THE EPIDEMIOLOGY OF JAPANESE ENCEPHALITIS ON TAIWAN DURING

Transcription

1 Am. J. Trop. Med. Hyg., 61(1), 1999, pp Copyright 1999 by The American Society of Tropical Medicine and Hygiene THE EPIDEMIOLOGY OF JAPANESE ENCEPHALITIS ON TAIWAN DURING YING-CHANG WU, YAW-SHYONG HUANG, LI-JUNG CHIEN, TSUEY-LI LIN, YI-YUNG YUEH, WUU-LUH TSENG, KWO-JIING CHANG, AND GONG-REN WANG National Institute of Preventive Medicine, Department of Health, Executive Yuan, Taipei, Taiwan, Republic of China Abstract. Japanese encephalitis (JE) is an endemic disease in Taiwan. A mass vaccination program of children against JE was first implemented in Along with general improvements in various aspects of living conditions over the years, the program has brought JE well under control. The main characteristics of JE epidemiology in Taiwan in the past 3 decades are as follows. The transmission mode remains unchanged that is, the amplification stage of the virus in pigs is followed by a human epidemic each year. The frequency of JE incidence has dropped significantly. The incidence rate of confirmed cases was 2.5 per 1, in 1967, the highest in record, and merely.3 per 1, in Confirmed cases occur sporadically all over the island. The peak of the epidemic season has shifted from August in the 196s to June since the 198s. The age distribution of confirmed cases has shifted gradually from mainly children to adults. Vaccine efficacy for those having received more than 2 doses of the vaccine is estimated to be about 85%. Japanese encephalitis (JE) is a pernicious infectious disease widely spread in the western Pacific region where Taiwan is located. 1 5 This disease formerly was quite prevalent in Taiwan, reaching epidemic proportions every summer. The earliest documented clinical record of JE in Taiwan can be dated back to 1931, although under the name of summer encephalitis. 6 Kobayasi 7 first succeeded in isolating JE virus (JEV) from fatal JE cases in The ecology of JE in Taiwan was first studied and addressed by Grayston and others 8 13 in It was promulgated in 1955 that JE was designated a notifiable disease on the island, but no systematic surveillance of any sort was set up at that time. Twelve years went by before an active surveillance was put into place in 1967 under the direction of Dr. T. Okuno, the then World Health Organization consultant control of JE. Beginning in 1968, a mass vaccination program against JE for children was implemented in Taiwan, which, along with a general improvement in living conditions as well as other social and environmental changes over the past decades, has made a huge difference in the impact caused by the annual JE epidemics. Judging from the severity of suffering the disease inflicts on society, JE now seems to be under control in Taiwan. A few cross-sectional studies have been published on JE epidemiology in Taiwan. 8,14 2 However, as part of the staff of National Institute of Preventive Medicine, whose routine responsibilities in part encompass the registration, laboratory confirmation, active surveillance, and data processing of reported JE cases, we feel obligated to make known all pertinent information we have gathered over the years in an orderly, sensible, and analytical fashion. We do this in the hope that we may provide a somewhat broader comprehensive view of the real situation of JE in Taiwan in the past as well as of its current status, and we thus hope be of some help in preparing ourselves and other concerned health workers to deal with forthcoming JE epidemics. MATERIALS AND METHODS Case definitions. The following definitions were adopted when we commenced the active surveillance system for JE cases in Suspected cases. Suspected cases were defined as having 78 fever of 38 C and above with one of the following symptoms: 1) meningeal signs and symptoms such as neck stiffness, opistotonus, and Kernig sign, accompanied or unaccompanied by nausea, vomiting, and headache; 2) cortical irritation, such as convulsion and twitching; 3) disturbance of consciousness, such as lethargy, stupor, coma, and delirium; 4) cranial nerve symptoms, such as facial palsy, trismus, diplopia, strabismus, nystagmus, and swallowing disturbance; 5) pyramidal and extrapyramidal signs, such as rigidity, spasticity, flaccidity, athetosis, and chorea. Reported cases. Reported cases were suspected cases reported by the physician to health authorities. Confirmed cases. Confirmed cases were reported cases that were verified through serology testing. Reporting system. Before 1967, health authorities monitored JE cases by merely following up physicians passive reports. The active surveillance system we have now was put into place in The system stipulates that before and during each epidemic season, health station personnel are to visit local hospitals and clinics, to provide information about case definition and surveillance statistics, to remind and exhort physicians to report suspected cases to health stations, and to collect serum samples for case confirmation. Serum collection and transportation. Paired serum samples, one of them taken in the acute phase and the other during convalescence, for each reported case have been acquired for serology testing purposes since All samples are refrigerated during shipping to the National Institute of Preventive Medicine and stored at 2 C once received at the laboratory. Serology testing. A hemagglutination inhibition (HI) test has been used in serology testing. The method and reagents used in this test have remained the same throughout these years. Sucrose-acetone purified JaGAr1 strain JEV from infected suckling mouse brain is used as the antigen, and an acetone extraction method is used as serum treatment before the test. 21 Paired sera are tested side by side simultaneously. Confirmation standard. A case is considered to be confirmed based on the fulfillment of either of 2 criteria: the HI titer of the convalescent serum is 1:16 and has at least a 4-fold rise from titer in the acute phase serum, or the HI titer of either single serum is 1: Mass vaccination program. Since 1968, the first year the

2 JAPANESE ENCEPHALITIS IN TAIWAN 79 mass vaccination program against JE was implemented, vaccine has been given to children of certain age groups from March to May. Before 1987, an inactivated vaccine derived from Nakayama-NIH strain JEV-infected mouse brain was the only one used. An inactivated freeze-dried Beijing strain version was introduced in 1988, yet the Nakayama strain vaccine has remained dominant in the market to this day. Between 1968 and 1974, the vaccine was not given unless the child was 2 years old. In later years, the first dose of the vaccine was given to younger children. After 198, the first dose was given at 1 year 3 months of age. Dosages of vaccine vary for different birth cohorts. For those born between 1963 and 1969, 2 doses were given 1 week apart. For those born between 197 and 1975, 3 doses were given. A booster dose was added 1 year after the 2- dose basal vaccination. For children born after 1975, 4 doses are given. A second booster (the added fourth dose) is administered when the child attends the first grade of primary school. Vaccination records are filed at local health stations for future reference. Sequelae investigation. At end of the yearly epidemic, our laboratory personnel pay a personal visit to each patient with a new confirmed case to follow up the patient s prognosis and to complete a standardized questionnaire. This is done to determine whether the patient has a total recovery or is left with some impediments of neurologic functions (including numbness of limbs, muscle stiffness, slow movement of limbs, choreoathetosis, tremor, epilepsy, convulsion, dyslogia, facioplegia, dysphagia, dysacousia, and urine incontinence) or mental impediments (including emotional instability, changes in personality, mental retardation, and psychiatric disorder). Other information. The vaccination history of the confirmed case is to be traced through the local health station file. The record of vaccination coverage rates cited in this study are provided by Taiwan Provincial Health Department, Taipei Municipal Health Department, and Kaohsiung Municipal Health Department. Anti-JEV seroprevalence of the amplifying host. Two months before the epidemic season starts, 3 samples of swine serum are collected every week from assigned slaughterhouses in different northern and southern localities in Taiwan. Seroprevalence is determined by HI test results on those samples. Serum collection and testing are on going until 9% of seroprevalence is reached. Data from sources in central and eastern regions were also incorporated between 1978 and RESULTS Coverage rate of serum sample collection from reported cases. The coverage rate was at its lowest level in 1966 and 1967 (49 and 55%, respectively). The coverage rate was above 8% in the remaining years. Variation of anti-jev seroprevalence in the amplifying host. Variation in the annual date on which anti-jev seroprevalence surpassed 5% in pigs in northern Taiwan is shown in Figure 1. Although fluctuating somewhat, the date tended to move forward gradually from July in the 196s to May since the 198s. The date that anti-jev seroprevalence exceeded 5% in pigs in southern Taiwan was usually about FIGURE 1. The dates on which seroprevalence exceeded 5% in pigs in northern Taiwan and the median onset dates of confirmed Japanese encephalitis cases in Taiwan, Two months before the epidemic season started, serum samples were collected from 3 pigs at assigned slaughterhouses located in northern Taiwan every week to detect anti Japanese encephalitis virus antibody by hemagglutination inhibition test. The collection is ongoing until swine seroprevalence exceeded 9%. Swine seroprevalence data for 1966 and 1971 are retrieved from earlier studies. 17,22 Data for are taken from a seminar report (Okuno T, unpublished data). 1 2 weeks earlier than the date associated with pigs in northern Taiwan. The results of investigation on pigs in central and eastern Taiwan were somewhat similar to that of pigs in northern Taiwan. Once anti-jev seroprevalence has exceeded 5% in pigs, it will most likely rise to above 9% within 1 month. Incidence rate. The annual incidence rates of JE since 1955 are depicted in Figure 2. The epidemic status of JE over the years can be roughly divided into 4 stages. Before 1968, the incidence rate was quite high. It was 2.5 per 1, for confirmed cases in 1967, which was the highest recorded year. From 1968 to 1975, the incidence rate noticeably was on the decline. The number of yearly confirmed cases ranged between 44 and 279. From 1976 to 1985, occurrence was low and steady. The annual number of confirmed cases was held between 36 and 67. After 1985, the incidence rate went on a further down slide. The number of confirmed cases varied from 6 to 35 each year. Six confirmed cases occurred in 1997, which is equivalent to an incidence rate of.3 per 1,. Mortality rate. The annual mortality rates of reported cases after 1954 are shown in Figure 3. The variation in the rate through the years bears a close resemblance to that of the incidence rate. The mortality rate in 1997 was.14 per 1,, also the lowest ever recorded. Residential locations of confirmed cases. Figure 4 indicates the residential distribution of confirmed cases in different epidemic stages in Taiwan. Although the incidence rate appears to be a little higher in northern and eastern Taiwan, confirmed cases took place sporadically in various districts. In 1997, for instance, the 6 confirmed cases came from 6 separate counties in central and southern Taiwan. The distribution of epidemic peak month. Figure 5 re-

3 8 WU AND OTHERS FIGURE 4. Residential distribution of confirmed Japanese encephalitis (JE) cases in Taiwan, Figures denote numbers of confirmed JE cases per 1, total population. FIGURE 2. Incidence rate of Japanese encephalitis (JE) in Taiwan, Japanese encephalitis was officially classified as a notifiable infectious disease in 1955, and laboratory diagnosis was established in Reported cases are defined as suspected JE cases that were reported to health authorities by physicians. Confirmed cases are those reported cases that were diagnosed positive as JE by hemagglutination inhibition test. veals the distribution of the month of onset of confirmed cases from 1966 to Over those 3 decades, the peak of the epidemic season shifted forward by approximately 2 months. The peak month was August in the 196s and shifted to July in the 197s. Since the 198s, however, the peak moved well into the month of June. Figure 1 also clearly demonstrates that the epidemic season shifted forward from the 196s to the 198s. Although the current epidemic season lasts from May to October, sporadic cases occur in other months as well. For example, confirmed incidents occurred in January (1 case), February (1 case), April (4 cases), November (7 cases), and December (1 case) between 1978 and This indicates that JEV is somewhat active in the winter as well as in other months of the year. Age distribution for confirmed cases. The incidence rates in all age groups were mostly on downward slopes between 1966 and 1975 (Figure 6). Figure 7 depicts the track of the incidence rate for each age group since The incidence rates for age groups 9 and 1 19 years declined, whereas that for the age 4 and above group remained low but level. The incidence rate for age group 2 29 years first increased gradually, but then decreased after The incidence rate for age group 3 39 years increased after 1978, and the upward trend became more prominent after Figure 8 shows the age-specific incidence rates of birth cohorts , , and , respectively, during the period The birth cohort had the highest incidence rates among the 3 birth cohorts for most age groups over 9 years of age. Vaccination coverage rate. The coverage rates of the fourth dose (the second booster) given at primary school to FIGURE 3. Mortality rate of Japanese encephalitis (JE) in Taiwan, Rates are given as the number of patients who died as reported JE cases in a population of 1,. FIGURE 5. Occurrence of confirmed Japanese encephalitis (JE) cases in Taiwan by month of onset. Rates are given as the proportion of cases occurring in the indicated months in 1966, 1976, 1986, and 1996, respectively.

4 JAPANESE ENCEPHALITIS IN TAIWAN 81 FIGURE 6. Age-specific incidence rates of confirmed Japanese encephalitis cases in Taiwan, first graders, reached above 95%. The third dose coverage rates were 5.3% in 1974, 74 8% between 1975 and 1986, and over 8% after The second dose coverage rates were mostly above 8%, except for the years 1968 (only 14.1%), 1969 (56.1%), and 197 (68.5%). These rates stayed over 9% after Vaccine efficacy. The third dose (the first booster) of JE vaccine has been offered since In our efficacy study, the confirmed cases were divided into 2 groups based on the dosage availability of JE vaccine: group 1, born between 197 and 1975, when 3 doses of vaccine were given, and group 2, born between 1976 and 1994, when 4 doses were given. The vaccine efficacy data for those who received more than 2 doses of vaccine are presented in Table 1. The efficacy is estimated to be about 85% on average and definitely exceeded 8%. No statistically significant differences were found among age groups. Sequelae investigation. Table 2 shows the prognoses (investigated in the year of onset) in patients with confirmed cases who were born after 1969 and had their onset in On average, 4% of the surviving patients had sequelae. The prognosis distribution of patients with confirmed FIGURE 8. Comparison of age-specific patterns of confirmed Japanese encephalitis incidence rates among 3 separate birth cohorts in Taiwan during cases between those who received more than 2 doses of vaccine and those who did not receive any was not different statistically ( , P.23). DISCUSSION This report describes the epidemic situation of JE in Taiwan based on confirmed cases rather than reported ones. This may miss some fatal cases that did not have the chance to go through the confirmation process; also excluded are non-je cases. Since 1966, confirmation of all reported cases by serology testing has been required. The coverage rate of paired serum sample collection from reported JE cases rose to above 8% after 1967, but more than 8% of the reported cases were diagnosed to be something other than JE after 1975 (Figure 2). That sufficiently explains why using the data of confirmed cases to evaluate the epidemiology of JE is better than using reported cases. TABLE 1 Efficacy of Japanese encephalitis vaccine in groups vaccinated with more than 2 doses in Taiwan Province, * Birth year FIGURE 7. Age-specific incidence rates of confirmed Japanese encephalitis cases in Taiwan, Age (years) Total Vaccination status No. cases Vaccine efficacy (%) No. cases Vaccine efficacy (%) * Vaccine efficacy (1 incidence rate in group vaccinated with 2 doses/incidence rate in unvaccinated group) 1%. People born between 197 and 1975 underwent the 3-dose vaccination program. People born after 1975 received 1 additional boost (the fourth dose) while attending first grade of primary school.

5 82 WU AND OTHERS TABLE 2 Prognosis of patients with confirmed cases of Japanese encephalitis in Taiwan, * Vaccination status Recovered (%) Prognosis Recovered with sequelae (%) No. of cases 96 (56.8) 73 (43.2) doses 3 doses 34 (61.8) 35 (66.) 21 (38.2) 18 (34.) Total 165 (59.6) 112 (4.4) 277 * Patients surviving confirmed Japanese encephalitis cases were all paid a visit to investigate their prognosis each December. Only those who were born after 1969 are included in this table. No statistically significant difference was found in distribution of prognosis between the unvaccinated group and the group vaccinated with 2 doses ( , P.23). Sequelae included neurologic function impediments, such as numbness of limbs, muscle stiffness, slow movement of limbs, choreoathetosis, tremor, epilepsy, convulsion, dyslogia, facioplegia, dysphagia, dysacousia, and urine incontinence; mental impediments, including emotional instability, changes in personality, mental retardation, and psychiatric disorder. In last 3 decades, Taiwan has experienced a general improvement in many aspects of the living environment, coupled with a zealous and lasting immunization campaign for children. The net effect of these has been a change in the epidemic curve, reducing the incidence rate, and a shift of the average age of infection from young to older persons. In summary, enough epidemiologic evidence exists to demonstrate each of the following characteristics. The mode of transmission has remained unchanged. Culex tritaeniorhynchus was and still is the main vector, and swine are the amplifying hosts. Swine amplify the virus in late spring, which causes the human epidemic that follows. Studies in the 196s indicated Cx. tritaeniorhynchus and Culex annulus were the 2 vectors for JEV transmission in Taiwan. 1,22 25 Later studies suggested Cx. triaeniorhynchus to be more prominent. 26 Surveys made in 1978 and during fortified the latter notion by affirmative isolation of JEV from Cx. tritaeniorhynchus. 26,27 In our laboratory, we were able to isolate JEV from field-caught Cx. tritaeniorhynchus in every summer from 1986 to 199 (Wu YC and others, unpublished data). According to the investigation in the 196s, peak densities of Cx. trittaeniorhynchus were reached in the months of July and August. 9,28 This pattern was found little changed in This evidence implies that both the vector identity and the peak distribution of its density had little disparity over the past 3 years. Hurlburt 3 established the pig mosquito amplifying cycle of JEV in Taiwan in Figure 1 indicates a close correlation between the median of onset dates of confirmed cases in any particular year and the date in that year when swine seroprevalence exceeded 5%, which suggests the pig mosquito amplifying cycle of JEV on Taiwan also remained the same despite the passing of 3 decades. The incidence rate of confirmed cases has dropped dramatically. We believe several reasons exist for the decrease. First, many changes have occurred in the living environment in Taiwan, such as urbanization, shrinkage of rice fields, the use of insecticides, and centralization of the pig feeding industry which relocalized pig farms away from residential areas. These changes effectively lessened the exposure of humans to vectors. Indeed, the incidence rate in nonvaccinated infants decreased from 6.4 per 1, in 1966 to.37 per 1, in 198, and stayed very low afterward (Wu YC and others, unpublished data). Second, the mass vaccination has had a definite impact. As shown in Figures 6 and 7, the incidence rate in children under 1 years of age dropped markedly after Third, the epidemic season has shifted forward (Figure 5). The peak density of Cx. tritaeniorhynchus invariably occurred in July to September over the years; however, the epidemic in humans has moved forward into June since the 198s, and thus the vector density at the time is comparatively lower. 29 Fourth, other minor factors exist, such as improvement in nutrition and a decrease in outdoor activities at night for most young children. The mortality rate also decreased along with the incidence rate. This might be a result of lowered JEV virulence or sheerly caused by the decrease in incidence rate. Another possibility is the increase in the average age of JE patients, who might have experienced cross-reacting antigens to JEV in their early life and acquired some protection. Confirmed cases are found all over Taiwan. The residential distribution of confirmed cases is shown in Figure 4. The wide distribution of cases might be due to the occurrence of both the vector (mosquito) and the amplifying host (pig) in neighborhoods of nearly the entire island. The current epidemic season lasts from May to October each year. The reasons for the forward shift in the epidemic peak month need to be elucidated here. Chang and others 31 noticed new infections in pigs in southern Taiwan during January and March. Also, sporadic confirmed human cases occur in winter all over the island. The timing of annual JEV emergence has been demonstrated to not exactly correlate with the density of vectors. 1,3,32 In a study done in , Rosen and others 26 found 2 peaks of positive JEV isolation from Cx. tritaeniorhynchus in northern Taiwan each year. These peaks were 4 months apart. Our investigation of anti-jev seroprevalence of swine also shows a regular pattern of JEV emergence every year across Taiwan. The virus emerges in southern Taiwan first, then appears across the rest of Taiwan without a fixed order. This suggests a high possibility of JEV presence during winter throughout Taiwan. Ecological studies in Hokkaido and Okinawa demonstrate that JEV does overwinter locally. 33,34 Several hypotheses exist for JEV overwintering; however, none of them can explain why the epidemic peak month has moved forward in Taiwan. 35 Although the clustering of pigs may lead to more rapid spread of JEV, the spread of JEV in the 196s was already at a comparable, if not greater, speed, and thus the influence of consolidation of pig farms should be limited. Japanese encephalitis virus can remain in the human nervous system and cause latent infection and recurrence. 36,37 If latent infection of JEV also occurs in pigs, the forward movement of the epidemic peak month would be explained to some extent. A female pig with JEV infection could become latently infected. If JEV reactivates when the pig is pregnant, the virus would pass to newborn pigs, as it does in mice. 38,39 Both the mother and the newborn pigs would then serve as reservoirs for the periodic occurrence of JEV. The advancement of pig feeding technology from the 196s to the 198s, which has apparently shortened the period between pregnancies of female pigs, might be responsible for earlier JEV emergence. The onset age of the average confirmed case shifted grad-

6 JAPANESE ENCEPHALITIS IN TAIWAN 83 ually from under 1 years of age toward adulthood. The birth cohort has the highest incidence rates for age groups over 9 years old throughout the study (Figure 8). Several reasons exist for the high incidence rates in the birth cohort. General improvement in the living conditions on the island decreased the odds of people s exposure to JEV. This cohort, with less childhood exposure and less immunity, would have a higher incidence rate at older ages when compared to the earlier, more exposed birth cohort. The fact that most members of the birth cohort did not receive any JE vaccine also results in the higher incidence rate at older ages in that group when compared with the younger vaccinated birth cohort. This result coincides with the findings of a 1991 JE seroepidemiologic study in Taiwan. 4 Similar shifting of onset age was also observed in Korea and China after initiation of JE vaccination programs. 41,42 Vaccine efficacy after more than 2 doses of vaccine is estimated to be about 85%. Hsu and others 43,44 reported the efficacy of the first 2 basal doses of JE vaccine in the same year after administration was about 8%, but dropped to 58.7% the next year. Therefore, to get a more complete picture, we conducted a survey and a calculation on vaccine efficacy in those who received more than 2 doses of JE vaccine in our study (Table 1). The efficacy for each age group did not seem to be very consistent, perhaps due to an overall incidence rate that was too low. However, on average, efficacy remained above 8% within 1 years after the third dose of vaccination (the 95% confidence interval for those born in was 76 9%, and the interval was 79 92% for those who were born after 1975). However, further investigation is called for to determine the exact endurance of vaccine protection. About 4% of all patients who survived confirmed JE cases had sequelae (Table 2). Although confirmed cases seem to include nonvaccinated cases, the difference is not statistically significant (P.23), which was consistent with findings in China. 45 This may indicate that JE vaccine is effective in blocking viremia by neutralizing antibodies, but the vaccine cannot modify the disease outcome once the virus has invaded the central nervous system. 46 Japanese encephalitis virus is still wide spread and eradication in Taiwan seems remote, unless some abrupt immense changes in living conditions that are capable of effectively eliminating or breaking the JE transmission cycle. For the time being, a more effective and convenient vaccine is needed to ensure that JE is controlled and suppressed to minimize its human impact. Acknowledgments: We would like to thank Dr. T. Okuno for his guidance in setting up the JE active surveillance system and Mr. C. C. Kuo for his efforts put in systematizing this work. Thanks also go to the personnel in the Taiwan Provincial Health Department, Taipei Municipal Health Department and Kaohsiung Municipal Health Department, for their relentless endeavor in collection of JE prevalence information and serum samples, and to colleagues at the National Quarantine Service and National Institute of Preventive Medicine, for their professional know-how and serology testing. Financial support: This study was financed solely by the National Institute of Preventive Medicine, Department of Health, Executive Yuan, Taipei, Taiwan, Republic of China. Authors addresses: Ying-Chang Wu, Yaw-Shyong Huang, Li-Jung Chien, Yi-Yung Yueh, Wuu-Luh Tseng, and Kwo-Jiing Chang, Division of Epidemiology, National Institute of Preventive Medicine, Department of Health, 161 Kun-Yang Street, Nang-Kang District, Taipei, Taiwan, Republic of China. Tsuey-Li Lin, Division of Virology, National Institute of Preventive Medicine, Department of Health, 161 Kun-Yang Street, Nang-Kang District, Taipei, Taiwan, Republic of China. Gong-Ren Wang, National Institute of Preventive Medicine, Department of Health, 161 Kun-Yang Street, Nang-Kang District, Taipei, Taiwan, Republic of China. Reprint requests: Ying-Chang Wu, Division of Epidemiology, National Institute of Preventive Medicine, Department of Health, 161 Kun-Yang Street, Nang-Kang District, Taipei, Taiwan, Republic of China. REFERENCES 1. Wakai S, Scourge of Japanese encephalitis in southwestern Nepal. Lancet 351: Hanna JN, Ritchie SA, Phillips DA, Shield J, Bailey MC, Mackenzie JS, Poidinger M, McCall BJ, Mills PJ, An outbreak of Japanese encephalitis in the Torres Strait, Australia, Med J Aust 165: Cardosa MJ, Hooi TP, Kaur P, Japanese encephalitis virus is an important cause of encephalitis among children in Penang. Southeast Asian J Trop Med Public Health 26: Paul WS, Moore PS, Karabatsos N, Flood SP, Yamada S, Jackson T, Tsai TF, Outbreak of Japanese encephalitis on the island of Saipan, 199. J Infect Dis 167: Vaughn DW, Hoke CH, The epidemiology of Japanese encephalitis: prospects for prevention. Epidemiol Rev 14: Sakai K, Summer encephalitis among children in northern Formosa, Taiwan. Igakkai Zasshi 34: Kobayasi H, 194. On the virus of encephalitis epidemica isolated in Taihoku and Sintiku provinces in the summer, Acta Jpn Med Trop 2: Grayston JT, Wang SP, Yen CH, Encephalitis on Taiwan. I. Introduction and epidemiology. Am J Trop Med Hyg 11: Hu SMK, Grayston JT, Encephalitis on Taiwan. II. Mosquito collection and bionomic studies. Am J Trop Med Hyg 11: Wang SP, Grayston JT, Hu SMK, Encephalitis on Taiwan. III. Virus isolations from mosquitoes. Am J Trop Med Hyg 11: Wang SP, Grayston JT, Encephalitis on Taiwan. IV. Human serology. Am J Trop Med Hyg 11: Wang SP, Grayston JT, Chu IH, Encephalitis on Taiwan. V. Animal and bird serology. Am J Trop Med Hyg 11: Chu IH, Grayston JT, Encephalitis on Taiwan. VI. Infections in American servicemen on Taiwan and Okinawa. Am J Trop Med Hyg 11: Hsieh WC, Wang SP, Rasmussen AF, Epidemiology of Japanese encephalitis (JE) on Taiwan in 196. J Formos Med Assoc 6: Fan KY, Hsu TC, Chen CL, Epidemiological study on cases of epidemic Japanese B. encephalitis occurring in 1961 epidemic in north Taiwan. J Formos Med Assoc 61: Green IJ, Wang SP, Yen CH, Hung SC, The epidemiology of Japanese encephalitis virus on Taiwan in Am J Trop Med Hyg 12: Hsu TC, Hung CT, Hsu ST, Epidemiology and control of Japanese encephalitis in Taiwan. Jpn J Trop Med 1: Okuno T, Tseng PT, Hsu ST, Huang CT, Kuo CC, Lin SY, Japanese encephalitis surveillance in China (province of Taiwan) during I. Geographical and seasonal features of case outbreaks. Jpn J Med Sci Biol 28: Okuno T, Tseng PT, Hsu ST, Huang CT, Kuo CC, Lin SY, Japanese encephalitis surveillance in China (province of Taiwan) during II. Age-specific incidence in con-

7 84 WU AND OTHERS nection with Japanese encephalitis vaccination program. Jpn J Med Sci Biol 28: Chang KJ, Tseng TC, Seroepidemiological investigation on Japanese encephalitis in Taiwan. Chin J Microbiol Immunol 26: Clarke DH, Casals J, Techniques for hemagglutination and hemagglutination inhibition with arthropod-borne viruses. Am J Trop Med Hyg 7: Okuno T, Michell CJ, Chen PS, Wang JS, Lin SY, Seasonal infection of Culex mosquitos and swine with Japanese encephalitis virus. Bull World Health Organ 49: Okuno T, Tseng PT, Liu SY, Hsu ST, Huang CT, Rates of infection with Japanese encephalitis virus of two culicine species of mosquito in Taiwan. Bull World Health Organ 44: Cates MD, McCroddan DM, Huang WC, Chiu SF, Lien SC, Lin PC, Cross JH, Japanese encephalitis virus surveillance in Taiwan. I. Isolations from mosquitoes J Formos Med Assoc 68: Cates MD, Detels R, Japanese encephalitis virus in Taiwan: preliminary evidence for Culex annulus Theob as a vector. J Med Entomol 6: Rosen L, Lien JC, Lu LC, A longitudinal study of the prevalence of Japanese encephalitis virus in adult and larval Culex tritaeniorhynchus mosquitoes in northern Taiwan. Am J Trop Med Hyg 4: Lien JC, Huang WC, Cross JH, 198. Japanese encephalitis virus surveillance in the Taipei area, Taiwan in Southeast Asian J Trop Med Public Health 11: Sun WKC, The seasonal succession of mosquitoes in Taiwan. J Med Entomol 1: Lin TH, Lu LC, Population fluctuation of Culex tritaeniorhynchus in Taiwan. Chin J Entomol 15: Hurlbut HS, The pig mosquito cycle of Japanese encephalitis virus in Taiwan. J Med Entomol 1: Chang CP, New AE, Chiang HS, Irving GS, Taylor JF, Chen WF, A surveillance for Japanese encephalitis (JE) in pigs in southern Taiwan. Chin J Microbiol 7: Detels R, Cross JH, Huang WC, Lien JC, Chen S, Japanese encephalitis virus in northern Taiwan, Am J Trop Med Hyg 25: Takashima I, Watanabe T, Ouchi N, Hashimoto N, Ecological studies of Japanese encephalitis virus in Hokkaido: interepidemic outbreaks of swine abortion and evidence for the virus to overwinter locally. Am J Trop Med Hyg 38: Tadano M, Kanemura K, Arakaki S, Makino Y, Fukunaga T, Hokama T, Kyan T, Makiya H, Taira Y, Ohama N, Iju C, Sakurai K, Epidemiological and ecological studies of Japanese encephalitis in Okinawa, subtropical area in Japan. II. Prevalence of Japanese encephalitis antibody in residents in Okinawa, Miyako and Ishigaki islands. Microbiol Immunol 38: Rosen L, The natural history of Japanese encephalitis virus. Annu Rev Microbiol 4: Ravi V, Desai AS, Shenoy PK, Satishchandra P, Chandramuki A, Gourie-Devi M, Persistence of Japanese encephalitis virus in the human nervous system. J Med Virol 4: Sharma S, Mathur A, Prakash V, Kulshreshtha R, Kumar R, Chaturvedi UC, Japanese encephalitis virus latency in peripheral blood lymphocytes and recurrence of infection in children. Clin Exp Immunol 85: Mathur A, Arora KL, Chaturvedi UC, Transplacental Japanese encephalitis virus (JEV) infection in mice during consecutive pregnancies. J Gen Virol 59: Mathur A, Arora KL, Rawat S, Chaturvedi UC, Persistence, latency and reactivation of Japanese encephalitis virus infection in mice. J Gen Virol 67: Shyu WRH, Wang YC, Chin C, Chen WJ, Assessment of neutralizing antibodies elicited by a vaccine (Nakayama) strain of Japanese encephalitis virus in Taiwan. Epidemiol Infect 119: Umenai T, Krzysko R, Bektimirov TA, Assaad FA, Japanese encephalitis: current worldwide status. Bull World Health Organ 63: Gu PW, Ding ZF, Inactivated Japanese encephalitis (JE) vaccine made from hamster kidney cell culture (a review). Jpn Encephalitis Haemorrhagic Fever Renal Syndr Bull 2: Hsu TC, Chow LP, Wei HY, Chen CL, Hsu ST, Huang CT, Kitaoka M, Sunaga H, A controlled field trial for an evaluation of effectiveness of mouse-brain Japanese encephalitis vaccine. J Formos Med Assoc 7: Hsu TC, Hsu ST, Effectiveness of Japanese encephalitis vaccine: study in the second year following vaccination. Hammon WMcD, ed. Immunization for Japanese Encephalitis. Tokyo: Igaku Shoin Ltd., Luo D, Song J, Ying H, Yao R, Wang Z, Prognostic factors of early sequelae and fatal outcome of Japanese encephalitis. Southeast Asian J Trop Med Public Health 26: Igarashi A, Epidemiology and control of Japanese encephalitis. World Health Stat Q 45: