Histoplasmosis and Paracoccidioidomycosis in a Non‐Endemic Area: A Review of Cases and Diagnosis

Abstract

Background

Histoplasmosis and paracoccidioidomycosis (PCM) have increased in Spain in recent years, due firstly to the migration from endemic regions and secondly to travelers returning from these regions. In non‐endemic areas, diagnosis of both diseases is hampered by the lack of experience, long silent periods, and the resemblance to other diseases such as tuberculosis and sarcoidosis.

Methods

A total of 39 cases of imported histoplasmosis and 6 cases of PCM diagnosed in the Spanish Mycology Reference Laboratory since 2006 were analyzed. Microbiological diagnosis was performed using classical methods and also a specific real‐time polymerase chain reaction (RT‐PCR) assay for each microorganism.

Results

We had 9 cases of probable histoplasmosis in travelers and 30 cases in immigrants, 29 of whom were defined as proven. Paracoccidioidomycosis (PCM) cases were either immigrants or people who had lived for a long period of time in endemic regions, all of whom were classified as proven cases. Cultures showed a good sensitivity in detecting Histoplasma capsulatum in immigrants with proven histoplasmosis (73%); however, growth was very slow. The fungus was never recovered in traveler patients. Paracoccidioides brasiliensis was isolated in a culture only in one case of the proven PCM. Serological methods were not very reliable in immunocompromized patients with histoplasmosis (40%). A PCR‐based technique for histoplasmosis detected 55.5% of the cases in travelers (probable cases) and 89% of the cases in immigrants (proven). The PCR method for PCM detected 100% of the cases.

Conclusions

These kinds of mycoses are increasingly frequent in non‐endemic areas, and newer and faster techniques should be used to reach an early diagnosis. The RT‐PCR techniques developed appear to be sensitive, specific, and fast and could be helpful to detect those mycoses. However, it is also essential that physicians perform differential diagnosis in individuals coming from endemic areas.

Endemic mycoses have risen in recent years in Spain due to both the increase in the immigrant population from endemic areas and travelers returning from these regions. At present, the immigrant population from South America constitutes 38% of the total (http://www.ine.es), and 1 million Spaniards visit tropical or equatorial areas every year. The number of diagnosis requests for these mycoses in our laboratory has increased seven times in 10 years (data not shown). Histoplasmosis is the most frequently reported endemic mycosis in Europe. ¹ In Spain, several cases of histoplasmosis have been described in travelers returning from endemic areas. ^2–5 Most cases occur in small clusters with a common source of infection. The individuals affected have a history of involvement in leisure and/or work activities. ² In immunocompetent hosts, diagnosis of histoplasmosis is difficult because of its nonspecific clinical manifestations. ⁴

More frequently, disseminated histoplasmosis has been described in human immunodeficiency virus (HIV)+ patients from Latin–American countries. ⁶ The clinical symptoms of disseminated histoplasmosis in HIV+ patients can imitate those of Pneumocystis jirovecii pneumonia, tuberculosis, and other fungal infections. Clinical suspicion and rapid detection are essential to reach a diagnosis, and to initiate the appropriate antifungal therapy. When untreated, the mortality rate in those patients is close to 100%. ⁷

Recently, Norman and colleagues ⁸ reported clinical and epidemiological data on 10 cases of imported histoplasmosis in Spain, showing the two distinct above‐mentioned profiles in travelers and immigrants.

Several cases of paracoccidioidomycosis (PCM) have been described in recent years in Europe ^9–11 associated with populations from endemic regions and travelers. The prevalence of PCM in HIV‐infected population is lower than that of histoplasmosis and has been estimated at 1.4% to 1.5% in some regions of Brazil. ^12,13 In imported cases, chronic multifocal PCM, which had been acquired many years earlier, is usually detected. The period of latency in cases diagnosed in Spain was long, varying between 10 and 25 years. ⁹

Both mycoses are difficult to diagnose outside endemic regions. Isolating the organism from cultures is considered the reference procedure; however, these fungi are fastidious and slow‐growth organisms, requiring 3 to 4 weeks of incubation. Sensitivity and specificity of microscopic examination of fluids and tissues are too low. Limitations of serological techniques are also significant, as serology is negative in up to 50% of immunosuppressed patients suffering from histoplasmosis, especially those with acquired immunodeficiency syndrome (AIDS), ⁷ and the test for antigen detection in urine and/or serum ¹⁴ is not accessible in the majority of non‐endemic areas. Several conventional PCR assays have been described to detect Histoplasma capsulatum DNA ^7,15–18 targeted to different genes. In recent years, quantitative PCR assays such as real‐time PCR (RT‐PCR) have been proposed for the diagnosis of H capsulatum, firstly because of their greater sensitivity, specificity, and shorter time to diagnosis compared to conventional PCR, and secondly because they avoid the need to handle the fungi. ^19–22

There are a number of techniques for detecting both specific antibodies and antigens of Paracoccidioides brasiliensis. Antibody detection methods have the problem of cross reactivity with other primary pathogenic fungi and have very variable sensitivity. Regarding antigen detection, the circulating antigen, gp43, is the one mainly used for diagnosis, ²³ although this antigen disappears from the circulation during the treatment. ²⁴ Methods based on the detection of nucleic acids have also been described. ^25,26

This review analyzes the epidemiology and diagnosis methods used in 39 cases of imported histoplasmosis and 6 cases of PCM diagnosed in the Spanish Mycology Reference Laboratory since 2006. The work focuses on the diagnostic usefulness of new molecular techniques of DNA detection developed by the Reference Laboratory and their implications for the clinical practice.

Materials and Methods

Criteria for Classification of Cases

The cases were classified following the EORTC/MSG Consensus Group criteria (European Organization for Research and Treatment of Cancer/Mycosis Study Group). ²⁷ Proven histoplasmosis or PCM was considered when the fungus was recovered in culture from a specimen or when the microorganism was observed using histopathology or direct microscopy. Cases were classified as probable when a consistent clinical picture was found and we had a positive result in an immunodiffusion test (ID Fungal Antibody System, Immuno‐Mycologics Inc, Norman, OK, USA). All cases except one fulfilled the proven or probable criterion. In Patient 11, there were only clinical suspicion and positive results by RT‐PCR (Table 2). This case was classified as possible. Using epidemiological criteria, we also classified the cases as either travelers or immigrants and people who had lived in an endemic region for a long period of time.

Table 2

Epidemiological characteristics of immigrants with histoplasmosis, results of diagnostic test performed and classification of cases based in EORTC/MSG criteria

Morphological Identification

In case of H capsulatum strains, mycelia were stained with lactophenol cotton blue dye (Difco, Soria‐Melguizo, Madrid, Spain). Characteristic macroconidia were observed microscopically.

Extraction of DNA from Clinical Strains and Clinical Samples

Extraction of nucleic acids from clinical strains was undertaken in Biosafety Level III facilities and in compliance with Spanish Laws (Royal Decree 664/1997). DNA extraction from strains and clinical samples was performed as described by Buitrago and colleagues. ²⁰

Molecular Identification of Clinical Strains

DNA extracted from clinical strains was used to amplify the internal transcriber spacer (ITS) region. ²⁸ Sequence analysis of amplified fragments was performed by comparing the DNA sequences with the ITS sequences of H capsulatum var. duboisii (ATCC 24295), H capsulatum var. capsulatum (CBS207.55 and CBS214.53), and P brasiliensis (ATCC32069 and ATCC60855) obtained from the GenBank database (http://www.ncbi.nlm.nih.gov/Genbank/).

Use of RT‐PCR Technique in Clinical Samples

RT‐PCR for the detection of H capsulatum was carried out following the protocol described by Buitrago and colleagues. ¹⁹ Primers and probes were designed on the basis of the nucleotide sequence of the ITS1 rDNA region. Probes were marked using fluoresce resonance energy transfer (FRET) technology, and the PCR reactions were performed in the Lightcycler 480 (Roche Applied Science, Madrid, Spain). An internal control was included in the RT‐PCR reaction following the Brugraff method. ^20,29 RT‐PCR for the detection of P brasiliensis DNA was performed as described by Buitrago and colleagues. ²⁵

Technique of Immunodiffusion

Detection of precipitating antibodies in patient's sera was performed by an immunodiffusion test following manufacturer's recommendations (ID Fungal Antibody System). This commercial test uses the antigens M and H against histoplasmosis sera and antigen gp43 against PCM sera.

Results

A total of 39 cases of histoplasmosis and 6 cases of PCM have been diagnosed in the Spanish Mycology Reference Laboratory in the last 3 years. The cases of histoplasmosis were classified on the basis of epidemiological data as histoplasmosis either in travelers, immigrants or in people who had lived in an endemic region for a long period of time. These cases were also classified as probable in nine cases, proven in 29 cases, and possible in one case. All cases of PCM were classified as proven. The patients were alive at the time of diagnosis using PCR with the exception of patient 21 whose samples were obtained post mortem.

Demographic and Epidemiological Characteristics

Histoplasmosis

The demographic characteristics and underlying conditions are summarized in Table 1. The median age of patients was 34 (22–54) although age was not reported in five cases. There was a total of 21 males (54%) and 18 females (46%). The majority came from South‐American countries (30/36) (83%), particularly Ecuador (42%), five patients came from African countries (14%) and one patient had visited several African and South‐American countries. The origin was not reported in four cases (Table 2). Only nine patients were travelers returning from endemic regions, without underlying diseases (23%) (Table 3). The remaining patients were immigrants and people who had lived in these regions for a long time (77%) (Table 2) and had underlying conditions, in most cases HIV+ (97%). One of them was an oncohematological case.

Table 1

Demographic and clinical characteristics of 39 patients with proven or probable histoplasmosis

Characteristics	Number of patients (%)
Age, years
Mean	35.6
Median	34
Range	22–54
Gender
Male	21/39 (54%)
Travelers returning from endemic regions
Total number	9/39 (23%)
Underlying condition
None	9/9 (100%)
Origin or long term permanence in endemic regions
Total number	30/39 (77%)
Underlying condition
AIDS	29/30 (96.6%)
Malignancy	1/30 (3.3%)

AIDS = acquired immunodeficiency syndrome.

Table 3

Epidemiological characteristics of travelers with histoplasmosis, results of diagnostic test performed and classification of cases based in EORTC/MSG criteria

Patient	Age	Sex	Traveled to	Samples	PCR result	ID result	Activity	Culture/HP	Classification	Clinical category
1*	—	Female	Ecuador	Serum	Negative	Positive	Tourism	Negative	Probable	Acute with rheumatologic manifestations
2*	—	Female	Ecuador	Serum	Negative	Positive	Tourism	Negative	Probable	Acute with rheumatologic manifestations
3*	—	Female	Ecuador	Serum	Positive	Positive	Tourism	Negative	Probable	Acute with rheumatologic manifestations
4*	—	Female	Ecuador	Serum	Positive	Positive	Tourism	Negative	Probable	Acute with rheumatologic manifestations
5	53	Male	Angola/Brasil/Peru	Serum	Negative	Positive	Physician	Negative	Probable	No clinical data
6	—	Female	Venezuela	Serum	Positive	Positive	Spelunker	Negative	Probable	Acute pulmonary
7	54	Male	Africa	Serum	ND	Positive†	Volunteer	Negative	Probable	Subacute pulmonary
				Blood	Negative	ND
				LB	Positive	ND
				Sputum	Positive	ND
8	54	Male	Nicaragua	Blood	Negative	Positive	Tourism	Negative	Probable	No clinical data
9	46	Male	Ecuador	Serum	Negative	Positive	Tourism	Negative	Probable	Acute pulmonary
				LB	Positive	ND
				Blood	Negative	ND

PCR = polymerase chain reaction; ID immunodiffusion test; HP = histopathology; ND = not done; LB = lung biopsy.

^*Described by Alonso and colleagues. ²

^†Weak positive.

Paracoccidioidomicosis

The median age of patients with PCM was 51 with a range from 31 to 67. Age was not reported in one case. All were males (100%) with the precedent of having stayed in South‐American countries. Four were immigrants and two were Spaniards who had lived long term in these areas (patients 2 and 4). No immunosuppression was reported in any case (Tables 4 and 5). The diagnosis was delayed because of the lack of clear symptoms in all the cases. ²⁵ In addition, the diagnosis was wrong in two of them, clinical manifestations suggested sarcoidosis in one case and histoplasmosis in the other.

Table 4

Demographic and clinical characteristics of six patients with proven paracoccidioidomycosis

Characteristics	Number of patients (%)
Age, years
Mean	50
Median	51
Range	31–67
Gender
Male	6/6 (100%)
Female	0
Underlying condition
AIDS	No
Other	No

AIDS = acquired immunodeficiency syndrome.

Table 5

Epidemiological characteristics of PCM patients, results of the different diagnostics test employed and classification of cases based in EORTC/MSG criteria

Patient	Age	Sex	Origin	Samples	PCR results	ID results	Culture/HP	Classification (method)	Clinical category
1*	67	Male	Ecuador	Serum	Negative	Light Positive	HP	Proven (DV)	CAF
				Blood	Positive	ND
				Sputum	Positive	ND
				Skin Biopsy	Positive	ND
2*	57	Male	Venezuela†	Serum	Negative	Positive	HP	Proven (DV)	CAF
				Blood	Negative	ND
				Sputum	Positive	ND
3	44	Male	Paraguay	Serum	Negative	Light Positive	HP	Proven (DV)	CAF
				PCB	Positive	ND
				FCB	Positive	ND
4	51	Male	Paraguay†	BAL	Positive	ND	HP	Proven (DV)	CAF
				Blood	Positive	ND
				Serum	Negative	Positive
5	31	Male	Ecuador	Lung Punction	Positive	ND	Culture	Proven (culture)	CAF
6	—	Male	—	OMB	Positive	ND	HP	Proven (DV)	CAF
				Serum	ND	Light Positive

PCR = polymerase chain reaction; ID = immunodiffusion test; HP = histopathology; ND = not done; DV = direct visualization; CAF = chronic adult form; PCB = paraffined cerebral biopsy; FCB = fresh cerebral biopsy; BAL = bronchoalveolar lavage; OMB = oral mucosa biopsy.

^*Described by Buitrago and colleagues. ²⁵

^†Spanish patients with a long stay in those countries.

Diagnosis of Histoplasmosis in Travelers

We had nine cases of histoplasmosis in travelers, all with a history of travel to an endemic area and a clinical picture consistent with histoplasmosis. We had a cluster of four females who had visited rural areas in Ecuador ² (patients 1–4, Table 3); a physician who had traveled through African and South‐American countries (patient 5, Table 3); one person who had visited a cave in Venezuela (patient 6, Table 3); a volunteer who had returned from Africa (patient 7, Table 3); and two other tourists traveling through rural areas in South‐America (patients 8 and 9, Table 3). All cases were defined as probable, and the microbiological diagnosis was based on the serological test. The immunodiffusion test was positive in all the cases and the RT‐PCR in five patients (5/9, 55.5%). The PCR technique was performed on seven sera, three blood samples, two lung biopsies, and on one sputum sample. By samples, the technique was positive in 43% of the sera (3/7) and 100% of biopsies and respiratory samples (3/3). No positive results were obtained in the three blood samples tested. The fungus was not isolated in any case.

Diagnosis of Histoplasmosis in Immigrants

We had 30 patients native to endemic regions or living there long term. All but one were immigrants with AIDS as underlying condition (97%). One patient was an oncohematological patient (Table 2, patient 11) and was classified as a possible case. The other 29 cases were classified as proven (97%). The culture was positive in 73% of patients (22 cases) but always several weeks after the onset of symptoms. In seven cases (23%) the fungi was not cultured and the yeast cells were visualized in the tissues. The immunodiffusion test was performed in sera from 20 patients and was positive in only eight patients (40%). RT‐PCR was performed in samples from 27 patients and was positive in 24 patients, showing a sensitivity of 89%. By samples, RT‐PCR was performed on 54 samples from these patients: 16 sera, 10 respiratory samples, 8 blood samples, 6 biopsies, 6 bone marrow samples, 4 plasma samples, 3 lymph node biopsies, and 1 cerebrospinal fluid. The RT‐PCR was positive in 11 sera (69%), 10 respiratory samples (100%), 3 blood samples (37.5%), 6 biopsies (100%), 4 bone marrow samples (67%), three plasma samples (75%), and two lymph nodes (67%). Results were obtained within 24 hours of receiving the samples.

When the fungus had been cultured, DNA was extracted from mycelia to perform PCR amplification and sequencing of ITS regions. All sequences matched with H capsulatum. We obtained the variety duboisii in three patients from African countries (Table 2; patients 7, 29, and 30).

Diagnosis of Paracoccidioidomicosis

We had six patients with proven PCM. The fungus was cultured only in one patient several weeks after receiving the sample (CNM‐CM5413). In the other cases characteristic budding yeasts were observed in clinical samples.

The immunodiffusion test was performed in sera from five patients and was positive in all cases (100%), although the signal was very weak in three of them (60%).

RT‐PCR was performed on samples from these six patients and was positive in all cases (100%). By samples, RT‐PCR was performed on four tissue biopsies, four serum samples, three blood samples, two sputum samples, one bronchoalveolar lavage (BAL), and one lung biopsy. RT‐PCR was positive in two blood samples (66%), two sputum samples (100%), four biopsies (100%), one BAL (100%), and one lung biopsy (100%). The RT‐PCR results were also obtained 24 hours after receiving the samples. DNA was extracted from the isolated strain (CNM‐CM5413) to perform a PCR amplification of the ITS region, followed by sequencing. The sequence matched with P brasiliensis. In two patients, we tested samples several weeks after starting the antifungal therapy, showing that the amount of DNA had either decreased or disappeared. ²⁵

Discussion and Conclusion

Diagnosis of histoplasmosis and PCM is very frequently hampered by a lack of experience in non‐endemic areas. Histoplasmosis in travelers should be suspected when a self‐limited benign pulmonary illness with fever, headache, malaise, dry cough, and chest pain is found ² and the patient has been involved in outdoor activities or has visited caves. Serology is useful since this kind of patient has not had any previous contact with the fungus. All traveler patients diagnosed in our laboratory had a positive immunodiffusion test. RT‐PCR was positive in only five of the nine patients studied, probably due to the limited amount of DNA circulating in immunocompetent patients. Respiratory samples provided better results than sera or blood samples. For most patients, only sera samples were available for reaching diagnosis, a fact which could explain the low sensitivity of RT‐PCR in the case of travelers. More studies should be performed on this kind of patient. Finally, the fungi were never cultured.

In immigrant cases, we found mainly disseminated histoplasmosis in immunosuppressed patients. Histoplasmosis occurred as a result of the reactivation of a latent focus of infection acquired years earlier. ³⁰ A total of 29 out of 30 immigrants had AIDS as an underlying disease. This figure matches previously reported studies. ³¹ Patients with disseminated histoplasmosis present fever, weight loss, anorexia, cough, vomiting, diarrhea, and abdominal pain. ⁶ Only 8 patients out of 20 had a positive result in a serological test. In 73% (22/30) of cases the fungus was isolated. Cultures showed good sensitivity in detecting H capsulatum; however, the average time needed to obtain positive cultures was 15 days. RT‐PCR showed good sensitivity (89%). The technique was performed in 27 patients and was positive in 24. Respiratory samples and biopsies were the most useful samples, with 100% sensitivity. Blood samples appeared to have lower sensitivity than sera samples (37.5% vs 69%); however, we obtained a positive result for sera sample and a negative result for blood only in patients 15 and 11 (Table 4).

In these cases there may be a partial inhibition which was reflected in a slightly lower melting curve for the internal control. In the other cases, sera and blood samples were either both negative (Table 3, patient 9; and Table 4, patients 1, 18, and 20) or both positive (Table 2, patients 19 and 21). These results may correlate with the clinical status of each patient. More blood samples should therefore be analyzed to reach a conclusion.

PCM in non‐endemic areas is rarely suspected because of the extremely long silent period of this disease. ⁹ Diagnosis was delayed in four of the six cases diagnosed in our laboratory; we have no data on the other two cases. In all cases described in this paper characteristic yeasts were visualized at the hospitals. The fungus was cultured in only one case (patient 5) and growth was very slow. Serology proved to be useful since it was positive in all patients. RT‐PCR showed good sensitivity as we obtained positive results for all patients. Respiratory and biopsy samples proved more suitable than blood samples.

In conclusion, these kinds of mycoses are increasingly frequent in non‐endemic areas, ^5,8 and newer and faster techniques should be used to reach an early diagnosis. The RT‐PCR techniques developed appear to be sensitive, specific, and fast and could be helpful to detect those mycoses. However, it is also essential that physicians consider histoplasmosis and PCM in individuals coming from endemic areas and that they perform differential diagnosis.

We are grateful to the Spanish National Health Hospitals listed below which have contributed by sending samples and data on their patients: Hospital Carlos III (Madrid), Hospital Clinico San Carlos (Madrid), Hospital Comarcal de Orihuela‐Vega Baja (Orihuela, Alicante), Hospital Donostia (San Sebastian), Hospital General de Asturias (Oviedo), Hospital General de Lanzarote (Lanzarote), Hospital General Universitario Gregorio Marañón (Madrid), Hospital General La Mancha Centro (Alcazar de San Juan, Ciudad Real), Hospital General Universitario Morales Meseguer, Hospital de Hellín (Hellín, Albacete), Hospital Marina Baixa (Villajoyosa, Alicante), Hospital do Meixoeiro (Vigo, Pontevedra), Hospital Mutua de Terrassa (Terrassa, Barcelona), Hospital Universitario Carlos Haya (Málaga), Hospital Universitario Clinic de Barcelona (Barcelona), Hospital Universitario Doce de Octubre (Madrid), Hospital Universitari La Fe de Valencia (Valencia), Hospital Universitario Miguel Servet (Zaragoza), Hospital Universitario de Mostoles (Mostoles, Madrid), Hospital Universitario Principe de Asturias (Alcala de Henares, Madrid), Hospital Universitario Ramon y Cajal (Madrid), Hospital Universitario Son Dureta (Mallorca), Hospital Universitario Virgen de la Arrixaca (Murcia), Hospital Universitario Virgen de la Macarena (Sevilla), Hospital Vall d’Hebron (Barcelona), Hospital Virgen del Camino (Pamplona), and Hospital Virgen de la Salud (Toledo).

L. B.‐M. has a research contract from REIPI (Red Española de Investigación en Patología Infecciosa, Project MPY 1022/07_1)

Declaration of Interests

The authors state that they have no conflicts of interest to declare.