A large-scale field trial of ultra-low-volume fenitrothion applied by a portable mist blower for the control of Aedes aegypti*

Bull. World Health Organ. 11974, 51, 409-415 Bull. Organ. mond. Sante A large-scale field trial of ultra-low-volume fenitrothion applied by a portable mist blower for the control of Aedes aegypti* C. P. PANT,1 H. L. MATHIS,2 M. J. NELSON,3 & BOONLUAN PHANTHUMACHINDA4 Long-term control of Aedes aegypti, the vector of dengue haemorrhagic fever in Thailand, was obtained by 2 thorough applications offenitrothion mist applied at a target dosage rate of 0.1 ml per m3 of room space. Perfect control lasted for 6-7 months after treatment and densities were substantially reducedfor up to a year after treatment. Recovery of the population was still slow up to almost 16 months after treatment. This degree of control was achieved by the immediate mortalities produced by 2 treatments spaced about 2 weeks apart, the larvicidal effect of the fenitrothion aerosol, and a limited residual effect that prevented oviposition for a period, so that the recovery potential was greatly diminished. It appears that aerosol and mist treatments designed as epidemic control measures can be adapted to long-term preventive control of A. aegypti. However, the equipment used and the method of application require further improvement. The WHO Aedes Research Unit, Bangkok, has developed several methods for the long- and short- term control of A. aegypti, the vector of dengue haemorrhagic fever in Thailand. Kilpatrick et al. (1) and Lofgren et al. (2) utilized aircraft fitted with special dispensing systems to apply malathion in ultra-low-volume (ULV) quantities for the emer- gency control of this species. Pant et al. (3), using a vehicle-mounted Leco ULV cold-fogger to apply malathion, found that 2 applications 4 days apart at about 438 ml/ha gave 99% control of A. aegypti and maintained good control up to 2 weeks after treat- ment. Bang & Pant (4) obtained 12-month control of this species in a 170-ha suburb of Bangkok by applying 4 treatments of a 1% sand-granule formu- lation of temephos a at 0.1 g/litre to all water containers. Pant et al. (5), applying ULV fenitro- * From the WHO Aedes Research Unit, Bangkok, Thailand. The unit's present address is: WHO Vector and Rodent Control Research Unit, P.O. Box 302, Jakarta, Indonesia. 1 Project Leader, 8Ecologist, WHO Vector and Rodent Control Research Unit, Jakarta. 2 Entomologist, WHO Anopheles Control Research Unit I, P.O. Box 503, Kaduna, Nigeria. ' Chief, Medical Entomology Research Unit, Department of Medical Sciences, Bangkok, Thailand. a Proposed ISO name for OO'-(thiodi-4,1-phenylene) O,O,O',O'-tetramethyl phosphorothioate. thion sequentially with the Leco machine, achieved 5-month control of the same species; Sumatrapongse & Pant (unpublished report to WHO, 1973) used a portable mist-blower to obtain similar control for 7-8 months. In the meantime, Pant & Mathis (6) demonstrated that portable mist-blowers used for house-to-house treatment not only killed adult mos- quitos immediately but also gave a limited residual effect and some larvicidal action. This opened a new field for the long-term control of A. aegypti, which is largely domestic or peridomestic in Southeast Asia. Immediate kill of adult mosquitos accompanied by the death of newly emerged mosquitos over the few days after treatment and some larvicidal effect de- pletes the environment of all stages of the insect and further disminishes the recovery potential of the population. Thus mist application, which was largely designed for the emergency treatment of adult mos- quitos, can be effective as a long-term control mea- sure, aided by its limited residual and larvicidal effectiveness and the domestic and peridomestic rest- ing, breeding and feeding habits of A. aegypti. With the other available method of long-term control, the use of temephos, every larval habitat must be found, properly treated, and periodically examined to ensure that breeding does not restart. This requires considerable labour and thorough 3283 - 409 410 C. P. PANT ET AL. supervision. Measures against adults applied to rooms or houses are less demanding and can be easily supervised. In view of this, the WHO Aedes Research Unit conducted a large-scale operational trial of fenitrothion aerosol applied by the Mity Moe Sra portable ULV machine in Bangkok in 1972. MATERIALS AND METHODS Test area The area selected for treatment is part of Sutisan, a northern suburb of Bangkok. It covers slightly under 20 ha and has a population of about 11 500 people in 1500 homes. As throughout Bangkok, there is considerable economic variation within the test area, and this is reflected in the dwellings, which include shop-houses, slums and high-income resi- dences. However, low-income wooden framed houses, often on stilts over standing water, are most numerous. They average 3.5 rooms per house with a wall surface of about 50 m2 per room. Two-storey and single-storey living quarters are common, but multi-storey structures rare. Method of treatment The test area was divided into sections that could be conveniently treated in 1 day. Three teams worked simultaneously; each team, with 1 Mity Moe Sr ULV machine, consisted of 3-4 spraymen, 1 field supervisor, and one of the authors. Because of the weight (8 kg) and vibrations of the machine the operator had to change hands frequently. An ad- vance team prepared each house for treatment by ensuring that all the inhabitants of the house moved out of the rooms to be treated and did not return until at least 30 minutes after spraying. Food was covered during the treatment. Each team began field preparations at 07 h 00 and actual spraying extended from about 07 h 30 to noon. One team normally treated about 100 houses a day; the entire test area was covered in just under 5 days. The sixth day was devoted to mopping-up operations and treatment of a barrier zone. The rooms were usually treated from the doorway by slowly swinging the nozzle of the mist-blower, held away from the operator, for a count of about 20 seconds. Since the buildings were usually imme- diately adjacent to each other, the machine normally ran continuously as the team passed from house to a Manufactured by the Buffalo Turbine Agricultural Equipment Co., Inc., 70 Industrial Street, Gowanda, New York 14070, USA. house; however, when larger open areas were en- countered, it was turned off. Detailed maps of the test area and exact routes for the teams were prepared ahead of treatment and records were kept of the houses sprayed, amount of insecticide used, etc. All houses were treated twice, at an interval of 14 days; the dates of the 2 rounds of spraying were 15-20 May and 29 May-3 June 1972. Weekly evaluations were made before, during, and after the treatments. Dosage rates Because of the method of treatment, a dosage in terms of quantity of insecticide used per unit of area would have had little meaning. Therefore a target dosage rate of 0.1 ml per m3 of room space was set. The actual dosage applied depends on certain esti- mates. If extraneous interiors-e.g., small sheds, separate storage areas, small cooking areas and bathrooms-accounted for 12.5% of the spray and outdoor treatment for a further 12.5%, the rooms themselves may have received some 20% more than the target dosage. Barrier zone Open fields protected the test area on 3 sides for a distance of at least 140 m. The northern boundary was formed by a road about 15 m wide across which were many houses, shops, and other buildings. A strip 60-70 m wide on the north side of the road was treated in the same fashion as the test area itself. All evaluations in the treated area were done at least 30 m from any boundary and at a minimum of 160 m from the closest untreated residence. Evaluation procedure Effect on natural populations. Relative population indices of A. aegypti were established and monitored once or twice weekly from at least 3 weeks before treatment to 18 months after it. A year or more after treatment the frequency of observations was reduced to once every 15-30 days. Adult population. Six field workers each collected all adult mosquitos landing on their bared lower legs for 20 minutes in each of 6 houses, making a total of 36 houses and 12 man/hours per sampling day. Beginning in April 1972, landing catches were made 3 times before treatment, twice weekly during the month after the first treatment, and once every 7-30 days until October 1973. Larval infestation. Containers in and around some 100 houses were examined 8 times for the presence ULV FENITROTHION FOR A. AEGYPTI CONTROL or absence of A. aegypti larvae between January and May 1972, and thereafter with the same frequency as the adult landing catches. The parameters estab- lished were the Aedes house index (percentage of houses with 1 or more Aedes breeding sources), the container index (percentage of water holding con- tainers producing Aedes) and the Breteau index (mean number of containers positive for Aedes larvae per 100 houses). Pupal production. Pupal production in the treated and control areas was measured by counting the number of pupae from a sample of water jars and ant traps before and after treatment. Pupal counts per container were multiplied by the number of positive containers per 100 houses to give an esti- mate of the number of pupae per 100 houses. Oviposition. At the same frequency as the landing rate catches, 2 CDC-type ovitraps were placed at each of about 15 houses, 1 outdoors and 1 indoors. After 48-hour exposure, the traps were examined under a dissecting microscope for Aedes eggs. Untreated control area. Makkasan, a district of Bangkok slightly over 5 km south of the Sutisan test area, was used as a control area. The A. aegypti sources and populations were very similar to those at Sutisan. Adult, larval and oviposition surveys were carried out in the control area in the same way and at the same time as in the treated area. Bioassays. Larval bioassays to determine any pos- sible larvicidal effect of the aerosol were carried out by placing third-instar Culex fatigans larvae in 9 water jars and 8 ant traps in floating cages prior to treatment, mortalities being recorded after 24 hours. Additional jars and ant traps were used as controls. Adult bioassays were conducted by placing 15 bloodfed 3-day-old A. aegypti in each of 20 small nylon netting cages in several houses prior to treat- ment. Additional cages were used as controls. The residual deposit of fenitrothion on the walls of 5-6 houses was checked up to 5 days after treatment using plastic cones as described by the WHO Expert Committee on Insecticides (7). Ten to 15 3-day-old bloodfed A. aegypti per cone were exposed for 30 minutes to the wall surfaces using 3 cones per house; 1-2 additional houses were used as controls. Droplet size. A slide coated with magnesium oxide was passed through the spray at about 2 m from the nozzle of the machine. The diameters of the resulting craters were measured and corrected and the volume median diameter was calculated according to the method recommended by the WHO Expert Com- mittee on Insecticides (8). Droplet density. Six oil-sensitive dye cards were exposed in each of 8 houses during treatment. Half of the cards were placed vertically and the others horizontally. The number of droplets in a sample of 12.9 cm2 was counted on each card using the low power of a stereoscopic dissecting microscope. Operational notes The principle of ULV application with the Mity Moe Sr machine was excellent and results were very good. However, machine breakdowns were common and became a routine part of the operation; they had many causes, but proved repairable. No unduly hard treatment was given to any machine, but a mechanic, a vehicle, tools and a spare machine were kept busy almost constantly during treatment days. Considerable time was spent in planning before the actual treatments. A series of detailed maps was made that included every house in the area to be treated. Before actual treatment each sprayman was trained in the handling of the Mity Moe Sr. These preparatory efforts were worthwhile; on every treat- ment day each team completed its assignment effi- ciently. Refusals to permit spraying were extremely rare, even in the high-income homes. Some householders who at first refused treatment later requested the spraymen to come back and treat their homes. Bangkok's hot climate prevents spraymen from wearing respirator masks and heavy or nonporous clothing. However, normal safety instructions were given, e.g., no smoking or eating during spraying, daily changes of clothing, frequent washing of face and hands, and care not to direct the nozzle towards a person. For their protection, house occupants vacated the rooms before treatment and all exposed food was covered. No baseline cholinesterase levels were established for spraymen before the spraying started. However, midway through the second treatment week, blood was taken and cholinesterase levels were measured according to the method of Biggs et al. (9). Twelve days later 9 men tested previously were checked again. RESULTS Droplet size and density Droplets from the mist blower when it was dis- charging insecticide at the rate of 15 ml/min ranged in size from 19 to 109 ,um in diameter. The volume median diameter was 60 ,um. 411 C. P. PANT ET AL. Considerably more droplets fell on horizontal than on vertical surfaces. The mean number of droplets on horizontal cards was 48.8/cm2; on verti- cal cards it was 8/cm2. In the 8 houses under observation the droplet density on horizontal cards ranged from a mean minimum of 28/cm2 to a mean maximum of 73.3/cm2. For vertical cards the mean minimum was 0.93/cm2 and the mean maximum 19.8/cm2. Bioassays Larvae. Test larvae placed in water jars and ant traps during treatment showed 79.5% mortality in the former and 97.5% in the latter. The higher percentage kill in ant traps was to be expected, as more droplets fall on the floor. Control mortalities were 7.5 % in both types of container. Adults. There was 100% kill of caged mosquitos in treated houses; however, control mortality was high (36 %). Bioassays of residual insecticide on the walls gave 100% kill 1 day after treatment (control mor- tality 14.2%) and 58.2% kill 5 days after treatment (control mortality 16.6 %). Natural A. aegypti population Adults. The mean numbers of adults landing on human bait in the treated and control areas before and after treatment are shown in Table 1. Pretreat- Table 1. Mean numbers of adult A. aegypti landing on human bait per man/hour (= adult landing rate) in an area treated with ULV fenitrothion and in an untreated control area Treated area Untreated control area Month No. of No. of Ma adn sampling Mean landing rate sampling Mean landing days a days rate Before treatment April-May 1972 3 25.2 1 22.1 Treatment 1 May 1972 6 0.1 4 19.8 Treatment 2 June 1972 2 0.1 4 19.8 July 1972 4 0.0 3 20.4 August 1972 4 0.0 4 20.4 September 1972 6 0.0 4 19.1 October 1972 4 0.0 4 18.6 November 1972 3 0.07 5 17.1 December 1972 4 0.0 4 17.8 January 1973 5 0.9 4 15.6 February 1973 4 2.0 4 18.0 March 1973 3 2.6 3 18.0 Eastern Western zone zone April 1973 4 0.1 4.0 2 19.0 May 1973 6 0.0 6.2 3 21.0 July 1973 1 0.4 2.9 1 19.8 August 1973 1 0.0 4.8 2 19.6 September 1973 1 0.0 5.9 1 16.3 October 1973 1 0.2 4.5 4 15.0 a On each sampling day one 20-minute landing catch was made in each of 36 houses. 412 ULV FENITROTHION FOR A. AEGYPTI CONTROL ment landing rates in the test area averaged 25.2 adults per man/hour. During the first month after treatment this was reduced to well below 0.1 per man/hour, a reduction of more than 99%. The few female mosquitos collected were nulliparous. With the exception of 1 fully gravid female collected resting indoors 121 days after treatment, no further adults were captured until the sixth month. During early November 1972, 6 months after treatment, the mean landing rates were slightly over 0.1 per man/hour. In late November and December, 7 months after treatment, no adult mosquitos were collected, the reduction remaining above 99 %. In the eighth month, the adult population showed obvious signs of recovery although it was still extremely low. The mean landing rate for the month was 0.9 per man/hour. From 9 to 17 months after treatment the mean landing rates in the treated area were 2.0-3.1 per man/hour, or 69 %-92% lower than in the control area. The eastern half of the treated area had much lower landing rates than the western half. From 11 to 17 months after treatment the rates in the eastern section were only 0.0-0.4 per man/hour, compared to 2.9-6.2 per man/hour in the western section. Landing rates in the control area were 15.0-21.0 per man/hour during the same period. The eastern section, bordered by the wide airport high- way to the east and vacant fields to the south, was much more isolated than the western section. Oviposition. Of the 62 ovitraps placed in the test area before treatment, 32.3 % were positive for A. aegypti eggs. Of the 1020 placed during the 8 months after treatment (72-230 traps per month) none was positive. Positive ovitraps were first re- corded again in February 1973; the rate continued to be much lower (5 %-10 %) than before treatment and in the control area, where 27°/-67% of traps were found to be positive. Larval infestation. The larval infestation indices were the only parameters that never reached zero. Thus it would appear that when A. aegypti popula- tion levels are low, these indices would be the most sensitive. Of the various larval indices, the Breteau index seems to be the most useful. The monthly means of Breteau indices obtained from weekly observations at 100 houses showed a reduction of over 90% during the first 7 months after treatment. From 8 to 17 months this reduction was 82.5y%-87.6% compared to the pretreatment obser- vations and 84.8%-90.7% as compared to the con- trol area (Table 2). As in the case of landing rates, the best results were obtained in the more isolated eastern section of the treated area where, from 11 to 17 months after treatment, reductions of 96%-100% were observed, as against reductions of 63Y%-76% in the western zone. Pupal production. There was a marked decrease in the mean number of pupae per larva-positive con- tainer after each treatment. In the Makkasan control area the pupal counts were more or less constant. In 15 observations extending over 5 months the means ranged from 1.0 to 10.0 in a sample of over 100 water jars and from 0.8 to 22.2 in a similar sample of ant traps. At Sutisan, however, during the week after the first treatment there was already a decrease of 66% in pupal production in water jars and of 97% in the more exposed and traps. This follows the pattern of the larval bioassays. For the next 3 months the mean reduction in pupal production was 98% in water jars and nearly 100% in ant traps. The further reduction was caused by lack of oviposition during the week following the treatments: adults were killed immediately on treatment days and the mist's resi- dual effect resulted in no oviposition in the area for several weeks. Before treatment it was estimated that nearly 1000 pupae per 100 houses were present in water jars and ant traps. After the 2 treatments the count fell to 17 per 100 houses. Toxicology There were no complaints from any occupant of the treated area during the entire work programme. During the first treatment week and for the first 3 days of the second treatment week there were no complaints from any spraymen. On the following day 2 of the 20 persons involved in spraying com- plained of headache, nausea and some vomiting. On the next day 1 sprayman and the following day another had similar complaints. Within 2 days, all the workers had completely recovered. As already mentioned, blood cholinesterase levels were first established for the spray teams midway through the second treatment week, 9 workers being tested again after a further 12 days. In this period the cholinesterase levels of the 9 men increased from a mean of 86% to 98% of normal. The greatest percentage increase, from 38% to 69% of normal, occurred in a sprayman who reported no symptoms. With rigidly enforced safety rules, it is strongly felt that fenitrothion can be used safely as an ULV insecticide. 413 C. P. PANT ET AL. Table 2. Mean number of containers per 100 houses positive for A. aegypti larvae or pupae (= Breteau index) in treated and untreated areas Treated area Untreated control area Month No. of No. of Mean landing sampling Breteau index samp;ling M aealndgdays days rt Before treatment January-May 1972 Treatment 1 May 1972 Treatment 2 May 1972 June 1972 July 1972 August 1972 September 1972 October 1972 November 1972 December 1972 January 1973 February 1973 March 1973 8 3 6 3 4 5 5 4 3 4 4 3 226 3 188 71.3 0.0 15.7 10.9 8.2 8.2 1.7 6.0 19.9 28.9 34.6 39.5 4 205 3 184 5 181 4 212 5 186 4 161 4 218 3 212 4 229 3 267 Eastern a Western a zone zone 1 2.5 72.8 2 0.6 55.1 9.0 47.8 0.0 63.0 0.0 72.9 1.2 64.2 1 270 3 223 1 1 268 286 1 392 3 283 a Based on 50 houses sampled per zone per sampling day. Pretreatment Breteau indices were 254 in the eastern zone and 197 in the western zone. DISCUSSION Insecticide mists and aerosols produce an immedi- ate reduction of the population of adult A. aegypti. However, the immature population-eggs, larvae and pupae-is generally not affected. The present trial showed that by increasing the dosage rate and applying the mist directly to the domestic and peridomestic environment of the vector, particularly when the insecticide, like fenitrothion, exhibits larvicidal effectiveness and limited residual activity on resting surfaces, a reduction of the immature population can also be achieved. A second treatment after a 10-12-day interval caused further adult mor- tality; thus, over a 3-week period, adults were exposed more or less continuously, either directly to the mist, or to the insecticide's residual action. This combination of factors reduced the population's potential for recovery to a minimum. The few emerging adults were probably so sparse that their opportunities for mating were considerably reduced. April 1973 May 1973 July 1973 August 1973 September 1973 October 1973 414 1 ULV FENITROTHION FOR A. AEGYPTI CONTROL 415 The Mity Moe Sr machine used in this trial was not ideal, because of the frequent mechanical fail- ures. However, the principle of the method was shown to work. In future uses of the method, a better machine and more rigorous toxicological safety measures may be required. ACKNOWLEDGEMENTS This trial was made possible by the kind collaboration of several agencies. Thanks are due to Dr Prakorb Tuchinda, Director-General, Department of Medical Sciences, Bangkok, and Dr Sujarti Jatanasen, Ministry of Health, Thailand. The strenuous work of insecticide application was done by spraymen provided by the Department of Medical Sciences, Bangkok, and the Div- ision of Health, Bangkok Municipality. Special thanks go to Mr Wirat Sumatrapongse and Mr Anek Raungsri, who repaired the broken and defective machines with great skill, and to Mr Poonyos Rielrengbunya, who super- vised the operations. Thanks are also due to Miss Prapaipuk Chattrabudhi, Mr Prakong Phan-Urai and Miss Pimpa Paoungpongkul for their assistance. This work could not have been carried out without the unfailing support and hard work of the mosquito scouts and staff of the WHO Aedes Research Unit. RtSUMti ESSAI PRATIQUE A GRANDE tCHELLE D'APPLICATION DE FtNITROTHION SOUS ULTRA-FAIBLE VOLUME A L'AIDE D'UN PULVJERISATEUR PORTABLE POUR LA LUTTE CONTRE AEDES AEGYPTI On a utilise un appareil portable pour appliquer le fenitrothion sous ultra-faible volume a la dose de 0,1 mI/m3 dans 1500 maisons d'un faubourg de Bangkok habitees par 11 500 personnes. Une bande de terrain large de 60-70 m, a la limite nord du secteur, a egalement e traitee. Deux traitements ont e effectues 'a 10-12 jours d'intervalle par 3 equipes, i la cadence de 300 habitations par jour. Le taux d'emission de l'insecticide etait de 15 ml/min avec des gouttelettes d'un diametre moyen de 60 ,um. La mortalite de larves de Culex fatigans placees dans les pots a eau et dans les pieges a fourmis de la zone trait&e a atteint 79,5 et 97,50% respectivement, contre 7,5% parmi les larves temoins. Tous les Aedes aegypti adultes encages ont ete tues pendant le traitement (36% de mortalite chez les insectes temoins). L'activite residuelle de l'insecticide a entraine une mortalite de 100% chez A. aegypti adulte 1 jour apres le traitement (temoins: 14,2%), et de 58,2% apres 5 jours (temoins: 16,6 %). Six mois apres le traitement, le nombre moyen d'A. aegypti captures sur piege humain a et abaisse de 99% par rapport au nombre enregistre avant le traitement, et de 69-92% apres 17 mois. Le nombre moyen, par 100 maisons, de recipients renfermant des larves ou des nymphes d'A. aegypti a e reduit de plus de 90% durant les 7 premiers mois suivant le traitement et de plus de 82% du 8e au 17e mois. La production de nymphes a d6cru de plus de 98% pendant les 3 pre- miers mois apres le traitement. L'elimination durable d'A. aegypti par le fenitrothion sous ultra-faible volume est le resultat d'une action combinee: directe sur les adultes, directe sur les larves, residuelle chez les adultes entrant en contact avec les surfaces traitees, ainsi que du choix d'un traitement en deux temps tuant les adultes recemment eclos avant toute ponte. Cependant, l'equipement et la methode d'application doivent encore etre ameliores. REFERENCES 1. KILPATRICK, J. W. ET AL. Bulletin of the World Health Organization, 42: 1-14 (1970). 2. LOFGREN, C. S. ET AL. Bulletin of the World Health Organization, 42: 15-25 (1970). 3. PANT, C. P. ET AL. Bulletin of the World Health Organization, 45: 805-817 (1971). 4. BANG, Y. H. & PANT, C. P. Bulletin of the World Health Organization, 46: 416-425 (1972). 5. PANT, C. P. ET AL. Bulletin of the World Health Organization, 48: 455-459 (1973). 6. PANT, C. P. & MATHIS, H. L. Southeast Asian journal of tropical medicine and public health, 4 (2): 231-237 (1973). 7. WHO Technical Report Series, No. 443, 1970, An- nex 16A. 8. WHO Technical Report Series, No. 465, 1971, An- nex 1. 9. BriGcs, H. G. ET AL. American journal of clinical pathology, 30: 181-186 (1958).