Non-Endemic Leishmaniases Reported Globally in Humans between 2000 and 2021—A Comprehensive Review

Leishmaniases are human and animal parasitic diseases transmitted by phlebotomine sand flies. Globalization is an important driver of the burden and in the current dynamics of these diseases. A systematic review of articles published between 2000 and 2021 was conducted using the PubMed search engine to identify the epidemiology and clinical management of imported human leishmaniases as a fundamental step to better manage individual cases and traveler and migrant health from a global perspective. A total of 275 articles were selected, representing 10,341 human imported cases. Identified drivers of changing patterns in epidemiology include conflict and war, as well as host factors, such as immunosuppression, natural and iatrogenic. Leishmania species diversity associated with different clinical presentations implies diagnostic and treatment strategies often complex to select and apply, especially in non-endemic settings. Thus, diagnostic and management algorithms for medical clinical decision support are proposed. Increased surveillance of non-endemic cases, whether in vulnerable populations such as refugees/migrants and immunocompromised individuals or travelers, could improve individual health and mitigate the public health risk of introducing Leishmania species into new areas.


Introduction
Leishmaniases are a group of diseases caused by protozoa belonging to the genus Leishmania. The parasites are transmitted by phlebotomine sand flies, and the disease is zoonotic in most settings. Leishmaniases are worldwide distributed and can be separated geographically into Old World (OW) and New World (NW) diseases, with different species occurring in different areas [1]. Over 20 species have been recognized as human pathogens and clinical manifestations of leishmaniases vary largely but are often divided into two clinically distinct forms: visceral leishmaniasis (VL) and cutaneous leishmaniasis (CL). VL is caused by parasites of the Leishmania donovani complex (L. donovani in the Old World and L. infantum in both the Old and New Worlds) being responsible for causing a severe disease which is lethal when untreated. While L. donovani transmission is anthroponotic, the life cycle of L. infantum (synonym L. chagasi in NW) is mostly zoonotic, with domestic dogs as the main reservoirs of human infection. CL, caused by several species of Leishmania and responsible for considerable morbidity in endemic foci, ranges from a benign form with spontaneous resolution, to a disfiguring skin condition involving mucosal tissues [2]. According to [3], 98 countries and territories are endemic to leishmaniasis, with more than 12 million infected people, and with an estimated annual incidence of 50,000 to 90,000 VL cases and 600,000 to 1 million CL cases [4].
Reporting policies and practices in endemic countries are inconsistent, with a lack of systematic reporting of all human clinical forms of leishmaniases, leading to an un-derestimation of the local and global burden of leishmaniasis [5]. On the other hand, the available information shows an increase in the number of imported cases in endemic and non-endemic countries [6][7][8][9][10][11][12], which can be explained by a combination of factors, such as increased human traveling, migration, or population displacement from or to endemic areas and an increase in the number of susceptible populations due to immunosuppressive factors, co-morbidities, and aging. Altogether, increased human mobility and globalization have expanded the at-risk population for leishmaniasis and, simultaneously, pose a risk of geographic expansion of Leishmania species.
The present study aimed to summarize and analyze the epidemiology, clinical presentation, diagnosis, and management of non-endemic human leishmaniasis, through a comprehensive review of the literature in the last 22 years (2000-2021) to raise awareness of the medical community regarding the challenges associated with the diagnosis and management of this parasitic disease.
Search results were saved as a comma-separated value (CSV) file and imported into Microsoft Excel ® . Study eligibility was manually assessed by two independent researchers in a blind manner. All records were screened according to the title, and abstract, if available. Only studies published between 2000 and 2021 were included, even if the cases reported were diagnosed in previous years. Only original research articles reporting humans with non-endemic Leishmania infection (i.e., reportedly infected by Leishmania parasites in a country different than the one they were living in at the moment of diagnosis) were retained, including those published in some languages other than English (Figure 1).
The presence of repeated cases in different articles was assessed-either confirmed, when explicitly mentioned in the text, or suspected, based on the authors, place of infection, year, and place of diagnosis (including hospital or center). Articles, where all or most cases reported, had (certainly or likely) been previously described in the literature were mostly discarded (except if they contained clinical or epidemiological details not published in previous works). This was the case for two GeoSentinel articles [13,14]. Articles, where some cases in a series had (certainly or likely) been previously described, were retained, but some cases were discarded, either entirely or in part of the information. This verification process of repeated cases was performed manually and for all the selected articles. The presence of repeated cases in different articles was assessed-either confirmed, when explicitly mentioned in the text, or suspected, based on the authors, place of infection, year, and place of diagnosis (including hospital or center). Articles, where all or most cases reported, had (certainly or likely) been previously described in the literature were mostly discarded (except if they contained clinical or epidemiological details not published in previous works). This was the case for two GeoSentinel articles [13,14]. Articles, where some cases in a series had (certainly or likely) been previously described, were retained, but some cases were discarded, either entirely or in part of the information. This verification process of repeated cases was performed manually and for all the selected articles.
Some records had missing data, and the denominators mentioned in the text and tables count only those where data was available. Articles, where the place of infection included a list of several countries, were counted for the region of infection, but not for the country of infection. The same principle was applied to the place of diagnosis. Regions of infection/diagnosis were defined based on the World Bank Group proposed regions. New World cases grouped all cases infected in the American continent; Old World grouped all the remaining cases. Activity (travel, migration, refugee) was classified based on original articles' information and considering travel as a broader category including military and missionary service, tourism, visiting friends and relatives, work stays and exchange student programs. Some records had missing data, and the denominators mentioned in the text and tables count only those where data was available. Articles, where the place of infection included a list of several countries, were counted for the region of infection, but not for the country of infection. The same principle was applied to the place of diagnosis. Regions of infection/diagnosis were defined based on the World Bank Group proposed regions. New World cases grouped all cases infected in the American continent; Old World grouped all the remaining cases. Activity (travel, migration, refugee) was classified based on original articles' information and considering travel as a broader category including military and missionary service, tourism, visiting friends and relatives, work stays and exchange student programs.
Only laboratory-confirmed leishmaniasis cases were included in this review. Methods and samples for diagnosis were included only when specified for each individual in the article. Besides counting the number of individuals in which each test was performed, the result of the test for each individual was registered: positive (suggestive or confirmatory of current Leishmania infection) or negative. Species/complex identification was only considered when the articles mentioned laboratory confirmation, even though the exact technique may not be specified. Even though some articles mentioned identification to the species level, for result analysis and discussion purposes, cases of species of the same complex are presented together, following the classification proposed by Clinical signs, symptoms, and laboratory findings were extracted, whenever available, using the terms contained in the original articles. Patients with splenomegaly, hepatomegaly or both were grouped under the same category. Patients with unspecific/constitutional symptoms were also grouped. For CL, lesion type was classified into four categories, following classical dermatological lesion classification nomenclature: ulcerated lesions; papular/nodular lesions; macular lesions, plaques and crusts; other (whenever this term was used in the original articles). Classification of mucosal (ML) and mucocutaneous (MCL) leishmaniasis cases was performed by the authors, following the definition proposed in the "Manual on case management and surveillance of the leishmaniases in the WHO European Region" [5], whenever clinical information provided allowed, and regardless of the article's original classification. In articles with insufficient clinical information, ML/MCL classification was assumed according to the article [11,15]. Immunosuppression status included diabetes mellitus, malignancy, transplant, HIV infection and pharmacological immunosuppression (not-transplant related). Post-Kala-azar Dermal Leishmaniasis (PKDL), disseminated and diffuse CL were considered specifically when this term was used in the original article.
Similarly, laboratory findings of 125 patients translated the natural history of the disease well known in endemic countries [286]: most patients had single or multiple lineage cytopenias, while elevation of liver enzymes and renal failure were less commonly described (Table 2). In 22 patients, laboratory findings were consistent with hemophagocytic lymphohistiocytosis (HLH) [17,26,44,49,50,52,59,68,206,214]. Although the association between leishmaniasis and HLH seems to be a rare finding in the pediatric population of endemic areas [288], cases occurring in infants have been reported in imported cases [44], so in children under 2 years of age traveling to endemic areas, leishmaniasis should be included in the differential diagnosis of secondary HLH.
The diagnostic approach was described for 343 patients (Table 3): microscopy and serology were the techniques most often used. Bone marrow aspiration or biopsy was the most commonly used biological samples for microscopy, polymerase chain reaction (PCR) and culture. Positivity in these samples was higher using PCR (96.6%), followed by microscopy (91.5%) and culture (88.6%). For PCR, blood was also commonly used (30.5%). Other samples where positive parasitological results were occasionally obtained included liver and lymph nodes, the last ones possibly representing part of the path for investigation of alternative diagnoses, such as lymphoma.
Among serological techniques, indirect fluorescent antibody test (IFAT), direct agglutination test (DAT) and immunochromatographic test (ICT, rK39-based) were employed with similar frequency (40-45%), probably reflecting their more widespread availability in general, and especially in non-endemic countries/areas. The sensitivity and specificity of these tests in immunocompetent patients are considered to be high and therefore good choices for initial diagnosis [5]; in addition, quantitative serological tests (such as IFAT and DAT) are also useful for follow-up, as antibody titers tend to decay after successful treatment [5]; qualitative tests (such as ICT), on the other hand, offer a fast, point-of-care alternative for serological diagnosis.
All cases in which Leishmania species/complex was identified (n = 169) belonged to the L. donovani complex, except for one case caused by the subgenus Mundinia [253] (previously identified by the nomen nudum L. "siamensis" and later suggested to be L. martiniquensis [213]) and a co-infection caused by L. infantum/L. major [40]. Visceral disease caused by L. (Mundinia) species [289] and L. major has been rarely reported [290]. Leishmania major/L. donovani complex co-infections have been associated with visceral [291] and cutaneous disease [292], including disseminated CL in the context of HIV infection [293,294]. The case identified in this review was a patient chronically medicated with steroids and methotrexate.
The diagnostic approach was described for 1690 patients (Table 3). Although skin biopsy allows for differential diagnosis in suspected cases of CL, the sensitivity of wellperformed skin scrapings is similar [299], and the procedure is less invasive, which could explain why these biological samples were used in most cases. The positivity rates reported with PCR (97.2%) were considerably higher than with microscopy (84.2%) or culture (86.5%). Other biological samples occasionally used for diagnosis included mucosal biopsies (n = 61, in ML/MCL cases), blood (n = 2) and lymph node (n = 2, in MCL/ML cases).  Even though the use of serological methods for the detection of antibodies against Leishmania in CL should be discouraged because of their low sensitivity and variable specificity [5], the performance of serological testing was reported in 115 patients, yielding positive results in 69.1% of the samples tested. IFAT was the technique most commonly used (73.7% of serological tests), followed by ICT (17.2%). Reporting of serological testing was proportionally more common in MCL/ML cases (21/61 versus 94/1629 in CL) and positivity rates were also higher (82.1%); this finding is in line with the WHO recommendation in the European region to include serological methods in the laboratory diagnosis of MCL/ML [300]. Infecting Leishmania species/complex was described in 3495 individuals and followed the relative distribution shown in Figure 3, for Old World and New World disease. The distribution of species/complex for each country of infection is represented in Figure 5 and Supplementary Table S3.
Overall, the results demonstrated that human movements have led to an increase in the number of imported CL cases due to non-indigenous Leishmania species in both endemic and non-endemic countries. Although the potential risk of introducing these exotic Leishmania species into non-endemic areas is low, since for most of them their main reservoirs hosts are absent, the vectorial competence of local sand fly species must be considered, as it may allow for successful adaptation of non-indigenous Leishmania species with important epidemiological consequences [301]. The most expressive group of imported cutaneous cases was represented by refugees from war zones in the Middle East diagnosed mostly in neighboring countries (such as Lebanon and Turkey) and reflecting how the ongoing Syrian war has dramatically increased the incidence of CL in these countries [302]. The refugee status of these people in the host countries could be an important factor deterring an early diagnosis of disease. Some countries and centers, such as the ECDC [303], have produced and implemented guidelines and recommendations for the initial healthcare screening of migrants and refugees. Although leishmaniasis is only briefly addressed in these documents, promoting a complete assessment that includes skin checks [304] will probably help shorten onset to treatment intervals in CL. This shortening could be particularly relevant in the European context since untreated CL lesions harbor vector infective parasites [305], which could infect competent/permissive vectors and allow the establishment of anthroponotic cycles for non-endemic Leishmania species. Phlebotomus sergenti, a specific vector of L. tropica, the species most imported with refugees, is widely distributed in Southern Europe [306] and new endemic foci could emerge through the introduction of infected humans in areas where the sand fly species are present [307]. Other phlebotomine species present in the Mediterranean region, namely P. perniciosus and P. tobbi, have also been shown to be susceptible to L. tropica infection under experimental conditions [308,309] and capable of transmitting viable parasites to vertebrate hosts (for P. perniciosus), representing an additional threat of its introduction in countries where these permissive sand fly species are present [310]. The movement of refugees across borders has also been linked to the detection of new L. tropica zymodemes in endemic areas. In Crete, where L. tropica infection is diagnosed sporadically [310], a new zymodeme was detected from an Afghan refugee and later on a local (non-traveler) dog [138]. DNA of non-endemic Leishmania species has been detected in phlebotomine sand flies in refugee camps in Greece [311].  Although the risk of introduction of L. major seems to be low, as its gerbillids reservoirs are not present in Europe [310,312], voles of the genus Microtus have recently been implicated as L. major reservoir hosts in a CL focus in northern Israel [313]. As such, the possibility that L. major, having adapted to voles, spreads north into Turkey and southern Europe, where reservoir hosts and vectors exist sympatrically should not be neglected. Furthermore, the presence of European sand fly species permissive to this parasite species, such as P. perniciosus [314] and P. tobbii [315], should also be kept in mind. Indeed, the presence of L. major in Europe has already been reported, namely in Portugal, through the detection of its DNA in sand flies [316] and a cat [317], and of L. infantum/L. major hybrids in HIV-infected patients [291] and a cat [317].
Phlebotomine vectors for L. infantum are also permissive for L. donovani and are widely distributed across Southern Europe, raising concern for the introduction of this parasite species, as has already been documented in Cyprus in humans and dogs [318]. Hybridization between L. infantum and L. donovani is also concerning and has been demonstrated in Turkey [319].
The clinical presentation for each of the three most common OW species/complex is summarized as follows: -L. donovani complex: predominantly single lesions; involving mostly the head/neck and upper limbs, often non-ulcerative (especially plaques/crusts); -L. major: predominantly multiple lesions; involving mostly the limbs, mostly ulcerative; -L. tropica: single or multiple lesions; predominantly in the head/neck and upper limbs, often non-ulcerative.
For New World disease, relevant findings by species/complex can be summarized as follows, for the three most common species/complexes: -L. braziliensis complex: predominantly single lesions; most often in the lower and upper limbs; commonly ulcerative; more frequently associated with lymphadenopathy; -L. guyanensis complex: predominantly multiple lesions; involving mostly the upper limbs; ulcerative; often associated with lymphadenopathy; -L. mexicana complex: predominantly single lesions; mostly in the head/neck; mostly ulcerative, but also frequently plaques/crusts.
For ML, the median age was 64 years old and 78.8% of patients were male. Infection likely occurred in the Old World for 76.9% of individuals (85.7% of these from Europe/Central Asia). Eleven different countries were identified as likely places of infection for 29 patients.
Of 27 patients where the description of the location of mucosal lesions was available, 10 (37.0%) had exclusively nasal involvement, eight had exclusively oral involvement (including four cases of tongue leishmaniasis), eight had laryngeal involvement and one intestinal involvement (no evidence of visceral disease) [26]. Hoarseness, nasal obstruction, and nasal discharge were frequently reported signs/symptoms (16-42%). In three cases, mucosal coinfection with Candida species was reported (oral/pharyngeal candidiasis, [158,177,204]).
The analysis of all the CL cases allowed for a comparison of clinical presentation and outcome, which differed not only between OW and NW cases but also between species, as outlined in previous works [15,320]. It is also important to note that travel between endemic regions for different Leishmania species/strains can potentiate coinfection and lead to the generation of new hybrids, with different pathogenicity [291], the different clinical presentations with the consequent increase in the difficulty of diagnosing and management [321,322]. By combining this information with the relative abundance of different species in imported cases from each country, an approach for the diagnosis and management of CL, MCL and ML cases in non-endemic settings is suggested ( Figure 6). This algorithm represents the authors' opinions based on the results of this review and aims to help clinicians to judge the likely infecting species and adapt treatment strategies accordingly, in situations where laboratory confirmation of the infecting species/complex cannot be performed in clinically relevant time. Additionally, it should be noted that recommended treatments are not always available in non-endemic settings and, when available, medical teams may not be very familiar with their use.

Particular Aspects in Immunosuppressed Patients
Data on the immunological status of patients was gathered whenever available, to emphasize the contribution of new immunosuppression factors to the progression of the disease and understand the changing epidemiology of leishmaniasis in non-endemic countries, similarly to endemic areas [323]. In immunosuppressed individuals, even when asymptomatic, Leishmania infection can be transmitted to phlebotomine vectors [324], and, in this way, these patients could play a role in the maintenance of the cycle of the parasite.
Even though almost half of the immunosuppressed patients with VL were people living with HIV/AIDS, approximately one-fourth were patients chronically medicated with immunosuppressive drugs for inflammatory and auto-immune disorders. Of these, methotrexate, steroid and anti-TNFα were the most commonly implicated. In ML cases, immunosuppressive therapy was the most common form of immunosuppression (over half of cases), reinforcing that these patients represent one of the groups at risk of developing this clinical condition [5]. A high proportion of these cases (~25% of VL and 50% of ML) were attributed to chronic medication with anti-TNFα. The association of anti-TNFα therapy with symptomatic Leishmania infection has been described in many case reports [325] and reviewed in previous articles [326]. Though the use of anti-TNFα has been increasing, no clinical trials have properly addressed the risk of progression to disease in previously and newly infected patients compared to non-medicated patients, and no strategies for the treatment of asymptomatic individuals have been prospectively researched, although clinical (and laboratory) monitoring could be suggested [323]. As such, no evidence currently supports the screening of individuals before starting treatment and there is no consensus on when and how to treat Leishmania infection in asymptomatic cases. Additionally, guidance on secondary prophylaxis in non-HIV immunosuppressed patients is lacking. Future research should address this gap of knowledge regarding appropriate prevention and management in emerging groups of immunosuppressed patients.
In addition, and as most of CL/MCL/ML in immunosuppressed patients was associated with infection in the Old World with the L. donovani complex, ear nose and throat clinicians should be aware of leishmaniasis as a differential diagnosis of oral and laryngeal lesions in immunocompromised people who traveled to Leishmania endemic areas, namely to the Mediterranean region.

Conclusions
More than 10,000 cases of non-endemic leishmaniasis were reported in humans between 2000 and 2021, reflecting the impact of this disease on global tourism and migration and the movement of people (Box 1).
Drivers of changing patterns in epidemiology included the same as in endemic settings, namely conflict and war, as well as host factors such as immunosuppression (both natural and iatrogenic). Increased clinical management and surveillance of non-endemic cases (by physicians, especially dermatologists) could improve individual health and mitigate the public health risk of introducing Leishmania into new areas where favorable environmental conditions and permissive vectors exist. Strengthening surveillance and systematically combining animal and human data into an integrated platform, following a One Health approach (as proposed in a recent ECDC report [327]), could be the key to addressing the risk of leishmaniasis introduction associated with increased human and animal mobility. Box 1. Main findings.

•
Over 10,000 non-endemic cases of human leishmaniasis were reported in the literature from 2000-2021, most commonly CL, followed by VL and ML/MCL. • VL resulted from travel to Europe in most cases; approximately half of the patients were children or elderly; fever, hepatosplenomegaly and pancytopenia were the most common findings; atypical presentations such as isolated lymphadenopathy, gastrointestinal and pulmonary involvement were described; the diagnosis was commonly made by microscopic examination of bone marrow biopsy/aspiration and/or serology; L. donovani complex was implied in almost all cases; LAmB was the drug most often used for treatment.

•
Most CL cases were diagnosed in refugees from the Middle East, migrants from Latin America and South Asia and military personnel deployed in Asia; diagnosis relied on skin scraping and/or biopsy, with positivity rates higher for PCR than microscopy or culture; L. tropica and L. major were the two most common species in the Old World, while L. braziliensis complex and L. guyanensis complex were predominant in the New World; the number, type and location of lesions differed between species/complexes, as well as the therapeutic strategies used and the relapse rates reported; L. aethiopica, L. naiffi, L. lainsoni and L. martiniquensis infections were rarely described. • MCL was reported in younger individuals, infected in the New World, most often by L. braziliensis complex; nasal mucosa was more often involved, and lymphadenopathy was common; the time between cutaneous and mucosal lesions varied from simultaneous to fifty years. • ML was diagnosed mostly in older patients, infected in the Old World, most often by L. donovani complex; oral and laryngeal mucosa involvement was frequently described.

•
Immunosuppressed patients represented a significant share of ML and VL cases; the two most common causes for immunosuppression were HIV/AIDS infection and chronic therapy, where anti-TNFα drugs represented the largest group; relapse/failure rates were higher in these patients. • Non-endemic leishmaniasis represents an individual health problem, especially for refugees and immunosuppressed people; but also, a public health concern, related to the risk of introduction of the disease in new areas.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/pathogens11080921/s1, Figure S1: Age distribution of visceral leishmaniasis cases; Figure S2: Age distribution of cutaneous, mucocutaneous and mucosal leishmaniasis cases; Table S1: References of cases of visceral leishmaniasis diagnosed, by country of travel or migration; Table S2: References of cases of cutaneous, mucocutaneous, and mucosal leishmaniasis diagnosed, by country of travel or migration; Table S3: References of identifications of species/complex by country in the Old and New Worlds.