Three years of insecticide resistance monitoring in Anopheles gambiae in Burkina Faso: resistance on the rise?

Abstract

Background and methods: A longitudinal Anopheles gambiae s.l. insecticide-resistance monitoring programme was established in four sentinel sites in Burkina Faso. For three years, between 2008 and 2010, WHO diagnostic dose assays were used to measure the prevalence of resistance to all the major classes of insecticides at the beginning and end of the malaria transmission season. Species identification and genotyping for target site mutations was also performed and the sporozoite rate in adults determined.

Results: At the onset of the study, resistance to DDT and pyrethroids was already prevalent in An. gambiae s.l. from the south-west of the country but mosquitoes from the two sites in central Burkina Faso were largely susceptible. Within three years, DDT and permethrin resistance was established in all four sites. Carbamate and organophosphate resistance remains relatively rare and largely confined to the south-western areas although a small number of bendiocarb survivors were found in all sites by the final round of monitoring. The ace-1R target site resistance allele was present in all localities and its frequency exceeded 20% in 2010 in two of the sites. The frequency of the 1014F kdr mutation increased throughout the three years and by 2010, the frequency of 1014F in all sites combined was 0.02 in Anopheles arabiensis, 0.56 in An. gambiae M form and 0.96 in An. gambiae S form. This frequency did not differ significantly between the sites. The 1014S kdr allele was only found in An. arabiensis but its frequency increased significantly throughout the study (P = 0.0003) and in 2010 the 1014S allele frequency was 0.08 in An. arabiensis. Maximum sporozoite rates (12%) were observed in Soumousso in 2009 and the difference between sites is significant for each year.

Conclusion: Pyrethroid and DDT resistance is now established in An. gambiae s.l. throughout Burkina Faso. Results from diagnostic dose assays are highly variable within and between rounds of testing, and hence it is important that resistance monitoring is carried out on more than one occasion before decisions on insecticide procurement for vector control are made. The presence of 1014S in An. gambiae s.l., in addition to 1014F, is not unexpected given the recent report of 1014S in Benin but highlights the importance of monitoring for both mutations throughout the continent. Future research must now focus on the impact that this resistance is having on malaria control in Burkina Faso.

Figure 1

Map showing geographical locations of the study sites and the percentage of adult mosquitoes positive for sporozoites (Y axis) during years of collection (X axis) of the study (data for Goundry 2010 not available).

Figure 2

Insecticide bio-assay results for Anopheles gambiae s.l from 2008–2010 in two rounds of monitoring in four sentinel sites in Burkina Faso (data from 2008 have been already published [6]and are included for the purpose of comparison). The percentage mortalities 24 hours following a one-hour exposure to the WHO diagnostic dose of insecticide (with 95 % CI) are shown. The minimum sample size for each test was 100 non-blood fed female mosquitoes, three to five days old.

Figure 3

Species identification within Anopheles gambiae s.l. for the four sentinel sites in 2008–2010. Data are presented as proportions of the total for each species, An. arabiensis (A), An. gambiae M form (M), An. gambiae S form (S) and the hybrid of M and S forms (H), by year and by site. Sample sizes are minimum of 75 per site for 2008 and 2009 and a minimum of 309 for 2010.

Figure 4

The percentage survival of Anopheles gambiae s.l. species/form 24 hours following exposure to DDT or permethrin. Data for 2010 and are stratified by species for each sentinel site. Statistical comparison between molecular forms and localities were performed using Z-test with significant differences (P < 0.05) showed in the figure (*).

Figure 5

The allele frequencies of 1014F (blue) and 1014S (red) by species and site. No statistical difference is observed in the allelic frequencies between sites for each of the species/molecular form for the 1014F mutation (P > 0.08). Differences between sites are significant when comparing for 1014S allele in Anopheles arabiensis.