• dengue;
  • vector control;
  • Bacillus thuringiensis israelensis ;
  • Aedes


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Cited References


To systematically review the literature on the effectiveness of Bacillus thuringiensis israelensis (Bti), when used as a single agent in the field, for the control of dengue vectors.


Systematic literature search of the published and grey literature was carried out using the following databases: MEDLINE, EMBASE, Global Health, Web of Science, the Cochrane Library, WHOLIS, ELDIS, the New York Academy of Medicine Gray Literature Report, Africa-Wide and Google. All results were screened for duplicates and assessed for eligibility. Relevant data were extracted, and a quality assessment was conducted using the CONSORT 2010 checklist.


Fourteen studies satisfied the eligibility criteria, incorporating a wide range of interventions and outcome measures. Six studies were classified as effectiveness studies, and the remaining eight examined the efficacy of Bti in more controlled settings. Twelve (all eight efficacy studies and 4 of 6 effectiveness studies) reported reductions in entomological indices with an average duration of control of 2–4 weeks. The two effectiveness studies that did not report significant entomological reductions were both cluster-randomised study designs that utilised basic interventions such as environmental management or general education on environment control practices in their respective control groups. Only one study described a reduction in entomological indices together with epidemiological data, reporting one dengue case in the treated area compared to 15 dengue cases in the untreated area during the observed study period.


While Bti can be effective in reducing the number of immature Aedes in treated containers in the short term, there is very limited evidence that dengue morbidity can be reduced through the use of Bti alone. There is currently insufficient evidence to recommend the use of Bti as a single agent for the long-term control of dengue vectors and prevention of dengue fever. Further studies examining the role of Bti in combination with other strategies to control dengue vectors are warranted.


Revue systématique de la littérature sur l'efficacité de B acillus thuringiensis israelensis ( B ti), lorsqu'utilisé comme agent unique sur le terrain, pour la lutte contre les vecteurs de la dengue.


Recherche systématique de la littérature publiée et grise dans les bases de données suivantes: Medline et EMBASE, Global Health, Web of Science, Cochrane Library, WHOLIS, ELDIS, les reports de littérature grise de la New York Academy of Medicine, Africa-Wide et Google. Tous les résultats ont été examinés pour éviter les doublons et pour l’éligibilité. Les données pertinentes ont été extraites et une évaluation de la qualité a été réalisée en utilisant la liste de contrôle CONSORT 2010.


14 études répondaient aux critères d’éligibilité, intégrant un large éventail d'interventions et de mesures des résultats. 6 études ont été classées comme études sur l'effectivité, les 8 autres ont examiné l'efficacité du Bti dans des cadres plus contrôlés. 12 études (dont les 8 études d'efficacité et 4/6 études d'effectivité) ont rapporté une baisse des indices entomologiques avec une durée moyenne de contrôle de 2 à 4 semaines. Les 2 études de rendement qui n'ont pas rapporté d'importantes réductions entomologiques étaient toutes deux des études randomisées en grappes qui ont utilisé les interventions de base tels que la gestion de l'environnement ou l’éducation générale sur les pratiques de contrôle de l'environnement dans leurs groupes témoins respectifs. Seule 1 étude a décrit une réduction des indices entomologiques ainsi que des données épidémiologiques, rapportant 1 cas de dengue dans la zone traitée, comparé à 15 cas de dengue dans la zone non traitée, pendant la période observée de l’étude.


Bien que le B ti peut être efficace dans la réduction du nombre d’ A edes immatures dans des conteneurs traités dans le court terme, il y a très peu d’évidences que la morbidité de la dengue peut être réduite grâce à l'utilisation du B ti seul. Les évidences actuelles sont insuffisantes pour recommander l'utilisation du B ti comme agent unique dans le contrôle à long terme des vecteurs de la dengue et pour la prévention de la dengue. Des études supplémentaires portant sur le rôle du B ti en combinaison avec d'autres stratégies pour contrôler les vecteurs de la dengue sont justifiées.


Realizar una revisión sistemática de la literatura existente sobre la efectividad del B acillus thuringiensis israelensis ( B ti) cuando se utiliza como único agente en el campo para el control vectorial del dengue.


Búsqueda sistemática en la literatura publicada y gris, utilizando las bases de datos de Medline, EMBASE, Global Health, Web of Science, Cochrane, WHOLIS, ELDIS, el informe de literatura gris de la Academia de Medicina de Nueva York, Africa-Wide y Google. Todos los resultados se revisaron en busca de duplicados y se evaluaron para determinar si eran elegibles. Se extrajeron los datos relevantes y se realizó una evaluación de calidad utilizando la lista de comprobación CONSORT 2010.


14 estudios cumplían los criterios de inclusión, incorporando un amplio rango de intervenciones y de resultados medidos. Seis estudios se clasificaron como estudios de efectividad, y los 8 restantes examinaban la eficacia de B ti en emplazamientos más controlados. 12 (los 8 estudios de eficacia y 4/6 estudios de efectividad) reportaron reducciones en los índices entomológicos con una duración promedio del control de 2–4 semanas. Los 2 estudios de efectividad que no reportaron una reducción entomológica significativa, estaban dentro de los estudios con un diseño de aleatorización por conglomerados que utilizaban intervenciones básicas tales como el manejo ambiental o una educación general sobre prácticas de control ambiental en sus grupos control respectivos. Solo 1 estudio describía una reducción en los índices entomológicos junto con datos epidemiológicos, reportando 1 caso de dengue en el área tratada, en comparación con 15 casos de dengue en el área sin tratar, durante el periodo de observación del estudio.


Mientras que B ti puede ser efectivo a la hora de reducir el número de A edes inmaduros en contenedores tratados a corto plazo, hay una evidencia limitada con respecto a la reducción de la morbilidad por dengue mediante el uso de B ti por sí solo. Actualmente no hay suficiente evidencia para recomendar el uso de B ti como un único agente para el control a largo plazo de los vectores del dengue y la prevención de la fiebre del dengue. Se requieren más estudios que examinen el papel del Bti en combinación con otras estrategias para controlar los vectores del dengue.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Cited References

Dengue is the most rapidly advancing vector-borne disease in the world (Special Programme for Research and Training in Tropical Diseases, WHO/TDR 2007). Overall, it is estimated that 2.5 billion people, or roughly one-third of the world's population, live in dengue-endemic areas. Annually, an estimated 50–100 million cases of dengue fever and several hundred thousand cases of severe dengue occur.

In the absence of any vaccine and preventive drugs, controlling the mosquito vectors of dengue (principally Aedes aegypti) is the only way to prevent and control dengue transmission. Chemical and biological agents remain important components of dengue vector control programmes. Unfortunately, the widespread use of chemical insecticides has contributed to increasing resistance to these agents among A. aegypti, especially in the Americas and the Caribbean (Rodriguez et al. 2005; Harris et al. 2010).

Biological control, ‘based on the introduction of organisms that prey upon, parasitise, compete with or otherwise reduce populations of the target species’, is considered a practical alternative to the application of chemical insecticides in controlling mosquito vectors of disease (Special Programme for Research and Training in Tropical Diseases, WHO/TDR 2007). In addition to reducing both toxicity to non-target species and environmental contamination, biological control also offers reduced potential for resistance development. However, biological control agents can be comparatively expensive and logistically more challenging to deploy and maintain compared to traditional chemical agents.

Bacillus thuringiensis var. israelensis (Bti) is a gram-positive, spore-forming entomopathogenic bacterium first isolated in 1976 (Goldberg & Margalit 1977). As a biological control agent, Bti has demonstrated high efficacy against target organisms, primarily mosquito and black fly larvae (Mittal 2003; Lacey 2007). Bacillus thuringiensis israelensis exerts its lethal effects through producing a variety of toxic proteins that are ingested by the larvae of susceptible organisms. These toxins are then activated in the gut of the larvae where they cause disruption of the cell membranes and death of the organism. The specificity of this mechanism has been demonstrated in multiple studies with no adverse effects on non-target invertebrates and vertebrates (Lee & Scott 1989; Merritt et al. 1989; Lacey & Mulla 1990; Saik et al. 1990). Because of the complex mechanism of action involving many proteins, the potential for resistance development is greatly reduced. Bacillus thuringiensis israelensis is available in a number of formulations that can be applied by hand or with conventional spray equipment (Lacey 2007), allowing Bti to be utilised in a variety of breeding habitats.

To date, no systematic review of the scientific literature has been undertaken to examine the evidence for the effectiveness of Bti against dengue vectors. The objective of this study was to provide a systematic review of the effectiveness of Bti, when used as a single agent in the field, for the control of dengue vectors and prevention of dengue fever.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Cited References

This review follows the reporting guidelines set forth in the PRISMA statement for systematic reviews and meta-analyses (Liberati et al. 2009).

Eligibility criteria

The eligibility criteria for the reviewed literature included the following: (i) research with an experimental design producing primary quantitative data, (ii) research conducted in the field, defined as any community or environment where dengue vectors naturally occur, (iii) the use of Bti as a single agent to control dengue vectors, (iv) clear information on Bti formulation and dosing, (v) outcome measures reported as immature indices (i.e. Stegomyia indices, oviposition indices and/or presence/absence of immature stages of Aedes) and (vi) a minimum follow-up period of at least 20 days. Only literature reported in English was included. Conference abstracts and proceedings from conferences were excluded.

Search strategy

The literature search and analysis was developed and carried out through March 2012, by two data extractors. The search terms derived from three major categories: (i) dengue disease, which included ‘dengue’ and ‘dengue hemorrhagic fever’, (ii) the Bti intervention and (iii) the outcome, which centred on reductions in dengue vector density as measured by various entomological indices. The search included both free text and subject heading terms.

The search strategy was applied to the following databases to locate peer-reviewed studies: MEDLINE, EMBASE, Global Health, Web of Science, the Cochrane Library and WHOLIS. The following databases were searched for grey literature: ELDIS, the New York Academy of Medicine Gray Literature Report, Africa-Wide and Google. Because of the limited search options available in many of the grey literature databases, a broad search strategy was employed, typically including only the terms ‘dengue’ and/or ‘Bacillus thuringiensis’.

Study selection, data extraction

All results were screened for duplicates by author, title, journal and publication date. In the first stage, results were screened based on the title and abstract only. The full text of those studies that were not excluded was then reviewed for final assessment. The reference section of each of the selected publications was reviewed to identify additional relevant studies. Relevant information, including study design, setting and Bti formulation, from each of the selected studies was extracted, and when information was unclear from the reported results, an attempt was made to contact the corresponding author. Studies were first classified as either efficacy studies, where Bti was placed in targeted containers that were observed for reductions in immature stages, or effectiveness studies, where Bti was used to treat containers or peridomestic areas, and surveyed for reductions in classical entomological indices. Study designs were subsequently divided into three categories: randomised or quasi-randomised controlled trials (RCT), cluster-randomised controlled trials (CRCT) and non-randomised controlled trials (NRCT).


Because the majority of the included studies were non-randomised, it was not possible to utilise an existing, validated instrument for assessing risk of bias. Instead, the CONSORT 2010 checklist served as a framework to describe limitations in the conduct and reporting of the included studies. No studies were excluded for quality reasons if the eligibility criteria were met, but limitations and possible biases are reported in the results section. Different Bti formulations and methods of application, together with varying statistical analyses and outcome measures across the studies, precluded any attempt at meta-analysis.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Cited References

Study selection

After screening out duplicates, the literature search identified 355 articles for assessment. Fourteen articles met the eligibility criteria. Of these, nine were identified from the published literature databases and three from the grey literature search (Phan-Urai et al. 1995; Haq et al. 2004; Lam et al. 2010). The most common reasons for exclusion were as follows: interventions did not use Bti (190 studies), the manuscripts were review articles and/or of non-experimental design (93 articles) or the studies took place in laboratory or semi-field settings (33 articles). A review of the references cited in the articles that met the eligibility criteria yielded one additional study (Dua et al. 1993). One more article (Tan et al. 2012) that was published during manuscript preparation was included in the review. No further studies were included after reviewing the references of this additional article. Figure 1 summarises the selection process.


Figure 1. Flow chart describing paper selection and inclusion/exclusion process.

Download figure to PowerPoint

The 14 included studies are summarised in Tables 1 and 2, in which they are assigned a unique identifier for reference purposes in the remainder of this manuscript.

Table 1. Summary of efficacy studies evaluating the use of Bti for the control of dengue vectors
No.ReferenceSettingObjectivesStudy design and durationBti formulationIntervention group(s)Control group(s)Principal outcomes and statistical analysisResultsStudy conclusions
1.Ansari and Razdan (1999) Delhi, IndiaEvaluate the Bt H-14 granule formulation under laboratory and field conditions against Aedes aegypti


4 weeks

Bti H-14 granules

Potency not reported

Manufacturer not reported

Single spray application of Bti H-14 applied to 20 evaporation coolers and 36 discarded tires at three different doses (0.25, 0.50 and 1.0 g/m2)No treatment of 4 evaporation coolers and 6 discarded tires

Larval density and per cent reduction (Mulla)

No statistical testing of results

Reduction at 4 weeks:

Evaporation coolers:

0.25 g/m2 = 46.3%

0.5 g/m2 = 100%

1.0 g/m2 = 100%

Discarded tires:

0.25 g/m2 = 47.7%

0.5 g/m2 = 100%

1.0 g/m2 = 100%

The formulation was most effective at a dose of 0.5 g/m2 with larvicidal activity persisting for up to 4 weeks. Bt H-14 may be used for control of dengue, but studies examining relative efficacy and cost-effectiveness are required
2.Batra et al. (2000)Delhi, IndiaEvaluate the effectiveness of 3 formulations of Bti against immature A. aegypti in coolers and tires


6 weeks

Bacticide (1000  ITU/mg)

Biotech International, Delhi

Vectobac DT (1900 ITU/mg)

Vectobac G (3500 ITU/mg)

Abbott Laboratories

Single application of Bti in 5 different formulations/doses:

(1) 15 desert coolers with Vectobac G at 2 g/cooler

(2) 65 desert coolers with Vectobac DT at 0.75 g/cooler

(3) 10 desert coolers with Bacticide powder at 1.0 g/cooler

(4) 35 discarded tires with Vectobac DT at one tablet (0.375 g) per tire

(5) 20 discarded tires with Vectobac DT at two tablets (0.75 g) per tire

No treatment of 10 desert coolers

Per cent of breeding sites positive for 3rd/4th stage instars and pupae; per cent reduction (Mulla)

No statistical testing of results

Per cent reduction

Intervention 1: Vectobac G 2 weeks = 100%

4 weeks = 66.7%

6 weeks = 16.7%

Intervention 2: Vectobac DT

2 weeks = 100%

4 weeks = 86.5%

6 weeks = 28.8%

Intervention 3: Bacticide

2 weeks = 100%

4 weeks = 25.0%

6 weeks = 12.5%

Intervention 4: Vectobac DT

2 weeks = 100%

4 weeks = 60.9%

6 weeks = 6.3%

Intervention 5: Vectobac DT

2 weeks = 100%

4 weeks = 100%

6 weeks = 0%

Results showed that Bti formulations provide complete control of A. aegypti larvae for 2–4 weeks. Study highlights the practical community-based application of Bti. Use of these formulations is effective and more user-friendly than conventional methods
3.Dua et al. (1993)Hardwar, IndiaEvaluate Bactoculicide for its efficacy to control mosquitos (Aedes, Anopheles, Culex) breeding in factory scraps in an industrial area


6 weeks


Potency not reported

Bordsk Chemical, Moscow

Single application of Bactoculicide sprayed over the standing water in 73 industrial scrap containers positive for mosquito larvae at a dosage of 0.5 g/m2No treatment of 10 industrial scrap containers positive for mosquito larvae

Larval density of 3rd/4th stage instars, per cent reduction (Mulla)

No statistical testing of results

Per cent Reduction:

24 hrs: 100%

10 days: 100%

14 days: 97.7%

24 days: 98.4%

5th week: 100%

6th week: 90%

Bactoculicide appeared to be the best solution for controlling mosquito breeding in such problematic habitats, controlling mosquito breeding for up to 5 weeks
5.Haq et al. (2004)Surat City, IndiaEvaluate spraying of two bacterial larvicide formulations for efficacy against Anopheles, Culex and Aedes mosquitoes under the operational conditions of an urban malarial control programme


20 days

Bacticide WP (1000 ITU/mg)

Biotech International, Delhi

Vectobac 12AS (1200 ITU/mg)

Aventis Crop Sciences

(1) Bacticide sprayed at 5 kg/ha in 5 cemented tanks and chambers at 15 construction sites; retreatment at 10 days

(2) Vectobac sprayed at 11 kg/ha in 5 cemented tanks and chambers at 16 construction sites; retreatment at 10 days

No treatment of 8 cemented tanks and chambers at 15 construction sites

Pupal/larval densities and per cent reduction (Mulla)

No statistical testing of results

(1) Bacticide provided 100% reduction for duration of the study

(2) Vectobac provided 100% reduction through day 3; 53.1% reduction on day 17 after retreatment

Study demonstrated that biolarvicides should be used at an interval of 7–10 days. Liquid Vectobac had a relative ease of operation compared to Bacticide. Biolarvicides can be incorporated as part of integrated vector control program
Table 2. Summary of effectiveness studies evaluating the use of Bti for the control of dengue vectors
No.ReferenceSettingObjectivesStudy design and durationBti formulationIntervention group(s)Control group(s)Principal outcomes and statistical analysisResultsStudy conclusions
  1. Bti, Bacillus thuringiensis israelensis; RCT, randomised controlled trials; CRCT, cluster-randomised controlled trials; NRCT, non-randomised controlled trials.

4.Favier et al. (2006)Brazilia, BrazilDetermine the influence of climate and of environmental vector control with or without insecticide on Aedes aegypti larval indices and pupal density


18 months

Mosquito Dunks (7000 ITU/mg)

Summit Chemical

Four intervention groups, each containing environmental management plus:

(1) Temephos (Abate) at 1 ppm in all containers

(2) Temephos (Abate) at 1 ppm in all containers in all premises

(3) Bti (Mosquito Dunks) in all containers in positive premises only

(4) Methoprene-S (Altosid) in all containers in positive premises only

Only environmental management, defined as incitation to container removal when possible or emptying water in storage containers in all premises

House index, container index, Breteau index, pupal density

Nonparametric tests with 95% CI for indices and Mann–Whitney U-test for proportions

No significant differences in larval indices and pupal density between control and intervention groupsIn moderately infested areas, insecticides do not improve upon environmental vector control. Infestations could be further reduced by focusing on residences and containers particularly at risk
7.Lam et al. (2010)Western SingaporeInvestigate the efficacy of Bti against Aedes albopictus in a forested military training ground where source reduction in natural breeding sites is difficult and non-specific insecticides may cause harm to the ecosystem


3 months

Vectobac WG (3000 ITU)

Valent BioSciences

130 ha sprayed with Bti at a dosage of 500 g/ha every 2 weeks using motorised back-pack and vehicle-mounted sprayers128 ha treated with standard control strategies comprised of weekly oiling of ground larval habitats and monthly treatment of permanent water bodies with temephos

Ovitrap index, larval density, per cent reduction (Mulla)

t-test for OI and larval density

Per cent reduction in ovitrap index at intervention site:

Month 1 = 14.2% (P < 0.05)

Month 2 = 47.9% (P < 0.05)

Month 3 = 66.0% (P < 0.05)

Per cent reduction in larval density at intervention site:

Month 1 = 59.3% (P < 0.05)

Month 2 = 53.2% (P < 0.05)

Month 3 = 80.0% (P < 0.05)

Wide application of Bti into vegetation to treat all natural breeding sites produced a significant decline of the adult A. albopictus population compared to the control This is an innovative approach that can be easily adapted to all communities to successfully suppress the A. albopictus adult population
8.Lee et al. (2008)Selangor State, MalaysiaDetermine the impact of larviciding with a Bti formulation, Vectobac WG, on the adult mosquito population in a dengue endemic site in Selangor State, Malaysia


12 weeks

Vectobac WG (3000 ITU/mg)

Valent BioSciences

Two residential areas (20 houses each) received a single indoor treatment of Vectobac WG in all water containers >50 l at a dosage of 8 mg/l combined with back-pack spraying of all outdoor larval habitats at a dosage of 500 g/ha every 2 weeksOne untreated residential area that was treated with extensive space spraying 6 weeks into the trial due to a dengue outbreak

Ovitrap index and larval density

t-test for OI and larval density

The OI significantly decreased in both study sites 4 weeks after initiating the intervention, declining from 57.5% to 19.1% (P < 0.05) at one site and 66.7% to 30.3% (P < 0.05) at another. This decline in the OI was paralleled by a similar decline in both A. aegypti and Aedes albopictus larval densities. An increase in OI and larval density was observed in both sites following cessation of the intervention. The OI at the control site remained high until the initiation of space spraying following a dengue outbreakWidespread application of Vectobac WG at targeted larval habitats is able to provide control of dengue vectors. Space spraying of Bti was found to be superior to traditional methods of application in terms of effectiveness, coverage and labour. Additionally, Bti did not produce any undesirable environmental consequences
11.Ocampo et al. (2009)Cali, ColumbiaEvaluate two control methods for A. aegypti that can be used by the community: lethal ovitraps and Bti briquettes


4 months

Bactimos briquettes (7000 ITU/mg)

Summit Chemicals

Four neighbourhoods, each had one block (40 houses) randomly selected to receive education and either (1) lethal ovitraps, (2) Bti briquettes in main breeding sites, (3) lethal ovitraps + Bti briquettes or (4) no insecticide (control); Bti dosage was 1/4 briquette in each water storage tank (76 total), which were replaced monthlyOne block from each of the 4 neighbourhoods was randomly selected to receive no insecticide, but only education (environmental control, emptying water jars)

House index, pupal index (mean pupae per house) and adult index (per cent of houses infested with adult Aedes)

Poisson regression

Entomological indices obtained during the intervention period were not significantly different between treatment groups and controls. During the entire study period (341 visits), only one water tank treated with Bti was positive for larvae. Positive containers consisted mostly of plants in water and small containers. Bti briquettes were not used routinely in 40% of housesPre-intervention education and environmental management may have contributed to the lack of effect seen with interventions. Decrease in pupal density did not eliminate presence of adult mosquitoes in homes to a level sufficient to prevent transmission suggesting larger buffer zone may be required to address breeding sites outside of the residential areas. Use of Bti briquettes may have induced people to change water more often
12.Phan-Urai et al. (1995)Chanthaburi Province, ThailandEvaluate the Bti H-14 formulated tablet for Aedes larvae in a rural village of Chanthaburi Province


17 weeks

Larvitab 1-g tablet (600 ITU/mg)

Manufacturer not reported

Bti at a dosage of 1 g per 200 l was applied to all potential breeding sites, including water storage containers, cement baths and ant guards, in a single village (61 houses) in Chanthaburi Province. Treatment was repeated when containers were reinfested with mosquito larvaeNo treatment of similar breeding sites in a single village (92 houses) in Chanthaburi province.

House index, container index, Breteau index, per cent reduction (Mulla), average larva free period, landing and biting rates

F-test, t-test

Significant reductions (P < 0.05) in the intervention area were observed, with averages of:

HI = 69.8%

CI = 84.1%

BI = 84.4%

Landing rate = 73.9%

Biting rate = 73.6%

No similar trend observed in control area. Average larval free period was longest in drinking containers (16.4 + /− 2.5 weeks), followed by ant guard, washing and bathing containers

The Bti formulation was very practical and effective for the control of A. aegypti larvae in Thai communities. The intervention was readily accepted by the community
14.Tan et al. (2012)Shah Alam, MalaysiaComparing the efficacy of BTI (Vectobac) against Aedes albopictus and A. aegypti


1 year

Vectobac WG (3000 ITU/mg)

Valet Biosciences Corporation

One residential area (300 houses) received 7 biweekly cycles of Vectobac WG, followed by 7 weekly cycles, followed by 4 biweekly cycles in all mapped potential and actual outdoor larval habitatsOne residential area without BTI application. However, intensive pyrethroid fogging operation in study weeks 37–54 due to a dengue outbreak

Ovitrap index and larval density

Routine epidemiological data on dengue cases

Chi-square test for OI

t-test for larval density

OI was suppressed to below 10% and maintainted up to 4 weeks post-treatment

Outdoor OI remained at 40% in the untreated site (P < 0.05)

1 dengue case in the treated area

15 dengue cases in the untreated area

BTI application can suppress A. aegypti and Aedes albopictus populations at a dengue endemic site with a temephos-resistant population and potentially interrupt dengue transmission in humans

General study characteristics

All studies were published between 1993 and 2012. Ten of the studies were conducted in South-East Asia, with India being the most common location (1, 2, 3, 5 in Table 1). Three studies were conducted in South America (4, 6, 11 in Tables 1 and  2) and one in the Caribbean (10 in Table 1). In most studies, the setting was selected because of a high incidence of dengue or elevated pre-intervention entomological indices. No study incorporated economic analysis or provided cost data for consideration.

Study designs

Six studies were classified as effectiveness studies; the remaining eight examined the efficacy of Bti in more controlled settings. The most common study design was the NRCT, which was used in 10 studies. Three were CRCTs (4, 10, 11 in Tables 1 and  2), often considered the ‘gold standard for such research’, and 1 was a RCT (13 in Table 1). Eight studies used various breeding containers (water tanks, water jars, discarded tires, plants) as the unit of allocation, 4 used households (4, 8, 11, 12 in Table 2), and 2 used land areas (7 and 14 in Table 2). All studies incorporated a control group, the majority of which received only entomological surveillance. Two of the control groups received education and/or environmental management interventions (4, 11 in Table 2), and one control group received ‘standard control strategies’ consisting of weekly temephos treatment (7 in Table 2). Two control areas were unexpectedly treated with extensive space spraying mid-way through the study due to a dengue outbreak (8, 14 in Table 2).

The sample size varied significantly across studies. The smallest study (5 in Table 1) examined only five cement tanks in each of the two intervention groups compared to eight control tanks, while the largest study (4 in Table 2) incorporated more than 1000 households. The duration of follow-up ranged from 20 days (5 in Table 1) to 15 months (4 in Table 2) with a median follow-up of 6 weeks. Three studies (9, 10, 14 in Tables 1 and 2) pre-specified the period of follow-up.

Nearly all studies provided a rudimentary description of the study setting, often limited to geographical location and previous estimates of dengue incidence. Less commonly reported was information on potential confounding factors such as the socio-economic status of residents, housing construction and infrastructure. Weather conditions, either historical or during the intervention period, were reported in only five studies (4, 7, 10, 11, 12 in Tables 1 and 2).

Outcome measures

The most commonly reported outcome measures in the efficacy studies were larval and pupal densities, defined as the mean number of larvae or pupae per container, and per cent reduction in infestation using Mulla's formula (Mulla et al. 1971). Four studies (6, 9, 12, 14 in Tables 1 and 2) also monitored the ‘average larval free period’ or the amount of time after application of the intervention that potential habitats remained free from Aedes larvae. In the community effectiveness trials, oviposition indices and the Stegomyia indices (the house index (number of positive houses per 100 houses), container index (number of positive containers per 100 containers) and Breteau index (number of positive containers per 100 houses)) were more commonly reported (WHO/SEARO 1999). Only one study reported clinical outcomes from routinely collected epidemiological data (14 in Table 2).

Entomological surveillance protocols were clearly reported in 10 of the studies, while four studies (2, 4, 6, 13 in Tables 1 and 2) provided no or limited information regarding entomological sampling methods. Five studies (1, 2, 3, 5, 9 in Table 1) did not incorporate statistical precision estimates or significance tests in the reporting of outcomes.


A wide range of formulations and application methods were utilised in the studies. Seven studies (2, 4, 6, 9, 10, 11, 12 in Tables 1 and 2) incorporated slow-release Bti tablets or briquettes, seven studies (1, 3, 5, 7, 8 14 in Tables 1 and 2) used manual or motorised spray equipment to apply Bti, and three studies (2, 8, 13 in Tables 1 and 2) applied Bti granules or powder directly into positive breeding sites. The majority of the efficacy trials utilised only a single application of Bti in the study containers. In contrast, all of the effectiveness trials repeated application of Bti in differing frequencies as outlined in Table 2. Seven studies incorporated multiple intervention groups, comparing the effect of various formulations or dosages of Bti (1, 2, 5, 6 in Table 1) or comparing the effect of Bti to other insecticides (4, 10 in Tables 1 and 2) or lethal ovitraps (11 in Table 2) with a standard control. The most common insecticide of comparison was temephos, which was used in two of the studies (4, 10 in Tables 1 and 2).

Efficacy and effectiveness of Bti interventions

Each of the eight efficacy studies reported a post-intervention reduction in the observed density of immature stages compared to the respective control group. In general, these were non-randomised studies using containers as the unit of allocation. In the community-based effectiveness studies, the results were less clear, with 4 of the 6 studies (7, 8, 12, 14 in Tables 1 and 2) reporting significant but time-limited reductions in entomological indices. The two effectiveness studies that did not show significant reductions employed basic interventions such as environmental management (4 in Table 2) or general education on environment control practices (11 in Table 2), in the respective control groups. The other study that used standard dengue vector control practices as the control comparison (7 in Table 1), did, however, report a relative reduction in entomological indices in the Bti intervention area.

In the efficacy studies looking only at targeted breeding sites, Bti demonstrated a rapid killing effect, typically eliminating all larvae from treated containers within 24 h. The vast majority of containers remained free of larvae for the first 2 weeks of observation. However, reinfestation of some containers did occur within 7–9 days of treatment (5, 9 in Table 1). Only one study (3 in Table 1) reported 100% reduction in larval density for more than 4 weeks with a single application of Bti. Similar results were demonstrated in another study (1 in Table 1), but the study period was limited to 4 weeks.

Five studies examined the effect of repeated Bti treatments (7, 8, 11, 12, 14 in Table 2). Two of these studies sprayed Bti every 2 weeks (7, 8 in Table 2), in 1 study, Bti briquettes were replaced every month (11 in Table 2), in another study, Bti briquettes were replaced when the containers were reinfested (12 in Table 2), and in 1 study (14 in Table 2), Bti was sprayed in biweekly and weekly cycles. In four of the studies (7, 8, 12, 14 in Table 2), the repeated application of Bti at various frequencies resulted in significant (P < 0.05) reductions in the density of immature stages compared to the control over the course of treatment. Each of these four studies also examined the effect on adult dengue vector populations, either through ovitrap indices or landing and biting rates. In each study, a decline in the adult mosquito population was observed, typically only after the second application of Bti, which occurred in the fourth to sixth week of the trial. This delayed effect is thought to be due to the survival of already existing adult mosquitoes, which would not have been affected by larvicidal applications of Bti. The study by Lee et al. (8 in Table 2) showed a rebound in the adult and larval indices within 6 weeks of cessation of treatment. Although the study by Ocampo et al. (11 in Table 2) did not observe any significant reductions in entomological indices, only one Bti-treated container was ever found to be positive for mosquito larvae.

The four studies that compared different formulations of Bti (1, 2, 5, 6 in Table 1) did not suggest evidence of superior efficacy of any one commercial product. The results did, however, suggest that higher doses of Bti provide a longer duration of effect (1, 2, 6 in Table 1). The per cent reduction in the density of immature stages at 4 weeks nearly doubled with a proportional increase in Bti concentration in these three studies. The study by Ansari and Razdan (1 in Table 1) demonstrated the decreasing marginal benefit of this approach, as both of the higher concentrations used (0.5 and 1.0 g/m2) achieved similar results at 4 weeks, suggesting that a similar effect could be achieved with less cost at the lower dose.

In the two studies where Bti formulations were compared with other insecticides (4, 10 in Tables 1 and 2), the results were mixed. The study by Favier et al. (4 Table 2) found no significant reductions attributable to any of the interventions, which included Bti, temephos and methoprene-S. Here, the authors concluded that insecticides may not improve on environmental control practices, which served as the control comparison. In contrast, the study by Marcombe et al. (2011) showed that both Bti and pyriproxyfen were effective in initially lowering larval densities but had a notably lower residual activity compared to diflubenzuron and spinosad.

Only one study (14 in Tables 1 and 2) linked routinely collected epidemiological data to the entomological data. When a dengue outbreak occurred in the study area, only one case was reported in the Bti intervention area and 15 cases were reported in the control area, suggesting that the lower mosquito densities in the Bti intervention area could have had a protective effect.

Acceptability of Bti

Very few of the included studies formally evaluated the acceptability of Bti application. The study by Haq et al. (5 in Table 1) surveyed field staff and noted some operational challenges and reports of skin irritation after contact with Bacticide. Lee et al. (8 in Table 2) monitored for potential effects on non-target organisms, but did not observe any undesirable consequences. Ocampo et al. (11 in Table 2) reported that the Bactimos briquettes left a visible residue on water storage containers and speculated that this may have induced residents to clean the containers more often.

In general, most studies reported that the Bti intervention was readily accepted and even preferred over other measures such as temephos applications because it did not affect water quality or taste. The study by Ocampo et al. (11 in Table 2) was the only one to quantify acceptance, reporting that 40% of the houses did not routinely use the briquettes, because householders did not like the residue the briquettes left in their water.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Cited References

The evidence presented from the efficacy studies suggests that Bti can be effective in controlling the immature stages of dengue vector mosquitoes in a variety of breeding sites. The killing effect is rapid, typically eliminating all immature forms from treated containers within 24 h, with a residual effect that ranges between 2 and 4 weeks. These results, however, only suggest that Bti is effective in specifically targeted containers that actually receive treatment. Given the large number of potential habitats, the widespread application of Bti to all potential breeding sites may not be practical.

A better estimate of the overall impact of Bti as single agent is provided by the effectiveness studies (4, 7, 8, 11, 12, 14 in Table 2). The findings from these studies are mixed and do not provide conclusive evidence that Bti, when used a single agent, produces significant reductions in entomological indices. There are a number of reasons that might explain the lack of effect. First, investigators may have failed to identify and treat all potential breeding sites within each household. Thus, while Bti may have achieved significant reductions in treated containers, the existence of even a few untreated and productive breeding sites may have masked the effect of Bti. The study by Ocampo et al. (11 in Table 2) highlights this phenomenon, where over the course of the study, only one of 76 water storage tanks that were treated with Bti was ever positive for mosquito larvae. However, untreated smaller containers and containers holding aquatic plants were often positive for larvae.

The comparison of experimental results with untreated control groups is also important in determining the impact of Bti as a sole control agent. In all studies where individual containers were the unit of allocation, the control containers received no treatment. However, in 2 of the 4 household and community-level studies (4, 11 in Table 2), the control group received an intervention in the form of education and/or environmental management. Both these studies found no significant differences between the Bti intervention groups and the control groups, suggesting that Bti alone provided an equivalent level of control as what was achieved through education and/or environmental management.

The study by Ocampo et al. (11 in Table 2) also raised the issue of buffer zones. While it may be possible to greatly reduce breeding sites within targeted residential areas, treatment must also extend beyond these areas to prevent the immigration of adult mosquitoes from non-treated areas. The flight range of A. aegypti is restricted: mosquitoes rarely disperse further than 100 m from their emergence location (Muir & Kay 1998; Harrington et al. 2005). Despite this limited flight range, Ocampo et al. (11 in Table 2) did not observe a reduction in the presence of adult mosquitoes in the intervention households. This result suggests that their buffer zone, which consisted of treatment of one house beyond the intervention block in every direction, was insufficient. Lee et al. (8 in Table 2) experienced a similar challenge because they were unable to treat a factory that was adjacent to one of their study sites. While they still observed a reduction in the ovitrap index at this location, the magnitude of the effect was less pronounced than that of another site, which was not adjacent to the untreated factory.

The reviewed studies did not provide a generalisable estimate of community acceptance and uptake of the Bti interventions. In the majority of the studies, vector surveillance and Bti application were carried out by the investigators rather than vector control authorities or members of the community. Given the limits on national budgets for vector control programmes, it is unlikely that routine control efforts will achieve levels of intensity comparable to those obtained in these studies.

The majority of the studies failed to assess and compare baseline characteristics between the control and intervention groups. Additionally, five studies (1, 2, 3, 5, 9 in Table 1) did not incorporate any statistical methods into their data analyses for the purpose of estimating precision or comparing results between groups. Without this information, it is difficult to interpret the significance of the reported results.

Finally, only one study (14 in Table 2) linked routinely collected epidemiological data from both the Bti intervention and control sites. Although fewer dengue cases were reported in the intervention area, no information was available regarding the method or quality of routine epidemiological data collection or possible sites of dengue infection outside of the study area.

Publication bias should be considered as a final limitation of this study, as likely more studies with a positive outcome are reported in the literature. The diversified search strategy limits this publication bias; however, it cannot be completely eliminated.

In summary, there is evidence that Bti is effective in reducing the density of immature dengue vectors when it is applied to targeted containers as demonstrated by the efficacy studies. However, the evidence to suggest that Bti is effective as a single agent, when used in a community setting, is limited. Given the increasing prevalence of insecticide resistance in dengue vectors in many parts of the world, understanding the control implications of using alternatives to chemical insecticides such as Bti is becoming increasingly important. However, there is a clear need for further studies that utilise cluster-randomised controlled designs to investigate the efficacy and effectiveness of Bti and to further link entomological outcomes to dengue transmission measures.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Cited References
  • Goldberg L & Margalit J (1977) A bacterial spore demonstrating rapid larvicidal activity against Anopheles sergentii, Uranotaenia unguiculata, Aedes aegypti, Culex pipiens, Culex univitattus. Mosquito News, 37, 355358.
  • Harrington C, Scott T, Lerdthusnee K et al. (2005) Dispersal of the dengue vector Aedes aegypti within and between rural communities. American Journal of Tropical Medicine and Hygiene 72, 209220.
  • Harris A, Rajatileka S & Ranson H (2010) Pyrethroid resistance in Aedes aegypti from Grand Cayman. American Journal of Tropical Medicine and Hygiene 83, 277284.
  • Lacey L (2007) Bacillus thuringiensis serovariety israelensis and Bacillus sphaericus for mosquito control. Journal of the American Mosquito Control Association 23, 133163.
  • Lacey L & Mulla M (1990) Safety of Bacillus thuringiensis (H-14) and Bacillus sphaericus to non-target organisms in the aquatic environment. In: Safety of Microbial Insecticides (eds M Laird, L Lacey & E Davidson) CRC Press, Boca Raton, pp. 169188.
  • Lee B & Scott G (1989) Acute toxicity of temephos, fenoxycarb, diflubenzuron, and methoprene and bacillus thuringiensis var. israelensis to the mummichog (Fundulus heteroclitus). Bulletin of Environment Contamination and Toxicology 43, 827832.
  • Liberati A, Altman D, Tetzlaff J et al. (2009) The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Medicine 6, e1000100.
  • Merritt R, Walker E, Wilzbach M, Cummins K & Morgan W (1989) A broad evaluation of B.t.i. for black fly (Diptera: Simuliidae) control in a Michigan river: efficacy, carry and nontarget effects on invertebrates and fish. Journal of the American Mosquito Control Association 5, 397415.
  • Mittal P (2003) Biolarvicides in vector control: challenges and prospects. Journal of Vector Borne Diseases 40, 2032.
  • Muir L & Kay B (1998) Aedes aegypti survival and dispersal estimated by mark-release-recapture in northern Australia. American Journal of Tropical Medicine and Hygiene 58, 277282.
  • Mulla M, Norland R, Fanara D, Darwazeh H & Mckean D (1971) Control of chironomid midges in recreational lakes. Journal of Economic Entomology 64, 300307.
  • Rodriguez M, Bisset J, De Armas Y & Ramos F (2005) Pyrethroid insecticide-resistant strain of Aedes aegypti from Cuba induced by deltamethrin selection. Journal of the American Mosquito Control Association 21, 437445.
  • Saik J, Lacey L & Lacey C (1990) Safety of microbial control agents to domestic animals and vertebrate wildlife. In: Safety of Microbial Insecticides (eds M Laird, L Lacey & E Davidson) CRC Press, Boca Raton. pp. 115131.
  • WHO/TDR (2007) Special Programme for Research and Training in Tropical Diseases/World Health Organization. Report of the Scientific Working Group meeting on dengue. Dengue Bulletin, 31, 183186.
  • WHO/TDR (2007) Special Programme for Research and Training in Tropical Diseases/World Health Organization. Dengue: Guidelines for Diagnosis, Treatment, Prevention, and Control. World Health Organization, Geneva.
  • WHO/SEARO (1999) Prevention and Control of Dengue and Dengue Haemorrhagic Fever: Comprehensive Guidelines. WHO Regional Office for Southeast Asia, New Delhi.

Cited References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  8. Cited References
  • Ansari M & Razdan R (1999) Laboratory and field evaluation of Bacillus thuringiensis H-14 (Bt. H-14) granule formulation against Aedes aegypti in Delhi, India. Dengue Bulletin 23, 9498.
  • Batra C, Mittal P & Adak T (2000) Control of Aedes aegypti breeding in desert coolers and tires by use of Bacillus thuringiensis var. israelensis formulation. Journal of the American Mosquito Control Association 16, 321323.
  • Dua V, Sharma S & Sharma V (1993) Application of Bactoculicide (Bacillus thuringiensis H-14) for controlling mosquito breeding in industrial scrap at BHEL, Hardwar (U.P.). Indian Journal of Malariology 30, 1721.
  • Favier C, Degallier N, Vilarinhos P, De Carvalho M, Yoshizawa M & Knox M (2006) Effects of climate and different management strategies on Aedes aegypti breeding sites: a longitudinal survey in Brasilia (DF, Brazil). Tropical Medicine, International Health 11, 11041118.
  • Haq S, Bhatta R, Vaishnavb K & Yadava R (2004) Field evaluation of biolarvicides in Surat city, India. Journal of Vector Borne Diseases 41, 6166.
  • Kroeger A, Dehlinger U, Burkhardt G, Athehortua W, Anaya H & Becker N (1995) Community based dengue control in Columbia: People's knowledge and practice and the potential contribution of the biological larvicide Bti (Bacillus thuringiensis israelensis). Tropical Medicine and Parasitology 46, 241246.
  • Lam P, Boon C, Yng N & Benjamin S (2010) Aedes albopictus control with spray application of Bacillus thuringiensis israelensis, strain AM 65–52. Southeast Asian Journal of Tropical Medicine and Public Health 41, 10711081.
  • Lee H, Chen C, Masri S, Chiang Y, Chooi K & Benjamin S (2008) Impact of larviciding with a Bacillus thuringiensis israelensis formulation, VectoBac WG, on dengue mosquito vectors in a dengue endemic site in Selangor State, Malaysia. Southeast Asian Journal of Tropical Medicine, Public Health 39, 601609.
  • Mahilum M, Ludwig M, Madon M & Becker N (2005) Evaluation of the present dengue situation and control strategies against Aedes aegypti in Cebu City, Philippines. Journal of Vector Ecology 30, 277283.
  • Marcombe S, Darriet F, Agnew P et al. (2011) Field efficacy of new larvicide products for control of multi-resistant Aedes aegypti populations in Martinique (French West Indies). American Journal of Tropical Medicine, Hygiene 84, 118126.
  • Ocampo C, Gonzalez C, Morales C, Perez M, Wesson D & Apperson C (2009) Evaluation of community-based strategies for Aedes aegypti control inside houses. Biomedica 29, 282297.
  • Phan-Urai P, Kong-Ngamsuk W & Malainual N (1995) Field trial of Bacillus thuringiensis H-14 (Larvitab®) against Aedes aegypti larvae in Amphoe Khlung, Chanthaburi Province, Thailand. Journal of Tropical Medicine, Parasitology 18, 3541.
  • Sulaiman S, Pawanchee Z, Wahab A, Jamal J, Sohadi A & Rahman A (1999) Evaluation of Abate 1-SG and Vectobac G in bromeliads infested with dengue vector Aedes albopictus (Skuse) in Kuala Lumpur, Malaysia. Medical Entomology and Zoology 50, 165167.
  • Tan A, Loke SR, Benjamin S, Lee HL, Chooi KH & Sofian-Azirun M (2012) Spray application of Bacillus thuringiensis israelensis (Bti Strain 65–52) against Aedes aegypti (L) and Ae. albopictus Skuse populations and impact on dengue transmission in a dengue endemic residential site in Malaysia. Southeast Asian Journal of Tropical Medicine and Public Health 43, 296310.