The 1970 yellow fever epidemic in Okwoga District, Benue Plateau State, Nigeria

Bull. Org. mond. SantE 1973, 49,113-121 Bull. Wld Hlth Org. The 1970 yellow fever epidemic in Okwoga District, Benue Plateau State, Nigeria 1. Epidemiological observations* T. P. MONATH,1 D. C. WILSON,2 V. H. LEE,3 G. STROH,4 K. KUTEYI,5 & E. A. SMITH' A focal epidemic ofyellow fever occurred in late 1970 in southern Benue Plateau State, Nigeria, I year after a much larger outbreak in northern Nigeria. Like its predecessor, the 1970 epidemic was associated with virus transmission by wild-breeding Stegomyia mosqui- tos. Epidemiological data on the outbreak are reported in this paper. In late 1969 a widespread outbreak of yellow fever was recognized on the Jos Plateau of Nigeria. Investi- gations by Carey et al. (1972) showed that this epidemic, the first in Nigeria in 17 years, ceased rather abruptly in December 1969, and was geographically confined to the plateau and immediately adjacent areas of North Central and North Eastern States. No further cases of yellow fever were recognized in Nigeria until late 1970, a year after the Jos Plateau epidemic. This second outbreak, which occurred more than 193 km south of the plateau in Okwoga District, Benue Plateau State, was more circum- scribed geographically and affected fewer individuals. Geography Okwoga District lies in Oturkpo Division, Benue Plateau State, 45 km south-west of the nearest township, Oturkpo (Fig. 1). The district is readily accessible by road from the Enugu-Oturkpo high- way. The population of 19 772 (1963 census) is divided into 12 villages, the largest being Aidogodo. * From the Epidemiologic Unit, Federal Ministry of Health, Lagos, Nigeria; The Ministry of Health, Jos, Benue Plateau State, Nigeria; and the Virus Research Laboratory, Faculty of Medicine, University of Ibadan, Ibadan, Nigeria. 1 Virologist, Vector-Borne Diseases Division, Center for Disease Control, US Department of Health, Education, and Welfare, P.O. Box 2087, Ft Collins, Col., USA. 'Virus Research Laboratory, Ibadan, Nigeria. Present address: Yale University, New Haven, Conn., USA. 'Entomologist, Virus Research Laboratory, Ibadan, Nigeria. ' Operations Officer, Smallpox Eradication Program, Ministry of Health & Social Welfare, Jos, Nigeria. I Medical Officer, Federal Ministry of Health, Lagos, Nigeria. Consultant Epidemiologist, Federal Ministry of Health, Lagos, Nigeria. Census data from Okwoga and Aidogodo, obtained during these investigations, indicated the following approximate distribution by age: 0-4 years, 20%; 4-9 years, 21 %; 10-14 years, 12%; 15 years or older, 47%. Okwoga District lies within the zone of derived Guinea savannah or savannah-woodland, at an alti- tude of 122-152 m, with an annual rainfall of 152-203 cm. Distinct dry and rainy seasons exist, the latter lasting from April or May to October. In 1970 the last rainfall noted by residents of Okwoga oc- curred on 5 November. The area is characterized by open grassland with stands of deciduous hardwood trees growing 1.5 to 7 m apart. Along permanent streams, heavier gallery forest is found, especially near Aidogodo. Euphorbia, the cactus-like hedging found in association with breeding sites for yellow fever mosquito vectors on the Jos Plateau (Lee, 1972), is not grown. Village compounds are generally quite dispersed and separated by cultivated fields ofyams, coco yams, cassava, tobacco, and other minor crops. Compounds consist of one or more mud-walled, thatched huts around a swept clearing. Storage of water for pro- longed periods is not practised. Water pots are emptied and refilled daily from streams or wells. Interviews with local hunters indicated that mon- keys were quite rare, and repeated hunts for them during these investigations failed. Although galagos probably exist in the area, they are unknown to the local people. A government dispenser is in attendance at Okwoga village, but the nearest physician and hospi- tal are 64 km away by road, at Oturkpo. Most illnesses, especially those associated with jaundice, 3095 -113 T. P. MONATH ET AL. Fig. 1. Okwoga District, Nigeria, site of epidemic yellow fever, 1970-71. are not brought to either the dispenser or the hospital, and the use of native herbal medicines is widespread. First information about the epidemic Investigations began following an informal report of the presence of an unusual febrile illimess with many deaths in several villages in Qkwoga District. The area involved was reached on 20 December 1970, when the epidemic was rapidly declining. Initial interviews with survivors and village chiefs provided a remarkably uniform description of the clinical fea- tures. The onset of illness was characteristically sudden, with high fever, chilliness or frank, shaking chills, and severe generalized or bitemporal headache. Epigastric pain, generalized body aches, and lumbo- sacral pain were frequent early symptoms. Anorexia, nausea, and, less often, vomiting were present; in a number of cases vomitus was blood-streaked. Black vomit was not common. In the few fatal cases, death occurred on the second to fourth day of illness. Jaundice was not noted in fatal cases, nor in a large number of survivors; when present, it developed within a week of onset and persisted for 7-10 days. Intensive efforts to find active clinical cases exhibit- ing the typical syndrome described were unsuccessful, and more indirect survey methods were adopted to detect atypical symptomatic cases and to reconstruct the chronology and extent of human infections. MATERIALS AND METHODS Fever clinics were held on successive days (21-24 December) in Okwoga, Okpudu, Aidogodo, and Adiga villages (Fig. 1). Persons claiming to be ill 114 YELLOW FEVER IN NIGERIA: I were interviewed briefly and examined, and blood samples were obtained. Because of the lack of diagnostic facilities, no specimens for haematologic examination or urinalysis were taken. Blood samples were also obtained and past histories taken from persons not actively ill. House-to-house surveys were conducted in Okwoga on 25-26 December and Aidogodo on 14-15 March 1971 in order to obtain census data, histories of recent illness, and numbers of deaths. When appropriate, blood samples were taken. These were not surveys of a random sample. The entire village of Okwoga (population 201, in juxtaposed compounds) was surveyed. In Aidogodo, which is much larger (population 1 530) and more widely dispersed, covering more than 26 kM2, the survey included only those compounds within an area of approximately 2.6 km2 extending from the village centre, together with compounds extending 4.8 km south to a village subunit (Ipole). When possible, paired sera were obtained from individuals giving a history of illness within 2 weeks of the initial (December) surveys. Convalescent phase sera were obtained in January or March 1971. Blood was collected by antecubital venepuncture into sterile vacuum tubes and allowed to clot at ambient temperature for 4-12 hours. Sera were decanted aseptically, without centrifugation, and stored in plastic tubes in liquid nitrogen. Upon return to Ibadan, the sera were transferred to mechanical freezers at -20°C for serological examination or -60°C for virus isolation attempts. The techniques used in performing the serological tests are described in a companion paper (Monath et al., 1973a). Criteria for a confirmed diagnosis of yellow fever were either (a) isolation of virus or (b) demonstration of a fourfold rise or fall in haemagglutination- inhibiting (HI) and/or complement-fixing (CE) anti- body in appropriately timed, paired sera. In most cases, only a single serum could be obtained; a presumptive diagnosis of yellow fever infection was then made if the CF titre to yellow fever was 1: 16 or greater, notwithstanding the presence of cross-reac- tions of equal titre with heterologous Group B arbovirus antigens (Monath et al., 1973a). Sera having a yellow fever CF titre of 1: 8 or less, but positive by HI and neutralization (N) tests, were considered to be inconclusive. Negative sera had no demonstrable CF or N antibody in paired or single specimens obtained from convalescent patients. In attempts to isolate the virus, undiluted sera obtained from 108 persons during the acute phase of illness (within I week of onset) were inoculated intracerebrally into litters of suckling mice. The techniques used in recovering and identifying viruses have been described elsewhere (Causey et al., 1961). RESULTS Laboratory confirmation ofyellow fever CF tests on single sera provided the first evidence for the presence of yellow fever virus. Table 1 shows CF test results listed by history of febrile illness with or without jaundice occurring within the preceding 2 months: 71 % of individuals with a history of jaundice and 61 % of those ill without jaundice had CF antibodies to yellow fever, compared with only 12% of persons not ill. These differences are highly significant (P < 0.001). Table 1. Relationship between clinical history and presence of yellow fever complement-fixing anti- bodies (Okwoga and Aidogodo) Clinical illness Age No. of CF-positive CF-negativewithin the pre- Ageas sera cFpstv cFngtvceding 2 months (years) tested (%) (%) fever without 0-9 24 13 (54) 11 (46)jaundice 10-19 20 12 (60) 8 (40) > 20 25 17 (68) 8 (32) Total 69 42 (61) 27 (39) fever and 0-9 13 7 (54) 6 (46)jaundice 10-19 5 4 (80) 1 (20) >20 30 23 (77) 7 (23) Total 48 34 (71) 14 (29) not ill 0-9 53 3 ( 6) 50 (94) 11-19 30 5 (17) 25 (83) > 20 43 7 (16) 36 (84) Total 126 15 (12) 111 (88) Twenty-two cases of yellow fever were confirmed by testing paired sera on the basis of a fourfold (or higher) rise (8 cases) or fall (14 cases) in CF titre. The number of infections (clinical and subclinical) con- firmed or presumed to be due to yellow fever is shown in Table 2. 115 T. P. MONATH ET AL. Table 2. Laboratory diagnosis of yellow fever by serological tests and virus isolation No. of Unsatisfac- Pre- Confirmed diagnosis Village persons tory for Negative b Incon- timSero- tested testing a clusivec diagnosis d So e Virus f anss logical e Okwoga-Okpudu 172 7 114 9 24 16 2 Aidogodo 96 1 37 7 45 6 0 Adiga 48 0 32 3 13 0 0 Total 316 8 183 19 82 22 2 a Anticomplementary. b CF < 1: 8, HI < 1 :10 or > 1 :10, yellow fever neutralization (N) test negative. c CF < 1:8 or 1: 8, HI > 1: 10, N-test positive. d CF> 1: 16 in single convalescent or paired sera (stable titres). e Fourfold rise or fall in HI and/of CF titres in paired sera. f Yellow fever virus isolated from serum using suckling mice. Virus isolations Two virus strains were recovered and identified as yellow fever virus by CF tests. Laboratory strains of yellow fever virus were not in use at the time of isolation attempts; re-isolation of virus from the original material was successful. Both of the viraemic cases were ambulatory and did not appear severely ill. They complained of fever, headache, nausea, and generalized body aches and malaise. The only abnormality noted on physical examination was increased skin temperature. Clinical histories Retrospective clinical histories were obtained be- fore serological results were known. The symptoms noted by individuals later shown to have sustained yellow fever infection are listed in Table 3, which emphasizes the difficulty in distinguishing yellow fever from other illnesses, both with and without jaundice. Severe headache and abnormal bleeding (blood-streaked vomitus, epistaxis, and bleeding gums) were noted more often in yellow-fever cases than in other illnesses. Black vomit was noted in one case with serologic evidence for yellow fever. Age-specific infection rates Age-specific and clinically apparent infection rates for Okwoga-Okpudu, Aidogodo, and Adiga villages are shown in Table 4. Only the results from sera obtained in the voluntary clinic surveys are presented, as sera obtained during house-to-house surveys were selected by positive history of illness. On the basis of CF tests, 24% of persons in Okwoga-Okpudu, 27% in Adiga, and 47% in Aidogodo were infected. The proportion of infections in children aged 0-9 years was lower than that in older individuals. Most infections resulted in a clinical illness, and the proportion of persons infected with yellow fever who also experienced clinical symptoms was the same for children and adults. The youngest case was 5 months of age. There was no difference in overall infection rates or clinically apparent infection rates between males and females. Chronology of the epidemic Surveys were continued until March 1971 in order to reconstruct the epidemic curve (Fig. 2) showing the weekly incidence of presumptive and confirmed clini- cal yellow fever cases. The earliest case occurred in August, no further cases being recorded until the second week in October. The epidemic reached its peak in November and early December, and no cases were found after January 1971. It appeared that the peak weekly incidence of cases in Aidogodo preceded that in Okwoga-Okpudu. Geographic confines of the epidemic The epidemic appeared to be quite focal, with the highest attack rate in Aidogodo village. No suspect cases were hospitalized at Oturkpo, nor were suspect cases seen by the rural health superintendent at Ankpa, Kwara State, 56 km north-west of Okwoga. Serological surveys performed at localities peripheral to Okwoga District are presented in a companion paper (Monath et al., 1973b). Case-fatality ratios Surveys to determine the number of deaths from yellow fever were conducted in Okwoga (December) and Aidogodo (March). 116 117YELLOW FEVER IN NIGERIA: 1 Table 3. Clinical symptoms in persons with confirmed or presumptive yellow fever and with negative sera No. (%) showing the following symptoms Age No. of myalgia, anra(years) persons fever headache abdominal vomiting ablendngl jaundice pain Confirmed or presumptive yellow fever 0-9 20 20 (100) 16 (80) 1 (5) 6 30) 3 (15) 7 (15) 10-19 15 14 (93) 12 (80) 6 (40) 7 (47) 1 (7) 4 (27) > 20 41 35 (86) 33 (80) 15 (37) 13 (32) 3 (7) 23 (56) Total 76 69 (91) 61 (80) 22 (29) 26 (34) 7 (9) 34 (45) Negative sera 0-9 18 17 (95) 11 (61) 3 (17) 11 (61) 1 (6) 6 (33) 10-19 9 8 (89) 5 (56) 1 (11) 3 (33) 0 (0) 1 (11) > 20 15 14 (93) 12 (80) 6 (40) 6 (40) 1 (7) 7 (47) Total 42 39 (93) 28 (67) 10 (24) 20 (48) 2 (5) 14 (33) Table 4. Yellow fever age-specific clinically apparent infection rates in Okwoga District, 1970 a Clinically Proportion Age No. of persons tiv eAggregate N ol CF-positive apparent withVillage (years) tested infection with illness infection apparentlVillage rate (%) rat (%) ~~~~~~~~~~~~~~~~~~~~~~~illness(% Okwoga-Okpudu 0-9 41 6 15 4 10 67 10-19 24 4 17 3 13 77 >20 66 22 33 14 21 64 Total 131 32 24 21 16 67 Aidogodo 0-9 26 10 39 8 31 80 10-19 18 10 56 8 44 80 >20 28 14 50 13 46 93 Total 72 34 47 29 40 85 Adiga 0-9 16 2 13 0 0 0 10-19 11 1 9 1 9 100 >20 21 10 48 5 24 50 Total 48 13 27 6 13 46 a Results of voluntary clinic survey only. T. P. MONATH ET AL. Fig. 2. Serologically confirmed and presumptive clinical yellow fever cases, Okwoga District, 1970-71. A total of 3 deaths (2 adults, 1 child) following illnesses clinically consistent with yellow fever were recorded. All three occurred in November: 2 in Aidogodo, 1 in Okwoga. Five other deaths occurred during the epidemic period (August-January), but they involved 4 infants and a young child for whom clinical details were unavailable. Table 5 shows low and high estimates of mortality, Table 5. Estimated yellow fever mortality, morbidity, and case-fatality ratios, Okwoga District, Benue Plateau State, Nigeria, 1970 Estimated total Clinically Case-fatalityPopulation No. of deaths yellow-fever aprn siae aiVillage sampled/total deaths c infecron total no. ratio % low a high b low high rate (%) d of cases e low high Okwoga-Okpudu 201/1091 1 5 5 25 16 174 2.9 14.4 Aidogodo 317/1530 2 3 10 15 40 612 1.6 2.5 a Yellow-fever compatible deaths recorded in sample population. b Total deaths during epidemic period recorded in sample population. c Recorded no. of deaths/population sampled x total population. d See Table 4. e Clinically apparent infection rate x total population. 118 18AIDOGODO 12 30 6 TOTAL. 24 I) E OKWOGA. OKPUDU : 18 12-0 0 0~~~~~~~~~~~~~12 0 -Z 6- 6 23 6 204 18 15 29 13 271024 7ADIGA. AUG. SEPT. OCT. NOV. DEC. JAN. .Date of onset 23 6 204 8 1 5 29 13 27 10 24 AUG. SEPT. OCT. NOV. DEC-. JAN. Date of onset WW 00 YELLOW FEVER IN NIGERIA: 1 Table 6. Adult Aedes mosquitos reared from larvae collected after artificial flooding of dry potential oviposition sites at Aidogodo (28-29 January, 1971) Nos. reared Water source Site Aedes species male female artificial tree hole 1 A. africanus 1 3 flooding A. argenteoventralis 2 0 2 A. ingrami 1 A. ingrami 1 1 3 A. africanus 1 4 A. ingrami 1 all bamboo stumps (± 30) A. africanus 11 4 A. simpsoni 1 1 A. argenteoventralis 3 4 Aedes sp. (damaged) I natural water bamboo stump no. 1 a A. africanus 1 1 a No adults were reared from the larvae of bamboo stump no. 2. morbidity, and case-fatality ratios computed for Okwoga and Aidogodo. Morbidity was taken as the product of estimated population and clinically appar- ent infection rate. Low and high estimates of the total number of deaths and of case-fatality ratios were derived from the observed number of yellow-fever- compatible deaths (low estimate) and total observed deaths (high estimate). An estimated 786 clinical cases occurred in Okwoga, Okpudu, and Aidogodo villages during the epidemic period, with a case-fatality ratio of 1.6-2.9% (low estimate) to 2.5-14.4% (high esti- mate). Entomological investigations Vector studies were conducted on 26-29 January, at least 2½/2 months after the last reported rainfall. Almost all oviposition sites for Stegomyia spp. (leaf axils, tree holes, bamboo stumps) and for ground- pool mosquitos were dry. Inspections for domestic Aedes aegypti larvae were made in Okwoga and Aidogodo villages. No mosqui- to larvae were seen in 69 water pots and numerous sacrificial cups in 57 compounds. Collections of resting mosquitos in sleeping quar- ters and catches from portable light traps yielded no Aedes and only a few Anopheles and Culex. Human bait captures attempted on 2 successive nights at Aidogodo in heavily wooded areas along streams resulted in the capture of 1 Anopheles sp., 1 Mansonia uniformis, and 1 A. nili. In an attempt to determine what species of Aedes may be present during times more favourable for lar- val development, we resorted to the flooding of tree holes and bamboo stumps. Four tree holes in the thick gallery forest at Aidogodo and many bamboo stumps near village compounds were flooded with stream water on 28 January and sampled the following day. In the process 2 bamboo stumps containing water and mosquito larvae were discovered. All other sites were dry. Results of this investigation are given in Table 6. Control measures In March 1971 the Benue Plateau State Ministry of Health carried out 24 573 vaccinations in Okwoga and a neighbouring district (Utonkon). Between June and August 1971 the Federal Ministry of Health performed 1 496 674 vaccinations in the 5 administra- tive divisions of Benue Plateau State lying south of the River Benue (Oturkpo, Gboko, Makurdi, Wukari, and Katsina Ala Divisions). A total of 1 521 247 yellow fever vaccinations were thus given in an area inhabited by 2 027 420 people (1963 census), the estimated coverage being 75 %. All vaccinations were performed by mobile teams using jet-gun vac- cinators and 17D vaccine. DISCUSSION Epidemics of yellow fever studied in West Africa prior to 1969, like those in tropical America, have been characterized by an " urban " transmission cycle involving Aedes aegypti mosquitos. The first rural outbreak studied in West Africa occurred on the Jos Plateau, Nigeria, in 1969 (Carey et al., 1972). The second such epidemic is described in this paper. It is likely that other epidemics associated with transmis- sion by wild-breeding Stegomyia mosquitos have 119 T. P. MONATH ET AL. occurred in West Africa but have passed undetected; the epidemic of jaundice on the Jos Plateau in 1951 may be an example (Carey et al., 1972). With the exception of a few fatal cases, the clinical illness associated with the Okwoga epidemic appears to have been relatively mild. Such anicteric forms of yellow fever, clinically not differentiable, have been noted in some other epidemics (Kerr & Patino Camargo, 1933; Causey & Maroja, 1959; Courtois et al., 1960; Satti & Haseeb, 1966). In Okwoga, the presence of jaundice was noted in 34 of 76 presumed or confirmed cases. In contrast to other epidemics, including that on the Jos Plateau in 1969 (Jones & Wilson, 1972), jaundice was apparently not asso- ciated with severity of infection and poor prognosis. In fact, fatal cases did not show signs of hepatic involvement, but died after a fulminating course of 2-4 days. This fatal form of the disease was noted in one area of Ethiopia (Bolosso) during the epidemic of 1960-62 (Serie et al., 1968). As in the rural outbreaks in Ethiopia (Serie et al., 1968) and on the Jos Plateau (Carey et al., 1972), adults in Okwoga were more often infected than children. This was possibly a reflection of the greater exposure of adult field workers to wild-breeding infected mosquitos. Children sustaining yellow fever infection gave a history of illness as frequently as adults. In children, however, jaundice appeared to be a less commonly recognized symptom. Previous authors have sug- gested that yellow fever infection may be milder or atypical in children (Chambon et al., 1967). Since a history of illness due to yellow fever could not be distinguished from one due to other diseases, the number of CF-positive individuals with a history of illness may have included persons with very mild or subclinical yellow fever infections who sustained other clinical illness during the epidemic. This might be especially true for the 0-9-year age group, and would lead to an overestimation of the clinically apparent infection rate. The aggregate morbidity in the Okwoga-Okpudu- Aidogodo focus (population 2 621) has been esti- mated as 786 cases on the basis of clinically apparent infection rates (Table 5). The case-fatality ratio was probably 1.6-2.9%, reflecting the mild course of yellow fever infection in this epidemic. Large populations of wild-breeding Stegomyia mosquitos were probably present during the latter part of October and early November (Kerr, 1933). After the cessation of the rains, the survival of infected adults may be expected for several weeks, especially since relative humidity did not markedly decline in November (Fig. 3). The epidemic reached its peak in mid-November, approximately 2 weeks after the last rainfall, and declined rapidly in Decem- ber, at a time when 24-47 % of the human population had CF and 30-60% had yellow fever N antibodies. Both the decline in wild-breeding mosquito popula- tions and the increasing proportion of immune individuals probably accounted for termination of the epidemic. Fig. 3. Rainfall and humidity during 1970 at Makurdi, 113 km NE of Okwoga District. Entomological investigations showed that domes- tic A. aegypti were not present. Other Aedes popula- tions surviving the dry season until the time of the investigations were assumed to be very low or nonexistent. From the results of artificial flooding experiments, 2 potential yellow fever vectors would be expected in appreciable numbers during the rainy season: A. africanus and A. simpsoni. Other sylvan vectors, such as A. aegypti and A. luteocephalus, should also be present during the rains. A. simpsoni is not considered to be anthropophilic in Nigeria (Lee& Moore, 1972). The origin of the epidemic is not known. Two possibilities exist: (1) the Okwoga outbreak, like that on the Jos Plateau, resulted from the introduction of yellow fever virus from a distant source at a time when ecological and entomological factors were favourable for interhuman transmission in an immu- nologically susceptible population or (2) yellow fever is endemic (or became endemic after 1969) in or near Okwoga District. The scarcity of non-human pri- mates and the entomological data tend to support the 120 YELLOW FEVER IN NIGERIA: 1 121 first hypothesis, but the source of virus introduction is not known. Enzootic foci of yellow fever activity, similar to those described in East Africa (Haddow, 1969), have not yet been clearly demonstrated in Nigeria. Moreover, the possibility of stable or wan- dering yellow fever endemicity dependent upon a human-human cycle of transmission (similar to dengue) cannot be easily dismissed. ACKNOWLEDGEMENTS We thank Dr S. L. Adesuyi, Chief Medical Adviser to the Federal Military Government, for permission to publish this paper. We also thank Dr Matthew Idoko, University College Hospital, Ibadan, who first reported the occurrence of an outbreak in his home village, Aidogodo; Dr Robert F. Addy, Chief Medical Officer, Benue Plateau State Ministry of Health and Social Welfare, who provided health personnel during the surveys; and Dr Donald E. Carey, Professor Akinyele Fabiyi, and Dr Graham E. Kemp, who gave valuable advice and critically reviewed the manuscript. RESUME L'EPIDEMIE DE FIEVRE JAUNE DE 1970 DANS LE DISTRICT D'OKWOGA, tTAT DU PLATEAU DE BENUE, NIGERIA: 1. OBSERVATIONS EPIDEMIOLOGIQUES La fin de 1969 a ete marquee par 1'eclosion d'une forte epidemie de fievre jaune, la premiere observee au Nigeria depuis 17 ans, sur le plateau de Jos. Elle s'est achev6e brusquement en decembre de la meme annee. Un an plus tard, a la fin de 1970, une seconde pouss6e de la maladie s'est manifestee a 190 km plus au sud, dans le district d'Okwoga. On y a enregistre au total 106 cas de fievre jaune pr6sumes ou confirmes par les epreuves serologiques et les isolements de virus. Le village d'Aidogo a ete le plus atteint: 39% des enfants de 0 a 9 ans, 56% des sujets de 10 a 19 ans et 50% des adultes de plus de 20 ans ont et6 infectes. Parmi les habitants du village porteurs d'anticorps antiamarils, 85% avaient et6 malades pendant l'epidemie, les enfants ayant presente des sympt6mes cliniques aussi souvent que les adultes. On estime que parmi les 2621 habitants de trois villages etudies, il y a eu au total 786 cas, avec un faible taux de letalite (1,6-2,9%) refletant la relative benignite de l'6pidemie. Gette demiere ne s'est pas 6tendue geogra- phiquement. L'enquete entomologique menee durant la saison seche a conduit a mettre en cause des moustiques selvatiques du sous-genre Stegomyia, notamment Aedes africanus et A. simpsoni. Une campagne de masse de vaccination antiamarile a et6 mise en ceuvre en 1971 dans l'Etat du plateau de Benue, au sud du fleuve Benue. REFERENCES Carey, D. E. et al. (1972) Bull. Wld Hlth Org., 46, 645-651 Causey, 0. R. et al. (1961) Amer. J. trop. Med. Hyg., 10, 227-249 Causey, 0. R. & Maroja, 0. (1959) Amer. J. trop. Med. Hyg. 8, 368 Chambon, L. et al. (1967) Bull. Wid Hlth Org., 36, 113-150 Courtois, Gh. et al. (1960) Ann. Soc. belge Med. trop., 40, 29-60 Haddow, A. J. (1969) Proc. roy. Soc. Edinb. B, 70, 191-227 Jones, E. M. M. & Wilson, D.C. (1972) Bull. Wld Hith Org., 46, 653-657 Kerr, J. A. (1933) Bull. ent. Res. 24, 493-510 Kerr, J. A. & Patino Camargo, L. (1933) Rev. Hig. Bogotd 2, 63-83 Lee, V. H. (1972) Bull. Wld. Hlth. Org., 46, 641-644 Lee, V. H. & Moore, D.L. (1972) Bull. Wid. Hlth. Org., 46, 669-673 Monath, T. P. et al. (1973a). Bull. Wld Hlth Org., 49, in press Monath, T. P. et al. (1973b) Bull. Wld Hlth Org., 49, 123 Satti, M. H. & Haseeb, M. A. (1966) J. trop. Med. Hyg. 69, 36-44 S&ri6, C. et al. (1968) Bull. Wid Hith Org. 38, 835-841