MATERIALS AND METHODS

Study area - The study was carried out over a period of 14 months (January 1994 - February 1995) in the location of Ocamo (02º50'N, 65º14'W) in the state of Amazonas, Southern Venezuela (Fig. 1). Ocamo comprises nine villages or "shabonos" (Santa Maria de los Guaicas-Ocamo, Boca Padamo, Dayaritheri, Lechoza, Kashora, Tumba, Shashana, Yohope, and San Benito) located along the Orinoco and Ocamo rivers, the distance between villages being about 2 km. The villages are about 116 m above sea level. The mean annual temperature was 24ºC and 80% relative humidity, with an annual rainfall of 2487 mm (MARNR 1995). The area is classified as lowland forest (Huber 1995) and interior evergreen and semi-evergreen forest (Osborn et al. 2004). From the eco-epidemiological point of view the area has been classified as lowland interior forest malaria (Rubio-Palis & Zimmerman 1997).

In Santa Maria de los Guaicas (Ocamo), the main identified anopheline larval habitats were two permanent large lagoons located about 400 m from dwellings, confirmed by subsequent studies (Rejmánková et al. 1999, Rubio-Palis et al. 2005). The population is Amerindian of the Yanomami ethnic group. The type of houses varies among "shabonos" from a large circular shelter with no walls and thatched roof to houses with mud walls and corrugated iron roofs. Members of the same family group hang their hammocks around a fire, and there may be up to eight family groups (eight fires) under the same roof. The fire burns day and night, so that houses with walls are very smoky.

The "shabonos" have not been subject to insecticide spraying since 1993.

Demographic surveillance - A population census was conducted during June 1994. In each village, dwellings were given a code number and numbers of fires or family groups recorded. In each family group every resident's Yanomami and/or "nape" (foreign) name, approximate age and sex were recorded. All births and deaths occurring during the study were recorded.

Malaria diagnosis and treatment - Under the Primary Health Care (PHC) system, active and passive case detection was routinely conducted in the villages. The active case detection was made twice a month in each "shabono". Thick and thin blood films were routinely taken from a person reporting fever at the moment of the visit or during the previous three days and children under six years of age with diarrhea or vomiting. After fixation with methanol, films were stained with Giemsa and observed by the authors. One hundred high-power fields of each thick film were examined and the parasite species determined. Confirmed malaria cases were treated according to the drug schedule approved by the National Malaria Control Program for the Upper Orinoco region: patients with uncomplicated P. falciparum malaria were given 25 mg/kg pyrimethamine-sulfadoxine-mefloquine, the dose calculation was based on the concentration of mefloquine; severe and complicated P. falciparum malaria cases were treated with a 20 mg/kg initial dose of quinine and subsequent doses of 10 mg/kg, 10 mg every 8 h for seven days; patients with P. vivax were given 25 mg/kg of chloroquine divided in three doses (10 mg/kg, 10 mg/kg, and 5 mg/kg) plus primaquine 0.5 mg/kg for five days. P. malariae cases were treated with 25 mg/kg of chloroquine divided in three doses. Primaquine and pyrimethamine-sulfadoxine-mefloquine was not given to infants and pregnant women. The later were treated with quinine. A new episode of slide-positive malaria was considered to be a relapse or recrudescence if it occurred within six months (P. vivax), or one month (P. falciparum) after a previous episode of acute malaria diagnosed during the study.

Malariometric parameters considered were: annual parasite index (API = malaria cases divided by the population under surveillance multiply by 1000), slide-positive rate (SPR = percentage of slides positives for malaria), and slide-positive percentage by species (percentage of parasite of each Plasmodium species). The association between malaria cases and age group was determined using a logistical regression model in Stata (version 7.0; Stata Corporation, College Station, TX, US). The age group of children under 10 years was considered as the base line.

Mosquito collections - Human landing catches of anopheline mosquitoes were carried out inside selected dwellings in the village of Santa María de los Guaicas. The houses had thatched roofs and incomplete mud walls. Collectors worked in pairs catching with a mouth aspirator all mosquitoes landing on a person resting inside his hammock, from 19:00 to 06:00 h, five to eight nights per month for 13 months between February 1994 and February 1995. Captured mosquitoes were placed in paper cups, a new cup being started every hour. After identification, mosquitoes were counted and stored over silica gel at room temperature until assayed by enzyme-linked immunosorbent assay (ELISA). Climatological data were obtain from the Ministerio del Ambiente y los Recursos Naturales Renovables (MARNR) weather station located in Santa Maria de lo Guaicas.

Mosquito preparation and ELISA - Abdomens, wings, and legs of dried female Anopheles were removed to reduce the risk of detection of circumsporozoite (CS) antigen from parts of the body other than the salivary glands. Mosquitoes of the same species and date of collection were combined in pools of up to 10; this criteria is accepted in areas where infections rates are less than 1%, because it is highly unlikely that more than one mosquito is infected in one pool. The ELISA protocol described by Wirtz et al. (1987, 1992) was followed to detect CS proteins of P. falciparum, P. vivax-210, and 247 polymorphs and P. malariae. Positive and negative controls were run on every ELISA test plate. Positive controls consisted of 0.1ng of recombinant P. fal-ciparum CS protein, 0.04 ng of P. vivax-210 CS protein, 1.25 ng of P. vivax-247 CS protein, and 250 ng of P. malariae CS protein. Laboratory-raised An. albi-manus males were used for negative controls in all assays. All readings were analyzed at 405 nm recorded with a Molecular Devices ELISA plate reader 30 min after the addition of the ABTS substrate. Samples were considered positive upon confirmation if absorbance values exceeded two times the mean of seven negative controls.