| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The first data suggesting JE virus transmission in Indonesia were from animal and human serosurveys conducted in the mid- to late 1960s, but because of cross-reactivity of hemagglutination inhibition (HI) assays, the results are considered inconclusive.3–5 The presence of JE virus was confirmed when it was isolated in 1972 from Culex tritaeniorhynchus mosquitoes near Jakarta.6 Subsequent isolations from mosquitoes were obtained from West and Central Java,7,8 West and East Nusa Tenggara provinces,5,9 and West Kalimantan (S. Ganefa, unpublished data). JE virus was also isolated from pigs, and > 20 mosquito isolates were obtained in further studies near Jakarta.10,11 Since the early 1970s, intermittent serologic surveys conducted by the Ministry of Health and other researchers have also suggested JE virus transmission in several areas of Indonesia.12–16
Despite studies showing JE virus presence, reports of confirmed human cases in Indonesia have been infrequent and from limited areas. One study reported possible cases in Jakarta in 1968; however, testing was conducted using an HI assay against JE and dengue 1 only, and most cases had elevated titers to both antigens.17 In 1981–1982 in Jakarta, about one in four (30/118) children with clinically suspected viral encephalitis in two hospitals showed a 4-fold rise in JE titer with HI and immune adherence hemagglutination tests.18 In 1985, seven JE cases with immunoglobulin (Ig)M in cerebrospinal fluid (CSF) were identified among hospitalized patients with encephalitis in Yogyakarta, Central Java (B. Herianto, unpublished data). Other sporadic reports have come from Timika, Irian Jaya,19 and Sumatera (D. Smith, unpublished data).
The only area where more complete studies of human disease burden have been undertaken is the island of Bali. From the late 1980s, reports of JE contracted in Bali by international travelers or expatriates brought initial attention to the disease.20–23 Between 1990 and 1995, 40 of 77 (52%) patients with suspected viral encephalitis were confirmed to have JE by detection of IgM in CSF (K. Kari, unpublished data). A hospital-based surveillance study from mid-2001 to 2003 further defined the epidemiology of JE in Bali and determined incidence was at least 7.1 per 100,000 children younger than 10 years of age.24
With very limited data available outside Bali, JE has not previously been considered a significant public health problem in Indonesia,25,26 although there has been recognition of the need for surveillance to better define JE disease burden.27 The strong perception of pigs as a necessary part of the transmission cycle has also led to uncertainty about the presence of virus in parts of Indonesia where pig-rearing is uncommon or absent. This study was undertaken to study the proportion of acute encephalitis cases caused by JE and define the epidemiology of JE in multiple geographic areas representing various risk levels across Indonesia.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The National Institute of Health Research and Development (NIHRD), Indonesia, oversaw implementation of the study in collaboration with PATH.
|
).
|
Inclusion criteria. Children from 1 month to 15 years of age who presented to one of the selected health facilities with AES and whose parent/guardian consented were included in the study. The World Health Organization (WHO) definition for AES was used: a person with an acute febrile illness and at least one of the following: 1) a change in mental status (including symptoms such as confusion, disorientation, coma, or inability to talk) or 2) the new onset of seizures (excluding simple febrile seizures, defined as single generalized convulsions lasting < 15 minutes with recovery of consciousness within 60 minutes).29 Cases of acute flaccid paralysis (AFP) were also included, because JE can present with AFP.30,31 A laboratory-confirmed case of JE was defined by the presence of JE virus-specific IgM antibody in CSF or serum detected by an IgM antibody capture (MAC) enzyme-linked immunosorbent assay (ELISA).29
Surveillance study procedures. Hospital and health center staff collected epidemiologic information and specimens from all cases who met inclusion criteria. A modified, shorter version of the hospital questionnaire was used at health center sites. At district hospitals, serum and CSF specimens for ELISA testing were initially stored in the freezer section of a refrigerator and were transported weekly to provincial level facilities where they were stored at –20°C. Each month, specimens and case report forms were shipped to the NIHRD in Jakarta to enable laboratory testing and data entry and analysis.
CSF and acute and convalescent sera were collected whenever possible. However, lumbar puncture (LP) was a routine practice at only five sentinel hospitals, infrequently done in six hospitals, and never conducted at four hospitals. In Papua province, because of difficulties in ensuring maintenance of the cold chain for frozen specimens shipped to Jakarta, only peripheral blood specimens on filter paper were collected during the first year of the study. In West Nusa Tenggara province (Lombok Island) hospitals, filter paper blood specimens were collected in addition to other specimens to determine the sensitivity of filter paper methodology for the detection of JE IgM under field conditions; in the health center sites, filter paper blood specimens alone were collected. In the other four provinces, filter paper blood specimens (from a finger prick) were only collected if venepuncture was not possible or failed. Filter paper specimens were dried and sent at room temperature to NIHRD. After 11 months of specimen collection and parallel testing of filter paper and serum specimens, the sensitivity of filter paper was determined to be poor (38%). As a result, from December 2005, Papua province started sera collection. However, at the health center sites in West Nusa Tenggara, filter paper blood specimens continued to be collected, because maintaining cold chain conditions for the transportation of specimens was not possible.
Laboratory methods. Testing was conducted at NIHRD using a MAC ELISA (Venture Technologies, Kuching, Malaysia).32,33 Specimens were tested for IgM antibodies to both JE and dengue viruses, because dengue virus also circulates in Indonesia and may elicit cross-reacting antibodies.5 A sample was considered to be JE IgM positive if the optical density against JE virus was higher than that against dengue virus.32
For quality control, ~10% of specimens were sent to Universiti Malaysia Sarawak for repeat ELISA testing. Confirmatory plaque reduction neutralization testing (PRNT) was also conducted in Malaysia on a sample of cases that had sufficient amounts of specimen available.
Malaria smears were conducted locally in malaria-endemic areas. Testing for other pathogens was done by physician request but in general was not routinely available.
Data analysis.
Data were entered into an Epi Info database v6.04d (Centers for Disease Control and Prevention) at NIHRD and PATH. Statistical analyses were conducted using SPSS 11.0 (SPSS, Chicago, IL). To assess concordance, a 
statistic was calculated. Median length of hospital stay was calculated using a survival analysis.
Ethics. Informed consent was obtained from the parent or legal guardian of each child. The surveillance study was approved by the NIHRD Committee on Health Research Ethics, Indonesia, and PATH’s Human Subjects’ Protection Committee.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
JE cases occurred in all six provinces, with a total of 82 cases confirmed during the 2-year surveillance period. Approximately equal numbers of JE cases occurred in both years (39 cases in the first year). Eighty cases (98%) were identified from hospitals, and two cases were detected at health center sites. Although the proportion of hospitalized AES cases laboratory confirmed as JE was 6% overall (80/1,401), the proportion varied widely among provinces. In West Kalimantan province, which was considered high-risk, 18% of all AES cases were JE IgM positive, whereas in West Sumatera, considered low-risk, only 2% of the cases were JE positive (Table 1
). In cases recruited at health centers, 2% were JE IgM positive.
Specimen collection and laboratory diagnosis. Thirty percent of the hospitalized AES cases (420/1,401) had CSF collected, 86% had an acute serum specimen, and 50% had a convalescent serum specimen. Overall, 18% of hospitalized patients had all three types of specimens collected. Thirteen percent of hospitalized patients had only filter paper specimens (one or two). Including the 95 health center patients from whom only filter paper specimens were collected, overall, 19% (283/1,496) of all AES cases had filter paper specimens only.
Diagnosis of JE was based on IgM antibodies in the CSF in 28% of cases (23/82). In cases where CSF was not collected or was negative, diagnosis was based on a seroconversion response (15%), IgM antibodies detected in both acute and convalescent specimens (28%), or IgM antibodies present in an acute serum or filter paper specimen (29%). Because IgM antibodies are only reliably detectable in sera of JE patients at Day 9 after illness onset,34,35 the days of sera collection were analyzed and showed that 35% (524/1,496) of AES patients had sera collected on Day 9 or later.
Among non-JE AES cases, dengue IgM antibody was detected in the CSF of nine patients (1%). It was also detected in the sera of 129 cases (9%) in whom CSF was negative (24/129) or not collected (105/129).
Laboratory testing quality control assessments carried out four times between July 2005 and December 2006 showed an agreement of 94% ( = 0.64; 95% confidence interval, 0.48–0.80). Nine JE cases that had sufficient amounts of specimen remaining underwent PRNT and were confirmed as JE infections. These specimens included at least one case from each province.
A malaria smear was undertaken in 75% of all patients. Five patients had Plasmodium falciparum in the smear and JE IgM in serum. An LP was not conducted in any of these patients, so presence of JE IgM in CSF was not able to be determined. One child with a positive smear seroconverted from JE IgM negative to positive in serum during his acute encephalitic illness. Without CSF results, it was not possible to determine which patients had encephalitic illness as the result of cerebral malaria, both JE and malaria, or JE with concurrent parasitemia.
Demographic characteristics of JE cases and patterns of disease.
JE occurred in children from 1.5 months to 13 years of age. In total, 95% of cases (78/82) were children younger than 10 years of age (Figure 2), including 71% younger than 5 years of age. Only four cases were detected in children 10 years of age or older. A higher percentage of JE cases were boys, with a male to female ratio of 1.3:1.
|
). There was some indication of a seasonal increase in cases in the fourth quarter of the year in West Kalimantan and in the first quarter of the year in West Nusa Tenggara (Lombok Island), but the limited period of surveillance and case numbers did not allow an accurate determination of any seasonal patterns in the different geographic locations.
|
The median length of inpatient stay for the 80 hospitalized JE cases estimated from a survival analysis was 14 days (range, 2–71 days). The clinical condition of the majority of patients was very serious, with the lowest level of consciousness noted to be "comatose" for 38% and "responding only to stimuli" for a further 25%. Complications of illness were not uncommon: 24% (19/80) developed a secondary pneumonia, and two patients developed decubitus ulcers.
Serology results were not available before hospital discharge, but doctors provided a suspected diagnosis, based on clinical impressions and any available laboratory information. In 44% (36/82) of patients who were later confirmed to be JE cases, the primary presumptive discharge diagnosis was viral encephalitis. For a further three cases (4%), the discharge diagnosis was meningoencephalitis. Bacterial meningitis was diagnosed in 21% and complex febrile seizures in 17% of the cases. Other presumptive discharge diagnoses included tuberculous meningitis (5%), cerebral malaria (3%), aseptic meningitis (1%), and other diagnoses (5%).
Death and disability rates.
The case fatality rate was 16% for JE cases, and 31% were considered disabled at the time of discharge. The most common disabilities recorded were aphasia, hemiparesis, and spasticity. Some patients (12%) had not recovered from acute illness on leaving the hospitals, and final outcomes were unknown. Only 41% of JE patients were reported to be alive and well at discharge (Table 2). Case fatality and disability rates were similar in the < 5 and 5–9 year age groups; there were too few cases for comparison in those 10 years of age and older.
|
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This study has some limitations. Most importantly, for reasons described below, the number of JE cases identified likely underestimates the true number that occurred. Pre-hospital deaths or children who died before specimen collection in hospital would not have been identified. Moreover, the WHO AES case definition may not have identified all potential JE cases. A recent assessment in patients hospitalized at a tertiary care infectious disease facility in Vietnam suggested that the case definition only detected 65% of JE cases in that setting, although including AFP cases increased the sensitivity to 87%.36 Collection of convalescent sera is important for diagnosis because JE antibodies rise relatively slowly, and only ~59–80% cases have IgM detectable at admission.34,35 However, convalescent sera were only collected from ~50% of cases in this study, and only 35% of cases had sera collected at least 9 days after illness onset, when IgM antibodies are reliably detectable. The sensitivity of testing filter paper blood specimens was poor, but this was the only type of specimen available from about one in five AES cases. CSF and sera were not able to be tested immediately and had to be stored in the province before transport, and in Jakarta were stored for up to 5 months in some cases, so there was the possibility of antibody degradation before testing. As a result of these limitations, the number of JE cases in this study can only be considered a minimum figure, with the true number likely to be higher. Accurate incidence rates could not be calculated because the catchment population of each sentinel site was not well defined.
The surveillance system was set up at the six sites to provide information on JE disease in Indonesia and be a foundation for future routine AES/JE surveillance. Hence, to the extent possible, the system was developed within routine surveillance systems and structures and not as a research project. Physicians were encouraged to collect CSF, but because LP was not routinely conducted at all sites, this was not always possible. CSF is the preferred specimen, but serum collected from acutely encephalitic patients is considered acceptable for confirmation of diagnosis for surveillance purposes.29 There is a chance that an encephalitis patient with JE IgM detected in serum actually has asymptomatic JE infection and the encephalitis is caused by another agent; however, this situation is likely to be rare. Regardless, detection of JE IgM antibodies in serum does indicate the patient has been exposed to JE virus, and therefore, it is an indication that the virus has been circulating in the region.
Agricultural activities, including rice growing, are an integral part of life for many families in Indonesia, especially in rural areas. Living in proximity to rice paddies or other Culex mosquito breeding grounds likely results in intense exposure to mosquitoes. Amplifying hosts for JE virus are also plentiful. Although JE is typically associated with the presence of pigs, these animals are not an essential part of the transmission cycle.37,38 Less than one half of JE cases in this study reported pigs within a 2.5-mi radius of their homes. Although information on pig proximity to houses was not verified, self-reporting could have introduced biases, and pigs could have been in the vicinity without the interviewee being aware, two of the sites had specifically been selected because there were no commercial piggeries and domestic pig-rearing was rare. Birds, including ducks and water birds, are equally important in the transmission and maintenance of JE virus28 and are likely to be important amplifying hosts in some parts of Indonesia.
Although pre-existing Indonesian JE disease burden data are limited, the state of Sarawak in Malaysia, which shares a border with West Kalimantan province, has collected JE surveillance data for > 10 years.33 JE has been recognized as a public health problem in Sarawak, and an immunization program was implemented in July 2001. In West Kalimantan, disease burden was also shown to be high—almost one in five AES cases were confirmed as JE. Epidemiologic patterns, including age group affected and year-round occurrence of cases with a fourth quarter seasonal peak, were also alike. This provides supportive evidence for the accuracy of the surveillance findings in this study.
In most JE-endemic countries, children are the group most commonly hospitalized with JE disease. The age distribution of cases found in this study, with the majority of cases occurring in children younger than 10 years of age and particularly high disease burden in children younger than 5 years of age, confirms local findings from previous studies in Bali.24,27 Although adults were not included in this study, < 5% of JE cases occurred in those 10–15 years of age, so it is likely the majority of children have developed immunity by this age and disease incidence in adults is likely to be low.
A high rate of disability, as documented in this study, is well recognized with JE disease.24,39,40 Children frequently suffer physical, cognitive, or psychiatric sequelae that generally persist throughout life.41 The impact of these long-term disabilities must be taken into account when considering the total burden of JE disease and the value of control measures. Even survivors not recognized as being disabled after illness often have behavioral problems or learning difficulties that may have serious consequences for them and their families.40
Ongoing surveillance in Indonesia is recommended to monitor disease incidence and trends. The most appropriate strategy, until diagnostics are more widely available, will be nationwide syndromic surveillance with reporting of clinical AES cases, and sentinel hospital sites selected where specimens can be collected and sent for laboratory confirmation. With limited resources, hospitals will be the most appropriate type of sentinel site. Although recognizing that some JE cases in health centers are not referred to hospitals—from the selected health centers on Lombok island, there were 2 JE cases among 95 AES cases that did not attend a hospital—collection and transportation of CSF and serum specimens for diagnostic testing are more feasible from hospitals than from the far more numerous and geographically scattered health centers. Adjustment for missed cases from health centers should, however, be taken into consideration when calculating overall disease burden rates.
A standardized AES case definition, combined with laboratory testing to confirm diagnosis, is needed for JE surveillance. Reporting and laboratory testing of patients with a clinical diagnosis of "viral encephalitis" is likely to be inadequate, because this study showed that at least one half of the patients who were subsequently shown to have JE were discharged from the hospitals with a diagnosis other than viral encephalitis. Many were considered clinically to have bacterial meningitis. It is well recognized that differentiation of JE from bacterial meningitis or other causes of central nervous system (CNS) infection based only on clinical criteria may be impossible.2
Efforts to strengthen surveillance should also focus on encouraging the collection of CSF. Not only is CSF more sensitive and specific than serum for diagnosis of JE,36 but LP with testing of CSF is considered beneficial for diagnosis and clinical management of patients with CNS infections.42 Collection of a convalescent serum specimen before discharge or death should also be encouraged because, for many patients, JE antibodies are only detectable in convalescent, and not acute, specimens.
These results build on findings from previous studies and provide a better understanding of the extent, patterns, and impact of JE disease in Indonesia. Based on the growing evidence of JE as a public health problem, JE control strategies are now being considered. JE is preventable by immunization. A safe, effective JE vaccine, the live, attenuated SA 14-14-2 JE vaccine, which has a simpler schedule and is more affordable than the older mouse brain–derived JE vaccine, is now available in Asia.43 Other JE vaccine candidates are also in late stage development.44 The Ministry of Health is planning a pilot program in Bali starting with an immunization campaign for children 10 years of age and under. This will pave the way for consideration of a national immunization program, including preventive campaigns and introduction of JE vaccine into the routine immunization program. Prioritization of introduction in areas considered highest risk for JE virus transmission may be warranted.
This study provides clear evidence of JE disease occurring in a wide geographic distribution across Indonesia. The development of the sentinel surveillance system provides a basis for future work to provide ongoing disease burden information and monitor the impact of vaccine introduction.
Received June 28, 2008. Accepted for publication August 31, 2008.
Acknowledgments: The authors thank all the members of the study team—national, provincial, and district staff—who were part of the surveillance work. In particular, we thank Gendro Wahyuhono, Djoko Yuwono, and Bambang Heriyanto (NIHRD, Indonesia). We thank the patients and their families for being agreeable to provide information. We also thank Jane Cardosa and Phaik Hooi Tio for providing technical assistance for transfer of the IgM ELISA technology, for provision of monoclonal antibodies and antigens, and for quality control and confirmatory testing (Universiti Sarawak Malaysia); Widyati Santoso, Mardiana Agustini, and Iwan Ariawan for data analysis (PATH Indonesia); Jodi Udd for assistance with manuscript preparation (PATH USA); and Asheena Khalakdina, Deborah Phillips, Chutima Suraratdecha, and Chris Victor for review of the manuscript (PATH USA).
Financial support: The JE project at PATH provided funding to the Indonesian Ministry of Health for this work. The Bill & Melinda Gates Foundation provided funding to the JE project at PATH.
Disclosure: The authors report no conflicts of interest.
* Address correspondence to Susan L. Hills, PATH, 1455 NW Leary Way, Seattle, WA 98107. E-mail: shills{at}path.org 
Authors’ addresses: Sahat Ompusunggu, Masri Sembiring Maha, Ni Ketut Susilarini, Agus Suwandono, and Endang R. Sedyaningsih, National Institute of Health Research and Development, Indonesia. Jl. Percetakan Negara 29, Jakarta 10560, Indonesia, Tel: 62-21-424-4375, Fax: 62-21-424-5386, E-mails: sahat{at}litbang.depkes.go.id, masri{at}litbang.depkes.go.id, susilarini{at}litbang.depkes.go.id, asuwandono{at}yahoo.co.id, and esedyani{at}indo.net.id. Susan L. Hills and Julie A. Jacobson, PATH, 1455 NW Leary Way, Seattle, WA 98107, Tel: 1-206-285-3500, Fax: 1-206-285-6619, E-mails: shills{at}path.org and Julie.Jacobson{at}gatesfoundation.org. Vanda A. Moniaga, Anton Widjaya, and Agus Sasmito, PATH, TIFA Building, 10th Floor, Suite 1001, Jl. Kuningan Barat 26, Jakarta 12710, Indonesia, Tel: 62-21-520-0737, Fax: 62-21-520-0621, E-mails: vanda{at}path.org, widjaya{at}path.org, and asasmito{at}hotmail.com.
Reprint requests: Japanese Encephalitis Project, PATH, 1455 NW Leary Way, Seattle, WA 98107, E-mail: jeproject{at}path.org.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
