SEARCH

SEARCH BY CITATION

Reference: Lindblade et al. (2005) Evaluation of long-lasting insecticidal nets after 2 years household use. Tropical Medicine & International Health 10, 1141–1150

  1. Top of page
  2. Reference: Lindblade et al. (2005) Evaluation of long-lasting insecticidal nets after 2 years household use. Tropical Medicine & International Health 10, 1141–1150
  3. Methods
  4. Study design
  5. Reference: J Gimnig et al. (2005) Laboratory wash resistance of long-lasting insecticidal nets. Tropical Medicine & International Health 10, 1022–1029
  6. Conclusion
  7. References

Dear Sirs,

We have carefully reviewed the above paper and have serious technical concerns over the methodology used and the results obtained. The conclusions reported in the paper may lead potential users of Olyset® mosquito nets to doubt the efficacy of the product and this could in turn negatively impact the substantial human health benefit that is currently being derived from this product's use. Such is our concern, that we have consulted worldwide experts in bednet technology whose opinions are presented in this letter with their permission.

Olyset nets have been evaluated by many scientists and experts in malaria and bednet technology with positive results. Olyset has also been sold and used in the field for many years, successfully contributing to large-scale malaria control programmes with demonstrable impact by reducing malaria incidence and mortality. Olyset nets contribute to malaria control through four modes of action against mosquitoes: mortality, knockdown (KD), repellency and bite inhibition. In the paper presented by Dr Lindblade only the first of these multiple modes of action is considered.

The first objective of any insecticide treated bednet that is specified for use in a malaria control programme is to prevent malaria. Much work has been published on evaluations of Olyset by measuring the eventual impact on the incidence of malaria. Contrary to the results of the studies carried out by Lindblade et al., in all these other cases, Olyset nets have shown a measurable impact on mosquitoes and malaria reduction. The intention of this letter is to set out the scientific reasons for disputing the results of work carried out by Lindblade et al., and the conclusions they have reached. Our comments are presented in the same order as they appear in the paper.

Methods

  1. Top of page
  2. Reference: Lindblade et al. (2005) Evaluation of long-lasting insecticidal nets after 2 years household use. Tropical Medicine & International Health 10, 1141–1150
  3. Methods
  4. Study design
  5. Reference: J Gimnig et al. (2005) Laboratory wash resistance of long-lasting insecticidal nets. Tropical Medicine & International Health 10, 1022–1029
  6. Conclusion
  7. References

It is stated by the authors that Olyset nets come with instructions to wash the net every 6 months and then place it in the sun in a clear plastic bag provided by the manufacturer to regenerate the insecticide. This statement is not correct. Olyset labelling does not currently carry such a recommendation. Labelling of Olyset is today consistent with the WHO Pesticide Evaluation Scheme (WHOPES, trials reference WHO/CDC/WHOPES/2001.4 page 15), which states:

‘‘a major advantage of Olyset net is that its biological efficacy is resumed by diffusion of the insecticide from inside of the yarn to the surface. This ‘regeneration’ occurs naturally but is accelerated when the nets are exposed to heat. However under tropical conditions, Olyset net washed 10 times recovered its efficacy in less than 15 days. In experimental huts, after 5 washes with local soap, Olyset net without heating remained as effective as non-washed net.’’

Additionally, we have been provided with the following personal communication:

SourcePersonal communication
Prof. CF Curtis, London School of Hygiene and Tropical Medicine (2004)‘We conclude from our study carried out for the WHOPES (2001) that, without any artificial heating, insecticide lost during the washes was replaced by diffusion to the surface of the fibres in the early part of the 6-week duration of the study so that the average performance over the whole 6-week period was indistinguishable, with or without washing or washing and heating’

Study design

  1. Top of page
  2. Reference: Lindblade et al. (2005) Evaluation of long-lasting insecticidal nets after 2 years household use. Tropical Medicine & International Health 10, 1141–1150
  3. Methods
  4. Study design
  5. Reference: J Gimnig et al. (2005) Laboratory wash resistance of long-lasting insecticidal nets. Tropical Medicine & International Health 10, 1022–1029
  6. Conclusion
  7. References

In the study design, the authors repeat the above error by inappropriately including the heating of nets in the sun. It is also reported that a check was made to determine if households reported washing their nets; yet there is no indication in the remainder of the report when nets were tested after washing – it may have been 1 day, 1 week or even longer. Since we know that re-generation continuously occurs, this factor is vital to the interpretation of any bioassay results. Testing too soon after washing, before the regeneration of surface insecticide has occurred would give a misleading result.

Determination of net failure

Our most substantive criticism of the study is on the single method used to determine efficacy. The researchers used the WHO cone bioassay method alone. While we recognize some value in this well-established scientific practice as a method to measure mosquito mortality, eminent experts and ourselves represent that is has severe limitations if solely relied upon to measure efficacy of the Olyset nets. We ask your consideration of the following personal communications:

SourcePersonal communication
Dr. M Zaim, WHOPES, Geneva, Switzerland (2005)‘There is strong evidence that the cone bioassay alone is not sufficient for the determination of the efficacy of Olyset and similar technologies’
Dr. Pierre F. Guillet, Vector Biology and Control Unit, WHO Regional Office for Africa (2004)‘There is with Olyset nets a general trend of discrepancy between results obtained with cones and results observed in experimental huts or in tunnel tests’
Guidelines for laboratory and field testing of long-lasting insecticidal mosquito nets, WHO/ CDS/WHOPES/GCDPP/ 2000.11‘The efficacy of treated nets can be underestimated if judged only on standard cone bioassays. This is specially the case with insecticides which have a high excito repellent effect, such as permethrin...’

In their work, the authors apparently failed to take into account that permethrin (the active ingredient present in Olyset) is far more repellent than deltamethrin (the insecticide used in other nets that were tested). Permethrin is also used in skin cream insect repellents. Its repellency causes mosquitoes to avoid resting on the netting during the brief 3-minute exposure time, resulting in them preferring to rest on the plastic cone away from the net. Again we refer the reader to personal communication:

SourcePersonal communication
Dr. Pierre F. Guillet, Vector Biology and Control Unit, WHO Regional Office for Africa (2004)‘The repellent effect (of Olyset) and the resulting blood feeding inhibition remained surprisingly high after 7 years with nets that according to our cone test criteria would have been rejected most probably long ago and were apparently totally ineffective. These results show persistence of insecticide and biological activity for up to 7 years’

Lindblade et al. also introduced a cardboard disk behind the netting. This non-standard practice deviates from the WHO test method and could have affected the results. Evaluation of this non-standard methodology by independent researchers showed consistently lower mortality and a longer time for knockdown of Olyset netting (Magoma, J, in Press). Significantly the authors seem not to have considered alternative test methods, such as the Tunnel test technique defined by Chandre et al. (2000), WHO (2005) or the WHO Ball bioassay method (WHO 1996; Curtis et al. 1998). See the following personal communications:

SourcePersonal communication
  1. *Markedly lower mortality in bioassays using cones, as compared to those using ball-shaped wire frames (‘ball tests’), is also very evident in Figure 1 of the letter by Gimnig et al., although the authors describe the results with the different methods as ‘‘similar’’.

Prof. CF Curtis, London School of Hygiene and Tropical Medicine (2004)‘In our experience with this method some mosquitoes spend time on the plastic cone. It seems likely that some mosquitoes would be able to stand on the cardboard without touching the fibers of the broad mesh Olyset netting, and thus avoid insecticide exposure. This is supported by data of J. Magoma in our laboratory, who showed significantly lower mortality with unwashed or washed Olyset nets bioassayed under cones with cardboard backing, than with the mosquitoes placed inside similar nets wrapped round wire frames*. It is therefore more reliable to wrap the net around a wire frame and put the mosquitoes inside so that they are forced to walk on the net. Lindblade et al. placed a cardboard disk on the opposite side of the net from their plastic cone. I wonder if the mosquitoes may stand on the cardboard without touching the fibres of the insecticidal netting and thus avoid insecticidal exposure’
Dr. Pierre F. Guillet, Vector Biology and Control Unit, WHO Regional Office for Africa (2004)‘7-year-old samples (Olyset net) from Senegal were submitted to a tunnel test without any heat reactivation. This test system is the closest to experimental hut conditions, using free flying host seeking An.gambiae females. Mortality has been >50% and as high as 78.2% with a net which did not have any detectable activity under cone test’

The authors’ protocol may have been more appropriate had they considered the practice followed by Graham et al. (2005), who evaluated long lasting insecticide nets (LLINs) by using two types of bioassay, WHO cone and WHO ball bioassay. They then supplemented their bioassay work with trials of nets in experimental huts and on enclosed platforms, which gave a more comprehensive picture of net efficacy.

Criteria for net failure and regeneration of Olyset nets

We draw the reader's attention to the authors’ definition of net failure and to the related subject of regeneration of Olyset nets and to the views expressed in the following personal communications:

SourcePersonal communication
Prof. CF Curtis, London School of Hygiene and Tropical Medicine (2004)‘Net ‘failure’ was defined by Lindblade et al. as a bioassay mortality of <50% based on three tests with ten mosquitoes. This is an arbitrary definition based on artificially confining the mosquitoes in the above-described assembly of a cone, the net and a cardboard disk. Tests in experimental huts with free flying, wild mosquitoes and measurement of prevention of biting and mosquito killing are a more realistic measure of whether there is enough insecticidal activity to have the two desired effects of treated bednets. These tests included the unexpected demonstration that insecticidal netting with 4 mm mesh (Olyset or conventionally treated) reduced biting on the sleeper approximately as well as 1.5 mm mesh. However, experimental hut tests of a few nets take weeks, and a judicious blend of bioassays and more realistic huts tests is desirable’
Prof. CF Curtis, London School of Hygiene and Tropical Medicine (2004)‘They report that ten Olyset nets, which had ‘failed’ by their definition, gave 85–100% mosquito mortality after washing them and ‘proper regeneration’ by leaving in bags in the sun until they had dried (usually for 2 days). This appears to contradict our data about lack of effect of (or need for) heating after washing. Unless there was only an interval of a few days between washing and bioassaying I cannot see how to reconcile the data of Lindblade et al. and our data on lack of effect of washing and heating and the fact that after 4 years domestic use an Olyset net performed in an experimental hut as well as a new Olyset net. In Tanzania we have an ongoing trial in 23 villages with four villages with Olysets, four with Permanets, four with Dawa nets, four with conventionally treated nets and seven with no nets. Data for the first year of this trial show that villages with any of the insecticidal nets have reduced Anopheles population densities and sporozoite rates, compared with the villages without nets’

Evaluation of house entering/exiting behaviour, feeding success

This very practical work is perhaps more relevant to the actual protection afforded by nets rather than from extrapolation from cone tests but sadly, no table of results has been reported in the paper. However, in the Results the authors report that they found no statistical difference between nets tested. This qualitative remark is indicative that Olyset nets performed comparably with the other nets and is directly contrary to the results from the inconclusive WHO cone tests, which show a large difference between net types.

Chemical analyses

It is of substantial concern that Gas Chromatography (GC) was used to measure insecticide concentration of deltamethrin extracted from the nets. GC analysis requires a column temperature to be high enough to vaporize the insecticide. At such high temperatures degradation of deltamethrin occurs with resulting inaccurate and misleading data. Therefore only HPLC should be used for such deltamethrin analyses. HPLC operates at room temperature with no loss of active ingredient and provides an accurate result. It should be noted, however, that GC is suitable for analyzing permethrin.

SourcePersonal communication
P Rich, Head Analytical Department (retired), AgrEvo Environment Health‘This is a known fact (inaccurate analytical results when using GC to analyze for deltamethrin) amongst Analytical Scientists’

Graham et al. (2005) also substantiate HPLC as the established method for deltamethrin analysis. Thus the estimates of insecticide content of the deltamethrin nets could be a considerable underestimate. The Olyset analytical results also provide cause for concern, since the authors report that the amount of insecticide found in ‘failed nets’ was higher than the original concentration. It is not known how insecticide concentrations can ‘increase’ from the original baseline figure.

In summary, we contend that the method used in this field trial was inappropriate for Olyset nets and has produced misleading results. This paper contradicts WHOPES field trials, which have showed Olyset nets to be very effective. The rigorous independent trials required by WHOPES resulted in the Olyset net achieving its full approval. At present Olyset remains the only LLIN to carry such approval; by contrast Permanet® has only interim approval awaiting further investigations.

Although, through no fault of the authors of the paper, it should be noted that the Olyset nets evaluated in this study were already 7 years old when distributed in the field. Therefore they were 9 years old at the end of the trial. These nets had been in storage, and it is not known to what extent this may have affected their efficacy.

Different studies conducted by other researchers using the same WHO cone tests showed positive data, which is contrary to the Lindblade paper. For this reason Sumitomo Chemical Co. Ltd. and other leading scientists recommend other test methods and hut trial data to confirm the efficacy of Olyset nets.

The following references are from published papers which show results totally contrary to those found by Lindblade et al. and therefore supports our submission that the results obtained by Lindblade et al. are questionable.

ReferenceAuthorTitle and publication referenceKey result
1Guessan et al. (2001)Olyset net efficacy against pyrethroid-resistant Anopheles gambiae and Culex quinquefasciatus after 3 years se in Côte d'Ivoire. Medical and Veterinary Entomology (2001) 15, 97–104Bioassays with 3 min exposure of susceptible Anopheles gambiae gave >99% mortality of female mosquitoes tested on the ‘new’ Olyset net. The used Olyset nets gave mortality rates averaging 83% for the washed net, 85% for the dirty net and 55% for the very dirty net. Thus Olyset nets were found to remain remarkably effective against susceptible A. gambiae for at least 3 years under field conditions. Note: these researchers did not rely on this methodology alone but also employed experimental hut data
2Jeyalakshmi, T. and Shanmugasundaram, R. (in press) Department Entomology, International Institute of Biotechnology and Toxicology, Tamil Nadu, IndiaComparative Efficacy and Persistency of Permethrin in Olyset net and conventionally treated nets against Aedes aegypti and Anopheles stephensiOlyset nets were evaluated after each wash up to ten washings and later assessment for efficacy was made after 15th and 20th washings. Olyset net gave 100% mortality at all stages including after the 20th wash. The conventionally treated permethrin net had failed after only two washes
3Tami, A et al. (2004)Evaluation of Olyset insecticide-treated nets distributed seven years previously in Tanzania. Malaria Journal 2004, 3:19Bioassay results indicated high KD rates and a functional mortality of 50% using WHO cone bioassay test method.
4Yeang Chheang and Lek Sandy 1994 National Malaria Centre, Phnom Penh, CambodiaFinal Report on a Field Trial of Olyset net for the Control of Malaria Transmitted by Anopheles dirus and Anopheles minimus in Rattanak Kiri Province, CambodiaIn 3 min exposure with Olyset net the mortality rate of A.dirus and A. minimus was 100% through the first to seventh month of use. Note: these researchers did not rely on this method alone but also employed many other monitoring methods including the impact on malaria. The malaria mortality rate in the Olyset study area dropped from 8.4 per 1000 population to 0, a 100% reduction

Reference: J Gimnig et al. (2005) Laboratory wash resistance of long-lasting insecticidal nets. Tropical Medicine & International Health 10, 1022–1029

  1. Top of page
  2. Reference: Lindblade et al. (2005) Evaluation of long-lasting insecticidal nets after 2 years household use. Tropical Medicine & International Health 10, 1141–1150
  3. Methods
  4. Study design
  5. Reference: J Gimnig et al. (2005) Laboratory wash resistance of long-lasting insecticidal nets. Tropical Medicine & International Health 10, 1022–1029
  6. Conclusion
  7. References

We submit our comment on the second paper which we also believe has serious flaws that may lead potential users of Olyset mosquito nets to doubt their efficacy, causing a possible negative impact on the significant human health benefit of these nets.

Laboratory wash resistance

The washing regime is a pure laboratory method, which does not realistically replicate what happens in the field; neither does it take into account the unique nature of Olyset technology. The authors report that they washed nets every week and bio-assayed at 6 days after washing, measuring mortality and KD. However, it is known that Olyset takes time to regenerate sufficient insecticide to kill mosquitoes after washing (c 15 days – ref. WHO/CDS/WHOPES/2001.4). Graham et al. (2005) state that ‘In accelerated trials, it is convenient to re-wash rapidly, even daily, which may be insufficient interval to allow reactivation of certain types of LLIN such as Olyset Net’.

It is not known why Lindblade et al. chose a washing protocol that was certain to show poor mortality at 6 days with Olyset nets. It should be noted that during the regeneration period Olyset nets still have sufficient surface insecticide to cause mosquito repellency and bite inhibition, thus continuing to afford the user protection.

Bioassays

As explained above, the WHO cone bioassay method is not ideal for this type of technology. Unlike in the previously referred paper –Evaluation of long-lasting insecticidal nets after 2 years household use, it is not stated whether cardboard disks were used behind the netting; we therefore must assume they followed the same methodology. If this is the case inaccurate results would have been obtained and the observations of Prof. Curtis (see above) would be relevant. Graham et al. (2005) comment: ‘Higher mortality levels of Anopheles in contact bioassays did not always translate to superiority in experimental hut or enclosed platform trials’.

The only difference in the methodology that Lindblade et al. reported is the introduction of a measurement of the KD time but even when measuring KD the authors used the WHO cone methodology. Other researchers (e.g. Prof. CF Curtis) have found that the WHO Ball bioassay is a better tool for measuring KD as the mosquitoes have to rest on the net, preventing the variability caused by mosquitoes avoiding the net which occurs when using the cone bioassay.

Again the authors relied solely on the WHO cone bioassay and mortality as the measure of efficacy and seem to have ignored the many other criteria, which affect mosquito behaviour and in the field conditions prevent man-vector contact. The results of bioassays shown in Table 1a report only mosquito mortality using the WHO cone method. Unsurprisingly, the mortality impact of Olyset drops to a very low level after a few washes since the authors washed all the permethrin from the fibre surface and allowed insufficient time for the net to regenerate insecticide from within the fibre.

Table 1b refers to KD data: mortality and KD returned to normal after the authors heated the nets at 60 °C, as would be expected from a practice that artificially forced a high level of permethrin to reach the surface of the fibres of the net. The KD figures shown in Table 1b follow the same pattern of rapid loss of effect as a result of repeated washes at short intervals without allowing sufficient time for surface insecticide replenishment. We therefore submit that the data presented in Table 1a, b are invalid.

Regeneration experiments

The authors report that they attempted to simulate and monitor regeneration in Olyset nets by testing them and then washing them three times. Repeated washing with no time allowed for insecticide replenishment is not a practice that replicates field conditions and additionally may adversely affect the performance of Olyset nets. It is of concern that it appears that the results shown in Table 2a, b are from just one sample each at the two temperatures of 30 and 35 °C. Each sample was 30 × 30 cm2. Using just one sample for each temperature means that the relatively small net sample is handled frequently, resulting in insecticide depletion. These samples are therefore not representative and should not be scientifically acceptable due to lack of replication.

The results reported show poor mortality, however, as was described fully in the comments submitted against paper 1 the test method used by the authors was inappropriate in that mortality of mosquitoes alone is not the key criterion to be measured, as communicated by Dr. P Guillet above.

In Table 2, we see some strange variability in results between the two net samples, which are difficult to explain and cannot be commented on further because replication was so small that any conclusion would be flawed. The substantial variations in KD results when there has been no washing of the net are also a puzzle that might be explained by variability in test mosquitoes or the test method.

The fact that mosquito KD is occurring confirms our contention that protection from mosquitoes would be ongoing. If mosquitoes were detecting insecticide and being KD then they would not attempt to enter or feed through a net. This phenomenon equals protection and shows that mosquito mortality alone is not the key criterion.

Chemical analysis

The authors repeated the error (see comments submitted against paper 1) of using GC as the analytical methodology to measure deltamethrin on treated nets. This almost certainly would result in misleading figures and underestimates of the amount of deltamethrin on those nets. GC may however be used to detect permethrin in Olyset nets. Underestimating the levels of deltamethrin on Permanet, e.g., may therefore indicate it has better efficacy than it in fact has if it actually contained a much higher dose than was detected.

Note that the figures reported in the paper show that Permanet lost 61% of its original dose rate after 20 washes whereas Olyset only lost 19.4%– perhaps indicating the substantial difference in the two techniques in respect to the ease with which the insecticide can be dislodged from the nets during washing and handling.

Conclusion

  1. Top of page
  2. Reference: Lindblade et al. (2005) Evaluation of long-lasting insecticidal nets after 2 years household use. Tropical Medicine & International Health 10, 1141–1150
  3. Methods
  4. Study design
  5. Reference: J Gimnig et al. (2005) Laboratory wash resistance of long-lasting insecticidal nets. Tropical Medicine & International Health 10, 1022–1029
  6. Conclusion
  7. References

The authors conclude by saying ‘The wash-resistance of the LLINs tested correlated well with their performance under field conditions and laboratory studies may serve as rapid tools for predicting long-lasting efficacy of candidate LLINs under field conditions’. After reviewing both papers we conclude that this statement is wrong. It has been well established that there is little to replace the evaluation of net technologies by hut trials or large-scale field trials where both entomological and epidemiological parameters are examined. Limiting evaluations and conclusions to a single inappropriate methodology is both scientifically flawed and misleading.

Yours sincerely,

References

  1. Top of page
  2. Reference: Lindblade et al. (2005) Evaluation of long-lasting insecticidal nets after 2 years household use. Tropical Medicine & International Health 10, 1141–1150
  3. Methods
  4. Study design
  5. Reference: J Gimnig et al. (2005) Laboratory wash resistance of long-lasting insecticidal nets. Tropical Medicine & International Health 10, 1022–1029
  6. Conclusion
  7. References
  • Chandre F, Darriet F, Duchon S et al. (2000) Modifications of pyrethroid effects associated with kdr mutation in Anopheles gambiae. Medical and Veterinary Entomology 14, 8188.
  • Curtis CF, Maxwell CA, Finch RJ, Njunwa K (1998) A Comparison of pyrethroid either for house spraying or for bednet treatment against malaria vectors. Tropical Medicine & International Health 3, 619631.
  • Graham K, Kayedi MH, Maxwell C et al. (2005) Multi-country field trials comparing wash-resistance of Permanent and conventional insecticide-treated nets against anopheline and culicine mosquitoes. Medical and Veterinary Entomology 19, 7293.
  • WHO (1996) Pesticide Evaluation Scheme Report of WHO informal Consultation on the evaluation and testing of insecticides. CTD/WHOPES/IC/96.1.
  • WHO (2005) Guidelines for laboratory and field testing of long lasting insecticidal mosquito nets. WHO/CDS/WHOPES/GCDPP/2005.11.