Insecticide impregnated curtains to control domestic transmission of cutaneous leishmaniasis in Venezuela: cluster randomised trial

Study area

Trujillo, Venezuela, is 800 metres above sea level, has 33 399 inhabitants, and is divided into 22 sectors. It has an annual average temperature of 23.3½C and two annual rainfall periods (July and November) of 750 mm each. The city has many green areas where opossum and other woodland reservoir hosts of Leishmania live.

The research centre of the university is the main provider of clinical services for cutaneous leishmaniasis. It has identified the vectors15 and Leishmania species 16 17 and has mapped sectors of high and low risk.16 We carried out this study in high risk areas only, where the estimated annual incidence of cutaneous leishmaniasis was above 0.5% according to the epidemiological data collected by the centre during the past three years.

Cluster randomised controlled trial (July 2000 to July 2001)

Baseline household survey (July 2000)—We included 14 sectors of the city, with 569 houses and 2913 inhabitants. The original sample was 578 and the non-response rate was only 1.6%. Researchers used a questionnaire at the end of the first annual rainy season to gather data on the demographic (age and sex) and socioeconomic characteristics (formal education, occupation, housing conditions) of the population as well as on people's individual experience with cutaneous leishmaniasis. Additionally, the interviewers examined any scars of lesions. They paid particular attention to “new” cases that had occurred during the 12 months before the interview. Subsequently interviewers checked for any new cases that had been reported in the leishmaniasis clinic in the past 12 months. The consistency of interview answers and clinic records was excellent. Only 25 additional cases, which had been reported in the interviews, could not be identified as these patients had gone to the public hospital. In an observational study researchers visited 50 sample households in the early evening to observe where men, women, and children tended to be at the time when the sandflies were expected to enter the houses.

Entomological baseline study (January to June 2000)—As the lutzomyia sandflies are phototactic 5 15 from 6 pm to 7 am we placed light traps in the main room of 565 houses (one trap per house per night) for 150 nights. The sandfly species were determined by professional entomologists.

Randomisation and provision of curtains—We followed the methods of evaluating health interventions at area level18 and paired the 14 sectors according to the incidence of cutaneous leishmaniasis in the 12 months before the baseline household survey. For each of the seven pairs we randomly allocated one sector (using computer created random numbers) to the intervention group and the other to the control group, the population being “blind” towards the group allocation. In the intervention group the windows of all 241 houses (with a total of 1336 inhabitants) were covered with loosely hanging polyester curtains impregnated with the pyrethroid insecticide lambdacyhalothrin (ICON 2.5CS, Syngenta, Basle) with a target concentration of 12.5 mg/m². The mesh size of the curtains was 0.05 mm. After six months all curtains in intervention households were impregnated again. In the 222 houses in six of the control sectors the windows were covered with non-impregnated curtains and in one randomly selected control sector (in pair 7, see table 1) with 106 houses no curtains were provided so we could compare abundance of sandflies in houses with non-impregnated curtains and no curtains at all.

Entomological follow up study (August to October 2000)—We placed light traps in the same houses and in the same way as in the baseline study. There were 239 houses with impregnated curtains (intervention group) and 220 with non-impregnated curtains and 106 houses without curtains (control groups).

Follow up household survey (August 2001)—Twelve months after the baseline survey (again at the end of the rainy season) we used the same questionnaire in the same houses using the same procedure of data collection and quality control. The study populations in the baseline and follow up studies were almost the same: only 15 additional people in the intervention group and 10 in the control group were included in the follow up study.

Data analysis—We entered the survey data on to a computer using EpiInfo v6.4 (CDC, Atlanta, GA, USA) and used EpiInfo, SPSS, and Stata v6 (Statacorp, TX, USA) for analysis. Before the main analysis we used Fisher's exact tests to compare cumulative incidence between intervention and control sectors for each pair. We used cumulative incidence rates of cutaneous leishmaniasis and the average number of flies per trap (house) for each sector as the units of analysis. We compared data at baseline and then at follow up between the intervention and control groups using a paired t test, weighting the data according to the sector size. We also used Wilcoxon's matched pairs test because the small number of pairs made it difficult to assess whether the underlying distribution of the differences was normal (necessary for the validity of the t test), and the Wilcoxon test does not require this assumption. Differences rather than ratios are presented as the estimates of effect because zeroes for the main outcome, cutaneous leishmaniasis, precluded the use of ratios. As one of the control sectors (pair 7) did not receive any curtains, whereas the other control sectors received curtains without insecticide, we carried out separate analyses including and excluding pair 7.

Characteristics of population

The demographic and socioeconomic indicators pointed to moderate levels of poverty: 31% of the population were children (<15 years old) and only 9% were 60 years and older; crowding was high (five people per household) but almost all houses had television. Of those people aged over 14 years, 23% had no school education. Within the same age group 21% were engaged in domestic activities, 21% were students, 13% were manual workers, self employed artisans, or secretaries, 7% were unemployed, 7% had an academic profession, and only 2% were farmers.

Baseline epidemiology

The cumulative rate of non-mucosal cutaneous leishmaniasis (that is, the proportion of people who reported having had cutaneous leishmaniasis once in their lives) was 14% (408/2913). The most common sites of lesions were the legs and arms (79%). In 88% of the cases the diagnosis was confirmed by a doctor.

All occupational groups, working either at home or in or around the city, showed only small differences in the cumulative rate of cutaneous leishmaniasis (roughly 23% in all groups) and the difference between males (13%) and females (15%) was not significant. This points to transmission being mostly within and around the domestic setting, with little protection according to different levels of housing. Only the small group of farmers on the edge of the urban areas seemed to have an additional risk of infection (15/35, 43%).

Infection rates were low among the children aged <5 years (3%) and schoolchildren (8%), probably because of the shorter exposure time and the particular protective measures taken with young children. Rates were higher in adults (17%) and in people aged over 60 years (21%) (P<0.01 for difference between children and adults).

The incidence rate of cutaneous leishmaniasis in the 12 months preceding the interview was 4% (126/2913). The highest number of new cases was registered in July and November, at the end or soon after the rainy season.

Entomological survey

In the baseline study we captured 8962 sandflies in the 565 houses (on average 16 per trap per night); 63% of these were principal vectors (Lutzomyia youngi and L ovallesi) and 37% other vectors (L scorzai, L trinidadensis, L venezuelensis, L walkeri, L otroclavata, L lichyi).

Impact of impregnated curtains—Table 1 shows the mean number of phlebotomine sandflies per trap before and after the placement of impregnated and non-impregnated curtains. There were no significant differences between the control and intervention groups before the placement of the curtains but afterwards significantly fewer sandflies per house were observed in the sectors with impregnated curtains (mean difference per sector 13 sandflies per trap (95% confidence interval 9 to 17; P<0.001) compared with those with unimpregnated curtains (see table 3).

Exposure to sandflies (observational study)—Family life in the early evening is centred in the living room around the television. Children under 5 years are put to bed between 8 pm and 9 pm and schoolchildren around 9 pm. Between 9 pm to 10 pm the main door is closed and the parents retire to bed. The windows are generally left open during the night, exposing people to the sandflies.

Protective efficacy of curtains—Our main outcome was the 12 month cumulative incidence of cutaneous leishmaniasis. Table 2 shows incidence for each pair of sectors at baseline and after the intervention. There were no significant differences in incidence of cutaneous leishmaniasis between the groups before placement of the curtains. After the intervention the incidence in the group that received impregnated curtains was 0 and significantly lower than the group which received unimpregnated curtains (mean difference in 12 month incidence 8 cases per sector; 4 to 12; P=0.001) (table 3).

Side effects

We observed no side effects of the pyrethroid impregnated curtains and none was reported in the interviews in the follow up study.