Distribution and severity of onchocerciasis in southern Benin, Ghana and Togo

ONCHOCERCIASIS DISTRIBUTION AND SEYERITY IN SOUTHERN BENIN, GHANA AND TOGO G. De Sole, S. Accorsi, O. Ba, H. Creusvaux, J. Remme, F. Walsh, J. Henderickx Onchocerciasis Control Programme in West Africa P.O. Box 549, Ouagadougou, Burkina Faso. Abstract The Onchocerciasis Control Programme in West Africa has recently extended its operation in Southern Benin, Ghana and Togo. To estimate the number of people infected and blinded by onchocerciasis and to describe the distribution and severity of the disease 99, villages were selected, using a stratified random sample procedure and surveyed. All the ecological and entomological information available was used in the sampling procedure and in the selection of 87 non representative villages surveyed to confirm the findings. The study estimated that 590,468 people are infected and I1,715 blinded out of a rural population of 1,828,234. The Pru, Asukawkaw and Mono river basins were areas with high risk of onchocercal blindness. The Oueme and Zou river basins in Benin and the mountainous areas between Ghana and Togo were classified as areas with medium risk of onchocercal blindness. The other parts of the study are presented low or no risk of onchocercal blindness by detecting the communities at risk of onchocercal blindness. This study permitted the selection of the population to be targeted for disease control based on mass treatment with ivermectin, a new microfilaricide. Page 1 INTRODUCTION The Onchocerciasis Control Programme in West Africa (OCP) started activities in 1974. A few years later it was realized that the Programme borders were invaded yearly by S. damnosum s.l. flies originating from outside the programme area. In order to achieve its objective, i.e. to eliminate onchocerciasis as a problem of public health and socio-economic importance, and to ensure that there is no recrudescence of the disease thereafter, the Programme extended its vector control activities to the areas identified as sources of invading flies. The success or failure of the vector control was asessed in the original OCP area by the prospective parasitological evaluation of indicator villages (Prost et al, 1975). The same evaluation technique was used in the Eastern extension of OCP, which includes the Sourthern portion of Benin, Ghana and Togo. In the Western extension, however, epidemiological mapping based on a stratified random sample of villages was conducted to provide a reliable estimate of the number of people blinded and infected with O. volvulus; to describe the distribution and severity of the diseaes; to select the river basins where the population is at high risk of onchocercal blindness for treatment with ivermectin, a new microfilaricide; and to collect baseline data for future evaluation of control activities (De Sole et al, in press). The mapping yielded so much important information for the planning and evaluation of control activities with larvicides and the new drug that it was utilised also in the Eastern extension. MATERIAL AND METHODS Sampling Procedure i. Stratification of the Eastern extension in zones. The Eastern extension was subdivided into 9 zones characterised by ecological, entomological and demographic criteria(see Fig. l). The characteristics of each zone are depicted in Table I and 2. On the basis of the information available, the ecological and entomological features considered were believed to be largely uniform within a zone. Estimation of the rural population of each zone The rural population of each zone was estimated from the most recent national census (see Table 4). The rural population of azone is defined as the total population living inside azorre minus the population living in the district capitals and villages with more than 1000 inhabitants. iii. Mapping of river sections inside each zone. The rivers of each zone are divided into sections with high vector production, low vector production and no vector production (see Fig. l). The differentiation of river sections as high productive, low productive and non productive was based on the judgement and knowledge of the two most experienced OCP entomologists working in the eastern extension. The information on the vectorproductionof eachriverof azor.e wererecordedonamapwithscale l:1,000,000. Anyarea of approximately 2,000 square kilometers without breeding sites was identified and called empty area. iv. Numbering of river stretches and empty areas. River sections were subdivided into stretches of about 50 km of length. The exact delimitation of the stretches was however influenced by the need for a clear landmark such as a tributary, or a major bend of the river. Zones boundary, always delimited a stretch; country borders do not. Change in the type of vector production from high to low or nil always marked the borders of a stretch. The stretches of each zone were numbered starting from one extreme point and following the river flow. Stretches were numbered by zone and by type of stretch. Empty areas were numbered by zone. Page 2 v. Random selection of river stretches and empty areas The river stretches and empty areas were selected using a table of random numbers. The selection was done by zone and the number of stretches and empty areas selected was related to the size of the rural population of the zones (see Table 3). Empty areas cover most of zone I and zone 8 (see Fig. l). Therefore in these zones, two empty areas were selected instead of two high and two low productive river stretches. vi. Boundary of the area allocated to a selected river stretch or empty area. Before proceeding to the choice of villages the geographical area allocated to each stretch or empty area was identified on maps with scale l/200.000 as follows. Stretches: The upstream and downstream borders of a selected river stretch were marked. A line was then traced from each border to the next river with breeding sites located to one side of the selected stretch. The area so delimited was then divided into half. The half adjacent to the selected river stretch constituted one side of the area allocated to it. The process was then repeated for the other side of the river stretch. Empty areas: Empty areas were areas without known breeding sites. They were bordered by areas assigned to river stretches or other empty areas. A zone boundary always delimited the area allocated to an empty area. vii. Random selection of villages Within the area allocated to a selected river stretch, all first line villages, i.e. the villages closest to the river banks, were numbered and two of them were randomly selected using a table of random numbers. Thereafter all the other villages within the allocated area were numbered and two were similarly selected. The location of the selected 99 villages is depicted in Fig. 2. Because of the inaccuracy of the maps a number of amendments to the selected villages were foreseen. The amendments were made using the following rules. Problem Rule l. A village has disappeared: The disappeared village should be replaced by the nearest village of at least 100 people of the same category. If in a river stretch selected there are only two villages in one category both are automatically selected but if one of them has disappeared it cannot be replaced 2. A village is too large: Villages that are administrative centers or with a population of more than 1000 people should not be part of the sample. If a village presents one or both these characteristics should be dropped or replaced following the rules expressed in point l. 3. A village is too small: If a village has a population of 100 or less people an extra village should be added. The village added should be the nearest one to the selected small village and of the same category. Page 3 Non rapresentative villages To clarify the distribution and severity of onchocerciasis in specific areas of the Eastern exstension 87 villages mainly located near major breeding were surveyed. Skin snip survey Each selected village was surveyed using the standard OCP methodology (Prost & Prod'hon, 1978). This involves taking the census of the village population, and carrying out a microscopic examination of two skin snips, taken from the iliac crests of each villager, for the presence and number of O. volvulus microfilariae (mf) after 30 minutes incubation in distilled water. Data analysis The skin snip and ophthalmological surveys were processed and analyzed on personal computers using routine programmes developed in the OCP. The villages are characterized by their standardized prevalence of infection and blindness and by their community microfilarial load (CMFL) which is the geometric mean number of microfilariae per snip in adults with an age of 20 years or above (Remme et al, 1986). Oncocercal infection and total blindness are estimated from the rural population of the zones as established by the most recent national census. The estimates of onchocercal blindness in all the extension area, excluding Sierra Leone are calculated from the CMFL observed using a regression factor between CMFL and onchocercal blindness developed from the analysis of the results of 33 villages with 10,870 inhabitants located in the savanna zone of the original OCP area (Remme et al, 1989). The estimate of ochocercal blindness in Sierra Leone is likewise calculated from l7 local villages with 3252 inhabitants. RESULTS Estimated oncocercal infection and blindness The estimation of the number of people infected and blinded with onchocerciasis in each zone is reported in Table 4. In zones I to 8 590,468 people are estimated to be infected and ll,7l5 to be blinded with onchocerciasis out of a rural population of 1,878,234. [n zone 9 no comparable estimation of the number of people infected and blinded with onchocerciasis was possible because this zone had been treated with ivermectin. However, the endemicity of this area was studied before the treatment and reported elsewhere (Remme et al, 1989). The population of zone 9 is 26,1 l7 with very high prevalence and intensity of infection in the villages located along the river Pru and far lower infection rates in the communities farther away from the river. Distribution and severity of onchocerciasis The distribution and severity of onchocerciasis in the study area is described in Fig. 4. The severity of the disease in a zone is determined by the estimated prevalence of onchocercal blindness (see Table 4). A zone with an estimated prevalence of onchocercal blindness equal or above I 0/o is classified at high risk of onchocercal blindness. A zone with a prevalence of onchocercal blindness between 0.2 o/o and 0.9 0/o is classified at medium risk of onchocercal blindness. A zone with a prevalence of onchocercal blindness below 0.2 0/o is classified at low or no risk of onchocercal blindness. Based on this classification the Mono river basin in zones 2 and 54 and the Asukawkaw river basin in zone 68 are classified as zones with high risk of onchocercal blindness. Zone 9 is likewise classified on the basis of the very high CMFLs, up to 72, recorded in the villages located along the Pru river. Zones l, 3, and 4 in Benin and the mountainous area between Ghana and Togo including zones 5C, 6A and 7 A are classified as zones with medium risk of onchocercal blindness. Zone 5B in Togo and zones 78 and 8 in Ghana are classified as zones with low or no risk of Page 4 onchocercal blindness. The westernmost part of zone 2 is classified as an area with low risk of onchocercal blindness because no breeding sites are reported from its rivers and the two villages surveyed had very low CMFLs, 1.4 and 0.6. The CMFLs of the non rapresentative villages fully support the above classification of the 9 zones. DISCUSSION This was the second epidemiological mapping undertaken by OCP. The same methodology used in the mapping of the Western extension was adopted. The Southern extension is definitely smaller than the Western extension and the entomological information available was more abundant and accurate. Furthermore, the mapping of the former area benefitted greatly from the previous experience. Therefore, the zones were smaller, more accurately defined and no mistake in the straification, as in the northern part of the western extension, was made. The uniform distribution and severity of the disease within the zones indicate that the assumpion of homogeniety of the zones was correct. The division of rivers into high and low vector production stretches was generally correct, with the villages of the high production stretches more infected than those of low production stretches. Only in the zones with low or no risk of onchocercal blindness there was no appreciable difference in the CMFLs of the sampled villages of both type of river stetches. The ratios between the rural population estimated from the sample and obtained from the the national census per zone were far better in the southern extension than in the western extension where a systematic underestimation of the sample population was noted. This is most probably due to the availability of better maps in the former area. The mapping exercise classified two limited foci along the Assukawkaw and Pru rivers and the Mono river basin of zone 2 as areas with high risk of onchocercal blindness. The western part of zone 2 is nonetheless classified as an area with low or no risk of onchocercal blindness, because it has no reported productive river stretches and the two sampled villages surveyed had very low CMFLs of 1.4 and 0.6. Considering that this area is definitely separated from the Mono river basin it would have been better to include it in zone 8. The Oueme and Zou river basin in Benin and the mountain area between Ghana and Togo are areas with medium risk of onchocercal blindness. In the Oueme and Zou rivers the disease is transmitted by savanna vector species as in the villages of the original area used to calculate the risk of onchocercal blindness. [n the mountain area between Ghana and Togo the disease is transmitted mainly by S. Squamosum a vector never associated with onchocercal blindness. Therefore, the risk of onchocercal blindness in this area is most probably overestimated. As in the western extension, the mapping of the southern extension provided key information for the planning and monitoring of onchocerciasis control activities. The sharp delimitation of areas at risk of onchocercal blindness described in figure 2 have permitted the concentration of control activities, based on the mass distribution of the new microfilaricide, ivermectin, in the most affected areas. The accuracy of the mapping depended not only on the quantity and quality of the information available but also on the good understanding of the transmission prosess and the major factors influencing the risk of onchocercal blindness. The recognition of the importance of the first line villages improved tremendously the reliability of the estimate of onchocercal infection and blindness ( Rolland & Balay, 1976;De Sole et al 1990; De Sole et al, in press). The CMFL value of a village and a transmission caused by savanna vectors were the major factors determining the risk of onchocercal blindness (Remme et al, 1989; Dadzie et al 1989; Dadzie et al 1990, Dadzie et al, in press). ACKNOWLEDGEMENTS We would like to thanks Dr. E.M. Samba, Director of OCP, for his support. We are indebted to Messrs J. Kafando, J. Tiga, O. Coulibaly and B. Kambou for training and supervising the national teams. And the members of the national teams of Benin, Ghana and Togo for executing the surveys. Page 5 REFERENCES Dadzie, K.Y. et al. Ocular onchocerciasis and intensity of infection. tI West African rainforest foci of the vector Simulium Yahense. Tropical Medicine and Parasitology, 40 (1989) 348-354 Dadzie, K.Y. et al. Ocular onchocerciasis and intensity of infection. III West African rainforest foci of the vector Simulium Sanctipauli. Tropical Medicine and Parasitology, (1990) Dadzie, K.Y. et al. Ocular onchocerciasis and intensity of infection. IV. Degraded forest of Sierra Leone. Tropical Medecine and Parasitology, in press. De Sole, G et al. Onchocerciasis distribution and severity in five West African country. Bulletin of the World Health Organization, in press. De Sole, G. et al. Detailed epidemiological mapping of three onchocerciasis foci in West Africa. Acta Tropica, (1990) De Sole, G. et al. Distribution of onchocerciasis in selected river basins of four West African country. Bulletin of the World Health Organization, in press. Prost, A. et al. Methodes d'evaluation epidemiologique de mass de I'onchocercose. Leur utilisation au cours d'un programme de lutte contre le vecteur. Communication to the Expert Commitee on Epidemiology of Onchocerciasis, WHO, Geneva, l0-18 November 1975. Document ONCHO/WP/75 14 (197 5) 2tpp Prost, A & Prod'hon, J. Le diagnostique parasitologique de I'onchocercose, revue critique des methodes en usage. Medecine Tropical 38 (1978) 519-532 Remme, J., O. Ba, K.Y. Dadzie, M. Karam : A force of infection model for onchocerciasis and its applications in the epidemiological evaluation of the Onchocerciasis Control Programme in the Volta River basin area. Bull. Hlth. Org. 64 (1986) 667-681. Remme, J., K.Y. Dadzie, A. Rolland, B. Thylefors: Ocular onchocerciasis and intensity of infection in the community. I. West African savanna. Trop.Med.Parasit.40 ( I 989) 340-347 . Rolland, A & Balay, G. L'onchocercose dans le foyer Bissa. Mimeographed document. OCCGE, No I I l/ONCHO, English translation published by OCP, Ouagadougou (1969). >+,LE .5 (ULEOG'CLEELo(uFO- [il o.EELooFO- Ege= = E CLu,t,/) (l).e OOL+,EEII (J G'E-F! c,oct(,a .r) v, .r't tr1 tt1 v, .r1 il[ilil EOrEg .J1 t/'l a +++++++++++++++++++++++++++++++++ ++++ + ++++++++++++++++++++++++++++++++++++ ++++*++++++++++++++++++++++++++ +qg+EEEEEEEEEEEEELLLLLLLL'oooeJ0.r0J0JooooooF F F t-- O_ 1 t_ o_ o_ o_ o_ o- o- e qo o o E Etr1 tn .n tJ1 tr) + i+ + + + + (, trr tE Er.r (,E'O!!'OE'AtntnsA=nt! rlttttltttttt NJ fi) st tr) lJ) rr1 \o ro N - cc| ol +++++++++++++++++++++++++++ + +lt ++++++++++++++ EJ UJ lrJ l! lrJ lrJFt-FFFF ,oooooo =2,==2,==FFF===1=1d,z.d,d,v)vt.n-A- IJ'I V' I'1 U1 tL E lJ. ooy, >z rrr J 9I =a=o== 4 E d, d, E d, d, x, (5 HliJ liJ UJ uJ = trJ U,J r! lrJ = =++===ooe=o-=ov,+ - a o o F{ = o- O o- 6 -,6 z.O F) o- o-F (5 z, av)<r!I.J FG: H@a tvl = trJ @H - (x:E<FF = U- UJ e, =o F Ft!(5 t! l4J =o d. l+J F d,() (-) C5o O =o =lrJ o z. J (J (l,o -JoT(J trJ (.0 z, a V, UJ z.O NI =o an =lrJ x EJ =d, lrJ t1 lrJ +F oV' lrJ u-o =o o. e,(J a,r'l lrJC! lrj co F t1EUJOHFL)(J lrJ lJ.J O- >.J1 ITABLE 2 : SPECIAL FEATURE OF THE SOUTH EASTERN EXTENSION ZONES II. Very short transmission season. Surprisingly low prevalence of blindness on the Terou river. No suitable breeding sites in most of the zone. Vector mostly damnosum with some soubrense. The Anie is a major tributary rather densely populated. The western part of the zone is without suitable breeding sites. III. Enormous breeding sites - human create breeding sites (fishing traps) South of Zagnanado. More densily populated in the South. a) l{ono below the dam is very regular and highly productive all yearlong. The epidemiological significance of the southernmost rapids should be assessed. The northern part presents savanna type of ocular lesions where the southern part presents mixed vector pattern. IV. vI. a) l4ountainous area. The site of Djodji present the highest ATPs of the Southern extension. VII. b) Presence of the unique Djodji form of the sanctipauli sp. At leastI village Dodo Fie present savanna patterns of ocular lesions. a) The upper Dayi val lcy remain well forested. b) The lower Daye valley has been recently almost totally deforested. At Kudzera 5 cyLospecics reyuiarly breeds. VIII. No suitable breeding sites in most of the zone. Treated with ivermectin and on-going study area for control of transmjs- sion by use of ivermectin. IX. v Table 3 RELATTONSHIP BETI{EEN IHE NTN{BER OF RANDO!{LY SELECTED RIVER STRETCITES AND EMPTY AREAS AND TtrE STZE OF TTTE ZONE POPTILATION Population Stretches Empty areas Size Less than 150 000 150 000 - 500 000 l,lore than 500 000 * if present High 1 2 3 Lov 1 2 3 1* 1 2 illl crll \o[.ll o ilI ilI illl lnll -{ll r-ll-ll -lll .r il ll illl vil.ll -rll coI il lt rl .Ill coll roil vll-ll oll o'll 'nlt il illl c.tll \oI il ilIIIIll -fll cnll c.tll-ll -{ll r-ll -lll.ll -'l ltIll oll c.lll coll-ll @ll r.ll coll.ll -r llll -lll ,Ull +Jll oIHI ilI sf \O c1 CO O \f O f- d O\ -l O\ai (Y) sl N ca O (n Ci r-l C! -l O O(rIOO-lOOOr{OOO tosf\Ocldf-OcnOO\$tr) NO\@\Olf)cOc!ottf)rtnFf-C!O.f Ol (n N-l N .<I-l (1 O-itOf-NcnOmnmlnO rtN-lOmf.c!o\Of-cOorN\o(nC!lf) cnm\orndrt c!r\c\o\ooc!\t v(nlr)sl\O @ CO (') \f, f\ co \O f.. Ot c1 ril\t t-- cc i- .+ co .+ or U-l - c\ st @ c! @ cn i \o s o\ c1 r,o o.| nsf r{f\(n\O-{(n rlri n c{ r{ \O @ \O C! Ol lO C! tr) r{ rt \O nO(nNCIC!N(\lc.llr)sf;C\lC! r{ F{ r{ r{ \O lO F{ O ri O\ .-{ C{ lO (n O't Crl(n(ni{Osfc!r{tOo\Onc,) CQO\(')@sfnr{C\:@e{@\O no\tf)sf@@sf!ilt-|f-f-rl ct \O \O r{ CO to C! (n m F{ [O f-$Gl \O\O(nf-\OSO\(nC!@Otf- r{ f- O\ C{ @ lO \O Ol C! N sf f\!S \O C! \O -l rl @ Cn $ C! Or rl - \o\ r{r{ON('lrlf-nC!F{str{ r{@to\OO(Y)Oc!c!$ OtOilFlNr{CnNd C! G{c\t(n$<OC)<trl4trcoIOtolf)to'or-F o\P -l6oo!o oo)oco6EtrO'-lC-ro.a d o 1J t) G)(}{ -1 tr a G) +)(0 E ._1 1) a 0) o a) +J U q) t.l o U o\o tn I{2Az2HO rlHpq u:2 zts frlzozHgHE:UHE] U) Z! Fr oCfqt{o(J ou)tsH(,Z20ON eco lr:HZ E H Hat{ ! o^ .rl o\o dx oc) -f +JO{dEF(U '.'llj.f1) U) o F{Od5caotr ..{ o +J C) .U e o .Fl +Jd Fl ag. og{ F{ d +) o E{ o tr oN .il 0,, dA oH Figure 1 Figure 2 stratification of the Eastern extension and vectcr production caracteriestics of the rivers. Distribution of the villages surveyed for the epidemiological mapping of the Eastern extension. Figure 3 Risk of onchocercal blindness in the Eastern extension. /!to oo t oa z oFo9 Ya yo o o EI - /\ -2- 0 uailta H Do L o(L z ttJ H 60 H -\-r"--I I \ i st v, H E o ";nollE s]l -o ll .oll ;6J :c5 o C) jlllh Eo:.F E is is 't " 'EE 'iE l.. -c @ -a c, = o 9o.t ?iE <lE E !E;Eiii:E #ilrigss ts{ a o l9 ?79 \ \ I!to I oo z .---/E(o \F o 9 v, oo o fa vo -u\\r- o taa0 oa H oo L0I o o oa0 o o il ono o o a H o -\--ra- oN \ I !Qlt at, (l,o : '= (,=;LEo a;: ,,cx;9, o 3; ac\:o= LEEb3 .(,zNo oo E ;n :il ao H o o o o o o\,s (9 +cE7e ; o >----- UJ /.tlo t oo z oFoI lito -/\-/- auao N \-//\ ttt 9, 9Coi Ig;c . ;n HF59i=t *Jl EEEE; =8""]IeEe€EB5 <o < o N N c) o .I o '-,, -oooCE r-!Oc i=o -Go .=L =$8€c,(,l-C<o %ll ;i H v o o t 'cu7, 3 v) il