Transmission dynamics of miracidia of Schistosoma haematobium in the Volta Lake*

Bulletin of the World Health Organization, 59 (4): 555-560 (1981) Transmission dynamics of miracidia of Schistosoma haematobium in the Volta Lake* K. Y. CHu,1 J. A. VANDERBURG,2& R. K. KLUMPP 2 Schistosoma haematobium miracidia were detected in sentinelsnailsplaced in 16huma water contact sites in the Volta Lake, each month from March 1973 to November 197: Resultsshowed that rates ofinfection wereseasonal, and that infectedsnails weremore ofte found in water contact sites sheltered by emergentplant growth than in exposed open beac sites with no emergent vegetation. Sentinel snail infection rates were correlated with natur snail iqfection rates and with epidemiological levels ofschistosomiasis in village inhabitant After two years of chemotherapy and mollusciciding, levels of disease and sentinel sn4 infection rates dropped in two-thirds ofthe villages. In the remaining villages, however, ti sentinel snail infection rates were not correlated with the fall in epidemiological leve because of ecological changes in the water contact sites. It is concluded that, unless control measures are continued, the constant changes in ti lake shore environment will lead to a rapid re-establishment ofprevious levels of diseaJ transmission. The study of the transmission dynamics of schisto- somiasis is essential to an understanding of the epidemiology of the infection. Increasing attention is being given to the dynamics of miracidia in the trans- mission cycle. Schistosome miracidia cannot be stud- ied directly in natural waters, and the only way to monitor changes in miracidial density is to study infec- ted snails. However, the constant fluctuation of natural snail populations precludes their use and the only practical approach is to expose fixed numbers of laboratory-bred (sentinel) snails in natural trans- mission sites at regular intervals. Such studies on Schistosoma mansoni in St Lucia have been reported by Upatham (1-4). This report presents the results of studies on the miracidial transmission of S. haemato- bium, which formed part of the WHO/UNDP Schis- tosomiasis Project on the Volta Lake, and which were intended to assess the effects of various factors, especially chemotherapy and mollusciciding. MATERIALS AND METHODS Snail breeding and rearing Wild Bulinus roh(fsi, the intermediate host of S. haematobium, were collected from the Volta Lake. * From the WHO Schistosomiasis Research Project, P.O. Box M190, Accra, Ghana. 1 Senior Biologist. 2 Biologist. Five adult snails were placed in each of sixty 10-litre plastic bowls containing 6 litres ofstream or lake water and kept in a shaded, airy room. Maximum water tem- perature ranged between 27 °C and 31 OC, and the water was changed every 2-3 days. The adult snails were allowed to lay eggs for a period of ten days, after which they were removed and the eggs left to hatch. Baby snails were fed tropical fish food or, in very hot weather, a mixture of fish food and dried lettuce powder. After two weeks of age, snails were fed only dried lettuce. Up to 50 snails could be raised in each plastic bowl. Sentinel snail cage The cages were copied from the design used in the Rockefeller Project in St Lucia (1), and were made of 2-mm mesh nylon screen, in the shape of an envelope, 10 x 15 cm. The edges of the cages were heat-sealed. A U-shaped piece of PVC tubing was used as an internal support. The cages were covered with a protective welded wire frame to prevent damage and thus improve sentinel snail survival. Location ofcages in the water In an attempt to determine the most suitable location for sentinel snail cages, an experiment on the dispersal of miracidia was conducted in the field. An area of the lake away from human water contact sites was selected, where the water was 1 m deep. Cages were placed in concentric circles 1-2 m from a central 4087 -555 556 K. Y. CHU ET AL. point. One-third of the cages were suspended 10 cm below the water surface, one-third 50 cm deep, and the remainder 100 cm deep. Urine containing about 150 000 S. haematobium eggs was then poured into the water at the central point. The snails were collected from the lake 24 hours later, kept for 27 days, and crushed to examine for cercariae. None of the snails kept at the 10 cm and 50 cm depths became infected, but 11 out of 68 (16.2%) placed at the lowest level were found to be positive. This result agrees with Shiff's finding (5, 6) that S. haematobium miracidia tend to be more concentra- ted near the bottom of shallow field water. It is poss- ible that the eggs hatched close to the bottom cages and therefore miracidia were most concentrated there, while local water currents probably reduced the con- centration at the higher levels. Hence, it was decided to place all sentinel snail cages near the bottom of the lake near shore during the monthly field surveys. Determination of infection It was initially intended to study the relationship between the number of daughter sporocysts and the number of successful miracidial penetrations of the sentinel snails. However, under field conditions, it was impossible to detect clearly the mother and daugh- ter sporocysts. Different rates of maturation of the daughter sporocysts made examination tedious, and the number of abnormal or amorphic forms ofdaugh- ter sporocysts (7-9) made it impossible to differen- tiate various types of trematode infection from S. haematobium. Assessment was therefore based simply on the presence of mature or immature schisto- some cercariae upon crushing of the snails 30 days after retrieval from the lake. Field studies Two-month-old laboratory-bred B. rohlfsi were exposed each month in the main human water contact sites of 16 selected villages in the project area. In each village, 10 cages containing a total of 100 snails were placed in the water and left in situ for two days. After retrieval, the exposed snails were kept in the labora- tory for 30 days before they were crushed and exam- ined with a dissecting microscope for the presence of schistosome cercariae. Data collection began in March 1973 in 8 villages and was extended to 16 villages during March and April 1974. Results obtained in this early period up to completion of the first round of selective population chemotherapy (SPC 1) in February 1976 were taken as baseline data. Two further rounds of chemotherapy were completed in 1976 and 1977, and data collected during this period should reflect the results of the campaign. No reliable data were obtained after November 1977, because of a lack of suitable snails. The level of disease in the human population is expressed as an epidemiological index. This is defined as the product of the disease prevalence rate and the geometric mean of eggs counted in 5 ml samples of positive urine, divided by 100. The pre-intervention epidemiological indices were based on data collected in a survey of all 26 villages in 1974. The post-inter- vention parasitological surveys were carried out 6-10 months after the start of SPC 1 and 4-9 months after SPC 2. Monthly sampling of field snails was also conducted in 14 of the 16 WCSs used for sentinel snail exposures. Snails were collected by the palm-mat sampling method in 8 villages and by the modified man-time method, using dip-nets, in 6 villages (10). All snails collected were brought back to the field laboratory and examined for both mature and immature cer- cariae. As in the sentinel snail programme, the pre- intervention baseline data period was from March 1973 to May 1975, when intervention by focal mollus- ciciding was started in all suitable WCSs.a RESULTS Seasonal variations The pre-intervention baseline data for the monthly sentinel snail infection rates and the annual lake water level fluctuation, between March 1973 and February 1976, are shown in Fig. 1. Infection rates were found to be seasonal, with high rates at periods of high water level and low rates at periods of low water level. a The mollusciciding programme did not affect the sentinel snail programme, since cages were always retrieved before the monthly spraying with niclosamide. Molluscicidal action in a WCS was short- lived and there was no residual effect because the WCSs shifted con- tinually with the rise and fall in the water level. 6626 1973 - 1974 10 61 2 84-4 0 2 3 1974 - 1975 -1Z 02.79 I - F -7969 94 -1it6917 -1975 -1976fIL 82 9 -9 A M J A S 0 9 D J F Fig. 1. Water level fluctuations (dashed line) and sentinel snail infection rates in water contact sites (solid line), in the Volta Lake, 1973-76. TRANSMISSION OF S. HAEMA TOBIUM MIRACIDIA Table 1. Number of water contact sites found positive in monthly sentinel snail exposures, 1973-76 Type March 1973- March 1974- March 1975- TYPe of WCS February 1974 February 1975 February 1976 Total No. positive/ No. positive/ No. positive/ No. positive/ No. tested % No. tested % No. tested % No. tested % Open beach 11/42 26.2 35/65 53.8 24/75 32.0 70/182 38.5 Pocket 14/31 45.2 45/60 75.0 33/55 60.0 92/146 63.0 Channel 7/23 30.4 18/34 52.9 18/35 51.4 43/ 92 46.7 Ecological types of WCS Water contact sites in the lake were subject to change because of the annual cycle of water level fluctuation (10). Three main types of WCS were observed: open beaches, pockets, and channels. Open beach sites were most common at low water level and were found on wide exposed stretches of shoreline where little or no emergent vegetation grew in the water. Channel-shaped sites, mainly cut through dense Polygonum senagalense from the shore to the open water, started to appear in late August when the water level began to rise, and predominated from high water level to the early drawdown phase. Pocket- shaped WCSs were also numerous at high water level and early to mid-drawdown, i.e., from September to March. Details of the formation of such WCSs are described elsewhere (10). If any sentinel snails became positive after exposure in a WCS, that site was considered positive for mira- cidial transmission. Table 1 shows the number of times each WCS was positive between 1973 and 1976, grouped according to the type of site. Over the three- year period, the number of sentinel snail infections in pocket-shaped WCSs was significantly higher than that in channel-shaped sites (x2= 6.09, P<0.05). There was no significant difference in infection rates between channel-shaped and open beach WCSs (x2= 1.73, 0.1<P<0.2). Natural snailpopulations In the main WCSs used for both sentinel snail exposures and ecological snail sampling, the highest numbers of field B. rohlfsi and of infected specimens were found in January, February, and March. Few snails were found in September and October because of the rapidly rising water level. At lowest water levels (April-July), B. rohlfsi were found almost exclusively in WCSs with moderate to heavy growths of Cerato- phyllum. From Fig. 1, it can be seen that peak sentinel snail infection rates occurred in March 1973, January, November, and December 1974, and October and November 1975. Except for the latter two months, these peak transmission periods coincided with the peak periods of field snail infections. Comparative results of sentinel snail and field snail infection rates in the WCSs of 14 villages from March 1973 to May 1975 are presented in Table 2. At Pawm- pawmnya No. 1, on the steeper eastern shore of the lake, 2157 sentinel snails were examined and only 12 were positive, an infection rate of 0.56%. During 27 consecutive months of field snail sampling at the same site, only one snail was found (in December 1974 when the site was pocket-shaped in emergent vegetation for one month), and it was positive for cercariae. The epidemiological index of Pawmpawmnya No. 1 has always been low because the WCSs in the village have almost always been open beaches devoid of weed growth. If this atypical village is omitted from the analysis, infection rates of sentinel and field snails in the remaining 13 villages were significantly correlated (r = 0.52, P<0.05). Epidemiological levels ofschistosomiasis The epidemiological indices of disease and the over- all sentinel snail infection rates in all 16 villages before the completion of SPC1 are shown in Table 3. The snail infection rates were significantly correlated with the epidemiological indices of schistosomiasis in the villages(r = 0.79, P<0.01). Effects of intervention Table 4 shows the sentinel snail infection rates before completion of SPC 1 and the post-intervention data collected after completion of SPC 1 and SPC 2. Of 16 villages studied, a significant reduction in miracidial transmission occurred in 9 villages follow- ing SPC 1 and in 11 villages after SPC 2. In the villages of Kasa and Asakeso, miracidial transmission rose following drug treatment because the villagers changed from their original WCS to a new, smaller one. In Pawmpawmnya 1, positive results were 557 558 K. Y. CHU ETAL. Table 2. Sentinel and field snail infection rates in the water-contact sites of 14 villages, March 1973-May 1975 Sentinel snails Field snails Village No. positive/No. tested % No. positive/No. tested % Pawmpawmnya No. 1 12/2157 0.56 1/1 100.0 Fatem 97/2303 4.21 11/145 7.59 Kasa 95/2272 4.18 8/140 5.71 Poakwe Pawmpawmnya 32/2327 1.38 14/783 1.79 Dawa Kofi 55/1308 4.2 12/97 12.37 Akokoma 31/1107 2.8 1/71 1.41 Kwabia 64/2177 2.94 1/31 3.23 Kuma Kuma 91/2201 4.13 19/106 17.92 Asakeso 45/2146 2.1 16/457 3.5 Akotui West 113/2295 4.92 171/1518 11.26 Tamayeso 87/1068 8.15 4/49 8.16 Nyafutu 29/965 3.0 0/32 0 Dukuase 67/974 6.88 32/367 8.72 Odortom II 87/1165 7.47 35/330 10.61 Total 905/24 465 3.70 325/4127 7.87 a If results from Pawmpawmnya No. 1 are omitted, r = 0.52, P<0.05. Table 3. Pre-intervention epidemiological index and sentinel snail infection rates in 16 villages, 1974-75 Epidemiological Snail infectionVillage index° rate (%) Odortom II 72.26 7.88 Tamayeso 67.90 6.56 Akotui East 54.85 3.44 Dukuase 54.74 8.11 Akotui West 49.65 4.54 Nyafutu 47.13 2.58 Akrusu 42.26 7.82 Dawa Kofi 37.98 3.77 Fatem 31.91 3.39 Kasa 27.64 3.35 Poakwe Pawmpawmnya 25.00 1.14 Kwabia 23.94 2.52 Kuma Kuma 23.32 3.58 Asakeso 22.56 2.04 Akokoma 21.52 1.91 Pawmpawmnya No. 1 5.79 0.45 ° Epidemiological index = disease prevalence x geometric mean egg count per 5 ml of urine/ 100. obtained in only two of the 33 pre-intervention expo- sures, providing further evidence that transmission was low and sporadic. At Akotui East, the WCS used for the sentinel snail exposures was frequently pocket- shaped before the start of chemotherapy, but later became an open beach habitat because of the continu- ing drop in lake level during 1976 and 1977. No infec- ted snails were found in this WCS after SPC 1, even though the epidemiological index in the village remained fairly high. At Odortom II, a significant reduction in miracidial transmission was obtained after SPC 1, but transmission increased after SPC 2. This occurred after the falling water level enabled off- shore Ceratophyllum to invade the WCS and to main- tain a submerged, semi-barrier around it, keeping the water calm and helping to concentrate miracidia closer to the sentinel snails. DISCUSSION The changes in the lake water level greatly affected the miracidial detection programme. Each year, sentinel snail infection rates were highest in the season of early lake drawdown and lowest at low water. At high water level, WCSs were smaller because of the TRANSMISSION OF S. HAEMA TOBIUM MIRACIDIA Table 4. Effects of the chemotherapy programme on sentinel snail infection rates in 16 villages Pre-intervention Post-intervention 1976w 1g77b Village No. positive/ % No. positive/ No. positive/ No. tested No. tested % No. tested % Pawmpawmnya No. 1 12/2697 0.45 0/934 0 0/569 0 Fatem 97/2858 3.39 3/919 0.33c 5/812 0.62c Kasa 95/2833 3.35 68/938 7.25c 101/812 12.44C Poakwe Pawmpawmnya 33/2898 1.14 9/943 0.95 1/824 0.12C Dawa Kofi 69/1831 3.77 24/854 2.81 6/632 0.95c Akokoma 31/1619 1.91 10/885 1.13 4/676 0.59C Kwabia 69/2735 2.52 2/883 0.23c 1/558 0.18C Kuma Kuma 103/2878 3.58 4/868 0.46C 3/826 0.36C Akrusu 82/1049 7.82 33/751 4.39C 20/716 2.79c Asakeso 57/2791 2.04 28/785 3.57C 25/705 3.55C Akotui West 134/2949 4.54 20/977 2.05c 5/585 0.85C Akotui East 41/1193 3.44 0/691 oc 0/577 oc Tamayeso 118/1799 6.56 17/791 2.15c 9/662 1.36c Nyafutu 43/1664 2.58 10/658 1.52 13/660 1.97 Dukuase 139/1714 8.11 28/734 3.81c 21/672 3.13C Odortom II 163/2069 7.88 6/654 0.92C 44/628 7.01 8 After one round of chemotherapy. b After two rounds of chemotherapy. c Significant difference pre- and post-intervention, P <0.05. wide zone of marginal emergent vegetation, and the majority of the sites were pocket-shaped. The side vegetation and the shore itself created barriers which concentrated the miracidia inside the WCS while keeping the water calm. Such well-defined sites made it very easy to pinpoint the centre of human water- contact activity. In this condition, both miracidial and cercarial transmission were focal and consistent (11). Channel-shaped WCSs were also common at high water levels, but because human activity was confined to narrow spaces near the shore, the sentinel snail cages in the water were often disturbed. The increased pollution near the shore also increased sentinel snail mortality and reduced miracidial infection rates. Almost all WCSs were of the open beach type at low water level each year. With no emergent vegetation in the water to confine human activity, water contact was more diffuse and miracidial density was considerably reduced. Sentinel snail infection rates in these WCSs were always very low except when moderate to heavy growths of Ceratophyllum were present. While it is often difficult to pinpoint human water- contact sites in other schistosome habitats such as irrigation canals, ponds, and swamps, theWCSs in the Volta Lake were usually very distinct and consistent. Intensity of water contact was high, not only because the villagers were mainly fisherfolk, but also because the lake water was of good quality and therefore attractive as a water source. Also, during the hot season, children often used the open beach WCSs for swimming and playing. With such intensive water con- tact in all villages, a correlation between sentinel snail infection rates and the epidemiological index of the disease in the villages would be expected. In the St Lucia schistosomiasis control project mira- cidial detection using sentinel snail exposure was used to help evaluate the efficacy ofchemotherapy (12, 13), and no infected snails were found after chemotherapy had been started. In the present project, infection rates of sentinel and field snails were found to be proportional to the epidemiological indices of schisto- somiasis in the villages in the pre-intervention period. However, the results after intervention were quite different. Sentinel snails were uninfected in only 2 of 16 villages studied. In 9 of the remaining 14 villages, the reduction in sentinel snail infection rates was 559 560 K. Y. CHU ET AL. proportional to the reduction in epidemiological levels of the disease. In the remaining 5 villages, however, there was no such relationship because of the eco- logical conditions of the WCSs changed thus favour- ing transmission. Control of schistosomiasis in this area is difficult because of: (1) the poor cure rate of metrifonate in a high endemic area, (2) the high level of migration of the population looking for more productive fishing and farming grounds, (3) the large number of un- treated people living in hinterland villages, 1-5 km from the lake, and (4) the large proportion of untreated people living in the villages undergoing intervention. Even where the project reduced epi- demiological indices and sentinel snail infection rates by 901o, the residual disease could lead to a build-up of cercarial transmission within 2-3 months if snails were present during the main transmission season. From this study, it can be concluded that chemo- therapy, mollusciciding, and limited well-water supply are not sufficient to stop transmission of S. haematobium, although they achieved excellent results in the short-term control of transmission and infection. Permanent control would require the virtual eradication of infection from the project area, the nearby hinterland villages, and a wide zone along the lake on each side of the project area. ACKNOWLEDGEMENTS We are grateful to Dr E. G. Beausoleil, Director of Medical Services, Ghana Ministry of Health, for his active support of the work, and for permission to publish the results. We want to thank Dr A. Davis, Dr G. Webbe, and Dr D. Scott for tech- nical and administrative support, Dr K. Senker and Mr E. C. England for permission to use their epidemiological data, Mr H. Dixon for data and statistical assistance, and Mr Ben Ocloo and Mr C. K. Agbezuke for technical assistance. The work was supported in part by grants from the Edna McConnel Clark Foundation. RtSUME DYNAMIQUE DE LA TRANSMISSION DES MIRACIDIES DE SCHISTOSOMA HAEMA TOBIUM DANS LE LAC VOLTA Pendant cinq ans, on a effectue des etudes sur la dynamique de la transmission des miracidies de Schistosoma haematobium tous les mois dans seize villages lacustres en plagant des mollusques sentinelles dans des lieux de contact homme-eau. On s'est aper,u que les taux d'infection des mollusques sentinelles avaient un caractere saisonnier, qu'ils etaient eleves lorsque le niveau de l'eau etait haut et peu eleves lorsque le niveau de l'eau etait bas. On a decouvert que les mollusques etaient plus souvent infectes dans les petites poches d'eau que dans les plages ou canaux largement ouverts. Les maximums dans les taux d'infection des mollusques sentinelles etaient comparables a ceux des mollusques vivant A l'etat naturel et correspondaient aux niveaux epidemiologiques de la schistosomiase parmi les habitants locaux. Apres deux ans d'intervention par chimiotherapie et l'emploi de molluscides, les taux d'infection des mollusques sentinelles n'avaient diminue que dans deux tiers des villages. Dans les autres, le changement dans le taux d'infection des mollusques sentinelles ne correspondant pas A la baisse des niveaux epidemiologiques, en raison de modifications ecologiques dans les lieux de contact avec l'eau. Certaines modifications de ces lieux de contact entraineront rapidement une reprise du taux de transmission de la maladie meme apres une baisse de 90%o des niveaux epid&miologiques. REFERENCES 1. UPATHAM, E. S. Journal ofhelminthology, 46: 297-306 (1972). 2. UPATHAM, E. S. Journal ofhelminthology, 46: 307-315 (1972). 3. UPATHAM, E. S. Internationaljournal ofparasitology, 3: 289-297 (1973). 4. UPATHAM, E. S. Internationaljournal ofparasitology, 6: 239-245 (1976). 5. SHIFF, C. J. Journal of parasitology, 54: 1133-1140 (1968). 6. SHIFF, C. J. Journal of parasitology, 55: 108-110 (1969). 7. CHERNIN, E. & DUNCAN, C. A. American journal of tropical medicine and hygiene, 11: 455-471 (1962). 8. DiCONZA, J. J. & BASCH, P. F. Journal of parasito- logy, 60: 550-55 1 (1974). 9. HUSSEY, K. L. & STAHL, W. B. Journal ofparasitology, 47: 445-446 (1961). 10. KLUMPP, R. K. & CHu, K. Y. Bulletin of the World Health Organization, 55: 715-730 (1977). 11. CHu, K. Y. & KLUMPP, R. K. In: Proceedings of the International Conference on Schistosomiasis, Cairo, 1975, Vol. 1, 1978. pp. 85-88. 12. COOK, J. A. ET AL. American journal of tropical med- icine and hygiene, 26: 887-893 (1977). 13. CHRISTIE, J. D. & UPATHAM, E. American journal of tropical medicine and hygiene, 26: 894-898 (1977).