Transmural Granulomatous Inflammation with Perforation of Small Intestine Due to Disseminated Infection with Histoplasma capsulatum
Stephen D. Allen
Indiana University School of Medicine
A 43-year-old man was transferred from an outside hospital for further evaluation and
management of abdominal pain. The patient had abdominal pain for approximately 10 days, and the pain had
been worsening. He noted difficulty in swallowing and had a 30 pound unintentional weight loss. He also
stated he had gotten surgery for hemorrhoids and "intestine hanging out of my anus" approximately three
weeks ago. The patient noted tenesmus with significant straining to produce a bowel movement. Although
he also gave a history of diarrhea in recent weeks, his last bowel movement was normal and occurred three
days ago. He had occasional nausea, but no emesis. He denied any chills but had been having high grade
fevers. Otherwise, he denied any shortness of breath or chest pain. His abdominal pain was cramping and
worse in his left lower quadrant and suprapubic regions. His past medical history was significant for
AIDS with his last CD4 count two weeks ago of 10/µL. In addition, the patient had been incarcerated in
northern Indiana and was a recovering narcotic and alcohol addict. He had immigrated to the US from
Puerto Rico in 1998. The patient denied a history of opportunistic infection and tuberculosis. He
denied HAART, but outside records indicated he was admitted on a regimen of emtricitabine, kaletra,
lamivudine, levoquin, and bactrim. On physical examination, his initial temperature in the emergency
room was 105.5° F. He was a cachectic Latino man with a mildly toxic appearance. HEENT examination was
unremarkable. Chest examination revealed rales bilaterally. His abdomen was soft, mildly distended, and
moderately tender diffusely, greatest in the left lower quadrant and suprapubic regions. There was a 2
cm, mobile mass in his scrotum. Laboratory data was notable for a white blood count of 1900/µL with 13%
bandemia, hemoglobin 9.6, platelets 85, and AST 223 (H). Blood and urine cultures were pending. A chest
x-ray demonstrated a large amount of free intraperitoneal air and a suspected small left pleural
effusion. A head CT was notable for mild generalized cerebral and cerebellar volume loss with a few
scattered hypodense foci within the central pons. A CT of the abdomen and pelvis, with oral and
intravenous contrast, demonstrated free intraperitoneal air over the upper abdomen with periportal,
retroperitoneal, and mesenteric lymphadenopathy as well as pleural effusion with bibasilar atelectasis,
left greater than right. The patient underwent urgent exploratory laparotomy for suspected small bowel
One scanned glass slide from the suspected perforation site in the jejunum (hematoxylin and eosin
[H&E]) and images (H&E and Gomori methenamine silver [GMS] stains as part of the handout [Figures
Shows a low power view of the inflammatory infiltrate within the mucosa and lamina propria of the small intestine specimen (H&E).
Is a low power view adjacent to that shown in Figure 1 revealing transmural inflammation (H&E). Adjacent to this area was a prominent ulcer which extended from the mucosa to the serosal surface.
Shows diffuse granulomatous inflammation in the submucosa (H&E).
Shows nonnecrotizing granulomas within the muscle layers (H&E).
Demonstrates a prominent necrotizing granuloma (H&E).
High magnification reveals clusters of small yeasts within macrophages (H&E x500).
Higher magnification reveals clusters of small yeasts within macrophages. Note the retraction of the fungus cytoplasm from the cell wall ("halo") giving the false impression of a capsule (H&E x1000). The halo effect is not seen using Gomori methenamine silver (GMS) stain as demonstrated in Figures 8 and 9..
GMS stain (x500) revealing myriads and aggregates of dark stained intracellular and extracellular budding yeasts.
GMS stain (x1000) high power view revealing details of small yeasts, 2-to-4 µm in diameter, having single buds with narrow necks.
Histopathologic examination revealed a combination of microscopic findings. These included a
prominent ulcer, diffuse lymphohistiocytic infiltrates, non-necrotizing granulomas, and necrotizing
granulomas. The inflammatory process extended from the ulcerated mucosal surface to the serosal surface.
In addition to lymphocytes and macrophages, there were many granulocytes and plasma cells.
Multinucleated giant cells were present but not in large number. H&E stained sections revealed
abundant small yeasts clustered within macrophages. Extracellular yeasts were also present in large
number. GMS stain confirmed the presence of myriads of small yeasts, 2-to-4 µm in diameter, having
single buds with narrow necks.
Transmural Granulomatous Inflammation with Perforation of Small Intestine Due to Disseminated Infection with
This 43-year-old Spanish-speaking-only male underwent a small bowel resection on 11/4/08. This was
for an acute abdomen, secondary to perforation approximately 150 cm distal to the ligament of Treitz
(thought to be in the mid-jejunum). Three diagnostic results became available within one day of each
other. First, the small bowel specimen revealed transmural granulomatous inflammation with perforation
due to Histoplasma capsulatum. Second, a test for urinary Histoplasma
antigen (MiraVista Diagnostics, Indianapolis, IN) was positive (> 39 units). Third, a fungal gel
immunodiffusion test for antibody against Histoplasma antigen was M band positive (H band not detected).
After the diagnosis was made, the patient received amphotericin B (Abelcet) intravenously for 13/14
days; he did not tolerate it well. He subsequently was to be treated with itraconazole (200 mg po BID)
for the remainder of his life. Following the surgery, however, the hospital course was complicated by a
bowel obstruction. This involved the duodenum due to lymphadenopathy (extrinsic). It was elected to
place a gastrostomy-jejunostomy (GJ) tube for enteral nutrition, and the patient was discharged back to
prison. At the time, he refused to take his HIV medication due to nausea. He returned to surgery clinic
on 12/9/08. He had continued to lose weight, despite being on approximately 1800 calories of tube feeds
through his J-tube. On exam, the GJ tube was in good position, his incision was well-healed, and he had
no abdominal tenderness. The patient subsequently removed his own gastric tube and had since noted pain
and drainage from his external abdominal wall. Radiographic studies (upper GI) during his last
appointment (1/23/09) revealed marked improvement with no evidence of duodenal or other obstruction.
More than a century ago, Samuel T Darling, a U.S. Army pathologist assigned to Panama, first
described a disease entity now recognized as disseminated histoplasmosis
Although the gross
lesions resembled disseminated tuberculosis, microscopic examination showed intracellular organisms
Darling first thought were Leishmania; however, they lacked kinetoplasts.
Darling observed the organisms in histiocytes, thought they resembled plasmodium-like protozoans and that
the clear spaces around the organisms were capsules. Thus, he proposed the name Histoplasma capsulata
. Although further
investigation showed this was not a protozoan, it took many years to demonstrate that H. capsulatum is a thermal dimorphic fungus that occurs as a filamentous mold in
soil or on an agar medium incubated at 25 °-30 ° C; that soil, especially soil fertilized by bird
droppings or bat guano, is the environmental reservoir; and that histoplasmosis is primarily a pulmonary
infection acquired through inhalation of microconidia
The yeast form is found in tissues
from infected warm-blooded animals and on enriched media (e.g., brain heart infusion agar) incubated at
35 ° -37° C in the laboratory.
In humans, cases of histoplasmosis caused by H. capsulatum var. capsulatum have been documented in ³ 60 countries on most continents (but not
. The endemic areas with highest prevalence are in the midwestern and south central
United States (e.g., the Ohio, and Mississippi River valleys) and much of
Latin America .
H. capsulatum var. duboisii, which causes "African
histoplasmosis", will not be addressed in this presentation. The remainder of the discussion is confined
to H. capsulatum var capsulatum, or simply
H. capsulatum and histoplasmosis caused by this organism.
Interestingly, ileal ulcerations were
noted at autopsy in one of Darling's first three patients
Confirmation of enteritis and colitis
in disseminated histoplasmosis was reported many years later .
At autopsy, H.
capsulatum can be demonstrated in the GI tracts of 70-90% of patients who have disseminated
During life, however, GI disease due to H. capsulatum
(GIH) is recognized in only 3-12% of patients, thus raising concern that GIH may be underdiagnosed
Microconidia of H. capsulatum may be inhaled when dispersed in the air
during various activities that disrupt soil, dirt, chicken coops, other dusts where birds have roosted,
or bat guano (e.g., in caves or attics). Once inhaled, the conidia attract neutrophils and macrophages,
causing pneumonitis. When ingested by macrophages, the conidia transform into yeasts, which multiply
unchecked in the non-immune individual. Before cellular immunity to H. capsulatum
develops, infection occurs within the lungs, and the organism disseminates hematogenously
throughout the reticuloendothelial system. About a month after infection, specific T cell immunity
develops, progression of the infection stops, and the organisms are eliminated or contained in > 99%
of individuals . After low inoculum exposure, most individuals remain asymptomatic. Only a few
develop symptoms (usually a flu-like respiratory illness). However, heavy exposure may lead to diffuse
lung involvement with more severe pulmonary illness proportional to the level of exposure. In addition,
while previously infected individuals may have milder illnesses because of pre-existing immunity, more
severe illness occurs in individuals who are immunosuppressed. Progressive dissemination of H. capsulatum with various clinical consequences occurs in immunocompromised
Reactivation of quiescent infection with progressive dissemination has been assumed to occur in some
individuals when they become immunosuppressed
though evidence for this is considered weak
(Wheat, LJ, Histoplasmosis; see recent review at www.miravistalabs.com).
Accordingly, an argument against reactivation as a common mode of pathogenesis is that the frequency of
histoplasmosis in immunosuppressed patients is very low . As reviewed elsewhere by Wheat, the
frequency of histoplasmosis is " ~ 1% of patients with AIDS , < 0.5% following
and < 0.1% after anti-TNF treatment
" When the T cell response works,
sterilizing immunity presumably leads to macrophage killing of intracellular Histoplasma
Data are lacking to support the concept of "latent"
infection in histoplasmosis. As Wheat also has pointed out, cultures of old calcified granulomas show no
growth, even though yeasts may be seen microscopically. In the presence of old calcified granulomas,
reinfection , rather than reactivation, is more likely to occur. Thus, at least in endemic areas, it
is postulated that progressive disseminated histoplasmosis in immunosuppressed individuals probably is
due to exogenous infection (Wheat, LJ, Histoplasmosis; see recent review at
GIH may be the result of mediastinal histoplasmosis through impingement on the esophagus by enlarged,
inflamed lymph nodes, or involvement of the esophagus in patients with fibrosing mediastinitis .
Factors predisposing individuals to progressive dissemination of H.
capsulatum include the following: AIDS, immunosuppressive medications, extremes of age,
idiopathic CD4 lymphocytopenia, Job's syndrome, common variable immunodeficiency, other immune disorders,
and biologic agents that interfere with the action of tumor necrosis factor µ (e.g., infliximab and
The pathologic spectrum of GIH was described by Lamps and
colleagues in an excellent review of 52 patients with disseminated disease from endemic and nonendemic
areas . Both gastrointestinal and hepatic lesions were present in 60% of their cases and about half
were immunocompromised. Ulcers (0.2 to 4.0 cm in diameter), the most common gross lesions in the GI
tract, were most frequently observed in the ileum but also found in the jejunum, colon, esophagus, and
stomach. Most of the ulcers were multiple and annular with raised borders and were associated with
hyperemia, hemorrhage, and a necrotic base. Microscopically, diffuse lymphohistiocytic infiltrates,
ulcers, lymphohystiocytic nodules involving the mucosa and submucosa, or minimal inflammatory reactions
were seen. The inflammatory cell infiltrates contained relatively large numbers of lymphocytes, plasma
cells, and macrophages as well as many neutrophils and eosinophils. Yeasts were usually seen within the
cytoplasm of macrophages but also extracellularly in severe GI infections. Well-formed granulomas were
A more recent review by Kahi and associates provides additional descriptions of GIH pathology and
clinical findings . In H&E stained sections, H. capsulatum is
likely to be seen when the yeasts are abundant (Figures 6 & 7). As in the present case, they are
small, 2-4 µm, uniform, ovoid-to-spherical, uninucleate yeasts with narrow-based buds. The yeasts are
often present in clusters within the cytoplasm of macrophages. The cytoplasm of H.
capsulatum retracts from its cell wall, leaving a "halo" or clear zone around the organism (Fig.
7). The Gomori (or Grocott) methenamine silver stain is especially useful for demonstrating H. capsulatum in tissue (Figures 8 & 9), even when only small numbers of
yeasts are present. In our experience, a periodic acid-Schiff (PAS) stain is not as reliable for
visualizing this organism.
For GIH, the differential diagnosis includes Crohn's disease, ulcerative colitis (or idiopathic
inflammatory bowel disease) and
Other microorganisms that could be confused with
H. capsulatum include the following: Candida
species, particularly C. glabrata, Cryptococcus neoformans, Blastomyces
dernatitidis, Pneumocystis jirovecii, Leishmania spp., and Trypanosoma cruzi.
Laboratory diagnosis of histoplasmosis.
In addition to histopathology, a
variety of other methods can be used to diagnose histoplasmosis. Histoplasma antigen can be detected in
the urine of over 90% of patients who have disseminated histoplasmosis . Antigen testing can also
aid in monitoring treatment. However, antigen testing has not been well studied specifically in patients
with GIH. In patients with localized pulmonary histoplasmosis, antigen is detected in urine in only
about 25% of cases. Serologic tests for antibodies to H. capsulatum are
recommended for all cases of suspected histoplasmosis. Complement fixation tests are positive (titers ³
1:8) in 90% of patients who have pulmonary histoplasmosis. Immunodiffusion tests are positive for M
bands in 76% and positive for H bands in only 23% of such patients
Serologic tests tend to
be less useful in patients with disseminated disease
Direct microscopic examination of
fresh clinical specimens (e.g., frozen sections and/or touch preparations of lymph nodes, liver biopsies,
or GI biopsies) stained with Diff-Quik, Giemsa, GMS, and calcofluor white, and peripheral blood buffy
coat smears stained with Wright –Giemsa, aid in rapid diagnosis of patients who have disseminated
infection. Direct microscopic exams are not as sensitive as culture. Cultures are positive in about 85%
of cases with disseminated histoplasmosis, though multiple cultures are recommended to improve the yield
(16). H. capsulatum tends to grow slowly; cultures must be held 4-6 weeks
before they are reported as showing no growth. If disseminated histoplasmosis is suspected, at least
three sets of fungal blood cultures (i.e., using lysis-centrifugation method) should be collected.
Molecular diagnostic methods, though not commercially available, are being developed and used in some
Treatment guidelines from the Infectious Diseases Society of America for
treatment of histoplasmosis were published in 2007 . Without treatment, mortality of disseminated
histoplasmosis is 80%; with amphotericin B treatment, mortality can be reduced to 25%
amphotericin B (AmBisome)
was shown to be more effective than standard amphotericin B
Itraconazole is the treatment of choice for patients with less severe histoplasmosis or following initial
treatment with amphotericin B
I thank Dr. Xiaoyan Wang who originally consulted with me on the case and provided further clinical
details and follow-up information. I also appreciate the clinical perspective and additional information
on the patient provided by Dr. Virginia Caine.
- Alberti-Flor, J. J., and A. Granda. 1986. Ileocecal histoplasmosis mimicking Crohn's disease in a patient with Job's syndrome. Digestion 33:176-80.
- Cappell, M. S., and N. C. Manzione. 1991. Recurrent colonic histoplasmosis after standard therapy with amphotericin B in a patient with Job's syndrome. Am J Gastroenterol 86:119-20.
- Chandler, F., and J. Watts. 1997. Histoplasmosis capsulati, p. 1007-1015. In D. Connor, F. Chandler, D. Schwartz, H. Manz, and E. Lack (ed.), Pathology of Infectious Diseases. Appleton & Lange, Stanford (CT).
- Cutler, J. E., G. S. Deepe, Jr., and B. S. Klein. 2007. Advances in combating fungal diseases: vaccines on the threshold. Nat Rev Microbiol 5:13-28.
- Darling, S. 1906. A protozoan general infection producing pseudo tubercles in the lungs and focal necrosis in the liver, spleen, and lymph nodes. JAMA 46:1283-1285.
- Darling, S. 1908. Histoplasmosis: a fatal infectious disease resembling kala-azar found among natives of tropical America. Arch Intern Med 2:107-123.
- De Monbreun, W. 1934. Cultivation and cultural characteristics of Darling's Histoplasma capsulatum. Am J Trop Med 14:93-125.
- Emmons, C. W. 1949. Isolation of Histoplasma capsulatum from soil. Public Health Rep 64:892-6.
- Goodwin, R. A., Jr., and R. M. Des Prez. 1978. State of the art: histoplasmosis. Am Rev Respir Dis 117:929-56.
- Goodwin, R. A., Jr., J. L. Shapiro, G. H. Thurman, S. S. Thurman, and R. M. Des Prez. 1980. Disseminated histoplasmosis: clinical and pathologic correlations. Medicine (Baltimore) 59:1-33.
- Goodwin, R. A., J. E. Loyd, and R. M. Des Prez. 1981. Histoplasmosis in normal hosts. Medicine (Baltimore) 60:231-66.
- Hage, C. A., L. J. Wheat, J. Loyd, S. D. Allen, D. Blue, and K. S. Knox. 2008. Pulmonary histoplasmosis. Semin Respir Crit Care Med 29:151-65.
- Haggerty, C. M., M. C. Britton, J. M. Dorman, and F. A. Marzoni, Jr. 1985. Gastrointestinal histoplasmosis in suspected acquired immunodeficiency syndrome. West J Med 143:244-6.
- Henderson, R., and L. Moore. 1942. Histoplasma capsulatum as a cause of chronic ulcerative colitis. JAMA 118:885-889.
- Johnson, P. C., L. J. Wheat, G. A. Cloud, M. Goldman, D. Lancaster, D. M. Bamberger, W. G. Powderly, R. Hafner, C. A. Kauffman, and W. E. Dismukes. 2002. Safety and efficacy of liposomal amphotericin B compared with conventional amphotericin B for induction therapy of histoplasmosis in patients with AIDS. Ann Intern Med 137:105-9.
- Kahi, C. J., L. J. Wheat, S. D. Allen, and G. A. Sarosi. 2005. Gastrointestinal histoplasmosis. Am J Gastroenterol 100:220-31.
- Kane, S., and T. Brasitus. 2000. Histoplasmosis capsulatum as a cause of lower gastrointestinal bleeding in common variable immunodeficiency. Dig Dis Sci 45:2133-5.
- Kauffman, C. A. 2007. Histoplasmosis: a clinical and laboratory update. Clin Microbiol Rev 20:115-32.
- Kumar, V., A. Abbas, and N. Fausto. 2005. Robbins and Cotran Pathologic Basis of Disease, 7 ed. Elsevier Saunders, Philadelphia.
- Kwon-Chung, K., and J. Bennett. 1992. Medical Mycology. Lea & Febiger, Philadelphia.
- Lamps, L. W., C. P. Molina, A. B. West, R. C. Haggitt, and M. A. Scott. 2000. The pathologic spectrum of gastrointestinal and hepatic histoplasmosis. Am J Clin Pathol 113:64-72.
- Lee, J. H., N. R. Slifman, S. K. Gershon, E. T. Edwards, W. D. Schwieterman, J. N. Siegel, R. P. Wise, S. L. Brown, J. N. Udall, Jr., and M. M. Braun. 2002. Life-threatening histoplasmosis complicating immunotherapy with tumor necrosis factor alpha antagonists infliximab and etanercept. Arthritis Rheum 46:2565-70.
- Peabody, J. J. 1956. Histoplasmosis: unraveling the Panamanian puzzle. NEJM 255:408-413.
- Rubin, H., M. L. Furcolow, J. L. Yates, and C. A. Brasher. 1959. The course and prognosis of histoplasmosis. Am J Med 27:278-88.
- Sarosi, G. A., D. W. Voth, B. A. Dahl, I. L. Doto, and F. E. Tosh. 1971. Disseminated histoplasmosis: results of long-term follow-up. A center for disease control cooperative mycoses study. Ann Intern Med 75:511-6.
- Sathapatayavongs, B., B. E. Batteiger, J. Wheat, T. G. Slama, and J. L. Wass. 1983. Clinical and laboratory features of disseminated histoplasmosis during two large urban outbreaks. Medicine (Baltimore) 62:263-70.
- Savides, T. J., F. G. Gress, L. J. Wheat, S. Ikenberry, and R. H. Hawes. 1995. Dysphagia due to mediastinal granulomas: diagnosis with endoscopic ultrasonography. Gastroenterology 109:366-73.
- Schwarz, J., and G. Baum. 1957. The history of histoplasmosis, 1906 to 1956. NEJM 256:253-258.
- Suh, K. N., T. Anekthananon, and P. R. Mariuz. 2001. Gastrointestinal histoplasmosis in patients with AIDS: case report and review. Clin Infect Dis 32:483-91.
- Vail, G. M., R. S. Young, L. J. Wheat, R. S. Filo, K. Cornetta, and M. Goldman. 2002. Incidence of histoplasmosis following allogeneic bone marrow transplant or solid organ transplant in a hyperendemic area. Transpl Infect Dis 4:148-51.
- Wallis, R. S., M. S. Broder, J. Y. Wong, M. E. Hanson, and D. O. Beenhouwer. 2004. Granulomatous infectious diseases associated with tumor necrosis factor antagonists. Clin Infect Dis 38:1261-5.
- Wheat, J. 1997. Histoplasmosis. Experience during outbreaks in Indianapolis and review of the literature. Medicine (Baltimore) 76:339-54.
- Wheat, J., M. L. French, R. B. Kohler, S. E. Zimmerman, W. R. Smith, J. A. Norton, H. E. Eitzen, C. D. Smith, and T. G. Slama. 1982. The diagnostic laboratory tests for histoplasmosis: analysis of experience in a large urban outbreak. Ann Intern Med 97:680-5.
- Wheat, L. 1996. Histoplasmosis in the Acquired Immunodeficiency Syndrome. Current Topics in Medical Mycology 7:7-18.
- Wheat, L. J., P. A. Connolly-Stringfield, R. L. Baker, M. F. Curfman, M. E. Eads, K. S. Israel, S. A. Norris, D. H. Webb, and M. L. Zeckel. 1990. Disseminated histoplasmosis in the acquired immune deficiency syndrome: clinical findings, diagnosis and treatment, and review of the literature. Medicine (Baltimore) 69:361-74.
- Wheat, L. J., A. G. Freifeld, M. B. Kleiman, J. W. Baddley, D. S. McKinsey, J. E. Loyd, and C. A. Kauffman. 2007. Clinical practice guidelines for the management of patients with histoplasmosis: 2007 update by the Infectious Diseases Society of America. Clin Infect Dis 45:807-25.
- Wheat, L. J., R. B. Kohler, M. L. French, M. Garten, M. Kleiman, S. E. Zimmerman, W. Schlech, J. Ho, A. White, and Z. Brahmi. 1983. Immunoglobulin M and G histoplasmal antibody response in histoplasmosis. Am Rev Respir Dis 128:65-70.
- Wood, K. L., C. A. Hage, K. S. Knox, M. B. Kleiman, A. Sannuti, R. B. Day, L. J. Wheat, and H. L. Twigg, 3rd. 2003. Histoplasmosis after treatment with anti-tumor necrosis factor-alpha therapy. Am J Respir Crit Care Med 167:1279-82.