Controlling and sampling adult sand flies with a fumigant containing permethrin and deltamethrin



The efficacy of a new smoke-generating formulation (fumigant, MidMos Solutions Ltd., GB), containing the active ingredients permethrin and deltamethrin, was evaluated against adult sand flies in an apartment (280 m3), a semi-open large animal shelter (enclosing an area of 300 m2), a closed Bedouin animal tent (104 m3), and a garden (141 m2) enclosed by a stone wall. In each location, four cages with approx. 100 Phlebotomus papatasi were exposed to the fumigant 0.5 m and 2.0 m above ground for 15 and 60 min. Controls were kept in untreated similar rooms and there were two repetitions. In the apartment and the animal tent, a single cartridge caused 100% mortality within 15 min. In the large animal shelter, one fumigant caused mortality of 86% in the lower cages and 75% in the upper cages after 15 min. After 60 min, mortality was 94 and 87%, respectively. With two fumigants, mortality was 98.5 and 91% after 15 min and after 60 min all sand flies were dead. In the garden, one fumigant caused mortality of 93% in the lower cages and 85.5% in the upper cages after 15 min. After 60 min the mortality was 98 and 92%, respectively. With two fumigants, all flies were dead within 15 min.


Phlebotomine sand flies are highly susceptible to insecticides (Alexander and Maroli 2003) and in the past, in some areas, effective sand fly control was achieved as a side effect of anti-malaria campaigns in which the walls of houses had been sprayed with DDT and malathion against mosquitoes (Vioukov 1987, Pandya 1983). DDT was also used, mostly in the Old World, to specifically control sand flies (Hertig 1949). Synthetic pyrethroids were successfully applied against endophilic sand flies in Colombia and Peru (Davies et al. 2000), while in other South American countries with predominantly sylvatic species operations were less successful (Le Pont et al. 1989, Marcondes and Nascimento 1993).

Insecticide-impregnated curtains, screens, bed nets, as well as several meter-wide barrier strips outdoors on the ground, showed varying or little success controlling sand flies (Basimike and Mutinga 1995, Elnaiem et al. 1999, Kroeger et al. 2002, Orshan et al., 2006).

At Tallil Air Base in Iraq, the U.S. military developed an integrated plan to reduce sand fly populations including truck-mounted ultra-low-volume insecticides and residual sprays in and around tents with additional rodent and canine control (Coleman et al. 2006). Additionally, soldiers and personnel were advised to use personal protective measures to prevent sand fly bites. However, little or no impact on sand fly populations was seen after the extensive activities at Tallil Air Base (Coleman et al. 2006).

Vector control programs are limited by the number of adulticides and their formulations available on the market (Xue 2006). Accordingly, evaluations of new formulations and products are of considerable interest. Fumigant insecticide is an old method used for malaria control programs in the early 1960s (Schoof et al. 1963, Smith et al. 1964) and recently, Masuh et al. (2003) tested a smoking formulation (fumigant) containing the insecticide beta-cypermethrin against mosquito vectors of dengue fever. The fumigant of MidMos Solutions is a new smoke-generating formulation containing permethrin and deltamethrin.

The objective of this study was to determine the efficacy of the new product to control and sample sand flies in confined spaces. Both types of experiments were conducted from mid-June to mid-July, 2010 in the absence of livestock and people.


Sand flies for exposure studies were collected overnight with unlighted, CO2-baited CDC traps (CDC trap model 512 and dry ice container, John Hock, Gainesville, FL, U.S.A.) in southern Israel at the Neot Hakikar oasis, where only one sand fly species, Phlebotomus papatasi Scopoli, is present (Müller and Schlein 2004). Collection bags of traps were supplied with water and 5% sugar solution on cotton swabs to ensure maximum survival of the trapped flies. Catches were recovered early in the morning and transferred to the laboratory, where they were kept for the next 36 h under standard insectary conditions (26–28º C and 70–80% relative humidity). The flies were used in the following evening (21:00) for experiments. Approximately 100 sand flies were transferred with an aspirator to cages (cylindrical cartons, 18 cm diameter, 20 cm high, covered on both sides with mesh screens). Four cages were used in each experiment (N = 400 sand flies per treatment group). Each experiment was replicated twice (N = 800 total sand flies exposed to each treatment). Control situations, also replicated twice, used the same number of cages that contained the same number of sand flies (N = 800 sand flies for each control situation).

Statistical analysis used GraphPad Prism 5 Statistical Analysis Software. Numbers of dead male and female sand flies in each treatment group and the controls were compared using the One-Tailed Z-Test to compare proportions. Significance was taken at p < 0.05.

Experimental sites consisted of an apartment, two animal shelters, and an enclosed garden. The selected sites are typical shelters where people and animals can be found and these areas require sand fly control. In the Middle East, gardens enclosed by stone walls are traditional places where families spend long periods of time after sunset and they are accordingly exposed to sand flies.

The apartment with four interconnected rooms (12×7×3.5 m; 280 m3) was furnished. At the time of the experiments, the internal doors were open, while the windows and the external door were closed. One animal shelter was rather large and open (0.6 m high walls with 2 m-high iron mesh fencing on top enclosing an area of 25×12 m; 300 m2) with a sloped metal roof 3.5–4 m above the ground. In this enclosure, about 150 sheep, ten goats, and four donkeys were regularly kept overnight. The second animal shelter was a closed tent construction (12×4×2 m; 104 m3) with several layers of heavy cloth carried by wooden poles; the single entrance was closed by a heavy textile flap. Here, about 50 goats were kept overnight. The garden (15×9 m, 141 m2) was partially enclosed on two sides by buildings, while the rest was enclosed by a stone wall varying in height from 1.8 to 2.5 m. The garden was planted with some ornamental bushes, vines, and olive trees.

In each of the four corners of the four experimental sites, one cage was suspended 0.5 m and another 2.0 m from the ground with bamboo sticks. The same numbers of cages were kept under similar circumstances in untreated control areas. One or two fumigants were placed in the center of the test areas where they were lit, and the space was then exposed to the vapors. The smoke generated by the fumigant lasted for 5–8 min. Surviving sand flies were transferred after 15 min to new laboratory cages and the mortality was checked again after 60 min.

The first trial at the four sites included two replicates with one fumigant cartridge per test area. The second trial was restricted to the large animal shelter and the garden. It also included two replicates but two combined fumigant cartridges were used. For both trials, the same basic procedure was used in the test areas. The cages were collected after 15 min and the mortality was read at the same time.

After the successful control experiments, sand flies and other biting insects were sampled during daytime with a single fumigant in eight caves (varying in size from approximately 15 to 65 m3) and small animal shelters used for different types of livestock. The shelters included three tent constructions, two wooden constructions, and three brick/ stone constructions varying in size from 38 to 92 m3. The ground of the caves and shelters was covered with white cloths and the fumigant was lit on a plate in the center of the sample areas. After the space was exposed to the vapors for 15 min, the cartridge was removed. After the experiments, knocked-down sand flies were identified with the key of Lewis (1982).


One fumigant cartridge, used for 15 min, killed 100% of the sand flies in cages hung 0.5 m from the ground in the apartment (Z = 41.64) and in the animal tent (Z = 36.04), as well as killing 100% of sand flies in cages hung 2.0 m from the ground (Z = 41.35 and Z = 39.06, respectively) compared to the controls hung at the same heights. There was no significant difference between the number of male and female sand flies killed by exposure to a single smoke cartridge for 15 min at both locations regardless the height of the cages (Table 1).

Table 1.  Mortality of sand flies (%) in cages hung 0.5 m and 2.0 m from the floor, after exposure to fumigants for 15 and 60 min, at four sites.
Site# of cartridges and exposure timeHeight of sand fly cage
  1. ND: Not done.

  0.5 m2.0 m
Apartment1 cartridge 15 min100%100%
 1 cartridge 60 minNDND
Animal tent1 cartridge 15 min100%100%
 1 cartridge 60 minNDND
Large animal shelter1 cartridge 15 min86%75%
 1 cartridge 60 min94%87%
 2 cartridge 15 min98%91%
 2 cartridge 60 min100%100%
Enclosed garden1 cartridge 15 min93%85%
 1 cartridge 60 min98%92%
 2 cartridge 15 min100%100%
 2 cartridge 60 minNDND

In the large animal shelter, single fumigant cartridges were tested for 15 min against cages of sand flies 0.5 and 2.0 m from the ground. Compared to the controls, a 15 min exposure killed a significant number of sand flies (86.0 vs 0.0%; Z = 38.23 at 0.5 m and 75.0 vs 0.0%; Z = 36.82 at 2.0 m). When 0.5 m and 2.0 m-distance cages treated for 15 min were compared to each other, significantly more sand flies were killed at 0.5 m from the floor than at 2.0m (86.0 vs 75.0%; Z = 5.98) in this larger building. Exposure to a single cartridge for 60 min killed more sand flies than exposure for 15 min (86.0 vs 94.0%; Z = 4.83 at 0.5 m from the floor and 75.0 vs 87.0%; Z = 5.97 at 2.0 m from the floor). Use of a single cartridge for 60 min killed significantly more sand flies when cages were 0.5 m from the ground than when cages were 2.0 m (94.0 vs 87.0%; Z = 4.12).

After 60 min of exposure, mortality in the large animal shelter was still less than 100.0% at both distances (0.5 m = 94.0% and 2.0 m = 87.0%), therefore, two fumigant cartridges were tried against cages at 0.5 m and 2.0 m from the ground. Two cartridges used against cages 0.5 m from the floor and 2.0 m from the floor (15 min exposure) killed significantly more sand flies than their respective controls (98.0 vs 0.0%; Z = 41.59 at 0.5 m and 91.0 vs 0.0%; Z = 34.18 at 2.0 m) as did two cartridges used against cages 0.5 m and 2.0 m from the floor (60 min exposure, 100% mortality vs 0% at both distances; Z = 40.27 and Z = 37.97, respectively). Two cartridges killed significantly more sand flies in cages 0.5 m from the ground than cages 2.0 m from the ground (15 min exposure; 98.0 vs 91.0%; Z = 6.64;) but mortality was not 100.0%. An exposure of 60 min to two fumigant cartridges resulted in 100% mortality regardless of the height of cages.

In the enclosed garden, a similar experiment was conducted. Single fumigant cartridges were tested for 15 min against cages of sand flies 0.5 and 2.0 m from the ground. Compared to the controls, both durations of exposure killed significantly more sand flies (93.0 vs 0.0%; Z = 37.86 at 0.5 m and 85.0 vs 0.0%; Z = 37.07 at 2.0 m). When 0.5 m and 2.0 m distance cages (15 min exposure) were compared to each other, significantly more sand flies were killed at 0.5 m from the floor than at 2.0 m (93.0 vs 85.0%; Z = 4.31). Exposure for 60 min killed more sand flies than exposure for 15 min (98.0 vs 93.0%; Z = 4.69 at 0.5 m from the floor and 92.0 vs 85.0%; Z = 4.15 at 2.0 m). Use of a single cartridge for 60 min killed significantly more sand flies when cages were 0.5 m from the ground than when cages were 2.0 m (98.0 vs 92.0%; Z = 5.54). Since 60 min of exposure to a single cartridge failed to kill 100% of the sand flies, we exposed the flies for 15 min to two cartridges. This resulted in 100% mortality, while there was no mortality in control cages.

While sampling sand flies in caves and small animal shelters, we had the impression that insects engulfed by the smoke dropped dead almost instantly on the white sheets. After the fumigant was removed and the animal shelters were ventilated, it was possible within minutes to work at the sites without disturbing smoke. In the eight animal shelters all together, 4,537 sand flies were caught. Here the dominant sand fly was Phlebotomus sergenti with about half of the catch (2,150) followed by P. papatasi with about one-quarter of the catch (1,162), P. syriacus with one-fifth (890), and P. tobbi with the rest of the catch (335). Additionally, in the shelters we caught almost 1,000 Sergentomia sp., a little more than 900 Culex sp., about 400 Anopheles sergentii, several dozen Aedes caspius, almost 100 Stomoxys calcitrans, and several hundred cat fleas, Ctenocephalides felis (Bouché).

In the eight caves, a total of 295 sand flies was caught. The dominant sand flies were P. syriacus, with about two-thirds of the catch (206) and P. sergenti, with about one-third (78) of the catch, while only seven P. tobbi and four P. papatasi were caught. Additionally, some Culex sp. (29) and An. claviger (25) were collected.

It is worthwhile noting that MidMos Solutions claims that a single smoke generator is required for 250 m3 in homes and offices while the same amount is sufficient for 500 m3 in agricultural buildings to efficiently control a variety of insects including ants and house flies. At least in our trials, we proved that for the tested sand fly P. papatasi, the claims of covered space are exceeded.

The fumigant trials with P. papatasi, an important nuisance and vector species, were highly effective and may in the future be suitable for adoption to integrated vector management programs, based on local residential household and peridomestic conditions. From economic and environmental views and concerns, one cartridge should be considered sufficient against vector mosquitoes in an area of similar size. Nevertheless, according to the producer, the fumigant has no residual effects and therefore frequent applications are necessary to eliminate sand flies from open spaces.

There are numerous ways to collect night active adult sand flies, and the use of different types of traps and human bait is the common approach (Burkett et al. 2007). Resting populations are traditionally collected with different types of aspirators and exit traps, while the use of pyrethrum spray catches is rather uncommon (Alexander 2000). The reluctance to use the latter technique might be connected to the fact that the carrier solvents often leave some residual effects which make repetitive collections in one location problematic (Service 1993). The lack of residual effects regarding the fumigant can, in this case, be seen as a clear advantage.

After further evaluations and comparison with pyrethrum spray catches, the MidMos fumigant might prove to be an easy and convenient tool for collecting resting plebotomines and other biting flies. Pyrethrum spray catches (PSC) are presently the method of choice to collect resting mosquitoes indoors, but PSC is often laborious and time-consuming. The presently available cartridge is 100 g but can easily be broken/cut into pieces and burned. The results suggest that all kinds of small biting insects were sensitive to the fumigant and, depending on the site, smaller quantities should be sufficient for control and especially sampling. The rapid and complete knock-down of sand flies in confined spaces up to 280 m3 suggested that the fumigant is also an excellent sampling device and an alternative for labor intensive pyrethrum spray catches (WHO 1975).


The authors thank Mr. Nadim Rishmawi, The Hashemite University Zarka, Jordan, for his assistance and the coordination of the control study, and MidMos Solutions Ltd., Brierley Hill, West Midlands, UK, for providing the fumigant samples. This is a report on a research study only and does not imply endorsement of this product by any of the participants.