Symptom-Based Diagnosis in Pediatrics (CHOP Morning Report) 1st Ed.

CASE 11-2

Ten-Year-Old Boy




A 10-year-old boy presented with a 4-day history of worsening cough with occasional episodes of hemoptysis. Three days before presentation, he developed chills and fever to 38.9°C. Two days prior to presentation he had increasingly frequent posttussive emesis. He complained of abdominal pain with coughing. He also complained of “not having any energy.” There was no weight loss or night sweats. There were no known contacts with a chronic cough or history of tuberculosis. No family members lived or worked in nursing homes. He had not traveled outside of the state of Pennsylvania. Prior to this illness, he was actively participating in soccer at school. He had also assisted with household chores that included washing dishes, mowing the lawn, and sweeping the chimney.


The boy’s birth history was unremarkable. He required hospitalization at 1 year of age for Salmonella gastroenteritis leading to dehydration. Epidemiologic investigation attributed a local Salmonella outbreak occurring during that time to a pet store engaging in improper import of reptiles. The family turtle, purchased from that store, was held culpable for this child’s illness and removed from the home at the family’s request. The only other pet was a healthy cat acquired 2 years ago. The patient did not require any medication. He received appropriate childhood immunizations. A paternal uncle suffered from adult-onset diabetes.


T 38.3°C; HR 108 bpm; RR 24-28/min; BP 111/72 mmHg; SpO2 97% in room air

Weight 95th percentile

On examination, the patient was observed to be alert and cooperative. His oropharynx was clear. There was no cervical lymphadenopathy. There were no crackles or wheezing on lung examination. Heart sounds were normal. There was no hepatomegaly or splenomegaly. There were two mildly tender erythematous nodules on the anterior aspect of his left tibia. There were no other rashes.


A complete blood count revealed the following: 20 400 WBCs/mm3 (83% neutrophils, 5% eosinophils, and 11% lymphocytes); hemoglobin, 12.2 g/dL; and 372 000 platelets/mm3.


Chest radiograph revealed ill-defined pulmonary nodules. A chest computed tomography was performed to better delineate the radiographic findings (Figure 11-2). Tuberculous skin testing was negative.


FIGURE 11-2. Chest computed tomogram.



The differential diagnosis of reticulonodular infiltrates on chest radiograph or chest CT includes tuberculosis as well as pulmonary disease caused by endemic fungi, such as blastomycosis, coccidi-oidomycosis, and histoplasmosis. Tuberculosis is possible, particularly given the hemoptysis, but is less likely with a negative tuberculin skin test. Knowledge of this patient’s travel history virtually excludes blastomycosis and coccidioidomycosis. In an adolescent, Mycoplasma pneumoniae may cause hilar adenopathy as well as diffuse lung infiltrates.

Hypersensitivity pneumonitis, sarcoid, and vasculitis (particularly Wegener granulomatosis) may cause similar findings. Sarcoidosis, a multisystem granulomatous disease, usually presents with generalized lymphadenopathy and prominent cervical involvement. Associated findings include erythema nodosum and uveitis. Granuloma formation occurs in the eyes, skin, liver, spleen, and parotid glands. Blacks are more commonly affected than whites. Wegener granulomatosis is relatively uncommon in children.


The chest CT revealed bilateral hilar lymphadenopathy (Figure 11-2). The largest hilar lymph node on the right measured 1.7 × 2.2 cm. The CT also revealed numerous pulmonary nodules ranging in size from a few millimeters to 1 cm. No acid-fast bacilli were detected in serial sputum samples. Mycoplasma pneumoniae polymerase chain reaction (PCR) of a nasopharyngeal aspirate was negative. Antigens to influenza A and B, parainfluenza types 1, 2, and 3, adenovirus, and respiratory syncytial virus were not detected by immunofluorescence of nasopharyngeal washings. The diagnosis of pulmonary histoplasmosis was confirmed by detection of Histoplasma capsulatum antigen in the urine and a fourfold increase in H. capsulatum antibody between acute and convalescent serum samples. The chimney cleaning likely contributed to the development of pulmonary histo-plasmosis. He improved clinically during the next 3 days without specific treatment. A repeat chest radiograph was normal 2 weeks later.


Histoplasma capsulatum is a dimorphic fungus that grows as a yeast-like organism at temperatures greater than 37°C and as a spore-forming mold at lower temperatures. It is endemic in certain areas of the United States and Latin America. Between 1958 and 1965, a total of 275 558 military recruits raised in different areas of the United States underwent Histoplasma skin testing. Recruits from states of the Ohio and Mississippi River valleys along with Maryland and Virginia had the highest rates of reaction. Histoplasma was also prevalent in parts of Pennsylvania and Texas. This study by Edwards et al. remains the largest study of histoplasmosis endemicity in the United States.

Histoplasma capsulatum resides in the soil in endemic areas. Excretions of birds and bats facilitate growth of the organism. As a consequence, H. capsulatum infections are associated with aero-solization of debris from sites where birds or bats roost, as may occur by cutting firewood, sweeping chimneys, or playing in hollow trees, barns, or caves. Infection occurs after inhalation of spores that transform to the yeast phase in the lung. Hematogenous dissemination may occur after primary pulmonary infection. Rarely, the skin or intestinal mucosa serves as the portal of entry.


Severity of illness depends on the intensity of exposure and host’s immune status. Low-intensity exposure usually results in asymptomatic infection in immunocompetent hosts. Higher intensity exposures lead to pulmonary infection, manifested as fever, cough, malaise, and poor appetite. Some patients experience pleuritic chest pain. Rales and wheezing may also occur. Erythema nodosum and other hypersensitivity reactions to infection occasionally develop. Symptoms are self-limited and last 2 or 3 days. In a small number of children, symptoms persist for more than 2 weeks. Symptoms persisting longer than 3 weeks after acute histoplasmosis suggest progressive disease or dissemination. In the immunocompe-tent host, extrapulmonary dissemination is rare. Infants younger than 2 years of age are at higher risk of disseminated disease than older children. Features of disseminated histoplasmosis include prolonged fever, failure-to-thrive, and hepato-splenomegaly. Pericardial and pleural effusions occur rarely.

In the immunocompromised host, the illness begins with fever and cough, followed by worsening respiratory distress. Disseminated histoplasmosis is more likely to occur in immunocompromised patients. These patients usually have diarrhea, weight loss, hepatomegaly, splenomegaly, and skin lesions. Less commonly, dissemination leads to bone marrow involvement, meningitis, pericarditis, or chorioretinitis.


Histoplasma capsulatum does not comprise the normal flora of humans, so its isolation from mucous membranes, skin lesions, deep organs, or body fluids usually indicates infection.

Chest radiograph. In those with respiratory symptoms, chest radiograph abnormalities include hilar adenopathy and localized or diffuse reticulo-nodular lung infiltrates. In patients with previous pulmonary infection (symptomatic or subclinical), single or multiple calcified nodules may be detected in the lungs, hilar lymph nodes, spleen, or liver. Cavitary lesions resembling tuberculosis may be seen in those with chronic pulmonary his-toplasmosis (rare in children). Among immuno-compromised children with disseminated disease, 40%-50% have normal chest radiographs.

Culture. Culture of the organism on standard mycologic media, including brain-heart infusion agar or broth, requires a 2- to 6-week incubation period, making culture methods less useful in the acute setting (but important for confirmation of the organism in certain cases). Specimens appropriate for culture depend on the site of infection (pulmonary, cutaneous, or disseminated) and include sputum (for pulmonary disease), skin lesion biopsy specimens, blood, bone marrow, and organ biopsy specimens. Culture is most sensitive in patients with disseminated infection, being positive in greater than 75% of cases.

Histoplasmin skin test. The histoplasmin skin test has several limitations that limit its usefulness as a diagnostic tool. First, in endemic areas, the prevalence of skin test positivity because of previous asymptomatic infection approaches 60% among young adults. Second, the skin test is unable to reliably distinguish asymptomatic past infection from symptomatic current infection. Third, administration of the skin test has been associated with falsely elevated antibody titers in 15%-25% of patients, meaning that use of the skin test may complicate the interpretation of other diagnostic methods. Its use is now limited to epidemiologic investigations.

Serum H. capsulatum antibody. This test is recommended for routine detection of infection in otherwise healthy children. Antibodies are detectable 2-4 weeks after infection. Antibody titers greater than 1:8 or a fourfold increase between acute and convalescent titers suggests acute infection. Titers revert to negative 12-18 months after infection. Serologic tests are positive in 90% of patients with symptomatic disease; however, their utility is limited in immunocompromised patients. Cross-reaction from Blastomyces or Coccidioides antibodies can occur, but the travel history usually differentiates these from H. capsulatum.

Histoplasma capsulatum urinary antigen detection. This test is most useful in diagnosing disseminated disease in young children or infection at any site in immunocompromised patients, in whom antibody titers can be falsely negative. It is also positive in 80% of patients with acute pulmonary infection. Sensitivity is increased when combined with serum antigen testing. In a study by Fojtasek et al., H. capsulatum antigenuria was detected in all 22 children with disseminated histoplasmosis. Declining antigenuria levels correlate with clinical improvement. Like antibody titers, false positive antigen tests have been reported in patients with Blastomyces and Coccidiodes.

Histologic examination of tissue. Severely immunocompromised patients with disseminated histoplasmosis commonly have negative antibody titers and normal chest radiographs. They occasionally have negative urine antigen tests as well. In the severely ill, immunocompromised patients with negative H. capsulatum testing but continued suspicion of infection, a bone marrow biopsy should be considered for early detection and management of this life-threatening illness. In these patients, ovoid yeast forms are frequently visible on microscopic examination of bone marrow and biopsy specimens. Grocott-Gomori methenamine-silver nitrate and periodic acid-Schiff (PAS) stains are most useful for identifying H. capsulatum infection.

Other studies. Any child with hilar adenopathy requires evaluation, including tuberculin skin testing, to exclude tuberculosis. Findings in disseminated infection may include pancytopenia, anemia, coagulopathy, elevated liver enzymes, and increased serum ferritin.


Antifungal treatment clearly benefits those with progressive forms of histoplasmosis (e.g., disseminated infection). Other manifestations for which antifungal therapy should be considered include pulmonary infection with protracted symptoms (4 weeks), severe acute pulmonary infection (e.g., hypoxia), and granulomatous adenitis obstructing critical structures such as blood vessels and bronchi. Antifungal therapy is generally not recommended in children with mild-to-moderate acute pulmonary disease.

Options for treatment include ketoconazole, itraconazole, and amphotericin B (deoxycholate or lipid preparations). Fluconazole is less effective than either itraconazole or amphotericin B. Ketoconazole, although effective, is poorly tolerated compared with the other antifungal agents and is associated with a higher rate of hepatoxicity. Voriconazole and posaconazole, newer triazole antifungal agents, demonstrate comparable or better in vitro activity than either itraconazole or amphotericin against H. capsulatum. These agents have also shown success in a small number of patients but require additional clinical evaluation. Amphotericin B deoxycholate is generally well tolerated in children and is preferred over lipid preparation. In general, patients with severe acute pulmonary histoplasmosis should receive a short course of amphotericin B, followed by a prolonged course of itraconazole while those with disseminated disease should receive a prolonged course of amphotericin B.

Duration of therapy depends on the type of histoplasmosis and the underlying host immunocompetence. Those with acute pulmonary disease who require treatment generally receive 1-2 weeks of amphotericin B, followed by 12 weeks of itraconazole. Disseminated infection requires 4-6 weeks of amphotericin B in otherwise healthy patients, but those with acquired immunodeficiency syndrome (AIDS) require lifelong suppressive therapy with itraconazole. Prolonged therapy may be necessary for those with serious H. capsulatum infections, immunosuppression, or primary immunodeficiency syndromes. For these patients, duration of therapy is usually determined in conjunction with an infectious diseases specialist.

Antigen levels should be monitored during therapy to evaluate the response to treatment and for 12 months following completion of therapy to monitor for relapse.


1. Edwards LB, Acquaviva FA, Livesay VT. An atlas of sensitivity to tuberculin, PPD-B, and histoplasmin in the United States. Am Rev Respir Dis. 99;1:1969.

2. Fischer GB, Mocelin H, Severo CB, Oliveira FM, Xavier MO, Severo LC. Histoplasmosis in children. Pediatr Resp Rev. 2009;10:172-177.

3. Flynn PM, Hughes WT. Histoplasmosis. In: Chernick V, Boat TF, eds. Kendig’s Disorders of the Respiratory Tract in Children. 6th ed. Philadelphia: W.B. Saunders Company; 1998:946-953.

4. Fojtasek MF, Kleiman MB, Connolly-Stringfield P, Blair R, Wheat LJ. The Histoplasma capsulatum antigen assay in disseminated histoplasmosis in children. Pediatr Infect Dis J. 1994;13:801-805.

5. Kleiman MB. Histoplasma capsulatum (Histoplasmosis). In: Long SS, Pickering LK, Prober CG, eds. Principles and Practice of Pediatric Infectious Diseases. 2nd ed. New York: Churchill Livingstone; 2003:1233-1238.

6. Leggiardo RJ, Barrett RD, Hughes WT. Disseminated histoplasmosis of infancy. Pediatr Infect Dis J. 1986;7:799-805.

7. Wheat J, Freifeld AG, Kleiman MB, et al. Practice guidelines for the management of patients with histoplasmosis. Clin Infect Dis. 2007;45:807-825.