Arboviruses in Mammals in the Neotropics: A Systematic Review to Strengthen Epidemiological Monitoring Strategies and Conservation Medicine

Arthropod-borne viruses (arboviruses) are a diverse group of ribonucleic acid (RNA) viruses, with the exception of African swine fever virus, that are transmitted by hematophagous arthropods to a vertebrate host. They are the important cause of many diseases due to their ability to spread in different environments and their diversity of vectors. Currently, there is no information on the geographical distribution of the diseases because the routes of transmission and the mammals (wild or domestic) that act as potential hosts are poorly documented or unknown. We conducted a systematic review from 1967 to 2021 to identify the diversity of arboviruses, the areas, and taxonomic groups that have been monitored, the prevalence of positive records, and the associated risk factors. We identified forty-three arboviruses in nine mammalian orders distributed in eleven countries. In Brazil, the order primates harbor the highest number of arbovirus records. The three most recorded arboviruses were Venezuelan equine encephalitis, Saint Louis encephalitis and West Nile virus. Serum is the most used sample to obtain arbovirus records. Deforestation is identified as the main risk factor for arbovirus transmission between different species and environments (an odds ratio of 1.46 with a 95% confidence interval: 1.34–1.59). The results show an increase in the sampling effort over the years in the neotropical region. Despite the importance of arboviruses for public health, little is known about the interaction of arboviruses, their hosts, and vectors, as some countries and mammalian orders have not yet been monitored. Long-term and constant monitoring allows focusing research on the analysis of the interrelationships and characteristics of each component animal, human, and their environment to understand the dynamics of the diseases and guide epidemiological surveillance and vector control programs. The biodiversity of the Neotropics should be considered to support epidemiological monitoring strategies.


Introduction
Tropical forests harbor much of the world's tree diversity [1,2] and more than 1617 recognized mammal species [3,4]. Mammals are important ecological components for have a great capacity to adapt to new habitats [65]. Similarly, the order Didelphimorphia presents a wide distribution in the Neotropics in almost all habitat types [66]. These orders involve special attention due to their biological characteristics, as they act as wild reservoirs of arboviruses [36] that affect humans and domestic animals [67,68].
The emergence and spread of emerging infectious diseases are associated with the way humans interact with animal species and the environment [69]. The importance of detecting the zoonotic spread of a vertebrate animal, beyond being a natural reservoir, is to understand a complex process that requires the intervention of environmental, pathogen, and host factors [70,71]. Habitat loss increases biodiversity loss [69]. Better-preserved habitats reduce spillover events, which is known as the dilution effect [72]. Higher diversity often leads to a lower prevalence of infection in hosts [73].
This research proposes to contribute to public health and zoonosis or re/emerging diseases prevention considering that the niche, vector dynamics, hosts, and viruses have been modified in the last decades. In certain habitats, accelerated population growth, quality of life, and sanitary conditions could favor the spread of arboviruses in various mammalian hosts [33,74], posing a global public health problem [47,75].
In order to achieve a better understanding of the ecology of diseases, it is necessary to detect, prevent and control them, and to approach them from a conservation medicine approach given the connection between wildlife and domestic animals, the ecosystem, and humans [76]. This approach will serve as a tool for the understanding, prevention, and management of health problems derived from environmental change [77] in one of the most biodiverse areas of the planet.
This systematic review will allow the identification of studies focused on mammalian arboviruses in the Neotropics excluding the human primate group. The analyses will show the diversity and geographic records of arboviruses, as well as the taxonomic groups of mammals that are most monitored and those that have an information gap. The information obtained on the association of wild and/or domestic mammals as potential hosts and reservoirs of arboviruses is intended to support epidemiological surveillance tasks with a focus on conservation medicine and/or the One Health approach [78,79] to achieve the integration of knowledge and apply it in favor of biodiversity.

Methods
The literature review of mammalian arbovirus records in the Neotropics was conducted between 1 June and 31 July 2021 under the Preferred Reporting Items for Systematic Reviews and Meta-Analyses PRISMA 2020 Checklist reporting guide and the PRISMA 2020 flowchart [80,81] that describe the process for literature exploration and justification for the selection of each investigation [82,83].
For the search of indexed articles, we used Google Scholar and PubMed databases by combining the following keywords and Boolean operators: Mammals AND arbovirus AND Neotropical countries AND Argentina OR Belize OR Bolivia OR Brazil OR Chile OR Colombia OR Costa Rica OR Ecuador OR El Salvador OR Guatemala OR Guyana OR French Guyana OR Honduras OR Mexico OR Nicaragua OR Panama OR Paraguay OR Peru OR Suriname OR Uruguay OR Venezuela, including all publications without distinction of the year of publication. Additionally, exclusion criteria included (1) a language other than Spanish, Portuguese, and English, (2) a focus on humans, vaccines, clinical cases, and laboratories, (3) duplicates between the two predetermined databases, (4) lack of information regarding the locality of registration, (5) not including the mammalian class, (6) books with extensive and generalized content, and (7) bibliographic reviews.
All the information obtained from the selected articles was organized in a database to determine the sampling effort, i.e., the number of publications as surrogate data, arbovirus records in neotropical countries (number of samples and prevalence), including orders of mammals (wild and domestic) most monitored, laboratory techniques for sample processing used in arbovirus detection, type of habitat (primary forest, secondary forest, intervened # positive records # total samples * 100 Based on these results, we structured the table in Section 3.1 according to each arbovirus identified, in which the fourth column "n" and the fifth column "prevalence %" contain the minimum and maximum values obtained as a result of the calculation presented. For the representation of the positive records of arboviruses in mammals of the Neotropics, we generated several maps to identify the areas in each country where the presence of arboviruses was recorded and the orders of mammals monitored in the Neotropics. The data of the geographic points of the sampling sites with positive records for arboviruses were entered into the ArcGIS 10.5 platform [89]. According to the specifications of each investigation, we used coordinates given in the original paper (n = 21) and for the other, we georeferenced based on the locality names (n = 23).
For the graphical representation of viruses recorded in mammalian orders according to habitat, we used GEPHI 0.9.2 software ( [90] in which one positive record was considered each time an article recorded one positive order for one arbovirus; e.g., if one article recorded three arboviruses in one order, three positive records are considered).
To determine whether arbovirus records are shared among the countries of the Neotropics, the Jaccard and Sorensen similarity indices were calculated to estimate the compositional similarity of arboviruses among the orders and countries of the Neotropics based on the presence/absence data of positive records identified, using the following formulas [91][92][93]: where JI = the Jaccard index; SI = the Sorensen index; A = unique species per site one; B = unique species per site two; and C = the number of species in common between two sites. The two indexes were compared using Pearson's correlation coefficient.
In addition, a Spearman correlation was applied between the two indices for the number of arboviruses per country, as well as for the number of arboviruses per order. This analysis made it possible to identify the countries and orders that are most closely related due to various factors such as: host mammal distribution, migrations, trade, and even health policies [94].
The World Wide Fund for Nature has compiled and analyzed in the shape file the global data on significantly increasing deforestation and degradation trends [95], which were used for visualization purposes [96]. We added the positive arbovirus records for the Neotropics identified in the literature review. The characteristics of the habitat described in the articles, either in front of deforestation or with vegetation cover, were taken into consideration.
Finally, based on the analysis of the publications obtained as a result of the literature review, the characteristics, conditions, or behaviors that increase the likelihood of encountering the top-3 arboviruses, the Saint Louis encephalitis virus (SLEV), the Venezuelan equine encephalitis virus (VEEV), and the West Nile virus (WNV) in the Neotropics were identified [97]. The parameters considered include factors such as deforestation, habitat use change, human and animal migration, climate change, behavioral patterns, altered interactions, surveillance, and conservation programs. The purpose was to provide relevant data to be used as a tool to support arbovirus management and prevention protocols, and to contribute to the knowledge of arbovirus diversity of public health concerns in the region. We identified the odds ratio of a mammal being exposed to an arbovirus according to the vegetation cover and deforested fronts, as well as the positive and negative records of arbovirus identified in the nine orders of mammals.

Prevalence and Detection Methodology
As for the sample size in the studies obtained, it ranges from one single case (case studies) to 2214 individuals. These results show different prevalences among the arbovirus-positive records ranging from 1-100% but, for a small sample size, the 95% confidence interval of the prevalence of arboviruses is wide, so it induced uncertainty.
For the identification of arbovirus-positive records, the most common sample type tested was blood serum (46.67%). Indeed, no study was recorded that performed tests in which animals were not trapped or disturbed. The most used laboratory techniques were hemagglutination inhibition (HI), with 40%, and a plaque reduction neutralization test (PRNT), with 40%, and a real-time polymerase chain reaction qPCR, with 33.33%. A small percentage of studies applied techniques such as a polymerase chain reaction (PCR), with 8.89%, an enzyme-linked immunosorbent assay (ELISA), with 6.67%, and a tissue-based study of histopathology, with 2.22%.

Prevalence and Detection Methodology
As for the sample size in the studies obtained, it ranges from one single case (case studies) to 2214 individuals. These results show different prevalences among the arboviruspositive records ranging from 1-100% but, for a small sample size, the 95% confidence interval of the prevalence of arboviruses is wide, so it induced uncertainty.
For the identification of arbovirus-positive records, the most common sample type tested was blood serum (46.67%). Indeed, no study was recorded that performed tests in which animals were not trapped or disturbed. The most used laboratory techniques were hemagglutination inhibition (HI), with 40%, and a plaque reduction neutralization test (PRNT), with 40%, and a real-time polymerase chain reaction qPCR, with 33.33%. A small percentage of studies applied techniques such as a polymerase chain reaction (PCR), with 8.89%, an enzyme-linked immunosorbent assay (ELISA), with 6.67%, and a tissue-based study of histopathology, with 2.22%.
The similarity analysis for mammalian orders evidenced that according to the Sorensen and Jaccard indexes, the orders that share more arboviruses are Chiroptera and Didelphimorphia (Sorensen index = 61.54% and Jaccard index = 44.44%) ( Tables 4 and 5).
The orders non-human Primates and Rodentia have the highest arbovirus richness (n = 20 and n = 18, respectively). However, no significant difference was found between orders (Kruskal-Wallis equality-of-populations rank test = 14.54, p > 0.05). The orders with the highest number of arbovirus-positive records are Primates (n= 54) and Perissodactyla (n = 45) ( Figure 4). Wildlife habitat conditions influence arbovirus richness in non-human primates while the domestic environment influences more in the order Perissodactyla.
The arboviruses recorded in most mammalian orders in the Neotropics are the Saint Louis encephalitis virus (SLEV) (n = 7), the Venezuelan equine encephalitis virus (VEEV) (n = 6), and the West Nile virus (WNV) (n = 6) ( Figure 5).    The orders non-human Primates and Rodentia have the highest arbovirus richness (n = 20 and n = 18, respectively). However, no significant difference was found between orders (Kruskal-Wallis equality-of-populations rank test = 14.54, p > 0.05). The orders with the highest number of arbovirus-positive records are Primates (n= 54) and Perissodactyla (n = 45) (Figure 4). Wildlife habitat conditions influence arbovirus richness in non-human primates while the domestic environment influences more in the order Perissodactyla.

Habitat Types with Positive Orders and Sampling Conditions According to Order
It was possible to identify that arbovirus records were found in a wild range of sampling habitat types, i.e., primary forest, secondary forest, disturbed areas, captivity, and even in populated areas ( Figure 7). In addition, all mammalian orders that were identified as hosts were found in the wild. However, there were also records both in captive conditions and in the domestic environment ( Figure 8).

Habitat Types with Positive Orders and Sampling Conditions According to Order
It was possible to identify that arbovirus records were found in a wild range of sampling habitat types, i.e., primary forest, secondary forest, disturbed areas, captivity, and even in populated areas (Figure 7). In addition, all mammalian orders that were identified as hosts were found in the wild. However, there were also records both in captive conditions and in the domestic environment ( Figure 8).

Habitat Types with Positive Orders and Sampling Conditions According to Order
It was possible to identify that arbovirus records were found in a wild range of sampling habitat types, i.e., primary forest, secondary forest, disturbed areas, captivity, and even in populated areas (Figure 7). In addition, all mammalian orders that were identified as hosts were found in the wild. However, there were also records both in captive conditions and in the domestic environment ( Figure 8).

Risk Factors
In the Neotropical region, vegetation cover and deforested fronts were identified, as well as the arbovirus positive and negative records that were identified in the nine orders of mammals. The map obtained showed that the arbovirus-positive and negative records for each publication analyzed were mostly found outside the areas with vegetation cover and others were recorded within deforested fronts ( Figure 9 and Table A2). The risk factor map includes positive and negative records. All but four studies (which had only positive records) had both negative and positive records. No studies were found with only negative records. The odds ratio of a mammal becoming being exposed to an arbovirus is 1.46 higher when its habitat is located in deforested fronts (95% confidence interval: 1.34-1.59) than if its habitat is with vegetation cover (p < 0.0001) ( Table 6).

Risk Factors
In the Neotropical region, vegetation cover and deforested fronts were identified, as well as the arbovirus positive and negative records that were identified in the nine orders of mammals. The map obtained showed that the arbovirus-positive and negative records for each publication analyzed were mostly found outside the areas with vegetation cover and others were recorded within deforested fronts ( Figure 9 and Table A2). The risk factor map includes positive and negative records. All but four studies (which had only positive records) had both negative and positive records. No studies were found with only negative records. The odds ratio of a mammal becoming being exposed to an arbovirus is 1.46 higher when its habitat is located in deforested fronts (95% confidence interval: 1.34-1.59) than if its habitat is with vegetation cover (p < 0.0001) ( Table 6).

Richness of Arboviruses
The results of the literature review reported that monitoring efforts have been concentrated in Argentina, Brazil, Mexico, and Panama, which may be associated with the diversity of arboviruses and vectors they harbor [142,143]. These countries share the occurrence of three arboviruses, i.e., VEEV, SLEV, and WNV. SLEV and VEEV are neotropical viruses, and WNV is introduced from Africa. They are viruses with complex transmission cycles, a variety of hosts, and a variety of vectors. In addition, in the Neotropics,

Richness of Arboviruses
The results of the literature review reported that monitoring efforts have been concentrated in Argentina, Brazil, Mexico, and Panama, which may be associated with the diversity of arboviruses and vectors they harbor [142,143]. These countries share the occurrence of three arboviruses, i.e., VEEV, SLEV, and WNV. SLEV and VEEV are neotropical viruses, and WNV is introduced from Africa. They are viruses with complex transmission cycles, a variety of hosts, and a variety of vectors. In addition, in the Neotropics, they are distributed in rural areas, jungle areas, and a few cities on the periphery. SLEV and WNV are Flaviviruses and in humans they exist as cross-protections with the antibodies of the population against dengue and yellow fever, which is why they do not occur in the Neotropics. Dengue and Chikungunya are introduced from Africa and transmitted by the vector Aedes aegypti; they are part of an exact transmission cycle in urban areas in the Neotropics [110,144].
Among the countries with arbovirus records are Guatemala, Costa Rica, French Guiana, Paraguay, and Uruguay. In most cases, the pathways for the introduction of arboviruses into new regions are unknown [145]. However, the emergence and spread of arboviruses rapidly and geographically may be due to the growth of global transportation systems [146][147][148][149][150][151] and the adaptation of humans and arthropods due to increasing urbanization [52]. The countries that obtained the highest number of arbovirus similarity are geographically close, such as Colombia and Panama, which share the biogeographic region of El Choco, and Mexico and Guatemala with shared ecosystems, thus creating the possibility of harboring similar arbovirus species in the same ecozone [152,153].
Recently, arbovirus diseases have been reported with increased frequency worldwide [154]. In the Neotropics, Brazil is characterized by encompassing a large land area covered by tropical forests and densely populated areas [155,156]. These characteristics are closely linked to the strengthening of research efforts according to the number of resources invested for research and development in arbovirus research areas [157]. Similarly, surveillance preferences for specific species and the ease of sampling sites favor the detection of arbovirus prevalence due to the high rates of the infected population [158]. These conditions coincide with the results obtained from the systematic review carried out, which reflect that Brazil is the country with the highest arbovirus richness in the region (n = 27). Due to the high biodiversity of environments and components, a large number of arboviruses have been isolated in Brazil [67,134,135], especially those involved in human diseases such as the Western equine encephalitis virus [98], Saint Louis encephalitis virus [125], Mucambo virus [159], Guaroa virus [99], Tacaiuma virus [102], and Guama virus [106].
The ability of transmission vectors to spread is a determining factor for arbovirus outbreaks worldwide. A previous systematic review of arboviruses in Western Europe reflected that current outbreaks are due to the spread of Aedes albopictus and Aedes japonicus [160]. Dengue is the most important emerging arboviral disease globally [161,162] due to the wide variety of ecosystems in which it is found and the ease of its spread. Studies have also focused on other arboviruses [163]. An example of this is the VEEV [164], whose natural and most efficient vector in Latin America [165] is Culex sp. Melanoconion [166,167]. The VEEV is recorded in human [168], equine, and bovine serology studies [126,169] and wild animals [101,170], so it tends to replicate in livestock animals and results in higher levels of contagion in rural environments [171].
Changes in ecological conditions favor the creation of new habitats for arbovirus vectors [132,172,173] and may cause arthropods to adapt to new mammalian hosts [52,174], leading to the emergence of new pathogens in the domestic environment, which is sometimes the main reservoir [175]. Infection in domestic animals can increase circulation and human exposure in peridomestic habitats [168,176].
In the Neotropics, the SLEV is found in a wide distribution from Mexico to Argentina via the mosquito vector Culex sp. [177]. Initially, outbreaks were localized in the United States with high human case fatality rates [178]. In Argentina, these records are attributed to the expansion of agricultural and urban habitats [179]. Based on the records of this study, the occurrence of SLEV infection was evidenced in countries of the region such as Costa Rica [107], Guatemala [120], Mexico [104], and Uruguay [125,177].
West Nile virus has the ability to infect a wide variety of wild and captive mammals in all regions of the world [180]. Historically, WNV has one of the broadest host ranges [181][182][183], and mosquitoes act as vectors, e.g., Culex sp. [184] of arboviruses in humans and equines [185]. However, very few studies associated with infection in mammalian hosts have been conducted during the last decades [181]. Due to the wide distribution of WNV and the association of records of increased prevalence in mammals in urban areas [186], there is a possibility of new outbreaks in most continents [187][188][189][190][191].

Prevalence and Detection Methodologies
The variability in the number of individuals monitored in this review may be linked to the objectives of each study, the sampling effort, the order of mammals sampled, the capture techniques [192], the type of sample analysis [193], and the characteristics of the site [194].
Different types of methodologies were used to capture the mammals studied. For rodents and marsupials, Sherman traps were used [113] and mist nets were used to capture bats [109,120]. In relation to NHP, dart immobilization [103,115] and manual capture [107] were used. These techniques were used to obtain blood, serum, or tissue samples from the vertebrate host, which in some cases studies involved the sacrifice of species, such as rodents and marsupials [101,113,121,132].
In the three arboviruses with the highest number of records, both serological virus isolation and molecular techniques were used. This allows a better confirmation of the presence of arboviruses in individuals [101,118,124,141] due to the materials and equipment necessary for the correct execution of the analysis protocols for each sample, either under culture, molecular, or serological techniques. The most applied detection methodologies for arbovirus detection were hemagglutination inhibition by its capacity or facility and plaque reduction neutralization test. The combined use of serological and molecular techniques facilitates the indirect detection of arboviruses. For example, the simultaneous application of techniques, such as ELISA and RT-PCR, allowed the identification of co-infection of dengue with other arboviruses, such as Chikungunya and Zika, respectively [195].
In the last decade with the significant development of new molecular detection technologies in epidemiological surveillance, there are still few studies using PCRs or qPCRs for arbovirus species determination (16/46) [100,101,196]. However, according to Mendoza-Ponce, Corona-Núnez, Galicia, and Kraxner [16] worldwide, qPCR is highly effective for diagnosing arboviruses in humans, even with low viremias. The accuracy of this technique [65] is important in wildlife, as samples cannot always be repeated or individuals have already died. This methodology is sensitive and specific and should be used in wildlife. Phylogenetic studies that would complement detection studies are not widely present [108,113]. Currently, with the COVID-19 pandemic, the importance of having more specific tools, such as sequencing and metagenomics, at hand for proper species detection and determination of outbreak origins were observed [197,198]. Additionally, advances in equipment mean that samples can be taken and analyzed in the field, reducing data loss [199].
The results of the literature review showed that most of the records were identified in secondary forest mammals, which is associated with previous research showing that areas with greater intervention have greater potential for zoonotic diseases, which could act as a potential danger to surrounding communities [55]. For example, in Singapore, it was reported mosquitoes of the genus Aedes could be both in the forest and in urban open areas that are highly frequented by people [200]. Similarly, in Brazil, it was reported there were a high abundance of vectors in urban forests and a dominance of vector species according to habitat [172].
The use of non-invasive samples is an opportunity that could facilitate the diagnosis and detection of arboviruses in animals using urine and saliva samples [201][202][203]. Detection using fecal samples is another cost-effective and non-invasive option to monitor wild populations that could be potential reservoirs of arboviruses [204]. The use of this type of sample is effective for arbovirus detection and is an option to avoid stress on monitored species. Similarly, research focused on this sampling protocol could increase the number of orders that currently have not been monitored and contribute to conservation genetics, as well as behavioral ecology and infectious and parasitic diseases [205][206][207].
It is important to note that, although this study reflects the reality of arbovirus monitoring in the neotropics, there is a lack of study of negative records (zero prevalence). Unfortunately, few studies on infectious and parasitic diseases publish negative results [208]. This study confirms what others have already suggested regarding the publication of positive results [198,209]. There are more publications with positive results than negative ones. In this study, we found no publications with completely negative results. This is called publication bias [210]. Negative results are important because they may have an ecological, behavioral, or management explanation [211], which can directly contribute to our knowledge of infectious and parasitic diseases and contribute to epidemiological monitoring plans or policies. Additionally, in some neotropical countries, most research remains unpublish for economic reasons or publication rejections [212], which favors the perception that it is better not to publish negative results. Publication of negative results is highly recommended.

Risk Factors
Climate change implies impacts on human health and vector-borne infectious disease burden [213]. Fluctuating climatic conditions such as precipitation [214], temperature [215], and humidity [216] impact infection rates in tropical regions [217,218] that facilitate arbovirus reproduction and transmission in a wide diversity of habitats [156]. Patterns of vector population distribution, reproduction, and competition [219] can be influenced by temperature [220], precipitation [221], abundance, and the affinity of the vector for a mammalian host species [222]. Variability among hematophagous arthropod species [75], mammalian hosts [223], and the environments or niches they occupy [224] impacts arbovirus transmission dynamics [37,192] and are subject to changes in temperature and precipitation [186,225].
The current co-circulation of three arboviruses, Zika, dengue, and Chikungunya, spreading globally in the Americas suggests the need for more integrative studies and the use of new approaches to identify the cause and risk posed by these combinations [226]. González-Salazar, Stephens, and Sánchez-Cordero [221] even created a model of possible mammalian hosts of Zika for the Neotropical region where seven of the nine species identified were bats. As for Chikungunya, this was not recorded in any order of nonhuman mammals in the records of the Neotropics, although a study in Brazil carried out serological and molecular analyses of bats without any detection [227]. Based on this background, the question now arises whether Chikungunya and Zika have been found in non-human hosts in America for their maintenance cycle, similar to what occurred with the yellow fever virus and the Mayaro virus, which originated in Africa. Currently, incorrect diagnoses are made based on known symptoms in areas where several arboviruses are circulating simultaneously [228].
On the other hand, in countries in other regions of the world there have already been records of this arbovirus in NHP [229], rats [230], bats [230], and horses [231]. Monitoring and detection should be continued in mammals such as primates, rodents, and bats that are susceptible to infection [232], even without the intervention of a vector [233].
Arbovirus circulation can occur in sylvatic or urban cycles [234]. Thus, altering the balance of natural systems can increase vector abundance, create new reservoirs, or induce arboviruses to adapt to new maintenance cycles [235]. Additionally, the relationship that humans maintain with various species of domestic mammals is considered another factor that favors the reproduction cycles of arboviruses and their vectors [100,236].

Socio-Ecological Aspects
Records of arboviruses in new areas are linked to the distribution and spread patterns of vectors, such as Aedes aegypti, from Africa to the Neotropics due to the influence of human trade routes [237]. Social phenomena, such as human migration and species trafficking, are key factors that have favored the spread of arboviruses [60]. Factors that determine vector trajectory conditions are established in various regions [238]. These factors are associated with climate change [239], deforestation [240], uncontrolled growth of urban areas [241], difficulty in accessing clean water sources [242], and population displacement [243]. In the Brazilian Amazon region, there are a greater number of positive records of sylvatic arboviruses associated with deforestation [155], mining [244], road expansion [245], and urbanization with the emergence and/or re-emergence of relevant arboviruses [172].

Ecological Aspects
Environmental factors and environments are determinants in the life cycles of arboviruses as they determine their distribution and dispersal patterns, as well as their transmission to mammalian hosts [40]. The interaction between vectors and mammalian hosts conditions the dynamics and impact of arboviruses in human and domestic animal communities [32].
Arboviruses have a high host-specific association, so changes in range or distribution significantly influence their adaptations in new areas [176]. For example, as shown in Figure 7, NHP is the order identified in the greatest variability of habitats in primary forests, secondary forests, disturbed areas, captivity, and populated areas. Non-human primates are reservoirs for a large number of blood-borne pathogens and ecological factors such as host density, climate change, and activities facilitate the transmission of these pathogens [207].
As a survival mechanism, arboviruses have the ability to develop adaptive mutations when they reach new territories new hosts, vectors, and environments to adapt quickly and improve transmission. Thus, positive records of arboviruses are reflected in variability of sampling conditions, including domestic, wild, and zoo, which can be seen in Figure 8, with the wild condition being the most frequent in all orders. However, in the face of changes in the environment, it is difficult to predict the speed of species response [225,246], so research on the ecology of interactions between arboviruses and mosquito vectors is needed to understand the dynamics of invasion and adaptation in new areas [165].
Ecological mechanisms are an important part of every stage and enzootic potential of the Neotropics, and we must consider within them the exposure of wildlife, the pressures of propagation, the enzootic infection that affects the exposed animals, and the persistence of enzootic transmission [63]. The impact of deforestation on the abundance of vectors, which facilitate arbovirus transmission, influences their movement from wild areas to urban or rural areas [247] where they can adapt to new domestic hosts that maintain direct contact with humans. Figure 9 reflects that the effects of deforestation in forested areas of the Neotropics are more evident over time and increase the contact of the wide variety of vectors with humans, and thus the risk of contagion, dispersal, and epizootics [248,249].

Health Policies
Population growth [250], the expansion of the agricultural frontier [251], the impact of anthropogenic activities [252], and climate change [253] contribute to the spread of arbovirus infections [254]. The 2015 Lancet Commission on Health and Climate Change mentions that the fight against climate change could be the greatest opportunity for global health in the twenty-first century [255].
The levels of epidemiological and entomological surveillance should be potentiated, as well as the joint analysis of the factors that condition the level of vulnerability of a certain area to arboviral diseases [256]. Success in the execution, monitoring, and evaluation of programs focused on the prevention and control of emerging diseases will mitigate the spread of arboviruses in the Neotropics.
At the international level, arthropod vector-borne arboviruses are part of the public health problem that requires cooperation and joint research to establish effective control strategies. All the parameters that are part of the biology of the vectors must be taken into account, as well as the dynamics of reproduction of the arboviruses [257]. Institutions such as the Fundação de Medicina Tropical Dr. Heitor Vieira Dourado FMT-HVD in Brazil work on the detection of arbovirus infections, considering that the country has suffered silent outbreaks due to problems with the identification of arbovirus [156].
The World Organisation for Animal Health (OIE) seeks to promote the importance of animal disease surveillance and communication systems. By generating the necessary knowledge, it will be possible to manage risks, evaluate priorities and generate policies according to the guidelines of each country [258]. Additionally, eradication programs [259,260] consider the social, economic, and even political part of each zone in order to mitigate or control arbovirus infections in the Neotropics.

Actions to Take
The need to strengthen research, surveillance programs, and public policies is indispensable in the Neotropics. The identification of possible regions and habitats with ecological conditions suitable for the circulation of arboviruses and/or specific areas with a high risk of infection is required [261]. In addition, climate change-based modeling associated with infectious diseases supports early warning systems [262].
There should be an increase in the use of non-invasive techniques. The protocols used for sampling in each publication analyzed in the systematic review do not establish noninvasive methodologies, even though several species of mammals that registered positive arboviruses fall within a threat category on the International Union for Conservation of Nature's (IUCN) red list.
Due to the complexity of transmission and contagion dynamics, as well as the biology of vectors and hosts, it is considered that an integrated solution would allow effective vector control through new technologies, adequate management of space and resources, and control and sanitation policies [263]. For example, in the last decade, the use of geographic information systems (GIS) has become an important tool for the detection, analysis, and prediction of epidemiological patterns that have contributed to the prevention and control plans of diseases caused by arboviruses [264]. In the case of vectors that transmit arboviruses, a biotechnological tool has been generated that is helping to control mosquito vectors by genetically modifying them. It is highly effective, does not harm the environment, is efficient, and mainly has a low production cost [265]. The use of insecticides, on the other hand, has shown that it has created resistance in mosquitoes of the genus Aedes [266] that have developed resistance against insecticides. These mosquitoes are transmitters of different arboviruses such as dengue [267], yellow fever [268], Chikungunya [269], and Zika [270].
Long-term monitoring of the interactions of nearby populations is necessary to prevent the spread of arboviruses to uncontaminated locations [271]. It is essential to strengthen epidemiological monitoring in the areas that maintain research initiatives for the control of arboviruses in the Neotropics. Considering that knowledge is the basis for the prevention and control of emerging diseases, in areas that do not yet maintain plans and/or policies of epidemiological monitoring, basic but transcendental measures should be considered, such as training the general population for the elimination of spaces that could serve as potential reservoirs for vectors [272]. Similarly, health education campaigns should be strengthened [273] and encourage research for the early detection of arboviruses, mainly in the areas most likely to be infected [274].

Conclusions
This study identified the diversity of arboviruses in the Neotropics region, within the list of the 17 countries that make it up. We recorded a greater sampling effort in 11 countries (Argentina, Brazil, Colombia, Costa Rica, Guatemala, French Guiana, Mexico, Panama, Paraguay, Uruguay, and Venezuela). However, there are countries in which no effort has been made, despite the great biodiversity of the area and the wide range of host mammals.
The Venezuelan equine encephalitis virus (VEEV), St. Louis encephalitis virus (SLEV), and the West Nile virus (WNV) share the most mammalian orders. In addition, it was identified that there are orders with greater sampling effort that can be associated with ease in terms of the sampling technique and the objectives of each study. The prevalence of arboviruses (1-100%) varies due to the sample size of each study, as cases range from 1 to 2214 individuals found in various habitats and conditions. As the present research highlights the record of the Mayaro virus in mammalian hosts and the influence of climate change that facilitates the creation of new environments and the adaptation of vectors, it would be important to focus the monitoring effort on determining if we can consider the Mayaro virus as the next emerging arbovirus given its phylogenetic closeness to the Chikungunya virus [228,275,276]. Furthermore, in the case of Mayaro, due to environmental changes, contact between peri-urban and urban areas is increasingly possible. Finally, experimentally, competition between three vectors (especially Ae. aegypti and Ae. albopictus, and to a lesser extent Cx. quinquefasciatus) has been observed, which means that these vectors may also play a role in the circulation of Mayaro [277][278][279].
Deforestation was observed as an important risk factor in terms of the observed records, as a large number of positive records for arbovirus number of publications are found outside forested areas and within deforested fronts of the Neotropics, which is associated with the fact that disturbed habitats increase the risk of infection [73].
The studies analyzed in the review maintain conventional analysis techniques, such as serology. However, the use of molecular detection tools, e.g., PCR and qPCR, is a priority to have an adequate response in the case of outbreaks. Response tools require monitoring in wild areas to be able to relate outbreaks to origins, as well as the application of techniques that provide more specific information, such as sequencing and metagenomics for the detection of arboviruses.
The orders with the greatest monitoring effort are non-human primates, Perissodactyla, Chiroptera, and Rodentia. In addition, research gaps were identified. Future research should focus on the orders Artiodactyla, Carnivora, Chiroptera, Didelphimorphia, Lagomorpha, Perissodactyla, Pilosa, NHP, and Rodentia, which are important sources of information for disease monitoring. In biodiverse areas, such as Amazonian ecosystems, there are information gaps, so research should focus on the dynamics of emerging diseases and local fauna as a monitoring tool for conservation [245,280].
Interactions between factors such as environment, hosts, and vectors are a potential risk to disease prevalence. Long-term and constant monitoring is required, accompanied by monitoring and sampling techniques focused on methodologies with non-invasive techniques that are cost-effective and provide the same results without altering the dynamics of populations or the health of individuals.
Health education through the generation of educational instruments and constant training for the population is a good option on a smaller scale that could complement the monitoring and prevention initiatives of public policies in each country. Anthropic effects have a direct impact on the factors that determine the trajectory of vectors, the distribution ranges of host mammals, and the distribution of arboviruses in the Neotropics.