Failure of Permethrin-Treated Military Uniforms to Protect Against a Laboratory-Maintained Knockdown-Resistant Strain of Aedes aegypti

The Department of Defense (DoD), like other militaries, implements the Insect Repellent System where there is a risk of acquiring vector-borne disease (AFPMB 1996). The wearable system has two major components: a permethrin-impregnated battle-dress uniform and the application of N,N-diethyl-meta-toluamide (deet) or picaridin-based repellent (minimum 30%) applied to exposed skin. Factory treatment with permethrin (0.52% w/w) became standard for US Army uniforms beginning in 2013. Many laboratory and field studies have demonstrated that the combined components are effective against mosquitoes and other arthropod vectors of human disease (Faulde et al. 2006, Lindsey and Logan 2015). However, the proper use of both components is critical to system efficacy (Eamsila et al. 1994).

Although the DoD Insect Repellent System has demonstrated efficacy, the effectiveness of the system was not originally tested with pyrethroid-resistant mosquitoes. Pyrethroid resistance has become increasingly common, and Aedes aegypti (L.), the primary vector of several arboviruses, has demonstrated widespread resistance, largely due to heritable genetic knockdown resistance (kdr) mutations (Smith et al. 2016). Agramonte et al. (2017) showed a significant increase in blood feeding when the strongly permethrin-resistant Ae. aegypti Puerto Rico (PR) strain, with high levels of kdr, was exposed to a dose range of permethrin-impregnated sleeves. In contrast, a more recent study by Bowman et al. (2018) reported no loss of efficacy when permethrin-treated uniform material was exposed to a different strain of “resistant” Ae. aegypti from Puerto Rico. We note that neither study tested the efficacy of repellent skin treatment with the resistant mosquitoes.

To address these differing results and determine the impact of permethrin-resistant mosquitos on efficacy, sleeve tests were performed with treated and untreated Fire-Retardant Army Combat Uniform (FRACU) Type III fabric at the DoD standard concentration. Limited arm-in-cage studies were also performed to examine the efficacy of deet against pyrethroid-resistant and pyrethroid-susceptible Ae. aegypti.

The Ae. aegypti adults used in this study were the well-characterized Orlando (ORL) 1952 susceptible strain and the PR pyrethroid-resistant strain (provided by Centers for Disease Control and Prevention [CDC] for distribution by BEI Resources, eggs, NR-48830), reared as previously published (Agramonte et al. 2017, Estep et al. 2017). We conducted allele-specific kdr genetic testing (n = 32 individual females) as well as the CDC bottle bioassay using the standard dose of 43 μg/bottle (n = 3) as described in the latest CDC protocol (CDC 2016, Estep et al. 2018) to verify resistance levels. Results indicated the strains were as previously described, with no kdr mutations and permethrin susceptibility in the ORL1952 strain (100% mortality < 15 min). The resistant PR strain was homozygous for both the 1016 and 1534 kdr mutations (genotype IICC) and exhibited no mortality at the CDC diagnostic time.

While no unified standard has been formulated by World Health Organization (WHO) for uniform or wearable fabric efficacy certification procedures, the US Environmental Protection Agency (EPA) has approved testing protocols for use during good laboratory practice (GLP) registration studies by the US Department of Agriculture (EPA 2014). This present study used these approved methods but not the number of replicates required for GLP registration because we were not registering a product and desired to minimize the number of bites received by the study volunteers. Uniform sleeves for this study were produced and treated with 0.52% w/w technical-grade permethrin according to Agramonte et al. (2017) and allowed to dry for 15 min at room temperature. Sleeves were individually stored in sealable bags at room temperature (22.5 ± 2°C), under standard laboratory fluorescent lighting, and used multiple times over the 6-month testing period to simulate the aging of a military uniform. Following the published protocols for arm-in-cage tests, both arms, with tightly fitted treated and untreated sleeves, were simultaneously inserted into separate cages of 150–200 host-seeking females selected by attraction to human odor (EPA 2014, Agramonte et al. 2017). Mosquitoes were then mechanically aspirated from cages after the 15 min exposure period, anesthetized by CO₂, and the visibly blood-fed mosquitoes from each treatment were separated and counted. The remaining mosquitoes were separated and then crushed between 2 pages of white paper to identify individuals with partial blood meals, which indicated bite-through, and were scored as blood fed. Assays were performed using the same control and permethrin-treated sleeves on 3 volunteers, but mosquitoes were used for only a single exposure. Protocols were approved by the University of Florida Institutional Review Board. Analysis consisted of an initial ANOVA and subsequent Tukey's multiple comparisons as implemented in Prism7 (GraphPad Software, San Diego, CA).

This testing indicated no significant differences in blood-feeding success between the ORL1952 (78 ± 6%, mean ± SD) and PR strains (74 ± 12%) on the untreated uniform fabric (Fig. 1A). Confirming previous work, the treated sleeves were highly effective against the susceptible ORL1952 strain with bite-through to 2 ± 1% (Miller et al. 2004, Agramonte et al. 2017). In contrast, the PR strain bite-through was not significantly reduced by pretreatment of the FRACU Type III material with permethrin (mean = 53 ± 8%). While additional replicates may have been able to detect a smaller significant effect size between the strains, 53% bite-through is still a failure to protect the wearer. During the exposure periods, numerous mosquitoes were observed resting on the fabric after taking a full blood meal without common indicators of pyrethroid toxicity such as leg autotomy or irritation that would tend to make them leave the treated surface (Fig. 1B).

The repellent efficacy of deet was measured using the same 2 mosquito strains and WHO testing protocols (WHO 2009). Complete protection time was measured to estimate the amount of time the repellent chemical (deet) protects an individual from susceptible and resistant strain bites. Each volunteer's hands were double gloved, and 1 g of 40% deet formula (Repel Insect Repellent Sportsmen MAX Formula Lotion, WPC Brands, Inc., Bridgeton, MO) was uniformly applied to approximately 600 cm² of 1 forearm from wrist to elbow, which was inserted into a cage of 150–200 female Ae. aegypti for 3 min. The exposure was repeated every 30 min for the first 2 h and then hourly until 5 h postapplication. This limited evaluation of repellency by arm-in-cage testing with deet indicated test organisms were repelled consistently through 5 h with only 1 volunteer receiving a single bite from each strain (2 h, ORL1952; 3 h, PR). The other 2 volunteers received no bites. We were encouraged that the deet lotion provided nearly complete protection from host-seeking females for at least the timeframe tested here, even with the resistant PR strain.

In the present preliminary study, permethrin pretreatment of military-grade fabric provided no significant bite protection against a strongly pyrethroid-resistant PR strain at the DoD standard dose. This is consistent with the findings of Agramonte et al. (2017), which required a greater than 60-fold increase in permethrin to reach the same effective dose as with the ORL1952 strain.

The results of this study and Agramonte et al. (2017) seem at odds with the recent report from Bowman et al. (2018). We suggest that the results presented in Bowman are as expected for the strains they described. They state that “no mosquitoes were homozygous for kdr mutations at both SNPs,” thus their “resistant” PR strain was not analogous to the PR strain used in this study or in Agramonte and, as they also noted, was possibly more susceptible than assumed. This demonstrates the importance of verifying the toxicology profile of strains used in comparative exposure tests, whether arm-in-cage studies with human volunteers or static assays (Agramonte et al. 2017, Richards et al. 2018), as a standard good laboratory practice.

We note that these results are limited because we did not examine the efficacy against resistant malaria vectors or Culex species, so we would be hesitant to generalize these findings. However, the data presented here, along with declines in efficacy noted by Eamsila and Agramonte, call into question the effectiveness not only of permethrin-treated military uniforms but by extension also that of permethrin-treated outdoor gear, which has been available since 2003. The modern environment of increasing pyrethroid resistance in numerous arthropod vectors should lead scientists and risk managers to reevaluate the utility of previous studies conducted with only pyrethroid-susceptible mosquitoes. Understanding the efficacy of the DoD Insect Repellent System is essential to defining the risk posed by permethrin-resistant vectors and assessing whether further mitigation is necessary.

The authors thank Erica Lindroth, Erin Wilfong, and Tal-Beth Cohen for comments and review of the manuscript. The views expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense, Department of Agriculture, or the U. S. Government. All authors are employees of the U.S. Government. This work was prepared as part of official duties.