Mayaro virus in Latin America and the Caribbean

ABSTRACT Objectives. To assess the distribution of Mayaro virus (MAYV) in Latin America and the Caribbean and evaluate existing country-level MAYV surveillance mechanisms. Methods. Research was conducted from May 2018 through May 2019 to collect data from academic literature on Mayaro fever in Latin America and the Caribbean. PubMed, ClinicalKey, Scopus, Nature, SciELO, LILACS, and Google Scholar were searched for peer-reviewed journal articles, and data from health authorities, including the Pan American Health Organization (PAHO) and ministries of health, was also sought. MAYV-related publications published from 1954 through 2019 were screened. Publications that added to the overall understanding of MAYV, including its geographical and epidemiological distribution, were included in this report. Results. A total of 901 MAYV cases have been reported in humans in countries in Latin America and the Caribbean. Since its discovery in 1954 in Trinidad and Tobago, MAYV has been isolated from individuals living in Argentina, Bolivia, Brazil, Ecuador, French Guiana, Haiti, Mexico, Panama, Peru, and Venezuela. Of those 901 cases, 42 of them were reported exclusively by health authorities. In contrast, 843 confirmed and presumptive autochthonous cases and an additional 16 imported cases were identified in academic literature. No country-level surveillance mechanisms for MAYV were recorded in academic literature or by health authorities. Conclusions. This report demonstrates that MAYV surveillance efforts are limited in comparison to the virus’s presence in Latin America and the Caribbean, highlighting the importance of enhancing arboviral surveillance systems in the affected countries.

Transmission of MAYV to humans primarily occurs by the bite of female mosquitoes from the Haemagogus genus in the dipteran family Culicidae, which are not known to be anthropophilic (1,4). In order to assess whether an urban cycle of MAYV could be initiated and sustained, laboratory experiments were conducted to assess the vector competence of anthropophilic mosquitoes (1, 5). Studies showed that MAYV can infect Aedes albopictus, but Aedes aegypti demonstrated limited opportunity for MAYV transmission to humans (1, 5); however, the vector competence of the species was demonstrated and suggests the potential for urban transmission (5). MAYV-infected Ae. Aegypti and Culex quinquefasciatus have also been found in the city of Cuiabá, in the state of Mato Grosso, Brazil (6). MAYV was first isolated from Mansonia venezuelensis collected from the Rio Grande Forest in Trinidad and Tobago (5). A small number of Psorophora ferox/albipes, Ae. (Ochlerotatus) serratus, Sabethes spp., Culex spp., and Haemagogus janthinomys have also contained isolates of the virus (5).
Further evidence of the virus's potential to emerge in new areas was provided when sheep, caiman, and equids were sampled from 16 cattle ranches across the Pantanal region of Brazil from October 2009 through January 2011 (7). Results from the study showed that 10 of the 748 equids were seropositive for MAYV, demonstrating the first evidence of MAYV circulation in the southern Pantanal. Since equids tested negative for MAYV in previous investigations, the authors of the study suggested that MAYV circulation in the Pantanal might be a recent occurrence. More research is needed to identify amplifier hosts and vectors, and to understand the ecology of the transmission of the virus (7).
Clinical symptom of Mayaro fever tend to be nonspecific and include high-grade fever, debilitating arthralgia, headache, retro-orbital pain, myalgia, vomiting, diarrhea, and maculopapular rash (8). Similar to Chikungunya, MAYV infection can also lead to the development of long-term arthralgia (9). To date, no studies have shown whether individuals infected with MAYV can present as asymptomatic.
Due to the generic nature of the symptoms of alphaviruses, misdiagnosis can occur. Diagnosis of MAYV relies on clinical diagnosis, which may contribute to the misclassification of MAYV cases for dengue, Zika, Chikungunya, or other arboviruses (3,4). The diagnostic procedures for MAYV include serological detection through the detection of antibodies in the host serum; molecular detection through reverse transcription polymerase chain reaction (RT-PCR); and viral isolation in a cell culture (4). The reference standard for MAYV diagnosis is viral isolation (4). Enzyme-immune assays, hemagglutination inhibition, immunofluorescence, or neutralization methods are useful serological methods for the detection of antibodies (4). Diagnostic misdiagnoses can occur among viruses of the Alphavirus genus due to cross-reactivity among viruses with common antigenic sites (1). Further, the short viremic period of MAYV may make it difficult to accurately isolate the virus from blood (4). As a result, the exact number of MAYV cases is difficult to determine in both clinical and serological detection outside of well-equipped laboratories (1).
Disease prevention relies on vector control and personal protection measures (10). Case management of MAYV relies solely on the treatment of symptoms with analgesics. Efforts to develop other control methods have been recorded. For example, the antiviral effects of thienopyridine derivatives have been evaluated and found to affect the late early and late stages of MAYV replication (11). One study demonstrated that these substances could potentially act as a new class of antiviral drugs due to their bioavailability and low cytotoxicity (11). A liveattenuated vaccine against MAYV has also been developed. The vaccine was found to protect against lethal challenge in murine models, was highly immunogenic, and noninfectious to mosquitoes, but still required further preclinical development (12). A synthetic DNA envelope vaccine has also been evaluated in mice (13). T cell immunity and antibody immune responses were induced by the vaccine, and mice challenged with live MAYV after receiving the vaccine were protected against the disease (13). Despite these developments, licensed vaccinations and antiretroviral therapy are not currently available for MAYV (1, 4).
The objectives of this report were to assess the presence of MAYV in Latin America and the Caribbean while simultaneously evaluating country-level MAYV surveillance efforts in Latin American and Caribbean countries.

MATERIALS AND METHODS
In order to assess the presence of Mayaro fever in Latin America and the Caribbean, databases that included PubMed, ClinicalKey, Scopus, Nature, SciELO, LILACS, and Google Scholar were searched periodically from May 2018 through May 2019. Materials in French, Spanish, Portuguese, and English were reviewed. The terms used and adapted to each database were "Mayaro fever," "Mayaro virus," "MAYV," "Mayaro," "Mayaro AND alphavirus," "Mayaro AND Semliki Complex," and "Mayaro AND epidemiology." In PubMed, 90 articles were identified for the search term "Mayaro fever," 221 articles for "Mayaro virus," 79 articles for "MAYV," 227 articles for "Mayaro," 158 articles for "Mayaro AND alphavirus," 9 articles for "Mayaro AND Semliki Complex," and 71 articles for "Mayaro AND epidemiology." (The other databases did not generate new, unique journal-article hits that were not in PubMed. Therefore, from this point forward, this report will discuss the journal-article hits only from PubMed.) In addition, health authorities were consulted to obtain outbreak data. Also, the archives of the Pan American Health Organization were examined, and epidemiological bulletins were investigated on the ministry of health websites of countries with a reported presence of Mayaro (Argentina, Bolivia, Brazil, Ecuador, French Guiana, Haiti, Mexico, Panama, Peru, Suriname, Trinidad and Tobago, and Venezuela). Lastly, references listed in academic articles and health authority reports were manually examined to further identify MAYV data for this report.
We included in our analysis journal articles and other publications that were published between 1954 and 2019 and that provided a historical overview of Mayaro, hosts, transmission mechanisms of the virus, clinical manifestation, geographical distribution and epidemiology of the disease, diagnostic methods, case management, surveillance efforts, known risk factors, and prevention and control techniques. Studies were excluded from the report if they were unable to add to the overall understanding of MAYV in Latin America and the Caribbean, including studies that were too broad or too heavily focused on the microbiological and virological components of MAYV or did not contribute data to the analysis.

RESULTS
Periodic searches were carried out between May 2018 and May 2019 to assess the distribution of Mayaro fever in Latin America and the Caribbean. The study flow diagram, including with the PubMed searching, is given in Figure 1.
In this special report, 50 publications were included in the analysis and 41 were excluded. Of those 50 included publications, 49 of them were journal articles and the other was a health authority alert issued by the Pan American Health Organization (PAHO) in May of 2019.
As shown in Figure 2 and   Guiana, Haiti, Mexico, Panama, Peru, Suriname, Trinidad and Tobago, and Venezuela made no mention of the presence of Mayaro. Table 2 includes a descriptive overview of reported confirmed and presumptive cases of Mayaro, as well as the publications in which these cases were identified. The table provides information on the number of cases reported in an article, the lead author of the article in which the case(s) were reported, the country the case(s) originated in, the year the case(s) occurred, and the diagnostic procedure used to confirm the case(s).  1 A case report revealed that an HIV-infected patient that returned from a work trip to the Amazon basin presented symptoms of febrile illness. After testing negative for DENV and CHIKV, MAYV was confirmed by RT-PCR and subsequently followed by nextgeneration sequencing and phylogenetic reconstruction. The test was developed in the in-house laboratory for the E1 gene. The interplay between arboviral diseases and HIV is poorly understood. In theory, HIV-induced immunosuppression can lead to more atypical and aggressive manifestations of arboviral infections.
(continued)  The lack of standardization in Mayaro case definitions and laboratory diagnostics weaken the robustness of reported case data; however, the data was incorporated in this report due to the limited availability of MAYV case data to create the first epidemiological database for Mayaro. To date, epidemiological data on Mayaro continues to be descriptive in nature. Given the lack of accurate and consistent case information, incidence rate calculations were not attempted. Rather, an overview of the number of reported cases and/or epidemics of MAYV was given. Data on MAYV mortality was extremely limited. Of the 901 reported MAYV cases, only one fatality was recorded in academic literature (27). Lastly, MAYV attack rates were not available in the existing literature.

Argentina
Three cases of Una virus (UNAV), a subtype of Mayaro, were discovered in the sera of three individuals from central Argentina in 2001 (35). Due to the low titers of neutralizing antibodies detected, it was difficult to ascertain when the individuals became infected with UNAV; nonetheless, the circulation of UNAV in Argentina was demonstrated (35).

Bolivia
A total of 60 cases were identified from 1955 through 2007 in Bolivia. In the first instance, Uruma virus, now considered to be a strain of Mayaro, was isolated from the blood of two febrile Okinawan colonists in 1955 (15). Another article reported that 12 cases of MAYV were found in 2007, but no further information was provided about the cases (14). The remaining 46 MAYV cases in Bolivia were identified from 2000 to 2007 in a study done in three departments of the country (16).

Brazil
The country with the highest number of recorded MAYV cases in the Americas is Brazil. Of a total of 495 cases, 199 of them were identified in the state of Goiás, and 194 took place in the state of Pará. Cases were also identified in the states of Acre, Amazonas, and Mato Grosso and the city of São Paulo. The virus is considered to be endemic in the northern, western, and central portions of the country (1).

Ecuador
The first case of MAYV in Ecuador was identified during a study that took place from 2000 to 2007 (16). The patient lived in the city of Guayaquil, in the province of Guayas (16). In May of 2019 an epidemiological bulletin released by PAHO reported that an additional five cases of Mayaro were detected in Ecuador (38). These cases were identified in samples that tested negative for leptospirosis, dengue, Chikungunya, and Zika (38) and that had originated from individuals in the cities of Guayaquil, Babahoyo, Portoviejo, and Santo Domingo (38).

French Guiana
The first case of MAYV in French Guiana was identified in 1996 (25). The patient was a female who had been living in French Guiana for several months and had presented with symptoms for two days (25).

Haiti
Only one case of MAYV has been recorded in Haiti. In 2015, an eight-year-old boy from a rural/semirural area of the Ouest department presented febrile symptoms and tested positive for MAYV (26). This case demonstrates an instance where the ecological settings that precipitated the emergence of this case differed from the sylvan Amazon regions where MAYV infections have generally occurred (26).

Mexico
A total of two cases of MAYV have been found in Mexico. These cases were identified in 2001, when 35 patients at the Mexican Institute of Social Security in the state of Veracruz underwent testing for MAYV (27). Two of the patients tested positive for the virus and displayed symptoms characteristic of hemorrhagic cases (27). Of the two cases, one patient died (27). This has been the only recorded case of fatality due to MAYV (27).

Panama
One case of MAYV was detected in a study carried out from 1966 through 1967 (33). In this instance, a citizen of the United States of America who was conducting studies assessing the feasibility of routes for a new, sea-level canal in Panama or Colombia demonstrated a significant rise in viral MAYV antibodies (33). Unfortunately, due to study group cross-over, it was unclear whether the virus was acquired in Panama or Colombia (33).

Peru
Peru had the second-largest number of reported cases. A total of 230 cases were identified, with the highest number reported in the region of Loreto. Mayaro was also recorded in the regions of Madre de Dios, Junín, Cusco, Huánuco, Ucayali, San Martín, Tumbes, and Ayacucho.

Trinidad and Tobago
Trinidad and Tobago was the location where MAYV was first isolated, from blood samples from one urban-dwelling female and four male forest workers (1, 37). Since the virus was first discovered in the county of Mayaro, it was subsequently named for that area.

Venezuela
There have only been two recorded instances of MAYV infection in Venezuela. In 2000, MAYV was identified in a family of four in the Barlovento subregion of the state of Miranda (30). The second instance of MAYV was in the village of La Estación, where 77 cases were reported, of which 19 of them were confirmed as being seropositive (14).

Mayaro virus surveillance efforts
To date, no standardized case definitions or laboratory diagnosis algorithms have been recorded for Mayaro fever at the regional or country levels in Latin America and the Caribbean. Similarly, information on country-level MAYV surveillance efforts in Latin America and the Caribbean were not available in academic literature or from health authorities. However, many of the cases reported in academic literature came about from the differential diagnosis of sera that were negative for other, more well-known arboviral diseases (1, 18, 21-23, 26, 34, 39, 42, 46-48).

DISCUSSION
MAYV was first isolated in 1954 in Trinidad and Tobago and has since been reported in Central and South America, circulating through high-density tropical forests (1, 4,8,. A total of 901 MAYV cases have been reported in 10 Latin American countries and 1 Caribbean country. Of these 901 cases, 843 autochthonous and 16 imported cases were reported in academic literature, and the remaining 42 cases were exclusively reported by health authorities (1, 4,8,. Given the extent of arboviral diseases in Latin America and the Caribbean, MAYV may pose a similar public health threat in the area. As with other arboviral diseases, such as yellow fever, that have an urban cycle through Ae. aegypti, the MAYV case in Haiti highlights the need for vector competency research and demonstrates the potential for the MAYV transmission cycle to shift into an urban cycle. Urbanization of MAYV could establish new outbreaks of greater number throughout Latin America and the Caribbean and result in a wider distribution of MAYV, including circulation in North America and Europe (9,26). Changes in the natural environment and the resulting impacts on interactions between host and vector populations could also expand the geographic distribution of MAYV in Latin America and the Caribbean (49). These changes could include the globalization of human and animal transportation, climate change, deforestation, human colonization, urbanization, mining, and infrastructure construction (dams and highways) (10,28,49). Increased air travel and globalization is expected to boost the number of imported cases of MAYV (9,44). As such, MAYV demonstrates the potential to emerge in new areas and reemerge in existing areas and become a significant public health issue in the Western Hemisphere (10). Findings from our research coincide with previous research.
Cases of MAYV may be going undetected due to a lack of awareness among the medical community. For example, since diagnostic testing is not widely available and little is known of MAYV outside of endemic areas, imported cases may be getting clinically misdiagnosed as dengue. Within Latin America and the Caribbean, coinfection with other arboviral diseases may also misguide the diagnosis of MAYV, further demonstrating that the number of reported cases of MAYV only represent a portion of those that have occurred (9). It is also important to note that many of the cases of MAYV that were identified in academic literature were only detected because of extensive testing for arboviral viruses such as dengue and Chikungunya, further pointing to the need for differential testing for arboviral diseases. Lastly, since clinical symptoms of MAYV are generic, treatment may not be sought after and cases could go undetected. This would present a large obstacle to timely outbreak detection.
Data constraints must be considered when evaluating the outcomes of this research. The first limitation to this research includes the potential overestimation of the total number of presumptive MAYV cases found in academic literature due to serological cross-reaction between alphaviruses. Cross-reaction could also be a concern since diagnostic techniques were not uniform throughout the studies we included. Finally, many cases of MAYV may be unaccounted for both due to similarities between the clinical symptoms of MAYV and other arboviral diseases and coinfection.
Findings from this research demonstrate that MAYV surveillance efforts are limited in comparison to the virus's epidemic potential, highlighting the importance of enhancing arboviral surveillance systems to include the differential diagnosis of MAYV. Given that the number of known cases of MAYV were primarily drawn from academic literature, country-level surveillance systems need to be strengthened to improve the data collection capabilities of health authorities.
Specific recommendations for MAYV surveillance include standardized lab capacities, the proactive seeking of cases, and the creation of standardized case definitions and laboratory diagnosis algorithms. While these recommendations are generally applicable to other arboviral diseases, the application of these recommendations specifically for MAYV will allow epidemics to be identified before they become problematic and also help the data collected to serve as a bridge between research and public health practice. Capacities and differential diagnosis algorithms need to be strengthened to ensure that labs are able to accurately distinguish among arboviral viruses. It is also important that labs be considered an integral data source for surveillance systems. Other recommendations include strengthening entomological surveillance and creating a global arboviral database. Public health practitioners and clinicians should also be made aware of the other arboviral diseases that exist, to be better able to discern the differences between them. Lastly, increased research interest will allow for the ongoing exploration of potentially epidemic diseases, such as Mayaro fever.
Author contributions. Dr. Ana Riviere-Cinnamond conceived the original idea for the special report, and Niloofar Ganjian collected the data, analyzed the data, interpreted the results, and wrote the paper. Both authors reviewed and approved the final version.