Background
Most malaria-endemic countries have adopted policies to promote universal distribution of long-lasting insecticidal nets (LLINs) free of charge across all age groups, and an estimated 49% of the population in sub-Saharan Africa had access to at least one LLIN in their household in 2013 [
1]. However, resistance to pyrethroid insecticides used in all LLINs is now widespread across vector populations and may reduce the level of community protection [
2,
3]. Another challenge is maintaining effective year-round protection as during the hot dry seasons when transmission may still occur, some individuals are deterred from sleeping under nets [
4]. The development of large holes through wear and tear of net fabrics during normal household use may compromise their protective efficacy despite LLINs retaining insecticidal potency for three years [
5]. In areas of hyper-endemic malaria transmission, even when universal coverage (UC) of LLINs is achieved and nets are in good condition, malaria prevalence can remain relatively high unless additional control tools are implemented [
6].
Indoor residual spraying (IRS) is a proven vector control method that has been used since the Second World War and was the central feature of the Global Malaria Eradication Campaign between 1955 and 1969, which successfully eliminated malaria from several countries and significantly reduced disease incidence in others [
7,
8]. In 2005 the US President’s Malaria Initiative (PMI) revived IRS in sub-Saharan Africa by funding an initial $1.2 billion programme in 15 countries [
9]. IRS coverage in sub-Saharan Africa increased substantially from <2% of the at-risk population protected in 2005 to 11%, or 78 million people, by 2010 [
1]. A major challenge facing IRS programmes is how to sustain such gains in the face of operational problems, such as vector resistance to insecticides, lack of affordable alternative insecticides and limited resources for recurrent annual campaigns. In addition to the labour costs associated with spraying, high levels of pyrethroid resistance among mosquito populations have necessitated the use of more expensive non-pyrethroid formulations [
10]. The commodity cost of PMI-supported IRS campaigns with the organophosphate Actellic CS 300 (pirimiphos methyl CS) has been estimated to be more than four times the expense of using the pyrethroid Icon CS 10 (lambdacyhalothrin CS) to cover the same area [
11].
A new product has been developed which mimics the effect of IRS but is designed to control insecticide-resistant mosquitoes for a minimum of three years. Insecticide-treated, durable wall lining (ITWL) is a material that can be fixed to the inner walls and ceilings of houses. The principle is the same as IRS, to kill mosquitoes that land on the ITWL either before or after blood-feeding. If the coverage of ITWL is high enough, the population density and longevity of mosquitoes in the area becomes substantially reduced, together with malaria transmission. There is also the possibility that ITWL could be used to block entry of mosquitoes if the material is extended from floor to ceiling, therefore covering eave spaces. While ITWL, like IRS, could be used to reduce transmission by itself, it is more likely to be an adjunct to LLINs, with the LLIN providing additional personal protection through the barrier and excito-repellent effect. ZeroVector
® is a first-generation ITWL containing deltamethrin incorporated into high-density polyethylene shade cloth, which has been evaluated in several countries in sub-Saharan Africa and Asia and consistently received high levels of household acceptability and provided prolonged insecticidal activity of greater than 12 months [
12,
13]. With pyrethroid resistance now widespread throughout sub-Saharan Africa, attention has switched to development of a new generation of non-pyrethroid ITWL.
Initial experimental trials of ITWL + LLINs were conducted using plastic sheeting that had been spray treated with a non-pyrethroid (organophosphate) insecticide [
14,
15]. These studies produced differing results in Côte d’Ivoire and Burkina Faso, which were attributed to variation in phenotypic resistance to organophosphates and pyrethroids among the respective vector populations [
14,
16]. A newer factory-produced product (PermaNet
® Lining) has been developed that consists of thin non-woven sheets of cloth made from high-density polypropylene containing a non-pyrethroid insecticide mixture of abamectin (avermectin) and fenpyroximate (pyrazole) which are slowly released together and migrate to the surface of the fibre; neither insecticide has been used in malaria control before. Abamectin is a macrocyclic lactone that acts through chloride channel activation and was discovered in 1981 [
17]. Contact bioassays have shown that abamectin is efficacious against house flies [
18], cockroaches [
19] and fire ants [
20] in terms of mortality, and there is evidence for oviposition suppression in blowflies [
21]. There are limited data for use against mosquitoes, but ivermectin (also an avermectin) is highly effective in terms of both mortality and oviposition suppression as a cattle parasiticide [
22]. Abamectin is widely used in mixtures for control of crop and ornamental pests associated with greenhouse and nursery operators, e.g., abamectin + trifosine (fungicide) is used to control the two-spotted spider mite [
23]. Fenpyroximate is a pyrazole in the mitochondrial complex 1 electron transport inhibitors (METI) group of insecticides, which disrupt insect respiration and are in widespread use globally. METI acaricides are extensively used to control
Tetranychus spp (spider mites) [
24]. To date, there are no published data demonstrating efficacy of these proprietary insecticides against mosquitoes.
Despite the promise of ITWL, the only existing data to support the efficacy of this new product are small-scale, unpublished studies conducted by the manufacturer. In response to the increasing problem of insecticide resistance, PMI has funded a large-scale, cluster-randomized controlled trial (CRT) in Muheza, Tanzania to investigate whether ITWL combined with UC of LLINs provides added protection against malaria compared with LLINs alone [
25]. To aid interpretation of the CRT results, an experimental hut trial of the ITWL with or without LLIN was conducted in Muheza against wild free-flying populations of
Anopheles funestus sensu stricto (s.s.) and
Anopheles gambiae sensu lato (s.l.).
Discussion
At present there are no recognized WHO standards for ITWL products. Before progressing to community trials, candidate IRS and LLIN products are evaluated in Phase II experimental huts against an existing gold standard positive control and comparative performance assessed [
29,
34]. In this trial the positive control was the WHOPES-recommended pyrethroid LLIN, which produced equivalent levels of mortality to the non-pyrethroid ITWL. However, due to pyrethroid resistance in both of the major vector species, the level of mortality for the LLIN was lower than previous reports from the same site [
28]. In 2006 when vector species were fully susceptible to pyrethroids, Interceptor
® LLIN (alphacypermethrin) killed 92% of
An. gambiae s.l. when unwashed [
35], while Olyset
® LLIN (permethrin) produced 61% mortality for
An. gambiae s.l. and 72% for
An. funestus s.s. [
36]. The mortality rate with Interceptor
® LLIN in this trial was considerably lower at just 26% for
An. gambiae s.l. and 29% for
An. funestus s.s. and was due to the significant increase in resistance across local vector populations [
26]. In this study the non-pyrethroid ITWL also elicited relatively low levels of mortality, between 40 and 50% for
An. funestus s.s. and
An. gambiae s.l. The level of mortality in free-flying mosquitoes remained consistent over the 7 weeks of the trial.
WHO cone and cylinder bioassays exposing laboratory-reared susceptible and wild resistant
An. gambiae s.l. to new pieces of non-pyrethroid ITWL for 30 min produced 90–100% mortality, demonstrating the inherent toxicity of the ITWL insecticides, with no evidence for cross-resistance to pyrethroids. In both assays, mosquito mortality remained low at 24 h, reaching the highest levels after 72 h. While new formulations would ideally produce more immediate mortality, a similar phenomenon has been reported for chlorfenapyr, a pyrrole insecticide which has demonstrated promise as an IRS and net treatment [
37,
38]; historically some organochlorines used successfully as IRS (e.g., dieldrin) were also characterized by delayed mortality.
To help interpret the low levels of mortality observed in the main trial, WHO cylinder assays with incremental exposure times were performed to assess the duration of contact time required to kill the field strain (F1 progeny) of An. gambiae s.l. Only 7.5–15 min of contact with the non-pyrethroid ITWL induced 80–100% mortality. In all assays, levels of mosquito mortality were comparable between different rolls of ITWL, excluding differences in manufacturer batch production as a potential confounder in the main trial.
With the demonstration of high mortality in the bioassays, the lower levels in the main trial may be attributable to wild, free-flying mosquitoes spending less time in contact with treated wall surfaces. Shorter resting times are expected when insecticides induce a degree of repellence or irritancy. To further characterize mosquito behaviour in relation to ITWL exposure, time to first flight was measured for An. gambiae Kisumu and F1 wild An. gambiae s.l. in comparison to untreated netting and a pyrethroid LLIN (Interceptor® LLIN). No significant irritability was observed on exposure to the non-pyrethroid ITWL, unlike the pyrethroid LLIN which is known to have excito-repellent properties. These results were supported by the low exiting rates of both An. funestus s.s. and An. gambiae s.l. in non-pyrethroid ITWL huts compared to pyrethroid treatments (Interceptor® LLIN and ZeroVector® ITWL) in the main trial.
In the main trial, only treatments containing pyrethroid interventions significantly reduced vector entry. By comparison, the low levels of deterrence in the non-pyrethroid ITWL huts and the supporting irritability bioassays indicate that the ITWL was not influencing mosquito entry. One possible explanation for the low mortality, exiting and deterrence in the non-pyrethroid ITWL huts, is that vectors were instead resting on the hessian sack cloth ceilings, which remained uncovered throughout the trial, and were not contacting the walls for sufficient time to obtain a lethal insecticide dose. During the trial no data were documented on the location of mosquitoes within the room. While few studies have characterized exactly where African malaria mosquitoes reside within experimental huts, a recent trial of pirimiphos-methyl IRS on wooden panelled walls performed at the same study site also reported unexpectedly low levels of mortality and mosquitoes were noted to be resting on sack cloth-lined ceiling (M. Rowland, unpublished data). Earlier experimental hut trials of insecticide-treated wall lining materials have demonstrated that efficacy is strongly correlated with intervention surface area, with increasing coverage affording higher rates of vector mortality, deterrence and blood-feeding inhibition [
39,
40]. If in this trial mosquitoes secured refuge on the ceiling over the non-woven polyethylene wall lining material, this could partly explain the low mortality in the huts. During IRS campaigns in Tanzania, ceilings are not usually sprayed, being too high and inaccessible to spray men, and yet such campaigns can have had a major effect on
An. gambiae population density and malaria transmission rates [
6].
It has been suggested that ITWL may also impact malaria transmission by functioning as a method of housing improvement, if used to cover eave gaps, preventing mosquito ingress [
25]. However, in the present trial, there was no significant differences in vector entry in huts with partially blocked or open eaves. It is possible that host-seeking mosquitoes are able to compensate for a partial restriction of ‘entry points’ if host odour is concentrated from those that remain. ITWL material is not currently designed to act an eave seal or withstand strong winds and house improvements would better look to other means of restricting access.
Conclusions
PermaNet® ITWL is a new malaria vector control strategy, containing two non-pyrethroid insecticides, which is designed to function as a long-lasting IRS, when fixed to the inner walls and ceiling of houses. An experimental hut trial was performed in Muheza, Tanzania, to evaluate performance during 2 months after installation in comparison with a WHOPES-recommended LLIN (Interceptor®) and pyrethroid ITWL (ZeroVector®). The level of mosquito mortality was lower than expected: in the pyrethroid-treated LLIN and pyrethroid ITWL this was explained by insecticide resistance; in the non-pyrethroid ITWL this was attributed to low efficacy; since the ceiling was uncovered some mosquitoes may have secured refuge on this untreated surface. A series of novel, supplementary bioassays characterized the toxicity and mode of action of the non-pyrethroid ITWL product, and its effect on mosquito behaviour. The findings represent the first 2 months after installation and do not necessarily predict longer-term residual efficacy of non-pyrethroid ITWL.
Authors’ contributions
RM, RMO, LAM, FWM, MWR, and WK designed the study and were responsible for data analysis and interpretation. RM, BB, WS, BE, and SW led the entomology field activities and laboratory assays and participated in data collection. GM and JPM were responsible for oversight, management and delivery of the trial. LAM, RMO and MWR drafted the manuscript which was revised by co-authors. All authors read and approved the final manuscript.