Histoplasmosis Diagnosed in Europe and Israel: A Case Report and Systematic Review of the Literature from 2005 to 2020

Human histoplasmosis is a mycosis caused by two distinct varieties of a dimorphic fungus: Histoplasma capsulatum var. capsulatum and H. capsulatum var. duboisii. In Europe, it is usually imported by migrants and travellers, although there have been some autochthonous cases, especially in Italy; however, most European physicians are unfamiliar with its clinical and pathological picture, particularly among immunocompromised patients without HIV infection. This systematic review of all the cases of histoplasmosis reported in Europe and Israel between 2005 and 2020 identified 728 cases diagnosed in 17 European countries and Israel described in 133 articles. The vast majority were imported (mainly from Central and South America), but there were also seven autochthonous cases (six in Europe and one in Israel). The patients were prevalently males (60.4%), and their ages ranged from 2 to 86 years. The time between leaving an endemic region and the diagnosis of histoplasmosis varied from a few weeks to more than 40 years. Progressive disseminated histoplasmosis was the most frequent clinical picture among people living with HIV infection (89.5%) or a different immunocompromising condition (57.1%), but it was also recorded in 6.2% of immunocompetent patients. Twenty-eight cases were caused by Histoplasma duboisii. Immunocompromised patients without HIV infection had the worst outcomes, with a mortality rate of 32%.


Introduction
Human histoplasmosis is a disease that is traditionally considered to be caused by two varieties of the fungus Histoplasma capsulatum var. capsulatum (Hcc) and H. capsulatum var. duboisii (Hcd, also known as African histoplasmosis) [1,2]. Although the number of species belonging to the Histoplasma genus is unknown, the findings of genome-wide population genetic and phylogenetic analyses suggest the existence of four species that are endemic in the Americas: Histoplasma capsulatum sensu stricto, H. mississippiense sp. nov (formerly known as NAm1), H. ohiense sp. nov (formerly known as NAm2), and H. suramericanum sp. nov (formerly known as LAm A) [3]. A fifth lineage, the African clade (H. capsulatum var. duboisii), is probably a separate species, but confirmation of its status will require further analyses of more samples [3].
The disease is highly endemic in the USA and Latin America [4] but is probably more widespread than previously thought [5,6]; a survey by the European Confederation of Medical Mycology Working Group has described the cases diagnosed in Europe between January 1995 and December 1999 [7], and Antinori et al. have reviewed cases observed in patients living with HIV infection in Europe between 1984 and 2004 [8]. Other publications include a systematic review of cases of acute histoplasmosis in immunocompetent travellers worldwide [9] and a review of the worldwide cases of African histoplasmosis diagnosed between 1993 and 2019 [10].
The aim of this systematic review is to describe the imported and autochthonous cases of histoplasmosis diagnosed in Europe and Israel between 2005 and 2020 as well as a new case of progressive disseminated histoplasmosis in a Brazilian trans-sexual male living with HIV/AIDS.

Materials and Methods
This systematic review was carried out in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) statement [11].

Eligibility Criteria and Study Selection
The populations considered were children and adults of any age and origin diagnosed with histoplasmosis in geographical Europe and Israel. All publications were considered regardless of their type (individual case reports, case series, observational studies, and reviews providing information regarding age, gender, nationality, travel history, affected organs, latency period, diagnosis, antifungal therapy, and outcome) or language (excluding Turkish, Finnish, Romanian, and Russian). The cases included travel-related, possible autochthonous, and veterinary cases diagnosed in the same geographical areas.
The patients were divided into three categories based on their immune status: people living with HIV (PLWH); patients with other immunocompromising conditions (OIC: solid organ or stem cell transplant recipients; cancer patients; patients treated with immunosuppressive drugs including corticosteroids, biological immunosuppressants, and non-biological, non-corticosteroidal drugs); and immunocompetent subjects (HIV-negative patients not meeting any of the OIC criteria).

Data Collection and Evidence Summary
The following information was extracted from each article and entered into pilot-tested evidence tables: author, year, study design, language, country of diagnosis, country of exposure, latency period, number of cases, patients' characteristics (age, gender, occupation, and affected organs), systemic antifungal therapy, and outcomes.

Case Report
A 27 year old trans-sexual male born in Brazil was admitted to our infectious diseases ward in April 2012 with a 2-month history of intermittent high fever, productive cough, dysphagia, and diarrhoea. He had been concomitantly diagnosed with human immunodeficiency virus-1 (HIV-1) infection and disseminated tuberculosis in February 2006, after which he underwent a successful 12-month course of treatment for tuberculosis and antiretroviral therapy (ART: initially tenofovir/emtricitabine plus efavirenz followed by tenofovir/emtricitabine plus atazanavir/ritonavir), which led to a good recovery of CD4+ lymphocytes from 39/µL to 442/µL and controlled HIV-1 viremia (HIV RNA copy numbers went from 500,000 to <50/mL) until February 2011, when he was lost to follow-up.
A physical examination at the time of admission revealed a high fever (39.7 • C), cervical lymphadenopathy, hepatosplenomegaly, and oral candidiasis. His white blood cell count was 2530/µL (85% neutrophils, 8% lymphocytes), haemoglobin level 8.8 g/dL and platelet level 63,000/µL. He had a CD4+ cell count of 8/µL, an HIV-1 viral load of 30,780 copies/mL, and a normal lactate dehydrogenase level of 247 IU/L. He was initially treated with amoxicillin/clavulanate because of suspected pneumonia revealed by chest X-ray and started ART with cotrimoxazole prophylaxis. He refused a bone marrow biopsy and, after defervescence for 6 days, developed hypotension, a high fever (up to 40 • C), and dyspnoea. Whole-body computed tomography (CT) revealed generalised bilateral cervical lymphadenopathy (nodes 1-3 cm in diameter), retroperitoneal lymph nodes, hepatosplenomegaly, and bilateral interstitial/alveolar infiltrates in both lungs. Following unsuccessful treatment with helmet continuous positive airway pressure (CPAP), the patient was transferred to our intensive care unit (ICU) and underwent orotracheal intubation. As his clinical picture aroused the suspicion of multicentric Castleman disease accompanied by high-level of human herpes virus 8 (HHV-8) viremia (3 × 10 5 copies/µL), he underwent a lymph node biopsy and started treatment with intravenous etoposide (150 mg). Histopathological examination of the lymph nodes showed a total replacement of follicular and paracortical areas by a histiocytic infiltrate and PAS/Grocott-positive fungal cells compatible with Histoplasma spp. (Figure 1). Pan-fungal polymerase chain reaction (PCR) followed by sequencing of the amplified products confirmed the diagnosis of Histoplasma capsulatum infection. Serum galactomannan (GM) antigen (Platelia Aspergillus EIA, BioRad, Marnes la Coquette, France) was positive (OD index 2.73), and the patient was started on liposomal amphotericin B (L-AMB) at 3 mg/kg/day. After 12 days, he was extubated and returned to our ward. He completed the 2-week course of L-AMB and was started on itraconazole 200 mg every 12 h. Before discharge, he was administered etoposide again because of the reappearance of fever associated with a new increase in HHV-8 viremia levels, and ART was resumed with tenofovir/emtricitabine plus raltegravir and oral etoposide every week. In the following months, he returned to Brazil and was lost to follow-up.
The patient was re-hospitalised because of fever, abdominal pain, and cough in November 2012, when he said he had discontinued itraconazole for 3 months and ART for 1 week. Blood cultures were negative, and whole-body CT revealed enlarged laterocervical, retroperitoneal, and inguinal lymph nodes. HHV-8 viremia was negative, HIV RNA was 6,707 copies/mL, and his CD4+ cell count was 81/µL. On the basis of GM antigen positivity (OD index 2.14), he was again treated with L-AMB, which led to the rapid disappearance of fever and abdominal pain and a reduction in his GM antigen level (OD index 1.13). He was discharged with a prescription of itraconazole 200 mg twice daily, cotrimoxazole prophylaxis, and ART.
He was hospitalised again in March 2013 because of a 2-week history of fever, chills, cough, and vomiting. A chest X-ray revealed an interstitial picture with mediastinal enlargement; GM antigen was negative (<0.5), and a Ziehl-Neelsen sputum test and PCR indicated Mycobacterium tuberculosis. Antitubercular treatment led to a good clinical response, and the patient was definitely lost to follow-up in June 2013.

Summary of the Literature
The literature search identified 133 articles about cases of human and veterinary histoplasmosis from 17 European countries and Israel ( Figure 2).

Study Selection and Characteristics
As shown in Table 1, five studies involving 505 subjects (295/436 males, 67.7%) described cumulative experiences of histoplasmosis: two from Spain (one regarding imported endemic mycoses from 1997 to 2014 [12], the other relating to histoplasmosis in Spanish travellers to Latin America [13]), a retrospective study from metropolitan France reporting all the cases of histoplasmosis in PLWH diagnosed between 1985 and 2006 [14], and two about cases of acute histoplasmosis diagnosed in travellers in a single centre in Italy between 2005 and 2015 [15] and in Israel between 2000 and 2012 [16].
The case reports and small case series of histoplasmosis diagnosed between 2005 and 2020  involved a total of 223 patients (145 males, 65%) with a median age of 38 years (range 2-86).
A majority of patients were treated with one of the amphotericin B formulations followed by itraconazole or other azoles (47/76, 61.8%). Of the 78 patients whose outcomes were recorded, 19 (24.3%) died.

Patients with Other Immunocompromising Conditions (OIC)
There were 24 case reports and 4 case series of 28 patients with OIC and a diagnosis of histoplasmosis [20,44,. The patients' median age was 59 years (range , and the latency between the last time in an endemic area and diagnosis ranged from 3 weeks to 42 years. The most frequent underlying clinical conditions were solid organ transplantation and autoimmune diseases (51.9%) ( Table 4).

Immunocompetent Patients
A total of 81 immunocompetent subjects with a diagnosis of histoplasmosis were described in 36 case reports and 13 case series (5 of which also included PLWH) [23,33,38,40,. Their median age was 43 years (range , and there was a prevalence of males (57/79, 72.1%) ( Table 6). Median latency was 14 days in the case of travellers (range 2-120) and 6.5 years (range  in the case of migrants or expatriates. Six cases were considered to be autochthonous (three in Italy and one each in Spain, Ireland, and Israel) [105,106,117,121,126,131]; the Spanish case involved a laboratory technician who accidentally inoculated himself while handling a sample containing H. capsulatum [106]. The geographical areas in which the imported cases were acquired were mainly Central and South America (64.5%), followed by Africa (26.3%). Most cases (46.3%) were diagnosed in Spain, France, and Germany. There were six clusters of acute histoplasmosis involving 3-10 travellers to Latin America who had visited bat-infested caves [98][99][100]108,109] and the members of a multinational student expedition to a Ugandan rainforest in which a hollow bat-infested tree was the possible source of infection [115]. Most of the immunocompetent patients had pulmonary histoplasmosis (60/83, 73.2%), which was associated with rheumatological manifestations (arthralgia and/or erythema nodosum) in 23.3% of cases (Table 6). There were eight cases (13.3%) of single or multiple nodular lung lesions mimicking neoplasia or severe sarcoidosis [95,97,107,[119][120][121]128,133], three cases of chronic cavitary histoplasmosis [101,102,127], and one patient with fibrosing mediastinitis [130]. Five patients (6.2%) presented with a picture of PDH [103,116,123,125,126], and two showed central nervous system involvement [103,134]. Oropharyngeal ulcerations (tongue, uvula, gingiva, tonsils, hard palate) were described in five patients, and three had primary cutaneous histoplasmosis. Most of the cases were diagnosed by means of serology (44.4%) followed by histology (12.3%); one patient was diagnosed post mortem [126] ( Table 6). The most frequently used drug was itraconazole (60.8% of cases); 25.7% of the patients did not require any treatment. The outcome was apparently favourable in 94.4% of cases.

Histoplasma Capsulatum from Animals in Europe: The One Health Concept
There were seven cases of veterinary histoplasmosis: one in a dorcas gazelle (Gazella dorcas neglecta) living in captivity in Spain [137]; three in domestic cats in Italy (two cases) and Austria [138][139][140], one in a European hedgehog (Erinaceus europaeus) in Germany [141], one in a badger (Meles meles) in Germany [142], and one in a dog in southern Italy [143]. A necropsy study of free-ranging mustelids in Switzerland relating to 566 reports written between 1958 and 2015 documented 6 cases of histoplasmosis in 249 badgers (Meles meles) [144]. Except for the gazelle and the cat in Austria (imported from Texas), all the cases can be considered autochthonous.
The disease was disseminated in four cases (the gazelle, two cats, and hedgehog) [137,138,140,141], limited to the skin in one cat [139] and the badger in Germany [142] and involving the spinal cord in the dog [143].

Discussion
This systematic review of diagnoses of histoplasmosis made in Europe and Israel over 16 years identified 728 cases. Of those, 505 were described in five retrospective studies, including three studies of acute histoplasmosis in travellers diagnosed in Spain [13], Italy [15], and Israel [16]; one study of imported AIDS-related histoplasmosis conducted by the French National Reference Centre for Mycoses and Antifungals (NRCMA) [14]; and one summarising imported endemic mycoses (including histoplasmosis) in hospitalised patients in Spain [12]. The remaining 223 cases were retrieved from single case reports or small case series .
The first important message arising from the review is that only seven cases diagnosed in Europe (1%) [23,70,79,105,106,121,126] and one diagnosed in Israel [131] were not associated with any disclosed history of travel or residence in areas considered to be endemic for the disease and can, therefore, possibly be classified as autochthonous: four in Italy [23,79,105,126], two in Spain [70,106], and one in Ireland [121]. Italy has long been considered a country with endemic foci of histoplasmosis, and a number of autochthonous cases have been previously reported [7,8,[145][146][147][148]. Moreover, we retrieved two autochthonous veterinary cases involving a cat and a dog living in Italy, thus confirming that Italy should be considered a low-level endemic region [138,143]. One of the two Spanish cases involved a kidney transplant recipient, and although donor-derived infection cannot be definitely excluded, both the donor and the recipient were born in Spain and had no history of travel to endemic areas [70]; the other involved a laboratory technician who accidentally inoculated himself with a biological sample taken from an Ecuadorian man with a diagnosis of histoplasmosis [106]. A third autochthonous case (described before the beginning of our study period) occurred in an immunocompromised patient treated with azathioprine in Andalusia [149]. We are not aware of any previous autochthonous cases of histoplasmosis diagnosed in Ireland, and the Israeli report (which was the first to be published in the literature) involved a patient who lived near the cave of Yodfat where H. capsulatum was isolated from bats [131,150].
Most of the imported cases were acquired in Central and South America (120/219, 54.8%), with Ecuador being the most represented country of acquisition among PLWH (16/50, 32%) and immunocompetent subjects (12/50, 24%). Ghana and Ivory Coast (16/47, 34%) are the two African countries contributing the most patients with histoplasmosis associated with HIV/AIDS, although the NRCMA study of patients diagnosed with histoplasmosis in metropolitan France found that 39% of the cases had been acquired in French Guiana and French West Indies [14]. Among immunocompetent travellers acquiring histoplasmosis in Africa, the most represented country is Uganda, but this is due to a cluster of cases following a study field trip to a rainforest [115].
Of the cases of histoplasmosis diagnosed in Africa, 28 were due to H. capsulatum var. duboisii, 35% of which were acquired in the Democratic Republic of Congo (DRC) and Congo Brazzaville, which confirms recent review data indicating a 44% cumulative prevalence of histoplasmosis in both countries [10] as well as the findings of a recent study showing that Hcd is mainly responsible for histoplasmosis in the DRC [151]. Interestingly, Ghana contributed 15% of the cases included in our review but only one in the review cited above [10].
The fact that Spain (59.9%) and France (20.1%) together account for 80% of the European diagnoses of histoplasmosis is due to large-scale immigration from Latin America to Spain, from French overseas departments (i.e., French Guiana and French West Indies) to France, and from former colonial territories to both [12][13][14]. It is also interesting to note that more than half of the cases diagnosed in The Netherlands have links to Suriname and the Dutch Antilles.
We confirmed that the latency period from the last time spent in an endemic area to the time of diagnosis can be extremely long, except in the case of travellers with acute pulmonary histoplasmosis: median latency is 66 months among PLWH, 6 years among immunocompetent expatriates/migrants, and between as short as 3 weeks [83] and as long as 42 years [78] among patients with OIC. As previously shown by epidemiological studies of endemic areas [152] and among travellers [8], a majority of our cases involved males (440/659 66.8%) [12,, although the male-to-female ratio was lower (2:1) than that observed in other studies (4:1) [152].
Underlying diseases or treatments responsible for immunosuppression were present in 59.7% of the subjects, with HIV infection being the main risk factor (83.9%), followed by solid organ transplantation (3.9%). Primary immunodeficiency was diagnosed in 1.4% of the patients [12,20,40,73,86], and was notable for two cases of idiopathic CD4+ lymphocytopenia [40,73] and one case of IFN-γ autoantibodies identified during the patient's follow-up [71,86]. In the case of travellers, the most frequent route of exposure to Histoplasma was the exploration of bat-ridden caves and/or contact with bat guano, which was described in 52.2% of the cases for which this information was available [15,33,99,100,108,109,112,114,115,124,133,134,136], thus confirming this well-known risk factor [6,8], which should always be investigated in subjects presenting with respiratory symptoms, especially if they have recently travelled to Latin America [153].
Progressive disseminated histoplasmosis (PDH) was the rule among the PLWH (90.3%) and was also observed in more than a half of the patients with OIC (57.1%) and 6.2% of the immunocompetent patients. This last prevalence is in line with the 8% found in a recent systematic review of acute histoplasmosis in immunocompetent travellers [9]. One recent retrospective study of 261 cases conducted in the United States found similar prevalence of disseminated disease among PLWH (78%) and patients with OIC (58%), but a higher prevalence among immunocompetent subjects (33%) [154]. As the exposure of the majority of the immunocompetent patients considered in our review was limited to travel, it can be speculated that the higher rate of disseminated disease in the American review may have been due to greater inoculum exposure.
A diagnosis of histoplasmosis was the AIDS-revealing condition in 52% of our PLWH and was associated with severe immunodepression as confirmed by the very low median CD4+ cell count and the observation of concomitant opportunistic infections in nearly 30% of the patients. Although gastrointestinal involvement was less frequent than in highly endemic areas such as French Guiana [155,156], it is worth noting that 5.3% of the patients had localised intestinal histoplasmosis, and that the fungus was visualised and/or cultured in 10.5% of the cases. Skin lesions were described in 32% of PLWH, 21.4% of the patients with OIC, and 4.9% of the immunocompetent patients; furthermore, primary cutaneous histoplasmosis was observed in an HIV-positive patient and an immunocompetent laboratory technician [39,106]. Oral ulcers in isolation or as part of disseminated disease [132] were relatively frequent among the PLWH (5.3%) and the patients with OIC (25%). Morote et al. found mucocutaneous involvement in 8.8% of their PLWH in French Guiana, which suggests an association with more profound immunosuppression and a risk of early death [157]. Interestingly, haemophagocytic lymphohistiocytosis (HLH) was described more frequently among the patients with OIC (14.2%) than the PLWH (2.6%), which seems to be in line with the finding of only 34 cases of secondary HLH associated with HIV infection in Cayenne Hospital [158].
Histology and cultures were the most frequently used means of diagnosing histoplasmosis in the immunocompromised PLWH and patients with OIC, whereas antibody detection was more frequently used in the diagnostic work-up of immunocompetent patients, thus confirming the data of Staffolani et al. [9]. Although a search for serum and urinary Histoplasma antigen is considered a very sensitive and rapid means of diagnosing histoplasmosis, it was used in none of the PLWH and only two of the patients with OIC [73,74], but in five immunocompetent patients (6.1%) [103,124,136]. In order to overcome the very limited availability of the Histoplasma antigen in most European laboratories, galactomannan (GM) antigen is used as a surrogate marker of infection on the grounds of its cross-reactivity with Histoplasma capsulatum antigen [159] and was positive in 87.5% of the PLWH ( [36,47,51,53,57,60,64] and PR) and 75% of the patients with OIC [77,79,82,84] who had disseminated histoplasmosis. The use of PCR was reported in 33.2% of the patients as a whole: 46/114 PLWH (40.3%) ( Table 2) and 11/28 patients with OIC (39.3%) ( Table 4). It was mainly used to complement the traditional methods of microscopy and cultures, and the diagnosis of only three PLWH was exclusively based on PCR [38]. As no details are given concerning PCR protocols and gene targets used, it is difficult to draw any definite conclusions about the role of PCR in the diagnosis of histoplasmosis, especially in non-endemic countries [160]; however, it is worth noting that a recent French comparative study has found that using real-time quantitative PCR and the ribosomal small subunit RNA (mtSSU) gene of H. capsulatum as a target has the advantage of a high sensitivity (97.7%) [161].
Most of the immunocompromised patients with PDH were treated with amphotericin B followed by itraconazole in accordance with the guidelines but, as shown in Tables 2 and 4, a number of other regimens were used, even though they are usually considered less effective [162]. Sixty percent of the immunocompetent subjects with acute pulmonary histoplasmosis were apparently successfully treated with itraconazole. It is difficult to evaluate clinical outcomes on the basis of single case reports and small case series, but mortality rates in the different categories of patients (PLWH 24.1%; patients with OIC 32%; and immunocompetent patients 5.6%) were sufficiently similar to those reported by Franklin et al. in the USA (15%; 24%, and 13%, respectively) [154].
This systematic review has two limitations: because of the exclusion of some languages (Turkish, Finnish, Romanian, and Russian), it is possible that some published cases were not considered, and the absence of some important information from the case reports and small case series may have biased our findings.

Conclusions
In conclusion, this review highlights the fact that histoplasmosis should be considered in the differential diagnosis of systemic diseases among immunocompromised subjects in Europe. The potentially long latency period in such patients requires a more extensive evaluation of their travelling history than that required in immunocompetent travellers presenting with respiratory symptoms. Informed Consent Statement: Written informed consent was obtained from the subject of our case study.

Data Availability Statement:
No new data were created or analysed in this study. Data sharing does not apply to this article.