[Risk evaluation in sleeping sickness: a mathematical contribution]. / Evaluation du risque dans la maladie du sommeil: apport de la modélisation mathématique.

The concept of "risk" is important in epidemiology but is often used in a confused way for sleeping sickness. Using a rigorous approach resulting from mathematical modelling, new virtual entomological risk indicators and parasitological transmission indexes derived from the basic reproduction rate R0 are proposed and discussed.

[New data on parasitization of Glossina palpalis palpalis (Diptera: Glossinidae) by Hexamermis glossinae (Nematoda: Mermithidae) in a forested area of Ivory Coast]. / Nouvelles données sur le parasitisme de Glossina palpalis palpalis (Diptera: Glossinidae) par Hexamermis glossinae (Nematoda: Mermithidae) en secteur pré-forestier de Côte d'Ivoire.

Observations on the parasitism of Glossina palpalis palpalis by Hexamermis glossinae were carried out over a period of one year by catching flies at Abengourou, Aboisso and Daloa (forested area of Ivory Coast). No parasite is observed out of 2,168 Glossina palpalis palpalis caught in Abengourou and 9,732 in Aboisso. At Daloa, dissections of 7,341 Glossina reveal 1.75% parasited flies. All the worms were located in the abdominal cavity, loosely intertwined with the internal organ. Males were more infested than females (2.68% versus 1.26%). The parasites were more abundant among the nulliparous (2.30%) than the young parous (1.19%) and than the old parous (0.52%). The majority of infected flies were caught at the beginning of the rainy season (5.17%) and few in the dry season (0.23%). The low parasitic infection rate observed here indicates a minimal effect on the population dynamics of the vector of sleeping sickness in Ivory Coast.

Estimating tsetse population parameters: application of a mathematical model with density-dependence.

A density-dependent model is used to describe the dynamics of an open population of tsetse flies (Diptera: Glossinidae). Immigration (or emigration) takes place when the total population is below (or above) a biologically determined threshold value. The population is also subjected to birth and death rates, as well as to the risk of being trapped (continuously or intermittently). During trapping the population decreases toward a 'low' equilibrium population and when trapping ceases the population starts recovering and increases toward a 'high' equilibrium. The model is fitted using data collected on trapped flies in four experiments. The first one was conducted with 'intermittent trapping' (i.e. several trapping-recovery cycles) on Glossina fuscipes fuscipes Newstead in the Central African Republic (Bangui area). In the other experiments, trapping data on Glossina palpalis palpalis (Robineau-Desvoidy) was collected in 'aggregate' form over several days at a time. Two of these were in Congo-Brazzaville (Bouenza area) and one in the Ivory Coast (Vavoua focus). Estimates are derived for the low and high equilibrium values as well as the trapping rate. The estimated effect of sustained trapping is to reduce the population to low equilibrium values that are 85-87% lower than the levels without trapping. The effects of the natural intrinsic growth and of the migration flows cannot be estimated separately because in the model they are mathematically indistinguishable.

[Reproductive disorders in Glossina palpalis palpalis (Diptera: Glossinidae) in forested areas of Ivory Coast]. / Anomalies de la reproduction chez Glossina palpalis palpalis (Diptera: Glossinidae) en zone forestière de Côte d'Ivoire.

Study of reproductive disorders were carried out through the dissection of 11,012 tsetse flies caught over a period of one year in forested different habitats of Glossina palpalis palpalis of Daloa in Ivory Coast. The proportion of females with reproductive disorders was very low and estimated at 0.79%. Out of 87 tsetse files with reproductive disorders, 93.10% were abortions, 5.77% were ovular blockage and 1.13% was uterine pupaison. Reproductive disorders were recorded from all age groups: 0.78% in young parous (out of 6,398 tsetse flies examined) and 0.80% in old parous (out of 4,614 tsetse flies dissequed). Our results show that reproductive disorders occur at any stage of the female pregnancy cycle. Amplifying these reproductive disorders using chemical compounds is proposed as a way of improving the efficacy of insecticide-impregnated targets (pour-on, traps and screens) of tsetse control in rain-forest areas.

A model of Gambian sleeping sickness with open vector populations.

A compartmental model of Gambian sleeping sickness is described that takes into account density-dependent migratory flows of infected flies. Equilibrium and stability theorems are given which show that with a basic reproduction number R0 below unity, then in the absence of reinvasion the disease goes to extinction. However, even a low prevalence rate among reinvading flies can then bring about significant equilibrium prevalence rates among humans. For a set of realistic parameter values we show that even in the case of a virulent parasite that keeps infected individuals in the first stage for as little as 4 or 8 months (durations for which there would be extinction with no infected reinvading flies) there is a prevalence rate in the range 13.0-36.9%, depending on whether 1 or 2% of reinvading flies are infected. A rate of convergence of the population dynamics is introduced and is interpreted in terms of a halving time of the infected population. It is argued that the persistence and/or extension of Gambian sleeping sickness foci could be due either to a continuous reinvasion of infected flies or to slow dynamics.

A density-dependent model with reinvasion for estimating tsetse fly populations (Diptera: Glossinidae) through trapping.

A simple density-dependent reinvasion model is described and used to estimate tsetse fly populations on the basis of removal trapping experiments. The model was tested on Glossina fuscipes fuscipes Newstead in the Central African Republic and G. palpalis palpalis (Robineau-Desvoidy) in the Republic of Congo (Brazzaville). The density-dependence is modelled by postulating that the inflow of flies each day is proportional to the deficit relative to the equilibrium population. Non-linear least square techniques were used to estimate the following parameters: the daily capture rate, the strength of the density-dependence, and the equilibrium fly population, at the beginning and at the end of the trapping experiment. The model ignores birth and death rates of flies and is applicable only when a rapid decrease in population occurs over a short period (between 10 and 20 days). Over longer periods one could not ignore the natural growth of the populations as well as other more complex density-dependent mechanisms.

[Persistence and resurgence of sleeping sickness caused by Trypanosoma brucei gambiense in historic foci. Biomathematical approach of an epidemiologic enigma]. / Persistance et résurgence de la maladie du sommeil à Trypanosoma brucei gambiense dans les foyers historiques. Approche biomathématique d'une énigme épidémiologique.

Since the end of the 19th century, historic endemic foci of Trypanosoma brucei gambiense sleeping sickness have proven very persistent. A five-compartment mathematical model with open vector populations was developed in order to study the dynamics of this disease in Central Africa. Of particular interest is the rate at which the disease spreads or goes to extinction at the beginning of an epidemic outbreak. A measure of this rate is the initial halving/doubling time T(o) of the numbers infected; T(o) is a doubling time when the basic reproduction number Ro > 1 and a halving time when Ro < 1. For realistic parameter values, T(o) can be quite large (i.e. several years or even decades) which corresponds to a persistent low-level endemic brought about by an Ro either just above 1 (slow spread) or just below 1 (slow extinction). A resurgence of historical foci can then be caused by a small shift in parameter values that brings Ro well above 1 and decreases T(o). In addition, when Ro is less than 1 (in the absence of vector migrations), simulations show that a very small percentage of infected immigrant flies can bring about high prevalence rates in the human population. The model is validated with field data from historical Congolese, Central and West African foci of the past.

A two-patch model of Gambian sleeping sickness: application to vector control strategies in a village and plantations.

A compartmental model is described for the spread of Gambian sleeping sickness in a spatially heterogeneous environment in which vector and human populations migrate between two "patches": the village and the plantations. The number of equilibrium points depends on two "summary parameters": gr the proportion removed among human infections, and R0, the basic reproduction number. The origin is stable for R0 < 1 and unstable for R0 > 1. Control strategies are assessed by studying the mix of vector control between the two patches that bring R0 below 1. The results demonstrate the importance of vector control in the plantations. For example if 20 percent of flies are in the village and the blood meal rate in the village is 10 percent, then a 20 percent added vector mortality in the village must be combined with a 9 percent added mortality in the plantations in order to bring R0 below 1. The results are quite insensitive to the blood meal rate in the village. Optimal strategies (that minimize the total number of flies trapped in both patches) are briefly discussed.

Control strategies for sleeping sickness in Central Africa: a model-based approach.

Vector control and the detection (followed by treatment) of infected individuals are the two methods currently available for the control of sleeping sickness. The basic reproduction rate of a compartmental model is used to analyse and compare the two strategies. The efficiency of each strategy will depend on two epidemiologic parameters; the intrinsic contamination rate Q (closely related to the index of new contaminations) that captures the potential spread of the disease, and the intrinsic removal rate from the first stage (intrinsic to the particular trypanosome strain and to the population's susceptibility). The model shows that when the intrinsic removal rate is low (that is, when there is a long first stage characteristic of an endemic situation) the detection of sick individuals is more efficient than vector control. The situation is reversed when the removal rate is high (in an epidemic situation). The conclusions of the analysis are shown to be in general agreement with results obtained in two different sleeping sickness foci of Central Africa.

[Insecticide treatment (Pour on and Spot on) of cattle against Glossina fuscipes fuscipes in the Central African Republic]. / Traitement épicutané (Pour on et Spot on) du bétail contre Glossina fuscipes fuscipes en République centrafricaine.

Four herds of Mbororo Zebu cattle (approximately 40 head each) in traditional situation of the Fulani breeders were treated over a 12 month period. Flumethrin Pour on was first used every three weeks during the rainy season and then deltamethrin Spot on was used every six weeks during the dry season. Pour on treatments might affect the age structure and feeding patterns of the neighbouring populations of G fuscipes. However, the treatments had no impact on the apparent density of flies. They changed neither the prevalence of trypanosome infection rates in cattle nor the hematocrit levels. This trial shows that, under these experimental conditions, insecticide treatments were not effective for the control trypanosomosis in cattle. Further trials should be carried out to assess the usefulness of this method when it is integrated with the trapping of tsetse flies.

Are stable flies (Diptera: Stomoxyinae) vectors of Trypanosoma vivax in the Central African Republic?

The epidemiology of Trypanosoma vivax infections was studied at a riverside site in the Ouro-Djafoun livestock area situated in the Central African Republic during the period between July 1991 and July 1992. This paper examines the possibility that stable flies (Diptera: Stomoxyinae) were also vectors of this trypanosome species in a non-cyclic way. Previous studies have revealed that the usual cyclic transmission by the tsetse fly Glossina fuscipes fuscipes was probably not the only transmission route. At the study site, at least five species or subspecies of stable flies were encountered: Stomoxys nigra nigra (approximately 60% of the sample), S taeniata, S sitiens, S omega omega and Haematobia spp. The hypothesis that stable flies could be good vectors of T vivax in this country is supported by three main observations: i) stable flies were very abundant at the cattle resting site; ii) an estimation of the 'contact index' between the cattle and stable flies demonstrated close interactions between cattle and stable flies at this site, particularly during the rainy season, and iii) there was a good correlation (P < 0.05) between the apparent stable fly densities at the resting site and the frequency of T vivax in the cattle. The relevance of this phenomenon in terms of epidemiology and combatting T vivax-caused nagana is discussed.

Trials of olfactory attractants to enhance trap catches of Glossina fuscipes fuscipes (Diptera: Glossinidae) in the Central African Republic.

Host odours increased the trap catches of Glossina fuscipes fuscipes in cattle breeding areas of the Central African Republic. The increase was significant with zebu urine (x 1.4) and the principal reptilian host, the monitor lizard (x 1.7). The greatest effect (x 4.2) was obtained for male G f fuscipes with zebu urine when the densities of flies were low (less than 5 males per trap per day). It seems that olfactory baits in urine could improve the control of G f fuscipes by trapping. Reptile odour contains attractants that should be identified.

[Control by traps of Glossina fuscipes fuscipes for protecting livestock in Central African Republic. IV. Entomological, parasitological and zootechnical impact]. / La lutte par piégeage contre Glossina fuscipes fuscipes pour la protection de l'élevage en République centrafricaine. IV. Impact entomologique, parasitologique et zootechnique.

Neighteen Mbororo zebu herds were monitored to evaluate the impact of a trapping campaign against Glossina fuscipes fuscipes restricted to the watering places. This programme is part of the strategy of an integrated campaign against bovine trypanosomoses in the Central African Republic. Trapping reduces the densities of G. f. fuscipes and causes trypanosome prevalence to fall. These effects are shown by the improved hematocrit values and the reduced number of trypanocidal treatments administered. The impact on productivity is more difficult to assess over a short period; nevertheless, it seems clear from the parameters calculated.

[Tsetse and livestock in Central African Republic: retreat of Glossina morsitans submorsitans (Diptera, Glossinidae)]. / Tsé-tsé et élevage en République Centrafricaine: le recul de Glossina morsitans submorsitans (Diptera, Glossinidae).

In the early 1960s, the most part of the Central African Republic was located inside the distribution area of Glossina morsitans submorsitans Newst. Since the last distribution studies of this tsetse in CAR (in 1963), the number of cattle has increased from 400,000 to approximately two millions, mainly of the Mbororo zebu breed. To set up the present distribution of G. m. submorsitans a study by trapping has been carried out in 27 livestock areas, regarding about 1,200 pastoralists' settlements. Furthermore, north-south transects have been made using both trapping (with bipyramidal traps laid every 2,000 meters) and net-catching (within a car driving slowly). The result have shown the disappearance of G. m. submorsitans from the main livestock areas in the west (Bouar, Bozoum, Bocaranga, Batangafo, Bossangoa, Paoua), centre (Bossembélé, Bouca, Dékoa) and east (Bambari, Grimari, Ippy). The southern border of the distribution area has been moved upwards to the north by up to 400 kilometres in the west of the country. This important retreat is discussed. It leads to a huge increase in the pastures available for the Mbororo pastoralists. Their access is still restricted in the centre-north and in the east by wild game reserves where G. m. submorsitans remains.

[The hosts of Glossina fuscipes fuscipes Newstead, 1910 (Diptera: Glossinidae) in 2 animal breeding zones in the Central African Republic]. / Les hôtes de Glossina fuscipes fuscipes Newstead, 1910 (Diptera: Glossinidae) dans 2 zones d'élevage de la République centrafricaine.

From 1987 to 1993, a survey on the feeding behavior of Glossina fuscipes fuscipes was conducted in the Central African Republic. A total of 556 blood-meal samples was analyzed by ELISA. According to the results, the number of blood meals from cattle was rather low (12% on average). During the rainy season, this number increased significantly and varied according to the sampling area. Along the riverine forests, this amounted to 5%, while blood meals from wild ruminants amounted to 87%. In the neighborhood of watering-places, the number of cattle blood meals reached 9-22%. Reptiles were found to be important hosts (17-35%). In all cases, man presented a non-negligible host (4-14%), similar to suidae (2-19%). The authors discuss the relevance of these results to risk of trypanosomes transmission.

Man-fly contact in the Gambian trypanosomiasis focus of Nola-Bilolo (Central African Republic).

A study using bipyramid tetse fly traps in the Nola-Bilolo sleeping sickness focus (Central African Republic) reveals ecological and behavioural differences between two vectors, Glossina palpalis palpalis and G. fuscipes fuscipes. The latter species inhabits mainly open water sites and surrounding forest, whereas G. p. palpalis occurs mainly in coffe plantations near villages. Consequently, the man-fly contact differs considerably according to the species. The intensity of trypanosomiasis transmission, estimated by the probable distribution of cases, showed significant positive correlation with the density of the flies. Analysis of the fly blood meals in two villages show that, unlike G. g. palpalis, G. f. fuscipes feeds on men more than on pigs. Trypanosoma vivax infection was observed only in G. fuscipes fuscipes. The differences in occupation of the environment between the two vectors must be taken in account in trapping programmes which may modify this distribution.

[An epidemiological survey to discover the probable places of infection with sleeping sickness in the Central African Republic]. / Enquête épidémiologique pour la recherche des lieux de contamination probables dans un foyer centrafricain de maladie du sommeil.

The sleeping sickness focus at Nola-Bilolo in the forest region was the subject of a survey of 142 patients in order to discover their probable places of infection. Seventy of these patients had been detected by active case-finding during a survey conducted in January-February 1991; the other 12 had been discovered by passive case-finding during 1990. The sample of actively detected patients is significantly younger (average 22 +/- 3 years) than the sample detected passively (30 +/- 3 years). This survey highlights the geographical heterogeneity of transmission in the focus. The places where infection is highest are M'Poyo, which with 18% of cases forms the epicentre of the focus and the main reservoir of infecting flies, followed by Bilolo (9%), Mékara (8%), Modigui-Kouna, Ziendi and Domissili (7%). The patients, 95% of whom belong to the M'Bimou tribe, all engage in agriculture; the main activities are coffee-growing for men and the steeping of casava in the river for women. These two activities determine the main places and times of human-tsetse contact. The information obtained by listing the cases detected passively from medical registers is inadequate for locating the places of transmission. The information gathered during active case-finding campaigns by the mobile teams offers a better approach. The method used here, which is recommended, consists of matching the replies given by the patients about their points of contact with the tsetse flies, not just with their places of residence but also with their successive movements and their places of work. The rigour demanded by this kind of approach can only be ensured by a specific survey. This survey also made it possible to demonstrate substantial movements from village to village, which appear to be characteristic of this focus. The results indicate that poorly targeted vector control is likely to be doomed to failure. The study makes it possible to specify priority areas for vector control by trapping and to avoid a "blind" control strategy whereby the whole focus is swamped with traps. The latter strategy is not financially feasible for many countries of tropical Africa.

PIP: In January-February 1991, an epidemiological survey took place in the Nola-Bilolo foyer (a forested zone), a focus of sleeping sickness in the Central African Republic, to learn where patients with sleeping sickness likely became infected. Active case finding during the survey detected 70 cases. Passive case finding in 1990 detected 72 cases. The principal activities which put cases at risk of acquiring African trypanosomiasis were those that took place almost always within a radius of 3 km, suggesting that a peridomestic tse-tse fly population was responsible. These activities included working in coffee plantations for men (31%) and the steeping of cassava in rivers for women (20%). Contact with the tse-tse fly within the village and in the forest contributed somewhat to acquisition of sleeping sickness (12%). Bathing in the rivers was the main activity placing children at risk. The epicenter of the foyer and the principal reservoir of the infective flies was M'poyo which had 18% of sleeping sickness cases. Other significant foci were Bilolo (9%), Mekara (8%), and Modigui-Kouna, Bindjo-Jeolo, Ziendi, and Domissili, all of which had 7% of cases. This survey revealed considerable movements from village to village, which seems to be characteristic of this foyer. It also showed that passive case finding from medical registers is adequate in obtaining information about the location of places of transmission and that mobile health teams are a better approach. The case finding approach allowed health workers to identify people's contact points with the tse-tse fly as well as their place of residence, which better depicts reality. It also allows health planners to identify specific priority areas for vector control by trapping which is financially more feasible for West African countries than is trapping the entire foyer. In conclusion, the case finding survey showed geographic heterogeneity of transmission of typanosomes.