Acta Tropica 153 (2016) 93100

Contents lists available at ScienceDirect

Acta Tropica
journal homepage: www.elsevier.com/locate/actatropica

Comparative study on the effectiveness of different mosquito traps in

arbovirus surveillance with a focus on WNV detection
Alex Pezzin, Victoria Sy, Arianna Puggioli, Rodolfo Veronesi, Marco Carrieri,
Bettina Maccagnani , Romeo Bellini
Centro Agricoltura Ambiente G. Nicoli, Department of Medical and Veterinary Entomology, Via Argini Nord 3351, 40014,Crevalcore, Italy

a r t i c l e

i n f o

Article history:
Received 27 March 2015
Received in revised form
29 September 2015
Accepted 5 October 2015
Available online 20 October 2015
Keywords:
Culex pipiens
Mosquito trap
West Nile
Arbovirus surveillance
Physiological age

a b s t r a c t
The selection of the ideal trap for arbovirus surveillance is an issue of primary importance to increase
the sensitivity of virus detection and the cost-effectiveness of the entomological surveillance. During
the summer 2011, the effectiveness of ve types of mosquito traps (CDC gravid trap, CO2- baited trap,
BG-SentinelTM and two experimental prototypes) to attract females potentially infected with West Nile
virus were assessed. The study was carried out in three natural wetland sites located in the EmiliaRomagna Region (Northern Italy), using a Latin square scheme. Single night collections of adult females
were performed and determination of species and physiological state (gravid, nulliparous or parous) was
made upon return to the laboratory. The species most frequently collected in the gravid trap was Culex
pipiens sl. L., being gravid females the large majority of the individuals. Species diversity was much higher
in CO2- baited traps, which may therefore enable a more comprehensive description of the vector species
composition and their role in arboviruses circulation. Our ndings indicate that gravid traps can be a
valid tool and should be integrated in the West Nile virus surveillance system in the Emilia-Romagna
region, mainly based on collections made with CO2 -baited traps.
2015 Elsevier B.V. All rights reserved.

1. Introduction
The increasing international trade and tourism, and the globalization of work are rapidly changing the distribution of arboviruses
worldwide, posing new concerns on public health due to the
increased risk of mosquito-transmitted arboviruses. In Europe,
the most dangerous mosquito-borne viruses are the four dengue
serotypes (DENV) causing the dengue fever, and the West Nile
virus (WNV) that can cause lethal encephalitis, as well as the Usutu
(USUV) and Bagaza viruses (BAGV) (Agero et al., 2011; Roiz et al.,
2012a; Vazquez et al., 2011) all belonging to the Flaviviridae family.
WNV lineage 1 has been responsible for repeated disease outbreaks
in the Mediterranean basin over the past 50 years. In 2004, and in
subsequent years, the WNV lineage 2 appeared to spread throughout Hungary and Austria, and subsequently emerged in Greece in
2010 and in Italy in 2011, involving outbreaks on the Italian mainland and Sardinia. Further spread through the Balkan countries is
also suspected (Hernndez-Triana et al., 2014). WNV is transmitted
in an avian cycle by ornithophilic mosquitoes, chiey of the genus

Corresponding author at: Via Argini Nord, 335140014, Crevalcore (BO) Italy.
E-mail address: bmaccagnani@caa.it (B. Maccagnani).
http://dx.doi.org/10.1016/j.actatropica.2015.10.002
0001-706X/ 2015 Elsevier B.V. All rights reserved.

Culex, being mammals dead end hosts because viraemia is generally

too low to infect mosquitoes (Reiter 2010). Aedes albopictus (Skuse)
recently established in Southern Europe, is the vector species of
Chikungunya and Dengue viruses (Togaviridae family, Alphavirus
genus), and it was responsible of an outbreak of chikungunya disease in Emilia-Romagna (Northern Italy) (Rezza et al., 2007), and
of several outbreaks in France (Delisle et al., 2015; Semenza et al.,
2014). In addition, in 2007, in Northern Italy insect avivirus DNA
sequences integrated in Ae. albopictus populations were found by
Roiz et al. (2009), while in 2008 a new insect avivirus was detected
in one pool of Ae. cinereus/geminus (Meigen) mosquitoes (Roiz
et al., 2012a), whose signicance for human health has still to be
addressed.
This scenario requires the development of effective surveillance programs, where the entomological surveillance should allow
detecting emerging viruses in eld-collected mosquitoes. It has
been demonstrated, both in the US and Europe (Dennet 2007; Hoel
et al., 2009; Hublek et al., 2010; Romi et al., 2004), that regular
mosquito surveillance programs can enable to detect virus circulation some weeks before the appearance of human cases (Unlu et al.,
2009). Depending on the arboviruses under surveillance, different
systems monitoring non-human hosts, vector species or human
categories at risk must conveniently be chosen in order to maximize

94

A. Pezzin et al. / Acta Tropica 153 (2016) 93100

virus detection. WNV can spread very quickly and easily, and may
threaten humans health before its detection (Kramer et al., 2008).
Bustamante and Lord (2010), using a model that simulates the process of mosquito sampling, pooling, and virus testing, found that
mosquito infection rates commonly underestimate the prevalence
of arbovirus infection in a mosquito population. They conclude that
other factors, like mosquito population size, age structure, weather
and historical baseline data have to be considered to assess the risk
of arbovirus transmission. According to Bellini et al. (2014a), the
development of surveillance programs and vector control strategies
in European regions based on data obtained from studies performed in North America is not entirely appropriated (and also
not feasible due to different legislative frameworks). The diversity of the susceptible bird fauna and the vector species involved
in the enzootic and tangential transmission of WNV in Europe,
pose a situation remarkably different to that in US (Bellini et al.,
2014a). The improvement of entomological surveillance methods
and programs can increase the reliability of the risk assessment and
contribute to reduce the gap between infection rate estimates and
the risk of arbovirus transmission to humans and animals (Gu et al.,
2008; Roiz et al., 2012b).
To develop an efcient vector surveillance system it is necessary to gain knowledge about the following issues: (i) which
environmental conditions are necessary for an outbreak to occur;
(ii) biology and ecology of the main vector species in the areas
potentially at risk; (iii) the efciency of the different types of
trap in attracting the vector species; (iv) the physiological age of
the captured females (i.e. nulliparous and parous females) (Hugo
et al., 2008). Recognition of parous females is important because
it implies females had completed at least one gonotrophic cycle
and had taken a blood meal, with the chance to be infected in
case the host is viraemic. In addition, the identication of gravid
females is also important, because having a batch of eggs ready
to be laid implies they already took a blood meal, independently
of the physiological age. In the same way, the presence of blood
in the stomach is an indication of a potentially infectious female.
Thus, mosquito traps capturing a higher proportion of potentially
infectious females will enhance the probability to predict a disease
outbreak (Bellini et al., 2014a,b). Concerning the efciency of different models of traps in attracting mosquito females, Kesavaraju
et al. (2011) and Allan and Kline (2004) compared some commercial models of gravid traps with structural differences and found
that several characteristics signicantly affect mosquito collection
efcacy. Commercial gravid traps differ in basic design, color and
size of the tank that contains the infusion, giving different capture results (Allan and Kline, 2004, Dennet, 2007; White et al.,
2009). Moreover, different infusions (aquatic grasses like Juncus
effusus L., Rhyncospora corniculata (Lamarck) and Typha latifolia L.,
cow manure, mix of grass clippings, wheat straw, rabbit chow)
can attract different mosquito species depending on the season
(Burkett and Mullen 2008; Jackson et al., 2005; McPhatter et al.,
2009). According to Williams and Gingrich (2007), the use of gravid
traps could give better results for West Nile virus surveillance over
light traps or resting boxes. Many types of traps use carbon dioxide as the primary attractant. The produced plume of CO2 mimics
human exhalation and thus makes these traps quite specic for
capturing blood-seeking females. CO2 traps allow the collection
of large numbers of mosquitoes and appear to be highly attractive to a wide variety of mosquito species. CO2 traps are widely
used in Italy for vectors monitoring and surveillance (Bellini et al.,
2003; Calzolari et al., 2010). Traps called resting boxes are passive devices that serve as shelters for mosquitoes during the day
(Crans 1995). They have been used to sample mosquito populations
since the time of the malarias control programs, and still they are
successfully used for monitoring many Anopheles species (Kweka
et al., 2010; Pombi et al., 2014). However, they perform very dif-

ferently depending on the technical aspects of construction and on

the richness of resting sites in the study area (LAmbert et al., 2012;
Panella et al., 2011). The BG-Sentinel mosquito trap mimics convection currents created by a human body and it is widely used
in many parts of the world especially for the collection of Aedes
mosquitoes (Maciel de-Freitas et al., 2006; Bhalala and Arias 2009;
Farajollahi et al., 2009; Bhalala et al., 2010). Different variations of
the BG-Sentinel can be used according to the target species. The
trap can be used with or without carbon dioxide, and with Biogents proprietary attractant for anthropophilic mosquitoes (such
as Aedes aegypti L. or Culex quinquefasciatus Say).
Our study was designed to compare the effectiveness of ve
mosquito traps in measuring species abundance and composition
in wetland habitats, and to analyze the attractiveness towards
potentially infectious females. The work was performed in the
perspective of a wide entomological surveillance program, with a
focus on Cx. pipiens s.l. L. being WNV the most widely distributed
arbovirus in Northern Italy.

2. Materials and methods

2.1. Study period and study areas
The study was run from June, 14 to September, 16, 2011
in three wetland sites in the Emilia-Romagna region: La Rizza
(44 39 41.82 N - 11 26 19.55 E), Le Vallette (44 44 33.18 N .11 57 19.95 E) and Oasi Val di Sole (44 56 28.09 N - 11 2 24.44 E)
(Fig. 1).
La Rizza is situated in the municipality of Bentivoglio (BO); it is a
natural protected area of about 1,500 ha. The dense vegetation hosts
many aquatic bird species, such as ducks, cormorants and herons.
This area includes permanent wetlands, wet meadows, reed beds,
copses and hedges, but also wetland tanks, shing lakes and two
observation sheds located in an expansion of the Navile canal. A
white stork Ciconia ciconia (L.) conservation center is also present.
Le Vallette is in the municipality of Ostellato (FE). It is a wetland
area of approximately 300 ha located between two canals, which
act as its boundaries. At least 150 bird species, mainly aquatic,
can be observed. Reeds are the predominant species, but trees like
poplar, elm and willow are also present.
Oasi Val di Sole is a natural protected area located in Concordia
sulla Secchia (MO) originated from the excavation of clay which
began in the 80s. It extends for an area of approximately 25 ha
between the Po and the Secchia rivers and consists of four main
basins, two ponds, ridges and canyons that make up a rest and nesting area for several bird species. More than 200 bird species have
been observed, including some quite rare species like the ferruginous duck Aythya nyroca (Guldenstadt), which is the symbol of the
oasis.

2.2. Mosquito traps

Five types of traps were compared in this study: the CDC
Gravid trap (John W. Hock Company, Gainesville, Florida, model
1712) (Fig. 2A), the CO2 -baited trap (CAA, Crevalcore, Italy, model
CAA2004) (Fig. 2B), BG-S trap (BG-SentinelTM , Biogents GmbH,
Regensburg, Germany) (Fig. 2C), and two experimental prototype
traps specically designed and manufactured for this study by the
authors (Fig. 2DE).
Gravid trap. The infusion was prepared as follows: 5 Lof tap
water with 2.5 g of dry brewer yeast and 30 g of dry grass hay. The
preparation was kept at 26 1 C in dim light inside a wide mouth
open tank for 3 days. The infusion was stirred once a day to enhance
the fermentation (Burkett 2005; Irish et al., 2012).

A. Pezzin et al. / Acta Tropica 153 (2016) 93100

95

Fig. 1. Map of the study sites.

CAA2004 CO2 -baited trap. The trap was constituted by a black

isolated container for dry ice 2 L capacity, with four holes which
open 8 cm above the air-sucking cylinder, so that the CO2 plume
could uniformly disperse around the trap. The trap was charged
with 500 g dry ice in a single block, and was hung on a tree branch
at 1.5 m from the ground (Bellini et al., 2003).
BG-S trap. The trap was baited with the BG-Lure attractant supplied by the manufacturer.
Resting box traps. The two experimental resting trap prototypes
were developed to catch mosquitoes without any kind of attractant, but offering them a suitable place to rest. The rst prototype
was a blue plastic drum of 200 L capacity lying horizontally, which
remained open overnight (RT001). The second one was a cylindershaped brown plastic basket 31 cm tall with a 30 cm diameter
opening and a downward-directed 12 V fan to suck approaching
mosquitoes (RT003). In both cases, the Resting boxes were quickly
closed in the early morning, and mosquitoes were killed by placing
inside a piece of dry ice.
2.3. Mosquito collection and classication
The trial was conducted in 2011, and in each of the three study
areas ve stations were established at a distance of at least 15 m
between each other. Every week the position of the traps was
changed according to a Latin square scheme. Five rotations were
done in 5 consecutive weeks, so that each station hosted the ve
trap types one time. Collections were performed one night per week

according to the experimental scheme reported in Table 1 (5 collections in total). In the ve stations the traps were simultaneously
activated at 6:00 PM and stopped on the next morning at 9:00
AM. The collected mosquitoes were taken to the laboratory to be
counted and prepared for analyses. Mosquito species determination was performed using the taxonomic keys of Schaffner et al.
(2001) and Becker et al. (2010).
2.4. Potentially infectious females
To the aim of this study, we considered at risk to be infectious
all the females with a hint of having taken a blood meal. 3070 Cx.
pipiens females per collection date per trap were analyzed, and in
instances where collections were less than 30 Cx. pipiens females,
all the specimens were used for the study. The pools were stored
at 20 C until dissection.
Females were classied as nulliparous or parous following
the ovarian tracheation technique described by Detinova (1962).
Mosquito dissections were performed in a drop of saline solution
on a glass microscope slide using a stereomicroscope. With the aid
of an entomological needle, ovaries were separated from the rest of
the surrounding tissues, placed in a drop of clean de-ionized water,
and allowed to air dry to reveal ovary tracheation. Using a compound microscope, females were classied as parous if they had
uncoiled tracheoles (Fig. 3A), as nulliparous if they had coiled tracheoles (skeins) (Fig. 3B). The presence of completely developed
eggs prevented tracheoles visualization, and it was not feasible
to determine the parity status, thus females were classied as
gravid females. When other causes did not allow establishing the
parous condition, females were categorized as non-determined.
In addition, we recorded the presence of blood in the stomach
Table 1
Experimental scheme.

Fig. 2. The ve types of traps. (A) Gravid trap, (B) CAA2004 CO2 -baited trap, (C)
BG-Sentinel trap, (D) Resting trap 1 RT001, (E) Resting trap 3 RT003.

Site

Period

Frequency of
collection

No.
collections

La Rizzaa
Le Vallette
Oasi Val di Sole

June 24July 8
July 15August 5
August
19September 16

Once every 46 days

Once every 46 days
Once every 46 days

5
5
5

CO2 -baited trap was not working on June, 29.

96

A. Pezzin et al. / Acta Tropica 153 (2016) 93100

Fig. 3. Ovaries of Culex pipiens s.l.. (A) parous females ovary, (B) nulliparous females ovary. 200X.

independently of the parous condition. Based on the observations,

females were categorized as uninfectious or potentially infectious.
Nulliparous females without blood in the stomach were assigned
to the uninfectious category, while all parous females, nulliparous
and non-determined females with blood in the stomach, and gravid
females were considered potentially infectious. To the aim of this
study, the possible presence of autogenic females, which could have
initiated the ovarian development without the blood meal, has been
considered of negligible impact.
2.5. Descriptive analysis and statistics
In each of the three sites and for each of the ve trap types, the
number of females and males of the most frequent species were
counted and percentages were calculated.
In all the study sites, the attractiveness of the ve trap types
to Cx. pipiens females, was analyzed by a non-parametric ANOVA
(KruskalWallis test) due to lack of normality and homogeneity of
variances. Z- values and p values for multiple bilateral comparisons
were calculated.
In two of the three sites, Le Vallette and Oasi Val di Sole, the
collection efciency of Gravid traps and CAA 2004 CO2 -baited traps
towards potentially infectious females was compared. Angular data

transformation was necessary to satisfy the variances homogeneity requirements for 2-way ANOVA. Study sites and types of trap
were set as the main factors. The signicance level for statistics was
stated at alpha = 0.05. Analysis were performed using Statistica 7.0.
3. Results
3.1. Species composition in the three study sites
The total number of mosquitoes collected in the three sites was
18,760, 18,036 of which were females and 724 males. Considering
all the mosquito species, 6,204 mosquito females were captured in
La Rizza, 5,796 in Le Vallette, 6,036 in Oasi Val di Sole. The mosquito
species were those typically found in the Po plain rural areas: Cx.
pipiens, Cx. modestus Ficalbi, Aedes caspius Pallas, Ae. vexans Meigen,
Ae. cinereus Meigen, Ae. albopictus Skuse and Anopheles maculipennis
s.l. Meigen. The large majority of the specimens (85.0% of the males
and 86.7% of the females) belonged to the species Cx. pipiens. In
Table 2, for each site of study and for each trap type, the mean number of females per species per sampling day is reported. In La Rizza
Cx. pipiens accounted for 96.7% of the total number of catches, while
2.1% were Ae. vexans; being present the other species cited above
with percentages well below 1%. This was the only site in which

Table 2
Collection capacity of different trap types to different mosquito species: mean number of females per site per trap.
Site

Species

CAA2004 CO2- baited trap Gravid trap

Mean

SD

Mean

La
Rizza

Ae. albopictus
Ae.caspius
Ae.cinereus
Ae.vexans
An. maculipennis
Cx. modestus
Cx. pipiens

0.0
6.6
4.8
27.8
0.8
0.8
536.0

0.0
3.1
10.7
36.7
1.8
1.3
368.3

0.5
0.0
0.0
0.0
0.0
0.0
724.0

Le
Vallette

Ae. albopictus
Ae.caspius
Ae.cinereus
Ae.vexans
An. maculipennis
Cx. modestus
Cx. pipiens

1.0
145.8
0.0
3.6
2.2
4.2
468.4

1.0
238.2
0.0
3.4
2.4
4.1
336.2

1.6
0.0
0.0
0.6
2.6
0.0
456.4

Oasi
Val
di
Sole

Ae. albopictus
Ae.caspius
Ae.cinereus
Ae.vexans
An. maculipennis
Cx. modestus
Cx. pipiens

0.0
59.4
0.0
0.8
0.8
191.8
344.6

0.0
46.2
0.0
1.3
1.3
98.4
129.1

0.0
0.0
0.0
0.4
1.6
0.8
589.2

BG-S
SD

RT001

Mean

SD

0.2
0.0
0.0
0.0
0.2
0.2
78.2

0.5
0.0
0.0
0.0
0.5
0.5
120.2

0.0
0.2
0.0
0.0
0.0
0.0
1.4

0.0
0.5
0.0
0.0
0.0
0.0
2.61

0.0
0.0
0.0
0.0
0.2
0.0
5.0

0.0
0.0
0.0
0.0
0.5
0.0
8.6

1.67
0.0
0.0
0.89
2.3
0.0
327.8

3.8
0.0
0.0
0.0
3.8
0.4
20.8

3.8
0.0
0.0
0.0
6.9
0.9
10.8

0.0
0.2
0.0
0.0
6.2
0.0
15.6

0.0
0.45
0.0
0.0
10.5
0.0
5.7

0.4
0.0
0.0
0.0
1.2
0.2
20.4

0.9
0.0
0.0
0.0
0.8
0.5
11.5

0.0
0.0
0.0
0.5
1.5
1.10
404.1

0.2
0.2
0.0
0.0
1.2
1.6
6.4

0.5
0.5
0.0
0.0
1.6
1.1
4.6

0.0
0.4
0.0
0.0
0.8
0.0
2.0

0.0
0.0
0.0
0.0
1.4
0.0
3.6

0.0
0.0
0.0
0.0
2.1
0.0
4.3

1.0
0.0
0.0
0.0
0.0
0.0
1,244.3

Mean

RT003
SD

0.0
0.5
0.0
0.0
1.1
0.0
2.0

Mean

SD

A. Pezzin et al. / Acta Tropica 153 (2016) 93100

4. Discussion
The entomological surveillance plays an important role in
vector-borne disease surveillance thanks to the capability to provide early detection of arboviruses circulation (Gu et al., 2008;
Almeida et al., 2008; Bellini 2014a,b). Many different mosquito

2000

No. individuals

Collection capacity towards Cx. pipiens. The most efcient traps

in attracting Cx. pipiens females were the CAA2004 CO2 -baited trap
and the Gravid trap, while the others were much less effective.
In two sites out of three, the CAA2004 CO2 -baited trap showed a
lower variability in the number of females captured with respect
to the Gravid trap (Fig. 4AC). In La Rizza (KruskalWallis: H (4,
N = 24) = 17.91 P = 0.0013), the CO2 -baited trap captured signicantly higher numbers of females with respect to RT001 (P = 0.0086)
and RT003 (P = 0.0149), while no statistically signicant differences
were detected with other traps. In Le Vallette (KruskalWallis: H
(4, N = 25) = 17.73, P = 0.0014) the CO2- baited trap and the Gravid
trap captured signicantly higher numbers of females than RT001
trap (respectively, P = 0.0245 and P = 0.0213), while no other differences were seen from the other trap comparisons. In Oasi Val di Sole
(KruskalWallis: H (4, N = 25) = 18.42, P = 0.0010) the Gravid trap
captured higher numbers of Cx. pipiens females compared to either
RT001 or RT003 (respectively, P = 0.0127 and P = 0.0282), while the
CO2 -baited trap effectiveness was higher than that of RT001 trap
(P = 0.0321). In none of the sites there was a statistically signicant
difference between the efcacy of the CO2- baited trap and of the
Gravid trap.
Collection capacity towards potentially infectious females. To this
aim, only the performances of the two most efcient traps (Gravid
trap and CAA2004 CO2 -baited trap) were compared between two
sites, Le Vallette and Oasi Val di Sole. In total, the pools of Cx. pipiens females formed for the assessment of their physiological age
(i.e. to determine if they were potentially infectious) counted 583
specimens captured by Gravid traps and 305 captured by CAA2004
CO2 -baited traps. In Oasi Val di Sole the percentage of potentially infectious females captured by the Gravid trap ranged from
98.4% to 100.0% in the four sampling dates, while in Le Vallette
it ranged from 0.0% to 77.8%, showing a wide variability. On the
contrary, the CAA2004 CO2 -baited trap showed much less variability between the two sites, ranging from 25.7 to 85.7% in Oasi
Val di Sole, and from 47.2 to 78.6% in Le Vallette. The two-way
ANOVA (site and type of trap as main factors) showed that the
interaction between the two main factors is statistically signicant, stating that collection capacity of the two trap types towards
Cx. pipiens potentially infectious females was different in the two
sites (F(3, 16) = 19.33, P = 0.0045) (Fig. 5). The percentages of potentially infectious females were 64.67 13.39 for the Gravid traps and
51.31 6.64 (means standard errors) for the CO2 -baited traps,
and the statistical analysis did not show any signicant difference between the two types of trap (F(1,16) = 1.88, P = 0.189). On
the contrary, a statistically signicant difference between the percentages of potentially infectious females was found between the
two sites (Le Vallette: 44.26 9.26; Oasi Val di Sole: 71.74 10.31)
(F(1,16) = 11.33, P = 0.004).

2400

1600
1200
800
400
0
BG S CO2 RT001 RT003 Gr T
Median
25%-75%
Non-Outlier Range

KW-H(4;25) = 17.728; p = 0.0014

1000
800

No. individuals

3.2. Trap collection capacity

KW-H(4;24) = 17.9103; p = 0.0013

600
400
200
0
-200

BG S CO2 RT001 RT003 Gr T

Median
25%-75%
Non-Outlier Range

KW-H(4;25) = 18.4186; p = 0.0010

1200
1000

No. individuals

Ae. cinereus was captured, on 29/06/2011 with the CO2- baited trap
CAA2004 (0.39%). In Le Vallette 84.7% of the females were Cx. pipiens, while Ae. caspius accounted for 12.6%, and An. maculipennis s.l.
for 1.4%; Ae. cinereus was not present, while the other species were
below 1%. In Oasi Val di Sole 78.4% of the collected females were
Cx. pipiens, 5.0% were Ae. caspius and 16.1% were Cx. modestus; all
the other species were present at percentages below 1%.

97

800
600
400
200
0
BG S CO2 RT001 RT003 Gr T
Median
25%-75%
Non-Outlier Range

Fig. 4. AC. Median number of Cx. pipiens females per trap per night in the three
study sites. (A) La Rizza, (B) Le Vallette, (C) Oasi Val di Sole. (Software: Statistica 7.0).

collecting traps have been developed to survey/monitor vector

mosquito species (Service 1993), which should accomplish two
main trait: (i) early detection capacity when focused on one specic
arbovirus which is known to occur in the area; (ii) large spectrum of
arbovirus detection when oriented to the surveillance of all the possible arboviruses that could be present in the area. In the rst case,
when the vector species is/are well known, the selection criteria for
the choice of the trap should conveniently be the performance in

98

A. Pezzin et al. / Acta Tropica 153 (2016) 93100

% Cx. pipiens females potentially

infected

Oasi Val di Sole

Le Vallette

140
120
100
80
60
40
20
0
-20

Gravid trap

CAA2004

Fig. 5. Percentage of potentially infectious females in the pools formed from the
captures of the Gravid trap and of the CAA2004 CO2 -baited trap in two study areas.
The following females categories were considered potentially infectious: all the
parous females, nulliparous females and females of undetermined parity status with
blood in the stomach, gravid females. (A) Le Vallette, (B) Oasi Val di Sole. (Software:
Statistica 7.0).

capturing the target vector species. In the second case, it might be

convenient to rely on traps that collect a larger number of mosquito
species (Hublek et al., 2010; Roiz et al., 2012b).
Our study was designed (i) to increase the efciency of the entomological surveillance plan in the Emilia-Romagna region, which
currently relies on CO2- baited traps and is mainly targeted to the
WNV surveillance (Calzolari et al., 2010; Carrieri et al., 2014; Bellini
et al., 2014a,b); (ii) to increase the efcacy of the plan in detecting the vector mosquito species composition, standing the risk of
other vector-borne human diseases, like Chickungunya, Dengue
and Usutu fever. According to our results, the best performing trap
in terms of absolute number of captured mosquitoes was the Gravid
trap, while the CAA2004 CO2- baited trap was more efcient in
attracting a wider spectrum of species. In fact, ve mosquito species
were detected by Gravid traps, (only two species at noticeable percentages) in comparison to the seven mosquito species captured by
the CAA2004 CO2 -baited trap. CAA2004 trap performed very well
either on Culex sp. or Aedes sp., and it was the only one capturing Ae.
cinereus, while it was less efcient in attracting Anopheles sp.. The
Gravid trap showed low efciency towards Aedes species as well
as towards Cx. modestus. The BG-S and the two prototypes of Resting trap showed much lower capture efciency in terms of species
diversity and total number of specimens. RT001 was someway
attractive to An. maculipennis s.l., while BG-S seemed to be more
attractive than the other traps for Ae. albopictus as expected, being
this trap designed to catch Aedes Stegomyia mosquitoes. As a conclusion, to the aim of improving the surveillance program on WNV
circulation in the region and the detection of other arboviruses,
Gravid traps and CAA2004 were the most suitable traps, while BGS, and the non-activated resting boxes RT001 and RT003 proved to
be not suitable for wide area monitoring plans in non-urbanized
and wetland areas.
Regarding species composition, Cx. pipiens was the dominant
species in all the sites. Aedes cinereus was captured only in La Rizza,
while in Le Vallette Ae. caspius showed a much higher percentage than in the other sites. The presence of An. maculipennis s.l.
was fairly homogeneously distributed among the three sites at
low percentages (below 1.0%), while Cx. modestus was captured at
noticeable numbers in Oasi Val di Sole. The number of specimens
of Ae. albopictus was low in all the sites, as expected on the basis of
the bioecology of the species, whose preferential breeding sites are
articial containers, in agreement with its adaptation to colonize
urbanized environments (Vallorani et al., 2015) but in Le Vallette it
was someway higher than in the other places, probably because of

the presence of factories nearby, with potentially active breeding

sites, likely catch basins.
Comparing the attractiveness of the two most effective traps,
Gravid trap and CAA2004 trap, towards potentially infectious
females (those females that have probably got in contact with the
virus) we found that the interaction between types of trap and
study areas was statistically signicant (Fig. 5). These ndings
lead to two considerations: (1) ecological factors can differently
affect the capacity to depict mosquito species diversity and
abundance depending on the type of trap. In particular, it is likely
that the availability of suitable larval habitats can reduce Gravid
trap attractiveness to females ready to lay eggs, in agreement
with (LAmbert et al., 2012), while it has a limited effect on the
trapping capacity of a generalist trap type like the CO2 -baited trap.
(2) Under ideal conditions, Gravid traps can collect the highest
proportion of gravid females in comparison to other types of trap,
but if we consider all parous females and those with blood in the
stomach as potentially infectious (in addition to the gravid ones),
the capacity of the CAA2004 CO2 -baited trap to early monitor the
potential risk of WNV outbreak is not different from that of the
Gravid trap. An attempt to improve the effectiveness of CO2 traps
in virus detection has been done by Hurk van den et al., (2014),
in Northern Australia by offering the insects honey-baited nucleic
acid preservation cards on which they can expectorate the virus.
In their study, overall 20/144 (13.9%) of traps from different weeks
contained at least one virus-positive card.
As a conclusion, we consider the CAA2004 CO2 -baited trap,
positioned at 1.5 m height, as the more appropriate trap for a reliable West Nile virus integrated surveillance system, thanks to its
attractiveness to a wide range of species. Its plasticity enables
researchers to gain information on the mosquito species composition, a baseline knowledge of increasing importance standing the
possibility of other mosquito vectored virus outbreaks (Semenza,
2015). However, in spite of the difculty to standardize the performance of the Gravid traps, compared to the CO2 traps, as the
Gravid trap collected the highest percentages of potentially infectious females, we believe they should be considered an effective
additional tool to increase the fraction of individuals suitable
for being analyzed for of WNV and other arboviruses, as other
authors suggested (Godsey et al., 2005). Their use could be recommended in particular after the detection of virus circulation
to nalize the mosquito pool analyses and to enable a more targeted blood unit testing strategy, thus reducing the risk of virus
transmission via blood, tissue and organ donation (Bellini et al.,
2014b).
In the Emilia-Romagna region, since 2009 (Angelini et al., 2010;
Regione Emilia-Romagna, 2015) a system integrating environmental (mosquitoes and birds) and human surveillance has been
implemented and progressively improved. The monitoring plan
relies mainly on CO2 -baited traps with the addition in 2013 and
2014, of 7 and 9 Gravid traps, respectively. The system has shown
highly satisfactory results in terms of early detection capacity (the
environmental surveillance component allowed detection of WNV
circulation 34 weeks before human cases of WN Neuroinvasive
Disease occurred) (Bellini et al., 2014b; Chancey et al., 2015), sensitivity (capacity to detect virus circulation even at the enzootic level)
and area specicity (capacity to indicate the spatial distribution of
the risk for WNND) (Bellini et al., 2014b).
Costs is another essential aspect of the arbovirus surveillance
efforts, because scarce economic resources must be used as effectively as possible, requiring a thorough analysis of the strategies
that a surveillance program will use (Scott et al., 2001; Gu et al.,
2008). The Department of Public Health of the Emilia-Romagna
Regional Government calculated that the surveillance program
helped to reduce costs EUR 2,560 million to EUR 2,093 million
(Bellini et al., 2014b). No data are available on the different con-

A. Pezzin et al. / Acta Tropica 153 (2016) 93100

tribution of CO2 -baited traps and Gravid traps to virus detection,

thus further effort should be made to systematically extend the
comparison between the two types of trap in different habitats on
the large scale.
Acknowledgments
This work was nanced by the Regione Emilia-Romagna, Public
Health Department DGR N. 2113/2010 2113/2010 Experimental
program for the integrated medical & veterinary monitoring of
arthropod transmitted diseases.
The study is part of the PhD research program of Alex Pezzin
Development of tools and methods for the surveillance and monitoring of Culicid species of sanitary importance at the Department
of Agroenvironmental Sciences and Technologies of the University
of Bologna.
References
Angelini, P., Tamba, M., Finarelli, A.C., Bellini, R., Albieri, A., Bonilauri, P., Cavrini, F.,
Dottori, M., Gaibani, P., Martini, E., Mattivi, A., Pierro, A.M., Rugna, G., Sambri, V.,
G, Squintani, Macini, P., 2010. West Nile virus circulation in Emilia-Romagna,
Italy: the integrated surveillance system 2009. Eurosurveillance 15 (April (16))
http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=19547.
Agero, M., Fernndez-Pinero, J., Buitrago, D., Snchez, A., Elizalde, M., San Miguel,
E., Villalba, R., Llorente, F., Jimnez-Clavero, M.A., 2011. Bagaza Virus in
Partridges and Pheasants, Spain. Emerg. Infect. Dis. 17 (8), 14981501.
Allan, S.A., Kline, D., 2004. Evaluation of various attributes of gravid female traps
for collection of Culex in Florida. J. Vector Ecol. 29, 285294.
Almeida, A.P.G., Galo, R.P., Sousa, C.A., Novo, M.T., Parreira, R., Pinto, J., Piedade, J.,
Esteves, A., 2008. Potential mosquito vectors of arboviruses in Portugal:
species, distribution, abundance and West Nile infection. Trans. R. Soc. Trop.
Med. Hyg. 102 (8), 823832.
D., Zgomba, M., Boase, C., Madon, M., Dahl, C., 2010. Mosquitoes
Becker, N., Petric,
and their control. Springer, Heidelberg, Dordrecht, New York.
Bellini R., Veronesi R., Gentile G., Alberani A., Pandol N., 2003. Presenting a new
standard CO2 trap model. The 14th European Conference of the Society of
Vector Ecology, September 36, 2003 Bellinzona, Switzerland. http://www.caa.
it/content/News/pubblicazioni.asp?page no=15&pagecount=10&CatID=1.
Bellini, R., Zeller, H., Van Bortel, W., 2014a. A review of the vector management
methods to prevent and control outbreaks of West Nile virus infection and the
challenge for Europe. Parasites Vectors 7, 323 http://www.
parasitesandvectors.com/content/7/1/323.
Bellini, R., Calzolari, M., Mattivi, A., Tamba, M., Angelini, P., Bonilauri, P., Albieri, A.,
Cagarelli, R., Carrieri, M., Dottori, M., Finarelli, A.C., Gaibani, P., Landini, M.P.,
Natalini, S., Pascarelli, N., Rossini, G., Velati, C., Vocale, C., Bedeschi, E., 2014b.
The experience of West Nile virus integrated surveillance system in the
Emilia-Romagna region: ve years of implementationItaly, 2009 to 2013. Euro
Surveill. 19 (44), pii=20953.
http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=.
Bhalala, H.V., Arias, J.R., 2009. The ZumbaTM mosquito trap and BG-SentinelTM trap:
novel surveillance tools for host-seeking mosquitoes. J. Am. Mosq. Control
Assoc. 25, 134139.
Bhalala, H.V., Smith, J.D., ODea, B.A., Arias, J.R., 2010. The efcacy of the
BG-SentinelTM CO2- baited nozzle in collecting host-seeking mosquitoes in
Fairfax County, Virginia. J. Am. Mosq. Control Assoc. 26 (2), 226228.
Burkett N.D., 2005. Comparative study of gravid trap infusion for capturing
blood-fed mosquitoes (Diptera: Culicidae) of genera Aedes, Ochlerotatus, and
Culex. Thesis, Auburn University, Alabama. http://etd.auburn.edu/bitstream/
handle/10415/771/BURKETT NATHAN 18pdf?sequence=1.
Burkett, N.D., Mullen, G.R., 2008. Comparison of infusions of commercially
available garden products for collection of container breeding mosquitoes. J.
Am. Mosq. Control Assoc. 24 (2), 236243, http://dx.doi.org/10.2987/5597.1.
Calzolari, M., Bonilauri, P., Bellini, R., Caimi, M., Delippo, F., Maioli, G., Albieri, A.,
Medici, A., Veronesi, R., Pilani, R., Gelati, A., Angelini, P., Parco, V., Fabbi, M.,
Barbieri, I., Lelli, D., Lavazza, A., Cordioli, P., Dottori, M., 2010. Arboviral survey
of mosquitoes in two Northern Italy regions in 2007 and 2008. Vec. Borne Zoon
Dis., http://dx.doi.org/10.1089/vbz.2009.0176.
Carrieri, M., Fariselli, P., Maccagnani, B., Angelini, P., Calzolari, M., Bellini, R., 2014.
2014, Weather factors inuencing the population dynamic of Culex pipiens
(Diptera: Culicidae): analysis of the period 1997- in the Po Plain Valley, Italy.
Environ. Entomol. 43 (2), 482490, http://dx.doi.org/10.1603/EN13173.
Chancey, C., Grinev, A., Volkova, E., Rios, M., 2015. The global ecology and
epidemiology of West Nile Virus. Hindawi Publishing Corporazion. BioMed.
Res. Int. 2015, Article ID 376230, http://dx.doi.org/10.1155/2015/376230.
Crans, W.J., 1995. Resting boxes as mosquito surveillance tools. In: Proceedings of
the 82nd Annual Meeting of the New Jersey Mosquito Control Association;
Atlantic City, NJ. 7-10 March 1995, New Brunswick, NJ: New Jersey Mosquito
Control Association Inc, pp.
5357.

99

Delisle, E., Rousseau, C., Broche, B., Leparc-Goffart, I., LAmbert, G., Cochet, A., Prat,
C., Foulongne, V., Ferr, J.B., Catelinois, O., Flusin, O., Tchernonog, E., Moussion,
I.E., Wiegandt, A., Septfons, A., Mendy, A., Moyano, M.B., Laporte, L., Maurel, J.,
Jourdain, F., Reynes, J., Paty, M.C., Golliot, F., 2015. Chikungunya outbreak in
Montpellier, France, September to October 2014. Eurosurveillance 20 (April
(17)).
Dennet, J.A., 2007. Description and use of the Harris county gravid trap for West
Nile virus surveillance 200306. J. Am. Mosq. Control Assoc. 23 (3), 359362.
Detinova, T.S., 1962. Age grouping methods in Diptera of medical importance with
special reference to some vectors of malaria. Monograph Series. World Health
Organiz. 47, 1216.
Bustamante, Dulce M., Lord, C.C., 2010. Sources of error in the estimation of
mosquito infection rates used to assess risk of arbovirus transmission. Am. J.
Trop. Med. Hyg. 82 (6), 11721184.
Farajollahi, A., Kesavaraju, B., Prince, D.C., Williams, G.M., Healy, S.P., Gaugler, R.,
Nelder, M.P., 2009. Field efcacy of BG-Sentinel and industry-standard trapsfor
Aedes albopictus (Diptera: Culicidae) and West Nile Virus surveillance. J. Med.
Entomol. 46 (4), 919925.
Godsey, M.S., Nasci Jr., R., Savage, H.M., Aspen, S., King, R., Powers, A.M., Burkhalter,
K., Colton, L., Charnetzky, D., Lasater, S., Taylor, V., Palmisano, C.T., 2005. West
Nile Virus-infected Mosquitoes, Louisiana, 2002. Emerg. Infect Dis. 11
(September (9)), 13991404, http://dx.doi.org/10.3201/eid1109.040443.
Gu, W., Unnascha, T.R., Katholib, C.R., Lampmanc, R., Novak, R.J., 2008. Fundamental
issues in mosquito surveillance for arboviral transmission. Trans. R. Soc. Trop.
Med. Hyg. 102, 817822, http://dx.doi.org/10.1016/j.trstmh.2008.03.019.
Hernndez-Triana, L.M., Jeffries, C.L., Manseld, K.L., Carnell, G., Fooks, A.R.,
Johnson, N., 2014. Emergence of West Nile Virus Lineage 2 in Europe: a review
on the introduction and spread of a mosquito-borne disease. Front. Public
Health 2 (December), 271, http://dx.doi.org/10.3389/fpubh.2014.00271,
Published online.
Hoel, A.F., Kline, D.L., Allan, S.A., 2009. Evaluation of six mosquito traps for
collection of Aedes albopictus and associated mosquito species in a suburban
setting in North Central Florida. J. Am. Mosq. Control Assoc. 25 (1), 4757.

O., Sikutov,
S.,
Hublek, Z., Rudolf, I., Bakonyi, T., Kazdov, K., Halouzka, J., Sebesta,
Juricov, Z., Nowotny, N., 2010. Mosquito (Diptera: Culicidae) surveillance for
arboviruses in an area endemic for West Nile (Lineage Rabensburg) and
Tahyiia Viruses in Central Europe. J. Med. Entomol. 47 (3), 466472.
Hugo, L.E., Quick-Miles, S., Kay, B.H., Ryan, P.A., 2008. Evaluations of mosquito age
grading techniques based on morphological changes. J. Med. Entomol. 45,
353369, http://dx.doi.org/10.1603/0022-258545[353:EOMAGT]2.0.CO;2 353369.
Hurk van den, A.F., Hall-Mendelin, S., Townsend, M., Kurucz, N., Edwards, J., Ehlers,
G., Rodwell, C., Moore, F.A., McMahon, J.L., Northill, J.A., Simmons, R.J., Cortis,
G., Melville, L., Whelan, P.I., Ritchie, S.A., 2014. Applications of a sugar-based
surveillance system to track arboviruses in wild mosquito populations. Vec.
Borne Zoon. Dis. 14 (1), 6673.
Irish, S.R., Moore, S.J., Bruce, J., Cameron, M.M., 2012. Effect of the volume of
organic infusion used in gravid traps for collecting Culex quinquefasciatus
(Diptera: Culicidae). J. Med. Entomol. 49 (5), 11181123.
Jackson, B.T., Paulson, S.L., Youngman, R.R., Scheffel, S.L., Hawkins, B., 2005.
Oviposition preferences of Culex restuans and Culex pipiens (Diptera: Culicidae)
for selected infusions in oviposition traps and gravid traps. J. Am. Mosq.
Control Assoc. 21, 360365.
Kesavaraju, B., Kiyoguchi, D., Dickson, S., 2011. Efcacy of gravid traps in trapping
Culex pipiens. J. Am. Mosq. Control Assoc. 27, 320322.
Kweka, E.J., Mwangonde, B.J., Mahande, A.M., 2010. Optimization of odour-baited
resting boxes for sampling malaria vector, Anopheles arabiensis Patton, in arid
and highland areas of Africa. Parasit Vectors 19 (August (3)), 75, http://dx.doi.
org/10.1186/1756-3305-3-75.
Kramer, L.D., Styer, L.M., Gregory, D.E., 2008. A global perspective on the
epidemiology of West Nile Virus. Annu. Rev. Entomol. 53, 6181.
LAmbert, G., Ferr, J.-B., Schaffner, F., Fontenille, D., 2012. Comparison of different
trapping methods for surveillance of mosquito vectors of West Nile virus in
Rhne Delta, France. J. Vector Ecol. 37 (2), 269275.
Maciel de-Freitas, R., Eiras, A.E., de-Oliveira, R.L., 2006. Field evaluation of
effectiveness of the BG-Sentinel, a new trap for capturing adult Aedes aegypti
(Diptera: Culicidae). Mem. Inst. Oswaldo Cruz, Rio de Janeiro. 101 (3), 321325.
McPhatter, L.P., Olsen, C.H., Debboun, M., 2009. Comparison of the attractiveness of
organic Infusions to the standard CDC Gravid mosquito trap. US Army Med.
Dep. J. (JulySeptember), 9196.
Panella, N.A., Rebekah, J.K., Crockett, B., Biggerstaff, J., Komar, N., 2011. The Centers
for Disease Control and Prevention Resting Trap: A novel device for collecting
resting mosquitoes. J. Am. Mosq. Control Assoc. 27 (3), 323325, http://dx.doi.
org/10.2987/09-5900.1.
Pombi, M., Guelbeogo, W.M., Kreppe, K., Calzetta, M., Traor, A., Sanou, A., Ranson,
H., Ferguson, H.M., Sagnon, NF., della Torre, A., 2014. The sticky resting box, a
new tool for studying resting behaviour of Afrotropical malaria vectors.
Parasites Vectors 7, 247, http://dx.doi.org/10.1186/1756-3305-7-247.
Regione Emilia-Romagna, 2015. http://www.zanzaratigreonline.it/Portals/
zanzaratigreonline/Piano%20regionale%20Sorveglianza%20Arbovirosi%202015.
pdf.
Reiter, P., 2010. West Nile virus in Europe: understanding the present to gauge the
future. Euro Surveill. 15 (10), pii=19508. Available online:
http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=19508.
Rezza, G., Nicoletti, L., Angelini, R., Romi, R., Finarelli, A.C., Panning, M., Cordioli, P.,
Fortuna, C., Boros, S., Magurano, F., Silvi, G., Angelini, P., Dottori, M., Ciufolini,

100

A. Pezzin et al. / Acta Tropica 153 (2016) 93100

M.G., Majori, G.C., Cassone, A., 2007. Infection with chikungunya virus in Italy:
an outbreak in a temperate region. Lancet 370 (9602), 18401846.
Roiz, D., Vazquez, A., Seco, M.P., Tenorio, A., Rizzoli, A., 2009. Detection of novel
insect avivirus sequences integrated in Aedes albopictus (Diptera: Culicidae)
in Northern Italy. Virol. J. 6, 93.
Roiz, D., Vzquez, A., Rosso, F., Arnoldi, D., Girardi, M., Cuevas Perez-Pastrana, E.,
Snchez-Seco, M.P., Tenorio, A., Rizzoli, A., 2012a. Detection of a new insect
avivirus and isolation of Aedes avivirus in Northern Italy. Parasites Vectors
5, 223, http://dx.doi.org/10.1186/1756-3305-5-223.

Roiz, D., Roussel, M., Munoz,

J., Ruiz, S., Soriguer, R., Figuerola, J., 2012b. Efcacy of
mosquito traps for collecting potential West Nile mosquito vectors in a natural
Mediterranean wetland. Am. J. Trop. Med. Hyg. 86 (4), 642648.
Romi, R., Pontuale, G., Ciufolini, M.G., Fiorentini, G., Marchi, A., Nicoletti, L., Cocchi,
M., Tamburro, A., 2004. Potential vectors of West Nile Virus following an
equine disease outbreak in Italy. Med. Vet. Entomol. 18 (1), 1419.
Scott, T.W., Wright, S.A., Eldridge, B.F., Brown, D.A., 2001. Cost effectiveness of
three Arbovirus surveillance methods in Northern California. J. Am. Mosq.
Control Assoc. 17 (2), 118123.
Schaffner F., Angel G., Geoffrey B., Hervy J., Rhaiem A., Brunhes J., 2001. The
mosquitoes of Europe. Montpellier: IRD & EID Mditerrane; CD-ROM.
Semenza, J.C., 2015. Prototype early warning systems for vector-borne diseases in
Europe. Int. J. Environ. Res. Public Health 12 (6), 63336351, http://dx.doi.org/
10.3390/ijerph120606333.
Semenza, J.C., Sudre, B., Miniota, J., Rossi, M., Hu, W., Kossowsky, D., Khan, K., 2014.
International dispersal of dengue through air travel: importation risk for

Europe. PLoS Neglec. Trop. Dis 8 (12), e3278, http://dx.doi.org/10.1371/journal.

pntd.0003278.
Service, M.W., 1993. Mosquito ecology: eld-sampling methods, 2nd ed. Elsevier,
London.
Unlu, I., Roy, A.F., Yates, M., Garrett, D., Bell, H., Harden, T., Foil, L.D., 2009.
Evaluation of surveillance methods for detection of West Nile virus activity in
East Baton Rouge Parish. J. Am. Mosq. Control Assoc. 25 (2),
126133.
Vallorani, R., Angelini, P., Bellini, R., Carrieri, M., Crisci, A., Mascali Zeo, S., Messeri,
G., Venturelli, C., 2015. Temperature characterization of different urban
microhabitats of Aedes albopictus (Diptera Culicidae) in Central-Northern
Italy. Environ. Entomol. 111, http://dx.doi.org/10.1093/ee/nvv067.
Vazquez, A., Jimenez-Clavero, M., Franco, L., Donoso-Mantke, O., Sambri, V.,
Niedrig, M., Zeller, H., Tenorio, A., 2011. Usutu viruspotential risk of human
disease in Europe. Euro Surveill. 31, 16.
White, S.L., Ward, M.P., Budke, C.M., Cyr, T., Bueno Jr., R., 2009. A Comparison of
gravid and under-house CO2 -baited CDC light traps for mosquito species of
public health importance in Houston, Texas. J. Med. Ent. 46 (6),
14941497.
Williams, G.M., Gingrich, J.B., 2007. Comparison of light traps, gravid traps, and
resting boxes for West Nile virus surveillance. J. Vector Ecol. 32, 285291.