. .f-1 •( ! (' WORLD HEAL111 ORGANIZATION ORGANISATION MONDIAL£ DE LA SANTE
!'
t.{
I .S f / ,.;.· ( \
WHQ/VBC/83.886 ENGLISH ONLY
LABORATORY AND FIELD TESTING OF INSECTICIDES AGAINST RHODNIUS PROLIXUS (REDUVIIDAE, TRIATOMINAE), THE VECTOR OF CHAGAS' DISEASE IN VENEZUELA1 by M. J. Nelson,2 J. 0. Williams,1 F, Nocerino,3 A. A. Arata4 and R. J. Tonn5
ABSTRACT From 1976 until 1980 the Pan American Health Organization Research and Reference Center for Vector Biology and Control collaborated with the Direccion de Malariologia to carry out laboratory and field trials with several formulations of 14 insecticides against Rhodnius prolixus (Reduviidae, Triatominae), the most important vector of Chagas' disease in Venezuela. All insecticides except organochlorines were effective against dieldrin-resistant R. prolixus. Fenitrothion gave the longest residual effectiveness of all.insec:icides evaluated when applied at 2 g/m2 in the field, and it was competitively pr1ced w1th dieldrin, the standard in Venezuela. Pirimiphos-methyl and jodfenphos both gave longterm control and have a low mammalian toxicity (see Table 1); bendiocarb gave also longterm control but is moderately toxic to mammals. Bromophos has low mammalian toxicity but it had little residual activity and fenitrothion and dieldrin gave adequate control but have a much higher vertebrate toxicity. The relative merits of wall bioassays, house searches, experimental hut tests, flushing agents and Gomez-Nunez box traps for measuring insecticide efficacy against R. prolixus were evaluated and are discussed. 1• INTRODUCTION
Control measures against Chagas' disease in Venezuela have been directed chiefly against the principal vector, Rhodnius prolixus, by means of intradomiciliary spraying of residual insecticides. Dieldrin is sprayed in houses at a dosage of 1 g/m2 and HCH at 0.8 g/m2 in animal shelters and other annexes (to avoid the toxic effect on livestock). Resistance of R. prolixus to dieldrin was first found in 1969 in the state of Trujillo (Gonzalez-Valdivieso et al., 1971; Nocerino, 1976); hence the need for alternative control materials or measures arose. Field trials were carried out with propoxur water-dispersible powders (wdp) (Gonzalez-Valdivieso & Sanchez Diaz, 1968; Nelson & Colmenares, 1979a; Pan American Health Organization Research and Reference Center for Vector Biology and Control, Apartado 2171, Maracay, Venezuela. 2
1
PAHO Aedes aegypti Ecology Research Project, Apartado A~reo 29668,, ·Bogota, Colombia.
Servicio de Evaluacion Biologica de Plaguicidas, Direccion de Malariologia, Maracay, Venezuela. 4 PAHO Centro Panamericano de Ecologia y Salud Humana, Apartado Postal 249, Toluca, Mexico. 5 HPD/HPT Pan American Health Organization, 525 23rd Street, N.W., Washington, D.C. 20037, United States of America. The issue of this document does not constitute formal publication. lt should not be reviewed, abstracted or quoted without the agreement of the World Health Organization. Authors alone are responsible for views expressed in signed articles. Ce document ne constitue pas une publication. 11 ne doit faire l'objet d'aucun compte rendu ou resume ni d'aucune citation sans l'autorisation de !'Organisation mondiale de la Sante. Les opinions exprimees dans les articles signes n'engagent que leurs auteurs.
3
WHO/VBC/83.886 page 2 Nocerino et al., 1975), dieldrin, HCH, fenthion (Nocerino et al., 1975), jodfenphos (Nocerino et al., 1976), fenitrothion (Williams et al., 1980a), and DDT (Williams et al., 1980b), and with emulsifiable concentrate formulations of propoxur, fenthion, malathion and pirimiphos-methyl applied by means of a backpack mistblower (Nocerino et al., 1976). Also, compression pumps, mistblowers and thermal foggers were compared for efficacy and control of R. prolixus; fenitrothion was used in this trial (Williams et al., 1980a). From 1976 until 1980 the Pan American Health Organization Research and Reference Center for Vector Biology and Control (PAHO/RRCVBC) collaborated with the Direccion de Malariologia of Venezuela in further insecticide trials in the laboratory and the field which are summarized in this report. The evaluation began with determination of the lethal dosage of the technical product by topical application toR. prolixus (Nelson & Colmenares, 1979b). Residual activity of the insecticides formulated, either as water-dispersible powders, emulsifiable concentrates or microencapsulated formulations, was evaluated first in the laboratory on insecticide-treated surfaces, typical of materials used in local houses. Further evaluation consisted of experimental hut trials, wall bioassays in rural houses, small field trials, and, finally, villagescale trials. Not all insecticides were tested at all stages and the tests were not always carried out in the above order, but they are described for a better understanding of the results. Although the trials are grouped together in this paper, they are not all strictly comparable with each other, as they were carried out at different times during the five-year project and by different investigators. In addition, some trials were terminated prematurely because of labour problems with field personnel, but their inclusion is considered worthwhile for comparison with the other trials. 2. 2.1 MATERIALS AND METHODS Laboratory tests
The required weight of formulated insecticide plus 10% for wastage was mixed with tap water and sprayed over 6-8 glass plates (25 cm x 25 cm) with an ordinary household hand operated compression sprayer. Several applications were made allowing the insecticide to dry after each, in order to ensure an even distribution of the deposit on the plate. The plates were held in an upright position in slotted wooden supports. Two colonized strains of R. prolixus were used for the bioassays~ (a) Cojedes dieldrinsusceptible strain, brought for the first time in 1976 from various villages in the state of Cojedes and (b) Santo Domingo dieldrin-resistant strain colonized in 1969, in Santo Domingo, State of Trujillo. For fixed-time bioassays, 10 blood-fed fifth-instar Cojedes strain nymphs were placed inside glass rings1 (5 cm high x 8 cm in diameter), on the insecticide-treated surface. The smooth surface of the walls of the rings prevented escape of the insects, and the open top prevented accumulation of insecticide vapours. Insects were exposed for 2 h and then held for 48 h in clean petri dishes for observation of mortality. Each position on the glass plate was used only once. Each replicate test was carried out one week after treatment and then every month until less than 70% of the insects died. For each compound, two to three replicates of fixed-time bioassays were carried out and for each compound the replicate series were begun on a different month. Continuous-exposure tests were carried out using the same kind of glass rings. One ring containinglOCojedes bugs and two rings with Santo Domingo bugs (both for replicate tests), were evaluated one week after each spraying of the glass plates. Hourly observations were made during the first eight hours of exposure and then daily until all insects were dead. Fumigant action tests were carried out by placing insects on netting placed over a glass ring separating the insecticide-treated plate. Another glass ring covered with a petri dish was placed on the netting to confine the insects. The vapours were trapped in the enclosed space and hourly observations were made as for the continuous-exposure tests. 1
Cut from 1-pint mayonnaise jars (1 pint (US)
0.47 1).
WHO/VBC/83. 886
page 3 Tests on surfaces other than glass were carried out by confinement of insects on the insecticide-treated surface under a petri dish, to simulate the technique of field bioassays. Exposure lasted 24 h, followed by 120 h (5 days) of observation for mortality in clean petri dishes. Mud and cement "bricks", 20 cm on a side and 2 cm thick were prepared in wood moulds, and both kinds of bricks were painted with either white-wash, a water-base paint or a rubber-base paint. Other surfaces such as tin, glass, wood and cardboard, were prepared as 20 cm x 20 cm square "sheets" and the insecticide was applied by means of a motorized compressor sprayer. 2.2 Field trials
Evaluation was carried out in several states of Venezuela with differing habitats, which are described separately for each trial. Treated houses were nearly all constructed of mud (bahareque) walls and corrugated metal roofs, the most common combination of construction materials now found in rural Venezuela. Before the government house-improvement programme, most rural dwellings had palm-thatch roofs, a very favourable habitat for R. prolixus. Houses were sprayed by staff of the PAHO Center, in collaboration with personnel from the appropriate Malariology Zone office; Hudson X-pert hand compression sprayers were used. Spraymen wore long-sleeved shirts, long trousers, metal helmets, rubber gloves and respirators. Prior to treatment, all furniture, utensils and food were removed from each house and the inhabitants were instructed not to enter houses for one hour after spraying, not to sweep floors and burn dead insects found, as well as not to allow pets, poultry or livestock to enter the houses. House infestation was monitored by Center and Zone crews. Detailed maps were prepared, every house was numbered, and for each evaluation every house was searched, starting with the bedroom, for one "man-hour" (one man for one hour or two men for one half-hour each). No "flushing" agent was used. Captured bugs were identified as to species, blood engorgement and sex and then released back into the house. For bioassay tests, 10 blood-fed fifth-instar Santo Domingo nymphs, resistant to dieldrin, were placed on treated mud walls and confined under a petri dish attached to the surface by tape. After 24 h, the bugs were removed from the surface, transported to the Maracay laboratory, and daily mortality was observed for five days. Usually two replicates per treated surface were carried out. The bioassay mortality was plotted against elapsed time after treatment on log-probit paper, a regression line was fitted by eye, and the time after treatment for survival to increase to 30% was read from the curve. Many workers, especially field operators, use the trade names of insecticides rather than the common names adopted by ISO. 1 In Table 1, 14 insecticides, described in this paper~ are listed and include common name, OMS number, trade name, formulation, lethal dosage to R. prolixus, and mammalian toxicity.
1
ISO
International Organization for Standardization.
WHo/VBr/83.886 page 4 3. 3.1 RESULTS Laboratory tests (Table 2)
Bendiocarb, at 1 gju{l. gave the longest residual action on glass (94 weeks) , but at 0.4 g/m 2 its residual activity decreased to 22 weeks and at 0.2 g/m2 to 9.5 weeks. Cypermethrin gave a residual activity for 93.5 weeks. Jodfenphos and fenthion were residually active for over a year (83.0 and 75.5 weeks respectively). Pirimiphos-methyl micorencapsulated formulation was active for 55.5 weeks, but the emulsifiable concentrate lasted only 3.0 weeks and the water-dispersible powder formulation only 2.5 weeks. The Sumithion brand of fenitrothion had similar residual activity (33.5 weeks) to the standard, dieldrin (31.5 weeks), but the Agrothion brand of fenitrothion lasted half as long (16.5 weeks). Pirimiphos-ethyl (18.0 weeks), bromophos (8.0 weeks), propoxur (6.0 weeks), HCH (2.5 weeks) and DDT (<1 week) had very short residual activity on glass. DDT never produced 70% mortality among insects exposed for two hours, even when freshly sprayed. There were notable differences between the duration of residual activity of the insecticides and the time it took for an insecticide to knock down 50% of the bugs by continuous exposure. The compound with the most rapid action was propoxur (1 h) which was among those insecticides with the shortest residual activity. Cypermethrin and bendiocarb also caused fast knockdown (1.5 and 2.0 h). Of the remaining compounds, the only insecticides with greater than 5-h 50% knockdown activity were dieldrin (16h), jodfenphos (24h), chlorphoxim (24h) and DDT (36h). The knockdown times for the dieldrin-resistant Santo Domingo strain were very similar to that of the susceptible Cojedes strain except, as expected, when exposed to the three organochlorines tested. Compared to the dieldrin-susceptible Cojedes strain, the resistant strain had a knockdown time 3.75 times longer after exposure to dieldrin, 5.8 times longer after exposure to HCH, and 1.3 times longer after exposure to DDT. Fumigant action varied greatly among the different compounds. HCH had the fastest toxic vapour activity, followed by pirimiphos-methyl and propoxur. The other insecticides had moderate fumigant action except for jodfenphos which had very low activity and cypermethrin and DDT which had no detectable vapour effect against R. prolixus. Residual activity of insecticides applied to other surfaces is shown in Table 3. Dieldrin, fenthion and deoxycarb ranked among the top three in residual activity for the unpainted surfaces of both mud and cement. Bromophos, jodfenphos, fenitrothion, chlorphoxim and HCH were active for a shorter time and propoxur, bendiocarb and DDT were the least active compounds. As anticipated, all insecticides gave longer residual activity on glass than any other surface. Insecticide-treated tin, wood and cardboard gave longer residual activity than cement or mud; cement painted with rubber-base paint had longer residual activity than cement or mud with or without whitewash or water-base paint. Both insecticide-treated cement and mud covered with water-base paint were better than either surface with whitewash. On whitewashed surfaces the residual activity of the insecticides was sometimes longer, sometimes shorter than on non-whitewashed surfaces; this did not depend on the kind of surface or class of insecticide (organophosphate or carbamate). Insecticide sprayed on cement surfaces usually produced longer activity than on mud surfaces. Water-dispersible powder lasted longer than emulsifiable concentrate (bromophOSJ and longer than the ULV formulation (fenitrothion). 3.2 Experimental hut trials
To fill the gap between laboratory testing of insecticides and field trials in villagers' houses, experimental huts were treated. Each week a colony of insects was introduced into the huts to simulate natural reinfestation. Four huts (2 m x 1.6 m x 2 m high) were constructed of ''bahareque" walls and galvanized corrugated iron roofs in the hacienda of El Charcote in Cojedes state. Sixty unfed
WHo/vBc/83 .886 page 5 fifth-instar R. prolixusnymphs were released each week for 9 weeks before treatment and for 14 weeks after treatment. Before each introduction of nymphs, each hut was searched for 30 minutes for presence of bugs from the previous releases. After the second release, the irritant Pironyl(R) (2.95% pyrethrins, 23.5% piperonyl butoxide) mixed 4:1 with the emulsifier Atlox(R) was applied in a 2.5% solution in water at 200 cc per hut with a household handoperated gaseous energy sprayer. Resultant concentration of pyrethrins was 0.015%. No blood source was provided for the nymphs. As can be seen in Table 4, deltamethrin, applied at the low dosage of 0.05 g/m2, gave lOO% control during the 14 weeks of the trial. While spraying, the irritant (flushing) effect of deltamethrin was observed. Pirimiphos-methyl, at 1 g1m2, maintained reasonable control during the same period, but its efficacy decreased towards the end of the spray period. Bendiocarb, at 0.4g/m2 lost its activity after two weeks. First reinfestation in both the "pirimiphos-methyl hut" and the "bendiocarb hut" was immediate - the week after application. In addition to the 60 nymphs introduced into each house, another group of 60 nymphs were left in a jar in the untreated hut, each week, to determine if there was any mortality due to the three-hour trip from the insectary to the field. Also, to determine mortality caused by the application of pyrethroids, all nymphs collected weekly from the control hut were placed in a jar next to the control1 jar. The first week after transport to the field, mortality among nymphs not exposed to pyrethrin was 4% of 958 and 7% of the 921 remaining during the second week. Of 388 nymphs collected in the control hut with pyrethrins, 35% died during the following week. When corrected by the control mortality of the second week, the net mortality due to pyrethrins was 30%. The pre-treatment recapture of nymphs without the utilization of pyrethrins was only 4% of 180 released in three huts; pre-treatment recapture with the use of pyrethrins in the same three buts was 6.5 times higher (26% of 1620 nymphs released). 3.2.1 Wall bioassay
After some field treatments, wall bioassays were carried out but the houses were not searched for triatomes. As can be seen in Table 5, dieldrin and fenthion gave good residual activity, whereas propoxur and pirimiphos-methyl had short residual lives. 3.3 Field trials in village houses Two-house trials in Palmaritas
3.3.1
Palmaritas, an isolated village in the municipality of Calderas, state of Barinas, at an altitude of 900m,with a high rainfall, was treated with five insecticides on 12-13 August 1978, two houses being treated with 1 g/m2 of a water-dispersible powder formulation of one of the five compounds. Table 6 shows that reinfestation occurred from 34 to 38 weeks after spraying in all of the treated houses. Bioassay mortality dropped to 70% at 7 weeks for bromophos and 26 weeks for fenitrothion. Mortality was still over 70% for fenthion, DDT and jodfenphos at 30 weeks when bioassays were discontinued. 3.3.2 Village-scale trials with five insecticides in Cojedes
Seven villages in northern Cojedes state, where virtually no insecticide spraying had been carried out since 1973, were selected on the basis of similar size, altitude and construction of houses. In November 1976, 153 houses were sprayed, 106 with the evaluation insecticides (14-36 houses per insecticide) and the rest with HCH to form a barrier zone. All insecticides 1 Jar containing 60 nymphs for observation of mortality attributable to transport from insectary.
WHO/VBc/83 .886 page 6 were sprayed at 2 g/m2 except dieldrin and HCH which were sprayed at 1 g/nll. to avoid deaths of domestic birds and animals. One village, Valle del Rio, was left unsprayed to serve as control. As can be seen in Table 7, there was no reinfestation of any village until 20 weeks after treatment when the village sprayed with bromophos became positive. The village treated with fenitrothion remained negative for 22 weeks and dieldrin maintained complete control for 26 weeks. When the trial was terminated at 26 weeks because of personnel problems, both jodfenphos and fenthion still maintained lOO% control. In bioassay tests jodfenphos and fenitrothion maintained 70% mortality the longest (7 weeks) and bromophos the shortest (2 weeks). No adverse effects among spraymen were observed. One to two weeks after treatment, householders of each treated village were interviewed as to illness in the family and deaths among pets or livestock. In the villages sprayed with fenitrothion, bromophos and jodfenphos, there were no reports of any cases of illness among the inhabitants or deaths of animals. In the barrier zone of Palambra del Doctor where 12 houses were sprayed with HCH, there was some mortality among young chickens in two houses. Where fenthion was sprayed, inhabitants of three houses reported headache and dead chickens. In Rio Claro and Aminta Suarez where dieldrin was applied, there was a very high mortality of chickens; there ~ere also dead parrots, dogs and cats (Table 8). Virtually all of the dead animals and birds were new-born or very young. The people of the two villages reported no unusual sickness and, in general, were happy to be free of triatomes, crickets and cockroaches in spite of the dead animals. 3.3.3 Trials with fenitrothion in three ecologically diverse villages
Field trials with fenitrothion were carried ou in the states of Barinas, Tachira and Trujillo in three localities with differences in altitude, rainfall, temperature, agricultural practices and resistance to dieldrin (Table 9). All inhabited houses were sprayed with fenitrothion 40% wdp at 1 g/m2; all abandoned houses and outbuildings were sprayed with HCH at 0.8 gjm2. Table 9 shows that recovery of R.prolixus after treatment was similar in the three villages. First reinfestation occurred from 8 to 13 weeks after spraying and 30% recovery from 23 to 30 weeks. Surprisingly, the most rapid recovery occurred in the coolest village (San Antonio), with the lowest pre-treatment mean, and the slowest recovery was in the warmest, driest village (Sabana Potrera) with the highest pre-treatment mean. In addition to the standard one man-hour searches in each house in Rio Morosmoy, two Gomez-Nunez traps were placed in each house from the third cycle (8 weeks) until the sixth cycle (19 weeks) of evaluation. These are small, perforated cardboard boxes that are attractive to triatomes as shelters (Gomez-Nunez, 1965). Positivity was defined as the presence of either R. prolixus adults, nymphs, eggs, exuviae or faecal droppings. The above method was discontinued after cycle 6 because of considerable pilferage and destruction by the inhabitants. As can be seen in Table 10, there was more positivity in the Gomez-Nunez traps than that found by one man-hour house searches. Nearly all houses infested with triatomes also had positive Gomez-Nunez traps, but several houses had positive traps without evidence of other signs of natural infestation. 3.3.4 A trial with bendiocarb in La Guaca
On 7 August 1979, the village of La Guaca, municipality of Pan-Pan, state of Trujillo, a humid locality with thick vegetation and many palm trees adjacent to the main highway, was treated with bendiocarb 20% wdp at 0~4g/m 2 . The pre-treatment mean in the 21 treated houses was 14.3 bugs per man-hour. After treatment indices remained nil until 13 weeks when the first bug was found. When the evaluation was discontinued at 42 weeks, 4 of the 13 houses still being evaluated were positive. R. prolixus was resistant to dieldrin in this village.
WHO/VBC/83.886 page 7 3.3.5 Trials with pirimiphos-methyl and fenitrothion in Trujillo
Another trial with fenitrothion was begun in Rio Morosmoy during June 1980 to compare the efficacy of 2 g/m2 with that observed with the 1 g/m2 applied previously in the same village. Simultaneously, a trial with the promising 20% microencapsulated formulation of pirimiphos-methyl at 1 g/m2 was carried out in the nearby village of Bucaral. In Rio Morosmoy, 43 houses were sprayed, but there was sufficient pirimiphos-roethyl for only 18 houses in Bucaral. Five were left as controls and 28 others formed a barrier treated with propoxur at 2 g/m2. Evaluations were carried out in 12 houses in each village. Although bioassay mortality on the mud walls sprayed with pirimiphos-methyl had dropped to 70% within four weeks, the first reinfestation did not occur until 34 weeks and reinfestation was only 17% at 53 weeks when the project was discontinued. In the village sprayed with fenitrothion, bioassay mortality remained above 70% for much longer - 15 weeks. After 59 weeks, when still no reinfestation had been detected in the 13 evaluation houses, a thorough search was made in 33 houses, of which only one was found infested. Table 11 suffiillarizes all of the 16 field trials with eight insecticides. Of the six tests carried out with fenitrothion under various ecological conditions and with two dosages, the compound applied at 1 g/m2 gave 5-7 months' control (<30% reinfestation) and at 2 g/m2 showed a residual activity for more than a year. Pirimiphos-methyl, applied at 2 g/m2, gave control for 53 weeks, which was surprising considering its poor results in wall bioassays and in experimental huts (see discussion below on reinfestation vs bioassay results). Bendiocarb, at the recommended dosage of 0.4 g/m2, maintained control for 43 weeks (another case of good field results but bad control in experimental huts). Recovery of bugs after bromophos spraying was faster than after any other insecticide tested. The field trials were incomplete, but showed that jodfenphos, fenthion and dieldrin can maintain control for over six months, although bioassay tests with these compounds indicated a shorter residual life. 4. 4.1 DISCUSSION Comparison of insecticides
Choice of an insecticide for the control of Rhodnius prolixus is a complex process depending not only on the residual effectiveness of the candidate compound but also on its mammalian toxicity 1 and cost. The cost depends on the price per kg of formulated wdp or emulsifiable concentrate (EC), the percentage concentration of the formulation, the grams of active ingredient applied per m2 and the frequency of treatments necessary to maintain control. An inexpensive insecticide of short residual action, requiring frequent treatments, may cost more to use than an expensive compound of longer duration, especially in a country where the cost of labour is high. More frequent usage also involves more expense in equipment maintenance and depreciation. The cost of formulated insecticide is quite variable, depending on the manufacturer or formulator, the quantity purchased, the availability, the shipping costs and the country involved. Therefore, the comments on cost below are of a relative nature only. Of all insecticides tested, fenitrothion gave the longest residual efficacy at 2 g/m produced an acceptable effect at 1 g/m2 • It was highly toxic to R. prolixus and competitively priced with dieldrin, although it is moderately toxic to vertebrates.
2
and
1 For data on mammalian toxicity refer to Table 1; see Table 8.
for toxicity to other vertebrates
WHO/ VBc/ 83.886 page 8 Fenthion is highly toxic to R. prolixus and showed good residual activity at 1 and 2 g/m2; however, it caused some deaths among chickens and thus is not recommended for residual spraying. Although its toxicity to R. prolixus is low , pirimiphos-methyl maintained long-term control when applied as a microencapsulated formulation at 1 g/m2 ; it has a very low mammalian toxicity. Jodfenphos has also a very low mammalian toxicity; it maintained long-term control at 2 g/m2 but was the most expensive insecticide to use at that dosage. Bromophos is another insecticide with low mammalian toxicity and is not very toxic to R. prolixus - even at 2 g/m2 it gave the shortest period of control of all the compounds tested. Dieldrin, the standard in Venezuela, is inexpensive and gives acceptable control at 1 g/m2, but it is unsuitable because of its very high vertebrate toxicity. Bendiocarb controls triatomes for an adequate period of time even when applied at the low dosage of 0.4 g/m2; it is also classified as being only moderately toxic to vertebrates. Deltamethrin gave excellent control in experimental huts at the low dosage of 0.05 gjm2. This compound should be tested further and the effect of its marked knockdown and repellent properties should be determined. The long-lasting activity with DDT in Palmaritas is notable. Although this insecticide has very low toxicity to triatominae (Nelson & Colmenares, 1969), continued exposure of insects to it in the laboratory does cause mortality, and it has given at least partial control in the field (Williams et al. 1980b). All of the insecticides except the organochlorines were equally effective against both the dieldrin-resistant Santo Domingo strain and the dieldrin-susceptible Cojedes strai~ and could be effective in areas where resistance to dieldrin occurs (as demonstrated in the state of Trujillo with fenitrothion, pirimiphos-methyl and bendiocarb). 4.2 Techniques
Although glass is not a typical surface where the bugs rest in a house, it is an ideal surface for the first laboratory tests of residual activity of formulated insecticides - it is smooth, non-porous and inert (e.g., it does not react with the insecticides); thus the above variables are excluded. Also, it can be washed and used repeatedly. The other, more representative, surfaces are difficult to standardize. Here enter such variables as pH, porosity, roughness and organic content, which are difficult to control and which were not quantified in our study. For future tests on glass, either exposure time or dosage should be reduced so that residual activity would be more comparable to that obtained in the field. In the case of some insecticides evaluated in the present study (2-h exposure of insects to glass treated with 1 gjm2), it was necessary to continue observations for up to two years after spraying. 4.2.1 Bioassay vs house "searches"
For evaluation of residual effect of an insecticide in the field, bioassays and house searches should not be the only methods used. Bioassays of sprayed wall surfaces indicate when the insecticide is losing its residual effect but do not show whether it is still controlling the triatome population. The exposure time, 24 h, is arbitrary - it is also not known whether under natural conditions triatomes are exposed continuously for 24 h
WHO/VBC/83.886 page 9 or intermittently (i.e., and thus affected by cumulative exposure to a compound), It is also not known if the surfaces selected for bioassays are representative of surfaces where triatomes come in contact with insecticides under natural conditions. Also, bioassays are only carried out with fifth-instar nymphs. Had first-instar nymphs been exposed, fewer would have survived. The greatest problem with bioassays is the great variation in percentage mortality, even on the same surface side by side, on the same day. This may be due to uneven distribution of insecticide during spraying, unknown differences in the surface, or differences among the insects. At best, bioassay tests give only an approximate index for comparison of relative duration of residual activity of different insecticides on different surfaces. House searches indicate when demonstrable reinfestation occurs, which is of primary interest. However, the time at which reinfestation occurs may not be directly related to duration of residual activity of the insecticide. Usually, reinfestation does not occur until long after residual activity (as determined by bioassays) has dropped to nil. For example, in the Cojedes trials (Table 7) houses were not reinfested until 15 to 22 weeks after bioassay mortalities had dropped below 70%. Reinfestation depends on the amount of movement of insects either from their sylvan habitat to houses or from house to house or village to village. This movement may be either active (adults or nymphs attracted to light) or passive (eggs, nymphs and adults being carried in boxes of goods from one village to another). Rate of reinfestation appears to depend on the characteristics of the village, its surroundings and probably season. In Palmaritas, for instance, although the residual effect varied for each of the five insecticides used, all the sets of two houses sprayed with the same insecticide became reinfested at approximately the same time (Table 6). In another trial (Williams et al., 1980a), it was shown that insecticides applied by thermal fog and mistblower gave control for almost as long as water-dispersible powder formulations although, presumably, application by use of the first two methods gave low residual activity. Unfortunately, for insecticide field trials, it is still not possible to quantify the degree of "immigration pressure" for each village in order to select strictly comparable villages for the trials. The best one can do is select villages that are similar with respect to size, altitude, proximity to palm trees, and triatome density, and carry out the trials simultaneously in the different villages or during the same season during different years. Ideally, several insecticides should be tested simultaneously on different houses in the same village. However, this involves .very careful supervision of the evaluation team to assure that houses treated with one insecticide are not confused with houses treated with another. This level of supervision was not possible in these trials. The utility of experimental huts was well demonstrated. Just as huts used in. anopheline research, they show the residual effect of insecticides more realistically than laboratory tests, without many of the variables found in natural houses. The weekly "challenge" with introduced triatomes is better than both bioassays and natural infestation because the former is an exposure of arbitrary duration and the latter depends on immigration pressure, as discussed above. In experimental huts the exposure depends on the normal activities of the liberated insects, and the "immigration" (introduction) is held constant. As it is often difficult to find houses with high natural infestation, it may be necessary to use experimental huts more often in the future. After the use of flushing agents, such as pyrethrins or pyrethroids, more triatomes may be found during house searches, but as seen in the experimental hut trials, these agents can also cause subsequent mortality. This is unimportant for one-time surveys, but if the population is being monitored over time, fewer bugs may be captured on each collection because of the effect of the flushing agent. The Gomez-Nunez box traps showed great promise for monitoring low density R. prolixus populations. They would be less useful for one-time surveys because two visits must be made to each house for a trap sample, once to place the trap and a second visit to retrieve it. In rural areas where houses are greatly dispersed, traps would therefore double the time nece_ssary for the survey.
WHO/VBq/83.886 page 10 ACKNOWLEDGEMENTS We gratefully acknowledge the arduous efforts of the personnel of the Malariology Zone offices in the various areas of Venezuela where the field trials were carried out, especially the zones of Cojedes, Trujillo, Tachira, Barinas and Portuguesa; the technicians of the Servicio de la Evaluacion Biologica de Pesticidas in Maracay who did the laboratory work with the various substrates and several of the field bioassays; the national staff of the PAHO/RRCVBC in Maracay that maintained the insect colonies; Mr Pablo Colmenares of Malariology who did all of the laboratory trials on glass; and Dr Helio Espinola of the RRCVBC who designed and built the experimental huts. We would also like to thank the following chemical companies for providing insecticide samples for the studies: Bayer (Baygon; Baytex; Baythion); Celamerk (Nexion); Ciba-Geigy (Nuvanol); FBC (formerly Fisons Co.) (Ficam); Imperial Chemicals (Actellic; Agrothion; Cymbush; Primicid); Roussel Uclaf (K-Othrin); and Sumitomo (Sumithion). REFERENCES Gomez-Nunez, J. C. (1965) Desarollo de un nuevo metodo para evaluar la reinfestacion intradomiciliaria por Rhodnius prolixus, Acta Cient. Venez., 16, 26-31 Gonzales-Valdivieso, F. E. & Sanchez Diaz, B. (1968) Ensayo de campo de la accion del insecticida OMS-33 (Bayer 39007, Baygon) sobre R. prolixus, Bol. Informative Direccion Malariologia y Saneamiento Ambiental, Venezuela, ~(6), 358-364 Gonzales-Valdivieso, F. E. & Nocerino, F. (1971) Susceptibility of R. prolixus to chlorinated hydrocarbon insecticides in Venezuela, WHO unpublished document WHO/VBC/71.264 Nelson, M. J. & Colmenares, P. (1979a) Insecticide susceptibility of vectors of Chagas' disease in Venezuela, WHO unpublished document WHO/VBC/79.736 Nelson, M. J. & Colmenares, P. (1979b) Topical application of insecticides to Rhodnius prolixus (Reduviidae: Triatominae), a Chagas' disease vector, WHO unpublished document WHO/VBC/79.737 Nocerino, F. (1976) Susceptibilidad de R. prolixus y T. maculata a los insecticidas en Venezuela, Bol. Informative Direccion Malariolo ia Saneamiento Ambiental Venezuela, J&(3), 276-283 (WHO unpublished document WHO VBC 75.565 in English) Nocerino, F., Rodriguez, M., Sanchez Diaz, B., Otero, M. A. & Tonn, R. J. (1975) Pequeno ensayo de campo empleando dieldrin, HCH, fenthion y OMS-33 para el control de R. prolixus, Bol. Informative Direccion Malariolo ia Saneamiento Ambiental Venezuela, 12(3-4), 98-109 (WHO unpublished document WHO VBC 75.527 in English) Nocerino, F., Valenzuela, J. V., Otero, M. A. & Tonn, R. J. (1976) Field trials using fenthion, propoxur, malathion, pirimiphos-methyl and jodfenphos for the control of Rhodni~s prolixus in Venezuela, WHO unpublished document WHO/VBC/76.606 Williams, J. 0., Tonn, R. J., Sanchez Diaz, B., Ortega, R. & Castillo, C. (1980a) Comparacion de tres diferentes tipos de equipos para applicar insecticidas en el control de los vectores de la enfermedad de Chagas con OMS-0043, Bol. Informative Direccion Malariologia y Saneamiento Ambiental, Venezuela, 19(2), 66-72 Williams, J. 0., Tonn, R. J., Sanchez Diaz, B., Ortega, R. & Castillo, C. (1980b) DDT como un insecticida en el control de Rhodnius prolixus: una prueba de campo, Bol. Informative Direccion Malariologia y Saneamiento Ambiental, Venezuela, ~(1), 40-44
WHO/VBC/83 .886 page 11
TABLE 1.
BACKGROUND INFORMATION ON 14 INSECTICIDES EVALUATED Lethal dosage to b R. 12ro lixus-
Insecticide
OMS No.
Trade name
Formulatio~
Maramaliag toxicity-
ORGANOCHLORINE 1. dieldrin 2. DDT 3. HCH ORGANOPHOSPHORUS 1. bromophos 2. chlorphoxim 3. fenitrothion 4. 5. 6. 7. fenthion jodfenphos pirimiphos-ethyl pirimiphos-methyl 0658 1197 0043 0002 1211 1424 Ne xi on Baythion Sumithion Agrothion Baytex Nuvanol Primicid Actellic wdp40 wdp50 wdp40 EC50 wdp40 wdp50 EC50 wdp25 EC50 EN20 1.3 o. 71 0.30 0.17 0,86 0.61 1600 500 503 330 2100 140 2018 0018 0017 wdp50 wdp75 wdp25 1.8 0.82 10 113 lOO D od lf
eft D 0 D 0 0 0
CARBAMATE 1. bendiocarb 2. propoxur PYRETHROID 1. cypermethrin 2. deltamethrin 2002 1998 Cymbush K-Othrin EC25 wdp5 1.4 0.12 >4000 >4000 1394 0033 Ficam Baygon wdp80 wdp20 wdp50 3.4 1.5 55 95 0 0
et
~
~ wdp = water-dispersible powder; EC = emulsifiable concentrate; encapsulated formulation; number = % concentration. - LD5o in pg per insect (from Nelson & Colmenares, 1979b) • b
EN =,·micro-
.£ LD5o of technical product in mg/kg to rats unless otherwise indicated; D = dermal; 0 = oral (from "Guidelines to the use of the WHO recommended classification of pesticides by hazard", WHO unpublished documen~ VBc/78.1 Rev.3).
- LD50 varies according to mixture of isomers.
d
~ Value within a wider than usual range. - LD5o depends on the cis-trans ratio and may be lower than quoted; figure quoted is for oral administration in aqueous suspension (much lower values obtained if oily carriers are used). f
WHO/VBC/83. 886 page 12
TABLE 2. LABORATORY BIOASSAY OF FORMULATED INSECTICIDES ON GLASS: 2-h CONTACT, CONTINUOUS CONTACT, AND CONTINUOUS EXPOSURE TO VAPOURS WITH COJEDES (COJ) AND SANTO DOMINGO (SD) STRAINS OF R. PROLIXUS, AT 1 gjm2
Exposure COIIDD.on name Trade a nameContinuous Contac~
Formulation
2-h contatt COJ-
c VapourCOJ
COJ Bendiocarb 0.2 gjm2 0,4 gjm2 1.0 gjm2 Cypermethrin Jodfenphos Fenthion Pirimiphos-methyl wdp EC EN Chlorphoxim Fenitrothion Dieldrin Pirimiphos-ethyl Bromophos Propoxur HCH DDT Sumithion Agrothion 9.5 22.0 94.0 93.5 83.0 75.5 2.5 3.0 55.5 56.0 33.5 16.5 31.5 18.0 8.0 6.0 2.5 2.0 2.3 2.0 1.5 24.0 3.5 5.0 3.0 5.0 24.0 2.5 2.5 16.0 3.5 24.0 1.0 2.5 36.0
SD
2.0 3.0 2.0 2.0 24.0 3.5 5.0 3.5 5.0 24.0 3.5 2.5 60.0 4.0 24.0 1.0 14.5 48.0
96.0 60.0 72.0 >96.0 96.0 24.0 14.5 14.5 5.0 84.0 36.0 24.0 48.0 36.0 60.0 18.3 3.0 >96.0
< 1.0 -Weeks after treatment until b
~ As listed in Table 1. 30% survival. £Hours until 50% mortality on plates one week after treatment.
TABLE 3.
RESIDUAL ACTIVITY AFTER 24-h EXPOSURE OF FORMULATED INSECTICIDES ON VARIOUS SURFACES IN THE LABORATORY
No. days until
~30%
survival i I
Insecticide g/m Tin Glass Whitewash Untreated 4 Waterbase paint Whitewash Waterbase paint Rubberbase paint
2
Mud Cement
Wood
Cardboard
Untreated
Chlorphoxim
2.0
4
Propoxur
1.5 <5 65 <15 77 37 72
<5 21 <15 30 79 4 69 <3 >32 16 >23
<5 14 52
<5
Dieldrin >98 49 <3 15 10 10
1.0 >98 ;:.98 91
27
DDT
2.0
<15
>59
Deoxycarb
2.5
>98
Fenthion
2.0
26
Bendiocarb 12 >10 3 10
0.4
0
>32 >32 18 >37 >37
>32 >32
>32 >32
>32 >32
Bendiocarb 11
0.6 >37 27 12 26 <1 <1 4
(3
HCH 3
0.8
Bromophos
2.0
wdp
17
Bromophos
2.0
EC
14
I
Jodfenphos
2.0
7
Fenitrothion 2.0
wdp
7
Fenitrothion 0.5
ULV
<1
<1 (1
Fenitrothion 1.0
ULV
0
'g ~ OQ
ro-...........
0
W-.Q.._ ~ . 00 00 0"1
t-'~
WHO/VBC/83. 886 page 14
TABLE 4. THREE EXPERIMENTAL HUTS TREATED WITH INSECTICIDES AND CHALLENGE~ WEEKLY WITH 60 UNFED FIFTH-INSTAR NYMPHS OF R. PROLIXUS. PER CENT. REDUCTION FROM PRE-TREATMENT LEVEL, CORRECTED BY 15% MEAN CONTROL HUT REDUCTION
Insecticide Formulation Dosage (g/m 2 a. i.)
Deltamethrin 5% wdp 0.05 15.4 9
Pirimiphosmethyl 20% EN 1.0 9.6 7
Bendiocarb 80% wdp 0.4 12.5 9
Pre-treatment mean No. bugs per hut No. weeks observed Post- treatment week 1 2 3 4 5 6 7 10 11 12
lOO 100 lOO 100 lOO 100 lOO lOO lOO 100 lOO lOO
0 88 75 88 88 75 51 100 88 27 75 64
91 72 6 0 15 34 0 0 0 0 0 0
13 14
a - Each week 60 unfed fifth-instar nymphs were introduced into each hut.
WHO/VBcj83. 886 page 15 TABLE 5. BIOASSAYS CARRIED OUT AFTER TREATMENT IN SEVERAL VILLAGES IN COJEDES STATE!:
Weeks until >30% survival Insecticide Formulation g/m2 Mud Untreated Chlorphoxim Propoxur Dieldrin Fenthion HCH Pirimiphos-methyl !: 24-h exposure; wdp wdp wdp wdp wdp EN 2.0 1.5 1.0 2.0 1.0 1.0 2 1
Whitewash 1 (1 21
Wood
Dung
Cardboard
>22 13
>22
12 3 3 (2
5
48-h exposure after dieldrin treatment.
TABLE 6.
SMALL-SCALE FIELD TRIALS OF FIVE WATER-DISPERSIBLE POWDER FORMULATIONS IN PALMARITA~
Insecticide
Pretreatment mean£ 13.5 10.0 10.5 15.0 12.5
Weeks until first re infestation 34 34 38 34 34
Weeks until )30% bioassay surviva~
Fenitrothion Fenthion Bromophos DDT Jodfenphos
26 >30 7
>30 >30
~ Each of five insecticides applied to a different set of two houses. b
-Number of R. prolixus per man-hour. - 10 nymphs tested per house. c
WHO/VBC/83.886 page 16
TABLE 7. TREATMENT OF FIVE VILLAGES IN COJEDES STATE WITH WATER-DISPERSIBLE POWDERS AT 2 gjm 2 ; DIELDRIN AT 1 gjm2 Insecticide Village Altitude (m) No. houses sprayed House searches: No. searched Pre-treatment mean Weeks until first reinfestation 30% reinfestation Bioassay: No. houses Weeks until 30% survival Bromophos Solano Jodfenphos Palambra 200 14 10 24.1 >26
Fenitrbthion Jiraco 400 22 12 15.4 22 >26 4 7
Fenthion Potrero Largo 400 15 8 11.4 >26 >26 4 4
Dieldrin Rio Claro 200 36 15 8.9 26 >26 3 3
zoo 19 10 22.1 20 22 7
>26 4 7
2
TABLE 8. MORTALITY OF DOMESTIC ANIMALS IN RIO CLARO 2 AND AMINTA SUAREZ AFTER ONE APPLICATION OF DIELDRIN AT 1 gjm TO HOUSES AND HCH AT 0 .8 gjm2 TO ANIMAL SHELTERS No. houses with dead animals 20 4 2 2 No. of dead animals No. of live animals
Animal
Per cent. mortality
Chickens Parrots Dogs Cats
172 5 2 3
389 NC NC NC
31
-
NC
= not
counted
WHO/VBC/83.886 page 17 TABLE 9. FIELD TRIALS WITH FENITROTHION IN THREE ECOLOGICALLY DIVERSE VILLAGES IN HOUSES WITH MUD WALLS AND GALVANIZED CORRUGATED IRON ROOFS State Municipality Village Altitude (m) Rainfall Temperature Crops Resistance to dieldrin Treatment date Houses searched Pretreatment mean~ % positive Weeks until first reinfestation 30% reinfestation Bioassay No. of houses Weeks l,lntil 30'7. survival High Cool Coffee No 14.8.1978 8 6.1 lOO 11
Barinas Calderas San Antonio 900
Tachira San Antonio Sabana Potrera 300 Low Wann Irrigated crops No 26.9.1978 23 33.6 87 13 30 23 13
Trujillo La Paz Rio Morosmoy 500 Moderate Warm Corn, beans, coffee Yes 28.ll.l978 23 8.5 87 8 26' 23 18
23 8 14
~ Number of R. prolixus per man-hour. TABLE 10. COMPARISON OF NATURAL REINFESTATION WITH "POSITIVITY" IN GOMEZ-NUNEZ TRAPs! DURING A 4-MONTH PERIOD AFTER TREATMENT WITH FENITROTHION wdp AT 1 gjm2 IN THE VILLAGE OF RIO MOROSMOY Infestation (a) (b) (c) (d) (e) a
Observed
ExpectedE.
Natural and Gomez-Nunez trap Natural only Gomez-Nunez trap only No infestation Total comparisons
5 1 6 70 82
0.8 5.2 10.2 65.8 82.0
- Two traps per house, left for four weeks. -Expected b
(a+ b) (a+ c). e
'g ~ OQ
1-'ta SUMMARY OF FIELD TRIALS IN VENEZUELA, 1976-1980
0 (1)"-.._
(X)~
TABLE ll.
. Bioassays (week) 30% bioassay survival 26 14 13
(X)
w
(X) (X)
0\
Reinfestation (week) Dosage gjm2 Village N
Insecticide R First reinfestation 30'7o reinfestation
Formulation
+
23 30 26 >26 >59
-
+
>26
22
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
+ + 13 34 26 34
34 13 ll 8 22 59 34 >26 38 20 34 >26
>26 43 >53 >26
18 7 15 >30 7 7 2 )30 4
13.
-
14. 15. 16.
Fenitrothion Fenitrothion Fenitrothion Fenitrothion Fenitrothion Fenitrothion Jodfenphos Jodfenphos Bromophos Bromophos Fenthion Fenthion Bendiocarb p-methyl Dieldrin DDT Palmaritas San Antonio Sabana Potrero Morosmoy Jiraco Morosmoy Palmaritas Palambra Palmaritas Solano Palmaritas Potrero Largo La Guaca Bucaral Rio Claro Palmaritas
wdp40 wdp40 wdp40 wdp40 wdp40 wdp40 wdpSO wdp50 wdp50 wdpSO wdp40 wdp40 wdp20 EN20 wdp50 wdp75
1.0 1.0 1.0 1.0 2.0 2.0 1.0 2.0 1.0 2.0 1.0 2.0 0.4 1.0 1.0 1.0
-
2 8 23 23 12 12 2 10 2 10 2 8 21 12 15 2
-
4 3 >30
---·-
---
---·-
...
I
R
=
Resistance to dieldrin.
N
= Number of houses searched.