HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Hoang, L. M.N. Articles by Roscoe, D. L. Search for Related Content PubMed PubMed Citation Articles by Hoang, L. M.N. Articles by Roscoe, D. L. Agricola Articles by Hoang, L. M.N. Articles by Roscoe, D. L.

Cryptococcus neoformans infections at Vancouver Hospital and Health Sciences Centre (1997–2002): epidemiology, microbiology and histopathology

¹Department of Pathology and Laboratory Medicine, The University of British Columbia, Vancouver Hospital and Health Sciences Centre, 855 W. 12th Ave, Vancouver, BC, Canada ²Epidemiology Services, British Columbia Centre for Disease Control, 655 W. 12th Ave, Vancouver, BC, Canada V5Z 4R4

Correspondence Diane L. Roscoe droscoe{at}vanhosp.bc.ca

Received August 13, 2003
Accepted May 25, 2004

An outbreak of infections due to a rare subspecies of Cryptococcus neoformans (var. gattii) was recognized on Vancouver Island (VI), British Columbia, in 2002, which had affected 59, mostly immunocompetent, individuals since 1999. The objectives of this study were to: (1) determine if the outbreak had spread to Vancouver and its surrounding communities and (2) review the epidemiological, clinical and pathological features of all cryptococcal infections in patients admitted to the Vancouver Hospital and Health Sciences Centre (VHHSC) over a 5 year period. VHHSC microbiology and pathology databases were searched for cryptococcal infections from 1 June 1997 to 31 December 2002. Hospital charts of all identified patients were reviewed. Available cryptococcal isolates and histopathological specimens were reviewed. Twenty-six cases of cryptococcosis were identified in both HIV-positive (n = 15) and HIV-negative (n = 11) patients. C. neoformans var. grubii was cultured from 13 patients, of whom 10 were HIV-positive. The outbreak strain, C. neoformans var. gattii, was detected in three patients; all had travelled to VI. C. neoformans var. neoformans was cultured from two patients, Cryptococcus laurentii was cultured from one, and seven patients had cryptococcosis based on histopathology alone, without cultures. The majority (10/15) of the HIV-positive patients developed systemic disease whilst HIV-negative patients (8/11) presented with pulmonary cryptococcosis. Lung biopsies revealed necrotizing and/or fibrosing granulomas, with cryptococcal cells in 5 of 10 specimens. Brain biopsies showed cryptococcal organisms within leptomeninges and deeper structures with minimal associated inflammation. This retrospective study demonstrated a sharp increase in the total number of C. neoformans infections in both immunocompromised and immunocompetent patients at the VHHSC in 2002. There was no evidence of spread of the outbreak strain to the Greater Vancouver area. This is the first correlation of clinical and investigational findings of cryptococcosis in a region in North America where C. neoformans varieties gattii and grubii are endemic.

	INTRODUCTION

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
In January 2002, the British Columbia Centre for Disease Control (BCCDC) identified an outbreak of Cryptococcus neoformans var. gattii on the east coast of Vancouver Island (VI), British Columbia (BC), the largest island off the west coast of North America. The outbreak affected at least 59 patients and caused two deaths between January 1999 and July 2002 (Fyfe et al., 2002). Unlike the ‘classic’ cryptococcal disease in immunocompromised patients (Kerkering et al., 1981), 75 % of these cases presented primarily with pulmonary disease, and 25 % presented with central nervous system (CNS) involvement. Both medical and media attention were subsequently focused on cryptococcal diseases: local and national newspaper headlines read ‘Tropical fungus a mystery in BC’ (Boei, 2002) and ‘Vancouver Island fungus scares off tourists’ (Jang, 2002).

In the setting of this unusual outbreak, the objective of this investigation was to review the epidemiology, clinical presentation, and microbiological and pathological features of cryptococcal infections in patients admitted between 1997 and 2002 to the quaternary care hospital in Vancouver, the Vancouver Hospital and Health Sciences Centre (VHHSC), the major referral centre for surgical and diagnostic procedures in the Lower Mainland and for the province of BC.

	METHODS

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Epidemiology and clinical investigations.

A retrospective patient chart review for cryptococcal infections was undertaken from 1 June 1997 to 31 December 2002. Ethical approval was granted by the University of British Columbia Ethics Committee. We identified the cases using in-house search codes for cryptococcal infections in both the VHHSC Microbiology and Pathology databases. Patients were included in this study if a clinical diagnosis of cryptococcal disease was made on or during the period of admission at VHHSC and confirmed by either positive culture or presence of cryptococcal organisms identified on cytological and/or histological studies. Patients were excluded when all available specimens were negative for cryptococcal organisms by culture, and cytological and/or histological studies. The patient charts and hospital computer patient information system, from time of index diagnosis, were reviewed and pertinent data were recorded.

The survey was designed to review the epidemiological, clinical, radiological, microbiological and histopathological features associated with cryptococcal disease in patients admitted to the VHHSC. Known risk factors were reviewed (HIV/AIDS, corticosteroid therapy, organ transplantation, cancer, diabetes, contact with birds or travel to VI). Patients admitted for diagnostic day procedures, patients transferred to other centres for continuation of care, and patients who left against medical advice were omitted from the calculation of mean hospital admission duration. The survey also focused on the salient clinical and investigative features, including radiology, microbiology and histopathology.

Microbiological and pathological investigations.

All specimens obtained from patients during hospitalization were processed at the VHHSC Medical Microbiology laboratory. Yeasts isolated from normally sterile sites are routinely identified using two commercially available systems, Auxacolour 2 (Bio-Rad) and Uni-yeast Tek Plate (Remel), with the exception of yeast recovered from urine and sputum specimens. From these latter two sites, yeasts growing in high numbers or predominating were identified when clinically indicated. Isolates from invasive infections are frozen at –70 °C and stored in the microbiology laboratory. For this study, all isolates obtained between January 1997 and December 2002 were plated onto 5 % sheep blood agar plates with Columbia base, incubated at 37 °C with 5 % CO₂ for 48 h and identifications were confirmed using the Auxacolour 2 yeast identification kit. Cryptococcal isolates from all sources were sent to BCCDC Laboratory Services, for further speciation and subspecies identification using a canavanine-glycine-bromthymol agar plate (Kwon-Chung et al., 1982), which differentiates C. neoformans var. gattii from C. neoformans var. neoformans and C. neoformans var. grubii, followed by serology using the Cryptocheck Iatron agglutination test kit (Iatron Laboratories).

Pathology reports and microscope slides of biopsies and autopsies from all identified cases were reviewed. Tissue sections were stained with H & E and special stains, which included mucicarmine, PAS, methenamine silver stain and Ziehl–Neelsen. Gram's stain was performed when indicated.

Statistical analyses.

Categorical data were analysed by Fisher's exact test (two-tailed) using Graph Pad Prism software (version 3). P < 0.05 was considered statistically significant.

	RESULTS AND DISCUSSION

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
In this retrospective analysis, 27 cases of cryptococcosis at the VHHSC were identified from the VHHSC Medical Microbiology and Pathology databases between 1997 and 2002. One case was excluded for not meeting the study inclusion criteria of positive cultures and/or histopathology. Of the remaining 26 cases, 77 % were identified through culture and 23 % through histopathological studies. The first 4 years of the study showed a fairly stable number of cryptococcosis cases, while a greater than threefold increase in cryptococcosis cases occurred in 2002. As shown in Fig. 1, the VHHSC admitted three cases of cryptococcal infections per year between 1997 and 2001. In 2002, 11 new patients were diagnosed with cryptococcosis. Isolates from six of these cases were identified as C. neoformans var. grubii and two were C. neoformans var. gattii. Isolates were not available for speciation in three patients with pulmonary nodules who were investigated to exclude an underlying malignancy and cultures were not performed. The two cases of C. neoformans var. gattii identified in 2002 were in an HIV-positive patient and an HIV-negative patient. Both individuals were either from or had travelled to VI where the infection was likely acquired. The three patients with cryptococcosis of unknown variety had lung biopsies performed with no specimens submitted for microbiological investigations. All had travelled or resided in VI and presented with pulmonary nodules consistent with malignant disease, but no malignancy was identified. Based upon the clinical profiles and presenting symptoms, it is possible that these may indeed be additional cases of C. neoformans var. gattii.

View larger version (34K): [in this window] [in a new window]   Fig. 1. A total of 26 cases of cryptococcosis were definitively diagnosed at the VHHSC between 1997 and 2002.

The increase in cases of cryptococcosis was mostly due to C. neoformans var. grubii in HIV-infected patients, an unexpected finding in the era of highly active antiretroviral therapy (HAART). Worldwide, studies have demonstrated a dramatic decline of cryptococcosis in HIV-positive patients with the introduction of HAART (van Elden et al., 2000; McNeill & Kan, 1995). Our findings may reflect an increase in awareness of cryptococcosis resulting from the extensive media coverage of the VI outbreak, an increase in the number of referrals to the quaternary care centre, or a true increase in the incidence of cryptococcosis in the Lower Mainland. Recently, Mirza et al. (2003) suggested that HIV-positive individuals in the United States who developed cryptococcosis in the era of HAART therapy had limited access to health care.

Demographics/epidemiology

The demographic distribution in this series reflects the endemic HIV disease and intravenous drug use in select regions of BC. Table 1 summarizes the demographic profile of both HIV-positive and -negative patients with cryptococcosis. The increase in cryptococcal disease in HIV-negative patients may reflect the presence of the new Cryptococcus variety in BC. Not surprisingly, 10 of the 13 C. neoformans var. grubii cases were isolated from the HIV-positive group, while a statistically significant majority (P = 0.02) of the ‘species-unknown’ cryptococcal cases and two of the three C. neoformans var. gattii cases were identified in HIV-negative patients. Ten of 11 HIV-negative patients in this review had underlying comorbidities, which may have been the predisposing factor for cryptococcal infections. Our data are consistent with studies in Australia and New Zealand where the HIV-positive patients were infected with C. neoformans var. grubii while C. neoformans var. gattii infected the majority of HIV-negative and immunocompetent patients (Chen et al., 2000).

A number of epidemiological factors associated with cryptococcal infections were specifically sought for in the chart review, but the data were, for the most part, not recorded. Some available details included exposure to pets (birds, 1; no birds, 1; dogs, 1; unknown, 23), travel outside BC (no travel, 3; SE Asia, 1; Ontario, 1; unknown, 21), travel within BC (VI, 4; Savory Island, 1; unknown, 21) and special hobbies (pigeon feeding, 1; unknown, 25). One particular case from this series, an elderly woman with diabetes mellitus who enjoyed feeding pigeons in her back balcony, succumbed to cryptococcal meningitis caused by C. neoformans var. grubii possibly aerosolized while she cleaned off pigeon droppings from her bird feeder.

Clinical features

The mean index admission duration was 31 ± 8 days (mean ±SD; range from 2 to 122 days). Six deaths (23 %) were attributed to cryptococcosis, while 18 (69 %) patients were discharged home. One patient was transferred to a peripheral hospital for continuing care, and one was still in hospital at the end of the study period.

Clinical presentation depended upon the immune status of the host. Details of the clinical presentation are summarized in Table 1. Six patients, all HIV-positive, presented with a temperature of greater than or equal to 38 °C. Seventeen had a documented temperature of less than 38 °C at time of admission while three patients did not have a temperature recorded. Eleven of the 26 patients were admitted for pulmonary problems. Eight of these presented on admission with pulmonary symptoms consisting of cough, haemoptysis or shortness of breath, while three were either asymptomatic or had no symptoms documented. Ten patients, the majority (80 %) being HIV-negative, were subjected to lung biopsies for the investigation of pulmonary nodules, which were thought to be suspicious for malignancy (see ‘Culture results’ section for details). CNS involvement was not sought in these patients. Pulmonary involvement was significantly more common in the HIV-negative patients (P = 0.02), while the HIV-positive patients presented primarily with CNS disease (P = 0.02). Twelve of the 26 patients were admitted with CNS symptoms of headache, visual changes, meningismus or confusion. Three of these patients had a documented elevated increased intracranial pressure at time of lumbar puncture.

Two HIV-positive patients were admitted with gastrointestinal symptoms. Specifically, one was a cocaine user who presented with an upper gastrointestinal bleed and the other presented with left upper quadrant pain diagnosed as gastroenteritis. Both were cryptococcaemic during the admission period. Although rare, this is not an unusual presentation for cryptococcosis, especially in the severely immunocompromised host (Washington et al., 1991; Daly et al., 1990; Bonacini et al., 1990).

One HIV-negative individual with end-stage liver disease was transferred from a peripheral hospital for liver transplantation and developed disseminated cryptococcal disease post-transplantation, possibly an activation of quiescent disease.

Radiological findings

Twenty-four cases had at least one chest X-ray (CXR) performed during the admission period. Eighteen CXRs were reported as ‘abnormal'. These findings ranged from diffuse ‘ground glass’ infiltrates bilaterally to, more commonly, solitary and multiple lesions, which were either solid or cavitating. Ten patients underwent a computed tomography (CT) scan of the chest; all scans were abnormal with findings ranging from diffuse interstitial disease to consolidated nodules. Fifteen CT scans of the head were performed and 10 of them were reported as abnormal. The abnormalities ranged from diffuse atrophy to multiple cystic lesions. The radiological investigations of the chest and CNS system in this series were consistent with those reported in cryptococcal patients, with no differences in the HIV-positive and HIV-negative groups (Zinck et al., 2002; Lacomis et al., 2001). The differential diagnosis of patients presenting with pulmonary nodules and/or meningitis should include cryptococcal infection, especially in those who are immunocompromised and/or have travelled to VI.

Microbiological findings

Latex agglutination. The array of diagnostic tools available for the diagnosis of cryptococcosis includes the latex agglutination (LA) test, an antibody agglutination test directed against the capsular protein of Cryptococcus. The LA test can be rapidly performed on cerebrospinal fluid (CSF) as well as serum with high sensitivity and specificity (90 %). Ten patients had a serum LA test performed during the index admission. Eight of the tests were positive: six patients had CNS disease and two had pulmonary disease. Two patients with pulmonary cryptococcal disease had negative serum LA tests. One was an HIV-negative patient with a solitary lung nodule and no microbiology studies; the other was an HIV-positive patient infected with Cryptococcus laurentii. Thirteen of the 15 CSFs were sent for LA tests. Twelve tested positive: all were patients with CNS cryptococcal disease. The three patients with the negative CSF LA test, however, had pulmonary cryptococcal disease and a positive serum LA test with no CNS manifestation.

In this series, LA tests performed on CSF diagnosed all cases of CNS cryptococcosis. Similarly, LA tests performed on patient serum identified all cases of pulmonary and disseminated cryptococcosis except for two patients. One was immunocompetent and HIV-negative, and the other was infected with C. laurentii, confirming the specificity of LA tests for C. neoformans.

Culture results. Thirteen patients had blood drawn for culture. Eight patients, all HIV-positive, had positive cultures: C. neoformans var. grubii was grown from seven of the blood cultures and C. neoformans var. neoformans was grown from one blood culture. One of three bronchial alveolar lavages was positive for C. neoformans var. neoformans. Lung biopsies obtained from 10 patients yielded three positive cultures: C. neoformans var. grubii was isolated from one specimen, C. neoformans var. gattii from another, and C. laurentii from the third. Of the remaining seven biopsies, cultures were not requested in three, cultures were negative in three, and the isolate cultured in one case was identified as C. neoformans, but was not available for serotyping studies. For these cases, the diagnosis of cryptococcosis was based on histological findings and positive serum LA testing.

Of the 15 CSF samples that were obtained, C. neoformans var. grubii was grown from 11, C. neoformans var. gattii was grown from two, C. neoformans var. neoformans was grown from one specimen, and one specimen was culture negative. The index case of the VI outbreak was an immunocompetent person from whom C. neoformans var. gattii was recovered from the CSF and is one of the patients in this review. CNS tissues were available for five cases (two autopsies and three surgical biopsies). Only two were culture positive for C. neoformans: one of the specimens grew C. neoformans var. grubii and the other grew C. neoformans var. gattii, the same index patient with the var. gattii in the CSF described above.

Speciation and subspeciation studies require yeast isolates from culture. In our series, seven (27 %) cases of cryptococcosis were not confirmed by culture. Methods of in situ diagnosis of Cryptococcus subtypes are currently being evaluated for further investigation of these tissue sections. In many of these cases, patients were referred for investigation of incidental pulmonary nodules that were ‘consistent with lung cancer’ and lung biopsies were sent for histological studies only. This was unfortunate and limited our findings, underlining the need to submit specimens to the microbiology laboratory for culture, especially those from patients who present with newly identified pulmonary lesions in endemic areas. In addition, specimens obtained from patients with cryptococcosis as a differential diagnosis must be presented to the microbiology laboratory for appropriate evaluation.

Histopathological features

Lung pathology. A total of 10 chest biopsies were performed. The common histological features were necrotizing and/or fibrosing granulomas. Five of the 10 cases showed histological evidence of yeast cells consistent with Cryptococcus species. There was a minimal inflammatory response in all of the pulmonary tissues examined, a possible reflection of the organism's ability to evade the host's immune response.

Neuropathology. Of two autopsy and three surgical specimens, histological evaluation of the brain tissues showed minimal inflammation with a predominance of lymphocytes, and virtually no polymorphonuclear lymphocytes (PMNs) (PMNs, 0/5; lymphocytes, 4/5; granulomas, 0/5; plasma cells, 1/5). There was no association between the patterns and distribution of the cryptococci and inflammation. Extensive cystic spaces, which were frequently perivascular, were observed, some obliterating basal ganglia. Four of the five brain tissues showed yeast cells consistent with Cryptococcus species. One patient had underlying neurosarcoidosis involving the spinal cord while another showed co-existing Toxoplasma of the CNS. The yeast cells were most commonly identified within the cortex (4/5), white matter (4/5) and leptomeninges (4/5), followed by the deep nuclei (3/5). Cryptococcus species were also identified, on mucicarmine staining, within the ventricles (2/5 cases), cerebellum (1/5), brainstem (1/5) and spinal cord (1/5) of the five cases reviewed.

Neuropathological studies revealed no discernible differences between the HIV-positive and -negative cases. There were no novel patterns of CNS cryptococcosis compared to those previously reported (Chen et al., 2000). The distribution of cryptococcosis correlated with the amount of cerebral blood flow. The observed minimal inflammation probably reflects the ability of this organism to effectively evade the host's immune system. As discussed in detail by Buchanan & Murphy (1998) and Chen et al. (2000), numerous properties of Cryptococcus contribute to its virulence. The best example is the thick, negatively charged, glycoprotein capsule, which evades phagocytosis (Kozel & Mastroianni, 1976), alters antigen presentation (Collins & Bancroft, 1991) and reduces cytokine production (Vecchiarelli et al., 1995) to name a few properties. These yeast cells are also capable of inhibiting leukocyte migration into an inflammatory site (Dong & Murphy, 1995), down-regulating cell-mediated immune response (Murphy & Moorhead, 1982), synthesizing melanin (Kwon-Chung, 1992) and, interestingly, producing mannitol (Wong et al., 1990).

In three of the six fatalities, death was probably a result of increased intracranial pressure (ICP). A total of four cases, all with presenting CNS disease, had a documented elevated ICP. We propose in these cases that mannitol production by cryptococci resulted in the raised ICP that contributed to the morbidity and mortality.

In summary, there was a sudden increase in the number of cryptococcal infections at the VHHSC in 2002, mostly due to C. neoformans var. grubii and partly attributable to the VI outbreak. This is the first correlative review of the clinical and investigational findings of cryptococcosis in a region in North America where C. neoformans varieties gattii and grubii are endemic. Our findings show that early presentation of cryptococcosis, particularly in HIV-negative individuals, may mimic lung cancer. Finally, this study emphasizes the importance of communication between diagnostic laboratories, clinicians, epidemiologists and scientists in improving vigilance in the detection of emerging and re-emerging infectious diseases.

	ACKNOWLEDGEMENTS

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
We thank Dr J. Isaac-Renton, Ms S. Mithani and the BC Centre for Disease Control Laboratory Services for providing the subspeciation results of our cryptococcal isolates. We also gratefully acknowledge the neuropathologists, anatomic pathologists and microbiologists at VHHSC for contributing patient information. Finally we thank Dr S. Schwarz for assistance with data analysis.

	REFERENCES

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES