����������
�������

Citation: Ortiz, D.I.; Piche-Ovares,

M.; Romero-Vega, L.M.; Wagman, J.;

Troyo, A. The Impact of

Deforestation, Urbanization, and

Changing Land Use Patterns on the

Ecology of Mosquito and Tick-Borne

Diseases in Central America. Insects

2022, 13, 20. https://doi.org/

10.3390/insects13010020

Academic Editor: Yuval Gottlieb

Received: 16 November 2021

Accepted: 16 December 2021

Published: 23 December 2021

Publisher’s Note: MDPI stays neutral

with regard to jurisdictional claims in

published maps and institutional affil-

iations.

Copyright: © 2021 by the authors.

Licensee MDPI, Basel, Switzerland.

This article is an open access article

distributed under the terms and

conditions of the Creative Commons

Attribution (CC BY) license (https://

creativecommons.org/licenses/by/

4.0/).

insects

Review

The Impact of Deforestation, Urbanization, and Changing Land
Use Patterns on the Ecology of Mosquito and Tick-Borne
Diseases in Central America

Diana I. Ortiz 1,*,†, Marta Piche-Ovares 2,3 , Luis M. Romero-Vega 4,5, Joseph Wagman 6 and Adriana Troyo 5,7,†

1 Biology Program, Westminster College, New Wilmington, PA 16172, USA
2 Laboratorio de Virología, Centro de Investigación en Enfermedades Tropicales (CIET),

Universidad de Costa Rica, San José 11501, Costa Rica; maria.piche.ovares@una.cr
3 Departamento de Virología, Escuela de Medicina Veterinaria, Universidad Nacional,

Heredia 40104, Costa Rica
4 Departamento de Patología, Escuela de Medicina Veterinaria, Universidad Nacional,

Heredia 40104, Costa Rica; LUIS.ROMEROVEGA@ucr.ac.cr
5 Laboratorio de Investigación en Vectores (LIVe), Centro de Investigación en Enfermedades Tropicales (CIET),

Universidad de Costa Rica, San José 11501, Costa Rica; adriana.troyo@ucr.ac.cr
6 Malaria and Neglected Tropical Diseases Program, Center for Malaria Control and Elimination, PATH,

Washington, DC 20001, USA; jwagman@path.org
7 Departamento de Parasitología, Facultad de Microbiología, Universidad de Costa Rica,

San José 11501, Costa Rica
* Correspondence: ortizdi@westminster.edu
† These authors contributed equally to this work.

Simple Summary: Central America is a region that possesses distinct ecological and socioeconomic
characteristics, making it increasingly vulnerable to vector-borne diseases. The emergence and
resurgence of these diseases has been linked to environmental changes driven by human activities,
particularly land use changes associated with deforestation, forest degradation, and urbanization.
However, the effects of these environmental modifications on the transmission dynamics and the
increase of infection risks are not well understood in Central America where information is limited
and scattered. In this article, we review and analyze the current knowledge and potential impacts of
deforestation and urbanization on the risk and transmission dynamics of the most relevant mosquito-
borne and tick-borne diseases in Central America. Disease events, such as the recent Zika and
dengue epidemics, and the uneven progress towards regional malaria elimination highlight the need
to increase awareness regarding the complex ecological interactions and environmental changes
taking place in this region and how this information could be used to improve prevention and
control strategies.

Abstract: Central America is a unique geographical region that connects North and South America,
enclosed by the Caribbean Sea to the East, and the Pacific Ocean to the West. This region, encom-
passing Belize, Costa Rica, Guatemala, El Salvador, Honduras, Panama, and Nicaragua, is highly
vulnerable to the emergence or resurgence of mosquito-borne and tick-borne diseases due to a com-
bination of key ecological and socioeconomic determinants acting together, often in a synergistic
fashion. Of particular interest are the effects of land use changes, such as deforestation-driven ur-
banization and forest degradation, on the incidence and prevalence of these diseases, which are not
well understood. In recent years, parts of Central America have experienced social and economic
improvements; however, the region still faces major challenges in developing effective strategies and
significant investments in public health infrastructure to prevent and control these diseases. In this
article, we review the current knowledge and potential impacts of deforestation, urbanization, and
other land use changes on mosquito-borne and tick-borne disease transmission in Central America
and how these anthropogenic drivers could affect the risk for disease emergence and resurgence
in the region. These issues are addressed in the context of other interconnected environmental and
social challenges.

Insects 2022, 13, 20. https://doi.org/10.3390/insects13010020 https://www.mdpi.com/journal/insects

https://doi.org/10.3390/insects13010020
https://doi.org/10.3390/insects13010020
https://creativecommons.org/
https://creativecommons.org/licenses/by/4.0/
https://creativecommons.org/licenses/by/4.0/
https://www.mdpi.com/journal/insects
https://www.mdpi.com
https://orcid.org/0000-0003-4809-0095
https://orcid.org/0000-0001-9513-9969
https://doi.org/10.3390/insects13010020
https://www.mdpi.com/journal/insects
https://www.mdpi.com/article/10.3390/insects13010020?type=check_update&version=1


Insects 2022, 13, 20 2 of 44

Keywords: arbovirus; malaria; deforestation; urbanization; Rickettsiales; Culicidae; Ixodidae;
Central America

1. Introduction

Vector-borne diseases (VBDs) remain an important public health problem worldwide,
particularly in tropical and subtropical regions, and they are becoming more prevalent in
recent years. Arthropod vectors are associated with the transmission of some of the most
significant infectious diseases affecting both animals and humans [1–3]. The global burden
of VBDs is significant, accounting for more than 17% of infectious diseases in humans
with more than three billion people currently inhabiting endemic areas and at risk of
exposure to these pathogens [4]. Most people affected by these diseases live in developing
countries under conditions that favor a greater burden of disease, especially in poor and
marginalized populations, including rural inhabitants, Indigenous populations, women
living in poverty, the elderly, and children [5,6]. Together, these diseases produce significant
mortality and morbidity, causing millions of deaths every year, long-term disabilities, and
life-long sequelae [5,7–9].

Emerging and resurging diseases are defined as recently evolved or newly discovered
pathogens that have demonstrated increased incidence in host populations in the past
20 years or poses a future threat. These pathogens are characterized by their expanding
geographic spread, increasing public health impact, changes in their clinical presentation,
or novel infection occurrence in humans [10,11]. Some of these pathogens are also char-
acterized by their resurgence after long periods of decline in infection incidence. About
60% of these diseases are zoonotic in origin and, to various degrees, dependent on ani-
mal reservoirs for survival and maintenance [10–14]. In many countries, the incidence of
these diseases has declined to low levels due mostly to effective prevention and control
programs; however, some were never satisfactorily controlled throughout their endemic
regions. Many are currently increasing in incidence and spreading beyond their previously
known geographical ranges. Some of them have reappeared only in limited regions while
others have become major global problems [9,12,15].

Most emerging zoonotic VBDs are transmitted by ticks (Family Ixodidae) and mosquitoes
(Culicidae) and caused by RNA viruses (Families Flaviviridae, Bunyaviridae, and Togaviri-
dae) and Rickettsiaceae bacteria [12,16]. The vectorial capacity of mosquitoes and ticks
is enhanced by their high environmental adaptability, which includes high reproductive
outputs under suitable conditions and high capacity to invade ecologically disturbed envi-
ronments, especially peridomestic habitats where human and domesticated animal hosts
are readily available [16–18]. In terms of human morbidity and mortality, malaria, dengue,
Rickettsial fevers, and Lyme disease are some of the most important of these resurgent
infections [2,15,16,19].

The relationship between arthropod vectors and the pathogens they transmit is partic-
ularly sensitive to anthropogenically driven global changes. Factors behind the dramatic
emergence and resurgence of VBDs are complex, vary geographically and temporarily, and
often have an additive effect on disease ecology and epidemiology [1,2,12,16,19]. Currently
recognized anthropogenic drivers of VBD emergence and resurgence include demographic
changes (e.g., global population movements and growth, unplanned and uncontrolled
urbanization), socioeconomic changes (e.g., modern transportation and commerce, human
encroachment on natural disease foci), illegal activities (e.g., illegal logging and cattle ranch-
ing, illegal drug trafficking), accelerated exploitation of natural resources (e.g., changes
in land use, forest degradation, reduction in biodiversity, agricultural practices), changes
in host susceptibility and pathogen adaptation (e.g., increased movement of humans and
animals, pathogen genetic variability), degradation of public health infrastructure (e.g.,
lack of effective vector control, disease surveillance, and prevention programs), and climate



Insects 2022, 13, 20 3 of 44

change (e.g., changes in regional temperature and rainfall patterns lead to alterations in
vector dynamics) [1,2,12,16,19–22].

There is increasing evidence that anthropogenic land use changes can directly and
indirectly influence vector-borne pathogen transmission dynamics [10,23]. Although land
use changes, such as urbanization as a result of forest degradation and deforestation,
have been associated with increased disease transmission, their direction, extent, specific
mechanisms, and persistence are not clearly understood. Often the effects of these factors
on VBD emergence and resurgence are interdependent, synergistic, and difficult to study.

Central America is the region that links North and South America comprising most of
the narrow isthmus that separates the Pacific Ocean from the Caribbean Sea. It consists
of the countries of Belize, Costa Rica, Guatemala, El Salvador, Honduras, Panama, and
Nicaragua, and is inhabited by about 44.5 million people. Generally, this region is character-
ized by a diversity of ecosystems, physical geographies, sociocultural and socioeconomic
structures, and public health profiles [24]. Over the past decades, this region has sustained
dramatic land use changes, including cattle ranching, large-scale commercial plantations,
illegal airstrips, mining, road construction, tourism infrastructure, illegal timber extraction,
and housing construction. These land use changes have significantly accelerated the rates
of deforestation and urbanization in the region [25–30]. Despite the high prevalence of
several VBDs in Central America, the effects of anthropogenic-driven deforestation and
urbanization on the transmission dynamics of these diseases are not well understood. The
presence and interactions of numerous physical and socioeconomic factors in the region
amplifies its vulnerability to the emergence and resurgence of several VBDs [6].

The purpose of this descriptive review is to profile the potential impact of defor-
estation, forest degradation, and urbanization on mosquito-borne and tick-borne disease
transmission dynamics in Central America and how these anthropogenic drivers could
affect risk for disease emergence and resurgence in the region. This review article focuses
specifically on mosquito-borne and tick-borne pathogens currently causing high disease
incidence and prevalence in humans or that have a high potential for emergence or resur-
gence in the region. To provide a proper context to these issues, our review also includes
a brief history on deforestation and urbanization in Central America and the mosquito-
borne and tick-borne diseases that have been recorded in the region. These diseases have
received less attention than other VBDs in Central America, such as leishmaniasis and
Chagas disease, which have been widely studied in the region and have been featured in
comprehensive literature reviews [31–35]. This review assesses direct and indirect evidence
and identifies knowledge gaps that could stimulate future research initiatives in the region.
Understanding the causes and potential impacts of deforestation and urbanization on
VBD transmission in Central America is essential to the improvement of current disease
prevention and control approaches.

2. Deforestation in Central America

The Central American region is about 2200 km in length, northwest to southeast,
and 600 km wide at its broadest point covering a total area of more than 550,000 square
kilometers. The topography and vegetation in this region are defined by several mountain
ranges that stretch extensive parts of the region’s length. Between these mountain ranges
lie fertile valleys where most of the populations reside and where most of the agricultural
activity occurs, such as raising livestock and the cultivation of coffee, beans, tobacco,
and other crops [24,25]. Central America is also part of the Mesoamerican Biodiversity
Corridor (MBC), a patchwork of diverse and protected biomes, established in 1997, that
connects North and South America. It represents the world’s third largest biodiversity
hotspot containing about 7–10% of the world’s known species [36]. Although Central
American forests currently cover about 200,000 square kilometers, they were once much
more extensive. About 12% of the MBC is protected land in the form of ecoregions and
nature reserves [36,37]. Mesoamerican forests are highly susceptible to destruction and
damage and have one of the highest rates of ecological degradation in the world [38].



Insects 2022, 13, 20 4 of 44

However, the history and drivers of deforestation have taken different forms depending on
the country and region [22,25,26].

In the 1960s and 1970s, Central America underwent the highest rate of deforestation
in the world with an increasing number of settlers clearing the land for cattle ranching (the
“hamburger connection”) and commercialization activities [25,39,40]. Over the last 30 years,
Central America has experienced a rising demand for food and energy driving national
authorities to exploit natural resources for energy generation and agricultural production.
Numerous drivers of deforestation and forest degradation have significantly accelerated
the pace of net forest loss in this region, including land settlements, logging, illegal cattle
ranching, large-scale agriculture (e.g., coffee and palm oil plantations), and subsistence
farming [1,22,27,41–45]. In the last two decades, the three largest surviving forest segments
in Central America have shrunk in size by about 23% and are limited to a few pockets of
old-growth forests mostly bound by international borders and heavily urbanized regions.
Within these forests, there are indigenous populations and biodiverse ecosystems; however,
as climate change intensifies, urban areas expand, and clearing for farmland continues,
these forests will continue to shrink [28,38].

From 2001 to 2010, an average of 5376 square kilometers (2076 sq mi) of forest dis-
appeared in the region. Currently, the percentage land area covered by forests in Central
America varies by country, with the highest in Costa Rica (58.8%), followed by Honduras
(57.2%), Panama (57.1%), Belize (57%), Guatemala (33.1%), Nicaragua (30%), and El Sal-
vador (28.6%) (Table 1) [46]. Most of the deforestation in Central America is in the moist
forest biome of the Caribbean slopes of Nicaragua [44]. Recent reports have revealed that
over 90% of forest loss in Central America is due to extensive illegal cattle ranching with
most of it occurring in Indigenous territories and protected areas. This illegal activity is
often connected to money laundering and drug trafficking [22,38,45]. Other agricultural
activities, such as the proliferation of oil palm plantations, have displaced cattle and people
into protected areas, which further accelerates deforestation in the region. Regions such
as La Mosquitia in Nicaragua and Honduras and the Maya Forest region located between
Mexico, Guatemala, and Belize are under the greatest threat [22,38].

Table 1. Current proportions of forest area, agricultural land, and urban population in Central
America, including changes over the last 30 years.

Country
a Total

Country Area
(sq. km.)

b Forest Area,
% of Land

Area
(2018)

c Change in
Forest Area, %
of Land Area
(1990–2018)

d Agricultural
Land, % of
Land Area

(2018)

e Change in
Agricultural
Land, % of
Land Area
(1990–2018)

f Urban
Population %

(2020)

g Change in
Urban

Population %
(1990–2020)

Belize 22,810 57 −13.2 7.5 +2 46 −1.44
Costa Rica 51,060 58.8 +1.9 34.9 −8.1 80.8 +30.8
El Salvador 20,720 28.6 −6.1 71.4 +6.2 73.4 +24.2
Guatemala 107,160 33.1 −11.5 36 −4 51.8 +9.8
Honduras 111,890 57.2 −5.2 30 +0.3 58.4 +17.9
Nicaragua 120,340 30 −23.2 42.1 +8.6 59 +5.9

Panama 74,177 57.1 −4.9 30.5 +1.9 68.4 +14.5
a Data source: The World Bank https://data.worldbank.org/indicator/AG.LND.TOTL.K2?view=map (accessed
on 18 August 2021). b Data source: The World Bank https://data.worldbank.org/indicator/AG.LND.FRST.ZS?
view=map (accessed on 18 August 2021). c Data source: The World Bank. Value calculated by determining the
difference in forest area (% of land) between 1990 and 2018 per country https://data.worldbank.org/indicator/AG.
LND.FRST.ZS?view=map (accessed on 18 August 2021). d Data source: The World Bank https://data.worldbank.
org/indicator/AG.LND.AGRI.ZS?view=map (accessed on 18 August 2021). e Data source: The World Bank.
Value calculated by determining the difference in agricultural land area (% of land) between 1990 and 2018 per
country https://data.worldbank.org/indicator/AG.LND.AGRI.ZS?view=map (accessed on 18 August 2021).
f Data source: The World Bank https://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS?view=map (accessed
on 18 August 2021). g Data source: The World Bank. https://data.worldbank.org/indicator/SP.URB.TOTL.IN.
ZS?view=map. Value calculated by determining the difference in urban population % between 1990 and 2020 per
country (accessed on 18 August 2021).

Deforestation rates in protected areas differ among environmental governance models
and intensity of human activities [47–49]. Recent studies indicated that drug trafficking and
related criminal activities have significantly contributed to forest loss in Central America

https://data.worldbank.org/indicator/AG.LND.TOTL.K2?view=map
https://data.worldbank.org/indicator/AG.LND.FRST.ZS?view=map
https://data.worldbank.org/indicator/AG.LND.FRST.ZS?view=map
https://data.worldbank.org/indicator/AG.LND.FRST.ZS?view=map
https://data.worldbank.org/indicator/AG.LND.FRST.ZS?view=map
https://data.worldbank.org/indicator/AG.LND.AGRI.ZS?view=map
https://data.worldbank.org/indicator/AG.LND.AGRI.ZS?view=map
https://data.worldbank.org/indicator/AG.LND.AGRI.ZS?view=map
https://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS?view=map
https://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS?view=map
https://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS?view=map


Insects 2022, 13, 20 5 of 44

since the early 2000s [22,45,50]. As a result of successful and disruptive U.S.-led interdiction
activities in the Caribbean and Mexico, illegal drug traffickers were forced to diverge
cocaine shipments through Central America, which is currently the primary trafficking
corridor for cocaine between South to North America [50–52]. Presently, about 86% of
the cocaine trafficked worldwide is transported across Central America, leaving billions
of dollars in annual illegal profit in the region. Moreover, approximately 10% to 14%
of the gross domestic product of Nicaragua, Honduras, and Guatemala, a major drug
corridor in the region, is linked to illegal drug trafficking [22,45,52]. In these three countries,
where most of the Central American forest loss has occurred, drug trafficking along with
other illicit trades are increasingly cited as principal drivers of environmental degradation,
accounting for about 25% of all forest loss since the mid-2000s [53–55]. These regions are
characterized by poor socioeconomic development located in remote forests that are highly
vulnerable to deforestation [45,50,52,54].

Effective drug smuggling territories are characterized by their remoteness and pro-
tected forests status cutting across international terrestrial and marine borders and are
seldom monitored by national drug enforcement agencies. Moreover, these smuggling
territories typically have weak civil governance structures, insecure land tenures, high
unemployment, and are frequently controlled by low-resource conservation groups and
agencies that are highly susceptible to undermining and exploitation by criminal orga-
nizations [22,45,54,56–58]. The increasing use of these protected areas for illicit activities
diminishes the region’s capacity for conservation of forest cover and biodiversity that
could help reduce the effects of climate change in the region [29,45]. When cocaine is
trafficked through Central America, money is laundered through the conversion of forests
to agricultural land in order to legitimize illicit profits in the legal economy. Money
laundering activities linked to deforestation include illegal cattle ranching and timber
extraction, clandestine airstrips, and mining. The most destructive of these illegal activities
is cattle ranching [22,38,45,50,53,55,59]. Protected and remote areas highly impacted by
drug trafficking and related illicit activities include the Petén and Nicaragua’s Caribbean
Coast, Guatemala’s Maya Biosphere Reserve, and Honduras’s Rio Plátano Biosphere Re-
serve [44,45,55,57,60–63].

In Central America, drug trafficking has produced distinctive patterns of extensive
deforestation and other forms of environmental degradation [45,62,63]. Deforestation in
this region is considered a large-scale, late-stage effect of “narco-degradation” that reflects
changes in smuggling routes and greatly varies in the time, space, type, and intensity of
these activities, typically emerging at a local level as drug trade inserts in specific locations
across Central American countries [22,45].

3. Urbanization in Central America

Central America is currently experiencing a major demographic transition along with
an accelerated growth of urban populations. National statistics suggest that, between
2000 and 2014, rural population growth in this region has been declining while urban
populations have been steadily increasing [45]. After Africa, Central America is the second-
fastest urbanizing region in the world [30]. Over the past 20 years, Central American
urban populations grew at an average rate of 3.8% per year, which is twice as fast as other
Latin American populations and 1.7 times faster than the global average. Although the
proportion of urban populations increased to 59% today, compared to 48% in 1990, Central
America remains the least urbanized region in Latin America. As a consequence of rural-to-
urban migration and natural population growth, it is expected that the urban population
of Central America will double by the year 2050, growing to more than 25 million new
urban inhabitants [30]. Central American countries with the highest-to-lowest percentage
of urban populations are Costa Rica (81%), El Salvador (73%), Panama (68%), Nicaragua
(59%), Honduras (58%), Guatemala (52%), and Belize (46%) (Table 1) [30,64].

Across countries, official definition differences of what “urban” constitutes make
comparisons between countries difficult. In Central America, the definition of urban varies



Insects 2022, 13, 20 6 of 44

widely. For example, in Guatemala and Honduras, any human settlements with a popula-
tion larger than 2000 residents with access to basic infrastructure, such as electricity and
piped water, are classified as urban. In Panama and Nicaragua, urban is defined as settle-
ments of 1000 and 1500 inhabitants, respectively, Moreover, in Costa Rica and El Salvador,
urban areas are defined as people that are living within municipal boundaries (locally
known as “cantones” or “cabeceras municipales”), regardless of population size [30].

In Central America, large numbers of poor, rural populations are migrating to cities in
search of better educational and employment opportunities and improving their quality of
life. The most dominant factors driving this migration include declining agricultural prices,
environmental degradation, vulnerability to natural disasters, food insecurity, violence, and
economic instability [30]. This large influx of migrants poses significant challenges for cities,
including the provision of adequate urban infrastructure and reliable basic services, such
as sewer, water, and waste management, the worsening of existing housing deterioration,
and increased vulnerability to natural disasters [30,65].

Although cities contain most of a country’s population, recent population expansions
in Central America are driven by growing urban agglomerations in secondary cities in
areas surrounding capital cities. For instance, most urban population growth in Costa Rica
and Guatemala has taken place outside the capital cities [30]. Moreover, recent trends show
that land development in the region has been intensifying faster than population growth.
This expansion of built-up areas results in sprawling urbanization, which accelerates
deforestation [30,66]. In rural areas, population growth has also been a major driver of
environmental change, which further exacerbates deforestation, impacts land use, and
changes animal husbandry practices [30].

Unplanned and uncontrolled urbanization, characterized by the development of
informal settlements in high-risk areas with deficient building standards and infrastructure
and increase flood risks, has led to increased vulnerability to natural disasters in the
region [67,68]. Large-scale flooding is the most common disaster with close to 40 events
occurring across the Central America between 2006 and 2010 [30]. Furthermore, storms
have frequently impacted the region, including Hurricane Mitch in 1998 which directly
affected about 6.7 million people, causing 14,600 deaths and over USD 8.5 billion in damages
in Honduras, El Salvador, Nicaragua, and Guatemala [30]. Recently, back-to-back category
4 storms, Eta and Iota, devastated much of the region and impacted nearly 7 million
people in November of 2019 [30,69,70]. It is expected that climate change will further
modify current weather patterns in Central America potentially leading to an increase
in the number and severity of extreme meteorological events. Increase frequency and
intensity of floods, droughts, and hurricanes could affect access and quality of water and
alter ecosystem services in affected areas [68,71].

4. Mosquito-Borne and Tick-Borne Diseases in Central America

The global emergence and resurgence of VBDs in the last three decades are closely
linked to demographic, economic, and societal changes. The decay in public health in-
frastructure required to prevent and control these diseases and the unprecedented pop-
ulation growth, primarily in rapidly growing cities, are factors that have facilitated VBD
transmission and their geographic spread [2,5]. In Central America, the most important
VBDs affecting humans and animals include Chagas disease, leishmaniasis, dengue fever,
malaria, Zika fever, chikungunya fever, West Nile fever, rickettsial diseases, Eastern equine
encephalitis, Saint Louis encephalitis, and Venezuelan equine encephalitis (Table 2) [6,72].



Insects 2022, 13, 20 7 of 44

Table 2. Description of the most important mosquito-borne and tick-borne diseases in Central America.

Disease Causative Agents Distribution of Infections in
Humans

Confirmed or Suspected Mosquitoes
and/or Tick Vectors

Confirmed or Suspected Non-Human
Vertebrate Hosts

West Nile fever West Nile virus (Flavivirus) Clinical, serosurveys (CR, N) Culex quinquefasciatus,
Cx. mollis/Cx. inflictus (G)

Equines, non-human primates, wild
birds, sentinel chickens (CR, B, ES, G)

Saint Louis encephalitis Saint Louis encephalitis virus
(Flavivirus) Clinical, serosurveys (P, B, G, H)

Sabethes chloropterus, Trichoposopon
spp., Wyeomyia spp., Haemagogus

lucifer, Deinocerites pseudes, Mansonia
dyari, Culex nigripalpus

(P, CR, G)

Wild rodents, wild birds, sentinel
rodents, sentinel chickens, non-human
primates, sloths, equines, pigs (P, CR, B,

H, G)

Venezuelan equine
encephalitis

Venezuelan equine encephalitis
virus (Alphavirus) Clinical, serosurveys (all countries)

Psorophora confinnis, Culex nigripalpus,
* Cx. (Melanoconion) taeniopus, other Cx.
(Melanoconion) spp., Mansonia titillans,

Ps. cilipes, Aedes taeniorhynchus, and
Deinocerites pseudes (P, CR, B, G)

Equines, wild rodent, opossum, birds,
and bats (CR, N, H, ES, G)

Eastern equine encephalitis Madariaga virus (Alphavirus) Clinical, serosurveys (P) Culex (Mel.) taeniopus (P) Horses, bats, wild lizards, wild birds (P,
CR, B)

Yellow fever Yellow fever virus (Flavivirus) Clinical (all countries)

Aedes aegypti, Haemagogus janthinomys,
Hg. leucocelaenus, Hg. lucifer, Hg.
equinus, Hg. spegazzinii, and Sa.

chloropterus, Hg. mesodentatus (P, CR,
N, G)

Non-human primates, marsupials
(P, CR, N, B, H, G)

Zika fever Zika virus (Flavivirus) Clinical and serological (all
countries)

** Aedes aegypti, Ae. albopictus
(all countries) Unknown

Chikungunya fever Chikungunya virus (Alphavirus) Clinical and serological (all
countries)

** Aedes aegypti, Ae. albopictus
(all countries) Unknown

Dengue fever Dengue viruses 1–4 (Flavivirus) Clinical and serological (all
countries)

** Aedes aegypti, Ae. albopictus
(suspected) (all countries) Bats, non-human primates (CR)



Insects 2022, 13, 20 8 of 44

Table 2. Cont.

Disease Causative Agents Distribution of Infections in
Humans

Confirmed or Suspected Mosquitoes
and/or Tick Vectors

Confirmed or Suspected Non-Human
Vertebrate Hosts

Mayaro fever Mayaro virus (Alphavirus) Clinical and serological
(P, CR, G)

Haemagogus janthinomys,
Psorophora ferox, Culex (Mel.) vomerifer

(P)

Non-human primates
(P, CR, H, G)

Malaria Plasmodium vivax,
P. falciparum

Clinical and serological (all
countries)

* Anopheles albimanus, An. darlingi, An.
punctimacula, other Anopheles spp. (all

countries)
Unknown

P. malariae Clinical and serological
(P, CR, B, ES, G) Unknown Unknown

Rickettsiosis Rickettsia spp. (species causing
spotted fevers was not identified) Clinical, serosurveys (all countries) Amblyomma mixtum (G) Wild rabbits, dogs, coyote

(P, CR)

R. rickettsii (Rocky Mountain
spotted fever) Clinical (P, CR)

* A. mixtum, Rhipicephalus sanguineus
s.l., A. varium, Dermacentor nitens,

Haemaphysalis leporispalustris (P, CR)
Dog, horse (P, CR)

R. akari (rickettsialpox) Serosurvey (CR) Unknown Unknown
R. parkeri Unknown * A. maculatum (B) Unknown

R. parkeri strain Atlantic Forest Unknown * A. ovale (B) Unknown
R. africae Unknown A. ovale (N) Unknown

Ehrlichiosis Ehrlichia chaffeensis (monocytic
ehrlichiosis) Clinical (CR)

Amblyoma mixtum, Amblyomma sp.,
Dermacentor nitens, Rhipicephalus

microplus (P)
Unknown

E. ewingii (monocytic ehrlichiosis) Unknown R. microplus (P) Unknown
E. canis (granulocytic ehrlichiosis) Clinical (P, CR) ** R. sanguineus s.l. (all countries) Dogs (all countries)

Anaplasmosis Anaplasma phagocytophilum (human
granulocytic anaplasmosis) Unknown Rhipicephalus sanguineus s.l., R.

microplus (P, CR) Dogs, bovines, equines, deer (CR, N, G)

Borreliosis Borrelia burgdorferi s.l. (Lyme
disease) Clinical (CR) Ixodes c.f. boliviensis (P) Dogs (CR)

Borrelia sp. (tick-borne relapsing
fever) Clinical (P, G) Ornithodoros talaje, O. rudis (P) Armadillos, opossums (P)

* Main vector. ** Suspected vectors; pathogen has not been detected/isolated in all countries from these vectors, but they are present and considered the main vectors worldwide.
Country abbreviation (Belize = B; Costa Rica = CR; El Salvador = ES; Guatemala = G; Honduras = H; Nicaragua = N; Panama =P).



Insects 2022, 13, 20 9 of 44

4.1. Mosquito-Borne Arboviral Diseases in Central America
4.1.1. West Nile Virus Disease

West Nile virus (WNV) (genus Flavivirus, family Flaviviridae), initially discovered in
Uganda in 1937, has become endemic in parts of the Americas since its introduction in
1999 [73]. In Central America, the first evidence of its active circulation in horses was
detected between November 2001 and April 2003 in El Salvador, where 203 horses died
from undetermined encephalitis [74], and Belize, where a single encephalitic horse was
diagnosed with WNV infection [75]. In addition, ten serum samples from horses in the
same area were positive for WNV. Although no human cases were reported during this
outbreak [74], another study carried out in 2003 in Guatemala, Belize, and El Salvador
detected antibodies against the virus in humans [76,77]. Moreover, in 2006, a single case of
WNV infection was detected in a Spanish missionary who was living in Nicaragua [78].

Several studies have been conducted in Central America to understand WNV transmis-
sion involving equines and mosquitoes. For instance, studies were conducted in Guatemala
to determine WNV transmission dynamics and seroepidemiology [79–81]. Of the seven
departments selected for monitoring by sentinel chickens in the country, one transmission
focus was identified in the eastern city of Puerto Barrios. Annual transmission at that
site was detected between the months of May and October of 2005–2008. Additionally,
great-tailed grackles (Quiscalus mexicanus [Gmelin, 1788]) have been identified as primary
amplifying hosts in the region [79,81]. Environmental factors, such as high temperatures
and low rainfall, are strongly associated with chicken seroconversions since these factors
influence mosquito population density and viral infection kinetics in the vector in both rural
and urban environments [79]. Another study, conducted between 2003–2004, evaluated
352 Guatemalan horses for WNV antibodies finding nine horses positive for WNV, 33 for
SLEV, and 21 positive for undifferentiated flaviviruses [80]. Recently in Panama, a study by
Carrera et al. (2018) found a WNV seroprevalence in equids with neurological disease of
2.6% [82].

In Costa Rica, WNV circulation was first detected in 2004 in seropositive horses from
the Guanacaste Province with a prevalence of 18–28%, followed by the first equine with
neurologic disease in 2009 in the same region [83]. Since then, new equine cases are reported
annually in the country, especially in the lowlands during the rainy season [84]. Serological
evidence of WNV infection in wildlife has also been reported, including non-human primates,
Hoffman’s two-toed (Choloepus hoffmanni Peters, 1858) and brown-throated sloths (Bradypus
variegatus Schinz, 1825), and birds [79,80,85–87]. The ecological distribution of these species
coincides with the geographic distribution of neurological cases in equines [84,86,87]. The
importance of wildlife in the enzootic transmission of WNV in Central America is largely
unknown. A more recent study by Piche-Ovares et al. (2021) during the rainy and dry seasons
reported additional serological evidence of neutralizing antibodies against WNV in equines,
humans, sentinel chickens, a local wild bird, and one seroconversion event in a horse [88].
However, the study did not find molecular evidence of active virus circulation in wild birds
and mosquitoes. Although human cases of neurological WNV infection have not yet been
recorded in Costa Rica, evidence suggests that the virus is widely distributed throughout the
region [88].

A limited number of studies have examined the incrimination of mosquito species
in WNV transmission in Central America. For instance, a study by Morales-Betoulle
et al. (2013) at a WNV transmission foci in Guatemala isolated the virus from Culex
quinquefasciatus Say, 1823 and Culex mollis Dyar and Knab, 1906/Cx. inflictus Theobald,
1901 mosquitoes; however, no isolates were obtained from the most abundant species, Cx.
nigripalpus Theobald, 1901 [79]. Laboratory vector competence using Central American
WNV strains and mosquitoes have found evidence of moderate-to-strong competency for
Cx. quinquefasciatus and Cx. nigripalpus in Honduras and Guatemala [89,90].

One of the most critical issues in Central America is the lack of data on the extent of
WNV human disease burden in the region since it can be very challenging to diagnose
these infections. Human WNV cases are often misdiagnosed or underdiagnosed due to



Insects 2022, 13, 20 10 of 44

several factors, including serological cross-reactivity with other flaviviruses circulating in
the region, virus genome diversity, low transient viremia, and the lack of laboratory testing
capacity [91]. Presently, the diagnosis of WNV infection is mostly based on serological
methods since molecular identification of virus RNA is often unreliable due to the short-
term transient viremia and low viral load at the time of the onset of symptoms [92]. As
seen in other areas where the WNV circulates, the serological diagnosis of WNV disease
in Central America is also problematic due to the active co-circulation of dengue virus,
Zika virus, and other flaviviruses that will likely produce cross-reactivity in serological as-
says [92]. Lastly, laboratory capacity to test for WNV and other VBDs varies widely among
Central American countries, ranging from some laboratory capacity and epidemiological
surveillance systems to a lack of critical resources, such as appropriate testing technologies,
reagents, facilities, epidemiological surveillance systems, and technical expertise [93].

One of the possible reasons for the lack of evidence for human and equine WNV
neurological disease and high avian mortality in Central America is that birds infected with
more virulent strains could not start their migration process to South America while only
birds infected with less virulent viral strains are able to migrate [94]. Another hypothesis
that seeks to explain the relatively low human, equines, and avian mortality in Central
America focuses on pre-existing neutralizing antibodies against other flaviviruses, such
as dengue virus (DENV) and Saint Louis encephalitis virus (SLEV), which might offer
partial protection from WNV disease [77,95,96]. Finally, environmental and host factors,
such as temperature, vector competence, and host susceptibility, may influence the genetic
selection of less virulent variants [95]. Although WNV disease has not been reported
in most of Central America, all the components that could support virus circulation are
present throughout the region maintaining the risk for future outbreaks in the region.

4.1.2. Saint Louis Encephalitis

Saint Louis encephalitis virus (SLEV) (genus Flavivirus, family Flaviviridae) was first
identified in St. Louis, Missouri, USA in 1933. SLEV can be classified into eight genotypes,
with genotypes I and II circulating mainly in the US and associated with outbreaks of
human encephalitis [97], while genotypes III–VIII have been found only in Central and
South America. In Central America, SLEV circulation was first documented in 1957 in
Buena Vista, Panama, when the virus was isolated from a pool of Sabethes chloropterus (von
Humbolt, 1819) [98]. Subsequently, serological evidence and isolates of SLEV were also
detected in Maje Island, Darien Province, and Panama Province in humans, wild birds (local
and migratory), several wild mammals (rodents, non-human primates, sloths), various
sentinel animals, and several mosquito species, including Haemagogus lucifer (Howard,
Dyar and Knab, 1913), Deinocerites pseudes Dyar and Knab, 1909, Mansonia dyari Belkin,
Heinemann and Page, 1970, Cx. nigripalpus, Trichoprosopon spp., and Wyeomyia spp. [98–101].
In the Pacific coast of Guatemala, the virus was also isolated from Cx. nigripalpus [102].

More recently, two studies in Costa Rica reported serological evidence of SLEV infec-
tion in free range and captive non-human primates and sloths. For example, a study by
Chavez et al. (2021) found homotypic and heterotypic neutralization reactivity to SLEV of
47.6% and 5.9%, respectively, in mantled howler monkeys (Alouatta palliata [Gray, 1849])
and spider monkeys (Ateles geoffroyi Kuhl, 1820), while another study found a seroposi-
tivity of 42% in Hoffman’s two-toed sloths (C. hoffmanni) and brown-throated sloths (B.
variegatus) [85,86]. Moreover, a study conducted between 2017 and 2018 in northwestern
and southeastern Costa Rica found neutralizing SLEV antibodies in humans, local and
migratory birds, and equines [88]. One seroconversion event to SLE, between the rainy and
the dry seasons, was also detected in a horse from the province of Guanacaste [88].

4.1.3. Venezuelan Equine Encephalitis

Venezuelan equine encephalitis virus (VEEV) (genus Alphavirus, family Togaviridae) is a
serocomplex of six antigenic subtypes (I–VI). Within subtype I, there are five antigenic vari-
ants (AB–F), with antigenic variants IAB and IC associated with epizootic/epidemic activity



Insects 2022, 13, 20 11 of 44

in equines and humans, while variants ID, IE, and IF and subtypes II–IV are associated
with enzootic forest cycles. These enzootic viruses circulate in sylvatic rodent populations
within tropical and subtropical forests or swamp habitats continuously transmitted by Cx.
(Melanoconion) mosquitoes [103,104]. Most enzootic variants are avirulent in equines, but
some strains could cause clinical disease in both humans and equines with a similar clinical
presentation to epizootic strains infections [105].

The first VEEV epizootic involving humans and equines in Central America took place
between 1969–1970 in Honduras, Guatemala, Nicaragua, El Salvador, and Costa Rica, as part
of a major intercontinental epizootic that spread from Central America to northern South
America and south Texas, U.S. [106,107]. The virus strain responsible for this epizootic was
identified VEEV subtype IB (later reclassified as IAB) and, at the time, was commonly isolated
from equines, humans, and several mosquito species, including Psorophora confinnis (Lynch
Arribalzaga, 1891), Cx. nigripalpus, Cx. (Melanoconion) sp., Ma. titillans (Walker, 1848), Aedes
taeniorhynchus (Wiedemann, 1821), and De. pseudes [106,108–110]. Following this epizootic,
numerous studies linked the origin of the VEEV epizootic subtype IAB to incompletely
innactivated vaccines [111]. Moreover, evidence also suggest that key mutations in enzootic
VEEV strains could potentially mediate the emergence of novel epizootic strains [112].

In recent decades, only two enzootic VEEV subtypes have been reported in Central
America. The VEEV subtype IE appears to be widely distributed throughout the region
while the VEEV subtype ID has been detected only in Panama [82,84,112–114]. Recent
data suggest that humans infected with these two enzootic subtypes could develop suf-
ficient viremia to potentially infect both Central American enzootic and epizootic VEEV
vectors [112,113]. Moreover, experimental competency studies suggest the potential trans-
mission of VEEV subtype ID in Central America by Ae. aegypti (Linnaeus, 1762) and Ae.
albopictus (Skuse, 1894), increasing the possibility that urban transmission cycles could
take place in the region [115,116]. It is possible that febrile illness linked to endemic VEEV
infections in Latin America are more widespread than previously observed but are vastly
unrecognized or misdiagnosed due to the presence of similar dengue-like febrile illnesses
and lack of effective public health surveillance systems [112,113].

The first reported human case of enzootic VEEV ID virus infection in Panama took
place in the early 1960s near Panama City [117]. Recently, an increase in human VEEV
cases in Panama have intensified ecological and epidemiological studies in the region,
with several studies reporting the active and stable circulation in humans of both VEEV
subtypes ID and IE throughout the country. For example, recent serosurveys have found
neutralizing VEEV antibodies levels between 8.5% and 78% across several communities
in the Darien region [82,113,114]. Moreover, subtype ID outbreaks in Panama also show
higher case fatality rates than those reported during previous subtype IAB epidemics [113].

Field studies in Panama, Costa Rica, Guatemala, Belize, and Honduras have also
detected the circulation of enzootic VEEV ID and IE viruses in a variety of vertebrates,
including wild rodents (Zygodontomys brevicauda [Allen and Chapman, 1893]), Transadino-
mys (=Oryzomys) bolivaris Allen, 1901, Proechimys semispinosus [Tomes, 1860], Melanomys
caliginosus [Tomes, 1860]; Oryzomys spp.), opossums (Didelphis marsupialis Linnaeus, 1758,
and Marmosa robinsoni Bangs, 1898; Philander spp.), sloths (Bradypus spp. and Choloe-
pus spp.), sentinel guinea pigs and hamsters, equines, wild and domestic birds, and
bats [82,84,114,118–124]. Moreover, entomological surveys in Central America have iden-
tified Cx. (Melanoconion) spp., particularly Cx. (Mel.) taeniopus Dyar and Knab, 1907, Cx.
(Mel.) ocossa Dyar and Knab, 1919, and Cx. (Mel.) panocossa Dyar, 1923; formerly known
as Cx. (Mel.) aikenii (Aiken and Rowland, 1906), Cx. (Mel.) vomerifer Komp, 1932, and Cx.
(Mel.) erraticus (Dyar and Knab, 1906) as the most important enzootic VEEV vectors in the
region [112,125].

4.1.4. Madariaga Virus

Maradiaga virus (MADV) is an emergent Alphavirus (family Togaviridae) in the Eastern
equine encephalitis (EEE) antigenic linage III strain complex. This virus was previously



Insects 2022, 13, 20 12 of 44

known as the EEE South American variant and it is maintained in stable, enzootic cycles
throughout Central and South America [126]. Although broad spectrum human and equine
infections linked to MADV have been reported [124,127,128], a recent outbreak in Panama
underscores concerns as an emergent virus in Latin America [114,129]. Although the enzootic
cycle of MADV remains unclear, its circulation has been detected in birds, rodents, marsupials,
reptiles, and bats [130]. In Panama, several rodent and bat reservoirs have been proposed,
including the black rat (Rattus rattus [Linnaeus, 1758]), short-tailed cane mouse (Z. brevicauda),
long-whiskered rice rat (T. bolivaris), Tome’s spiny rat (P. semispinosus), Seba’s short-tailed bat
(Carollia perspicillata [Linnaeus, 1758]), and pale spear-nose bat (Phyllostomus discolor Wagner,
1843) [114]. The primary mosquito vectors for this virus are Cx. (Melanoconion) mosquitoes,
especially Cx. (Mel.) taeniopus, which could serve as an enzootic and epizootic vector [130]. In
Central America, MADV isolates have been obtained also from Cx. (Mel.) taeniopus in Panama
during field surveys and equine outbreaks [127,131].

The first report of an outbreak of MADV-related neurologic disease took place in 2010
in Darien, Panama, where seven humans and 210 horses developed encephalitis and were
confirmed positive for the virus [129]. In 2017, another outbreak was reported in the same
region of Darien. A serosurvey conducted in the area showed a higher seroprevalence
than during previous investigations [132,133]. Human activities, such as horse and cattle
ranching, fishing, farming, pasture, and poor housing conditions, have been identified as
risk factors associated with MADV infections [114]. Moreover, people living near or having
vegetation around the house had higher seroprevalence to MADV [133]. Therefore, the
increased exposure of people to MADV in this region could have resulted from ecological
changes, primarily deforestation, which increased human contact with enzootic transmis-
sion cycles. The co-circulation of MADV and VEEV makes diagnosis difficult in regions
where these viruses are endemic [129].

4.1.5. Yellow Fever

Yellow fever virus (YFV) is a member of the genus Flavivirus (family Flaviviridae)
primarily transmitted by the bite of Aedes (Stegomyia) spp., Haemagogus spp., and Sabethes
spp. mosquitoes in tropical and subtropical regions of Africa and South America [134].
In the Americas, YFV is currently distributed between southern Panama and northern
Argentina [134,135]. In Central America, outbreaks have been recognized as far back as
the mid-1600s in the Yucatan Peninsula and through the construction of the Panama Canal
in the late 19th century [136–138]. From 1905 to 1948, there was a period of relative quiet
with no autochthonous cases of yellow fever reported; however, between 1949 and 1954, a
large human outbreak of sylvatic YF, which originated in Panama, spread northward to
Costa Rica, Nicaragua, Honduras, Guatemala, and the Guatemala–Mexico border [139].
During this period, a high mortality in non-human primates was also reported in the
region [138,140]. At the time, several entomological surveys conducted in Costa Rica and
Nicaragua identified a high abundance of Haemagogus spp. and Sabethes spp. mosquitoes
in the region affected by the epizootic [141–143]. In addition, the virus was isolated in
Panama from Hg. lucifer, Hg. equinus Theobald, 1903, Hg. spegazzinii Brethes, 1912, and
Sa. chloropterus, and in Guatemala from Hg. mesodentatus Komp and Kumm, 1938, Hg.
equinus, and Sa. chloropterus during the same time period [144,145]. The vector competence
of Guatemalan and Panamanian Haemagogus and Sabethes species for YFV was established
via a mouse inoculation experiment [146]. The presumptive reservoir of YFV in Central
America is the howler monkey (Alouatta spp.), which is highly susceptible to the infection
and has shown high mortality levels during epizootics [138]. After the 1950s, no other
urban outbreaks of YF were documented in Central America; however, sporadic outbreaks
of sylvatic yellow fever took place during the 1960s in the region [147].

4.1.6. Zika Fever

Zika disease (or Zika fever), caused by the Zika virus (ZIKV, family Flaviviridae,
genus Flavivirus), first detected in Uganda in 1947 [148], has received intense scrutiny



Insects 2022, 13, 20 13 of 44

since its emergence as a significant human pathogen in recent years [149–151]. ZIKV is
now considered endemic throughout Latin America [152]. In Central America, the first
autochthonous cases of Zika fever were documented in November 2015 in El Salvador
and Guatemala [153]. The virus rapidly spread through the rest of Central America, with
cases reported in Honduras and Panama and later in Costa Rica and Nicaragua [154,155].
By April 2016, ZIKV was present in all Central American countries, with Belize being
the last in which introduction was documented [156]. Early into the epidemic, two ZIKV
strains obtained in Guatemala in 2015 were sequenced and identified as belonging to the
Asian lineage, the same lineage detected in Brazil that was spreading to other countries
that year [157]. However, phylogenetic analyses have suggested that ZIKV was likely
introduced from Brazil to Honduras as early as late 2014 and spread undetected to other
Central American countries [158]. Overall, the rapid introduction and spread of ZIKV
through Central America probably resulted from the constant movement of people between
these countries as well as environmental and climatic conditions that promote high densities
of Ae. aegypti and can drive transmission dynamics [159,160]. Most cases of Zika fever in
the region have been reported in El Salvador, Belize, Nicaragua, and Honduras [161].

Although Zika fever incidence has decreased in the past few years, it is still one
of the most frequent arboviral diseases in Central America. The overlapping signs and
symptoms of dengue fever, Zika fever, and chikungunya fever, as well as the unavailability
of widespread laboratory confirmation in many areas of Central America, make diagnosis
and epidemiological surveillance of these diseases challenging [162,163]. Moreover, several
studies suggest that ZIKV infection rates during the recent American epidemic provided
adequate herd immunity to lessen the risk of another large epidemic for at least another
10 years [164]. Nonetheless, the scale of ZIKV transmission has remained patchy and
widely variable in the Americas [165].

The establishment of sylvatic cycles involving non-human primates has not been reported
yet in the Americas; however, its possibility cannot be ruled out [166]. There is consensus
that ZIKV can potentially become established in sylvatic cycles between non-human primates
and mosquitoes; however, field evidence is still inconclusive [167]. At least three primate
species present throughout Central and South America [168], Ma’s night monkey (Aotus
nancymaae Hershkovitz, 1983), Guianan squirrel monkey (Saimiri sciureus [Linnaeus, 1758]),
black-tufted marmoset (Callithrix penicillata [Geoffroy, 1812]) are susceptible to ZIKV infection.
Although they usually do not develop clinical symptoms, their viremia can potentially support
transmission based on experimental infections [169,170]. Regarding potential sylvatic vectors
for ZIKV in Central America, experimental studies have shown that Sa. cyaneus (Fabricius,
1905) is a competent vector, but less competent than Ae. aegypti, while Ha. leucocelaenus (Dyar
and Shannon, 1924) have shown low rates of dissemination. [171,172].

4.1.7. Chikungunya

Chikungunya virus (CHIKV) (family Togaviridae, genus Alphavirus), the etiological
agent of chikungunya fever, was first identified in Tanzania in 1952, and has recently spread
throughout tropical and subtropical regions of the world, including the Americas [173,174].
In 2014, the first cases of CHIKV in Central America were reported in El Salvador followed
by Guatemala, Costa Rica, Honduras, and Nicaragua [175–177]. In Panama, recent studies
reported CHIKV seroprevalence strongly associated with densely populated urban and
periurban areas, poor socioeconomic conditions, and high infestation indices of the vectors
Ae. aegypti and Ae. albopictus [178,179]. Similar results have been found in Nicaragua,
where a serosurvey involving over 11,000 participants, found 39% CHIKV seroprevalence
associated with sites containing high vector infestation indices [180].

The potential introduction of sylvatic cycles of CHIKV in the Americas is still under
investigation. Although serological evidence is weak, there is potential for the introduction
of CHIKV sylvatic cycles through spillback events [181]. For instance, experimental studies
have shown that some reptiles and amphibians maybe susceptible to infection [182], while
there is limited evidence on the role of neotropical non-human primates in the establishment



Insects 2022, 13, 20 14 of 44

of CHIKV sylvatic cycles [183]. Regarding potential vectors, Ha. leucocelaenus and Ae. terrens
(Walker, 1856), two sylvatic species of mosquitoes in the neotropics, appear to be competent
experimental vectors of CHIKV [184]. It has been suggested that the level of herd immunity
recently observed throughout much of the Americas could limit the occurrence of major new
epidemics until the next population generation provide additional amplifying hosts [165].

4.1.8. Dengue Fever

Dengue fever, a disease caused by four dengue viruses (DENV, genus Flavivirus,
family Flaviviridae) serotypes, emerged and evolved from sylvatic cycles in Asia and are
primarily transmitted to humans by Ae. aegypti mosquitoes. Aedes albopictus could also
act as a DENV vector, particularly in areas where Ae. aegypti is absent; however, it has
been difficult to directly incriminate Ae. albopictus as a DENV vector during autochthonous
arbovirus outbreaks [179,185–187]. Moreover, recent evidence suggests that this species
maybe effective as a natural reservoir of DENV via transovarial transmission [188].

The first reports of dengue fever in Central America were made in the early 20th century
in Panama [189], although the occurrence of human cases could go as far as the 1600s [190].
The failure to eradicate Ae. aegypti populations in the 1960s and 1970s triggered a resurgence of
dengue fever in the Americas [190]. Between 1978 and 1980, an increase in dengue fever cases
was observed in Guatemala, Belize, and El Salvador [191,192]; then, in 1985, a major epidemic
in Nicaragua caused by DENV 1 and DENV 2 affected over 17,000 people [193]. In 1993, Costa
Rica and Panama confirmed local transmission and autochthonous cases of dengue fever
for the first time in 40 to 50 years [194]. Since then, the virus has become endemic in both
countries [195,196]. Further details on the first reports of different DENV subtypes in Central
America can be found elsewhere [190,191,197]. The re-emergence of dengue fever in Central
America, just as in other parts of the Americas, has been attributed to several factors, including
diminished political importance in countries where eradication was achieved, reduction in
surveillance and other public health resources, development of insecticide resistance and
resurgence of Ae. aegypti populations, establishment of Ae. albopictus, social disparities, and
increased urbanization in the region [179,190,198–200].

Today, there is little evidence that establishes the existence of sylvatic DENV trans-
mission cycles in the Americas. Although several neotropical non-human primate species,
including white-faced capuchin monkeys (Cebus capucinus [Linnaeus, 1758]), spider mon-
keys (Ateles spp.), and mantled howler monkeys (A. palliata), are susceptible to DENV
infection, serological surveys of Panamanian monkeys failed to show evidence of enzootic
circulation [201,202]. Recently, serological and molecular evidence of DENV infection
was reported in several species of non-human primates in Costa Rica, which suggest po-
tential bidirectional exposures due to the presence of bridging vectors or an increase in
human–wildlife contacts [85,87].

4.1.9. Mayaro Fever

Mayaro virus (MAYV) is a neotropical Alphavirus (family Togaviridae) member of the
Semliki Forest antigenic complex initially isolated in Trinidad in 1954 [203]. It has been de-
tected throughout the Americas, from Mexico to Brazil and the Caribbean [204], and it is con-
sidered a neglected viral disease in humans [203,204]. Infection with MAYV is characterized
by a self-limiting febrile illness accompanied by long term incapacitating arthralgia [203];
however, severe cases and even deaths have also been reported [204]. The virus circulates
in continuous sylvatic cycles between canopy dwelling Hg. janthynomis Dyar, 1921 and
non-human primates, including howler monkeys (A. seniculus [Linnaeus, 1766], A. caraya
(Humbolt, 1812), and A. villosa [Gray, 1845]), silvery marmosets (C. argentata [Linnaeus,
1771]), and capuchin monkeys (Sapajus spp.). Human cases are typically associated and
restricted to the edges of neotropical rain forests causing limited outbreaks [204,205]. Syl-
vatic cycles may also include other animals, such as sloths, sheep, rodents, horses, reptiles,
agoutis, and birds, but their role in transmission is still unclear [206,207]. Vector compe-
tence for MAYV and numerous field isolates have also been reported with other mosquito



Insects 2022, 13, 20 15 of 44

species, including Anopheles spp., Culex spp., Sabethes spp., Mansonia spp., and Psorophora
spp. [206,208]. In Central America, very little is known about the epidemiology and ecology
of MAYV and its potential for emergence as an important pathogen. Only a small number
of field studies have been conducted in Central America. In Panama, Guatemala, and
Costa Rica, MAYV has been detected in humans, sentinel monkeys, wild howler monkeys
(A. villosa), and agoutis (Dasyprocta punctata Gray, 1842) [82,132,205,206,209–211], while in
mosquitoes, the virus has been isolated in Panama from Ps. ferox (Humbolt, 1819) and Cx.
(Mel.) vomerifer [118,212].

Disease spillover into rural and peri-urban areas has been reported and there is
increasing concern that MAYV could become urbanized since it could adapt to replicate
in the urban vectors Ae. aegypti and Ae. albopictus, whose competence has been reported
in both the laboratory and the field [204,213,214]. The number of true MAYV cases in
the Americas is potentially higher than what has been reported due in part to potential
misdiagnosis and underdiagnosis, clinical similarities to other arboviral infections, and
coinfections with other endemic arboviruses, such as DENV and CHIKV [204].

4.2. Malaria in Central America

Malaria, a febrile disease caused by Plasmodium spp. parasites transmitted by Anopheles
spp. mosquitoes, causes the highest morbidity and mortality compared to any other VBD
worldwide [215]. While less than 1% of the 2019 global malaria burden was recorded in
the Americas, the region remains endemic for the disease. In Central America, there were
nearly 20,000 autochthonous malaria cases recorded in 2019, with cases reported in five
out of seven countries within the region: Nicaragua (13,200)), Guatemala (2100), Panama
(1600), Honduras (330), and Costa Rica (91) [215]. Almost all these cases were caused by
P. vivax (Grassi and Feletti, 1890) (74%) and P. falciparum (Welch, 1897) (23%), which are
transmitted mainly by An. albimanus Wiedemann, 1820, An. pseudopuntipennis Theobald,
1901, An. darlingi Root, 1896, An. marajoara-Galvao and Damasceno, 1942, An. aquasalis
Curry, 1932, An. albitarsis Lynch Arribalzaga, 1878, and An. vestitipennis Dyar and Knab,
1906, reflecting a remarkable diversity of competent vectors with diverse ecologies and
bionomics [215–221].

For broader context, these 2019 regional case estimates represent more than a 50%
reduction from 2010 estimates, when nearly 40,000 cases were recorded, and over 80%
reduction from 2000 estimates when more than 340,000 cases were reported [222]. Despite
some challenges that persist, current case numbers highlight the remarkable progress
and numerous successes achieved by malaria control and elimination programs in the
region [223]. For example, El Salvador was recently certified as malaria free while Belize,
Costa Rica, Guatemala, Honduras, and Panama have the potential to eliminate malaria
within the next five years [224].

While it is important to reflect on this encouraging progress and celebrate recent
malaria control successes across Central America, it is equally important for the region not to
let its collective guard down. With multiple competent malaria vectors naturally occurring
throughout its geography [216], Central America remains vulnerable to the re-introduction
of malaria across its entire range. Furthermore, the influx of people from malaria endemic
regions of the Americas, both travelers and migrants, as well as international travelers
from other malaria endemic regions, creates a low but constant risk of Plasmodium parasite
re-introduction into areas with greatly reduced local transmission [225–228]. This risk
is exacerbated by the prospects of a changing climate, deforestation, changing land-use
patterns, and increasing drug and insecticide resistance [229,230].

4.3. Tick-Borne Diseases in Central America

In Central America, recent studies on the ecology of tick-borne diseases are limited
for most countries, and outbreaks and case reports in humans are mostly sporadic and
infrequent. However, there is evidence of possible zoonotic human pathogens in ticks and/or
vertebrate animals, such as Anaplasma phagocytophilum (Foggie, 1949), Ehrlichia canis (Donatien



Insects 2022, 13, 20 16 of 44

and Lestoquard, 1935), E. chaffeensis Anderson et al., 1992, E. ewingii Anderson et al., 1992,
Rickettsia rickettsii (Wolbach, 1919), R. parkeri Lackman et al., 1965; including the strain Atlantic
Rainforest), R. akari Huebner et al., 1946, R. africae Kelly et al., 1996, Borrelia burgdorferi s.l.
Johnson et al., 1984, and Borrelia sp. (causing tick-borne relapsing fever), as well as several
other microorganisms that may infect ticks and/or vertebrate animals for which pathogenicity
is unknown, or that are not known to cause human disease [231–264]. In humans, the most
common tick-borne diseases reported in Central America in the past few decades are spotted
fever group rickettsioses and ehrlichioses, followed by isolated reports of probable Lyme
disease in which the bacteria were not directly detected or identified [250,252,265–271]. In
addition, there are records of tick-borne relapsing fever and serological evidence of exposure
to R. akari [235,248,272], while human anaplasmosis and R. parkeri spotted fevers have yet
to be confirmed in the region. In most cases of outbreak descriptions of human tick-borne
diseases, the main vectors and vertebrate reservoirs implicated in local transmission cycles
have not been clearly identified.

4.3.1. Rickettsioses

Bacteria of the genus Rickettsia (Rickettsiales: family Rickettsiaceae) are responsible
for various clinical rickettsioses in humans; in the Americas, R. rickettsii spotted fever is
the most relevant in terms of morbidity and mortality [273]. Tick-borne rickettsioses have
been known to occur in Central America since the 1950s [252,274]. The first cases were
identified and confirmed by isolation of R. rickettsii from humans and ticks in Panama
and later in Costa Rica [252,271]. Ecological investigations to identify possible vertebrate
hosts and tick vectors have been carried out in these countries throughout the decades. For
instance, Amblyomma mixtum Koch, 1844 (previously referred to as A. cajennense) has been
implicated as a probable vector of R. rickettsii to humans in Panama and probably Costa
Rica [252,271]. This agrees with the information available about transmission of rickettsiae
in South America, where several tick species belonging to the “A. cajennense species group”
are considered vectors of R. rickettsii [275,276]. Moreover, the brown dog tick, Rhipicephalus
sanguineus s.l. (Latreille, 1806), has also been investigated as a possible vector of urban
human cases of R. rickettsii infection in Panama [252,277]. As for other spotted fever group
rickettsiae, there are reports of human disease and outbreaks in Guatemala, Honduras,
Nicaragua, although the species was not identified in these cases [278–280]. In addition,
there are historical records in which antibodies against R. akari (or an antigenically similar
species) have been detected in humans in Central America, but there are no records of
direct detection of the bacterium in ticks or human cases of rickettsial pox [272]. Rickettsia
akari transmission is usually associated with hematophagous mites, but it has been detected
in humans, dogs, and ticks in neighboring Yucatan, Mexico [281,282]. Recently, DNA of R.
africae was detected in A. ovale Koch, 1844 ticks from Nicaragua, but there are no human
cases of African tick-bite fever confirmed to this date [254].

4.3.2. Ehrlichiosis and Anaplasmosis

The most relevant species of Ehrlichia and Anaplasma (Rickettsiales: family Anaplasmat-
aceae) in terms of morbidity in humans are E. chaffeensis and A. phagocytophilum, which cause
human monocytic ehrlichiosis and human granulocytic anaplasmosis, respectively [283,284].
Different species of Ehrlichia and Anaplasma are known to occur in Central America, espe-
cially in domestic animals (e.g., E. canis, E. chaffeensis, Ehrlichia sp. H7, A. platys (Dumler
et al., 2001), A. phagocytophilum), while E. ewingii have been detected only through DNA in
ticks [231,234,236–239,241–244,246,253,255,257,285]. In humans, there are several reports of
possible ehrlichiosis/anaplasmosis diagnosed by observation of morulae in stained blood
smears, by indirect antigen or antibody detection, without isolation, or by molecular identi-
fication of the species [250,266,267,269]. In 2015, E. chaffeensis was confirmed in humans by
polymerase chain reaction (PCR) and DNA sequencing in Costa Rica [245]. In addition, E.
canis DNA has also been reported in humans in Costa Rica and in one case in Panama, which
was the only one associated with severe disease [251,258]. Anaplasma phagocytophilum DNA



Insects 2022, 13, 20 17 of 44

has been detected repeatedly in ticks and domestic animals, but this species has not been
confirmed in human infections [234,236,239,249,255,257]. In areas of North America where
human ehrlichiosis and anaplasmosis are common, the principal vectors are Amblyomma
americanum (Linnaeus, 1758) and Ixodes spp. (I. scapularis Say, 1821; I. pacificus Cooley and
Kohls, 1943; and others), respectively, but these ticks are not found in Central America or
South America, where the possible vectors are still being determined [286–288]. As for E.
canis, studies have confirmed that R. sanguineus s.l. is the vector within the dog population in
Central America, in correspondence to what is known for other regions [249,286,289].

4.3.3. Borrelioses

The genus Borrelia (Spirochaetales: family Borreliaceae) includes spirochaetal bacteria
that are mostly associated with ticks and reptiles, but also lice [290,291]. Currently, half
of the named species (21 of 42) belong to the “relapsing-fever associated” group, and
almost all of the other half (20 species) to the “Lyme borreliosis associated” group (Borrelia
burgdorferi s.l.) [291].

In Central America, the first descriptions of Borrelia spp. spirochetes in human blood
samples were reported in Panama in 1909 [292]. Although a species of Borrelia was not assigned,
subsequent studies confirmed that tick-borne relapsing fever was common in humans, and
transmission cycles in Panama were associated with argasid ticks that can sometimes feed
on humans, such as Ornithodoros talaje (Guérin-Méneville, 1849) and O. rudis Karsch, 1880
(=O. venezuelensis), and on animals such as armadillos and opossums [248,292–294]. Later, the
bacterium implicated in Panama was presumed to be the same species transmitted by O. rudis
infecting humans in South America, referred to as B. venezuelensis (Brumpt, 1921), although
this has not been confirmed with certainty [295]. Unfortunately, no other cases have been
reported in the last decades in Panama and the precise identification of the pathogen has not
been determined. However, a case of relapsing fever was diagnosed more recently in a traveler
who visited areas of Guatemala and the border with Belize, confirming that infections with
Borrelia spp., and causing relapsing fever, are probably present throughout the region and may
be unreported [235,248].

In contrast, B. burgdorferi s.l. has been suspected in a few imported and autochthonous
human cases in Honduras and Costa Rica, although there is only indirect serological
evidence of infection [265,268,270]. The principal vectors of Lyme disease in North America,
Europe, and Asia are ticks of the I. ricinus species complex, including I. ricinus (Linnaeus,
1758), I. scapularis, I. pacificus, and I. persulcatus Schulze, 1930 [296]. These species are not
present in neotropical areas and most Ixodes spp. in Central America do not frequently
bite humans, which supports the idea that Lyme disease does not occur or it is rare in the
region [287,297]. Recently, bacterial DNA identified as B. burgdorferi s.l. was detected in
Ixodes c.f. boliviensis Neumann, 1904 from Panama, although the ability of this bacterium to
infect humans and cause Lyme disease is currently unknown [263].

5. Impact of Deforestation, Urbanization, and Changing Land Use Patterns on the
Ecology of Mosquito and Tick-Borne Diseases in Central America
5.1. Impact on Mosquito-Borne Arboviral Diseases

Major knowledge gaps persist on the enzootic transmission cycles of most arboviral
diseases endemic to Central America, including information on geographic distribution,
vertebrate reservoir species, mosquito vectors, and human disease risk. Furthermore, little
is known about the potential effects of increasing deforestation, forest fragmentation, and
urbanization on the ecology and epidemiology of these diseases in the region. While dengue
fever has received the most attention throughout Central America [125,298], research on
other arboviral diseases in the region have been conducted mostly in Panama and Costa
Rica, the countries with highest economic development and significant public health
investments in the region [30]. Arboviral diseases present in Central America are of
increasing public health concern due to their recent emergence or resurgence and most are
considered neglected [6,112,129,203].



Insects 2022, 13, 20 18 of 44

Changes in land use can significantly affect mosquito population dynamics, oviposi-
tion, abundance, and host-seeking behaviors. Numerous studies have shown that mod-
ifications in land use could result in the loss of hosts, predators, and mosquito habitats,
which may affect vector population dynamics, abundance, oviposition, and host-seeking
behaviors. These environmental modifications could drive mosquito vectors to search for
new blood-feeding sources and alternative breeding habitats [3,23,299], promote higher
host contact rates, and initiate disease spillover events, introducing new infections into
susceptible human populations [300–302]. Other factors, such as human migration and
urbanization, can significantly impact the distribution and occurrence of arboviruses by
driving their emergence or resurgence. Moreover, different sociodemographic factors
associated with urbanization, such as social inequality, health care capacity, food safety,
population density, and inadequate infrastructure, could be associated with human disease
outbreaks [301,303].

Recent studies in the Amazon region have focused on the relationship between de-
forestation, vector mosquito abundance, and arbovirus outbreaks [299,302,304]. Forest
fragments and growing agricultural areas show higher abundance, richness, and diversity
of mosquito species. Conversely, mosquito abundance and richness decreased in the urban
environment [304,305]. Consequently, anthropophilic species, such as Ae. aegypti, could
become very efficient vectors in urban areas compared to Culex spp. and Ae. albopictus,
which feed on a broader range of hosts [303]. Moreover, several species that serve as vectors
of multiple human pathogens appear to benefit from deforestation, including species found
in Central America, such as An. darlingi, Ae. aegypti, and Cx. quinquefasciatus [300].

Recent studies in central Panama found that mosquito diversity peaks in pristine forest
habitats, such as old-growth forests, while the abundance of colonist mosquito vectors (e.g.,
disturbed-areas specialists) increased significantly in highly disturbed forest sites [306,307].
These differences in mosquito abundances and diversity across various levels of forest
disturbance could be attributed to changes in ecological conditions that maybe affecting
the quality and availability of larval breeding sites. All together, these results suggest that
forest disturbance could drive VBD emergence and increase risk for disease transmission
in recently disturbed tropical regions due to the abundance of colonist-vector species.
These mosquito vectors tend to be opportunistic feeders, targeting hosts that are readily
available [308]. Moreover, a study by Bayles et al. (2020) in Costa Rica found that areas
with a high proportion of anthropogenic-altered landscapes, especially in areas with a high
degree of agricultural intensification, have the highest transmission risk for VBDs, such as
ZIKV, compared to protected areas [309].

The relationship between yellow fever and deforestation was initially established in
the first half of the 20th century when ecological observations were made regarding the
ability of Haemagogus mosquitoes to survive in forests with some degree of deforestation
pressure [310]. In recent years, the number of cases of sylvatic yellow fever cases in Brazil
have increased significantly in epizootic or transition areas and have been linked to large-
scale deforestation and forest fragmentation within urbanized settings [311,312]. The main
vectors of YFV in the Americas, Haemagogus spp. and Sabethes spp., share similar ecological
and bionomic characteristics, including acrodendrophilous behavior and opportunistic
feeding habits [313]. Studies on these vectors have shown that they are more abundant
in sites with lower forest cover suggesting that forest fragmentation could be a critical
factor in determining their presence [314]. For example, Sa. chloropterus was recently
found in deforested areas adjacent to a primary forest in Costa Rica [315], indicating that
deforestation could increase microhabitats for mosquito colonization, such as phytotelma
and tree holes, typically used by some enzootic yellow fever vectors [142,316,317]. Other
studies suggest that deforestation could be pushing YFV vectors to adapt their feeding
habits due to pressure from habitat disturbance. For instance, Hg. janthinomys Dyar,
1921 and other YFV vectors in Brazil were frequently observed descending to ground
level in the presence of humans conducting wood extraction activities or when the non-
human primate population number is small [318]. Moreover, others have found that their



Insects 2022, 13, 20 19 of 44

feeding behavior is more prolonged and aggressive at ground level than in the forest
canopy [319]. Haemagogus spp. and Sabethes spp. are also eclectic feeders able to shift their
host seeking among various wild or domestic animals and human hosts according to their
local availability, frequently moving vertically from the top of trees to feed at ground level.
However, it is not clear if this eclectic feeding behavior is innate or if it has been driven by an
increase in deforestation in regions where this species occurs [134,313,320,321]. Although
there is no current evidence of active YFV transmission by Ae. aegypti or Ae. albopictus [322],
experimental studies have determined that local domestic and peridomestic mosquito
populations may be competent vectors for YFV strains circulating in South America [323].

The recent reemergence and dispersion of yellow fever in Brazil, potentially driven
forest fragmentation near urbanized areas [312], raises the possibility of its resurgence
elsewhere in the continent, including parts of Central America, where these sylvatic vectors
are also present and similar ecological disturbances are taking place. The urbanization of
yellow fever is one of the biggest infectious disease threats in Latin America [324]. The
establishment of urban cycles, via Ae. aegypti and Ae. albopictus, could be catastrophic since
these vectors are widely distributed throughout the region [318]. Most people living in
urban areas of Latin America are not vaccinated, which could lead to high disease incidence
that could further spread transmission [325]. It has been estimated that the proportion of
infected people during an epizootic could be up to 29% under these conditions [326].

Deforestation can also affect non-human primate reservoirs of YFV and other ar-
boviruses in the Americas. Recent studies have found a strong association between non-
human primate diversity, their mobility patterns through forests, and the presence of
human yellow fever cases [327,328]. The importance of howler monkeys (Alouatta spp.)
as main reservoir host for several arboviruses, such as YFV, and their behavior must be
further explored. In Costa Rica, Alouatta monkeys spend most of their day (77%) in an
inactive state while a smaller proportion of their time is spent moving through the forest,
feeding, or engaging in social behavior. In contrast, capuchin monkeys (Cebus spp.) spend
most of the time (70–80%) day foraging or conducting other active behaviors [329]. The low
activity budget of Alouatta monkeys could make them more susceptible to mosquito bites
and YFV transmission, given that host movement has an important influence on the R0 in
vector-borne disease systems [330]. Furthermore, Alouatta spp. can adjust their behavior to
accommodate different feeding strategies as their forest habitat changes. These monkeys
are found in both modified and undisturbed habitats throughout their distribution in
Mexico and Central America. Their ability to adapt to changing environments is evidenced
in their continued presence in regions where white-faced capuchins and spider monkeys no
longer exist [331]. However, recent studies indicate that this behavior may not be spatially
consistent. A recent study by Schreirer et al. (2021) in Costa Rica found that A. palliata do
not appear to adjust their activity or spatial cohesion patterns in response to anthropogenic
edge effects due to forest fragmentation, suggesting that these monkeys exhibit less be-
havioral flexibility than A. palliata at some other sites [332]. The high susceptibility of A.
palliata to YFV infection and movement within disturbed or fragmented habitats could
increase the risk of arbovirus transportation and exposure to humans in urbanized forest
edges [333–336].

In Latin America, the relationship between DENV and deforestation has not been
clearly established or evidence is scarce. For instance, studies conducted in the Brazilian
Amazon have found no association [337,338]. Studies conducted in other endemic DENV
regions suggest that land use changes following deforestation (e.g., agriculture, settlements,
or road construction) have been identified as significant dengue fever risk factors [300,339].
An increase in human population densities benefits DENV and its vectors through the
establishment of artificial breeding habitats (i.e., water storage) and more frequent contact
with susceptible populations, thus increasing the risk for virus transmission in rural and
urban settings [340]. Other factors, such as proximity to paved roads and house clustering,
could also promote breeding habitats [341]. The expansion of urban centers and bordering
rural areas closer to the forest edge increases the chance of Ae. aegypti dispersal and



Insects 2022, 13, 20 20 of 44

colonization. For example, a study in the Peruvian Amazon by Guagliardo et al. (2014)
found that the geographic spread of Ae. aegypti is driven by human transportation networks
along rivers and highways in proximity to the city [342]. In this region, urban development
and the availability of oviposition sites appear to contribute to the colonization of Ae.
aegypti along roads. Moreover, unintentional transport of mosquitoes on boats disperses
their populations over long distances into rural, riverine communities [342].

Forest fragmentation driven by land use changes could also facilitate movement of
adult mosquito vectors between communities. For instance, a mark-release-recapture study
by Russell et al. (2005) reported that released Ae. aegypti exhibit nonrandom patterns of
dispersal with larger proportion of mosquitoes being recaptured along a corridor with
heavy shading from trees and vegetation [343]. In Central America, several studies have
been conducted using several vegetation indexes to evaluate correlations between vege-
tation coverage and dengue incidence. For example, a study by Fuller at al. (2009) was
able to explain up to 83% of the variability in weekly cases of dengue fever and dengue
hemorrhagic fever in Costa Rica between 2003 and 2007 based on vegetation indexes [344].
In another study, high dengue fever incidence correlated spatially with high temperature,
low altitude, and a high vegetation index [345]. Moreover, ovitrap egg counts are also
associated with vegetation indexes, which are dependent on temperature and rainfall, well
known factors affecting vector abundance [346–348]. Interestingly, other studies conducted
in Central America revealed that dengue cases may be directly associated with tree cover
and non-forested areas and inversely associated with built areas, probably because more
larval habitats are available in larger, tree-covered outdoor areas with vegetation during
rainfall episodes [349–351].

In Central America, there are increasing concerns regarding the expansion of Ae.
albopictus throughout the region. This important arbovirus vector species was initially
detected in El Salvador, Honduras, and Guatemala in 1995; later in Panama and Nicaragua
in 2002–2003; and Costa Rica and Belize between 2007 and 2009 [352–356]. During the
last decade, its distribution in Central America has expanded which is evidenced by
surveillance and field study reports [357–360], as well as by population genetic studies
in Panama and Costa Rica [361]. Recent studies in Costa Rica demonstrate the adaptive
capacity of this species to changes in land use and expansion of urbanized areas. For
example, a study by Calderon-Arguedas et al. (2019) detected all four DENV serotypes
in both Ae. albopictus female and male adults, and larvae associated with commercial
pineapple farming, suggesting local horizontal and vertical transmission of DENV in the
region [362]. Moreover, during these studies, most adult Ae. albopictus were collected within
forest galleries bordering pineapple fields, which may indicate that forest edge habitats
could serve as ecological refuges for this species in DENV endemic areas [186,362].

The movement of pathogens from sylvatic to amplified human transmission cycles in
rural and urban settings (e.g., disease spillover) have been commonly reported in the literature;
however, important questions on specific mechanisms still linger [363]. These events are typi-
cally associated with human and animal populations near forest edges, leading occasionally
to urban cycles involving peridomestic vectors [304,364]. Arboviral disease spillover events in
Central America are possible considering that these viruses share several ecological charac-
teristics, including vectoring by anthropophilic mosquitoes in urban and rural transmission
cycles, use of humans as main reservoirs, and their occurrence in habitats associated with
expanding agriculture and urbanization near forest edges [150,173,365]. Several studies in
Central America suggest the potential for disease spillover of sylvatic arbovirus transmission
cycles into rural and urbanized areas. For example, in Panama, evidence suggests that human
cases of VEEV infections are associated with spillover infections from an enzootic cycle involv-
ing sylvatic rodents and Cx. (Melanoconion) spp. mosquitoes [82]. Recent clusters of human
cases have occurred in Darien and Panama Provinces near rainforest and swamp habitats. In
addition, entry of humans into forest galleries appears to be a risk factor for VEEV transmis-
sion [82,113]. Other studies suggest frequent occurrence of spillover events associated with
other sylvatic arboviruses in Central America, such as MADV and MAYV [82,118,129,203,211].



Insects 2022, 13, 20 21 of 44

Conversely, spillback (e.g., reverse zoonosis) involves the movement of pathogens
from urban/rural transmission cycles to sylvatic cycles between wild non-human primates
and forest mosquitoes [366]. In the Americas, spillback events have been reported in
several countries, including Argentina, Brazil, Colombia, and French Guyana, involving
DENV, ZIKV, YFV, CHIKV, neotropical wild primates, and known mosquito vectors [150,
363,366,367]. A recent study carried out in Costa Rica found evidence of SLEV, WNV,
and DENV seroprevalence in the primate species A. palliata, A. geoffroyi, and S. oerstedii
(Reinhardt, 1872). Based on the collection of seropositive samples coinciding temporarily
and spatially with peaks of infections in human populations, their study concluded that
DENV exposure in these monkeys occurred through bidirectional human–wildlife contact
or bridging vectors [85].

The ecological mechanisms by which forest disturbance triggers disease spillback
events are still poorly understood, which makes it difficult to predict disease risk scenarios
for future outbreaks in human-altered forest habitats and the potential establishment of
sylvatic transmission cycles [363,366]. In Central America, where arbovirus-susceptible
wild primates and known mosquito vectors are present (e.g., Haemagogus spp., Sabethes
spp.), the occurrence of any one or a combination of epidemiological factors that could
drive transmission is highly plausible, considering the high level of deforestation, extensive
land use changes, and accelerating development of human settlements near forest edges. It
is also possible that immunity of monkeys against active sylvatic arboviruses, such as YFV,
could inhibit infections by DENV and ZIKV and, thus, evade the emergence of sylvatic
cycles [366]. However, considering that YFV epizootics have not been reported in Central
America in decades, cross-immunity to flaviviruses in non-human primates in the region
may not be sufficient to suppress other arboviral infections.

Several mosquito species distributed throughout Central America could serve as
bridging vectors and initiate spillback events due to their bionomic plasticity. For example,
Ae. albopictus is found predominantly in urban areas, but also spreads into rural, semi-
rural, and forest areas, which could potentially drive arboviruses, such as ZIKV, DENV,
and CHIKV, into sylvatic transmission cycles [363,366]. Moreover, sylvatic vectors like
Haemagogus spp. could also serve as bridging vectors since they can be opportunistic feeders,
frequently imbibing on humans and monkeys, in environmentally disturbed regions along
forest edges [140,313,321,368]. These mosquito vectors tend to be opportunistic feeders,
targeting hosts that are immediately available [308]. This evidence demonstrates that
disease spillover is not a random process but may be the result of forest degradation
increasing the likelihood of contact between humans and mosquito vectors in forest-altered
sites [306,307].

Both spillover and spillback events occur due to increasing human activities near adja-
cent forest areas and where humans closely interact with wild animals and their pathogens,
facilitating the “jump” or “shift” of new pathogens between different host species [302].
Despite our current knowledge of these diseases, it is still difficult to predict the risk for
future outbreaks since the ecological mechanisms that drive spillovers and spillbacks in
human-modified forest habitats are not well defined. Considering the accelerated rate of
deforestation, unplanned urbanization and agricultural expansion near forest edges, high
endemicity of multiple arboviruses, and close human contact with wildlife and vectors
taking place in Central America, there is a tangible possibility that spillback and spillover
events may be more frequent than reported and may surge in the future. Therefore, it is crit-
ical that more intense epidemiological and ecological monitoring systems are established
in the region to help predict future epizootics and epidemics.

5.2. Impact on Malaria

In recent decades, compelling evidence linking deforestation and land use changes
with malaria transmission and anopheline vector ecology in Central America have been
demonstrated [229,369–371]. Deforestation and land use changes can have profound
anthropogenic environmental effects on malaria transmission. In general, evidence shows



Insects 2022, 13, 20 22 of 44

that, in the Americas, deforestation leads to an increased potential for malaria transmission,
particularly in communities where the vector An. darlingi is present [372–376]. Working
specifically in the Darien Province of Panama, Loaiza et al. (2017) correlated increased
malaria incidence rates with extensive changes in landscape, including deforestation, and an
associated expansion of An. darlingi habitat. These findings were recently supported when
An. darlingi was implicated as an important vector of P. vivax in the Darien region [306,377].
Studies in other parts of Latin America have found that the human-biting activity of An.
darlingi is more intense in areas associated with deforestation and road development [378],
while a study by Loaiza et al. (2008) found that the Central American malaria vectors,
An. vestitipennis and An. neivai Howard, Dyar and Knab, 1913 are closely associated with
specific forms of vegetation and land-use practices in Panama [217].

Other studies have suggested that agricultural land use changes and environmental
pollution can significantly alter natural malaria vector habitats and even affect mosquito
diversity driving disease prevalence and more dominant vector populations. For instance,
a study by Chapin and Wasserstrom (1981) showed that, as early as the 1970s, expanding
the acreage used for cotton cultivation, and the associated increases in pesticide use in
Guatemala, Nicaragua, and El Salvador, correlated with both the emergence of DDT resis-
tance in local malaria vectors and rapid increases in annual malaria case incidence rates, in
some cases three times greater than previously recorded [379]. Similar findings were noted
in Belize, where malathion use in sugarcane cultivation was associated with malathion
resistance in local An. albimanus populations [380]. Another compelling case study from
Belize highlights how phosphorous runoff from sugarcane cultivation in proximity to
marshlands increases the amount of dense cattail (Typha domingensis) marsh habitat favored
by An. vestitipennis, the most efficient malaria vector in Northern Belize, and is associated
with higher larval densities [381,382]. Interestingly, a strategy to reduce this cattail habitat
through environmental management (e.g., mowing and burning) to control An. vestitipennis
populations in the area was marginally successful, with one important caveat: while the
habitat management strategies significantly reduced An. vestitipennis larval populations,
the resulting altered marshland habitats proved suitable for another important local vector
species, An. albimanus, whose populations increased [383]. Changes in landscape structure,
in the form forest cover percentage and forest density, could also lead to dominant vector
populations, such as An. darlingi, increasing the risk for malaria transmission [384].

Another aspect of environmental change, like urbanization, is the global expansion
of invasive malaria vectors, like An. stephensi Liston, 1901, and its potential implications
for malaria control and elimination throughout Central America. Often described as a
highly competent malaria vector able to breed in human-made containers, An. stephensi
is known to establish and sustain outbreaks of urban malaria in previously malaria-free
regions [385–387]. While the arrival and establishment of An. stephensi in Central America
is probably less likely than in Africa or other parts of Asia [385], it is nonetheless realistic to
consider the possibility. Regional receptivity to the vector and its potential to impact local
malaria control and elimination initiatives is poorly understood but potentially worrisome
given the rapid urbanization of population centers, the high abundance of peridomestic
habitats, its invasive potential, and intense human migration from malaria-endemic regions
through Central America [388,389].

Although the risk for malaria transmission in Central America has been steadily de-
clining in recent years, the intensification of human migration and illegal drug trafficking
throughout the region, which are driving deforestation and land use changes, have the po-
tential to reignite the resurgence of malaria in the region [370]. Currently, the La Mosquitia
tropical forest region, located between Honduras and Nicaragua, is one of the most malaria
endemic locations in Central America [229]. Although this region is part of the protected
MBC, it continues to be threatened by illegal drug smuggling, cattle ranching, logging,
and land grabbing [22,38,50]. The recent increase in malaria cases in Panama, Costa Rica,
and Nicaragua, especially of the more severe form, P. falciparum, raises concerns about
the possibility of eradicating malaria in those countries in the next few years. This recent



Insects 2022, 13, 20 23 of 44

resurgence has been linked to imported cases from foreign migrants passing through the
region [390]. While numerous findings highlight the effects of land use and micro and
macro habitat changes have on of malaria vectors and disease risk across Central America,
it is also evident that more research into these complex relationships and interactions is
desperately needed. An improved understanding of anopheline vector ecologies will help
inform on how malaria transmission dynamics might change over time and guide future
malaria control and elimination activities in the region.

5.3. Impact on Tick-Borne Diseases

Studies in different areas of the world have demonstrated that deforestation practices
and forest fragmentation could both decrease and modify biodiversity, which may lead
to a greater abundance of fewer vertebrate species and an increased prevalence of their
associated ticks that can thrive in modified habitats [391–397]. In addition, some domestic
animals, such as dogs and horses, may act as bridges for ticks and zoonotic pathogens from
wildlife to humans, resulting in disease spillover events into urban environments [398–401].
It is also important to consider environmental factors, such as temperature, precipitation
and humidity, critical to host and tick survival, which could also change in modified
environments. Each host, vector, and pathogen has its own requirements, which must be
met to ensure their establishment and survival [394,402–406].

Zoonotic pathogens may resurge depending on the conditions present, such as avail-
ability of humans and other domestic animals, tick vectors, and environmental conditions.
When conditions are optimal and disease vectors and suitable vertebrate reservoir hosts
become abundant, pathogen transmission risk may increase. Decreased contact between
ticks and vertebrate hosts not relevant in pathogen transmission can also increase disease
risk, which could cause a dilution effect. However, the mechanisms of this effect depend
on scale and overall context, including local land use changes, climate/microclimate condi-
tions, host communities, and their interactions [395,397]. Therefore, if humans encounter
tick vectors in these deforested or disturbed areas, there may be a greater risk of zoonotic
disease transmission.

In general, the scientific literature concerning tick-borne diseases in Central America
focuses on the distribution or presence of tick species and their associations with vertebrate
hosts and reports of human infections by specific pathogens (for examples see [233,234,236,
238,241,243,244,247,249,253–255,257,259,261–264,278,285,293,404,407–413]). However, very
few studies in the region have focused on the influence of land use changes on pathogen
and vector ecology. Except for recent studies in Panama [405,414,415], no other published
reports in the scientific literature have directly and specifically investigated the effects of
forest modification, deforestation, or land use changes on tick species, or on pathogen
transmission dynamics.

Among the few studies conducted in the region, one conducted in the Chiriquí
province of Panama aimed to identify environmental predictors of tick burdens on dogs, as
well as environmental predictors of pathogens in these ticks, including vegetation cover
and land use change [405]. Although the most relevant predictor of tick prevalence and
abundance was elevation, a decrease in vegetation cover linked to increased urbanization,
was also associated with the highest tick prevalence and abundance. This is probably due
to the close