Nets used in the study
Four brands of LLINs were evaluated. The nets were randomly selected from a consignment supplied by different manufacturers for field distributions. A total of 20 net samples were used as follows, four from each treatment group:
(i) Olyset (Sumitomo Corp.) production licensed to A - Z Tanzania. These are polyethylene-based nets. Treatment is done at the factory, with a synthetic permethrin @ 1,000 mg/m²
(ii) PermaNet® 2.0 (Vestergaard-Fransden, Denmark). These are polyester based nets and treatment is applied topically at the factory with deltamethrin @ 55 mg/m²
(iii) Interceptor (BASF, Germany) also polyester based. Treatment is done at the factory with alphacypermethrin @ 200 mg/m².
(iv) NetProtect (TNT) (Intelligent Insect control Co. France). They are polyethylene based and treatment is accomplished at factory with deltamethrin @ 65 mg/m². The production of these nets is licensed to Siva Enterprises of India.
A conventionally treated net with deltamethrin @ 25 mg/m² was used as a control. Nets were tested for bioactivity before washing to determine the baseline efficacy and thereafter they were tested after every 5th wash through wash 20.
Washing of nets by hand
The nets were hand washed and dried outdoor at KEMRI Centre for Global health research in Kisian village, western Kenya. Four field assistants from the local community were hired to do the washing. Hand washing was done by immersing the netting in a measured volume of water using a measured local detergent Omo. The field assistants were randomly assigned to wash the four brands of nets by hand rubbing in a 10-litre water bowl. Nets were washed for 10 minutes in 2 litres of cold rain water mixed with 5 g of detergent. After washing each net was rinsed twice for 5 minutes in same amount of clean water. Each net was washed twice a week. Washing was done mid-morning between 9 to 10 am.
Net drying procedures
The four different drying regimens were tested only on nets that were washed by hands. It is important to note that this is the most common washing method used in the local villages. After washing, the nets were air-dried as follows: one net from each treatment group was air-dried in the sun by hanging, while the second net was air-dried in direct sunlight spread on the ground. The third net from each group was air-dried under the shade spread on the ground and fourth net from each treatment group was dried under the shade hanging on a line. The nets were left in position to dry for a fixed period of 4 hrs. Initially before adopting a standard drying time, nets were inspected hourly to ascertain their drying status. It was established that 4 hours was the adequate period in the study area for nets to dry completely, whether in direct sunlight or under the shade. This period was adopted as the optimal time nets needed to dry and was used throughout the study
Washing of nets by machine
In this method, the net samples were washed using WHO protocol as follows: 1 gram of soap OMO powder was thoroughly dissolved in 500 ml of rain water in 1 L Erlenmeyer conical flask. A single sample of netting 30 by 30 cm was placed in the soap solution. The nets and soap solution were shaken for 10 min on an orbital shaker bath (C76 Water Bath Shaker; New Brunswick Scientific Co., Edison, NJ, USA) at 155 rotations per minute at room temperature. After washing with soap, the nets were rinsed twice by shaking for 10 min in 500 ml of rain water each. After the nets were rinsed a second time they were hung on a line indoors to dry for 4 hrs.
Washing by beating on rocks
In this experiment, 5 grams of OMO detergent was thoroughly mixed with 2 litres of rain waster. Each net was individually immersed in soap solution in a 10-litre water bowl. Nets were then removed from the water bowl then beaten against a concrete slab several times for a total of 10 minutes. Between these beatings, the nets were momentarily soaked in same soap solution for a few seconds when they appeared to run out of dipping water. This was to simulate what happens in the local village. After washing with soap, each net was then rinsed twice for five minutes in same amount of clean water. Each net was washed twice a week. Washing was done mid-morning between 9 to 10 am. After washing and rinsing, nets were dried by hanging under the shade for 4 hours.
Laboratory procedure
Four replicate net samples in templates measuring 30 × 30 cm were cut from each net brand and tested. In total, 20 samples were tested. Each net sample was subjected to a baseline bioassay before washing was started. After the baseline bioassay, samples were washed according to either the standard WHO protocol or a cording to local methods. Thereafter, the samples were washed twice a week. Subsequent bioassays were carried out after every 5th wash on each 2nd day after the previous washing till the 20th wash. Tested samples were stored until after the end of the study for chemical residue analysis. The same samples were used for bioassay and chemical residue analysis. Tested samples were taken as follows; one from the top and three from each of the four sides at random
Bioassays procedures
Four WHO cones were attached to each net sample and a total of 10 mosquitoes were introduced into each cone. 2-5 day old, unfed female laboratory reared susceptible Kisumu strain of Anopheles gambiae mosquitoes was used. At least 40 mosquitoes were exposed on each net for 3 min and then transferred to holding paper cups and provided with 5% sugar solution soaked in absorbent cotton pad. Knockdown was recorded after 30 minutes and 60 minutes after exposure. Mortality was recorded 24 hrs after exposure. Mosquitoes were considered knocked down or dead if they could not fly and could not stand upright on either the side or the bottom of the paper cups. Untreated net was used as a control and was tested each day the bioassay was performed. Tests were done at temperatures of 25°C ± 2 and 80% ± 10 relative humidity.
Residual insecticide quantification
After bioassay the second piece cut from each net were individually labelled with the name of treatment group and number of washes and stored in the dark for subsequent residual insecticide quantification by high performance liquid chromatography (HPLC), [19]. For ease of identification the nets were marked with indelible ink using permanent markers before washing was started. The insecticide content of each net was determined by cutting from each net piece, smaller pieces of 2 by 2 cm then extracting the insecticide into a solution using a mixture of solvents. For deltamethrin and alphacypermethrin iso-octan plus 1, 4 dioxan with 0.15% HPLC grade water was used. Dibutyl phthalate was added as the internal standard. For permethrin acetone, 99.9% HPLC grade was used with 99.93% methyl alcohol as the internal standard. After extraction samples were thoroughly shaken to mix and then filtered by water pump suction on 0.45 micrometre membrane filter into a vial. An aliquot of 1 µL of the filtered solution was then injected onto a normal phase isocratic HPLC machine with a UV detector. The insecticide quantification was achieved using an internal calibration curve.
Statistical analysis
Statistical analyses were conducted with SAS, version 9.2. Insecticide residue on the nets after repeated washing was modelled with linear regression. Covariates which were included in this model were the number of washes, drying regimens (shade and spread on ground, sun and spread on ground, shade while hanged and sun while hanged). Mosquito mortalities were calculated using probit regression analysis with machine wash as the reference. Covariates which were included in the model were the number of washes and the washing procedures (washing by hand, machine washing and washing by beating on rocks).