Abstract
This review examines the published laboratory and field tests where the repellents DEET and picaridin have been compared for their efficacy as repellents against mosquitoes. The review is limited to an assessment of whether the duration of protection afforded by picaridin is similar to or better than DEET.
Identification and analysis of laboratory and field-based trials published in peer-reviewed journals that compared DEET to picaridin efficacy.
Only eight field studies and three laboratory studies met the review criteria for inclusion and most were considered to be of high risk of bias and of lower quality when judged against evidence-based principles. Overall, the studies showed little potential difference between DEET and picaridin applied at the same dosage, with some evidence pointing to a superior persistence for picaridin.
Applied dosage is one important variable in determining the persistence of a repellent experienced by users but the maximum concentration in current picaridin formulation is <30%w/v. Therefore, where only 30% DEET or lower concentrations are available, then on current evidence, it is reasonable to offer DEET or picaridin as a first choice. Where >50% DEET products are available then the protection time advantage associated with these formulations reasonably can be invoked to consider them as first choice repellents.
Introduction
Preventing bites is an important disease risk reduction strategy for travellers to areas where mosquito-associated disease are prevalent. The approach is widely recommended by public health authorities1–3 and consists of complimentary methods, e.g. repellents applied to the exposed skin, barriers such as appropriate clothing and bed nets (ideally treated with insecticide), and area control using devices such as insecticide vaporizers.
The relative benefit of a given bite prevention approach depends on a number of variables, including mosquito behaviour. For example, topical repellent is ‘better’ than a bed net for a mosquito that happens to bite outdoors, whereas the opposite holds true for a mosquito that seeks its meal indoors and at night. Unfortunately, not all mosquitoes follow strict behavioural rules. For example, malaria mosquitoes do not always bite inside at night, and dengue mosquitoes do not always bite during daylight hours. This variability is a key reason why travellers should employ several bite prevention methods, i.e. to provide a hedge against within and between species behavioural variability.4
In this review, our focus is on topical repellents, specifically DEET (N,N-diethyl-3-methylbenzamide, diethyltoluamide) and picaridin ((2-(2-hydroxyethyl)-piperidinecarboxylic acid 1-methyl ester), WHO designation: icaridin, trade name Bayrepel®, development reference code KBR 3023 also known as Salitidin). The first has, for some time, been the de facto repellent recommendation for travellers; the second, more recently on the scene, is also widely recommended.1–3,5 The question we consider is: should picaridin replace DEET as the preferred repellent for travellers? In answering this question, we limit ourselves to a technical assessment of whether the duration of protection (also called protection time) afforded by picaridin is similar to or better than DEET? In so doing, we make the assumption that the duration of protection provided by a repellent is important to travellers.
In addressing the question raised in this review it is important to consider how repellent efficacy is determined. Generally, the approach is to use volunteers exposed to naturally occurring populations of mosquitoes in the field, or in a laboratory where an arm is exposed to a cage containing mosquitoes (arm in cage tests). Because these tests have been performed in many different ways, comparison between trials is difficult. Partly for this reason, the World Health Organization (WHO) has developed and recommends a specific approach for assessing repellent efficacy in the field and laboratory.6 Whether this will bring greater consistency to the field is uncertain, not least as major regulators (e.g. the United States Environmental Protection Agency)7 do not adhere to the WHO standard.
Irrespective of the protocol used, the fundamental question for a study is how much repellent is required to repel the target mosquitoes. In dose finding studies, this often is expressed as the effective dose 90 (ED90), which is the dose, in mg of product applied to each cm2 of skin, that reduces the number of bites (compared to a negative control) by 90%. If an impractical amount of repellent needs to be applied to achieve the ED90, the product would not be deemed suitable for use. In general, the ED90 should be achieved with at most, 1–2 mg product/cm2, which is when the skin would start to feel wet and product might run off. Importantly, because most repellent products are usually not entirely composed of active ingredient, the ED90 of an active ingredient usually needs to be substantially less than this threshold. This certainly applies to DEET and picaridin which have ED90 values far less than 1 mg/cm2 for Anopholese and Aedes species, and hence can be administered in a variety of concentrations.8
Given that DEET and picaridin have acceptable ED90s, the crucial question is how long will the repellent maintain its activity at or above the 95% efficacy level – commonly referred to as ‘time to first bite’ or ‘Complete Protection time’. This could be influenced by the formulation, e.g. a micro-encapsulated long acting preparation, the initial skin concentration or a property of the active ingredient. Figure 1 describes the time profile of activity after application for DEET, which follows broadly first-order (exponential) kinetics. The most important variable is the applied dose expressed as the amount of active ingredient (ai) mg/cm2. The model illustrates that the higher the applied dose, the longer the protection period up to a plateau at around 2 mg/cm2.
Persistence of DEET as predicted by a model based on Rutlidge et al.23
While dose response models are important for establishing the repellency potential of a product, they should not be equated with actual protection time as would be experienced by an end user. This is because field performance in impacted by a variety of factors external to the repellent itself, for example the types, numbers and physiologic state of the targeted mosquitoes, the application technique of the user, physical removal of the repellent (e.g. sweating) and individual-level variability in the user.
Therefore, this review will examine in detail those studies that compare the duration of protection afforded by picaridin and DEET. Our null hypothesis would be that if DEET and picaridin are applied at the same concentration of active ingredient there would be no significant difference in the length of protection they afforded.
Methods
For this review, we carried out a focused systematic review comparing DEET and picaridin in the field setting and also reviewed laboratory work with arm-in-cage tests.
Systematic search for relevant field studies
One of the authors (SS) previously (unpublished) undertook a systematic literature search (2013) and evidence analysis of field studies comparing picaridin and DEET. This focused review was updated to include literature published up to and including October 2017.
The electronic databases Ovid, MEDLINE and Scopus were searched using the terms DEET AND picaridin (or icaridin). Only articles in English and/or French were retained. After removing duplicate records, titles and abstracts were screened for relevance. Only articles meeting the following inclusion criteria were kept: contemporaneous comparison of DEET and icaridin; field studies using human test subjects; non-contemporaneous use of treatment and control on the same subject; and endpoint of proportional reduction in biting.9 Retained articles were assessed for risk of bias in the following domains: sequence generation, allocation concealment; performance (blinding of subjects); detection (blinding of assessors); attrition (completeness of reporting); and selective reporting approach based on the Cochrane risk of bias tool.10
Laboratory arm-in-cage studies
LG performed a literature search in 2016 to identify published arm-in-cage tests comparing DEET to picaridin.
The databases PubMed, Web of Knowledge and Science Direct were searched. The search terms were DEET AND picaridin (or icaridin). Inclusion criteria were: laboratory tests using human subjects; comparison of, minimally, DEET and picaridin; direct assessment of protection, i.e. bite protection on human skin; a cage test; performed using mosquitoes; and, endpoint of time to first bite.
Results
Our focused systematic search identified 116 potentially relevant field studies. The large majority (108) of these did not meet our inclusion criteria. The remaining eight studies11,12,13,14,15,16,17,18 summarized in Table 1 were done in the Americas,11,18 Australia,13,14 Africa and Asia.15,16 Amongst them, they included multiple mosquito genera including those that contain important vectors, e.g. Anopheles, Aedes, Culex.
Summary and risk of bias assessment for included field studies
| Study description | Products tested* | Results | Quality assessment (risk of bias) |
|---|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Picaridin and DEET at doses of 0.1, 0.3, 0.6 and 0.8 mg/cm2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Study description | Products tested* | Results | Quality assessment (risk of bias) |
|---|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Picaridin and DEET at doses of 0.1, 0.3, 0.6 and 0.8 mg/cm2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
*DEET and picaridin treatments are included in table, some studies also evaluated other active ingredients.
Summary and risk of bias assessment for included field studies
| Study description | Products tested* | Results | Quality assessment (risk of bias) |
|---|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Picaridin and DEET at doses of 0.1, 0.3, 0.6 and 0.8 mg/cm2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Study description | Products tested* | Results | Quality assessment (risk of bias) |
|---|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Picaridin and DEET at doses of 0.1, 0.3, 0.6 and 0.8 mg/cm2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
*DEET and picaridin treatments are included in table, some studies also evaluated other active ingredients.
The literature search by LG identified 26 laboratory based studies, of which 10 were not performed using mosquitoes (mostly ticks), five did not involve human skin, three examined adverse effects and five did not assess repellent longevity. The three included studies were by Scott et al. with non-disease transmitting wild Psorophora species,19 Massetti and Maini20 with Aedes albopictus, and Barnard and Xue21 using several mosquito species (Aedes albopictus, Culex nigripalpus and Ochlerotatus triseriatus).
Repellent performance
There was not a consistent difference in performance between DEET and picardin in field studies (Table 1). In two studies picaridin appeared to outperform DEET,11,17 in two studies DEET appeared to outperform picardin12,18 and in four studies these active ingredients performed equally well13,14,15,16. It is, however, noteworthy, that in the two studies where DEET appeared to provide enhanced protection, the concentration of active ingredient in the tested DEET product was substantially higher (~1.5 or 8 times) than in the tested picaridin product.
The duration of protection afforded by DEET and picaridin varied between studies, from a minimum of ~4 h to more than 12 h. Importantly, when the duration of protection was relatively short for picaridin, it also was relatively short for DEET.11,14 Finally, there was some evidence of variable performance against different genera of mosquitoes. For example, the duration of protection afforded by DEET and picaridin in a study done in Australia was shorter for Anopheles spp. than for Culex spp.14 A similar result was obtained in Cambodia,17 with the duration of protection being relatively shorter for Anopheles or Aedes spp. than for Culex or Mansonia spp.14
Regarding the laboratory work, the study by Barnard and Xue21 compared an applied dose of 0.65 mg/cm2 of a number of products including low dose formulations of DEET (15% and 7%) and picaridin (10%). The mean protection (SE) times expressed as time to first bite were; 7% DEET 5 h (1.5), 15% DEET 7.2 h (0.5) and 10% picaridin 5.7 h (0.5). Massetti and Main22 also assessed time to first bite in volunteers applying the same volume of a formulation; 15% DEET or 20% picaridin. The mean protection times were similar with very wide standard deviations; DEET 250 min (SD 94), and picaridin 220 min (SD 121). These studies suggest that similar applied ai dosages of DEET and picaridin result in the same duration of effect as measured by time to first bite. The Scott et al. cage test used the same applied doses of a 15% DEET and 5% picaridin. Three volunteers were used, testing DEET on one arm and picaridin on the other and exposing them alternately. The average time to first bite was 3.5 h for picaridin and 5.5 h for DEET. This result illustrates the importance of ai dose, as DEET seems to offer superior performance in this study, but also was applied at a rate that was 3-fold higher than picaridin.
Quality
We judged that seven of the included field studies were at high risk of bias (Table 1), and a single one15 was at low risk of bias. In particular, the included studies generally did not allocate treatments randomly and/or participants/assessors were not blinded. The quality of these studies, i.e. our confidence that they represent the true state of affairs, is thus low. The exception was a trial done in Burkina Faso primarily against Anopheles spp (including important malaria vectors). In it, both active ingredients provided long-lasting protection against bites, with picaridin showing a modest protection period advantage based on decreased loss rate from the skin. Importantly, both DEET and picaridin were superior to another active ingredient (IR3535) tested in this study, i.e. at higher concentrations, the protection afforded by IR3535 fell below 95% by ~5–6 h after application.
The included field studies do not permit a robust assessment as to whether picaridin is equal to or better than DEET in terms of protection period. This reflects, inter alia, the: low confidence we have in the effect estimates generated by the majority of the included studies; and, our uncertainty about the generalizability of the relatively higher quality study15 to the diverse mosquito environments that travellers experience.
Of the included laboratory studies, one used the non-disease transmitting wild Psorophora species,19 another22 used three volunteers but a lower than usual number of mosquitoes in the cage and the final study21 appeared to be well conducted but three of the five subjects did not complete the tests. Further, none of the included studies were blinded or adhered to the methods recommended by the WHO.6 Therefore, as with field studies, our confidence in the estimates of effect for these studies was low.
Discussion
The current body of evidence does not allow us to conclude with confidence that picardin offers a mosquito bite protection period advantage when compared to DEET. Not only are there insufficient numbers of such studies directly comparing the two active ingredients, but as we have shown those that have been published in peer reviewed journals tend to be of a lower quality than would be desired to provide a high level of confidence in their findings. Certainly, more work could be done to increase our confidence in the effect estimates for these active ingredients. On the other hand, it is not unreasonable to make a judgement based on current evidence. In particular, DEET and picaridin, from a protection period perspective, could be considered equally preferred repellent entities. Indeed, it is unlikely that new evidence will reveal operationally meaningful differences in protection period associated with these active ingredients. The rationale for this is:
– despite quality-associated limitations, the reviewed evidence is consistent in that it not does not identify a clear superiority for one of the active ingredients. This enhances our confidence that differences in performance, if they exist, are modest;
– even if there was a modest difference in performance, e.g. picaridin ‘lasts’ longer than DEET, its significance would be muted by pragmatic aspects of repellent use and their impacts on protection period. These include, but are not limited to: user preferences; variation in adherence; variation is application rate/frequency/technique; formulation-based differences; subject-based variation; environmental factors; and/or, mosquito-based factors.
Despite the above, we acknowledge that the number of studies conducted using DEET over its long history as the ‘Gold Standard’, might tend to outweigh the relatively few peer reviewed studies conducted on picaridin when considering a recommendation of a preferred repellent.
Overall, there is therefore an emerging picture that providing the same amount of ai is applied and efficacy is measured as time to first bite, whether in a field or cage test, the two chemicals will provide near to equivalent periods of protection. However, the results reported should not be taken as the actual length to a first bite that might be experienced by an individual due to the multitude of factors as already discussed above. An important consideration when considering the ‘preferred’ repellent is that in some countries such as the United Kingdom, products can currently contain more DEET (up to 50%) than picaridin (up to 30%). In this situation, there might be advantage in terms of protection period to using the higher concentration products, although the advantage is likely to be muted given repellent kinetics (Figure 1) and the observation that picaridin is likely to be somewhat more persistent on the skin than DEET.15
Conclusion
In conclusion we feel that where 50% DEET products are available then it can be argued that the protection time advantage associated with these formulations reasonably can be invoked to consider it as a first choice repellent. Where only 30% DEET or lower concentrations are available, then on current evidence it is reasonable to offer DEET or picaridin as a first choice. While new evidence or registration of higher concentration picaridin products might alter the nuance of these recommendations, it is unlikely that it will change importantly the conclusion that DEET and picaridin offer long periods of protection against mosquito bites. Indeed, given the vagaries of repellent performance, including the clear impact of adherence, it is also reasonable to consider these active ingredients as near to equals and that, based on traveller values and preferences, can be considered the preferred option.
Conflict of interest: None declared.
