy; (4) 5 patients had 2 negative biopsies each; (5) 1 patient had 3 specimens, only 1 of which showed a single CMV intranuclear inclusion by IHC; (6) 1 patient had only adenovirus in 1 biopsy specimen and CMV in another; and (7) 1 patient had 8 negative biopsies.


Several observations stand out from this study of 103 consecutive colonic biopsies from 82 HIV-infected patients undergoing colonoscopy between late 1991 and mid-1994, prior to the advent of highly active antiretroviral therapy.

1. There was a high percentage of specimens that even after expert review still had no specific diagnosis. This could be attributed, at least in part, to the exclusion of patients with positive stool specimens. It could also mean that the patient had a small bowel and not a large bowel process. Stool studies should have detected most pathogens that involved the large bowel, such as Clostridium difficile toxin, Campylobacter, Shigella, Yersinia, Cryptosporidium, and Mycobacterium avium complex.[4,5,26]

2. The diagnosis of adenovirus proctocolitis requires experience to recognize its classic amphophilic intranuclear inclusion.[27,28] In this study, it appears that the intranuclear inclusions formed by adenovirus were not originally confused with those caused by CMV, since none of the cases initially diagnosed as CMV turned out to have only adenovirus on review.

3. Bacterial colitis is also an overlooked histologic diagnosis.[29-31] When present, the attaching and effacing bacteria usually involve, to some degree, all of the pieces of tissue in the biopsy specimen. Giemsa or Gram stains greatly enhance visualization of the bacteria. It is not routine, in most laboratories, to culture stool for pathogenic types of Escherichia coli other than O157H7.

4. Spirochetosis can be a difficult histologic diagnosis that can be confused with a prominent brush border or totally overlooked.[32] Giemsa or silver stains dramatically enhance the spirochetes.

5. The tissue diagnosis of cryptosporidiosis is relatively uncommon when stool studies are performed first.[33]

6. Cytomegalovirus infections may be overdiagnosed in the bowel, since instead of the classic targetoid type of intranuclear inclusion they often produce a smudgy type of intranuclear inclusion, which can be confused with the amphophilic nucleus of reactive macrophages[34] and ganglion cells.

There are also several questions that this study poses.

1. How much effect did using recuts as opposed to the original slides have on the results? Focality of infection could explain our failure to detect a number of infections, especially CMV, but also perhaps spirochetosis and cryptosporidiosis. What is the significance of a "few" scattered positive CMV-infected cells, usually endothelial cells, in an otherwise "normal-appearing" specimen?

2. What is the difference between ulcerative and non-ulcerative CMV colitis, as it often does not appear to be simply a matter of the burden of CMV-infected cells?

3. Did any of these patients actually have IBD, and therefore might have benefited from standard steroid and salicylate therapy?[22-24] There is some evidence for the efficacy of prednisone for HIV-positive patients with idiopathic colitis and ulceration.

4. What is the significance of adenovirus detected by histologic examination as opposed to stool studies?

Cytomegalovirus is a ubiquitous pathogen in HIV, often targeting the gastrointestinal tract.[25,35] Sampling is always a consideration when the CMV burden is low. We have seen enteric biopsy specimens cut at 2 or 3 levels that still have only rare endothelial cells containing diagnostic CMV intranuclear inclusions. These specimens typically have little if any epithelial damage and are usually random biopsies, as the colon had appeared normal. We have wondered whether this could represent a low-grade systemic CMV endothelialitis and not simply a colonic infection. Should such a patient be treated for CMV without further searching for other intestinal pathogens (eg, upper gastrointestinal endoscopy)? We have seen patients treated for CMV colitis under these circumstances who have become asymptomatic. Endothelial cells appear to be the primary target for CMV (especially those located immediately beneath the basement membrane), and as the CMV burden increases, in addition to seeing more and more infected endothelial cells, infected macrophages, smooth muscle cells, and crypt (never surface) epithelium appear.[35] It has been our impression that although IHC and/ or DNA probes for CMV can give a dramatic picture that more likely reflects the true infected cell burden, they are rarely necessary to render a CMV diagnosis.

Adenovirus proctocolitis is an almost totally overlooked histologic diagnosis.[27,28,36,37] In our experience with more than 25 cases, adenovirus has been accompanied by CMV infection about one third of the time. In this study, 5 of 12 cases of adenovirus colitis had coinfections: 2 with CMV alone and 1 each with CMV plus Mycobacterium avium complex, CMV plus enteropathogenic bacterial infection, and CMV plus spirochetosis. The presence of necrotic epithelial cells and cellular debris, along with neutrophil and eosinophil infiltrate, in association with adenovirus-infected cells supports the pathogenic role of adenovirus.[37] Since there is no proven therapy for adenovirus infection, this cannot be confirmed. Its underappreciation may help explain why some cases diagnosed as CMV colitis alone do not completely respond to specific CMV therapy. All of our patients with adenovirus proctocolitis had diarrhea. Adenovirus infection should not be confused with CMV in biopsies. In the colon, adenovirus infects surface colonocytes, particularly goblet: cells (the infected cells are often "upside down" with their mucous contents below the nucleus), and forms an amphophilic intranuclear inclusion that completely fills the nucleus and is often irregular in shape. The irregularity of the inclusion is due to the formation of crystalline arrays of virions in the nucleus. The intranuclear inclusion is not accompanied by a cytoplasmic inclusion, as is characteristic of CMV. Immunohistochemical staining can be used to detect adenovirus-infected cells in paraffin sections.

Colitis due to bacteria such as E coli are considered to be within the domain of the microbiology laboratory and not that of the histopathologist.[29-31] Enteropathogenic and enteroaggregative E coli are major causes of traveler's diarrhea and diarrhea in developing countries, especially in children.[2] In patients with HIV/acquired immunodeficiency syndrome, we have regularly made this diagnosis histologically and confirmed, by TEM, the attaching of the bacteria to the brush border and its effacing of the microvilli. Infection is associated with the development of defects in the layer of surface colonocytes (does not affect crypt epithelium) and the presence of sloughing cells covered with bacteria. In addition to the histopathology, response to antibiotics such as ciprofloxacin hydrochloride supports the pathologic role of attaching and effacing E coli in HIV/acquired immunodeficiency syndrome.

In summary, colonoscopy with biopsy reveals a range of microscopic diagnoses to the experienced histopathologist that may be overlooked by a less experienced surgical pathologist. Some of these diagnoses, such as bacterial colitis and spirochetosis, are treatable and are therefore especially important to diagnose.


[1.] Conlon CP, Pinching AJ, Perera CU, et al. HIV-related enteropathy in Zambia: a clinical, microbiological, and histological study. Am J Trop Med Hyg. 1990; 42:83-88.

[2.] Thea DM, St Louis ME, Atido U, et al. A prospective study of diarrhea and HIV-1 infection among 429 Zairian infants. N Engl J Med. 1993;329:1696-1702.

[3.] Gradon JD, Timpone JG, Schnittman SM. Emergence of unusual opportunistic pathogens in AIDS: a review. Clin Infect Dis. 1992;15:134-157.

[4.] Mayer HB, Wanke CA. Diagnostic strategies in HIV-infected patients with diarrhea. AIDS. 1994;8:1639-1648.

[5.] Lew EA, Poles MA, Dieterich DT. Diarrheal diseases associated with HIV infection. Gastroenterol Clin North Am. 1997;26:259-290.

[6.] Schmidt W, Schneider T, Heise W, et al. Stool viruses, coinfections, and diarrhea in HIV-infected patients. J Acquir Immun Defic Syndr Hum Retrovirol. 1996;13:33-38.

[7.] Durepaire N, Ranger-Rogez S, Gandji JA, et al. Enteric prevalence of adenovirus in human immunodeficiency virus seropositive patients. J Med Virol. 1995;45:46-60.

[8.] Hierholzer JC, Wigand R, Anderson LJ, et al. Adenovirus from patients with AIDS: a plethora of serotypes and a description of five new serotypes of subgenus D (types 43-47). J Infect Dis. 1988;158:804-812.

[9.] Khoo SH, Bailey AS, de Jong JC, et al. Adenovirus infections in human immunodeficiency virus-positive patients: clinical features and molecular epidemiology. J Infect Dis. 1995;172:629-637.

[10.] Grohmann GS, Glass RI, Pereira HG, et al. Enteric virus and diarrhea in HIV-infected patients. N Engl J Med. 1993;329:14-20.

[11.] Orenstein JM, Chiang J, Steinberg W, et al. Intestinal microsporidiosis as a cause of diarrhea in human immunodeficiency virus-infected patients: a report of 20 cases. Hum Pathol. 1990;21:475-481.

[12.] Orenstein JM, Tenner M, Cali A, et al. A microsporidian previously undescribed in humans, infecting enterocytes and macrophages and associated with diarrhea in an AIDS patient. Hum Pathol. 1992;23:722-728.

[13.] Cali A, Kotler DP, Orenstein JM. Septata intestinalis, N.G.,N.SP, an intestinal microsporidian associated with chronic diarrhea and dissemination in AIDS patients. J Eukaryot Microbiol. 1993;40:101-112.

[14.] Molina J, Sarfati C, Beauvais B, et al. Intestinal microsporidiosis in human immunodeficiency virus-infected patients with chronic unexplained diarrhea: prevalence and clinical and biologic features. J Infect Dis. 1992;167:217-221.

[15.] Kotler DP, Reka S, Clayton F. Intestinal mucosal inflammation associated with human immunodeficiency virus infection. Dig Dis Sci. 1993;38:1119-1127.

[16.] Fox CH, Kotler D, Tierney A, et al. Detection of HIV-1 RNA in the lamina propria of patients with AIDS and gastrointestinal disease. J Infect Dis. 1989;159: 467-471.

[17.] Schmidt W, Fackler OT, Schafer M, et al. Similar proviral load but increased HIV-1 p24 in the intestinal mucosa compared to the peripheral blood in HIV-infected patients. Ann N Y Acad Sci. 1998;859:276-279.

[18.] Kotler DP, Reka S, Borcich A, et al. Detection, localization, and quantitation of HIV-associated antigens in intestinal biopsies from patients with HIV. Am J Pathol. 1991;139:823-830.

[19.] Jarry A, Cortez A, Rene E, et al. Infected cells and immune cells in the gastrointestinal tract of AIDS patients: an immunohistochemical study of 127 cases. Histopathology. 1990;16:133-140.

[20.] Kotler DP, Reka S, Orenstein JM, et al. Chronic idiopathic esophageal ulceration in the acquired immunodeficiency syndrome. J Clin Gastroenterol. 1992; 15:284-290.

[21.] Wilcox CM, Schwartz DA. Idiopathic anorectal ulceration in patients with human immunodeficiency virus infection. Am J Gastroenterol. 1994;89:599-604.

[22.] Bach MC, Howell DA, Valenti AJ, et al. Aphthous ulceration of the gastrointestinal tract in patients with the acquired immunodeficiency syndrome (AIDS). Ann Intern Med. 1990;112:465-467.

[23.] Bernstein BB, Gelb A, Tabanda-Lichauco R. Crohn's ileitis in a patient with long-standing HIV infection. Am J Gastroenterol. 1994;89:937-939.

[24.] Silver S, Wahl SM, Orkin BA, et al. Changes in circulating levels of HIV, CD4, and tissue expression of HIV in a patient with recent onset ulcerative colitis treated by surgery: a case report. J Hum Virol. 1999;2:52-57.

[25.] Roskell DE, Hyde GM, Campbell AP, et al. HIV associated cytomegalovirus colitis as a mimic of inflammatory bowel disease. Gut. 1995;37:148-150.

[26.] Nelson MR, Shanson DC, Hawkins DA, et al. Salmonella, Campylobacter and Shigella in HIV-seropositive patients. AIDS. 1992;6:1495-1498.

[27.] Janoff EN, Orenstein JM, Manischewitz JF, et al. Adenovirus colitis in the acquired immunodeficiency syndrome. Gastroenterology. 1991 ;100:976-979.

[28.] Maddox A, Francis N, Moss J, Blanshard C, Gazzard B. Adenovirus infection of the large bowel in HIV positive patients. J Clin Pathol. 1992;45:684-688.

[29.] Kotler DP, Orenstein JM. Chronic diarrhea and malabsorption associated with enteropathogenic bacterial infection in a patient with AIDS. Ann Intern Med. 1993;119:127-128.

[30.] Kotler DP, Giang TT, Thiim M, et al. Chronic bacterial enteropathy in patients with AIDS. J Infect Dis. 1995;171:552-558.

[31.] Orenstein JM, Kotler DP. Diarrheogenic bacterial enteritis in acquired immunodeficiency syndrome: a light and electron microscopy study of 52 cases. Hum Pathol. 1995;26:481-492.

[32.] Guccion JG, Benator DA, Zeller J, et al. Intestinal spirochetosis and acquired immunodeficiency syndrome: ultrastructural studies of two cases. J Ultrastruct Pathol. 1995;19:15-22.

[33.] Connolly GM, Dryden MS, Shanson DC, et al. Cryptosporidial diarrhea in AIDS and its treatment. Gut. 1988;29:593-597.

[34.] Schwartz DA, Wilcox CM. Atypical cytomegalovirus inclusions in gastrointestinal biopsy specimens from patients with the acquired immunodeficiency syndrome: diagnostic role of in situ nucleic acid hybridization. Hum Pathol. 1992: 23:1019-1026.

[35.] Goodgame RW. Gastrointestinal cytomegalovirus disease. Ann Intern Med. 1993;119:924-935.

[36.] Blanshard C, Francis N, Gazzard BG. Investigation of chronic diarrhoea in acquired immunodeficiency syndrome: a prospective study of 155 patients. Gut. 1996;39:824-832.

[37.] Strickler JG, Singleton TP, Copenhaver CM, et al. Adenovirus in the gastrointestinal tracts of immunosuppressed patients. Am J Clin Pathol. 1992;97:555-558.

Accepted for publication March 15, 2001.

From the Department of Pathology, George Washington University Medical Center, Washington, DC (Dr Orenstein); and the Department of Medicine, Division of Gastroenterology, New York University School of Medicine, New York, NY (Dr Dieterich).

Reprints: Jan Marc Orenstein, MD, PhD, Ross 502, Department of Pathology, George Washington University Medical Center, 2300 Eye St NW, Washington, DC 20037 (e-mail:
COPYRIGHT 2001 College of American Pathologists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Orenstein, Jan Marc; Dieterich, Douglas T.
Publication:Archives of Pathology & Laboratory Medicine
Article Type:Statistical Data Included
Geographic Code:1USA
Date:Aug 1, 2001
Previous Article:CD30 Expression in Follicular Lymphoma.
Next Article:The Small Blue Cell Dilemma Associated With Tamoxifen Therapy.

Related Articles
Targeting hospital screening for HIV.
Video Technology Detects Polyps.
Generalized Tuberculosis in the Acquired Immune Deficiency Syndrome.
Enterocolonic Mycobacterium avium-intracellulare. (Images in Pathology).

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