The resurgence of lymphatic filariasis in the Nile delta.

The resurgence of lymphatic filariasis in the Nile delta M. Harb,1 R. Faris,2 A.M. Gad,3 O.N. Hafez,4 R. Ramzy,5 & A.A. Buck6 A study of 325 000 residents of 314 villages in six governorates of the Nile delta area of Egypt revealed that the prevalence of lymphatic filariasis increased from <1% in 1965 to >20% in 1991, especially in the governorates of Qalyubiya, Monufiya, Dakhaliya, and Giza. The distribution of the communities with endemic filariasis is focal. Clusters of villages with high prevalences are surrounded by others in which the disease is absent, although their environmental, social, and agricultural features appear similar. The article analyses why the significant decline in filariasis between 1945 and 1965 in Egypt has been followed by a resurgence of the disease. Introduction Periodic lymphatic filariasis caused by Wuchereria bancrofti has been endemic in Egypt since Pharaonic times. However, before the discovery of the W. ban- crofti life-cycle all evidence of the occurrence of the disease was based on sporadic observations of its clinical manifestations, i.e., elephantiasis of the limbs and genitals (1). A comprehensive account of the history of filariasis in Egypt up to 1978 has appeared previously (2). Although clinical, pathological, and parasitolo- gical research on filariasis was carried out in Egypt during the early part of this century, significant epi- demiological and entomological studies were not carried out before the 1930s (3, 4). These investi- gations led to the identification of Culex pipiens as the main vector of the disease and also revealed that the distribution of filariasis in the country was highly focal (1, 5). 1 Director, Division of Malaria and Filariasis Control, Ministry of Health, Cairo, Egypt. 2 Professor and Chairman, Department of Community Medicine, Faculty of Medicine, Ain Shams University, Ramses St., Abbas- sia, Cairo, Egypt. Requests for reprints should be sent to this author. 3 Professor, Team Leader, Filariasis Module, Regional Centre for Epidemiology and Control of Vector-borne Diseases, Ain Shams University Cairo, Egypt. 4 Instructor, Department of Community Medicine, Ains Shams University, Cairo, Egypt. 5 Senior Scientist, Immunology and Biochemistry Section, Regional Centre for Epidemiology and Control of Vector-borne Diseases, Ain Shams University, Cairo, Egypt. 6 Professor, Department of Immunology and Infectious Diseases, Johns Hopkins University, Baltimore, MD, USA. Visit- ing Professor of Community Medicine, Ain Shams University, Cairo, Egypt, and resident Scientist, NIH/NIAID, Bethesda, MD, USA. Reprint No. 5354 In Egypt, surveys of filariasis using combined measurements of the microfilarial (mf) rate and of the frequency of clinical manifestations were conducted in many non-randomly selected communi- ties. The results provided a sketchy panorama of the distribution of the disease. Foci with high levels of endemicity were scattered in the densely populated area of the Nile delta, and mf rates of 20% or more occurred, especially in the govemorates of Qalyubiya, Sharkiya, Dakhaliya, and Damietta. In contrast, filarial infections appeared to be absent in most parts of middle and upper Egypt, with the exception of a small hypoendemic area in the govemorate of Asyilt. By 1936, filariasis was recognized as a major public health problem in Egypt and measures for its control were taken by the Ministry of Health. Between 1950 and 1965, a large-scale filariasis control programme was carried out in the endemic areas. The control methods comprised mass treat- ment with diethylcarbamazine citrate, vector control by residual house-spraying and larviciding, and source reduction through the elimination of artificial C. pipiens breeding sites. General public health measures with a broad impact on disease prevention were also introduced on a national scale. These measures included the creation of safe water supply systems, as well as better sanitation, waste disposal, and agricultural irrigation practices. The improvements in the health status resulting from these changes were further strengthened by a nationwide health education programme. The effects of widespread application of insecticides against agricultural pests in the rapidly expanding areas of land that were cultivated for cash crops also helped to reduce both the size of the vector population and the intensity of filariasis transmission. The impact of all these activities on filariasis was mirrored by a steady decline of the disease in all Bulletin of the World Health Organization, 71 (1): 49-54 (1993) © World Health Organization 1993 49 M. Harb et al. previously identified endemic areas of the country (5-8). Between 1955 and 1964, the decrease in the mf rates and in the prevalence of clinical filariasis was investigated in a series of epidemiological sur- veys in most of the affected areas. A sample consist- ing of more than 500 000 persons was involved, from whom blood specimens were taken and who underwent a brief physical examination (9, 10). Although the decrease in the prevalence was impres- sive in all parts of the country, the residual preva- lence was never reduced to zero in many of the old endemic foci in the Nile delta. The public health importance of these communi- ties with low endemicity levels of filariasis appeared insignificant to the authorities and the medical pro- fession. This impression was enhanced by health sta- tistics which used the small numbers of reported cases of filariasis in individual communities as numerators to calculate the prevalence for popula- tions of whole districts and govemorates. Because of the focal distribution of the disease, the filariasis rates reported for the larger administrative divisions were too low and not representative of the real situa- tion. Eventually, use of this approach led to the erro- neous conclusion that filariasis was a rapidly disap- pearing disease in Egypt. This opinion was also expressed by the WHO Expert Committee on Filaria- sis in 1984 when it stated that the microfilaraemia rate in Egypt was less than 1% in the population of the endemic areas (11). Since 1974, a number of small spot surveys for microfilaraemia and clinical signs of filariasis have been carried out in several parts of the Nile delta. The results revealed that the downward trend of the disease had stopped. Moreover, in some of the vil- lages of Qalyubiya govemorate there were clear indi- cations that both the prevalence and the intensity of microfilaraemia had increased (12-15). These obser- vations prompted the Ministry of Health to reorgan- ize and intensify its active surveillance of filariasis. The results of these surveys and of some intensive epidemiological studies in selected villages are reported in this article. Materials and methods Between 1985 and 1991, Ministry of Health sur- veillance teams took blood samples from 324 552 individuals who lived in 314 villages and towns in the Nile delta area. From each person a 20-jl finger- prick sample of blood was taken during the peak of periodic microfilaraemia, preferably between 22 h 00 and 02 h 00. The blood samples were stained with Giemsa and examined for W. bancrofti microfilariae; the results were recorded as positive or negative. Physical examinations for clinical manifestations of filariasis were not carried out by the survey teams. The population samples ranged from 1550 to 27 837, and included only persons above 1 year of age who resided in the administrative districts shown in Table 1. Although attempts were made to obtain representative samples from all residents in each community, the necessity of visiting the families late at night reduced the compliance rate and introduced an unknown degree of bias. Follow-up examinations in some of the villages by teams from Ain Shams University 1-2 years after the Ministry of Health visits yielded mf rates that were similar to those found previously. Table 1 summarizes the demogra- phic data for the population in the districts that were covered by sample surveys up to July 1991. The demographic data were obtained from the official report of the 1986 census (16). Estimates of the prevalence of filariasis based on mf rates in fingerprick samples of blood are insensi- tive, since this approach does not permit detection of patent infections where the mf density is low, and the results are negative also for the majority of clini- cal cases. Efforts were therefore made to improve the prevalence estimates by applying an adjustment factor to the crude mf rates determined in the Ministry of Health surveys. This factor was calcula- ted using the results of a study of 427 residents of the village of Kafr Tahoria, where the following diagnostic tests were compared by independent groups of investigators (19): physical examination for the clinical lesions caused by filariasis; standard microscopic examination for microfilariae in 50-gl samples of stained, thick blood smears; Nuclepore ® filter concentration using 1-ml venous blood samples to detect microfilariae; and a monoclonal antibody assay that had been thoroughly evaluated in the labo- ratory and under field conditions in India and Egypt (17, 18). Fig. 1 summarizes the estimated frequency of filariasis obtained using each of these methods. A detailed description of the evaluation has been pub- lished previously (19). The antigen test had the highest relative sensitivity, covering the entire spec- trum of filariasis. However, both the parasitological and the antigen tests failed to detect the majority of clinical cases. To date, there have been no indica- tions of false-positive reactions in the continuing validation of the filarial antigen test using blood samples from persons with patent Schistosoma haematobium, S. mansoni, Ascaris lumbricoides, Trichuris spp., Oxyuris spp., hookworm, or intestinal protozoal infections. The high specificity of the monoclonal antigen assay is further supported by reports of significant reductions in antigen titres following effective antifilarial treatment (19). The systematic validation of this assay is continuing. WHO Bulletin OMS. Vol 71 199350 Lymphatic filarlasis In the Nile delta Table 1: Census populations and sample sizes of the populations included in the filarlasis surveillance programme Population Governorate Census Sample and district totala size Qalyubiya Shibin 192 751 21 213 (11.0) Khanka 76 281 14 486 (19.0) Qalyub 160 992 15 076 (9.4) Kaireiha 148 686 7702 (5.2) Benha 217 161 16 727 (7.7) Tukh 246 944 27 288 (11.1) Shubeira 300 416 5898 (2.0) Kafr Shukr 76 282 1550 (2.0) Subtotal 1 419 514 109 940 (7.7) Dakhaliya Mit Gahmr 303 981 11 575 (3.8) Monofiya Qisna 204 600 14 221 (7.7) Ashmun 319 673 13 866 (4.3) Bagur 167 505 15 243 (9.1) Monofiya 213 137 6156 (2.9) Shuhada 130 956 10 725 (8.2) Subtotal 1 035 871 60 211 (5.8) Gharbiya El Santa 100 000 5587 (5.6) Zifta 100 000 18 437 (18.4) Subtotal 200 000 24 024 (12.0) Giza Badrashein 163 824 13 469 (8.2) Ayat 170 979 10 505 (6.1) Saff 130 729 4050 (3.1) Giza 79 227 23 416 (29.6) Imbaba 595 482 27 837 (4.7) Subtotal 1 140 241 79 277 (7.0) Sharkiya Zkalik 58 696 4053 (6.9) Abu Hammad 214 802 11 157 (5.2) Bilbeis 100 887 10 615 (10.5) Minya El Kam 86 182 13 700 (15.9) Subtotal 460 567 39 525 (8.6) Total 4 560 174 324 552 (7.1) a Census totals are for 1986; see ref. 16. b Figures in parentheses are percentages. Determination of the mf rate has been the basic method used in the surveillance programme in Egypt and has remained the most widely used screening Fig. 1. Estimated frequency of filariasis in the village of Kafr Tahoria, according to the results of various diag- nostic tests. Figures in parentheses are the number of positive tests for filariasis in a sample of 421 persons. 50 (189) 40 (186) 0~~~~~~~10>30 (10 o (93) - 20- 0.3 1.0 1.3 17 20 Standardized test ratio (mf rate 1.0) Clinical ~~~~~~~~~~mffilterCinical S mf fingerprick blood (1ml) * Antigenaemia Antigenaemia + clinical test elsewhere (20). Therefore, in calculating the adjustment factor for estimating the prevalence of filariasis we assigned a relative test ratio of 1.0 to the crude mf rate (Fig. 1). By comparison, the corres- ponding value of the relative test ratio for the combined tests was 2.0. Accordingly, the crude mf rates for the Ministry of Health surveys were adjus- ted by this factor to obtain a more realistic estimate of the prevalence of filariasis in the Egyptian com- munities studied. The adjusted figures obtained are conservative estimates of the prevalence, because in the Ministry of Health surveys the mf rates were based on 20-gl fingerprick blood samples instead of the 50-,ul samples used in the evaluation study of the four diagnostic tests. To obtain a comprehensive overview of all the results of the surveillance programme in the southem Nile delta, we determined the exact latitude and longitude of each village from detailed maps of the area. The estimated prevalence ratios were then coded into four categories of endemicity. Using a digitizer and Generic Cadd 5 ® software,a we plotted these values on a computerized scale map of the area, using the exact geographical coordinates of each village. Results Of the 324 552 persons examined by the Ministry of Health teams, 8191 (2.5%) had W. bancrofti micro- a Available from: Generic Software Inc., Bothwell, WA 98011, USA. WHO Bulletin OMS. Vol 71 1993 51 M. Harb et al. filariae in the 20-,ul fingerprick blood samples. The crude mf rates for the 314 villages examined by the teams ranged from 0% to 23%. The frequency distri- butions of these rates are shown separately for each of the six govemorates in Fig. 2. A comprehensive map with the adjusted esti- mates for the prevalence of filariasis in each village is depicted in Fig. 3. The observed distribution of filariasis in the southem Nile delta is prominently focal, with clusters of high endemicity in the gover- Fig. 2. Frequency distribution of crude microfilarial (mf) rates in 314 villages in six governorates in the Nile delta. Figures in parentheses are the number of villages surveyed in each governorate. .I ........1~~~~~~ I1 I-- ------------'-------------i-i- . .. .---..-... .--e;- -- I-. *........ 0 5 10 mf rate (%) 15 M Qalyubia (76) M Gharbiya (35) E3 Sharkiya (28) MDakhaliya (22) E: Monufiya (68) = Giza (85) Fig. 3. Map showing the geographical distribution and prevalence of filariasis caused by Wuchereria bancrofti in the southern Nile delta. norate of Qalyubiya. Individual foci of high preva- lence exist also in the govemorates of Giza, Monufiya and Dakhaliya. The clustering of filari- asis in Egypt has previously been noted by Khalil et al. (5), Baz (4), and Shawaby et al. (9). The reasons for these large differences in prevalence between apparently similar villages that are located only a few miles apart are not clear. A tendency for familial aggregation of filariasis cases has also been observed within individual communities. The main vector of filariasis in Egypt in the Nile delta, C. pipiens, has an uneven distribution in natural and artificial breeding sites in endemic and non-endemic villages. Recent findings have identified the relative importance of indoor breeding of the vector as a significant risk factor in the transmis- sion of W. bancrofti (A.M. Gad, unpublished find- ings, 1992). Using the value for the prevalence of filariasis together with the latest census data for the districts in which sample surveys were performed, we estimated the total number of cases expected and the popula- tion at risk. The number of filariasis cases was esti- mated by applying the prevalence ratios for the indi- vidual districts to their corresponding populations. For this purpose, the population data were trimmed to include only the rural segments of the district and only those aged >1 year. The numbers of the expect- ed cases in the districts were then combined for each governorate; the results are shown in Table 2. Based on the available data, the total size of the population at risk was estimated to be 3.65 million. These calculations are confined to the current- ly known endemic areas and provide a conserva- tive assessment of the magnitude of the recurrent filariasis problem in the Nile delta. In many villages where polyparasitism has prevailed, filariasis is now the most prevalent parasitic infection, surpassing even schistosomiasis. Table 2: Estimated number of filariasis cases in six governorates in the southern Nile deltaa No. of cases estimatedb Governorate Qalyubiya Dakhaliya Monufiya Gharbiya Giza Sharkiya Total 300 East Longitude 31° East 1 986c 87 100 28 500 24 700 3300 19 200 49 900 212 700 1991 99 300 32 500 28 100 3700 21 900 56 900 242 400 a Includes only persons >1 year of age from districts with recent surveillance records. I Based on a crude annual growth rate of 2.65%. c Based on the 1986 national census report. WHO Bulletin OMS. Vol 71 199352 Lymphatic filariasis in the Nile delta Discussion As late as 1990, when it was already known that there were foci of filariasis with high levels of endemicity in Egypt (1, 9, 12, 16), the general belief still prevailed that filariasis was a disease of little public health importance in the country. This opinion was held also by many doctors who no longer in- cluded filariasis in their differential diagnostic con- siderations (A. Khjer, personal communication, 1991). Of the many factors that have been associated with the decline and subsequent resurgence of lymphatic filariasis in Egypt, a few have been identified as major determinants in the epidemiology of the disease. The environmental improvements that took place between 1936 and 1965 in Egypt, combined with specific measures to control filariasis over this period, led to sizeable reductions in the prevalence of the disease (9). However, even in 1965 there were districts in the eastem Nile delta where the mf rates were >1% (9). Over the past 25 years significant environmental and demographic changes have taken place that have had a profound effect on the epidemiology of filaria- sis in Egypt. There have been substantial increases in C. pipiens breeding sites in most rural and urban areas, which have considerably increased the prob- ability of vector contact in the resident populations. Many of the environmental changes that have contri- buted to the explosion of the vector population are directly related to the creation of the Aswan High Dam. There has been an extension and intensification of irrigation for increased agricultural production, and this has raised the groundwater table resulting in poor and delayed drainage of ditches, heavily pol- luted stagnant pools, and wastewater puddles where the vector mosquitos breed. At the same time, the extensive agricultural use of pesticides has intro- duced resistance to a broad range of vectors, includ- ing C. pipiens (21, 22). In the affected areas there have been large increases in the populations of rural and urban com- munities caused by a natural growth rate of 2.65% per annum and migration. The impact of the environmental and demogra- phic changes on the quality of sanitation has been considerable, especially in rural communities and in the rapidly growing periurban fringes of metropoli- tan areas. The ensuing increase in water consump- tion, accompanied by inadequate disposal of human waste and wastewater have enormously increased the vector breeding sites. The problem is further aggra- vated by the large number of residents in rural and periurban locations who commute between their homes and places of work. This has led to a steady influx of infected persons from endemic areas close to Cairo into the metropolitan area. The recent report of C. pipiens caught carrying W. bancrofti L3 larvae in the outskirts of Cairo (A. Gad, unpublished obser- vations, 1989-90) suggest that active transmission of urban filariasis has probably already occurred. Acknowledgements The study was supported by the project Epidemiology and Control of Arthropod-borne Diseases in Egypt (USAID/ NIAID regional project, No. NOI-AI-22667) and by the proj- ect Urban Filariasis in Cairo (USA-Egypt Cooperative Health Plan, project No. E-03-N). Resume R6surgence de la filariose lymphatique dans le delta du Nil Le d6clin spectaculaire de la filariose de Bancroft en Egypte entre 1935 et 1965 a ete le r6sultat d'un ensemble de mesures g6nerales de sante publique (adduction d'eau, assainissement, 6va- cuation des d6chets et des eaux usees, irrigation) et d'un programme de lutte contre la maladie combinant les traitements de masse et la lutte antivectorielle. A partir de 1965, la surveillance de la filariose est devenue fragmentaire et inefficace. Les resultats d'enquetes ponctuelles limit6es, effectuees dans des villages apres 1975, ont montr6 que la tendance a la regression de la fila- riose end6mique ne s'observait pas partout en Egypte et que la pr6valence de l'infestation 6tait en augmentation dans certains foyers anciens. Afin d'evaluer l'ampleur du probleme, le Ministere egyptien de la Sante a remis en vigueur son pro- gramme de surveillance de la filariose dans le delta du Nil. En 1991, 314 villages de 6 gouverno- rats (population totale: 4,5 millions d'habitants) avaient ete couverts et des prelevements de sang nocturnes avaient ete faits sur 350 000 personnes pour rechercher la pr6sence de microfilaires de Wuchereria bancrofti par examen au microscope. Les r6sultats ont confirm6 les observations ant6rieures qui faisaient 6tat d'une augmentation de la pr6valence de la filariose dans le sud du delta du Nil, ou le taux brut de microfilar6mie (mf) atteignait ou d6passait 20% dans certains vil- lages. La distribution geographique des foyers d'end6mie 6tait in6gale, les gouvernorats de Qalyubiya et Dakhaliya 6tant les plus touch6s. Etant donne la relative insensibilit6 du test de depistage utilise dans ces enquetes, a savoir l'examen d'un prelevement nocturne de 20 gI de sang, un facteur de correction a e applique au WHO Bulletin OMS. Vol 71 1993 53 M. Harb et al. taux brut pour obtenir une estimation plus r6aliste de la prevalence de la filariose. Le facteur de cor- rection a 6t6 determine lors d'une etude pr6alable effectu6e en aveugle dans un village d'end6mie, 6tude au cours de laquelle 4 tests independants portant sur les manifestations cliniques, la microfi- laremie et le dosage des antigenes circulants ont ete pratiqu6s. L'analyse des causes possibles de la r6sur- gence de la maladie a montr6 que la forte aug- mentation du nombre et de l'6tendue des lieux de reproduction de Culex pipiens, principal vecteur de W. bancrofti, 6tait le facteur le plus important dans l'epid6miologie de la filariose. Les recherches actuellement en cours portent sur les d6terminants comportementaux, parasitologiques, environnementaux et entomologiques qui inter- viennent dans la dynamique de la transmission et qui r6gissent les facteurs de risque d'infestation et d'apparition de la maladie. References 1. Khalil, M. The role of Arabic countries and Egypt in the discovery of elephantiasis and filariasis. Journal of the Egyptian Medical Association, 22: 86-106 (1939). 2. Southgate, B.A. Bancroftian filariasis in Egypt. Tropical diseases bulletin, 76: 1045-1068 (1979). 3. Khalil, M. et al. On the transmission of filariasis bancrofti in Egypt. Journal of the Egyptian Medical Association, 15: 317-322 (1932). 4. Baz, I.l. Distribution of filariasis in Egypt. Journal of the Egyptian Medical Association, 29: 280-287 (1946). 5. Khalil, M. Filariasis and elephantiasis in Rosetta and the means of their effective control. Journal of the Egyptian Medical Association, 19: 701-716 (1936). 6. Mahdi, A.H. et al. Bancroftian filariasis in the United Arab Republic; dynamics of complications under natural versus restrained disease patterns; field trial areas Qalyubiya Governorate, 1957-1967. Journal of the Egyptian Public Health Association, 43: 501-510 (1968). 7. Mahdi, A.H. et al. Bancroftian filariasis in the United Arab Republic; natural versus differentially chemical- ly restrained population dynamics of Culicini spp.; field trial areas, Qalyubiya Governorate 1963-1968. Journal of the Egyptian Public Health Association, 43: 441-451 (1968). 8. Halawani, A. et al. A preliminary report on the treatment of ambulant cases of bancroftian filariasis with Hetrazan in Egypt. Journal of the Egyptian Medical Association, 32: 395-403 (1949). 9. Shawaby, A.A. et al. Incidence of filariasis in Egypt. Journal of the Egyptian Public Health Asso- ciation, 40: 267-282 (1965). 10. Shawaby, A.A. et al. Bancroftian filariasis in United Arab Republic. Assessment of control measures 1963-66. Journal of the Egyptian Public Health Association, 43: 79-99 (1968). 11. WHO Technical Report Series No. 702, 1984 (Lymphatic filariasis: fourth report of the WHO Expert Committee on Filariasis). 12. Preliminary report on filariasis survey at Manshat El- Bakary, Giza. Ain Shams medical journal, 28: 319-323 (1977). 13. Hassan, Z.A. & Rifaat, M.A. Comparative efficacy of the stained blood-film and counting chamber technique for the diagnosis of Wuchereria bancrofti. Journal of the Egyptian Public Health Organization, 51: 223-228 (1976). 14. Rifaat, M.A. et al. Studies on filariasis in Egypt. Entomological and epidemiological studies in an endemic village. Journal of the Egyptian Society of Parasitology, 8: 95-99 (1978). 15. Feinsod, F.M. et al. Clinical manifestations of Wuchereria bancrofti filariasis in Egypt in an endem- ic village in the Nile delta. Annales de la Soci6t6 belge de M6decine tropicale, 67: 259-265 (1987). 16. 1986 Census report. Cairo, Central Agency for Mobilization and Statistics, 1989. 17. Ramzy, M.R. et al. Evaluation of a monoclonal- antibody-based antigen assay for diagnosis of Wuchereria bancrofti in Egypt. American journal of tropical medicine and hygiene, 44: 691-695 (1991). 18. Faris, R. et al. Community diagnosis of bancroftian filariasis. Transactions of the Royal Society of Tropi- cal Medicine and Hygiene (in press). 19. Weil, G.J. Changes in circulating parasite antigen levels after treatment with diethylcarbamazine and ivermectin. Journal of infectious diseases, 164: 814-816 (1991). 20. Control of lymphatic filariasis: a manual for health personnel. Geneva, World Health Organization, 1987. 21. Gad, A.M. & Kamel, O.M. Susceptibility of the fila- ria vector mosquito Culex pipiens to insecticides in Egypt. Journal of the Egyptian Public Health Association, 43: 1-12 (1968). 22. WHO Technical Report Series No. 821, 1992 (Lym- phatic filariasis: the disease and its control. Fifth report of the WHO Expert Committee on Filariasis). 54 WHO Bulletin OMS. Vol 71 1993