- Email: [email protected]
The epidemiology of human schistosomiasis in the Senegal river basin
TRANSACTIONS OF THE ROYAL SOCIETY OF TROPICAL MEDICINE AND HYGIENE (1996) 90,340-346
Royal Society of Tropical Medicine and Hygiene Meeting at Manson House, London, 18 May 1995 The epidemiology
of human schistosomiasis
in the Senegal river basin
M. Picquet1y3, J. C. E rnould2, J. Vercruysse3*, V. R. Southgate4, A. Mbayel, B. Samboul, M. Niangl and D. Rollinson IProgramme Espoir, Rigion Mt%cale de St Louis, BP. 394, St Louis, S&gal; 2G.P. ‘Eau et Sante’: ORSTOM, B.P. 1386, Dakar, SJnbgal; 3Department of Parasitology, Faculty of Veterinary Medicine, Salisbuylaan 133, B-9820 Merelbeke, Belgium; 4The Natural History Museum, CromwellRoad, London, SW7 5BD, UK Abstract
Extensive water development has taken place in the north of Senegal over the last decade,resulting in a large increase in the amount of fresh water for irrigation. The objectives of the present study were to determine the prevalence and intensity of Schistosoma mansoni and S. haematobium in the Senegal river basin (SRB), and to ascertain the distribution of the snail speciesacting as intermediate hosts for both speciesof schistosomes.The schistosomiasis survey started in January 1994and was completed in March 1995.Compared to studies before the construction of the Diama dam, there was a significant increase in both the prevalence and intensity of urinary and intestinal schistosomiasis in the human population in parts of the SRB. From the 9014 people who were registered from 180 villages and 4 towns (10 districts), 7750 were examined. S. mansoni was found in the lower valley (lower delta-Senegal river, lower delta-Lampsar river, upper delta, and dieWe’)but not in the middle valley. The mean prevalence ranged from 4.4% in the lower delta-Senegal River to 71.8%in the zone of Lac de Guiers, where prevalence and intensity of infection were higher on the eastern side of the lake (81.3% with a mean number of 2088 eggs/gof faeces)compared with the western side (50.3% with a mean 1111 eggs/g).S. haematobium was recorded throughout the area of study, ranging from a mean prevalence of 0.37%in d&e (lower valley) to 415% in the lower valley (Lampsar river), where the mean egg count was 313/10mL of urine. Physical and chemical changesto the environment have favoured the spreadand increasein the populations of freshwater snails. The only snail involved in the transmission of S. mansoni was Biomphalaria pfeifiri. Five speciesof bulinid snails were present-Bulinus globosus, Bu. umbilicatu!, Bu. senegalensis,Bu. forskalii and Bu. truncatus-but only the first 3 specieswere involved in the transmission ofS. haematobium in the lower and middle valleys. Keywords:
Schistosoma
haematobium, S. mansoni,Senegal,epidemiology
Introduction
Water development projects in Africa are essential but all too often are associated with an increase in waterborne diseasessuch as schistosomiasis. New freshwater habitats for snail intermediate hosts of schistosomesmay be created by the construction of dams and irrigation projects. Greater opportunities for water contact may lead to increases of both urinary and intestinal schistosomiasis, as has been observed in villages along the shores of the Volta (Ghana), Kariba (Zimbabwe), Kainji (Nigeria) and Nasser (Egypt and Sudan) lakes (HIRA, 1969; PAPERNA, 1969; HUNTER et al., 1993). A similar situation is emerging in the Senegal river basin (SRB), where 2 dams have been constructed; one at Diama in Senegal, 40 km from the mouth of the Senegalriver and another in Mali on the Bafing river at Manantali. The Diama dam becameoperational in 1986 and is primarily designed to prevent salt water intrusion from the sea, in order to provide fresh water for irrigation and a supply of water to Lac de Guiers for municipal use in Dakar. The Manantali dam, which becameoperational 2 years later, will supply hydroelectricity and influences the flow of water downstream (VERCRUYSSE et al., 1994). Before the construction of the dams in the SRB, the prevalence of schistosomiasis and certain aspects of transmission were assessed.CHAINE & MALEK (1983) and VERCRUYSSE et al. (1985) showed that villages in the lower and middle vallevs of the SRB (Figure) were free of intestinal schistosomiasis.Biomphal&&pfe$,feri, an intermediate host of Schistosoma mansoni, was found in very small numbers at 2 isolated sites in the delta, and it was considered that this was the northern limit of its distribution in West Africa (CHAINE & MALEK, 1983). Generally, the prevalence of haematuria and/or urinary schistosomiasis in the SRB delta, in wale (villages close to the river) and around Lac de Guiers, was very low *Author for correspondence.
(CHAINE & MALEK 1983; CISSE et al., 1983; VERCRUYSSE et al., 1985; MALEK & CHAINE, 1989). However, the prevalence of S. haematobium and/or urinary schistosomiasis in 2 walo villages (Lampsar and Guede Chantier) and die% (villages situated further away from the river) was much higher than in the other wale villages examined (CHAINE & MALEK, 1983; VERCRUYSSE et aZ.,
1985). Extensive rice fields in Guide Chantier and rainfed iaterite pools are known to support populations of Bulinus seneealensis. This snail is an efficient intermediate host for “s. haematobium in parts of West Africa and is able to withstand long periods of desiccation and to tolerate high water temperatures (SOUTHGATE et al., 1985; DIAW & VASSILIADES,
1987; BROWN, 1994). Bu. globosus
is an important host of S. haematobium in parts of the SRB. Examination of 352 subjects from Mbodiene in July 1992 showed that 87% were infected with S. haematobium and, of the 366 Bu. globosus collected, 106were infected with furcocercous cercariae(VERL~ et aZ., 1994). Before the construction of the dams, funding agencies and their experts assessedthe situation but the risk of increased levels of schistosomiasis was not fully appreciated. A feasibilitv studv carried out bv USAID and OMVS (OrganisaAon pour la mise en naieur du Fleuve S&&al) reported that ‘... one cannot state with assurance howthe planned expansion of irrigated perimeters in the SRB will affect the orevalence of human schistosomiasis but the possibility &existsthere will be little or no increase.This bodes well for the proposed further development of rice crop on completion of the Diama barrage’ (MILLER, 1981). On the contrarv, VERCRUYSSE et al. (1985) g&e warning of the likely increase of urinary schistosomiasis and recommended regular surveillance for new transmission foci. An outbreak of intestinal schistosomiasis reported in 1988 at Richard-Toll. 130 km from the dam at Diama. was the first indicatioh that ecological changescausedb; the dam were having an impact on the prevalence of schistosomiasis in the region (TALLA et aZ., 1990). In a
SCHISTOSOMIASIS
IN THE SENEGAL
RIVER BASIN
341
LOWER VALLEY
i
MIDDLE
I
VALLEY
>
MAURITANIA POLIOR
LAC DE GUIERS TO MATAM 150 km
ST LOUIS 0
OliRi 50
e
100 km
NORTH
SENEGAL Figure. Sketchmapof the lower and middle valleysof the Senegalriver basin(R-Toll = Richard-Toll). Table 1. Description
of ecological zones in the Senegal river basin
Zone Lower valley Lower delta-Senegal river Lower delta-Lampsar river Upper delta-Senegal river Lac de Guiers D&k
Physical environment
Water supply
Population density
Ethnic group
Salt soils Irrigated land and river Irrigated land and river Irrigated land and lake
Senegalriver Lampsar river Senegalriver Lac de Guiers
Low High High High
Maure/Peul Wolof Wolof Wolof
Sahelian Savannah
Temporary pools
Low
Animal husbandry Rice culture Rice/sugar cane culture Rice/sugar cane culture/fishing Maure/Peul Animal husbandry
Alluvial plain Senegalriver Sahelian tree Savannah Temporary pools
High Low
Toucouleur Animal husbandry Peul Animal husbandry
Activity
Middle valley Walo DiM
period of 3 years, more than 60% of the population of Richard-Toll became heavily infected with S. mansoni (see TALLA et al, 1990, 1992). The first signs of changing distribution of S. haematobium infection in the delta were noticed in 1989(VERLI? et al., 1994). The aims of the present study were to determine the prevalence and intensity of schistosomiasis caused by S. mansoni and S. haematobium in the SRB and to relate these observations to the distribution of the snail species acting as intermediate hosts for both species of schistosomes. Materials and methods Study site
The Senegal river is the second largest river in West Africa; it forms the border between Senegal and Mauritania. Geoerauhicallv the SRB may be divided into 3 regions: the-lower vafiey, middle valley and upper valley (Figure). Only the first 2 regions were included in the present study. Lower uallev. The main road from St Louis to Matam passesthrough this vast area, dividing it into 2 distinct ecological areas. (i)Walo, stretching between the main road and the Senegalriver, with a dense sedentary population, mainly Wolof, who traditionally practise floodwater recession agriculture. In this region the area of land under irrigation, mainly for rice culture, increased considerably. In 1984 there were 21 769 ha under irrigation; this increased in 1994 to 67 788 ha and the intention is a yearly increase of 3000 ha (Societt d’Amenagement et d’Exploitation des Terres du Delta [SAED], St Louis, personal communication). (ii) Dikr& stretching
on both sides of the Lac de Guiers and limited at the north by the main road and at the south by the boundary of the Louga region. The population is mainly nomadic, essentially of Peul and Maure origin, and depends on rain-fed agriculture. In the present study, 7 zones were considered according to geographical localization, type and density of the population, and agricultural activities (irrigation associated with rice and sugar cane culture or animal husbandry) (Table 1, Figure). Five zones were included in the lower valley, which extends from St Louis to Dagana. (i} Lower delta-Senegal river (wale). Before the construction of the dam there was little agriculture in this region becauseof salt water intrusion, but this changed after the dam was built and new irrigated areaswere developed. This region is situated between St Louis and Debi (west and east) and between the Senegal and Djeuss rivers (north and south). (ii) Lower delta-Lampsar river (waZo). This region has been protected for some time against salt-water intrusion by small dams which provide fresh water for the population of St Louis. (iii) Upper delta-Senegal river (wale). This region was influenced by salt water intrusion before the construction of the Diama dam but has become an important area for the cultivation of rice and sugar cane. The area, which includes the towns of Ross Bethio and RichardToll, lies between Debi and Dagana and between the Senegalriver and the main road. (iv) Lac de Guiers. An intake canal at Richard-Toll is linked to the Lac de Guiers, which is the main water
M. PICQUET ETAL.
342
supply for Dakar. Agricultural activities include rice culture and sugar cane (irrigated sugar plantations extend to 8000 ha) (Compagnie Sucriere SCnCgalaise[CSS], Richard Toll, personal communication); fishing is also important. (v) D&X?,situated to the south of the main road. Middle valley. This stretches east, then south-east, for about 450 km from Dagana to Matam near the Mali border. It lies in a shallow alluvial plain 10 to 20 km wide. The present survey extended to the village of Salde (200 km west of Matam). The middle valley can be divided into 2 zones: the ZuaEo, where irrigated rice culture is developing, and the drier d&6 to the south of the main road. Upstream from Matam, the river basin extends southeast to south in a narrow deeperbasin, termed the upper valley. The region is a typical sahelian area with less than 300 mm of rainfall per year. Population survey
The study commenced in January 1994 in the lower delta, and was completed in March 1995.Before the survey the objectives were discussed with the population. The sampling method was based on a 2-degree cluster sample survey (villages and households) with random drawing proportional to the population and stratification for the various geographical zones (BENNETT et al., 1991).From the estimated750 000 inhabitants of the SRB (Recensement General de la Population et de 1’Habitat de 1988, Rapport regional, St Louis), approximately 238 000 persons were living in the area covered by the present survey. A total of 9014persons(3.4%)from 180villagesand 4 towns (10 districts) were registered(Table 2). All occupants of a household over 4 years of age, who had lived permanently for more than 6 months in the area, were invited to participate in the study. Each subject was asked to provide one urine and one stool sample. Ten mL of the urine were filtered with NytrelB tilters (WHO, 1983). The filters were stained with iodine. Samples were examined the following day; 10% of the slides, chosen at random, were re-counted within 24 h. Stool specimens were processedby a double Kato tech-
nique (POLDERMANet al, 1985). Sampleswere examined within 24 h and, again, 10% of the samples were recounted (KATZ et al.. 1972). No svstematic stool examination was carried out on ieople from the Podor area as preliminary results (not published) showed the complete absenceof S. mansoni infections and Bi. pfez@ri. All subjects infected were offered treatment with 40 mg/kg praziquantel. Mean prevalences for infected villages and mean egg counts for infected people were calculated. Arithmetic means (with standard deviations) were used to assessegg counts (FULFORD, 1994). Results are presented for the total population and, separately, for children aged 7-14 years. 9
Snail survey
During 1983 and 1993, snail surveys were carried out in parts of the river, temporary rain-fed pools, irrigation canals and drains, rice fields, and Lac de Guiers, to determine the geographical distribution within the SRB of the potential intermediate hosts of schistosomes. Morphological examination and enzyme and deoxyribonucleic acid analysesof the snails were carried out to determine their specific identity (ROLLINSON& SOUTHGATE, 1979; GOLL, 1981; MIMPFOUNDI & GREER,1989). Water quality
A Hanna water tester was used to measurepH, oxidation-reduction potential, dissolved solids content, and temperature of water in natural habitats. Results Human schistosomiasis
Of the 9014 people registered, 7750 were examined, an overall participation rate of 86%. Fewer stool samples than urine samples were collected (Table 2). The prevalence and intensity of S. haematobium and S. mansoniinfections are summarized in Table 2. Lower valley. {i) Lower delta-Senegal river. Twenty clusters of 3 households distributed over 17 villages constituted a sample of 390 persons. The prevalence of urinary
Table 2. Details of the schistosomiasis survey and prevalence and intensity infections in the lower and middle valleys of the Senegal river basin
Zone Survey Population (no.) Sample (no.) No. of village? Pa;rt;$eation (%) Stool
Lower Valley Lower Lower Upper delta-Senegal delta-Lamnsar delta-Senegal river river river -
of S. haematobium and S. mansoni
Middle valley Lac de Guiers
D&b
Walo
D&6
4772 390 17
10207 849 21
50396 1681 23
10887 1139 18
19615 1359 24
110865 2585 59
31696 1011 28
851 74.6
85.4 77.0
89.9 73.1
85.3 80.5
805 64.7
85.5 NDb
87.8 ND
S. haematobium
Infected villages Prevalence (%)’ Load (eggs/l0 mL)C Children Prevalence (%)’ Load (eggs/l0 mL)C
4 16.4~11.0 82.1213.6
4&14.2 312.8&169
4 1.9k2.1 6.026
27.:12.0 7.557
I.:_tO.5 1.4&l
&5.2 31.9228
23.8rt27.9 93.9?188
66.1k14.2 452.22260
2.6? 3.9 2.3-t-l
26.3kl9.8
24k2.7 1.41-4
18.9k7.6 69.6k33.9 39.1-t-47 22.3-113
4& 1.0 221.92299
Zzt:? 16.5 43?* 13.0 7ll.88kl1.3 2Olrt5.5 720.62390 1219.7-c308 1793.32848 384.42 162
1.0-t-o
S. mansoni
Infected villages Prevalence (%J’ Load (eggs/g)
Children Prevalence (%J g Load (eggs/ )
1.912.8 100~0-1-0
iIncluding 10districts of 4 towns. Not done. ‘MeanAstandarddeviation.
41.5k23.4
56.5k15.3
79.4? 129 24.1-c10.3 244321372 570.1?343
1 -
51&26*7 24.1+-5
-
-
343
SCHISTOSOMIASISIN THE SENEGALRIVER BASIN schistosomiasis was low, only 4 villages being infected (235%). S. mansoni infections were present in only 3 villages (17.6%). (ii) Lower delta-Lampsar river. Thirty clusters of 3 households, distributed over 21 villages, including 849 persons, were surveyed. S. haematobium infections were present in 18 villages (857%). Intestinal schistosomiasis was detected in 13 villages (61.9%); the highest mean egg count was observed in Diagambal(1292 eggs/g). (iii) Upper delta. Thirty clusters of 6 households, distributed among 16 villages and 7 districts of 3 towns, constituted a sample of 1681 persons. Only a few S. haematobium infections were detected in 4 villages (17.4%). The prevalence of intestinal schistosomiasis was high, infections occurring in all towns and villages apart from one. (iv) Lac de Guiers. Around the lake, 19 clusters of 6 households distributed among 18 villages were surveyed, 1139 persons were registered. Only one village (Nder), at the western side of the lake, was infected with S. haematobium but all villages were infected with S. mansoni. Prevalences and intensities of infections were higher on the eastern side of the lake (81.3% with a mean egg count of 2088 eggs/g) than on the western side (50.3%, with 111 eggsk). (v) Diere. The 25 clusters of 6 households distributed over 24 villages included 1359 people, of whom 1094 were examined. The low average participation rate (80%) was probably related to the nomadic way of life led by these people. Urinary schistosomiasis was rare, only 4 persons among the 1094 examined being infected, with low egg counts. Nineteen villages (79.2%) were infected with intestinal schistosomiasis, but the overall prevalence was low. Large variations (3.3-57.6%) occurred between the villages. As in the region around the lake, there was a difference in prevalence and intensity of infection between the east and west d&e. The western part, which had the lower population, was less affected (prevalence of 16% compared with 21.3% for the eastern part, and mean egg counts of 350 and 394 eggs/g, respectively). Table 3. The localities Senegal river basina
Snail species Biomphalaria B. pfezj@ri
Bulinus B. senegalensis
of intermediate
Lower delta-Senegal river ND
-
B. globosus
-
B. umbilicatus
-
Middle valley. (i) Wale. A total of 61 clusters of 6 households distributed among 56 villages and 3 districts of Podor constituted a samnle of 2585 persons. Infections with S. haematobium were observed in 20 villages (33.9%). (ii) Diere. In this area, 28 clusters of 6 households, distributed among 28 villages, constituted a sample of 1011 persons. Only 2 villages were infected, with S. haematobium. Snail surveys Bi. pfeifferi, Bu. senegalensis, Bu. globosus and Bu. umbilicatus were collected from the SRB during 3 field missions conducted in June 1983, February 1993 (dry season) and October 1993 (wet season) (Table 3). The collections made in 1983 were made before establishment of the dams at Diama and Manantali, whereas those made in 1993 were 7 years after the establishment of the former and 5 years after the latter. The data on the distribution of Bu. trucatus and Bu. forskalii are not included because neither species is involved in the transmission of human schistosomiasis. Although the collections made were not comprehensive, some important points are worthy of comment. Bi. pfetfj’eri and Bu. globosus had significantly extended their range by colonizing new habitats, and increased in population size. For example, Bi. pfezsri was not recorded from Richard-Toll in 1983, nor was Bu. globosus recorded from the artificial irrigation canals near Lac de Guiers or Lac de Guiers itself. There are both experimental and field data (SOUTHGATE et al., 1985; VERLE et al., 1994; unpublished observations) to indicate that Bi. pfeifferr and Bu. globosus are intermediate hosts for S. mansoni and S. haematobium, respectively, in the SRB. The absence of Bu. globosus from the diem of the middle valley and the widespread occurrence of Bu. senegalensis clearly indicate the importance of the latter species in the epidemiology of urinary schistosomiasis in this part of the SRB. On the other hand, the apparent absence of Bu. senegalensis from the lower delta implies that Bu. globosus is the main intermediate host for S. haematobium in this zone. Furthermore, the
hosts of human schistosomiasis
in the lower and middle valleys of the
Lower valley Lower Upper delta-Senegal delta-Lampsar river Lac de Guiers river D&e’ Ross Bethio (1789); Canal principal, Mbodiene (1701) Richard-Toll ND (1713); Ndombo (1703); Bouton Batt (1707, 1786); Temeye Salaane (1711,1712) -
Mbeyssous (1755) -
Mbodiene (1715, 1705,1708,1784) -
Bouton Batt (1705, 1708,1785); Nder -
Middle valley
Walo -
D&e’ -
ND
-
-
-
Matam (1135,1136, 1137,1174); Diatar (1746) Nianga (1748, 1749); Podor (1038,1072) Diamar Diery (1745) -
-
Ouro-sogui (1172)
-
aNumbers in parentheses are the live collection reference numbers in the Experimental Taxonomy Division, The Natural History MUscum, London, SW7 5BD, UK; ND = not determined. Nos. 1038-1175refer to collections made in 1983,nos. 1699-1717to February 1993,and nos. 1782-1790to October 1993.
M. PICQUET ETAL.
344
Table 4. Physical and chemical quantities
Locality Type of habitat February 1993(drf season) Irrigation canal Mbodiene Irrigation canalC
Richard-Toll
Nianga, Podor Irrigation canaId October 1993(wegseason) Irrigation canal Mbodiene
in different transmission
Snail speciesa
Temperature”C
pH
Conductance (pS)
Bu. globosus Bi. pfe$f& Bu. pfeifferi Bu. truncatus Bu. truncatus
226
74
180
11.8
7.5
75
17.2
7.9
153
Bu. globosus Bi. pfei@i Bi. pfeifferi Bu. truncatus Bu. truncatus
31
6.4
210
28.4
6.9
117
30.5
7.0
70
32.7
7.2
70
Irrigation canalC
Richard-Toll
Irrigation canald
Nianga, Podor Dianga, near Podor Bu. senegalensis
Rice fieldse
sites/snail habitats in the Senegal river basin
:Bi. = Biomphalaria; Bu. = Bulinus.
16”13.77’N,16”14.698’W. ‘16”27.922’N 15”41.805’W. d16”35922’N) lS”41.805’W. e16”37.622’N:16”54.038’W.
fact that Bi. pfeifferi was not present in the middle valley was the factor preventing S. mansoni from becoming established. Somephysical and chemical measurementswere made in various habitats during the 2 missions in 1993 (Table 4). The main difference between the seasonsrelated to water temperature, but the range of temperatures observed would not have prevented transmission occurring; similarly, the pH and electrical conductance remained within the levels of tolerance for bulinid snails and for transmission to occur. Discussion
It is now clear that there has been a significant increase in both the prevalence and intensity of urinary and intestinal schistosomiasis in the human population in parts of the SRB, less than 10 years after the construction of the Diama dam. Before the construction of the dams at Manantali, Mali and Diama, a feasibility study concluded that there would be little or no increase in schistosomiasis once the dams becameoperational, for a variety of reasons(MILLER, 1981).At the time of the feasibilitv studv, onlv urinary schistosomiasis occurred in the Sl?B and it was suggestedthat the human population had little contact with snail-infected waters. The most abundant bulinid snail of the 5 speciespresent was Bu. truncatus, and this was considered an inefficient intermediate h&t for the local strain of S. haematobium (see SOUTHGATE et al.. 1985: VERCRUYSSE et al.. 19851.The river water was gcidic,’ with pH measurements’ commonly 6 or below (MILLER, 1981). It is known that low pH levels are unfavourable for the survival of miracidia and their penetration into the snail. It was also suggested that mean daily temperatures of 30°C or more were too high for the intermediate host of S. mansoni, Bi. pfeifferi, __ __ to become established. Finally, the intrusion of sea water from the Atlantic ocean into the river system UD to Richard-Toll. and bevond when the river levels wherelow, resulted ih river w’ater of too high salinity for freshwater snails to become established. COCELSet al. (1993) showed that the Diama dam causeda gradual softening of the Lac de Guiers water: water mineralization has decreased20% in the northern part and as much as nearly 50% in the southern region, indicating profound environmental change. The differences between areas and villages in prevalence and intensitv of S. haematobium infections reflect the variable levels-of transmission within different parts of the SRB. The highest number of infected villages. the highest prevalence-per infected village, and the Ihiihest average .egg load, were observed in the lower delta-Lampsar river area and, to a lesser extent, in the
waZo area of the middle valley. In the former area,87% of
the examined villages were infected and the mean prevalence was 41%; in the latter 36% of the villages were infected, with a mean prevalence of 11%. In the other areas,the number of infected villages was lower. Only 6 of the 52 d&i villages examined were infected despite similar living conditions! emphasizing the extremely focal nature of schistosomlasis transmission. In the 2 surveys conducted before the construction of the dam, much lower prevalences of urinary schistosomiasis were recorded (CHAINE & MALEK, 1983; VERCRUYSSE et al., 1985). High prevalences were then also seen in the waZo villages of Lampsar and GuCdCChantier (near Podor) and somed&e’ villages. ” S. mansonihad never been recorded in the delta before 1988,when S. mansonieggswere seen in the stool specimens of a patient from Richard-Toll (TALLA et al:, 1990). A year later the situation had changed dramatltally, when S. mansoni was found in stools of 1906 patients. A survey in Richard-Toll in 1990 demonstrated that 60% of the subjects examined were infected with S. mansoni. Consequently, in the short period of 3 years since it was first recorded, S. mansonihad become firmly established in the population of Richard-Toll, with a high prevalence of heavy infections (TALLA et al., 1992; STELMA et al., 1993). Interestingly,
the increase in preva-
lence and intensity. was clearly correlated with increases in the oouulations of Bi. dzifferi (see TALLA et al.. 19901. Our data demonstrated*“& i. mansoni had ‘spreah throughout much of the delta. Prevalenceswere still low in the lower delta-Senegal river, ranging between 3.6% and 6.6%, but somewhat higher in the lower delta-Lampsar river area, with a mean prevalence of 28.8% and mean egg counts of 1292 in one village. A high prevalence rate(95.6%) existed in the upper delta, and all the villages in Lac de Guiers were infected with S. mansoni with”an overall prevalence of 71.8%. As yet, Bi. pfeifiri has not becomeestablished in the middle valley and, consequently, neither has S. mansoni. This situation should be monitored carefully over the ensuing years. What are the major changes that have taken place in recent years which have contributed to the increase in urinary schistosomiasis in the SRB? In the first place there has been a huge increase in rice farming, from 12 000 ha at the end of 1983to 67 788 ha in 1994.Associated with this has been the creation of additional irrigation channels and artificial temporary pools, providing additional snail habitats, mainly for Bu. senegalensis. Secondly, there have been several physical and chemical changes to the environment since the construction and operation of the dams at Diama and Manantali which fa-
SCHISTOSOMIASISIN THE SENEGALRIVER BASIN voured the spread and increase of the populations of bulinid snails. The dam at Manancali in the Bafing river, a tributary of the Senegal river, controls the flow of water and this results in greater stability of water level which is favourable for most species of freshwater snail, particularly Bu. globosus. The water levels showed annual fluctuations of 13 m before the dam was operational, and such instability is harmful to snail populations. The dam at Diama prevents the intrusion of sea water into the Senegal river, which previously occurred at times of low water levels up to and beyond Richard-Toll. Above salinities of one part per thousand there is a progressive elimination of gastropod species and a progressive reduction in snail and cercarial infectivity with increasing salinity. DONNELLY et al. (1983) demonstrated that the abilitv of eges to hatch and t-he fecunditv and survival of adult*Bu. a&anus, which is closely relited to Bu. globosus, was adversely affected by salinities as low as 1/103, wi;h significant ieductions bccurring between 35 and 4W103. DONNELLY et al. (1984) showed that survival of miracidia decreased progcessivily above 7/103. Furthermore, CHRISTENSEN et al. (1979) showed that the skin penetrative capacity of cercariae of S. mansoni was unaltered up to a salinity of 2.4/103, while at salinity levels above 2.4/103 the ability to penetrate decreased. Therefore, reduction in levels of salinity is not only beneficial to snails but also to schistosomes, in that the efficiency of transmission is increased. Concomitant with the reduction in salinity in the Senegal river, there appears to have been a change in the pH of the water from an acidic environment, with a pH of about 6, to a more alkaline environment with pH 7-8: an alkaline environment is more favourable to freshwater snails from the point of view of growth and fecundity, whereas an acid environment results in a low reproduction potential for snails (BROWN, 1994). The recent outbreak of S. haematobium infection in Mbodiene (lower delta, near the Lampsar river) (VERLB et al., 1994) and the new foci in the lower valley (present study) are probably related to changes in environmental conditions allowing Bu. globosus to become more widely established. Before the construction of dams at Diama and Manantali, much smaller dams were in use on the Lampsar river to prevent the intrusion of sea water and so providing drinking water for the town of St Louis. Bu. globosus, an intermediate host of S. haematobium, was able to survive in this particluar niche. Surprisingly, Bu. senegulensis was not reported to occur in the rice fields around the Lampsar river (BELOT et al., 1993), and it was not observed in our survevs. It is apparent that neither &. truncatus nor Bu. forskaiii are involved in the euidemioloav of human schistosomiasis in the middle and lower valleys, but, interestingly, both species of snail transmit S. bovis. B. truncatus is-an imporiant intermediate host for S. haematobium elsewhere in Africa (BROWN. 1994). Hence. onlv 3 of the 5 species of Bulinus ‘are respbnsibie for thd tra&mission of S. haematobium: that is, Bu. globosus, Bu. senegalensis and Bu. umbilicatus. Both field and laboratory data indicate that Bu. globosus is highly compatible with S. haematobium, as is Bu. umbilicatus. However, Bu. umbilicatus has a somewhat restricted distribution in the SRB and therefore is less important than Bu. globosus in the epidemiology of S. haematobium infection in that particuar region. Bu. senegulensis is commonly found in some of the rice fields and temporary laterite pools and is able to aestivate for prolonged periods diring the dry season. Bu. senepaalensisis comuatible with S. haematobium. but is invol;ed in transmi$sion only during and immddiately after the wet season, when the snails are active. It is highlv probable that Bu.~senegalensis is the intermediate host df S. haematobium in the Podor region. To the east of Podor, in the department of Mat&, both Bu. senegalensis and Bu. umbilicatus occur in temporary water bodies. Bi. Pfezjj&ri is solely responsible for the transmission of S. mansoni.
345
Thus it is concluded that the recent increases in prevalence and intensity of urinary schistosomiasis in the SRB may be correlated with ecological changes which have provided habitats facilitating the spread and increase of Bu. senegalensis (middle valley) and Bu. globosus (lower valley). The outbreak of intestinal schistosomiasis at Richard-Toll, first reported by TALLA et al. (1990), is undoubtedly linked with the establishment of Bi. pfeifiri in the main canal in the environs of RichardToll. P. H. Go11 (personal communication) failed to observe any specimens of Bi. pfeifferi in surveys conducted in 1978. although he noted that some snecimens had been fo;nd at an earlier date. It seems probable that small, unstable populations of Bi. pfeifiri occurred in some localities in the environs of Richard-Toll before the construction of the dam at Diama, but the change in ecological conditions, particularly the reduction in levels of salinity, has allowed this species to flourish, increasing substantially in population size and extending its distribution. The establishment of Bi. pfe#iri is the key to the creation of new foci of S. mansoni. As there is movement of people along the SRB, the parasite can be easily introduced and become established. The same pattern is repeating itself to varying degrees along the SRB with both S. mansoni and S. haematobium. For example, the recent outbreak of S. haematobium in Nder, on the western side of Lac de Guiers, is probably related to changes in environmental conditions allowing Bu. globosus, an intermediate host, to become established. The present study showed that intestinal schistosomiasis is now highly endemic in the lower valley and it has become an &p&ant health problem since-it was first noted in 1988 (TALLA et al.. 1990). Amone the 96 villages examined in the lowe; valley, 75 (7&) were infected with S. mansoni and in certain villages prevalences as high as 94% and egg counts of more than 4200/g were observed. The epicentres are Richard-Toll and Lac de Guiers, and it seems that the parasite is spreading both east and west. People living in the infected die% villages of the lower valley probably contracted the disease elsewhere, since intermediate hosts were not found in potential transmission sites near these villages. An interesting finding in the present study was the absence of the clear differences in age-related patterns of prevalence and intensity of infection seen in numerous similar studies throughout Africa. The prevalence and intensity of S. haematobium and S. mansoni infections were comparable in all age groups, suggesting that the whole population, which has been exposed to Schistosoma infect&s only in the last few years; lacks immunity to infection. This situation is likelv to channe with time. These data demonstrate that ;he previence and intensity of both urinary and intestinal schistosomiasis are increasing in parts of the SRB at an alarmingly rapid rate. It is of vital importance that this changing situation is carefully monitored over the forthcoming years,; such information will be of major importance in relation to the implementation of control programmes. Acknowledgements We gratefully acknowledge the help of Dr Cheikh Harm, Dr Bocar Daft, Malick Sy, Moussa Ba, Cheikh Thiam, Sidi Ndiongue, Khahlou Ba and the teams of Grandes Endbnies, Centre d’lnvestigations Biologiques and ORSTOM. Special thanks go to J. P. Piau, F. Rogerie, Dr J.-P. Her&, Dr J. F. Molez and Dr G. Hebrard for logistic support. This work was supported by ORSTOM (lower delta), the Commission of the European Communities Research Programme ‘Life Science and Technologies for Developing Countries’ (TS3-CT92-0117), the VLIR-Flemish Inter-University Council (Belgium), and is associated with the ESPOIR programme for research and control of schistosomiasis in the SenegalRiver Basin. References
Belot, J., Geerts, S. & Diouf, M. (1993). Observations on the population dynamics of snail hosts for schistosomes in the delta of the Senegal River Basin. Journal of Molluscan Studies,
346
59,7-13. Bennett, S.? Woods, T., Liyanage, W.M. & Smith, D.L. (1991). A simphtied general method for cluster-sample surveys of health in developing countries. World Health Statistics Quarterly, 44,98-106. Brown, D. S. (1994) Freshwater Snails ofAfrica and their Medical Importance, 2nd edition. London: Taylor & Francis. Chaine, J.P. & Malek, E.A. (1983). Urinary schistosomiasis in the sahelian region of the Senegal River Basin. Tropical and Geographical Medicine, 35,249-256. Christensen, N. @., Frandsen, F. & Nansen, P. (1979). The effects of some environmental conditions and final-host and parasite-related factors on the penetration ofSchistosoma mansoni cercariae into mice. Zeitschriti_ ”fur Parasitenkunde. , 59,I 267-275. Cisse, F., Diallo, S. & Dieny, M. (1983). Bilan actuel de la bilharziose urinaire chez les populations riveraines du lac de Guiers. Dakar Medical, 28, 343-350. Cogels, F. X., Thiam, A. & Gac, J. Y. (1993). Premiers effects des barrages du Fleuve Senegal sur le Lac de Guiers. Revue d’Hydrobiologie Tropicale, 26,105-l 17. Diaw, 0. T. & Vassiliades, G. (1987). Epidtmiologie des schistosomes du b&ail au Senegal. Revue de 1’Elevage et de Medecine V&&inaire desPays Tropicaux, 40,266-274. Donnelly, F. A., Appleton, C. G. & Schutte, C. H. J. (1983). The influence of salinity on certain aspects of the biology of Bulinus (l’hysopis) afiicanus. International Journal for Parasitology, 13.539-545. Donnelly, F. A., Appleton, C. G. & Schutte, C. H. J. (1983). The influence of salinity on the ova and miracidia of three species of Schistosoma. International Journal for Parasitology, 14, 113-120. Fulford, A. J. C. (1994). Dispersion and bias: can we trust geometric means? Parasitology Today, 10,446-448. Go11 P. H. (1981). Mixed populations of Bulinus senegalensis (Muller) and B. forskah [sic] (Ehrenberg) (Mollusca: Planorbidae) in The Gambia. Transactions of the Royal Society of Tropical Medicine and Hygiene, 75,576-578. Hira, P. R. (1969). Transmission of schistosomiasis in Lake Kariba, Zambia. Nature, 224,672-676. Hunter, J. M., Rey, L., Chu, K. Y., Adekolu-John, E.O. & Mott, K.E. (1993). Parasitic Diseases in Water Resources Development: the Need for Intersectional Negotiation. Geneva: World Health Organization. Katz, N, Chaves, A. & Pellegrino, J. (1972). A simple device for quantitative stool thick smear technique in schistosomiasis mansoni. Revista do Instituto de Medicina Tropical de Sao Paula, 14,397-400. Malek, E. A. & Chaine, J. P. (1989). Effects of the developments in the Senegal River Basin on the prevalence and spread of schistosomiasis. In: Demography and Vector Borne Diseases, Service, M. W. (editor). Boca Raton, Florida: CRC Press, pp. 181-192. Miller, M. (1981). Parasites of man and arthropod disease vectors in communities of a water development program on the Senegal River Basin. Report. Calgary, Alberta: University of Calgary. Mimpfoundi, R. & Greer, G. J. (1989). Allozyme comparisons among species of the Bulinus forskalii group (Gastropoda: Planorbidae) in Cameroon. rournal of Molluscan Studies, 55,
M. PICQUET
ETAL.
405410. Paperna, I. (1969). Study of an outbreak of schistosomiasis in the newly formed Volta Lake in Ghana. Zeitschr$&r Tropenmedizin und Parasitologic, 21,339-353. Polderman, A. M., Kayiteshonga Mpamilla, Manshande, J. P. and Bouwhuis-Hoogerwert, M. L. (1985). Methodology and interpretation of a parasitological surveillance of intestinal schistosomiasis in Maniema, Kivu province, Zaire. AnnaZes de la Societe Belge de Medecine Tropicale, 65,243-249. Rollinson, D. & Southgate, V. R. (1979). Enzyme analyses of Bulinus afiicanus group snails (Mollusca: Planorbidae) from Tanzania. Transactions of the Royal Society of Tropical Medicine and Hygiene, 73,667-672. Southgate, V. R., Rollinson, D., Ross, G. C., Knowles, R. J. & Vercruysse, J. (1985). On Schistosoma curassoni, S. haematobium and S. bovis from Senegal: development in Mesocricetus auratus, compatibility with species of Bulinus and their enzymes.Joumal of Natural Histoy, 19, 124991267. Stelma, F., Talla, I., Polman, K., Sturrock, R. F. & Gryseels, B. (1993). Epidemiology of Schistosoma mansoni infection in a recently exposed community in northern Senegal. American Journal of Tropical Medicine and Hygiene, 49,701-706. Talla, I, Kongs, A., Verle, P., Belot, J., Sarr, S. & Coll, A. M. (1990). Outbreak of intestinal schistosomiasis in the Senegal River Basin. Annales de la Societe’Belge de Medecine Tropicale, 70,173-180. Talla, I, Kongs, A. & Verle, P. (1992). Preliminary study of the prevalence of human schistosomiasis in Richard-Toll (the Senegal river basin). Transactions of the Royal Society of Tropical Medicine and Hygiene, 86, 132. Vercruysse, J., Southgate, V. R. & Rollinson, D. (1985). The epidemiology of human and animal schistosomiasis in the Senegal river basin. Acta Tropica, 42,249-259. Vercruysse, J., Southgate, V. R., Rollinson, D., De Clercq, D., Sacko, M., De Bont, J. & Mungomba, L.M. (1994). Studies on transmission and schistosome interactions‘in Senegal, Mali and Zambia. Tropical and Geographical Medicine, 46,220-226. Verb?, P., Stelma, F., Desreumaux, P., Dieng, A., Diaw, O., Kongs, A., Niang, M., Sow, S., Talla, I:, Sturrock, R. F., Gryseels, B. & Capron, A. (1994). Preliminary study of urinary schistosomiasis in a village in the delta of the Senegal river basin, Senegal. Transactio& of the Royal Society of Troiical Medicine and Hygiene, 88,401-405. WHO (1983). Urine filtration technique for Schistosoma haematobium. Geneva: World Health Organization, mimeographed document PDPi83.4. -Details of the prevalence and intensity of schistosomiasis in wale and d&e (lower and middle vallevs) and the environs of Lac de Guiers,‘and a list of villages in which schistosomiasis was not detected, can be obtained on request from Professor J. Vercruysse, Department of Parasitology, Faculty of Veterinary Medicine, Salisburylaan 133, B-9820 Merelbeke, Belgium. Received 3 January ceptedfor publication
1996; revised 14 February 14 February 1996
1996; ac-
Royal Society of Tropical Medicine and Hygiene Meeting at Manson House, London, 18 May 1995 The epidemiology
of human schistosomiasis
in the Senegal river basin
M. Picquet1y3, J. C. E rnould2, J. Vercruysse3*, V. R. Southgate4, A. Mbayel, B. Samboul, M. Niangl and D. Rollinson IProgramme Espoir, Rigion Mt%cale de St Louis, BP. 394, St Louis, S&gal; 2G.P. ‘Eau et Sante’: ORSTOM, B.P. 1386, Dakar, SJnbgal; 3Department of Parasitology, Faculty of Veterinary Medicine, Salisbuylaan 133, B-9820 Merelbeke, Belgium; 4The Natural History Museum, CromwellRoad, London, SW7 5BD, UK Abstract
Extensive water development has taken place in the north of Senegal over the last decade,resulting in a large increase in the amount of fresh water for irrigation. The objectives of the present study were to determine the prevalence and intensity of Schistosoma mansoni and S. haematobium in the Senegal river basin (SRB), and to ascertain the distribution of the snail speciesacting as intermediate hosts for both speciesof schistosomes.The schistosomiasis survey started in January 1994and was completed in March 1995.Compared to studies before the construction of the Diama dam, there was a significant increase in both the prevalence and intensity of urinary and intestinal schistosomiasis in the human population in parts of the SRB. From the 9014 people who were registered from 180 villages and 4 towns (10 districts), 7750 were examined. S. mansoni was found in the lower valley (lower delta-Senegal river, lower delta-Lampsar river, upper delta, and dieWe’)but not in the middle valley. The mean prevalence ranged from 4.4% in the lower delta-Senegal River to 71.8%in the zone of Lac de Guiers, where prevalence and intensity of infection were higher on the eastern side of the lake (81.3% with a mean number of 2088 eggs/gof faeces)compared with the western side (50.3% with a mean 1111 eggs/g).S. haematobium was recorded throughout the area of study, ranging from a mean prevalence of 0.37%in d&e (lower valley) to 415% in the lower valley (Lampsar river), where the mean egg count was 313/10mL of urine. Physical and chemical changesto the environment have favoured the spreadand increasein the populations of freshwater snails. The only snail involved in the transmission of S. mansoni was Biomphalaria pfeifiri. Five speciesof bulinid snails were present-Bulinus globosus, Bu. umbilicatu!, Bu. senegalensis,Bu. forskalii and Bu. truncatus-but only the first 3 specieswere involved in the transmission ofS. haematobium in the lower and middle valleys. Keywords:
Schistosoma
haematobium, S. mansoni,Senegal,epidemiology
Introduction
Water development projects in Africa are essential but all too often are associated with an increase in waterborne diseasessuch as schistosomiasis. New freshwater habitats for snail intermediate hosts of schistosomesmay be created by the construction of dams and irrigation projects. Greater opportunities for water contact may lead to increases of both urinary and intestinal schistosomiasis, as has been observed in villages along the shores of the Volta (Ghana), Kariba (Zimbabwe), Kainji (Nigeria) and Nasser (Egypt and Sudan) lakes (HIRA, 1969; PAPERNA, 1969; HUNTER et al., 1993). A similar situation is emerging in the Senegal river basin (SRB), where 2 dams have been constructed; one at Diama in Senegal, 40 km from the mouth of the Senegalriver and another in Mali on the Bafing river at Manantali. The Diama dam becameoperational in 1986 and is primarily designed to prevent salt water intrusion from the sea, in order to provide fresh water for irrigation and a supply of water to Lac de Guiers for municipal use in Dakar. The Manantali dam, which becameoperational 2 years later, will supply hydroelectricity and influences the flow of water downstream (VERCRUYSSE et al., 1994). Before the construction of the dams in the SRB, the prevalence of schistosomiasis and certain aspects of transmission were assessed.CHAINE & MALEK (1983) and VERCRUYSSE et al. (1985) showed that villages in the lower and middle vallevs of the SRB (Figure) were free of intestinal schistosomiasis.Biomphal&&pfe$,feri, an intermediate host of Schistosoma mansoni, was found in very small numbers at 2 isolated sites in the delta, and it was considered that this was the northern limit of its distribution in West Africa (CHAINE & MALEK, 1983). Generally, the prevalence of haematuria and/or urinary schistosomiasis in the SRB delta, in wale (villages close to the river) and around Lac de Guiers, was very low *Author for correspondence.
(CHAINE & MALEK 1983; CISSE et al., 1983; VERCRUYSSE et al., 1985; MALEK & CHAINE, 1989). However, the prevalence of S. haematobium and/or urinary schistosomiasis in 2 walo villages (Lampsar and Guede Chantier) and die% (villages situated further away from the river) was much higher than in the other wale villages examined (CHAINE & MALEK, 1983; VERCRUYSSE et aZ.,
1985). Extensive rice fields in Guide Chantier and rainfed iaterite pools are known to support populations of Bulinus seneealensis. This snail is an efficient intermediate host for “s. haematobium in parts of West Africa and is able to withstand long periods of desiccation and to tolerate high water temperatures (SOUTHGATE et al., 1985; DIAW & VASSILIADES,
1987; BROWN, 1994). Bu. globosus
is an important host of S. haematobium in parts of the SRB. Examination of 352 subjects from Mbodiene in July 1992 showed that 87% were infected with S. haematobium and, of the 366 Bu. globosus collected, 106were infected with furcocercous cercariae(VERL~ et aZ., 1994). Before the construction of the dams, funding agencies and their experts assessedthe situation but the risk of increased levels of schistosomiasis was not fully appreciated. A feasibilitv studv carried out bv USAID and OMVS (OrganisaAon pour la mise en naieur du Fleuve S&&al) reported that ‘... one cannot state with assurance howthe planned expansion of irrigated perimeters in the SRB will affect the orevalence of human schistosomiasis but the possibility &existsthere will be little or no increase.This bodes well for the proposed further development of rice crop on completion of the Diama barrage’ (MILLER, 1981). On the contrarv, VERCRUYSSE et al. (1985) g&e warning of the likely increase of urinary schistosomiasis and recommended regular surveillance for new transmission foci. An outbreak of intestinal schistosomiasis reported in 1988 at Richard-Toll. 130 km from the dam at Diama. was the first indicatioh that ecological changescausedb; the dam were having an impact on the prevalence of schistosomiasis in the region (TALLA et aZ., 1990). In a
SCHISTOSOMIASIS
IN THE SENEGAL
RIVER BASIN
341
LOWER VALLEY
i
MIDDLE
I
VALLEY
>
MAURITANIA POLIOR
LAC DE GUIERS TO MATAM 150 km
ST LOUIS 0
OliRi 50
e
100 km
NORTH
SENEGAL Figure. Sketchmapof the lower and middle valleysof the Senegalriver basin(R-Toll = Richard-Toll). Table 1. Description
of ecological zones in the Senegal river basin
Zone Lower valley Lower delta-Senegal river Lower delta-Lampsar river Upper delta-Senegal river Lac de Guiers D&k
Physical environment
Water supply
Population density
Ethnic group
Salt soils Irrigated land and river Irrigated land and river Irrigated land and lake
Senegalriver Lampsar river Senegalriver Lac de Guiers
Low High High High
Maure/Peul Wolof Wolof Wolof
Sahelian Savannah
Temporary pools
Low
Animal husbandry Rice culture Rice/sugar cane culture Rice/sugar cane culture/fishing Maure/Peul Animal husbandry
Alluvial plain Senegalriver Sahelian tree Savannah Temporary pools
High Low
Toucouleur Animal husbandry Peul Animal husbandry
Activity
Middle valley Walo DiM
period of 3 years, more than 60% of the population of Richard-Toll became heavily infected with S. mansoni (see TALLA et al, 1990, 1992). The first signs of changing distribution of S. haematobium infection in the delta were noticed in 1989(VERLI? et al., 1994). The aims of the present study were to determine the prevalence and intensity of schistosomiasis caused by S. mansoni and S. haematobium in the SRB and to relate these observations to the distribution of the snail species acting as intermediate hosts for both species of schistosomes. Materials and methods Study site
The Senegal river is the second largest river in West Africa; it forms the border between Senegal and Mauritania. Geoerauhicallv the SRB may be divided into 3 regions: the-lower vafiey, middle valley and upper valley (Figure). Only the first 2 regions were included in the present study. Lower uallev. The main road from St Louis to Matam passesthrough this vast area, dividing it into 2 distinct ecological areas. (i)Walo, stretching between the main road and the Senegalriver, with a dense sedentary population, mainly Wolof, who traditionally practise floodwater recession agriculture. In this region the area of land under irrigation, mainly for rice culture, increased considerably. In 1984 there were 21 769 ha under irrigation; this increased in 1994 to 67 788 ha and the intention is a yearly increase of 3000 ha (Societt d’Amenagement et d’Exploitation des Terres du Delta [SAED], St Louis, personal communication). (ii) Dikr& stretching
on both sides of the Lac de Guiers and limited at the north by the main road and at the south by the boundary of the Louga region. The population is mainly nomadic, essentially of Peul and Maure origin, and depends on rain-fed agriculture. In the present study, 7 zones were considered according to geographical localization, type and density of the population, and agricultural activities (irrigation associated with rice and sugar cane culture or animal husbandry) (Table 1, Figure). Five zones were included in the lower valley, which extends from St Louis to Dagana. (i} Lower delta-Senegal river (wale). Before the construction of the dam there was little agriculture in this region becauseof salt water intrusion, but this changed after the dam was built and new irrigated areaswere developed. This region is situated between St Louis and Debi (west and east) and between the Senegal and Djeuss rivers (north and south). (ii) Lower delta-Lampsar river (waZo). This region has been protected for some time against salt-water intrusion by small dams which provide fresh water for the population of St Louis. (iii) Upper delta-Senegal river (wale). This region was influenced by salt water intrusion before the construction of the Diama dam but has become an important area for the cultivation of rice and sugar cane. The area, which includes the towns of Ross Bethio and RichardToll, lies between Debi and Dagana and between the Senegalriver and the main road. (iv) Lac de Guiers. An intake canal at Richard-Toll is linked to the Lac de Guiers, which is the main water
M. PICQUET ETAL.
342
supply for Dakar. Agricultural activities include rice culture and sugar cane (irrigated sugar plantations extend to 8000 ha) (Compagnie Sucriere SCnCgalaise[CSS], Richard Toll, personal communication); fishing is also important. (v) D&X?,situated to the south of the main road. Middle valley. This stretches east, then south-east, for about 450 km from Dagana to Matam near the Mali border. It lies in a shallow alluvial plain 10 to 20 km wide. The present survey extended to the village of Salde (200 km west of Matam). The middle valley can be divided into 2 zones: the ZuaEo, where irrigated rice culture is developing, and the drier d&6 to the south of the main road. Upstream from Matam, the river basin extends southeast to south in a narrow deeperbasin, termed the upper valley. The region is a typical sahelian area with less than 300 mm of rainfall per year. Population survey
The study commenced in January 1994 in the lower delta, and was completed in March 1995.Before the survey the objectives were discussed with the population. The sampling method was based on a 2-degree cluster sample survey (villages and households) with random drawing proportional to the population and stratification for the various geographical zones (BENNETT et al., 1991).From the estimated750 000 inhabitants of the SRB (Recensement General de la Population et de 1’Habitat de 1988, Rapport regional, St Louis), approximately 238 000 persons were living in the area covered by the present survey. A total of 9014persons(3.4%)from 180villagesand 4 towns (10 districts) were registered(Table 2). All occupants of a household over 4 years of age, who had lived permanently for more than 6 months in the area, were invited to participate in the study. Each subject was asked to provide one urine and one stool sample. Ten mL of the urine were filtered with NytrelB tilters (WHO, 1983). The filters were stained with iodine. Samples were examined the following day; 10% of the slides, chosen at random, were re-counted within 24 h. Stool specimens were processedby a double Kato tech-
nique (POLDERMANet al, 1985). Sampleswere examined within 24 h and, again, 10% of the samples were recounted (KATZ et al.. 1972). No svstematic stool examination was carried out on ieople from the Podor area as preliminary results (not published) showed the complete absenceof S. mansoni infections and Bi. pfez@ri. All subjects infected were offered treatment with 40 mg/kg praziquantel. Mean prevalences for infected villages and mean egg counts for infected people were calculated. Arithmetic means (with standard deviations) were used to assessegg counts (FULFORD, 1994). Results are presented for the total population and, separately, for children aged 7-14 years. 9
Snail survey
During 1983 and 1993, snail surveys were carried out in parts of the river, temporary rain-fed pools, irrigation canals and drains, rice fields, and Lac de Guiers, to determine the geographical distribution within the SRB of the potential intermediate hosts of schistosomes. Morphological examination and enzyme and deoxyribonucleic acid analysesof the snails were carried out to determine their specific identity (ROLLINSON& SOUTHGATE, 1979; GOLL, 1981; MIMPFOUNDI & GREER,1989). Water quality
A Hanna water tester was used to measurepH, oxidation-reduction potential, dissolved solids content, and temperature of water in natural habitats. Results Human schistosomiasis
Of the 9014 people registered, 7750 were examined, an overall participation rate of 86%. Fewer stool samples than urine samples were collected (Table 2). The prevalence and intensity of S. haematobium and S. mansoniinfections are summarized in Table 2. Lower valley. {i) Lower delta-Senegal river. Twenty clusters of 3 households distributed over 17 villages constituted a sample of 390 persons. The prevalence of urinary
Table 2. Details of the schistosomiasis survey and prevalence and intensity infections in the lower and middle valleys of the Senegal river basin
Zone Survey Population (no.) Sample (no.) No. of village? Pa;rt;$eation (%) Stool
Lower Valley Lower Lower Upper delta-Senegal delta-Lamnsar delta-Senegal river river river -
of S. haematobium and S. mansoni
Middle valley Lac de Guiers
D&b
Walo
D&6
4772 390 17
10207 849 21
50396 1681 23
10887 1139 18
19615 1359 24
110865 2585 59
31696 1011 28
851 74.6
85.4 77.0
89.9 73.1
85.3 80.5
805 64.7
85.5 NDb
87.8 ND
S. haematobium
Infected villages Prevalence (%)’ Load (eggs/l0 mL)C Children Prevalence (%)’ Load (eggs/l0 mL)C
4 16.4~11.0 82.1213.6
4&14.2 312.8&169
4 1.9k2.1 6.026
27.:12.0 7.557
I.:_tO.5 1.4&l
&5.2 31.9228
23.8rt27.9 93.9?188
66.1k14.2 452.22260
2.6? 3.9 2.3-t-l
26.3kl9.8
24k2.7 1.41-4
18.9k7.6 69.6k33.9 39.1-t-47 22.3-113
4& 1.0 221.92299
Zzt:? 16.5 43?* 13.0 7ll.88kl1.3 2Olrt5.5 720.62390 1219.7-c308 1793.32848 384.42 162
1.0-t-o
S. mansoni
Infected villages Prevalence (%J’ Load (eggs/g)
Children Prevalence (%J g Load (eggs/ )
1.912.8 100~0-1-0
iIncluding 10districts of 4 towns. Not done. ‘MeanAstandarddeviation.
41.5k23.4
56.5k15.3
79.4? 129 24.1-c10.3 244321372 570.1?343
1 -
51&26*7 24.1+-5
-
-
343
SCHISTOSOMIASISIN THE SENEGALRIVER BASIN schistosomiasis was low, only 4 villages being infected (235%). S. mansoni infections were present in only 3 villages (17.6%). (ii) Lower delta-Lampsar river. Thirty clusters of 3 households, distributed over 21 villages, including 849 persons, were surveyed. S. haematobium infections were present in 18 villages (857%). Intestinal schistosomiasis was detected in 13 villages (61.9%); the highest mean egg count was observed in Diagambal(1292 eggs/g). (iii) Upper delta. Thirty clusters of 6 households, distributed among 16 villages and 7 districts of 3 towns, constituted a sample of 1681 persons. Only a few S. haematobium infections were detected in 4 villages (17.4%). The prevalence of intestinal schistosomiasis was high, infections occurring in all towns and villages apart from one. (iv) Lac de Guiers. Around the lake, 19 clusters of 6 households distributed among 18 villages were surveyed, 1139 persons were registered. Only one village (Nder), at the western side of the lake, was infected with S. haematobium but all villages were infected with S. mansoni. Prevalences and intensities of infections were higher on the eastern side of the lake (81.3% with a mean egg count of 2088 eggs/g) than on the western side (50.3%, with 111 eggsk). (v) Diere. The 25 clusters of 6 households distributed over 24 villages included 1359 people, of whom 1094 were examined. The low average participation rate (80%) was probably related to the nomadic way of life led by these people. Urinary schistosomiasis was rare, only 4 persons among the 1094 examined being infected, with low egg counts. Nineteen villages (79.2%) were infected with intestinal schistosomiasis, but the overall prevalence was low. Large variations (3.3-57.6%) occurred between the villages. As in the region around the lake, there was a difference in prevalence and intensity of infection between the east and west d&e. The western part, which had the lower population, was less affected (prevalence of 16% compared with 21.3% for the eastern part, and mean egg counts of 350 and 394 eggs/g, respectively). Table 3. The localities Senegal river basina
Snail species Biomphalaria B. pfezj@ri
Bulinus B. senegalensis
of intermediate
Lower delta-Senegal river ND
-
B. globosus
-
B. umbilicatus
-
Middle valley. (i) Wale. A total of 61 clusters of 6 households distributed among 56 villages and 3 districts of Podor constituted a samnle of 2585 persons. Infections with S. haematobium were observed in 20 villages (33.9%). (ii) Diere. In this area, 28 clusters of 6 households, distributed among 28 villages, constituted a sample of 1011 persons. Only 2 villages were infected, with S. haematobium. Snail surveys Bi. pfeifferi, Bu. senegalensis, Bu. globosus and Bu. umbilicatus were collected from the SRB during 3 field missions conducted in June 1983, February 1993 (dry season) and October 1993 (wet season) (Table 3). The collections made in 1983 were made before establishment of the dams at Diama and Manantali, whereas those made in 1993 were 7 years after the establishment of the former and 5 years after the latter. The data on the distribution of Bu. trucatus and Bu. forskalii are not included because neither species is involved in the transmission of human schistosomiasis. Although the collections made were not comprehensive, some important points are worthy of comment. Bi. pfetfj’eri and Bu. globosus had significantly extended their range by colonizing new habitats, and increased in population size. For example, Bi. pfezsri was not recorded from Richard-Toll in 1983, nor was Bu. globosus recorded from the artificial irrigation canals near Lac de Guiers or Lac de Guiers itself. There are both experimental and field data (SOUTHGATE et al., 1985; VERLE et al., 1994; unpublished observations) to indicate that Bi. pfeifferr and Bu. globosus are intermediate hosts for S. mansoni and S. haematobium, respectively, in the SRB. The absence of Bu. globosus from the diem of the middle valley and the widespread occurrence of Bu. senegalensis clearly indicate the importance of the latter species in the epidemiology of urinary schistosomiasis in this part of the SRB. On the other hand, the apparent absence of Bu. senegalensis from the lower delta implies that Bu. globosus is the main intermediate host for S. haematobium in this zone. Furthermore, the
hosts of human schistosomiasis
in the lower and middle valleys of the
Lower valley Lower Upper delta-Senegal delta-Lampsar river Lac de Guiers river D&e’ Ross Bethio (1789); Canal principal, Mbodiene (1701) Richard-Toll ND (1713); Ndombo (1703); Bouton Batt (1707, 1786); Temeye Salaane (1711,1712) -
Mbeyssous (1755) -
Mbodiene (1715, 1705,1708,1784) -
Bouton Batt (1705, 1708,1785); Nder -
Middle valley
Walo -
D&e’ -
ND
-
-
-
Matam (1135,1136, 1137,1174); Diatar (1746) Nianga (1748, 1749); Podor (1038,1072) Diamar Diery (1745) -
-
Ouro-sogui (1172)
-
aNumbers in parentheses are the live collection reference numbers in the Experimental Taxonomy Division, The Natural History MUscum, London, SW7 5BD, UK; ND = not determined. Nos. 1038-1175refer to collections made in 1983,nos. 1699-1717to February 1993,and nos. 1782-1790to October 1993.
M. PICQUET ETAL.
344
Table 4. Physical and chemical quantities
Locality Type of habitat February 1993(drf season) Irrigation canal Mbodiene Irrigation canalC
Richard-Toll
Nianga, Podor Irrigation canaId October 1993(wegseason) Irrigation canal Mbodiene
in different transmission
Snail speciesa
Temperature”C
pH
Conductance (pS)
Bu. globosus Bi. pfe$f& Bu. pfeifferi Bu. truncatus Bu. truncatus
226
74
180
11.8
7.5
75
17.2
7.9
153
Bu. globosus Bi. pfei@i Bi. pfeifferi Bu. truncatus Bu. truncatus
31
6.4
210
28.4
6.9
117
30.5
7.0
70
32.7
7.2
70
Irrigation canalC
Richard-Toll
Irrigation canald
Nianga, Podor Dianga, near Podor Bu. senegalensis
Rice fieldse
sites/snail habitats in the Senegal river basin
:Bi. = Biomphalaria; Bu. = Bulinus.
16”13.77’N,16”14.698’W. ‘16”27.922’N 15”41.805’W. d16”35922’N) lS”41.805’W. e16”37.622’N:16”54.038’W.
fact that Bi. pfeifferi was not present in the middle valley was the factor preventing S. mansoni from becoming established. Somephysical and chemical measurementswere made in various habitats during the 2 missions in 1993 (Table 4). The main difference between the seasonsrelated to water temperature, but the range of temperatures observed would not have prevented transmission occurring; similarly, the pH and electrical conductance remained within the levels of tolerance for bulinid snails and for transmission to occur. Discussion
It is now clear that there has been a significant increase in both the prevalence and intensity of urinary and intestinal schistosomiasis in the human population in parts of the SRB, less than 10 years after the construction of the Diama dam. Before the construction of the dams at Manantali, Mali and Diama, a feasibility study concluded that there would be little or no increase in schistosomiasis once the dams becameoperational, for a variety of reasons(MILLER, 1981).At the time of the feasibilitv studv, onlv urinary schistosomiasis occurred in the Sl?B and it was suggestedthat the human population had little contact with snail-infected waters. The most abundant bulinid snail of the 5 speciespresent was Bu. truncatus, and this was considered an inefficient intermediate h&t for the local strain of S. haematobium (see SOUTHGATE et al.. 1985: VERCRUYSSE et al.. 19851.The river water was gcidic,’ with pH measurements’ commonly 6 or below (MILLER, 1981). It is known that low pH levels are unfavourable for the survival of miracidia and their penetration into the snail. It was also suggested that mean daily temperatures of 30°C or more were too high for the intermediate host of S. mansoni, Bi. pfeifferi, __ __ to become established. Finally, the intrusion of sea water from the Atlantic ocean into the river system UD to Richard-Toll. and bevond when the river levels wherelow, resulted ih river w’ater of too high salinity for freshwater snails to become established. COCELSet al. (1993) showed that the Diama dam causeda gradual softening of the Lac de Guiers water: water mineralization has decreased20% in the northern part and as much as nearly 50% in the southern region, indicating profound environmental change. The differences between areas and villages in prevalence and intensitv of S. haematobium infections reflect the variable levels-of transmission within different parts of the SRB. The highest number of infected villages. the highest prevalence-per infected village, and the Ihiihest average .egg load, were observed in the lower delta-Lampsar river area and, to a lesser extent, in the
waZo area of the middle valley. In the former area,87% of
the examined villages were infected and the mean prevalence was 41%; in the latter 36% of the villages were infected, with a mean prevalence of 11%. In the other areas,the number of infected villages was lower. Only 6 of the 52 d&i villages examined were infected despite similar living conditions! emphasizing the extremely focal nature of schistosomlasis transmission. In the 2 surveys conducted before the construction of the dam, much lower prevalences of urinary schistosomiasis were recorded (CHAINE & MALEK, 1983; VERCRUYSSE et al., 1985). High prevalences were then also seen in the waZo villages of Lampsar and GuCdCChantier (near Podor) and somed&e’ villages. ” S. mansonihad never been recorded in the delta before 1988,when S. mansonieggswere seen in the stool specimens of a patient from Richard-Toll (TALLA et al:, 1990). A year later the situation had changed dramatltally, when S. mansoni was found in stools of 1906 patients. A survey in Richard-Toll in 1990 demonstrated that 60% of the subjects examined were infected with S. mansoni. Consequently, in the short period of 3 years since it was first recorded, S. mansonihad become firmly established in the population of Richard-Toll, with a high prevalence of heavy infections (TALLA et al., 1992; STELMA et al., 1993). Interestingly,
the increase in preva-
lence and intensity. was clearly correlated with increases in the oouulations of Bi. dzifferi (see TALLA et al.. 19901. Our data demonstrated*“& i. mansoni had ‘spreah throughout much of the delta. Prevalenceswere still low in the lower delta-Senegal river, ranging between 3.6% and 6.6%, but somewhat higher in the lower delta-Lampsar river area, with a mean prevalence of 28.8% and mean egg counts of 1292 in one village. A high prevalence rate(95.6%) existed in the upper delta, and all the villages in Lac de Guiers were infected with S. mansoni with”an overall prevalence of 71.8%. As yet, Bi. pfeifiri has not becomeestablished in the middle valley and, consequently, neither has S. mansoni. This situation should be monitored carefully over the ensuing years. What are the major changes that have taken place in recent years which have contributed to the increase in urinary schistosomiasis in the SRB? In the first place there has been a huge increase in rice farming, from 12 000 ha at the end of 1983to 67 788 ha in 1994.Associated with this has been the creation of additional irrigation channels and artificial temporary pools, providing additional snail habitats, mainly for Bu. senegalensis. Secondly, there have been several physical and chemical changes to the environment since the construction and operation of the dams at Diama and Manantali which fa-
SCHISTOSOMIASISIN THE SENEGALRIVER BASIN voured the spread and increase of the populations of bulinid snails. The dam at Manancali in the Bafing river, a tributary of the Senegal river, controls the flow of water and this results in greater stability of water level which is favourable for most species of freshwater snail, particularly Bu. globosus. The water levels showed annual fluctuations of 13 m before the dam was operational, and such instability is harmful to snail populations. The dam at Diama prevents the intrusion of sea water into the Senegal river, which previously occurred at times of low water levels up to and beyond Richard-Toll. Above salinities of one part per thousand there is a progressive elimination of gastropod species and a progressive reduction in snail and cercarial infectivity with increasing salinity. DONNELLY et al. (1983) demonstrated that the abilitv of eges to hatch and t-he fecunditv and survival of adult*Bu. a&anus, which is closely relited to Bu. globosus, was adversely affected by salinities as low as 1/103, wi;h significant ieductions bccurring between 35 and 4W103. DONNELLY et al. (1984) showed that survival of miracidia decreased progcessivily above 7/103. Furthermore, CHRISTENSEN et al. (1979) showed that the skin penetrative capacity of cercariae of S. mansoni was unaltered up to a salinity of 2.4/103, while at salinity levels above 2.4/103 the ability to penetrate decreased. Therefore, reduction in levels of salinity is not only beneficial to snails but also to schistosomes, in that the efficiency of transmission is increased. Concomitant with the reduction in salinity in the Senegal river, there appears to have been a change in the pH of the water from an acidic environment, with a pH of about 6, to a more alkaline environment with pH 7-8: an alkaline environment is more favourable to freshwater snails from the point of view of growth and fecundity, whereas an acid environment results in a low reproduction potential for snails (BROWN, 1994). The recent outbreak of S. haematobium infection in Mbodiene (lower delta, near the Lampsar river) (VERLB et al., 1994) and the new foci in the lower valley (present study) are probably related to changes in environmental conditions allowing Bu. globosus to become more widely established. Before the construction of dams at Diama and Manantali, much smaller dams were in use on the Lampsar river to prevent the intrusion of sea water and so providing drinking water for the town of St Louis. Bu. globosus, an intermediate host of S. haematobium, was able to survive in this particluar niche. Surprisingly, Bu. senegulensis was not reported to occur in the rice fields around the Lampsar river (BELOT et al., 1993), and it was not observed in our survevs. It is apparent that neither &. truncatus nor Bu. forskaiii are involved in the euidemioloav of human schistosomiasis in the middle and lower valleys, but, interestingly, both species of snail transmit S. bovis. B. truncatus is-an imporiant intermediate host for S. haematobium elsewhere in Africa (BROWN. 1994). Hence. onlv 3 of the 5 species of Bulinus ‘are respbnsibie for thd tra&mission of S. haematobium: that is, Bu. globosus, Bu. senegalensis and Bu. umbilicatus. Both field and laboratory data indicate that Bu. globosus is highly compatible with S. haematobium, as is Bu. umbilicatus. However, Bu. umbilicatus has a somewhat restricted distribution in the SRB and therefore is less important than Bu. globosus in the epidemiology of S. haematobium infection in that particuar region. Bu. senegulensis is commonly found in some of the rice fields and temporary laterite pools and is able to aestivate for prolonged periods diring the dry season. Bu. senepaalensisis comuatible with S. haematobium. but is invol;ed in transmi$sion only during and immddiately after the wet season, when the snails are active. It is highlv probable that Bu.~senegalensis is the intermediate host df S. haematobium in the Podor region. To the east of Podor, in the department of Mat&, both Bu. senegalensis and Bu. umbilicatus occur in temporary water bodies. Bi. Pfezjj&ri is solely responsible for the transmission of S. mansoni.
345
Thus it is concluded that the recent increases in prevalence and intensity of urinary schistosomiasis in the SRB may be correlated with ecological changes which have provided habitats facilitating the spread and increase of Bu. senegalensis (middle valley) and Bu. globosus (lower valley). The outbreak of intestinal schistosomiasis at Richard-Toll, first reported by TALLA et al. (1990), is undoubtedly linked with the establishment of Bi. pfeifiri in the main canal in the environs of RichardToll. P. H. Go11 (personal communication) failed to observe any specimens of Bi. pfeifferi in surveys conducted in 1978. although he noted that some snecimens had been fo;nd at an earlier date. It seems probable that small, unstable populations of Bi. pfeifiri occurred in some localities in the environs of Richard-Toll before the construction of the dam at Diama, but the change in ecological conditions, particularly the reduction in levels of salinity, has allowed this species to flourish, increasing substantially in population size and extending its distribution. The establishment of Bi. pfe#iri is the key to the creation of new foci of S. mansoni. As there is movement of people along the SRB, the parasite can be easily introduced and become established. The same pattern is repeating itself to varying degrees along the SRB with both S. mansoni and S. haematobium. For example, the recent outbreak of S. haematobium in Nder, on the western side of Lac de Guiers, is probably related to changes in environmental conditions allowing Bu. globosus, an intermediate host, to become established. The present study showed that intestinal schistosomiasis is now highly endemic in the lower valley and it has become an &p&ant health problem since-it was first noted in 1988 (TALLA et al.. 1990). Amone the 96 villages examined in the lowe; valley, 75 (7&) were infected with S. mansoni and in certain villages prevalences as high as 94% and egg counts of more than 4200/g were observed. The epicentres are Richard-Toll and Lac de Guiers, and it seems that the parasite is spreading both east and west. People living in the infected die% villages of the lower valley probably contracted the disease elsewhere, since intermediate hosts were not found in potential transmission sites near these villages. An interesting finding in the present study was the absence of the clear differences in age-related patterns of prevalence and intensity of infection seen in numerous similar studies throughout Africa. The prevalence and intensity of S. haematobium and S. mansoni infections were comparable in all age groups, suggesting that the whole population, which has been exposed to Schistosoma infect&s only in the last few years; lacks immunity to infection. This situation is likelv to channe with time. These data demonstrate that ;he previence and intensity of both urinary and intestinal schistosomiasis are increasing in parts of the SRB at an alarmingly rapid rate. It is of vital importance that this changing situation is carefully monitored over the forthcoming years,; such information will be of major importance in relation to the implementation of control programmes. Acknowledgements We gratefully acknowledge the help of Dr Cheikh Harm, Dr Bocar Daft, Malick Sy, Moussa Ba, Cheikh Thiam, Sidi Ndiongue, Khahlou Ba and the teams of Grandes Endbnies, Centre d’lnvestigations Biologiques and ORSTOM. Special thanks go to J. P. Piau, F. Rogerie, Dr J.-P. Her&, Dr J. F. Molez and Dr G. Hebrard for logistic support. This work was supported by ORSTOM (lower delta), the Commission of the European Communities Research Programme ‘Life Science and Technologies for Developing Countries’ (TS3-CT92-0117), the VLIR-Flemish Inter-University Council (Belgium), and is associated with the ESPOIR programme for research and control of schistosomiasis in the SenegalRiver Basin. References
Belot, J., Geerts, S. & Diouf, M. (1993). Observations on the population dynamics of snail hosts for schistosomes in the delta of the Senegal River Basin. Journal of Molluscan Studies,
346
59,7-13. Bennett, S.? Woods, T., Liyanage, W.M. & Smith, D.L. (1991). A simphtied general method for cluster-sample surveys of health in developing countries. World Health Statistics Quarterly, 44,98-106. Brown, D. S. (1994) Freshwater Snails ofAfrica and their Medical Importance, 2nd edition. London: Taylor & Francis. Chaine, J.P. & Malek, E.A. (1983). Urinary schistosomiasis in the sahelian region of the Senegal River Basin. Tropical and Geographical Medicine, 35,249-256. Christensen, N. @., Frandsen, F. & Nansen, P. (1979). The effects of some environmental conditions and final-host and parasite-related factors on the penetration ofSchistosoma mansoni cercariae into mice. Zeitschriti_ ”fur Parasitenkunde. , 59,I 267-275. Cisse, F., Diallo, S. & Dieny, M. (1983). Bilan actuel de la bilharziose urinaire chez les populations riveraines du lac de Guiers. Dakar Medical, 28, 343-350. Cogels, F. X., Thiam, A. & Gac, J. Y. (1993). Premiers effects des barrages du Fleuve Senegal sur le Lac de Guiers. Revue d’Hydrobiologie Tropicale, 26,105-l 17. Diaw, 0. T. & Vassiliades, G. (1987). Epidtmiologie des schistosomes du b&ail au Senegal. Revue de 1’Elevage et de Medecine V&&inaire desPays Tropicaux, 40,266-274. Donnelly, F. A., Appleton, C. G. & Schutte, C. H. J. (1983). The influence of salinity on certain aspects of the biology of Bulinus (l’hysopis) afiicanus. International Journal for Parasitology, 13.539-545. Donnelly, F. A., Appleton, C. G. & Schutte, C. H. J. (1983). The influence of salinity on the ova and miracidia of three species of Schistosoma. International Journal for Parasitology, 14, 113-120. Fulford, A. J. C. (1994). Dispersion and bias: can we trust geometric means? Parasitology Today, 10,446-448. Go11 P. H. (1981). Mixed populations of Bulinus senegalensis (Muller) and B. forskah [sic] (Ehrenberg) (Mollusca: Planorbidae) in The Gambia. Transactions of the Royal Society of Tropical Medicine and Hygiene, 75,576-578. Hira, P. R. (1969). Transmission of schistosomiasis in Lake Kariba, Zambia. Nature, 224,672-676. Hunter, J. M., Rey, L., Chu, K. Y., Adekolu-John, E.O. & Mott, K.E. (1993). Parasitic Diseases in Water Resources Development: the Need for Intersectional Negotiation. Geneva: World Health Organization. Katz, N, Chaves, A. & Pellegrino, J. (1972). A simple device for quantitative stool thick smear technique in schistosomiasis mansoni. Revista do Instituto de Medicina Tropical de Sao Paula, 14,397-400. Malek, E. A. & Chaine, J. P. (1989). Effects of the developments in the Senegal River Basin on the prevalence and spread of schistosomiasis. In: Demography and Vector Borne Diseases, Service, M. W. (editor). Boca Raton, Florida: CRC Press, pp. 181-192. Miller, M. (1981). Parasites of man and arthropod disease vectors in communities of a water development program on the Senegal River Basin. Report. Calgary, Alberta: University of Calgary. Mimpfoundi, R. & Greer, G. J. (1989). Allozyme comparisons among species of the Bulinus forskalii group (Gastropoda: Planorbidae) in Cameroon. rournal of Molluscan Studies, 55,
M. PICQUET
ETAL.
405410. Paperna, I. (1969). Study of an outbreak of schistosomiasis in the newly formed Volta Lake in Ghana. Zeitschr$&r Tropenmedizin und Parasitologic, 21,339-353. Polderman, A. M., Kayiteshonga Mpamilla, Manshande, J. P. and Bouwhuis-Hoogerwert, M. L. (1985). Methodology and interpretation of a parasitological surveillance of intestinal schistosomiasis in Maniema, Kivu province, Zaire. AnnaZes de la Societe Belge de Medecine Tropicale, 65,243-249. Rollinson, D. & Southgate, V. R. (1979). Enzyme analyses of Bulinus afiicanus group snails (Mollusca: Planorbidae) from Tanzania. Transactions of the Royal Society of Tropical Medicine and Hygiene, 73,667-672. Southgate, V. R., Rollinson, D., Ross, G. C., Knowles, R. J. & Vercruysse, J. (1985). On Schistosoma curassoni, S. haematobium and S. bovis from Senegal: development in Mesocricetus auratus, compatibility with species of Bulinus and their enzymes.Joumal of Natural Histoy, 19, 124991267. Stelma, F., Talla, I., Polman, K., Sturrock, R. F. & Gryseels, B. (1993). Epidemiology of Schistosoma mansoni infection in a recently exposed community in northern Senegal. American Journal of Tropical Medicine and Hygiene, 49,701-706. Talla, I, Kongs, A., Verle, P., Belot, J., Sarr, S. & Coll, A. M. (1990). Outbreak of intestinal schistosomiasis in the Senegal River Basin. Annales de la Societe’Belge de Medecine Tropicale, 70,173-180. Talla, I, Kongs, A. & Verle, P. (1992). Preliminary study of the prevalence of human schistosomiasis in Richard-Toll (the Senegal river basin). Transactions of the Royal Society of Tropical Medicine and Hygiene, 86, 132. Vercruysse, J., Southgate, V. R. & Rollinson, D. (1985). The epidemiology of human and animal schistosomiasis in the Senegal river basin. Acta Tropica, 42,249-259. Vercruysse, J., Southgate, V. R., Rollinson, D., De Clercq, D., Sacko, M., De Bont, J. & Mungomba, L.M. (1994). Studies on transmission and schistosome interactions‘in Senegal, Mali and Zambia. Tropical and Geographical Medicine, 46,220-226. Verb?, P., Stelma, F., Desreumaux, P., Dieng, A., Diaw, O., Kongs, A., Niang, M., Sow, S., Talla, I:, Sturrock, R. F., Gryseels, B. & Capron, A. (1994). Preliminary study of urinary schistosomiasis in a village in the delta of the Senegal river basin, Senegal. Transactio& of the Royal Society of Troiical Medicine and Hygiene, 88,401-405. WHO (1983). Urine filtration technique for Schistosoma haematobium. Geneva: World Health Organization, mimeographed document PDPi83.4. -Details of the prevalence and intensity of schistosomiasis in wale and d&e (lower and middle vallevs) and the environs of Lac de Guiers,‘and a list of villages in which schistosomiasis was not detected, can be obtained on request from Professor J. Vercruysse, Department of Parasitology, Faculty of Veterinary Medicine, Salisburylaan 133, B-9820 Merelbeke, Belgium. Received 3 January ceptedfor publication
1996; revised 14 February 14 February 1996
1996; ac-