Cryptococcus neoformans is an important fungal pathogen that is recognized for its ability to cause subacute central nervous system (CNS) disease, especially among immunocompromised patients. The initial presentation of C. neoformans infection is usually CNS disease, but infection is acquired as a result of inhalation of infectious particles from the environment. Accordingly the lungs are the most commonly involved site of disease after the CNS.
Cryptococcal disease occurs less frequently in children than in adults. Two studies have determined a prevalence of 0.5 to 1% in children with AIDS. 1,2 Reports of children with cryptococcosis in the context of other underlying immunodeficiencies including acute lymphocytic leukemia, sarcoma, hyper-IgE syndrome, hyper-IgM syndrome and Bruton’s agammaglobulinemia have been published. 3–7 These generally consist of individual case reports or small series, reflecting the infrequency of this disease in children. Most reports describe CNS or disseminated cryptococcosis; reports of pulmonary cryptococcosis among children are particularly uncommon.
Nevertheless C. neoformans can be easily isolated form the environment and its association with pigeon excreta has been well-documented. 8 Several recent studies suggest that exposure to this fungus is common among children, especially those living in an urban area. In a serologic survey of children from the Bronx, Abadi and Pirofski 9 found that immunocompetent and HIV-infected children had significantly elevated titers to cryptococcal polysaccharide, with median IgG titers of 1/1000 and 1/325, respectively. In a study of 185 immunocompetent children living in the Bronx, we found antibodies to cryptococcal proteins in the majority of sera from children older than 2 years. 10 Cryptococcal polysaccharide at a concentration of 10 ng/ml was present in the serum of one patient from this cohort, a 5-year-old who presented with vomiting. These studies indicate that C. neoformans infection may be common in children, but that this infection is subclinical or not recognized.
In this report we describe a 10-year-old boy with a malignant fibrous histiocytoma of bone and pulmonary nodule secondary to C. neoformans infection. This patient’s diagnosis was suggested by the appearance of a pulmonary lesion on a computerized tomographic study (CT) done as surveillance for metastatic disease and ultimately confirmed by immunohistochemical techniques on the excised lesion with the use of a monoclonal antibody directed against the capsular polysaccharide. Using this case as a background, we review the reported clinical characteristics of pulmonary cryptococcosis in children. The approach to evaluation for systemic disease and utility of various diagnostic assays, including immunohistologic techniques to establish a diagnosis of pulmonary cryptococcosis, are also discussed.
A 10-year-old boy, diagnosed with malignant fibrous histiocytoma of bone in May 2000, completed chemotherapy and radiation therapy 1 year later. He remained symptom-free for months during which time he traveled from his home in western New York to southern California for a brief vacation. In August 2001 he had a routine chest CT which was negative for pulmonary disease. A subsequent physical examination several months later was normal. A routine chest CT scan done at this time to evaluate for potential metastases revealed right upper and lower lobe nodules which were subsequently resected (Fig. 1). In retrospect his family noted that the patient did have a persistent, moderate, nonproductive cough for the previous month. They denied any other signs or symptoms of infection including fever, nuchal rigidity or changes in mental status. On physical examination he appeared healthy. Respirations were regular and nonlabored. His lungs were clear to auscultation and percussion. The absolute neutrophil count was 3675 cells/mm3, and the the absolute lymphocyte count was 686 cells/mm3.
Histologic examination of the lower lobe nodule revealed a metastatic tumor consistent with malignant fibrous histiocytoma, whereas the upper lobe lesion demonstrated yeast-like organisms within a microabscess surrounded by granulomatous inflammation. No tumor cells were noted in the upper lobe lesion. Special stains including acid-fast, Gomori methamine silver and mucicarmine were used, and the morphologic features of the observed organism were consistent with C. neoformans. Cultures of surgical specimens were negative. A lumbar puncture was performed, and CSF indices including protein, glucose, white blood cell count and cryptococcal antigen were normal. CSF culture was negative and serum cryptococcal antigen was not detected.
Immunohistochemical studies using a murine monoclonal antibody, 18B7, which specifically recognizes the cryptococcal polysaccharide capsular material confirmed the presence of encapsulated C. neoformans organisms and demonstrated the presence of soluble polysaccharide within the tissue (Fig. 2). 11 Given the history of travel to southern California, the possibility of infection with C. neoformans var. gattii was considered. This variety of C. neoformans is typically native to Australasia and epidemiologically linked to eucalyptus trees. Nevertheless var. gattii infections have been reported in California and are more likely to cause disease in immunocompetent individuals than var. neoformans. To rule out this possibility immunostaining was done with a monoclonal antibody (302) that specifically recognizes C. neoformans variety neoformans (serotypes A and D). 12 The antibody for this assay was donated by T Kozel (Reno, NV). Staining with monoclonal antibody 302 produced strong capsular staining (not shown), indicating that this organism was in fact of the variety neoformans. Serum IgG and IgM titers to the cryptococcal polysaccharide were determined by ELISA 9 and were 1/3200 and 1/800, respectively.
The child completed a 2-week course of amphotericin B followed by prolonged (∼18 months) fluconazole therapy that coincided with the duration of his immunosuppressive therapy. His cough gradually resolved during a few months while receiving therapy. Follow-up chest CT scans done while the patient was receiving fluconazole and after therapy revealed no evidence of recurrent cryptococcal disease, but a nodule secondary to recurrent malignancy was ultimately resected. Follow-up encounters have revealed no symptoms or signs of recurrent pulmonary cryptococcosis.
Twelve other cases of pulmonary cryptococcosis have been described in children ranging in age from 5 to 17 years (Table 1). We identified these cases by searches of the English medical literature using the PubMed and Medline databases with Cryptococcus and children as key words. Cases were included if the primary presentation of disease was respiratory and the appropriate clinical and laboratory data were provided. Three cases of pulmonary disease in children were identified from reports that predated these databases. 17,18 These three cases were diagnosed by histologic examination but are included for the sake of completeness.
The majority of affected children were immunocompromised (Table 1), whereas about one-third (4 of 13) were immunocompetent. Most children (10 of 13 including our patient) were male. In addition to our patient, several cases of pediatric pulmonary cryptococcosis in the context of soft tissue malignancy have been previously described (Table 1). 4 Similar to our patient all patients with soft tissue malignancies had completed various courses of chemotherapy and were lymphopenic when pulmonary cryptococcosis was diagnosed. Although it is difficult to conclude an increased prevalence of pulmonary cryptococcosis among children with soft tissue malignancy, it seems reasonable to hypothesize that the immunosuppression associated with soft tissue malignancies or the chemotherapy that these patients receive makes them more susceptible to cryptococcal infection. Alternatively it may be that pulmonary cryptococcal infection is more commonly recognized in these children as a result of routine imaging done to evaluate for potential metastatic disease. Possibly a combination of both factors is involved, i.e. somewhat enhanced susceptibility and increased detection.
An interesting aspect of pulmonary cryptococcosis is the potential for asymptomatic disease. Six of the 13 cases in this review were asymptomatic. As many as one-third of adult patients with recognized pulmonary cryptococcosis are asymptomatic at the time of presentation. 19 In fact cryptococcal disease is often recognized as a result of imaging done for other reasons. These findings, along with serologic studies, the ubiquitous nature of C. neoformans in the environment and the asymptomatic nature of infections caused by other endemic fungi, have led some to conclude that the majority of pulmonary infections caused by C. neoformans in immunocompetent individuals are asymptomatic. 20,21 When symptoms are present they are usually subacute or chronic. A diagnosis of cryptococcal pneumonia is often made in the context of a failure to respond to standard antimicrobial therapy. Reported symptoms of pulmonary cryptococcal infection in this series include cough, fever, chest pain and weight loss (Table 1). Other symptoms reported in adults include increased sputum production, hemoptysis and dyspnea.
The radiographic features of pulmonary cryptococcosis are related to the underlying immune status of the patient. Isolated pulmonary nodule(s) are often found in immunocompetent individuals. 22 An apparent predisposition for lower lobe involvement was present in this series (Table 1), which is similar to the findings of some adult series. 21 Other radiographic findings include: mass-like lesions, consolidative lesions (involving single or multiple lobes) and pleural effusions. Diffuse interstitial infiltrates are more common among immunocompromised individuals and especially among adult patients with AIDS. 23–25
The sensitivity of sputum culture in the diagnosis of pulmonary cryptococcosis is limited, particularly among individuals without AIDS. 19,26 Furthermore the significance of isolating C. neoformans from respiratory secretions is not always clear. Several studies have demonstrated the potential for airway colonization by C. neoformans among patients with chronic lung disease and or lung cancer. 27,28 In addition current serologic assays do not distinguish between colonization and infection. Our patient had elevated IgG (1/3200) and IgM (1/800) to the cryptococcal polysaccharide, although given the high level of antibodies to cryptococcal polysaccharide in the general population the significance of these findings is unclear. The detection of cryptococcal polysaccharide by a latex agglutination assay in the sera of patients is often indicative of disseminated disease but might have limited sensitivity in the diagnosis of patients with isolated pulmonary cryptococcosis. 29
Given the limited sensitivity of these diagnostic methods and the extensive differential diagnosis associated with pulmonary cryptococcosis, invasive testing is often necessary. In the context of nodular pulmonary cryptococcosis, biopsy or fine needle aspiration may be the most appropriate approach. Nevertheless false negative results might occur. 13,30,31 In some cases of pulmonary cryptococcosis, the tissue load of viable fungi may be relatively low. Regardless of how the specimen is obtained, C. neoformans is a slow growing organism in culture, and specimens may required 7 to 8 days before becoming positive. 30
Our report demonstrates the utility of an immunohistochemical approach in the diagnosis of cryptococcal pneumonia in the context of a negative culture. This technique is based on the ability of specific monoclonal antibodies to recognize the main cryptococcal capsular polysaccharide (GXM) antigen (in situ). The cryptococcal polysaccharide antigen is relatively stable and can be detected in both frozen and formalin-fixed paraffin-embedded tissues. Also this assay can be easily performed within 1 day. This approach has been used extensively to study the pathogenesis of experimental cryptococcosis in mice and rats. 32,33 These studies indicate that polysaccharide capsule of C. neoformans is quickly shed into the tissue of animals with experimental pulmonary cryptococcosis and may persist for prolonged times. More recently Krockenberger et al. 12 has shown that this approach may allow for the distinction between the two varieties of C. neoformans known as var. neoformans (serotypes A, D and AD) and var. gattii, (serotypes B and C) in animal tissue.
Given the small number of pediatric cases, treatment recommendations for pulmonary cryptococcosis are based primarily on extrapolation of experience in adults. Although spontaneous resolution of pulmonary cryptococcal infection is the usual course in immunocompetent individuals, subsequent dissemination can occur. 34 In light of this potential risk and the low toxicity associated with fluconazole therapy, we recommend that immunocompetent children with pulmonary cryptococcosis be treated with fluconazole. For all immunocompromised patients a workup for extrapulmonary disease including a lumbar puncture is indicated. In these cases the type and duration of therapy should be guided by the immune status of the host and the extent of disease. 28
1. Abadi J, Nachman S, Kressel AB, Pirofski LA. Cryptococcosis in children with AIDS. Clin Infect Dis 1999; 28: 309–13.
2. Gonzalez CE, Shetty D, Lewis LL, Mueller BU, Pizzo PA, Walsh TJ. Cryptococcosis in human immunodeficiency virus-infected children. Pediatr Infect Dis J 1996; 15: 796–800.
3. Leggiadro RJ, Barrett FF, Hughes WT. Extrapulmonary cryptococcosis in immunocompromised infants and children. Pediatr Infect Dis J 1992; 11: 43–7.
4. Allende M, Pizzo PA, Horowitz M, Pass HI, Walsh TJ. Pulmonary cryptococcosis presenting as metastases in children with sarcomas. Pediatr Infect Dis J 1993; 12: 240–3.
5. Stone BD, Wheeler JG. Disseminated cryptococcal infection in a patient with hyperimmunoglobulinemia E syndrome. J Pediatr 1990; 117: 92–5.
6. Iseki M, Anzo M, Yamashita N, Matsuo N. Hyper-IgM immunodeficiency with disseminated cryptococcosis. Acta Paediatr 1994; 83: 780–2.
7. Wahab JA, Hanifah MJ, Choo KE. Bruton’s agammaglobulinaemia in a child presenting with cryptococcal empyema thoracis and periauricular pyogenic abscess. Singapore Med J 1995; 36: 686–9.
8. Emmons CW. Saprophytic sources of Cryptococcus neoformans
associated with the pigeon (Columba livia
). Am J Hyg 1955; 62: 227–52.
9. Abadi J, Pirofski LA. Antibodies reactive with the cryptococcal polysaccharide glucuronoxylomannan are present in sera from children with and without human immunodeficiency virus infection. J Infect Dis 1999; 180: 915–9.
10. Goldman DL, Khine H, Abadi J, et al. Serologic evidence for Cryptococcus neoformans
infection in early childhood. Pediatrics 2001; 107: e66.
11. Casadevall A, Cleare W, Feldmesser M, et al. Characterization of a murine monoclonal antibody to Cryptococcus neoformans
polysaccharide that is candidate for human therapeutic studies. Infect Immun 1998; 42: 1437–46.
12. Krockenberger MB, Canfield PJ, Kozel TR, et al. An immunohistochemical method that differentiates Cryptococcus neoformans
varieties and serotypes in formalin-fixed paraffin-embedded tissues. Med Mycol 2001; 39: 523–33.
13. Pan AH, Janner D. Lung mass in a sixteen-year-old young man. Pediatr Infect Dis J 2003; 22: 293–4, 296.
14. Gonzalez CE, Shetty D, Lewis LL, Mueller BU, Pizzo PA, Walsh TJ. Cryptococcosis in human immunodeficiency virus-infected children. Pediatr Infect Dis J 1996; 15: 796–800.
15. Cohen I. Isolated pulmonary cryptococcosis in a young adolescent. Pediatr Infect Dis 1985; 4: 416–8.
16. Meighan JW. Pulmonary cryptococcosis mimicking carcinoma of the lung. Radiology 1972; 103: 61–2.
17. Siewers CMF, Cramblett HG. Cryptococcosis (torulosis) in children. A report of four cases. Pediatrics 1964; 34: 393–400.
18. Haugen RK, Baker RD. The pulmonary lesions in cryptococcosis with special reference to subpleural nodules. Am J Clin Pathol 1954; 24: 1381–90.
19. Campbell GD. Primary pulmonary cryptococcosis. Am Rev Respir Dis 1966; 94: 236–43.
20. Casadevall A, Perfect JR. Human cryptococcosis. In: Cryptococcus neoformans. Washington, DC: ASM Press, 1998: 407–56.
21. Littman ML, Zimmerman LE. Cryptococcosis, torulosis or European blastomycosis. Mew York: Grune & Stratton, 1956.
22. Khoury MB, Godwin JD, Ravin CE, Gallis HA, Halvorsen RA, Putman CE. Thoracic cryptococcosis: immunologic competence and radiologic appearance. AJR Am J Roentgenol 1984; 142: 893–6.
23. Miller WT Jr, Edelman JM, Miller WT. Cryptococcal pulmonary infection in patients with AIDS: radiographic appearance. Radiology 1990; 175: 725–8.
24. Wasser L, Talavera W. Pulmonary cryptococcosis in AIDS. Chest 1987; 92: 692–5.
25. Cameron ML, Bartlett JA, Gallis HA, Waskin HA. Manifestations of pulmonary cryptococcosis in patients with acquired immunodeficiency syndrome. Rev Infect Dis 1991; 13: 64–7.
26. Kerkering TM, Duma RJ, Shadomy S. The evolution of pulmonary cryptococcosis. Ann Intern Med 1981; 94: 611–16.
27. Duperval R, Hermans PE, Brewer NS, Roberts GD. Cryptococcosis, with emphasis on the significance of isolation of Cryptococcus neoformans
from the respiratory tract. Chest 1977; 72: 13–9.
28. Aberg JA, Mundy LM, Powderly WG. Pulmonary cryptococcosis in patients without HIV infection. Chest 1999; 115: 734–40.
29. Pappas PG, Perfect JR, Cloud GA, et al. Cryptococcosis in human immunodeficiency virus-negative patients in the era of effective azole therapy. Clin Infect Dis 2001; 33: 690–9.
30. Kontoyiannis DP, Peitsch WK, Reddy BT, et al. Cryptococcosis in patients with cancer. Clin Infect Dis 2001; 32: E145–50.
31. Lee LN, Yang PC, Kuo SH, Luh KT, Chang DB, Yu CJ. Diagnosis of pulmonary cryptococcosis by ultrasound guided percutaneous aspiration. Thorax 1993; 48: 75–8.
32. Feldmesser M, Casadevall A. Effect of serum IgG1 to Cryptococcus neoformans
glucuronoxylomannan on murine pulmonary infection. J Immunol 1997; 158: 790–9.
33. Goldman D, Lee SC, Casadevall A. Pathogenesis of pulmonary Cryptococcus neoformans
infection in the rat. Infect Immun 1994; 62: 4755–61.
34. Nunez M, Peacock JE, Chin R. Pulmonary cryptococcosis in the immunocompetent host. Chest 2000; 118: 527–34.