• 7404 malaria professionals are enjoying the free benefits of MalariaWorld today
Home › Forums › Scientific Topics › Critique of article; The effect of repeated washing on long –lasting treated nets (LLINs) on the feeding success....

Critique of article; The effect of repeated washing on long –lasting treated nets (LLINs) on the feeding success....

2 replies [Last post]
November 5, 2010 - 18:14
Rune Bosselmann
Offline
Joined: 17 Mar 2010
Posts:

Critique of an article using an improper test method to arrive at an invalid discussion and results; “The effect of repeated washing on long –lasting treated nets (LLINs) on the feeding success and survival rates of Anopheles gambiae; Malaria Journal 2010, 9:304 (Francis K Atieli et al). The article by Francis Atieli et al bases its conclusions on test methods that are not valid for all types of LLINs. The authors state that this work shows that when nets are washed according to local wash method with washes every 3 days, two polyethylene (PE) LLIN have a diminishing efficacy effect while two polyester (PES) LLIN have a less diminishing effect. However, the statement is not correct. Nobody washes their nets every 3 day so a washing regime with washes every 3 days this cannot local net washing practices. This is the basic, and undermining for a comparison of nets, error of the study. As shown in the article "Median knock-down time as a new method for evaluating insecticide-treated textiles for mosquito control", Malaria Journal 2008:7:114, nets of the regenerating type (PE LLIN’s) takes more time to regenerate the more they are washed. Therefore, when the authors tested Netprotect and Olyset 3 days after 15 washes with 3 day intervals, the nets were not fully regenerated. Of course this is even more expressed after 20 washes. Had the authors read the above mentioned article, they would know that after 20 washes and 7-8 days, Netprotect kills more than 80 % with just 3 min exposure and around 100 % with 10 min exposure time. Had this been respected, the conclusion of the Francis Atieli et al article would have been the opposite than the one presented. Since the PES LLIN's do not regenerate what you find after 3 days is what you will find after 10. This does not mean that in practice the nets are inactive for 10 days after 20 washes. The Median Knock Down study mentioned above shows that a single wash does not remove all insecticide but successive washes with very short intervals do exactly that. Since this not a practice exhibited by LLIN users the test used in the Francis Atieli et al study fails at estimating the efficacy of the regenerative nets after local washing practices. Further, mosquito behavior studies of interaction with LLIN’s show a mosquito take off rate that is proportional to surface dosage. This indicates that with reduced surface concentration of insecticide, the mosquito will rest longer on the net. Therefore, laboratory studies with fixed exposure time do not reveal the realities of LLIN effect in the field. The speculations presented by the authors on net failure (all types) due to washing are therefore not very well founded. The proper way to study the real effect of washing and aging of LLIN’s are field studies or controlled field studies such as the hut studies carried out by WHO in the WHOPES phase II and in WHOPES phase III trials (for the effect of aging). Rune Bosselmann/ Intelligent Insect Control

‹ Who needs mosquitoes? When do we stop sending pyrethroid treated LLITNs into areas where there is pyrethroid resistance? ›
Top
November 8, 2010 - 12:01
#1
Stephen Munga
In reply, overall, our paper

In reply, overall, our paper captures all the concerns raised by Rune. First, the article explains that this is an accelerated washing procedure using local washing method and detergent not the standard WHO protocol that uses specific machine and soap. We have also explained that it is not a normal practice for villagers to wash nets at 3 days interval, our article presents a situation where nets deployed in single roomed houses (typical of rural African houses) where cooking also takes place, are prone to accumulate dirt and soot faster causing the need to wash nets at shorter time interval than is the normal practice. Ole complains that this method is unfair for his nets which require 4 - 10 days to regenerate. Since the washing guidelines are provided with the nets the villages may wash the nets at short time intervals in complete disregard of what is recommended. Our paper captures this concern very well and explains it with relevant references. Further, we have explained that polyethylene based nets takes longer time to regenerate. But if that is the situation, what happens in between those many days required for regeneration? In this case the mosquitoes may be able to feed during this regeneration period. It means sleepers will not be protected within this window period. There is a lot of literature the article sites regarding regeneration of polyethylene based nets. Previously nets were to be placed in polyethene bags in direct sun light for regeneration to occur but this requirement was withdrawn. We also referenced one article by Gimnig and Atieli 2005, where polyethyne nets were left in an oven set at 30 degree Celsius for more than 7 days after repeated washing and regeneration did not occur, but when the same nets were exposed to temperatures of 60 degrees Celsius, regeneration occurred only after 4 hours. In fact the reviewers of the article pointed out that we should explain that the method used is not the normal practice commonly used in the field, which we adequately responded to. No attempts have been made to measure feeding success of field mosquitoes through treated nets after washing. In the next set of experiments we anticipate to determine the feeding success at 7 day washing interval. We strongly believe that field mosquitoes might be able to feed through these nets only after 5 washes.

Top
November 29, 2010 - 10:19
#2
Rune Bosselmann
Offline
Joined: 17 Mar 2010
Posts:
Reply and elaboration of critique

I agree strongly with an essential premise for the investigations described in the article; to investigate the level of personal protected is provided by an LLIN when washed according to local norms and practices. In general I feel that such studies on LLIN performance in settings that are as close to real life as possible are highly relevant, important and should be carried out across regions to better evaluate and anticipate the effect of LLIN campaigns.

The problem is that the investigation diverges from normal real use on a parameter (frequency of washing) is quite important for the evaluation of LLIN performance. As pointed out, in text and reference, a PE net will lose a percentage of its surface concentration after a wash and regenerate its surface levels of pyrethroid only after some days (number of days depending on brand). For sakes of simplicity, lets say it loses 50% of all pyrethroid on the surface but regenerates to original level after 6days. That means that after 3 days the net is only half way regenerated. Not counting for the absolute loss at each washing that means that at 3washes, you will have a net with 75% of the original surface dose. Wash it at this time and you may arrive at 37,5%. Since we do not account for changes in the absolute reservoir inside the fiber and we also pretend that migration speed from the interior to the exterior is independent of relative concentrations (surface vs interior), 3 days will still allow for a migration of 25% of original level, meaning we will arrive at 62,5 %. Repeat the loop and you are at 56,25% etc. The point is you never arrive at the level that net would regenerate to if washing intervals reflected what is common in real life. As such, what you find in your survey does not well reflect the level of protection eg an Olyset offers after 5 real washes. In real life this product may exhibit the same level of bioactivity 12 days after its 20th wash as it did 3days after your 2nd wash. Also, do recall how well the product did in its 5year study.

I must admit I am not aware of what regeneration time or under what conditions the current formulation of the Olyset net should be left to regenerate. The other PE nets have not prescribed the black bag treatment. I should point out though, that the WHOPES recommended LLIN’s are allowed some manipulation in formulation (recipe) without having to resubmit to WHOPES.

I agree with you that villagers may not wash the nets as prescribed. The actual number of washes presumably varies wildly across countries and communities and even for nets within the household. If LLIN’s were really washed at three day intervals, I would agree with you that the PE nets would be the first to arrive at critically low insecticide concentration at the surface. But you could also after 3months put a PE net aside for a week and it would still be effective within the prescribed performance parameters. A polyester net would be completely exhausted. However, this does not reflect reality. If nets were washed that often, you would reach 60 washes in half a year and that alone may affect your malaria control planning for that community. I do not agree with you that it would be interesting to see what happens with the nets when washed every 7days unless this truly is the norm somewhere. For all the allure of accelerated time test designs you really need to observe for what the acceleration does to the validity of the results. Your 7day interval suggestion may display a reduced personal protection for some PE nets but it would likely also show a lifetime for polyester nets of half a year – which would prompt the question, what are the practical implications of this result?
I agree with you that nets used near the kitchen become dirty a lot faster and in consequence are washed more frequently. However, it should be noted that the texturized polyester yarn tends to become dirty much faster than the smooth and hard polyethylene yarn and because of this is likely to be washed more frequently. It may be that LLINs that are to be used in rooms where there is a kitchen should be designed so that they are easy to get out of the way and install on a daily basis and at little inconvenience to the user. After all, no LLIN can withstand any thinkable level of abuse. Better make it easy and likely to treat it right.

A second and fundamental question I would like to raise is whether the exposure time of 10 min used by Atiele, Munga et al is relevant for practical evaluations. When mosquitoes enter a house they tend to fly toward the sleeping person and when they come upon a barrier in the form of a LLIN they then try to find a way to reach the target. This may be via a hole in the net or a body part in contact with the net. They may search for hours or even the whole night with intervals until they finally either got a blood meal or got a lethal dosage of insecticide. Therefore, a lower dosage of insecticide does not automatically implicate that the net does not protect. The net no longer works in the Atiele, Munga et al 10 min bioassay, but that again that may mean little in the real life scenario. As a parallel, take the many kdr (resistance) studies from West Africa. When the kdr gene is alone as pyrethroid resistance mechanism the LLIN fails the WHOPES 3 min cone test but works well in practice; the kdr gene basically moving the tox tolerance threshold, but is not removing toxicity. The mosquito "overcomes" the elevated threshold by spending more time on the LLIN. This is also demonstrated in "time to first take off" studies that measure for how long a kdr resistant mosquitoes will accept to sit on a pyrethroid treated net compared to a wild type mosquito.

So, instead of repeating the study with a 7 day interval between washes and between wash and bioassay it may better for real life relevance to make the evaluation in hut studies as in WHOPES II. If possible, study how long a mosquito will touch a net before giving up and go somewhere else. Imagine that an average mosquito will spend 1 hr on a non-treated net trying to find a hole. Then it does not really matter if you use a bioassay of 3 min forced exposure as a WHO cone test or 10 min voluntarily exposure as in your test or if the mosquito has a kdr gene that will make it survive a 3 min exposure, but not a 6 min exposure. These exposure times will underestimate effect of a treated net. Knowing this average exposure time and search pattern of mosquitoes on a net may also help us in modeling and understanding impact of holes on net performance.

Top