Post on 26-Apr-2023
PR
OO
F
Acta Tropica xxx (2005) xxx–xxx
Effect of irrigated rice agriculture on Japanese encephalitis,including challenges and opportunities for integrated
vector management
3
4
5
Jennifer Keisera,∗, Michael F. Malteseb, Tobias E. Erlangera, Robert Bosc,Marcel Tannera, Burton H. Singerd, Jurg Utzingera
6
7
a Swiss Tropical Institute, P.O. Box, CH-4002 Basel, Switzerland8b St. Antony’s College, Oxford University, Oxford OX2 6JF, UK9
c Water, Sanitation and Health, World Health Organization, Avenue Appia 20, CH-1211 Geneva 27, Switzerland10d Office of Population Research, Princeton University, Princeton, NJ 08544, USA11
Received 29 October 2004; received in revised form 31 March 2005; accepted 5 April 2005
12
Abstract13
Japanese encephalitis (JE) is a disease caused by an arbovirus that is spread by marsh birds, amplified by pigs, and mainlytransmitted by the bite of infectedCulex tritaeniorhynchusmosquitoes. The estimated annual incidence and mortality rates are
fe yearsbetweenigated1963 to
illionludingtalluated dryfungi
egratedlations,
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
vector30
31
32
1 02 d
CO
RR
EC
TED
ACTROP 1666 1–18
30,000–50,000 and 10,000, respectively, and the estimated global burden to JE in 2002 was 709,000 disability-adjusted lilost. Here, we discuss the contextual determinants of JE, and systematically examine studies assessing the relationshipirrigated rice agriculture and clinical parameters of JE. Estimates of the sizes of the rural population and population in irrareas are presented, and trends of the rural population, the rice-irrigated area, and the rice production are analyzed from2003. We find that approximately 1.9 billion people currently live in rural JE-prone areas of the world. Among them 220 mpeople live in proximity to rice-irrigation schemes. In 2003, the total rice harvested area of all JE-endemic countries (excthe Russian Federation and Australia) was 1,345,000 km2. This is an increase of 22% over the past 40 years. Meanwhile, the torice production in these countries has risen from 226 millions of tonnes to 529 millions of tonnes (+134%). Finally, we evathe effect of different vector control interventions in rice fields, including environmental measures (i.e. alternate wet anirrigation (AWDI)), and biological control approaches (i.e. bacteria, nematodes, invertebrate predators, larvivorous fish,and other natural products). We conclude that in JE-endemic rural settings, where vaccination rates are often low, an intvector management approach with AWDI and the use of larvivorous fish as its main components can reduce vector popuand hence has the potential to reduce the transmission level and the burden of JE.© 2005 Published by Elsevier B.V.
Keywords: Japanese encephalitis; Geographical distribution; Global burden; Vector control; Rice agriculture; Irrigation; Integratedmanagement; Environmental control; Biological control
∗ Corresponding author. Tel.: +41 61 284 8218; fax: +41 61 284 8105.E-mail address:jennifer.keiser@unibas.ch (J. Keiser).
001-706X/$ – see front matter © 2005 Published by Elsevier B.V.oi:10.1016/j.actatropica.2005.04.012
UN
T P
RO
OF
8
2 J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx
1. Introduction1
Japanese encephalitis (JE) is a mosquito-borne viral2
disease. The JE virus is a member of the familyFla-3
viviridae. Clinically apparent infection takes place in4
one out of 200–300 infected patients (Pugachev et al.,5
2003). The disease is characterized by a wide range of6
presentations, as both the symptoms and the clinical7
course can differ broadly among patients. They range8
from mild flu-like symptoms to considerable neuro-9
logic symptoms, such as rigors, convulsions, polio-like10
flaccid paralysis, seizures or encephalomyelitis. Severe11
clinical cases are likely to have life-long neurological12
sequelae. Mostly children and young adults are affected13
(Solomon et al., 2000; Tsai, 2000; Ding et al., 2003;14
Halstead and Jacobson, 2003). The annual incidence15
and mortality estimates for JE are 30,000–50,000 and16
10,000, respectively (Solomon, 2004). However, there17
is considered to be severe under-reporting of JE and18
one study estimated the annual incidence at 175,000 per19
year (Tsai, 2000). JE outbreaks occur in cycles that may20
be linked to climatic patterns and the immune status of21
the populations. The great majority of cases and death22
occur in World Health Organization (WHO) regions of23
South-East Asia and the Western Pacific. In 2002, the24
estimated global burden of JE was 709,000 disability-25
adjusted life years (DALYs) lost (WHO, 2004). At26
present there are no established antiviral treatments27
against JE. Interferon alpha was the most promising28
drug in small open-label trials, but it failed to affect the29
o30
iol-31
o sive32
v ated33
i sia34
a and35
T ly re-36
d d37
J ns-38
m -East39
A ex-40
p , as41
w vel-42
o pro-43
d auses44
f a,45
1 ce46
o past47
50 years (Solomon et al., 2000). At present, the geo- 48
graphical distribution of JE ranges from Japan, mar-49
itime Siberia and the Republic of Korea in the North,50
to most parts of China and the Philippines in the East,51
Papua New Guinea in the South, and India and Nepal to52
the West (Broom et al., 2003). Recent outbreaks of JE 53
have been reported Southward in Australia, and west-54
ward in Pakistan (Solomon et al., 2000). 55
Currently, approximately 90% of the world’s rice 56
is produced in Asia (Consultative Group on Interna- 57
tional Agricultural Research et al., 1998). In most of 58
the countries, where JE outbreaks have been reported,59
rice is not only a staple food, but rice growing also is60
a major economic activity and key source of employ-61
ment and income generation (Consultative Group on 62
International Agricultural Research et al., 1998). Ef- 63
forts to further enhance the high annual rice production64
in these areas are essential to maintain food security.65
It has been estimated that in the next 25 years the de-66
mand for rice will rise by 65% in the Philippines, 51% 67
in Bangladesh, 45% in Viet Nam and 38% in Indonesia68
(http://www.biotech-info.net/riceexpert.html). Thai- 69
land, for example, is already in the planning stages70
of designing new rice-irrigation schemes for year-71
round irrigation (Consultative Group on International 72
Agricultural Research et al., 1998). Hence, there is con- 73
siderable concern in public health circles, as the in-74
tensification of rice production systems as well as the75
extension of the flooded surface area, particularly in76
semi-arid areas, contributes greatly to increased fre-77
q 78
the79
e JE;80
a ns 81
a rated82
v per83
i ex-84
t ole85
o Sec-86
o be-87
t s of88
J ing89
i in 90
J ions91
o e 92
c and93
r oun-94
t re-95
UN
CO
RR
EC
utcome in children with JE (Solomon et al., 2003).There has been a changing pattern in the epidem
gy of JE. On the one hand, primarily due to extenaccination campaigns, JE has been almost eliminn many economically advanced countries of East And South-East Asia (i.e. Japan, Republic of Koreaaiwan) and the burden of JE has been substantialuced in many other endemic countries (Halstead anacobson, 2003). On the other hand, intensified traission has been observed in other parts of Southsia and the Western Pacific, most likely due to anansion of irrigated agriculture and pig husbandryell as changing climatic factors. Water resource depment and management, in particular flooded riceuction systems, are considered among the chief c
or several JE outbreaks (Amerasinghe and Ariyasen991; Akiba et al., 2001). Conversely, the occurrenf the disease has changed considerably over the
ED
ACTROP 1666 1–1
uencies and intensities of JE outbreaks.The objectives of this paper were (i) to evaluate
ffect of irrigated rice agriculture on the burden ofnd (ii) to review different vector control interventiond discuss challenges and opportunities for integector management (IVM). The remainder of this pas structured as follows. First, we put forward contual determinants of JE, highlighting the important rf rice agroecosystems on JE vector populations.nd, we review studies assessing the relationship
ween irrigated rice growing and clinical parameterE. Third, we present estimates of the population liv
n proximity to irrigation and rice-irrigation schemesE-prone areas, stratified by relevant WHO sub-regf the world (WHO, 2004). Fourth, we quantify thhanges of the rice-irrigated area, rice production,ural population sizes over the past 40 years, in cries, where JE is currently endemic. Finally, we
CT
PR
OO
F
8
J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx 3
viewed studies that employed different environmental96
and biological control interventions to reduce larval JE97
vector populations, which have been implemented in98
irrigated rice production systems.99
2. Contextual determinants100
Fig. 1 depicts the contextual determinants of JE.101
The most important epidemiological features that gov-102
ern the transmission of JE include (i) environmental103
factors, i.e. agricultural practice, altitude, climate and104
the presence of pigs and marsh birds; (ii) human im-105
munization rates and vector control measures; and (iii)106
socioeconomic parameters.107
The principal vector of JE isCulex tritae-108
niorhynchus. Female specimens are infective 9–10109
days after having taken the viraemic blood meal,110
having undergone three gonotropic cycles (Gajanana111
et al., 1997). Other culicine mosquitoes that can trans-112
mit JE includeCx. bitaeniorhyncus,Cx. epidesmus,Cx.113
fuscocephala, Cx. gelidus, Cx. pseudovishnui, Cx. si-114
tiens,Cx. vishnuiandCx. whitmorei(Sehgal and Dutta,115
2003). In Australia,Cx. annulirostriswas found to be116
the major JE vector species (Hanna et al., 1996). In117
a recent study in Kerala, South India, JE was isolated118
fromMansonia indiana(Arunachalam et al., 2004).119
Although JE vectors are able to breed in ground wa-120
ter habitats, sunlit pools, roadside ditches, tidal marshes121
of low salinity, or man-made containers, one of their122
major preferred larval habitats are rice fields (Mogi, 123
1984; Sucharit et al., 1989). The ecology ofCulex 124
spp. in rice fields has been studied and reviewed in125
great detail (Suzuki, 1967; Lacey and Lacey, 1990). 126
Since very high densities of JE vector species were127
found consistently in rice fields, it was concluded128
that the impact of these man-made breeding sites is129
much more important than that of natural breeding130
places. For example, a significant increase in the abun-131
dance ofCx. tritaeniorhynchusand increased human-132
vector contacts have been noted following completion133
of the large rice-irrigation scheme in the Mahaweli134
project, Sri Lanka (Amerasinghe and Ariyasena, 1991;135
Amerasinghe, 1995). 136
JE vector abundance is closely related to agro-137
climatic features (Peiris et al., 1993; Phukan et al.,138
2004), most notably temperature and monthly rain-139
fall (Suroso, 1989; Solomon et al., 2000; Bi et al.,140
2003). In addition, potential JE vectors were rarely141
found at altitudes above 1200 m (Peiris et al., 1993). 142
However, the most important causative factors of JE143
is the management of paddy water, and the peak peri-144
ods of mosquito abundance are associated with cycles145
in local agricultural practices. In Thailand, the highest146
rminan
UN
CO
RR
E
Fig. 1. Contextual dete
ED
ACTROP 1666 1–1
ts of Japanese encephalitis.
T P
RO
OF
8
4 J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx
numbers of larvae and pupae of JE vectors were col-147
lected when the rice fields were ploughed with the water148
in the fields (also termed puddling). The vector popula-149
tion decreased after transplanting when the fields were150
flooded, and stayed low until harvesting (Somboon151
et al., 1989). In Malaysia, small plots in the rice fields,152
which are common before planting and contain veg-153
etation, were found to be conductive factors to facili-154
tate enhanced breeding of JE vectors; up to 40 pupae155
were collected per m2 (Heathcote, 1970). Furthermore,156
the height of the rice plants, water temperature, dis-157
solved oxygen, ammonia nitrogen and nitrate nitrogen158
strongly influence the abundance of immatures (Sunish159
and Reuben, 2001). The practice of paddy cultivation,160
proximity of houses to water bodies and suitable cli-161
matic factors were the most important environmental162
factors associated with several recent JE outbreaks in163
Northeast India (Phukan et al., 2004).164
The presence of pigs and marsh birds is crucial in165
the etiology of JE, as the virus is carried by birds and166
amplified by pigs (Broom et al., 2003). The latter are167
the most important natural host for transmission of JE168
to humans. Pigs have high and prolonged viraemias,169
are often common in endemic countries, and are gen-170
erally reared in open and unroofed pigpens, which171
are located near houses (Mishra et al., 1984; Solomon172
et al., 2000). In 2004, the world’s pig population was173
estimated at 951 million, of which 60% are found in174
Asia (www.faostat.fao.org). In this part of the world175
pig farming has increased considerably over the past176
1 on177
i178
s es are179
c180
e ith181
e has182
b cat-183
t 0184
T ian185
v in186
r ion187
r and188
5 ith189
h 56%,190
r191
ants192
c tion193
s ects194
of different environmental and biological control in-195
terventions in rice fields. Though JE vectors tend to196
bite and rest outdoors, self-protection behaviour, such197
as sleeping under insecticide-treated nets (ITNs), us-198
ing insect repellants, and insect-proofing homes and199
work places, might also assist in reducing JE trans-200
mission (WHO, 1997). For example, a population- 201
based case-control study in China, which evaluated202
the protective effect of ITNs against JE, showed that203
the risk of infection among children below 10 years204
was greatly reduced (Luo et al., 1994). On the other 205
hand, all 187 serologically confirmed JE cases in re-206
cent outbreaks in Northeast India reported that they207
had slept under a bed-net (Phukan et al., 2004). Fur- 208
thermore, application of deet-permethrin soap (“Mos-209
bar”) let to an 89–100% reduction in man-vector con-210
tact, including vectors transmitting JE (Mani et al., 211
1991b). 212
The third broad category of determinants that gov-213
ern JE transmission is people’s socioeconomic status.214
In Central China, more JE cases were observed among215
children living in poor quality houses, and whose par-216
ents had lower income. However, the sample size of the217
study was small, which might explain why the associa-218
tions were not statistically significant (Luo et al., 1995). 219
Religion, exposure to domestic animals, and household220
crowding were also described as risk factors associated221
with JE (Halstead and Jacobson, 2003). 222
3 223
an-224
a ans-225
m ct a226
s ork227
w e ir-228
r vid229
T ap-230
p ali-231
t 232
r a- 233
p in-234
c Chi-235
n e pa-236
p oned237
d are238
s 239
UN
CO
RR
EC
0 years, from 490 million pigs in 1994 to 574 millin 2004 (+17.1%) (www.faostat.fao.org). In Asia pigstocks increased. Humans, goats, cattle and horsonsidered dead-end hosts (Reuben et al., 1992). Forxample, in the Thanjavur district, India, an area wxtensive rice agriculture, a very low JE incidenceeen reported, which has been explained by a high
le to pig ratio (400:1) (Vijayarani and Gajanana, 200).he important role of birds was demonstrated in Indillages with or without herons in close proximity;ice-growing villages without herons, seroconversates in children aged 0–5 and 6–15 years were 0%%, respectively. In ecologically-similar villages werons, the corresponding rates were 50% andespectively (Mani et al., 1991a).
The second category of contextual determinomprises vaccination and transmission interruptrategies. Later in this paper, we review the eff
ED
ACTROP 1666 1–1
. Rice irrigation and JE incidence
As summarized in the previous section, the mgement of paddy water strongly influences the trission of JE. Hence, we were motivated to condu
ystematic literature review to identify published with an emphasis on the relationship between ric
igation and JE. We searched Biosis previews, Oechnologies, Medline and the Web of Sciencelying the following keywords: “Japanese enceph
is” and “water”, or “rice”, or “irrigation”, or “rice ir-igation and agriculture”, or “paddy”, or “field”. Pers published in English, French or German wereluded. We also considered manuscripts written inese, Japanese or Korean, if an abstract of thesers was accessible in English on the above-mentiatabases. The key findings of our literature reviewummarized here.
CT
PR
OO
F
8
J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx 5
Four studies in India analyzed the presence of rice240
irrigation and vector abundance in relation to the inci-241
dence of JE. In the Gorakhpur district in Uttar Pradesh,242
and the Mandya district of Karnataka, areas extensively243
developed for irrigated rice agriculture, the occurrence244
of JE was closely associated with high vector densi-245
ties, breeding in the fields or the canal system. The246
highest numbers of JE cases were observed shortly af-247
ter the mosquito densities peaked (Mishra et al., 1984;248
Kanojia et al., 2003). In addition, in the Mandya dis-249
trict, a high incidence of JE was found in extensively ir-250
rigated areas, while few cases occurred in villages with251
less irrigation or no irrigation systems (Geevarghese252
et al., 1994). In Assam, 78.6% of the JE cases occurred253
in families practicing rice cultivation (Phukan et al.,254
2004).255
Based on an epidemiological, serological and clin-256
ical study of 54 JE patients in the Northern Philip-257
pines, 41 cases (76%) were associated with irrigated258
rice fields and only one patient (2%) could not be linked259
to rice irrigation (Barzaga, 1989). In Taiwan, JE was260
monitored over 4 years and the highest numbers of261
cases were observed in the counties with the highest262
number of rice paddies (e.g. in Ilan county in 1969 a263
morbidity rate of 5.5 per 100,000 inhabitants). Fewer264
cases occurred in areas where dry farming had been265
adopted (e.g. in Yunlin county in 1969 a morbidity rate266
of 0.64 per 100,000 inhabitants) and no cases were re-267
ported in non-rice cultivated provinces (Okuno et al.,268
1975).269
ka,270
a han271
4 d out-272
b aths.273
T us-274
p high-275
e iga-276
t rted277
f ,278
2279
4280
k of281
J JE282
e CDC,283
2 is/284
risk-table.htm). The countries were grouped into differ-285
ent epidemiological sub-regions of the world, accord-286
ing to recent classifications of WHO, which is based287
primarily on child and adult mortality rates (WHO, 288
2004). The JE-endemic countries are located in seven289
of the 14 WHO sub-regions (Table 1). 290
Though sporadic JE cases have been reported291
from peri-urban areas, disease occurrence is largely292
restricted to rural settings (Self et al., 1973; Solomon 293
et al., 2000). Hence, inTable 1 we show data on 294
the rural population for the individual JE-endemic295
countries only (United Nations, 2004). For those 296
countries, where JE is endemic only in certain parts297
of the rural areas (e.g. in the Russian Federation JE298
outbreaks have been observed only in far Eastern299
maritime areas South of Chabarovsk (CDC, 2004;300
http://www.cdc.gov/ncidod/dvbid/jencephalitis/risk- 301
table.htm), we determined the rural endemic popula-302
tion of these areas consulting Encarta Encyclopedia303
(Microsoft Corporation, 2004). 304
We find that approximately 1.9 billion people cur-305
rently live in rural JE-prone areas of the world, the306
majority of them in China (766 million) and India (646307
million) (Tables 1 and 2). This figure is in line with a 308
recent “population at risk estimate”, quoting a popu-309
lation of 2 billion, 700 million of these children under310
the age of 15 years living in JE-endemic areas (Tsai, 311
2000). 312
As discussed above surface irrigation is an impor-313
tant risk factor in the epidemiology of JE. Conse-314
q ir-315
r ar-316
e rom317
t the318
r ir-319
r ar-320
e nal321
i 322
b n’s323
p rea,324
h area325
a rone326
a emic327
i esti-328
m op-329
u 330
t ies,331
r 332
UN
CO
RR
E
In 1985–1986 in the Mahaweli System H, Sri Lann epidemic of JE occurred, resulting in more t00 cases and 76 deaths. In 1987–1988 a seconreak took place with >760 cases and 138 dehe promotion of smallholder pig husbandry was sected to be responsible for these outbreaks. Thest number of cases occurred in areas with irr
ion and pig husbandry, while no cases were reporom non-irrigated areas with few pigs (Amerasinghe003).
. Population at risk
In order to estimate the current population at risE, we first compiled a list of all countries wherepidemics or sporadic cases have been reported (004;http://www.cdc.gov/ncidod/dvbid/jencephalit
ED
ACTROP 1666 1–1
uently, inTable 2we present the size of the totaligated and the rice-irrigated land in JE-endemicas, together with data on populations at risk f
hese areas and estimated DALYs, stratified byelevant WHO sub-regions. The numbers on theigated and the rice-irrigated areas in JE-proneas were calculated by multiplying the total natio
rrigated and rice-irrigated area (http://www.fao.org)y the “endemic fraction” (e.g. 17.7% of Pakistaopulation is estimated to live in the endemic aence we assume that 17.7% of both the totalnd the rice-irrigated areas are located in JE-preas). The sizes of the populations in JE-end
rrigated areas and JE-endemic rice areas wereated by multiplying the average national rural plation densities (United Nations, 2004) by the to-
al endemic area under irrigation and rice paddespectively.
UN
CO
RR
EC
TED
PR
OO
F
ACTROP 1666 1–18
6 J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx
Table 1Rural population/population in irrigated areas in JE-endemic countries stratified by relevant WHO sub-regions of the world (all numbers inthousand; n.d.: no data currently available)
The AmericasWHO sub-region 3
United States of America: Guam (150/n.d.) and Saipan (40/n.d.)
Eastern MediterraneanWHO sub-region 7
Pakistana (16,750/4,001)
EuropeWHO sub-region 10
Russian Federationb (614/n.d.)
South-East AsiaWHO sub-region 11
Indonesia (119,589/3033), Sri Lanka (15,057/1461), Thailand (42,796/4114)
WHO sub-region 12Bangladeshc (111,165/28,958), Bhutanc (2,065/17), Democratic People’s Republic of Korea (8,818/1,066), Indiad
(646,100/100,873), Myanmar (34,927/809), Nepale (12,500/329), Timor-Leste (719/51)
Western-PacificWHO sub-region 13
Australiaf (15/n.d.), Brunei Darussalam (85/<1), Japang (25,113/2376), Singapore (<100/n.d.)
WHO sub-region 14Cambodia (11,514/172), Chinah (766,757/24,796), Lao People’s Democratic Republic (4489/30), Malaysia (8814/99), PapuaNew Guinea (4958/n.d.), Philipines (31,182/1612), Republic of Korea (9,395/1081), Viet Nam (60,441/5460)
a Areas around Karachi and in the Lower Indus Valley, province of Sind (CDC, 2004,http://www.cdc.gov/ncidod/dvbid/jencephalitis/risk-table.htm); population estimates taken from (United Nations, 2004) and (Microsoft Corporation, 2004).
b Far Eastern maritime areas South of Chabarovsk (administrative district of Primorskij Kraij) (CDC, 2004,http://www.cdc.gov/ncidod/dvbid/jencephalitis/risk-table.htm); population estimates taken from (United Nations, 2004) and (Microsoft Corporation, 2004).
c Bangladesh and Bhutan are potential JE-endemic countries, but due to lack of data the situation has to be clarified (CDC, 2004,http://www.cdc.gov/ncidod/dvbid/jencephalitis/risk-table.htm).
d Reported cases from all states except Arunachal, Dadra, Daman, Diu, Gujarat, Himachal, Jammu, Kashmir, Lakshadweep, Meghalaya, NagarHaveli, Orissa, Punjab, Rajasthan and Sikkim (CDC, 2004,http://www.cdc.gov/ncidod/dvbid/jencephalitis/risk-table.htm); population estimatestaken from (United Nations, 2004) and (Microsoft Corporation, 2004).
e Hyperendemic in Southern Terai lowlands (BBIN, 2004).f Outbreaks on the Islands of Torres Strait and on mainland Australia at Cape York Peninsula (Hanna et al., 1996); at-risk to JE assumed for
all inhabitants at Cape York Peninsula; population estimates from (United Nations, 2004) and (Microsoft Corporation, 2004).g Rare-sporadic cases on all islands except Hokkaido; population estimates taken from (United Nations, 2004) and (Microsoft Corporation,
2004).h Cases in all provinces except Xizang (Tibet), Xinjiang, Qinghai (CDC, 2004,http://www.cdc.gov/ncidod/dvbid/jencephalitis/risk-table.htm);
population estimates taken from (United Nations, 2004) and (Microsoft Corporation, 2004).
According to our estimates 1,025,000–1,080,000333
km2 land is irrigated in JE-prone areas. We find that334
currently 180–220 million people are living in prox-335
imity to irrigation or rice-irrigation schemes in the336
JE-endemic regions, and thus are at risk of con-337
tracting the disease (Table 2). Irrigated agriculture is338
most pronounced in WHO sub-region 12. Accord-339
ing to our estimates 132–167 million people live in340
JE-endemic irrigated areas in this WHO sub-region341
(Table 2). 342
The estimated global burden of JE in 2002 was343
709,000 DALYs. WHO sub-regions 12 and 14 currently344
bear 84% of this burden (597,000 DALYs). The remain-345
ing 16% are thought to occur in WHO sub-regions 7
UNCORRECTED PROOF
AC
TR
J.Keise
reta
l./Acta
Tropica
xxx(2005)xxx–
xxx
Table 2Burden of JE and population at risk in endemic areas and living in close proximity to irrigation, stratified by relevant WHO sub-regions
WHOsub-region
DALYs lostdue to JE in2002a
Total area(km2)b
TotalJE-endemicareas (km2)c
Irrigated land inJE-endemic areasin 2002 (km2)d
Rice paddies inJE-endemic areasin 2003 (km2)d
Total populationin 2003 (×103)e
Rural endemicpopulation(×103)f
Population inJE-endemic irrigatedareas (×103)g
Population inJE-endemic riceareas (×103)g
3 n.d. 663 663 (100%) n.d. n.d. 1905 190 (100%) n.d. n.d.7 83,000 796,095 140,914 (17.7%) 31,507 (4.0%) 3912 (0.5%) 153,578 16,750 (10.9%) 4,001 (2.6%) 497 (0.3%)
10 n.d. 17,075,400 165,760 (1.0%) n.d. n.d. 143,246 614 (0.4%) n.d. n.d.11 29,000 2,483,300 2,483,300 (100%) 104,100 (4.2%) 233,888 (9.4%) 301,782 177,442 (58.8%) 8,609 (2.9%) 18,44812 277,000 4,437,432 3,401,672 (76.7%) 506,195 (11.4%) 521,384 (11.7%) 1,312,548 816,294 (62.2%) 132,103 (10.1%) 167,647 (12.8%)13 n.d. 8,125,410 330,670 (4.1%) 20,317 (0.25%)h 12,971 (0.16%)h 151,996 25,313 (16.6%) 2,376 (1.6%)h 1,517 (1.0%)h
14 320,000 11,539,430 6,902,490 (59.8%) 363,520 (3.2%)i 308,844 (2.7%) 1,560,475 897,550 (57.5%) 33,249 (2.1%)i 32,749 (2.1%)
Total 709,000 44,457,730 13,425,469 (30.1%) 1,025,639 (2.3%)h,i 1,080,999 (2.4%)h 3,623,815 1,934,154 (53.3%) 180,338 (4.9%)h,i 220,859 (6.1%)h
n.d.: no data.a WHO (2004).b Food and Agriculture Organization 2004,http://www.fao.org/waicent/portal/statisticsen.asp, Rome.c Estimated with aid of CDC, 2004 (http://www.cdc.gov/ncidod/dvbid/jencephalitis/risk-table.htm) andMicrosoft Corporation (2004).d Data for the whole country obtained from the Food and Agriculture Organization 2004 and multiplication by the endemic fraction.e United Nations (2004).f Microsoft Corporation (2004).g The size of the endemic irrigated population and endemic population in rice areas was estimated by multiplying the average national rural populationdensities (United Nations,
2004) by the total area under irrigation/of rice paddies in JE-endemic areas.h Omitting Singapore and Australia.i Omitting Papua New Guinea.
O
P16661–18 7
T P
RO
OF
8
8 J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx
and 11, whereas no information is currently available346
for WHO sub-regions 3, 10 and 13.347
5. Trend of rice agriculture in endemic WHO348
sub-regions349
We compiled data for the relevant WHO sub-regions350
on the rice harvested area and rice production between351
1963 and 2003, which is available athttp://faostat.352
fao . org / faostat / form?collection = Production .Crops.353
Primary&Domain=Production&servlet=1&hasbulk=0354
&version=ext&language=EN. No distinction could355
be made between rain-fed rice and irrigated rice and356
no information was available on the intensity of the357
irrigated rice production (single, double or triple358
cropping).359
No data was obtainable for the JE-endemic coun-360
tries of WHO sub-regions 3 (Guam and Saipan) and361
Singapore. We excluded data on the rice harvested area362
and rice production in the Russian Federation, because363
the current size of the JE-endemic area there are small364
compared to the rest of the JE-endemic countries. We365
furthermore excluded Australia as outbreaks have so366
far only be reported Torres Strait Islands and on main-367
land Australia at Cape York Peninsula (Hanna et al.,368
1996).369
As mentioned above rice agriculture accounted for370
up to 1,080,000 km2 in all JE-prone areas in 2003. In371
the same year the total rice harvested area of these372
c was373
1 er374
t rice-375
i in-376
v area377
i378
2 ted379
a apan,380
f lved381
(382
w over383
t sub-384
r ficant385
p 12,386
a ppar-387
e rew388
u ars.389
M d in390
WHO sub-region 13. And finally, in WHO sub-region391
14, a substantial expansion of the rice-irrigated area has392
occurred between 1963 and 1973, but has decreased393
thereafter. 394
The total rice production has risen from 226 mil-395
lions of tonnes in 1963 to 529 millions of tonnes in396
2003 (+134%). The most significant increases of rice397
production occurred in WHO sub-regions 7, 11, 12 and398
14, as shown inFig. 2. On the other hand, in WHO sub-399
region 13 (mainly Japan) a reduction from 16.6 to 9.7400
millions of tonnes (−43%) has taken place. 401
We also depict inFig. 2the change of the rural pop- 402
ulation of the five most relevant WHO sub-regions,403
which overall increased from 1325 million inhabitants404
in 1963 to 2197 million inhabitants in 2003 (+66%). In405
the past decade a decrease in the rural population has406
been observed in WHO sub-regions 11 and 14. 407
6. Intervention strategies in rice fields 408
Our aim was to systematically review the litera-409
ture to identify published work on biological control410
strategies againstCx. tritaeniorhynchuslarvae in rice 411
fields. We did not include studies on the application of412
synthetic larvicides and insecticides againstCx. tritae- 413
niorhynchusin rice fields, as the use of chemical control414
was found to be of no operational value due to the short415
activity of these products, high costs and resistance de-416
velopment (Wada, 1988). We searched the same elec-417
t e 418
f 419
t t 420
a 421
l , 422
o r- 423
v s”,424
o 425
6 426
6 427
es428
i s, in-429
c of ir-430
r and431
d can432
b ch-433
UN
CO
RR
EC
ountries, including the non-JE-endemic regions,,345,000 million km2. This is an increase of 22% ov
he past 40 years. In some of these countries therrigated area has almost doubled over the periodestigated. For example, in Pakistan the irrigatedncreased from 12,861 km2 in 1963 to 22,100 km2 in003 (+72%). Marked declines in the rice-irrigarea have been observed in other countries. In J
or example, the rice-irrigated area has been ha32,760 km2 in 1963 to 16,650 km2 in 2003). InFig. 2e present the changes of the rice-irrigated area
he past 4 decades for 5 of the 7 relevant WHOegions, where JE has been reported to be a signiublic health problem. In WHO sub-regions 11 andsubstantial increase in the rice-irrigated area is ant. In WHO sub-region 7 (Pakistan) the rice area gntil 1993, but remained stable over the past 10 yeeanwhile, the rice-irrigated area has decrease
ED
ACTROP 1666 1–1
ronic databases as mentioned in Section3and used thollowing keywords: “Japanese encephalitis” or “Culexritaeniorhynchus” in combination with “alternate wend dry irrigation”, or “Azolla”, or “Bacillus”, or “bio-
ogical control”, or “bacteria”, or “control”, or “fungi”r “intermittent irrigation”, or “invertebrates”, or “laivorous fish”, or “natural products”, or “nematoder “predator”, or “water management”.
.1. Environmental control strategies
.1.1. Alternate wet and dry irrigation (AWDI)Traditionally, rice fields are flooded, which provid
deal breeding places for several mosquito specieluding those that transmit JE. The managementigation water that leads to the alternate wettingrying of fields, including the canals and ditches,e active or passive. An important feature of this te
UN
CO
RR
EC
TED
PR
OO
F
ACTROP 1666 1–18
J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx 9
Fig. 2. Changes of rice growing area, rice production and rural population at risk of JE in 5 WHO sub-regions between 1963 and 2003.
TED
PR
OO
F
ACTROP 1666 1–18
10 J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx
nique is that the soil can dry out, which in turn curtails434
the life cycle development of the mosquito from lar-435
vae and pupae to adult. In order to achieve a significant436
reduction of mosquito larvae, AWDI (also termed inter-437
mittent irrigation) has to be applied during the entire438
cropping season and should cover all rice fields that439
are connected by irrigation canals over a large area.440
This method is particularly feasible in places where441
control of the water supply and drainage is possible,442
hence where land and climatic conditions are suitable443
(Mogi, 1988). Growing water shortages in many areas444
create an incentive to better control irrigation water445
management. AWDI is one such strategy. The poten-446
tial of AWDI is, however, limited in areas where there447
is a threat of insufficient resources to re-flood the fields448
and where farmers perceive a risk of reduced yields by449
letting their fields dry out.450
The effects of AWDI on the abundance ofAnophe-451
les, which are the vector species of malaria, rice yield,452
water consumption and methane emission have been453
reviewed recently (van der Hoek et al., 2001; Keiser454
et al., 2002). In brief, this method had been introduced455
in Asia about 300 years ago, primarily to obtain higher456
rice yields. The first study of AWDI as a potential tool457
to reduce mosquito vectors was conducted in Bulgaria458
on mosquitoes of theAn. maculipenniscomplex in the459
1920s (Keiser et al., 2002). AWDI was compulsory in460
Portugal in the 1930s, when malaria was still a problem461
there.462
We found four studies analyzing the effect of AWDI463
o um-464
m465
m ply-466
i DI467
o h468
d ined469
i ere470
o od471
h ted.472
6473
6474
c-475
t -476
l re-477
s es478
h inst479 Tabl
e3
Alte
rnat
ew
etan
ddr
yirr
igat
ion
(AW
DI)
agai
nstCulextrita
eniorhyn
chusla
rvae
inric
efie
lds
Stu
dysi
te;r
efer
ence
Spe
cific
ityof
AW
DI
Out
com
e
C.tritaeniorhyn
chus
imm
atur
esC.tritaeniorhyn
chus
adul
tsR
ice
yiel
dW
ater
cons
umpt
ion
Hen
an,C
hina
(Lu-B
aulin
,198
8)Ir
rigat
ion
inte
rval
:5da
ysD
ecre
ase:
81–9
1%D
ecre
ase:
55–7
0%In
crea
se:1
3%D
ecre
ase:
50%
Tam
ilN
adu,
Indi
a(R
ajen
dran
etal
.,19
95)Ir
rigat
ion
inte
rval
:3–5
days
Dec
reas
e:75
–88%
Not
know
nIn
crea
se:4
%N
otkn
own
Kin
ryu,
Japa
n(M
ogi,
1993
)Ir
rigat
ion
inte
rval
:Sev
eral
days
Dec
reas
eN
otkn
own
Not
know
nN
otkn
own
Tsu
City
,Jap
an(Ta
kagi
etal
.,19
95)M
idse
ason
dryi
ngD
ecre
ase
offo
urth
inst
ars:
14.3
–48.
2%N
otkn
own
Not
know
nN
otkn
own
UN
CO
RR
EC
n densities of JE vectors and the key findings are sarized inTable 3. Overall,Cx. tritaeniorhynchusim-atures were reduced by 14–91% in rice fields ap
ng AWDI. One study investigated the effect of AWn theCx. tritaeniorhynchusadult population, whicecreased by 55–70%. The crop yield was exam
n two trials and increases between 4 and 13% wbserved in AWDI rice fields. The effect this methas on the incidence of JE remains to be investiga
.2. Biological control strategies
.2.1. BacteriaThe larvicidal activity of the spore forming ba
eria Bacillus sphaericusandB. thuringiensis israeensis have been discovered in 1965 and 1976,pectively (Mittal, 2003). Subsequently, these biocidave been evaluated in various formulations aga
CO
RR
EC
TED
PR
OO
F
ACTROP 1666 1–18
J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx 11
Table 4Application ofBacillusspp. againstCulex tritaeniorhynchuslarvae in rice fields
Study site; period (if known);reference
Intervention strategy Outcome
India, 1983–1984 (Kramer, 1984) Application ofBacillus sphaericusandBacillus thuringiensisH-14 inrice fields
91–99% reduction with doses of 0.5–1.5 kg/haCulextritaeniorhynchusandAnopheles subpictus, howeveractivity did not subsist beyond a few days
India, 1982 (Balaraman et al., 1983) Application ofBacillus thuringiensisH-14 in rice fields
100% reduction with doses of 27× 105 spores/ml, but firstand second instars reappeared 3 days after application
India (Sundararaj and Reuben, 1991) Application of microgel dropletformulation ofBacillus sphaericusinrice fields
44–79% reduction of early instar and 82–100% reduction oflate instar ofCulex tritaeniorhynchus,Cx. vishnuiandCx.pseudovishnuifor at least 5 weeks applying a dose of4.3 kg/ha
Korea (Rhee et al., 1983) Application ofBacillus thuringiensisisraelensisH-14 in rice field
95–98% reduction ofCulex tritaeniorhynchus. The residualeffects lasted only 24 h at doses of 0.8–1.2 l/ha ofBacillusthuringiensis israelensisH-14, SAN 402 I suspensionconcentrate containing 600 IU/mg
Rourkela city, India, June 1993 toOctober 1994 (Yadav et al., 1997)
Application ofBacillus sphaericus(strain B-101, serotype H5a, 5b) inrice field
Application ofB. sphaericus, when sprayed at 1 g/m2
significantly reduced larval and pupal counts (P< 0.0001) inrice fields. Duration of effect could not be determined asfields received periodicallyBacillus sphaericustreatedwastewater
mosquito vectors worldwide (Lacey and Lacey, 1990).480
B. thuringiensisproduces four key insecticidal pro-481
teins, whileB. sphaericusproduces a single binary482
toxin (Federici et al., 2003). These protein toxins bind483
to cells of the gastric caecum and posterior midgut of484
the mosquitoes causing intoxication, which eventually485
leads to death (Mittal, 2003). The advantages of these486
bacterial insecticides in terms of efficacy, specificity487
and environmental safety are well documented (Zahiri488
et al., 2004). A major drawback of these products,489
however, is their high cost of production, the labor-490
intensive delivery, as well as first reports of resistance491
(Sundararaj and Reuben, 1991; Federici et al., 2003).492
The effect of the individual biolarvicides depends on493
water temperature, pH, aquatic vegetation, the formula-494
tion applied, type of habitat, and the target mosquitoes495
(Mittal, 2003).B. sphaericus, for example, is known to496
exhibit a high activity againstCulexmosquitoes, while497
certainAedesspecies are not affected (Mittal, 2003;498
Zahiri et al., 2004). Hence, no general conclusion499
can be drawn about whetherB. sphaericusand B.500
thuringiensis israelensisformulations are suitable for501
JE vector control in rice fields. The results on the502
application of these bacterial formulations in rice503
fields againstCx. tritaeniorhynchusare summarized504
in Table 4. In Tamil Nadu, India, the application505
of 4.3 kg/ha of a microgel droplet formulation of506
B. sphaericus1593 M resulted in a 44–79% reduc-507
tion of early instar and 82–100% reduction of late508
instar culicinae larvae (C. fuscanus, C. pseudovish- 509
nui and C. tritaeniorhynchus) for at least 5 weeks 510
(Sundararaj and Reuben, 1991). Similarly, up to 511
95–98% ofC. tritaeniorhynchuslarvae were reduced 512
in three other field sites evaluatingB. sphaericusor 513
B. thuringiensisformulations (Balaraman et al., 1983; 514
Rhee et al., 1983; Kramer, 1984). However, the larvi- 515
cidal activity did not persist in these rice fields beyond516
a couple of days; in the Republic of Korea the residual517
effect ofB. thuringiensisH-14 was found to last only 518
24 h (Rhee et al., 1983). 519
6.2.2. Nematodes 520
Romanomermis culicivoraxis probably the best- 521
studied nematode parasite of mosquitoes. The prepar-522
asitic nemas are applied to the fields, where they locate523
a host, penetrate the cuticle and develop within the524
mosquito larvae (Lacey and Lacey, 1990). Although 525
this method has not been widely studied, we identified526
three publications assessing the effect of nematodes on527
JE vectors, which are summarized inTable 5. In Tai- 528
wan, application of 7000 nematodes per m2 rice field 529
yielded a reduction of 11–18%C. tritaeniorhynchus 530
(Lacey and Lacey, 1990). In two studies in China the 531
nematodeR. yunanensishas been distributed in rice 532
UN
EC
TED
PR
OO
F
8
12 J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx
Table 5Application of nematodes againstCulex tritaeniorhynchuslarvae in rice fields
Study site; period (if known); reference Intervention strategy Outcome
China, 1986–1995 (Peng et al., 1998) Predation efficacy ofRomamomermisyunanensisin rice fields
2000–4000 nematodes/m2 yielded parasitism rates of52.2–96.7% forCulex tritaeniorhynchus
China (Song and Peng, 1996) Predation efficacy ofRomamomermisyunanensisand a secondRomamomermisspecies in rice fields
1000–3000 nematodes/m2 resulted in parasitism rates of60.8–95.5% ofCulex tritaeniorhynchusandCx.quinquefasciatusandAnophelesspp.
Taiwan, 1974 (Lacey and Lacey, 1990;Mitchell et al., 1974)
Predation efficacy ofRomamomermisculicivorax in rice fields
Larvae reduction of 11–18% ofCulex tritaeniorhynchus
fields (2000–4000 nematodes/m2), which resulted in533
parasitism rates of 52.2–96.7% and 60.8–95.5%, re-534
spectively, ofC. tritaeniorhynchuslarvae (Song and535
Peng, 1996; Peng et al., 1998).536
6.2.3. Invertebrate predators537
Invertebrate predators, i.e. Coleoptera, Hemiptera538
or Odonata, though less common than the use of fish,539
are also known to substantially reduce mosquito lar-540
val populations in rice fields (Lacey and Lacey, 1990).541
However, they are highly sensitive to temperature, pres-542
ence of vertebrates, growth of rice and chemical pollu-543
tants (Lacey and Lacey, 1990). In India, the presence of544
notonectids was negatively associated with larval abun-545
dance ofCx. pseudovishnui,Cx. tritaeniorhynchusand546
Cx. vishnui(Sunish and Reuben, 2002).547
6.2.4. Larvivorous fish548
The use of larvivorous fish to reduce mosquitoes has549
a history of over 100 years (Lacey and Lacey, 1990).550
The mosquitofish,Gambusia affinis,is the most widely551
used predator. Other fish species includeTilapia spp.,552
Poecilia reticulataor Cyprinidae (Lacey and Lacey,553
1990). A detailed summary of the most common preda-554
cious fish is given in (Lacey and Lacey, 1990).555
After stocking rice fields with 1–10 natural preda-556
tor fishs per m2, larval populations ofCx. tritae-557
niorhynchuswere reduced by 55.2–87.8% (Table 6).558
Larvivorous fish cannot be applied in rice fields, where559
i ted560
t by561
t ides562
o563
a iting564
m re-565
s fish566
( -567
duction of exotic predators such asGambusiamight 568
displace the native fish populations to reduce their nat-569
ural value, as being observed in Japanese rice fields570
(Wada, 1988). Therefore, the compatibility of the cho-571
sen fish with local fauna and flora is of high importance572
(Lacey and Lacey, 1990). 573
6.2.5. Fungi 574
Fungi that have been studied extensively for their575
potential as biological mosquito control agents include576
Coelomomycesspp. andLagenidium giganteum. The 577
former have been investigated how they impact the578
development of JE vectors in China. Field observa-579
tions there showed a strong effect of the fungusCoelo- 580
momyces indicaon Cx.tritaeniorhynchus, as infected 581
larvae were unable to develop into adults (Liu and Hsu, 582
1982). However, fungi have not been applied for bio-583
logical control of JE vectors on a large scale so far, as584
practical problems, for example their production, have585
yet to be solved (Lacey and Lacey, 1990). 586
6.2.6. Other natural products 587
Natural products might also have high potential for588
reducing the proliferation of culicine mosquitoes in589
rice fields. Our literature search on the control of JE590
vectors in rice fields yielded in-depth investigation of591
only two natural products. In Tamil Nadu, India ap-592
plication of the floating water fernAzolla microphylla 593
greatly reduced immature mosquito populations. How-594
e 595
d - 596
c tion,597
l rol598
a 599
d or600
s 601
i t a602
UN
CO
RRrregular irrigation is practiced. It should also be no
hat predator populations are strongly influencedemperature, rice growth, vegetation, use of pesticr chemical pollutants (Lacey and Lacey, 1990). Inddition, recent research has shown that ovipososquitoes may move to other breeding sites in
ponse to the stocking of rice fields with predatoryAngelon and Petranka, 2002). Furthermore, the intro
ACTROP 1666 1–1
ver, the infestation of the rice field withAzollawasifficult to achieve, 80% coverage byAzollawas acomplished only 13–14 days after rice transplantaimiting its wider use as a biological mosquito contgent (Rajendran and Reuben, 1991).
Neem cake powder, made from freshly harvestetored neem kernels, from the neem tree (Azadirachtandica) rich in the active principle azadirachtin. A
UN
CO
RR
EC
TED
PR
OO
F
ACTROP 1666 1–18
J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx 13
Table 6Application of fish predators againstCulex tritaeniorhynchuslarvae in rice fields
Study site; period (if known); reference Intervention strategy Outcome
Bulkyo, Bosong-gun, Chollanamdo,Korea 1995–1996 (Lee, 1998)
Presence ofMisgurnus mizolepisinrice fields
Coefficients of correlation betweenMisgurnusmizolepisand abundance of mosquito larvae showednegative correlations inAnopheles sinensis(−0.66) andCulex tritaeniorhynchus(−0.47)
Suwon, 1992–1993 (Kim et al., 1994) Moroco oxycephalusandMisgurnusanguillicaudatusin rice fields
4 fish/m2 gave a reduction of 55.2% of immatures ofAnopheles sinensis, Culex pipiens pallensandCulextritaeniorhynchuscompared to control field after 8weeks
South Delhi, May–June 1980 (Mathuret al., 1981)
Presence ofGambusia affinisin ricefields
10 fish/m2 gave a mean reduction of 63.9% ofimmatures ofCulex tritaeniorhynchusin 4 weeks
China (Liao et al., 1991) Predation efficacy ofCtenopharyngdon idella
78 days after treatment with 1.9 fish/m2 a reduction of80%Culex tritaeniorhynchuslarvae was observed
India, June to October 1991 (Prasadet al., 1993)
Predation efficacy ofGambusiaaffinis in rice fields
87.8% mosquito (6 Anopheline and 4 Culicine) larvalcontrol in well submerged rice fields, no effect ofGambusia affinisin rice fields applying AWDI
Quanzhou county, China, 1988–1989(Wu et al., 1991)
Predation efficacy ofCypriniuscarpio,Ctenopharyngdon idellaandTilopia spp.
Population size of larvae and adultCulextritaeniorhynchusdecreased
South Korea, 1979 and 1981 (Yu et al.,1982)
Predation efficacy ofAphyocyprischinensis
Fish stocking of 1.5/m2 in Wondang-Ni rice paddyresulted in mosquito larval reduction of 98.9% in thethird week after fish introduction againstAnophelessinensis, Culex pipiens, Culex crientalisandCulextritaeniorhynchus
Jindo Island, Chollanam-do Province,South Korea, 1989 (Yu and Kim,1993)
Predation efficacy ofAphyocyprischinensisin combination withBacillus thuringiensis(H-14)
Aphyocypris chinensis(1.5 fish/m2) achieved areduction of 60.8-77.0% against both,AnophelessinensisandCulex tritaeniorhynchusin the first 2month. In the third monthBacillus thuringiensis(H-14)treatment was made, which yielded a satisfactory degreeof control maintained above 93.1% for 2 weeks
dosage of 500 kg/ha it yielded an 81% reduction of late603
instar culicine larvae and a 43% reduction in the pupal604
production (Rao et al., 1992). Stable product neem cake605
coated urea and the combination of neem cake coated606
urea and water management practices (i.e. AWDI) led607
to a 70–95% reduction of culicine immatures. The ben-608
efits were also found to be cost-effective as neem treat-609
ment gave higher values for mean number of produc-610
tive tillers, plant height and mean number of grains per611
panicle (Rao et al., 1995).612
7. Discussion and conclusion613
Water resource development and management, in614
particular the construction and operation of small and615
large dams and irrigation schemes, occurred at an enor-616
mous pace over the past 50 years (Gujja and Perrin,617
1999). The effects of these water projects and accom-618
panying ecological transformations are manifold, and619
inherently difficult to assess and quantify. Negative ef-620
fects include increased frequencies and transmission621
dynamics of water-based (e.g. schistosomiasis) and622
water-related vector-borne diseases, including malaria623
(Keiser et al., in press), lymphatic filariasis (Erlanger 624
et al., in press) and JE. On the other hand, such water625
projects are key for hydroelectric power production and626
food security; hence they can stimulate social and eco-627
nomic development. In this review, we have focused628
on irrigation schemes, with particular emphasis on rice629
growing. Currently, there is a paucity on high quality630
data on the effects of large and small dams on JE. 631
Over the last four decades rice production has ex-632
panded considerably in most countries where JE is cur-633
rently endemic. This growth will most likely continue634
for food security reasons. We find that in 2003 approx-635
imately 180–220 million people live near irrigation636
schemes in JE-prone countries, hence have a poten-637
T P
RO
OF
8
14 J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx
tial risk of contracting JE. These figures are unavoid-638
ably subject to a level of uncertainty. The possibility639
that we have overvalued the population at risk cannot640
be ruled out, which is justified on four grounds. First,641
pigs, which are the amplifying hosts, are not present642
in all rural irrigated sites. Second, the disease affects643
mainly children under the age of 15 years. This segment644
of the rural population in Asia accounts for approx-645
imately one-third of the total rural Asian population646
(Tsai, 2000). Third, it is presently not possible to pre-647
cisely map the population of irrigated areas in partially648
JE-endemic countries. Forth, no data are available on649
vaccination coverage in rural irrigated areas. However,650
it is also conceivable that we might have underesti-651
mated the actual population at risk of JE due to irriga-652
tion, as we have applied a rather conservative proxy,653
namely the national rural population density. Irrigated654
areas, however, are often characterized by a large pop-655
ulation density as they attract a great number of people656
(Keiser et al., in press). In addition, due to the predicted657
expansion of irrigation schemes, the rural population in658
JE-endemic areas living in proximity to irrigated rice659
agro ecosystems will further rise in the years to come.660
Unfortunately and in contrast to malaria (Keiser661
et al., in press), very few studies are presently available662
that investigated the effect of an introduction or an ex-663
pansion of rice-irrigation schemes on the incidence of664
JE. The few studies reviewed here clearly link JE inci-665
dence with rice agriculture. There are important gaps666
in our understanding of the effects of different types of667
w mis-668
s ies669
t d JE670
i ater671
p ent672
s urce673
d674
ice675
f sia,676
a cam-677
p lin-678
i ama679
o .g.680
J by681
s ieve682
1 one683
b quent684
b685
side effects occur in 10–30% and allergic reactions oc-686
cur in 0.6% of vaccinated adults (Jones, 2004). Fur- 687
thermore, clinically significant neurological adverse688
events, such as a meningoencephalitis, have been re-689
ported. In addition, these vaccines are expensive and690
production can hardly keep up with demand (Jones, 691
2004). In China a live attenuated vaccine (SA14-14-692
2) was developed in 1988 and the efficacy of a single693
dose was estimated to be 99.3% (Bista et al., 2001). 694
However, its use in other affected countries has been695
restricted by regulatory concerns over manufacturing696
and control (Solomon et al., 2003). 697
There is ongoing research to develop improved JE698
vaccines. These efforts received a boost in December699
2003, when the Bill and Melinda Gates Foundation700
donated US$ 27 million to the Children’s Vaccine Pro-701
gram at the Program for Appropriate Technology in702
Health for this purpose (UNO, 2003). For example, a 703
newly developed live attenuated vaccine (ChimeriVax-704
JE) that uses yellow fever 17D as a live vector for the705
envelope genes of SA14-14-2 virus has been tested in706
phase 2 clinical trials and appears to be well tolerated707
(Monath et al., 2003). It is currently undergoing a 2- 708
year clinical study in Australia, to assess the duration709
of immunity and to gain further knowledge on safety710
and immunogenicity (Jones, 2004). 711
It is important to recognize that strategies other than712
vaccination may play an important role for prevention713
and control of JE, particular in rural areas, where vac-714
cination coverage are sometimes low or where there715
i re-716
c ., 717
2 te-718
g ses719
a body720
l be-721
h nets722
( on723
( of724
a they725
h 726
s nted727
a ts to728
r ized729
a 730
ental731
a lds:732
B is 733
UN
CO
RR
EC
ater-related projects on the frequency and transion dynamics of JE. In our view longitudinal studhat track both a change in vector abundance annfectivity rates in humans, for example at large wrojects in the framework of health impact assessmtudies, covering the entire course of a water resoevelopment project, are warranted.
Clearly, vaccination is the current strategy of choor prevention and control of JE outbreaks in And extensive government-supported vaccinationaigns will remain the mainstay of control. Forma
nactivated vaccines, based on wild type Nakayr Beijing-1 strains grown in adult mouse brain (eE-Vax), are licensed for immunization against JEeveral Asian countries. However, in order to ach00% seroconversion, 2–3 primary doses andooster dose after a year are required, with subseoosts every 3–4 years (Pugachev et al., 2003). Minor
ED
ACTROP 1666 1–1
s no history of immunization against JE at all, asently documented for Northeast India (Phukan et al004). Vaccination of the far rural population is straically difficult and costly, in particular as three dore required to achieve adequate neutralizing anti
evels. As demonstrated in China self-protectionaviour, such as sleeping under insecticide-treatedITNs) can significantly reduce the risk of infectiLuo et al., 1994). On the other hand, all JE patientsrecent outbreak in Northeast India reported thatad slept under a bed-net (Phukan et al., 2004). Con-equently, irrigation schemes should be implemend maintained in a way that adverse health effecice growers and residents of the area are minimnd social and economic development improved.
We evaluated and discussed several environmnd biological vector control measures in rice fieacillus sphaericusand B. thuringiensis israelens
CT
PR
OO
F
8
J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx 15
were found to greatly reduce JE mosquito larvae. How-734
ever, it is currently not economical to apply these bac-735
terial toxins in rice fields, as they are costly, labor-736
intensive and often only have a short duration of activ-737
ity. Few studies are available on invertebrates, fungi,738
nematodes or the neem cake powder in rice fields,739
which allows assessing and quantification of their po-740
tential as JE mosquito control agents. Reviewing the741
literature has shown that in settings where the irriga-742
tion water in the rice fields can be managed, AWDI743
has considerable potential to reduce JE vector densi-744
ties. A similar conclusion can be made for the use of745
larvivorous fish. An integrated vector management ap-746
proach with AWDI and the use of larvivorous fish as747
its main components can reduce JE vector populations,748
and hence has the potential to reduce the transmission749
level and the burden of JE. It should be emphasized750
that these intervention strategies must be tailored care-751
fully to a specific setting, which renders it difficult to752
generalize reported experiences and results.753
Consequently, there is a pressing need to implement754
and monitor the performance of well-designed stud-755
ies to further strengthen our understanding of the con-756
textual determinants of environmental and biological757
control methods on vector abundance and clinical out-758
comes of JE in different ecological, epidemiological759
and socio-cultural settings.760
Acknowledgements761
en762
o is of763
t re-764
s par-765
t J.766
K ger767
( wiss768
N ort.769
W is770
m771
R772
A Ku-773
e en-774
26,775
776
Amerasinghe, F.P., 1995. Mosquito vector ecology: a case study from777
Sri Lanka and some thoughts on research issues. In: Tropical778
Diseases, Society and Environment, Proceedings from a WHO-779
TDR/SAREC Research Seminar, SAREC Documentation: Con-780
ference Reports 1995, pp. 135–156. 781
Amerasinghe, F.P., 2003. Irrigation and mosquito-borne diseases. J.782
Parasitol. 89, 14–22. 783
Amerasinghe, F.P., Ariyasena, T.G., 1991. Survey of adult784
mosquitoes (Diptera: Culicidae) during irrigation development785
in the Mahaweli Project Sri Lanka. J. Med. Entomol. 28, 387–786
393. 787
Angelon, K.A., Petranka, J.W., 2002. Chemicals of predatory788
mosquitofish (Gambusia affinis) influence selection of oviposi- 789
tion site by Culex mosquitoes. J. Chem. Ecol. 28, 797–806. 790
Arunachalam, N., Samuel, P.P., Hiriyan, J., Thenmozhi, V., Gajanana,791
A., 2004. Japanese encephalitis in Kerala, South India: can Man-792
sonia (Diptera: Culicidae) play a supplemental role in transmis-793
sion? J. Med. Entomol. 41, 456–461. 794
Balaraman, K., Balasubramanian, M., Jambulingam, P., 1983. Field795
trial of Bacillus thuringiensisH-14 (VCRC B-17) against Culex 796
and Anopheles larvae. Ind. J. Med. Res. 77, 38–43. 797
Barzaga, N.G., 1972–1985. A review of Japanese encephalitis cases798
in the Philippines Southeast Asian. J. Trop. Med. Public Health799
20, 587–592. 800
Bi, P., Tong, S., Donald, K., Parton, K.A., Ni, J., 2003. Climate801
variability and transmission of Japanese encephalitis in Eastern802
China. Vector Borne Zoonotic Dis. 3, 111–115. 803
Bista, M.B., Banerjee, M.K., Shin, S.H., Tandan, J.B., Kim, M.H.,804
Sohn, Y.M., Ohrr, H.C., Tang, J.L., Halstead, S.B., 2001. Efficacy805
of single-dose SA 14-14-2 vaccine against Japanese encephalitis:806
a case control study. Lancet 358, 791–795. 807
Broom, A.K., Smith, D.W., Hall, R.A., Johansen, C.A., Macken-808
zie, J.S., 2003. Arbovirus infections. In: Cook, G., Zumla, A.809
(Eds.), Manson’s Tropical Diseases, 21st ed. Saunders, London,810
pp. 725–764. 811
C hni-812
port813
w of814
aly.815
D 03.816
nese817
81,818
819
E nner,820
ment821
opu-822
823
F J.J.,824
Biol.825
826
G i, T.F.,827
litis in828
di-829
Med.830
831
G tudies832
ndya
UN
CO
RR
E
This work was part of a project entitled “Burdf water-related vector-borne diseases: an analys
he fraction attributable to components of waterources development and management”, whichially funded by the World Health Organization.eiser (Project no. PMPDB-10622) and J. Utzin
Project no. PPOOB-102883) are grateful to the Sational Science Foundation for financial suppe thank Dr. B. Kay for carefully reviewing thanuscript.
eferences
kiba, T., Osaka, K., Tang, S., Nakayama, M., Yamamoto, A.,rane, I., Okabe, N., Umenai, T., 1997. Analysis of Japanescephalitis epidemic in Western Nepal in. Epidemiol. Infect. 181–88.
ED
ACTROP 1666 1–1
onsultative Group on International Agricultural Research, Teccal Advisory Committee, and CGIAR Secretariat, 1998. Reof the Fifth External Programme and Management RevieInternational Rice Research Institute (IRRI), FAO, Rome, It
ing, D., Kilgore, P.E., Clemens, J.D., Wei, L., Zhi-Yi, X., 20Cost-effectiveness of routine immunization to control Japaencephalitis in Shanghai. Chin. Bull. World Health Organ.334–342.
rlanger, T.E., Keiser, J., Castro, M., Bos, R., Singer, B.H., TaM., Utzinger, J., in press. Effect of water resource developand management on lymphatic filariasis, and estimates of plations at risk. Am. J. Trop. Med. Hyg.
ederici, B.A., Park, H.W., Bideshi, D.K., Wirth, M.C., Johnson,2003. Recombinant bacteria for mosquito control. J. Exp.206, 3877–3885.
ajanana, A., Rajendran, R., Samuel, P.P., Thenmozhi, V., TsaKimura-Kuroda, J., Reuben, R., 1997. Japanese encephaSouth Arcot district, Tamil Nadu India: a three-year longitunal study of vector abundance and infection frequency. J.Entomol. 34, 651–659.
eevarghese, G., Mishra, A.C., Jacob, P.G., Bhat, H.R., 1994. Son the mosquito vectors of Japanese encephalitis virus in Ma
T P
RO
OF
8
16 J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx
District, Karnataka India. Southeast Asian J. Trop. Med. Public833
Health 25, 378–382.834
Gujja, B., Perrin, M., 1999. A place for dams in the 21st Century.835
Discussion Paper, World Wildlife Fund, p. 108.836
Halstead, S.B., Jacobson, J., 2003. Japanese encephalitis. Adv. Virus837
Res. 61, 103–138.838
Hanna, J.N., Ritchie, S.A., Phillips, D.A., Shield, J., Bailey, M.C.,839
Mackenzie, J.S., Poidinger, M., McCall, B.J., Mills, P.J., 1995.840
An outbreak of Japanese encephalitis in the Torres Strait Aus-841
tralia. Med. J. Aust. 165, 256–260.842
Heathcote, O.H., 1970. Japanese encephalitis in Sarawak: studies on843
juvenile mosquito populations. Trans. R. Soc. Trop. Med. Hyg.844
64, 483–488.845
Jones, T., 2004. A chimeric live attenuated vaccine against Japanese846
encephalitis. Expert Rev. Vaccines 3, 243–248.847
Kanojia, P.C., Shetty, P.S., Geevarghese, G., 2003. A long-term study848
on vector abundance & seasonal prevalence in relation to the849
occurrence of Japanese encephalitis in Gorakhpur district Uttar850
Pradesh. Indian J. Med. Res. 117, 104–110.851
Keiser, J., Castro, M., Maltese, M., Bos, R., Tanner, M., Singer, B.H.,852
Utzinger, J., in press. The effect of irrigation and large dams on853
the burden of malaria on global and regional scale. Am. J. Trop.854
Med. Hyg.855
Keiser, J., Utzinger, J., Singer, B.H., 2002. The potential of inter-856
mittent irrigation for increasing rice yields, lowering water con-857
sumption, reducing methane emissions, and controlling malaria858
in African rice fields. J. Am. Mosq. Control Assoc. 18, 329–340.859
Kim, H.C., Kim, M.S., Yu, H.S., 1994. Biological control of vector860
mosquitoes by the use of fish predatorsMoroco oxycephalusand861
Misgurnus anguillicaudatusin the laboratory and semi-field rice862
paddy. Korean J. Entomol. 24, 269–284.863
Kramer, V., 1984. Evaluation ofBacillus sphaericus& B. thuringien-864
sisH-14 for mosquito control in rice fields. Indian J. Med. Res.865
80, 642–648.866
Lacey, L.A., Lacey, C.M., 1990. The medical importance of rice-867
ical868
869
L onal870
rice871
872
L eed-873
heng874
875
L a876
g877
878
L l of879
ease880
ent.881
the882
883
L mic884
k of885
sian J.886
887
L Y.,888
ated889
with pyrethroid insecticide and vaccination against Japanese en-890
cephalitis. Trans. R. Soc. Trop. Med. Hyg. 88, 632–634. 891
Mani, T.R., Rao, C.V., Rajendran, R., Devaputra, M., Prasanna, Y.,892
Hanumaiah, Gajanana, A., Reuben, R., 1991a. Surveillance for893
Japanese encephalitis in villages near Madurai, Tamil Nadu, In-894
dia. Trans. R. Soc. Trop. Med. Hyg. 85, 287–291. 895
Mani, T.R., Reuben, R., Akiyama, J., 1991b. Field efficacy of “Mos-896
bar” mosquito repellent soap against vectors ofbancroftian filar- 897
iasisand Japanese encephalitis in Southern India. J. Am. Mosq.898
Control Assoc. 7, 565–568. 899
Mathur, K.K., Rahman, S.J., Wattal, B.L., 1981. Integration of lar-900
vivorous fish and temephos for the control ofCulex tritae- 901
niorhynchusbreeding. J. Commun. Dis. 13, 58–63. 902
Microsoft Corporation, 2004. Encarta Encyclopedia, Redmond,903
USA. 904
Mishra, A.C., Jacob, P.G., Ramanujam, S., Bhat, H.R., Pavri, K.M.,905
1983. Mosquito vectors of Japanese encephalitis epidemic in906
Mandya district (India). Indian J. Med. Res. 80, 377–389. 907
Mitchell, C.J., Chen, P.S., Chapman, H.C., 1974. Exploratory tri-908
als utilizing a mermithid nematode as a control agent for Culex909
mosquitos in Taiwan. Taiwan Yi Xue Hui Za Zhi 73, 241–254. 910
Mittal, P.K., 2003. Biolarvicides in vector control: challenges and911
prospects. J. Vector Borne Dis. 40, 20–32. 912
Mogi, M., 1984. Mosquito problems and their solution in relation to913
paddy rice production. Protect. Ecol. 7, 219–240. 914
Mogi, M., 1988. Water management in rice cultivation and its rela-915
tion to mosquito production in Japan. In: Vector Borne Disease916
Control in Humans Through Rice Agroecosystem Management.917
International Rice Research Institute in collaboration with the918
WHO/FAO/UNEP Panel of Experts, pp. 101–109. 919
Mogi, M., 1993. Effect of intermittent irrigation on mosquitoes920
(Diptera: Culicidae) and larvivorous predators in rice fields. J.921
Med. Entomol. 30, 309–319. 922
Monath, T.P., Guirakhoo, F., Nichols, R., Yoksan, S., Schrader, R.,923
Murphy, C., Blum, P., Woodward, S., McCarthy, K., Mathis, D.,924
ccine925
clini-926
ose927
nacti-928
213–929
930
O 975.931
iwan)932
f case933
934
P L.P.,935
.A.,936
nka:937
atic938
939
P , Q.,940
941
ina.942
943
P halitis944
ublic945
946
UN
CO
RR
EC
land mosquitoes and their control using alternatives to cheminsecticides. J. Am. Mosq. Control Assoc. 2 (Suppl.), 1–93.
ee, D.K., 1998. Effect of two rice culture methods on the seasoccurrence of mosquito larvae and other aquatic animals infields of Southwestern Korea. J. Vector Ecol. 23, 161–170.
iao, S., Xu, B.Z., Chen, C.Y., Liang, Z.P., Zhang, H.F., 1991. Bring grass carp against mosquitoes in rice field. Zhongguo Ji SChong Xue Yu Ji Sheng Chong Bing Za Zhi 9, 219–222.
iu, S.L., Hsu, Y.C., 1982. Effect of the fungusCoelomomyces indicon the viability ofCulex tritaeniorhynchuslarvae. Kun ChonXue Bao 25, 409–412.
u-Baulin, 1988. Environmental management for the controricefield-breeding mosquitoes in China. In: Vector Borne DisControl in Humans Through Rice Agroecosystem ManagemInternational Rice Research Institute in collaboration withWHO/FAO/UNEP Panel of Experts, pp. 111–121.
uo, D., Ying, H., Yao, R., Song, J., Wang, Z., 1995. Socio-econostatus and micro-environmental factors in relation to the risJapanese encephalitis: a case-control study. Southeast ATrop. Med. Public Health 26, 276–279.
uo, D., Zhang, K., Song, J., Yao, R., Huo, H., Liu, B., Li,Wang, Z., 1994. The protective effect of bed nets impregn
ED
ACTROP 1666 1–1
Johnson, C., Bedford, P., 2003. Chimeric live, attenuated vaagainst Japanese encephalitis (ChimeriVax-JE): phase 2cal trials for safety and immunogenicity, effect of vaccine dand schedule, and memory response to challenge with ivated Japanese encephalitis antigen. J. Infect. Dis. 188, 11230.
kuno, T., Tseng, P.T., Hsu, S.T., Huang, C.T., Kuo, C.C., 1Japanese encephalitis surveillance in China (Province of Taduring 1968–1971. I. Geographical and seasonal features ooutbreaks. Jpn. J. Med. Sci. Biol. 28, 235–253.
eiris, J.S., Amerasinghe, F.P., Arunagiri, C.K., Perera,Karunaratne, S.H., Ratnayake, C.B., Kulatilaka, TAbeysinghe, M.R., 1993. Japanese encephalitis in Sri Lacomparison of vector and virus ecology in different agro-climareas. Trans. R. Soc. Trop. Med. Hyg. 87, 541–548.
eng, Y., Song, J., Tian, G., Xue, Q., Ge, F., Yang, J., Shi1998. Field evaluations ofRomamomermis yunanensis(Nema-toda: Mermithidae) for control of Culicinae mosquitoes in ChFundam. Appl. Nematol. 21, 227–232.
hukan, A.C., Borah, P.K., Mahanta, J., 2004. Japanese encepin Assam Northeast India. Southeast Asian. J. Trop. Med. PHealth 35, 618–622.
CT
PR
OO
F
8
J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx 17
Prasad, H., Prasad, R.N., Haq, S., 1993. Control of mosquito breeding947
throughGambusia affinisin rice fields. Indian J. Malariol. 30,948
57–65.949
Pugachev, K.V., Guirakhoo, F., Trent, D.W., Monath, T.P., 2003. Tra-950
ditional and novel approaches to flavivirus vaccines. Int. J. Para-951
sitol. 33, 567–582.952
Rajendran, R., Reuben, R., 1991. Evaluation of the water fernAzolla953
microphylla for mosquito population management in the rice-954
land agro-ecosystem of South India. Med. Vet. Entomol. 5,955
299–310.956
Rajendran, R., Reuben, R., Purushothaman, S., Veerapatran, R.,957
1995. Prospects and problems of intermittent irrigation for con-958
trol of vector breeding in rice fields in Southern India. Ann. Trop.959
Med. Parasitol. 89, 541–549.960
Rao, D.R., Reuben, R., Nagasampagi, B.A., 1995. Development of961
combined use of neem (Azadirachta indica) and water manage-962
ment for the control of culicine mosquitoes in rice fields. Med.963
Vet. Entomol. 9, 25–33.964
Rao, D.R., Reuben, R., Venugopal, M.S., Nagasampagi, B.A.,965
Schmutterer, H., 1992. Evaluation of neem,Azadirachta indica,966
with and without water management, for the control of culicine967
mosquito larvae in rice-fields. Med. Vet. Entomol. 6, 318–324.968
Reuben, R., Thenmozhi, V., Samuel, P.P., Gajanana, A., Mani, T.R.,969
1992. Mosquito blood feeding patterns as a factor in the epidemi-970
ology of Japanese encephalitis in Southern India. Am. J. Trop.971
Med. Hyg. 46, 654–663.972
Rhee, H.I., Shim, J.C., Kim, C.L., Lee, W.J., 1983. Small scale973
field trial withBacillus thuringiensis israelensisH-14 for control974
of the vector mosquito (Culex tritaeniorhynchus) larvae in rice975
fields. Korean J. Entomol. 13, 39–46.976
Sehgal, A., Dutta, A.K., 2003. Changing perspectives in Japanese977
encephalitis in India. Trop. Doct. 33, 131–134.978
Self, L.S., Shin, H.K., Kim, K.H., Lee, K.W., Chow, C.Y., Hong,979
H.K., 1973. Ecological studies onCulex tritaeniorhynchusas a980
vector of Japanese encephalitis. Bull. World Health Organ. 49,981
982
S 351,983
984
S ghn,985
Neu-986
987
S M.,988
.,989
e en-990
trial.991
992
S uwan-993
n the994
Mai995
alth996
997
S ies of998
ding999
1000
S anese1001
tors.1002
1003
Sundararaj, R., Reuben, R., 1991. Evaluation of a microgel droplet1004
formulation ofBacillus sphaericus1593 M (Biocide-S) for con- 1005
trol of mosquito larvae in rice fields in Southern India. J. Am.1006
Mosq. Control Assoc. 7, 556–559. 1007
Sunish, I.P., Reuben, R., 2001. Factors influencing the abundance1008
of Japanese encephalitis vectors in ricefields in India-I. Abiotic1009
Med. Vet. Entomol. 15, 381–392. 1010
Sunish, I.P., Reuben, R., 2002. Factors influencing the abundance1011
of Japanese encephalitis vectors in ricefields in India-II Biotic.1012
Med. Vet. Entomol. 16, 1–9. 1013
Suroso, T., 1989. Studies on Japanese encephalitis vectors in In-1014
donesia. Southeast Asian J. Trop. Med. Public Health 20, 627–1015
628. 1016
Suzuki, T., 1967. Bibliography on Culex tritaeniorhynchus in Japan1017
1945–1966. Jpn. Environ. Sanit Cent. 1018
Takagi, M., Sugiyama, A., Maruyama, K., 1995. Effect of rice cultur-1019
ing practices on seasonal occurrence ofCulex tritaeniorhynchus 1020
(Diptera: Culicidae) immatures in three different types of rice-1021
growing areas in Central Japan. J. Med. Entomol. 32, 112–1022
118. 1023
Tsai, T.F., 2000. New initiatives for the control of Japanese encephali-1024
tis by vaccination: minutes of a WHO/CVI meeting, Bangkok,1025
Thailand, 13–15 October 1998. Vaccine 18 (Suppl. 2), 1–25. 1026
United Nations, 2004. World Urbanization Prospects: The 2003 Re-1027
visions, Population Division Department of Economics and So-1028
cial Affair of the United Nations, New York. 1029
UNO, 2003. Gates foundation grant to help fight Japanese Encephali-1030
tis. United Nations Wire Service, Washington,http://www. 1031
unwire.org/UNWire/20031209/44911099.asp. 1032
van der Hoek, W., Sakthivadivel, R., Silver, J.B., Konradsen, F., 2001.1033
Alternate wet/dry irrigation in rice cultivation: a practical way1034
to save water and control malaria and Japanese encephalitis?1035
International Water Management Institute; Colombo, Sri Lanka.1036
Research Report 47. 1037
Vijayarani, H., Gajanana, A., 2000. Low rate of Japanese encephalitis1038
an1039
111,1040
1041
W in rice1042
orne1043
Man-1044
ation1045
1046
W als1047
eva,1048
ol 1049
1050
W Re-1051
1052
W he1053
h in1054
h 22,1055
1056
Y acil-1057
riasis1058
ontrol1059
1060
UN
CO
RR
E
41–47.olomon, T., 2004. Flavivirus encephalitis. N. Engl. J. Med.
370–378.olomon, T., Dung, N.M., Kneen, R., Gainsborough, M., Vau
D.W., Khanh, V.T., 2000. Japanese encephalitis. J. Neurol.rosurg. Psychiatry 68, 405–415.
olomon, T., Dung, N.M., Wills, B., Kneen, R., Gainsborough,Diet, T.V., Thuy, T.T., Loan, H.T., Khanh, V.C., Vaughn, D.WWhite, N.J., Farrar, J.J., 2003. Interferon alfa-2a in Japanescephalitis: a randomised double-blind placebo-controlledLancet 361, 821–826.
omboon, P., Choochote, W., Khamboonruang, C., Keha, P., Sphanit, P., Sukontasan, K., Chaivong, P., 1989. Studies oJapanese encephalitis vectors in Amphoe Muang, ChiangNorthern Thailand. Southeast Asian J. Trop. Med. Public He20, 9–17.
ong, J., Peng, Y., 1996. Field trials of combined use of two specmermithid nematodes to control Anopheles and Culex breein China. Indian J. Malariol. 33, 161–165.
ucharit, S., Surathin, K., Shrestha, S.R., 1989. Vectors of Japencephalitis virus (JEV): species complexes of the vecSoutheast Asian J. Trop. Med. Public Health 20, 611–621.
ED
ACTROP 1666 1–1
infection in rural children in Thanjavur district (Tamil Nadu),area with extensive paddy cultivation. Indian J. Med. Res.212–214.
ada, Y., 1988. Strategies for control of Japanese encephalitisproduction systems in developing countries. In: Vector BDisease Control in Humans Through Rice Agroecosystemagement. International Rice Research Institute in collaborwith the WHO/FAO/UNEP Panel of Experts, pp. 153–160.
HO, 1997. Vector Control—Methods for Use by Individuand Communities. World Health Organization, WHO, Genhttp://www.who.int/docstore/watersanitationhealth/vectcontr/ch08.htm.
HO, 2004. The World Health Report 2004, The World Healthport 2004, Geneva, World Health Organisation.
u, N., Liao, G.H., Li, D.F., Luo, Y.L., Zhong, G.M., 1991. Tadvantages of mosquito biocontrol by stocking edible fisrice paddies. Southeast Asian J. Trop. Med. Public Healt436–442.
adav, R.S., Sharma, V.P., Upadhyay, A.K., 1997. Field trial of Blus sphaericus strain B-101 (serotype H5a, 5b) against filaand Japanese encephalitis vectors in India. J. Am. Mosq. CAssoc. 13, 158–163.
T P
RO
OF
18 J. Keiser et al. / Acta Tropica xxx (2005) xxx–xxx
Yu, H.S., Kim, H.C., 1993. Integrated control of encephalitis vector1061
(Culex tritaeniorhynchus) with native fishes (Aplocheilus latipes1062
andAphyocypris chinensis) andBacillus thuringiensis(H-14)1063
in marshes in Jindo Island of Korea. Korean J. Entomol. 23,1064
221–230.1065
Yu, H.S., Lee, D.K., Lee, W.J., 1982. Mosquito control by the re-1066
lease of fish predatorAphyocypris chinensisin natural mosquito
breeding habitats of rice paddies and stream seepage in South1067
Korea. Korean J. Entomol. 12, 61–68. 1068
Zahiri, N.S., Federici, B.A., Mulla, M.S., 2004. Laboratory and1069
simulated field evaluation of a new recombinant ofBacillus 1070
thuringiensisssp. israelensis andBacillus sphaericusagainst 1071
Culex mosquito larvae (Diptera: Culicidae). J. Med. Entomol.1072
41, 423–429. 1073
UNC
OR
RE
C
ED
ACTROP 1666 1–18