INTRODUCTION

Histoplasmosis is an infection caused by Histoplasma capsulatum that affects humans and several other animal species ^18,20 . The organism grows profusely in soil that is rich in the droppings from birds and bats ¹⁸ . Old abandoned buildings and caves often contain high concentrations of H. capsulatum ¹⁸ . Demolition of buildings, movement of soil, cleaning of bridge structures, and spelunking have been implicated as the point source for dispersal of the organism ¹ .

Prior to the AIDS epidemic, histoplasmosis was most frequently reported in outbreak events ^20,23 , and the disseminated form was found primarily in patients with lymphoma or other cancers and sporadically in renal transplant patients ¹³ . However, since the 1980s, the disseminated form has been observed mainly in HIV-infected patients with advanced immunosuppression ^13,16 . Several Brazilian states have reported cases of histoplasmosis associated with HIV infection ^2,26 , however, the state of Ceará in Northeastern Brazil has had the highest incidence of AIDS/histoplasmosis association in recent years ^7,24 and one of the highest in the world ⁶ . In a retrospective study that was conducted from 1995 to 2004 in Fortaleza, the capital of Ceará, 164 patients were diagnosed with concurrent HIV infection and disseminated histoplasmosis ⁸ .

Histoplasmin skin-test surveys carried out in the general Brazilian population in different regions have reported a positivity of 4.4-63.1% in the Midwest and 6.3-89.0% in the South ¹⁵ . Previously, another survey of several cities throughout the country identified a prevalence ranging from 2.60% to 61.50% in populations living in Northeast Brazil ¹² . Studies from Ceará performed several decades ago indicated a high prevalence (23.6% to 61.5%) of infection in individuals from rural areas ^9,11 .

Based on the paucity of studies concerning the prevalence of histoplasma infection in Ceará, mainly in urban areas, and the high prevalence of disseminated histoplasmosis in HIV patients from this state, studies to address the epidemiology of this fungal infection in the immunosuppressed population in this area are urgently needed.

METHODS

A cross-sectional study was performed at the Center of Medical Specialties José de Alencar (Centro de Especialidades Médicas José de Alencar - CEMJA), a public outpatient clinic in Fortaleza (estimated population in 2012: 2.5 million inhabitants), the capital of Ceará State. CEMJA began to offer HIV outpatient assistance in 2006 and has received most of the newly diagnosed patients from the area since then (communication from the local health department). This study enrolled HIV-positive patients (enzyme-linked immunosorbent assay positive, confirmed with an indirect immunofluorescence test) of both sexes over 18 years old, residents in Fortaleza, and included patients with a CD4 count ≥ 350 cells/mm ³ regardless of their use of highly active antiretroviral therapy (HAART). Patients with a previous history of histoplasmosis and those who did not return for skin-test reading were excluded.

After providing written informed consent, the patients completed a questionnaire concerning socio-economic and demographic data and known risk factors associated with Histoplasma infection, such as recent (up to a year before) or past (more than a year ago) activities involving soil (gardening, civil construction or agriculture) or visits to farms or caves, or the presence of birds, bats, or mango trees (Mangifera indica) in the home or neighborhood. Clinical and laboratory data, such as time of AIDS and HIV diagnosis, CD4 cell count, and highly active antiretroviral therapy (HAART) use, were obtained from patient records.

Initially, 5 mL of blood was collected and stored at -20 °C for subsequent analysis. Skin tests were subsequently performed by intradermally injecting 0.1 mL of histoplasmin antigen (HMIN) ⁹ , in a 1:1000 dilution ¹² , into the inside of the left forearm of each selected individual ⁷ . HMIN is a metabolic antigen from the filamentous phase of H. capsulatum produced in-house, isolated from sputum of a case of chronic pulmonary histoplasmosis, (IPEC3356), in Rio de Janeiro provided by the Mycology Laboratory at FIOCRUZ - Rio de Janeiro. Tests that produced induration ≥ 5 mm in transverse diameter after 48 or 72 hours were considered to be histoplasmin positive ⁹ . The intradermal tests and readings were performed by the same investigator using the same instrument (gauge).

All patients who had a positive reaction to the intradermal test were analyzed with a double immunodiffusion test; the negative histoplasmin sera selected for testing were taken from patients with the highest CD4 counts. To control for possible cross-reactivity between H. capsulatum and Coccidioides immitis antibodies, 36 sera were randomly selected and tested with commercial antibodies and antigens for H. capsulatum IDCF and anti-IDCF, (Immy Immunodiagnostics, OK, USA) and C. immitis (IDCF and anti-IDCF, Immy Immunodiagnostics, OK, USA). The double radial immunodiffusion assay was performed as described by SIDRIM & OLIVEIRA ²⁵ .

Statistical analyses were performed with STATA version 8 (Stata Corporation, College Station, USA) and adopted a statistical significance level of 5%. This study was approved by the Ethics in Research Committee of the São José Hospital for Infectious Diseases in Ceará under protocol 018/2008.

RESULTS

A total of 344 patients were invited to participate in the study, and 161 met the inclusion criteria; the remaining 183 had CD4 < 350 cells/mm ³ (169) or were excluded because they did not return for skin-test readings (14). The studied population averaged 35.1 years of age (29-38 years), and the majority of the patients were men (76.4%). As demonstrated in Table 1, approximately 61.4% were born in Fortaleza, and 85.1% lived in houses.

Nineteen patients (11.8%) were positive (≥ 5 mm) for the skin test. All of these patients were men, with a mean age of 36 years. Twelve (63.1%) had an average monthly income of US$ 244.11 and had completed approximately 10.5 years of school. Most lived in downtown Fortaleza (three cases) and in two areas in the periphery of the city (two patients each). Seventeen of these patients (89.4%) were born in Ceará State. The distribution of the skin test results according to patient CD4 values is shown in Table 2. Although all participants have CD4 counts ≥ 350 cell/mm ³ , 53.41% of the studied patients have already met the Brazilian Ministry of Health criteria of AIDS.

There was no significant difference in the average time since HIV diagnosis between patients with positive (40.4 months) and negative (28.0 months) histoplasmin tests, as well as in the average time since AIDS diagnosis (36.8 months versus 40.7 months for positive and negative results respectively). The mean CD4 counts were also not different between the positive (590.4 cell/mm ³ ) and negative (577.5 cell/mm ³ ) groups. Nine (47.3%) of the histoplasmin-positive participants used highly active antiretroviral therapy. There was no significant difference (p = 1.0) in patients reactive (52.63%) and no reactive (53.52%) to HMIN considering the AIDS diagnosis.

The presence of mango trees (M. indica) in the current neighborhood (p = 0.040), activity involving soil in the past (p = 0.045), and having visited a farm in the past (p = 0.033) were significantly associated with HMIN reactivity (Table 3). The cleaning of the attic lining of the home or a chicken house in the past or present was not significantly correlated with histoplasmin status.

Of the sera tested with the double immunodiffusion tests (76), only two showed M-line precipitation, and these sera were obtained from histoplasmin-positive patients (2.63%). No cross-reaction was observed in the 36 sera tested with the commercial C. immitis antigen and antibody.

DISCUSSION

The vast majority of immunosuppressed patients with disseminated histoplasmosis have AIDS ¹⁶ . The reactivation of a latent focus of infection is considered to be a possible mechanism of disease manifestation in patients from endemic areas ²¹ . Therefore, studies on the prevalence of histoplasmosis infection in specific areas are important to understand the local epidemiology and risk factors associated with the acquisition of Histoplasma.

Most infected patients in this series were men; consistent with the results of a histoplasmin prevalence study (95.1%) performed by McKINSEY et al. (1997) in AIDS patients in the United States ²¹ . Considering the paucity of studies on HIV positive patients, comparisons will be performed mostly with histoplasmin intradermal tests performed on the general population.

A survey performed in three cities in Paraiba State (Northeastern Brazil) found histoplasmin reactivity in 18.5% to 31.5% of 1,957 tests applied in participants from two to > 60 years old ⁵ . Another serological survey by FAVA & FAVA NETTO (1998) carried out in two localities of Bahia (Northeast Brazil) and 29 Southeast areas of the country revealed histoplasmin reactivity in 13% to 19.6% and 3.0% to 93.2%, respectively ¹² ; the highest positivity was recorded in the city of Angra dos Reis on the coast of Rio de Janeiro State. In the southwestern Amazon, a study conducted in three native communities (individuals aged one to > 40 years old) observed positive histoplasmin test rates that varied from 5.8% to 80.5% ⁴ . Another study in native populations from the reservation of Xacriabá in the state of Minas Gerais in Southeast Brazil found a reactivity of 3.9% in 180 participants ¹⁹ .

Studies performed in Ceará by DIOGENES et al. in the early 1990s ⁹ found a high prevalence of histoplasmin positivity in Serra do Pereiro in the south of the state (61.5%). Similarly, FAÇANHA et al. (1991) found a 38.93% reactivity in residents from a mountainous area in the municipality of Palmácia ¹¹ . In 1983, another survey, this time with 138 patients from a rural area admitted to the Walter Cantídio University Hospital, showed a histoplasmin positivity rate of 23.6% ³ .

The prevalence (11.8%) observed in the current series was lower than that observed in a study ²¹ conducted on AIDS patients residing in histoplasmosis-endemic areas of the United States (16.7% in a total of 274 patients). This difference, however, was not large, suggesting that Ceará is an area with significant Histoplasma transmission.

The AIDS epidemic in Brazil affects mainly young adults ²² , which explains the lower average age observed in this study. In the study by PONTES et al., the average age of the 134 hospitalized patients with HD and AIDS in a referral hospital for infectious diseases was 35.8 years.

Based on information from the Ministry of Health of Brazil, only 7.7% of AIDS patients have 12 or more years of education ²² . The current study found that 61.49% of patients had 12 or more years of education. This result may be explained by one of the inclusion criteria, which selected cases with higher CD4 counts. Higher CD4 counts are associated with patients with better HAART adherence and access to health care facilities, which in turn is associated with better education ¹⁰ . HAART use and time since AIDS diagnosis were not associated with intradermal test response; a similar result was found in a previous study ²¹ .

Most of the patients enrolled in the current study lived in houses, which could have facilitated fungal infection due to the increased prevalence of tree cultivation (fruit trees or landscaping) or soil activities (cleaning and maintenance). In a retrospective series of 30 patients with deep systemic mycosis living in southern Brazil (2005-2010) analyzed by FREY et al., 26.3% of patients had reportedly performed some type of activity involving soil (farmer or gardener) ¹⁴ . The importance of M. indica trees reported in the current study was most likely due to its status as a bat habitat; these animals enjoy its fruit and are attracted by their presence. Risk variables such as living or working on a farm or direct interactions with birds were not mentioned by any of the patients, possibly because the individuals were from an urban area.

More than 50% of the samples studied already had the AIDS diagnosis. The low prevalence of Histoplasma infection found in this study compared with other surveys in Brazil might be explained by the low immune system response of the HIV patients; it is possible that although good levels of CD4 cells were observed in this study, the CD4 cells may not be functionally competent.

Only two patients were positive (prevalence 2.6%) for ID in the current study, and they also reacted to the intradermal histoplasmin test. It is not possible to determine whether or not the antibodies detected at the time of blood collection, resulted from previous illness with H. capsulatum. The patients in the current survey had their medical records investigated, and there were no reports (complaints, symptoms or signs) suggestive of histoplasmosis.

KUBERSKI et al. ¹⁷ reported that patients with Coccidioides may have a false-positive reaction to HMIN ID tests due to cross-reactivity, especially in areas where both fungi are endemic. SIDRIM & OLIVEIRA ²⁵ reported similar results for patients with paracoccidioidomycosis. Because almost 30% of the patients in this study were born in the countryside, and Ceará State is an endemic area for coccidioidomycoses, it was necessary to evaluate the patient sera for possible cross-reactivity to Coccidioides; however, none of the sera tested were positive for coccidioidin.

It can be concluded that patients with HIV/AIDS living in Fortaleza have an expressive prevalence of infection with Histoplasma. This study also identified several risk factors associated with Histoplasma infection, which will be useful in the implementation of preventive measures for patients who are susceptible to severe forms of this disease.