Vector mosquitoes of filariasis in Japan.

1 Species of vector mosquitoes

The main vector of Brugia malayi in the small endemic focus on Hachijo-koshima Island was determined to be Ochlerotatus togoi (as Aedes togoi) [1]. As to Wuchereria bancrofti, many species of mosquitoes were experimentally examined for susceptibility in the laboratory as by Mochizuki in 1911 and 1913 [2, 3], Yamada in 1927 [4], Fujisaki in 1958 and 1959 [5, 6], and Omori in 1962 [7]. The results indicated that among 26 mosquito species tested Culex pipiens pallens, Cx. pipiens molestus, Cx. Pipiens quinquefasciatus, Cx. vagans, Cx. whitmorei and Ochlerotatus togoi (as Aedes togoi) were highly susceptible (the microfilaria could develop to the infective stage in the body of mosquitoes in high rates), while Cx. bitaeniorhynchus, Cx. pseudovishnui (as Cx. vishnui), Cx. sinensis, Cx. tritaeniorhynchus, Anopheles sinensis and Armigeres subalbatus had extremely low susceptibility [8].

Of the highly susceptible six mosquitoes, Cx. pipiens molestrus, Cx. vagans and Cx. whitmorei were excluded from important vectors of bancroftian filariasis in Japan, because they were rather rare at least in filariasis endemic areas and no natural infections were observed [9]. Oc. togoi, which breeds almost exclusively in rock pools near the sea, was common in some coastal areas, for example, in Goto Islands of Nagasaki prefecture, and the natural infection with the parasite was not rarely observed. However, because of rather poor anthropophily in blood feeding habit [10] this mosquito was regarded only as the secondary vector even in particular coastal areas where the density was quite high. The remaining two mosquitoes, Cx. pipiens pallens and Cx. pipiens quinquefasciatus, were the most important vectors transmitting bancroftian filariasis in Japan. They are taxonomically different only in subspecific status, and the former is distributed in Kyushu Island and further north and the latter in Amami-Oshima Island and further south. The both mosquitoes with similar biological characteristics were always very common in dwelling houses in areas endemic for bancroftian filariasis and very anthropophilic in blood feeding, and the natural infections with W. bancrofti were frequently found. The main breeding places were drains receiving domestic waste water and manure pits.

2 Natural decrease of filariasis endemicity before the start of national drug treatment

Filariasis was obviously a serious health problem in many parts of Japan since old times [11]. In blood surveys of army recruits from all over Japan in 1912 reported by Army Medical Surgeon, the overall microfilaria positive rate was 1.86%, or 2,090 positive cases out of 112,353 recruits tested [12]. It is noteworthy that infected persons were found even in Aomori, the northernmost prefecture of Honshu Island, where vector mosquitoes could be active only in very short summer. The prevalence was naturally higher in more southern parts of Japan with higher vector density.

It is considered that the microfilaria positive rate decreased year by year after that time and filariasis in many of its endemic foci particularly in northern and central Japan disappeared in about a half century, though there still existed a number of small isolated foci in rural areas. This decrease in the endemicity of filariasis is apparently caused by the decrease in man/vector contacts by changes in living style including the popularization of mosquito nets and by reduction of vector density itself due to improvements of sanitary environments.

On the other hand, the prevalence of filariasis in southern Japan including Kyushu Island and the Ryukyu Archipelagoes was still one of the most serious public health problems for the inhabitants even in early 1960s. It should be noted that the prevalence was generally much higher in rural areas than in urban areas, probably because of the delay in the change of living style and of the reduction of vector density in the former areas.

3 Field experiments toward the control of vector mosquitoes

Several groups of workers took up the area control of filariasis in Japan on an experimental basis. One of the most systematic trials carried out was that by Sasa and others [13], aiming to eradicate malayan filariasis from Hachijo-Koshima Island. Extensive use was made of the anti-filarial drug (diethylcarbamazine) with simultaneous applications of anti-mosquito measures including residual spraying of dieldrin and DDT.

The Filariasis Control Committee with workers from the universities of Tokyo, Nagasaki and Kagoshima was organized in 1958 with financial aid from the Ministry of Education. The members of the Committee conducted field studies on the control of filariasis at some 10 different localities, to find out the most effective and economic measures and to establish a standard method to be applied to all endemic areas [13-15]. Anti-mosquito measures were included in the principle adopted by the Committee.

The effectiveness of insecticides available at that time was evaluated in the control of vector mosquitoes. The insecticides examined included chlorinated hydrocarbaons such as DDT, lindane and dieldrin, and organophosphorus compounds such as diazinon and malathion. Indoor residual spraying with those insecticides was apparently effective in reducing the mosquito population for quite a long period. Such household insect pests as flies, fleas and cockroaches were also greatly affected by the residual spraying.

Developments of the resistance to any of the above insecticides in larvae and adults of Cx. pipiens pallens were already demonstrated in a few areas according to the previous history of the insecticide applications, but the insecticide resistance was regarded as little practical importance in the control at that time [13, 15].

4 Anti-mosquito measures in national filariasis control program

On the basis of the information and experience accumulated during the period of ten years, the government of Japan decided to support a country-wide filariasis control program starting in the fiscal year 1962 [12]. A text drafted by Dr. M. Sasa was issued by the Ministry of Health and Welfare as the guide for standard method of filariasis control. The prefectural health departments were responsible for selecting the areas for conducting the control, while the municipalities within each prefecture actually carried out the drug administration and the mosquito control.

At the start of the program in 1962, four prefectures in southwestern Japan, namely Ehime, Kochi, Kagoshima and Nagasaki, accepted the government proposal. Additional five prefectures, Niigata, Tokyo (the Izu Islands), Oita, Miyazaki and Kumamoto joined the project in 1963 and 1964. Some prefectures terminated the project within one to several years as the number of microfilarial carriers had been reduced nearly to zero, but the project continued for approximately ten years in such very endemic prefectures as Kagoshima, Kumamoto and Nagasaki. The filariasis control program in Okinawa started in 1966 under different administrative background but with the same technical method. Several kinds of leaflets made by the Ministry of Health and Welfare and the prefectures conducting the filariasis control program were distributed to inhabitants in endemic areas to enlighten the importance of the program. Examples of such leaflets are shown in Figs. 1 and 2.

Fig. 1.

The back page of a leaflet made by Nagasaki prefecture around 1962 for distribution to inhabitants in endemic areas of bancroftian filariasis, appealing the importance of receiving blood examination for filariasis. A microfilaria (Wuchereria bancrofti) and a mosquito (Culex pipiens pallens, the principal vector) are drawn.

Fig. 2.

The front page of a leaflet made by the Ministry of Health and Welfare around 1962 for distribution to inhabitants in endemic areas of bancroftian filariasis, appealing the cooperation with local governments to create a new healthy native land without filariasis. The importance of vector control is emphasized.

The survey and the control of vector mosquitoes for the prevention of filariasis transmission were included in the text for the filariasis control program issued by the Ministry of Health and Welfare [12]. Procedures in the endemic areas concerned were the survey and the removal of breeding places of mosquitoes with emphasis on sewage ditches near houses, and the use of insecticides and natural enemies to control mosquitoes according to the direction in the text for the program. As a use of natural enemies it was recommended to release fishes into breeding places of mosquitoes.

Prefectural governments were asked to organize necessary units of filariasis control teams, each composed roughly of one health officer, three technicians and one entomologist. By involvement of entomologists in filariasis control teams, it was assumed that survey and control of mosquitoes would be very actively carried out. Expenses of insecticides used mostly for residual spraying were considerable as seen in the example of Nagasaki prefecture in Table 1 [16].

Table 1.

Vector control in Nagasaki prefecture in 1962-1967 as a part of the National Filariasis Control Project

Year	No. houses sprayed with insecticide	Total expense of fiilariasis campaign in thousand Yen	Cost in thousand Yen of insecticide used (% to the total expenses)
1962	40,534	11,564	4,568 (39.5)
1963	35,858	12,614	5,272 (41.8)
1964	30,220	11,398	4,842 (42.5)
1965	18,852	7,284	3,779 (51.9)
1966	12,718	5,744	2,755 (48.0)
1967	11,257	5,329	2,580 (48.4)
Total	149,439	53,933	23,796 (44.1)

However, well-trained entomologists were not always available in each control team. The residual spraying was actually troublesome to the inhabitants, since all furniture had to be taken out as seen in Fig. 3 for residual insecticide to be effectively sprayed on inside walls. Although the rate of houses sprayed was fairly high, it was not recorded whether or not the insecticide was properly applied. In either case, it seems very probable that the residual spraying was effective for temporal interruption of transmission, but not for eradication from a particular area unless it was repeated at least for many years, because the adult worm continues to produce microfilariae for nearly 10 years [17]. Thus, the effects of residual spraying upon the prevalence of filariasis in human populations is much less marked than in the case of malaria [1].

Fig. 3.

Nagate village, Nagasaki prefecture, on the day of residual spraying with insecticide in 1961. Furnture was taken out from houses for insecticide to be sprayed effectively on inside walls

5 Filariasis eradication experiment by vector control alone

Since bancroftian filariasis is transmitted from man to man only by vector mosquitoes, it is supposed that without vector mosquitoes no transmission will occur and adult filariae in infected persons will die out some years after infection. Since investigations along this line were thought to be very useful for understanding the epidemiology, a filariasis eradication experiment involving only the control of the principal vector mosquito Cx. pipiens pallens was planned in 1962 at Nagate village, Nagasaki prefecture, Kyushu without treatment of persons by drugs [17].

From 1962 to 1972, residual spraying of insecticide in all dwellings and cowsheds was carried out once a year, and larvicide applications were repeated in all the possible breeding places of Cx. pipiens pallens in the village once a week during the breeding season of the vector mosquito. As a result of the control work, the density of the vector mosquito was suppressed to the level at which new filaria infections were scarcely found within the village. In 1961 the number of villagers found positive for microfilariae was 81 out of 577 (14.0%). After the vector mosquito control was started in 1962, microfilarial prevalence and density decreased gradually year by year, and in 1970 the positive rate was 0.5% (Table 2). In 1971, all remaining microfilaria carriers became negative. Those results are chiefly attributable to the natural death of adult filaria worms in carriers. From the above, it can be said that the eradication of bancroftian filariasis solely through the control of vector mosquitoes is possible in about ten years. It was also found that the period for continuous production of microfilariae by adult worms was an average of about seven years, not exceeding 10 years.

Table 2.

The reduction in the microfilaria positive rate in all persons in Nagate village, Nagasaki prefecture, durng a ten year period under continuous vector control starting in 1962

Year	No. persons examined	No. positives	% positives
1961	577	81	14.0
1962	571	71	12.4
1963	567	62	10.9
1964	541	53	9.8
1965	493	39	7.9
1966	515	31	6.0
1967	491	20	4.1
1968	441	9	2.0
1969	447	5	1.1
1970	430	2	0.5

Although bancroftian filariasis can be eradicated by vector control alone, long time and much manpower are needed in doing so, and practically for this reason intensive drug treatment has the priority in the control of filariasis.

6 The recent situation of filariasis transmission by vector mosquitoes

It has been a great concern for epidemiologists studying filariasis to evaluate the relative efficiency of chemotherapy, vector control and a combination of both in a campaign against lymphatic filariasis. In this context, the phenomena of ‘facilitation’ and ‘limitation’ have recently been discussed. WHO (1984) stated “It has been shown that, as the number of microfilariae ingested increases, there is a decrease in the proportion passing through the wall of the mosquito stomach and developing further [18]. This phenomenon has been termed limitation ... The reverse situation, termed facilitation, in which the proportion of microfilariae that develop increases as the numbers of ingested microfilariae increase, is exhibited by Anopheles mosquitoes”. Then, WHO (1992) stated that the situation of facilitation will give rise to transmission thresholds that could lead to the ultimate extinction of lymphatic parasites from the human population, when the microfilarial reservoir in the human population is lowered to a certain level by chemotherapy [19].

If the above is justified, then the facilitation would be a very attractive idea in planning a control of filariasis. However, careful examination of previously reported papers revealed that there was no clear evidence to support the existence of facilitation and facilitation-based unstable equilibrium, in relation to microfilaria prevalence and density in human population, below which filariasis would spontaneously disappear even when the vector was Anopheles mosquitoes [20].

Natural reductions in microfilaria prevalence and density were commonly encountered with W. bancrofti transmitted by Cx. pipens pallens in many areas of Japan in the past before mass chemotherapy was implemented in 1962 [11]. It seems apparent that the natural reduction in microfilaria prevalence and density in Japan was due to the decrease of vector density, or more accurately of man/mosquito contacts. This strongly suggests the existence of critical man/mosquito contacts, but not of the unstable equilibrium of filaria parasites, below which filariasis tends to disappear.

The decrease in the vector density in recent years is very remarkable in Japan. In 1994, mosquitoes were collected by the human bait method in a village of Nagate, Nagasaki prefecture, where bancroftian filariasis was formerly endemic (microfilaria positive rate was 14.0% in 1961) but all inhabitants became free of microfilariae by the control of vector mosquitoes carried out from 1962 to 1971. The abundance of mosquitoes collected in 1994 [21] was compared with that in 1961 in previously published report [10]. A remarkable decrease of the principal vector mosquito, Cx. Pipiens pallens, was noted from 1961 to 1994 (Table 3). Reduction in the mean number per collection night of the principal vector from 71.2 in 1961 to 0.6 in 1994 was chiefly due to the reduction in the breeding place of this mosquito. The main breeding place in 1961 was standing water in open ditches collecting household waste water [10], but nearly all ditches have been changed to underdrains, thus there are scarcely any breeding mosquitoes. In this way, there is now no risk of filariasis infection at Nagate.

Table 3.

Comparison of mean numbers per night of the main vector of filariasis (Culex pipiens pallens) and other mosquitoes by the human bait trap between 1961 and 1994, in Nagate village, Nagasaki prefecture, southwestern Japan. Data are from Wada, Endem Dis Bull Nagasaki [10] and Sasa et al., Trop Med [21]

	1961	1994
Cx, pipiens pallens	71.2	0.6
Others	5.6	9.7

The situation in other parts of Japan was apparently similar to Nagasaki, and man/mosquito contacts were considered far below the critical level for the existence of filariasis. This was closely related to the movement for controlling flies and mosquitoes through community participation under the strong support of the government of Japan. The movement was implemented in 1955 in almost all communities throughout Japan, and this contributed greatly to the improvement of general public health situations. Such booklets as in Fig. 4 were published by many communities for use in the movement for controlling flies and mosquitoes.

Fig. 4.

The front page of a booklet “Ecology and control methods of mosquitoes and flies” published in 1959 for use in the community participation in Kagoshima prefecture.