Tsetse Control and the Elimination of Gambian Sleeping Sickness.

Sleeping sickness, or Human African Trypanosomiasis (HAT), is caused by two distinct parasites. In East and Southern Africa, Trypanosoma brucei rhodesiense causes the Rhodesian form of the disease (about 2% of all reported cases [1]). In Central and West Africa, T. b. gambiense causes the Gambian form of the disease (G-HAT—about 98% of all reported cases [1]). The disease normally affects remote rural communities. The people most at risk are those working outdoors for long periods, as they are most exposed to the bite of the tsetse fly (Glossina spp.: Diptera), which transmits the parasites. The comparable diseases which occur in livestock, collectively termed African Animal Trypanosomiasis (AAT), are a significant brake on African development [2]. Among the 31 tsetse species, the most important vectors of G-HAT are Glossina fuscipes and Glossina palpalis, which are riverine tsetse species (Palpalis group).Since the start of the 20th century, HAT has occurred in three huge epidemics. The most recent was in the 1990s when the annual cases officially reported to WHO peaked at 37,385 in 1998. It is widely acknowledged this severely underestimated actual numbers infected, which may have been as high as 450,000 in 1999 [3]. Untreated disease is normally fatal, so undoubtedly, many people infected in these epidemics died as a result. Although treatments for the disease have improved [4], they are still complex and difficult to administer particularly in the resource-poor settings where the disease thrives. There is no vaccine or chemoprophylaxis to prevent HAT and little prospect of either being developed in the near future. Vector control therefore remains the only means of protecting people from infection.Rhodesian HAT (R-HAT) is a zoonosis. As a consequence, vector control plays a key part in its control, and medical interventions are only used for humanitarian purposes. In contrast, G-HAT is generally considered to be an anthroponosis, and control has relied heavily on active and/or passive case detection and treatment programmes [5]. However, modelling [6], historical investigations [7], and practical interventions [8,9] have clearly demonstrated the role that vector control can play in control of G-HAT, but it was considered too expensive and difficult to deploy in the resource poor settings of HAT foci. In consequence, a study was started in 2006 to try to find a simpler and cheaper alternative for vector control suitable for G-HAT foci.The original hypothesis was that modifying insecticide-treated targets was the most likely means of producing a more cost-efficient vector control method for use in G-HAT foci. Two separate approaches were tried—to develop odours for use with targets or to change the visual characteristics of the target. The crucial finding was that a tiny target consisting of a small square of blue cloth flanked by a similar sized piece of black netting (Fig 1) was highly effective and would be about ten times more cost-effective than traps or large targets in control campaigns for the Palpalis (riverine) group tsetse flies responsible for the transmission of the vast majority of HAT [10–12]. This is in very strong contrast to Morsitans (savanna) group tsetse flies, which require much larger targets (1–2 m2). Importantly, it was found that all of the major G-HAT vectors responded well to tiny targets [13]. In addition, vegetation growth around tiny targets is a much smaller problem [14] than is the case for the large targets used against Morsitans group flies. In contrast to Morsitans group flies, odours seem to play only a minor role in the attraction of Palpalis group flies [15,16]. A modelling approach suggests that habitat geometry is the reason why Palpalis group flies are more dependent on sight than odour [17]. The general expectation is that relatively immobile insects in restricted habitats are more dependent on a thorough, vision-based search of their environment and that they are more wide-ranging in their diet.Fig 1A tiny target in a typical setting in Uganda.Inevitably, the targets are gradually degraded by challenges in the environment, and the worst problems are floods, fallen targets, and the 6-month effective life of the insecticide in the tiny target [18]. As a consequence, current practice has been to deploy tiny targets once or twice per year, and the method has been successful in practice [9,18].The aim in HAT foci is not to eradicate tsetse (although eradication should be embraced if feasible), but to stop transmission by reducing tsetse—human contact, and modelling suggests that this does not require complete removal of tsetse flies [6]. In addition, the reported time course of disease in humans is typically 3–4 years so that a fixed period of interrupted transmission may be sufficient to eliminate HAT in a focus. This approach is basically similar to the successful World Bank-funded OCP programme, which has led to the elimination of onchocerciasis as a public health problem in West Africa [19]. The approach had also been applied successfully in HAT foci of Ivory Coast in the 1980s and 1990s by Laveissiere and colleagues [8], although at that time the control techniques used were not considered to be sustainable and cost effective.Consequently, to test the utility of tiny targets, studies were started in G-HAT foci (typically 500–3,000 km2). To re-emphasise, the goal is to reduce tsetse numbers below a threshold for transmission for a defined period to either eliminate or reduce transmission in a HAT focus, thereby giving screen-and-treat programmes a far greater chance of success [20]. For example, a previously published model [6] has been used along with figures from West Nile, Uganda to calculate the impact of various levels of vector control on transmission in that region (Fig 2) [18]. In practice, the level of control actually achieved in that region was u003e90%, which exceeds the levels required to interrupt transmission (Fig 2) [18]. How long control must continue is a researchable question but, given the time course of the disease in humans, it is likely to be several years. Presumably, it is also dependent on the distribution of the parasite in the human population and/or the existence of reservoir hosts. Current discussions have been focusing on 4–5 years of control.Fig 2To obtain an estimate of the level of tsetse control required to stop transmission, a published model was rearranged [6].