Clinical Presentation of Nipah Virus Infection in Bangladesh

MAJOR ARTICLE Clinical Presentation of Nipah Virus Infection in Bangladesh M. Jahangir Hossain, 1 Emily S. Gurley, 1 Joel M. Montgomery, 4 Michael Bell, 4 Darin S. Carroll, 4 Vincent P. Hsu, 4 P. Formenty, 5 A. Croisier, 5 E. Bertherat, 5 M. A. Faiz, 2 Abul Kalam Azad, 3 Rafiqul Islam, 3 M. Abdur Rahim Molla, 3 Thomas G. Ksiazek, 4 Paul A. Rota, 4 James A. Comer, 4 Pierre E. Rollin, 4 Stephen P. Luby, 1 and Robert F. Breiman 1 1 International Centre for Diarrhoeal Disease Research and 2 Dhaka Medical College Hospital, Dhaka, and 3 Institute of Epidemiology, Disease Control and Research, Ministry of Health and Family Welfare, Government of Bangladesh, Bangladesh; 4 Centers for Disease Control and Prevention, Atlanta, Georgia; and 5 World Health Organization, Geneva, Switzerland Background. In Bangladesh, 4 outbreaks of Nipah virus infection were identified during the period 2001 2004. Methods. We characterized the clinical features of Nipah virus infected individuals affected by these outbreaks. We classified patients as having confirmed cases of Nipah virus infection if they had antibodies reactive with Nipah virus antigen. Patients were considered to have probable cases of Nipah virus infection if they had symptoms consistent with Nipah virus infection during the same time and in the same community as patients with confirmed cases. Results. We identified 92 patients with Nipah virus infection, 67 (73%) of whom died. Although all age groups were affected, 2 outbreaks principally affected young persons (median age, 12 years); 62% of the affected persons were male. Fever, altered mental status, headache, cough, respiratory difficulty, vomiting, and convulsions were the most common signs and symptoms; clinical and radiographic features of acute respiratory distress syndrome of Nipah illness were identified during the fourth outbreak. Among those who died, death occurred a median of 6 days (range, 2 36 days) after the onset of illness. Patients who died were more likely than survivors to have a temperature 137.8 C, altered mental status, difficulty breathing, and abnormal plantar reflexes. Among patients with Nipah virus infection who had well-defined exposure to another patient infected with Nipah virus, the median incubation period was 9 days (range, 6 11 days). Conclusions. Nipah virus infection produced rapidly progressive severe illness affecting the central nervous and respiratory systems. Clinical characteristics of Nipah virus infection in Bangladesh, including a severe respiratory component, appear distinct from clinical characteristics reported during earlier outbreaks in other countries. Nipah virus is a recently identified paramyxovirus that is closely related to Hendra virus [1]. The first recognized outbreaks of Nipah virus illness in humans occurred in Malaysia and Singapore from September 1998 through June 1999; 283 persons, mostly pig farm and abattoir workers were infected through contact with sick pigs [2 6]. A case-fatality rate of 40% was observed in Malaysia and Singapore; patients presented primarily with CNS symptoms [2, 7, 8]. A second outbreak of Received 8 July 2007; accepted 19 November 2007; electronically published 27 February 2008. Reprints or correspondence: Dr. M. Jahangir Hossain, Clinical Sciences Div., ICDDR,B, 68 Shaheed Tajuddin Ahmed Sarani, Mohakhali, Dhaka 1212, Bangladesh (jhossain@icddrb.org). Clinical Infectious Diseases 2008; 46:977 84 2008 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2008/4607-0003$15.00 DOI: 10.1086/529147 Nipah virus infection, with a case-fatality rate of 68%, occurred from January through February 2001 in Siliguri, India, a town close to the northern border of Bangladesh. Patients affected by this outbreak presented with both encephalitis and respiratory symptoms [9]. In Bangladesh, 7 outbreaks of Nipah virus infection were identified during the period 2001 2007. In Bangladesh, Nipah virus infection was associated with contact with a sick cow, consumption of fresh date palm sap (potentially contaminated with pteropid bat saliva), and person-to-person transmission [10 12]. The Malaysian outbreak was associated with a single strain [13]. By contrast, viruses isolated in Bangladesh represent diverse strains [14]. Hypothetically, a single strain of Nipah virus could result in a narrower range of clinical presentations than those found during epidemics associated with genetically diverse strains. Thus, Nipah virus illnesses occurring in Bangladesh potentially pro- Clinical Presentation of Nipah Virus CID 2008:46 (1 April) 977

Table 1. Clinical characteristics of Nipah virus infection during different outbreaks in Bangladesh. Characteristic Age, years Meherpur, 2001 (n p 13) Naogaon, 2003 (n p 12) Outbreak Rajbari, 2004 (n p 31) Faridpur, 2004 (n p 36) All (n p 92) Mean SD 40.2 16.2 17.4 13.5 15.3 10.5 35.4 14.7 27.0 17.0 Median (range) 38 (4 60) 12 (4 42) 12 (2 50) 35 (5 60) 26 (2 60) Fever 13/13 (100) 12/12 (100) 31/31 (100) 36/36 (100) 92/92 (100) Altered mental status a 13/13 (100) 10/12 (83) 30/30 (100) 29/36 (81) 82/91 (90) Unconsciousness b 3/13 (23) 9/12 (75) 28/30 (93) 27/36 (75) 67/91 (74) Headache 8/13 (62) 10/12 (83) 16/27 (59) 30/36 (83) 64/88 (73) Severe weakness NC 3/12 (25) 23/28 (82) 25/36 (69) 51/76 (67) Cough and/or cold 10/13 (77) 6/12 (50) 14/29 (48) 26/36 (72) 56/90 (62) Respiratory difficulty 9/13 (69) 7/12 (58) 19/29 (66) 27/36 (75) 62/90 (69) Vomiting 7/13 (54) 6/12 (50) 20/30 (67) 20/36 (56) 53/91 (58) Diarrhea 2/13 (15) 1/12 (8) 8/29 (28) 15/36 (42) 26/90 (29) Convulsion 3/13 (23) 3/12 (25) 14/30 (47) 1/36 (3) 21/91 (23) Planter reflex Absent ND ND 1/12 (8) 0/5 (0) 1/17 (6) Equivocal ND ND 2/12 (17) 2/5 (40) 4/17 (24) Extensor ND ND 6/12 (50) 0/5 (0) 6/17 (35) Flexor ND ND 3/12 (25) 0/5 (0) 3/17 (35) Deep tendon reflex Absent ND ND 4/11 (36) 0/3 (0) 4/14 (29) Diminished ND ND 5/11 (45) 0/3 (0) 5/14 (36) Exaggerated ND ND 0/11 (0) 0/3 (0) 0/14 (0) Normal ND ND 2/11 (18) 3/3 (100) 5/14 (36) Case fatality 9/13 (69) 8/12 (67) 23/31 (74) 27/36 (75) 67/92 (73) NOTE. Data are proportion (%) of patients, unless otherwise indicated. NC, not collected; ND, not done. a Altered mental status was defined as lack of orientation to time, place, and person. Unconsciousness was included as altered mental status. b Unconsciousness was defined as inability to give meaningful response to any stimuli. vide insight into broader clinical manifestations of Nipah virus infection. We describe the clinical presentation of 92 Nipah virus infected patients identified during the first 4 outbreaks in Bangladesh during the period 2001 2004. METHODS We investigated cases of Nipah virus infection from 4 outbreaks in the following regions: Meherpur District (from April through May 2001), Naogaon District (in January 2003), Rajbari (Goalanda subdistrict) and 7 other northwestern districts (from January through April 2004), and Faridpur District (from February through April 2004) [10, 11, 15 20]. The investigations for the first and second outbreaks, in Mehepur and Naogaon, took place after the outbreaks occurred. Because of increased awareness of Nipah virus among health officials, the third and fourth outbreaks, in Rajbari and Faridpur, were reported to authorities and investigated while the outbreaks were ongoing. Therefore, the case definitions and case detection methods differed slightly because of differences in the timing of investigations. Outbreaks of Nipah virus infection in Meherpur and Naogaon. The investigation of the Meherpur outbreak was conducted 2 years after the outbreak, and the investigation of the Naogaon outbreak was performed 2 months after the outbreak. Patients with suspected cases of Nipah virus infection were persons residing in the outbreak areas who experienced fever with either headache or altered mental status during the time of the outbreak [11]. Field research assistants identified patients with suspected cases during house to-house case-finding efforts. Study physicians collected illness histories from either patients with suspected cases or their caregivers (when the patients were minors or decedents) using a standardized case report form. We also collected information from hospital records, if available. A 5-mL blood specimen was collected from each living patient with a suspected case, and the serum was transported to the International Centre for Diarrhoeal Disease Research, Bangladesh (Dhaka), on ice for storage at 70 C. Serum samples were then shipped to the Centers for Disease Control and Prevention (Atlanta, GA) for Nipah virus serologic 978 CID 2008:46 (1 April) Hossain et al.

Table 2. Relationship between onset of Nipah virus illness and seroconversion in patients with acute Nipah virus infection. No. of days after onset of Nipah virus illness Total collected No. of specimens IgM antibody positive IgG antibody positive 1 1 1 1 2 1 1 0 3 1 0 0 4 2 1 0 5 4 3 1 6 8 7 2 7 4 4 1 8 4 4 3 9 2 2 1 10 2 2 2 11 14 9 9 6 15 30 4 4 4 45 60 4 3 4 11.5 years 4 0 4 testing. Patients with suspected cases of Nipah virus infection who survived and had evidence of the infection, demonstrated by the presence of either IgM or IgG antibodies, were considered to have laboratory-confirmed cases. Patients with suspected cases who died during the outbreak, experienced fever with altered mental status, and were linked to patients with laboratory-confirmed cases by place of residence and timing of symptom onset were considered to have had probable cases of Nipah virus infection. Outbreaks of Nipah virus infection in Rajbari and Faridpur. The Rajbari and Faridpur cases were investigated while the outbreaks were ongoing. A similar definition for a suspected case (fever and headache or altered mental status) was used during these investigations. However, because many severely ill patients with suspected cases also presented with respiratory symptoms, patients with a history of cough and fever were also considered to have suspected cases. Patients with suspected cases of Nipah virus infection were identified by house-to-house and hospital visits in the affected area. Physicians working with the investigation team collected illness histories from patients with suspected cases or their caregivers (when the patients were minors, decedents, or unable to provide personal histories because of altered mental status) using a standardized case report form. They verified clinical information by physical examination of the surviving patients (when possible) and by reviewing hospital records (when available). Case-finding efforts were expanded to 7 northwestern districts from January through April 2004. Tertiary care hospitals in the region were visited, and all patients hospitalized with fever and altered mental status during this period were investigated. We identified 19 additional cases of Nipah virus infection during this effort; some were clustered, and some were isolated. We included these 19 cases with the cases identified in Rajbari District for this report, because the cases were identified beginning with the Rajbari outbreak in January. We obtained acute serum samples from all living patients with suspected Nipah virus infection and convalescent serum samples at least 10 days after onset of illness from patients surviving acute illness. In addition to blood samples, we obtained throat swab, urine, and when possible, CSF samples from hospitalized patients with altered mental status. Serum and CSF samples were assayed for the presence or absence of Nipah virus specific IgM and IgG antibodies, as described elsewhere [21]. CSF and throat swab specimens were tested for presence of Nipah virus RNA by RT-PCR using a primer set to detect the nucleocapsid gene, as described elsewhere [7]. Attempts were also made to isolate virus from CSF, throat swab, and urine specimens by placing 100 ul of specimen in cell culture, according to methods described elsewhere [7, 22]. The samples were transported and stored in the same manner as in the first 2 outbreaks. Patients with suspected cases who had evidence of Nipah Table 3. Analysis of CSF specimens from 6 patients with laboratory-confirmed Nipah virus infection in 2004. Patient Age, years WBC count, cells/mm 3 RBC count, cells/mm 3 Glucose level, a mmol/l Protein level, b mg/dl Culture result Outcome 1 10 c 1 2 No growth Died 2 12 2 0 5.5 47 No growth Survived 3 35 d 4 3 3.3 30 No growth Survived 4 25 5 0 4.3 58 Not done Died 5 10 e 9 27 4.8 64 No growth Survived 6 18 f 2500 Moderate 3.8 181 No growth Died a Normal glucose level range, 2.8 4.2 mmol/l. b Normal protein level range, 20 40 mg/dl. c Virus isolated from a CSF specimen. d Virus isolated from a urine specimen. e Results of PCR of throat and CSF specimens were positive. f Results of PCR of urine and CSF specimens were positive. Clinical Presentation of Nipah Virus CID 2008:46 (1 April) 979

Table 4. Clinical features of Nipah virus infection in patients who died, compared with those who survived. Characteristic Age, years Patients who died a (n p 67) Patients who survived a (n p 25) OR (95% CI) P Mean SD 29.6 17.4 20.0 13.7.018 b Median (range) 28 (2 60) 15 (4 50) Male sex 44/67 (66) 13/25 (52) 1.8 (0.69 4.5).232 Altered mental status c 64/66 (97) 18/25 (72) 12.4 (2.1 128.3)!.001 Unconscious d 55/66 (83) 12/25 (48) 5.4 (1.7 16.9)!.001 Cough 44/65 (68) 12/25 (48) 2.3 (0.8 6.5).084 Respiratory difficulty 52/65 (80) 10/25 (40) 6.0 (2.0 18.5)!.001 Vomiting 39/66 (59) 14/25 (56) 1.1 (0.4 3.2).791 Convulsion 14/66 (21) 7/25 (28) 0.7 (0.2 2.4).493 Headache 48/64 (75) 16/24 (67) 1.5 (0.5 4.6).434 Severe weakness 38/55 (69) 13/21 (62) 1.4 (0.4 4.4).551 Temperature 137.8 C 27/30 (90) 4/9 (44) 11.2 (1.4 96.0).003 Abnormal plantar response (diminished, extensor, or equivocal) 8/8 (100) 3/9 (50) Undefined (NA).004 Deep tendon reflex (absent or diminished) 3/5 (60) 6/9 (67) 0.75 (0.0 14.1).803 NOTE. Data are proportion (%) of patients, unless otherwise indicated. NA, not applicable. a Difference in size of denominator is because of missing information or because the information was unknown or not collected. b P value by Mann-Whitney U test. c Altered mental status was defined as lack of orientation to time, place, and person. Unconsciousness was included as altered mental status. d Unconsciousness was defined as inability to give meaningful response to any stimuli. virus infection, demonstrated by the presence of Nipah virus IgM antibodies or by isolation of Nipah virus, were considered to have laboratory-confirmed cases [21]. Although other laboratory tests were performed, they were not considered to be reliable enough to use for the case definition. Patients with probable cases were those who had fever with altered mental status or breathing difficulty and whose cases were part of a cluster of laboratory-confirmed cases in the outbreak region; adequate specimens (including convalescent-phase specimens) for detection of Nipah virus antibodies could not be obtained from these patients because of fatal outcomes. Those with specimens (including convalescent-phase serum specimens) shown to be Nipah virus negative using all Nipah virus tests were considered to not be infected with Nipah virus. Statistical analysis. Clinical findings were contrasted between patients with fatal outcomes and patients who survived. Pearson s x 2 test or Fisher s exact test and univariate logistic regression were performed to determine ORs. Normally distributed continuous variables were analyzed using Student s t test, and nonnormally distributed continuous variables were analyzed using the Mann-Whitney U test. Associations were considered to be statistically significant at P!.05. Stata, version 8.0 (Stata), was used for statistical analysis. RESULTS We identified 92 patients with confirmed and probable cases of Nipah virus infection, including 13 patients in Meherpur, 12 in Naogaon, 31 in Rajbari (and other districts), and 36 in Faridpur. Forty-seven patients (51%) had laboratory-confirmed cases, 5 (5%; 4 from Meherpur and 1 from Naogaon) of whom were IgG positive only; all other patients with laboratory-confirmed cases ( n p 42) were IgM positive. Nipah virus was iso- lated from 4 patients (all of whom had detectable IgM antibodies to Nipah virus), and each isolate was identified by indirect fluorescent study test with Nipah virus antibodies and by RT-PCR [14]. Nipah virus was isolated from 2 patients associated with the third outbreak in Rajbari and 1 patient infected during the fourth outbreak in Faridpur; the fourth isolate was from a patient with an isolated case identified from Rajshahi District. Among 15 patients, all of whom had IgM antibodies, PCR results were positive for 8 throat swab, 6 urine, 5 CSF, and 4 saliva samples. Twenty-two (47%) of 47 laboratory confirmed-cases resulted in death. Of the 45 patients (49%) with probable cases, 41 (91%) died before collection of any specimens. Four patients had 1 specimen obtained during the early phase of illness (median time after onset of illness, 3.5 days; range, 2 5 days), and each patient had negative results of tests for Nipah virus and died before additional, potentially confirmatory specimens could be obtained. The age distribution of patients varied between outbreaks. The outbreaks in Naogaon (median patient age, 12 years; range, 4 42 years) and Rajbari (median patient age, 12 years; range, 2 50 years) occurred primarily among younger patients, com- 980 CID 2008:46 (1 April) Hossain et al.

Figure 1. Chest radiographs for 4 patients, showing diffuse bilateral opacities covering majority of the lung fields, consistent with acute respiratory distress syndrome. pared with the outbreaks in Meherpur (median patient age, 38 years; range, 4 60 years) and Faridpur (median patient age, 35 years; range, 5 60 years). Fifty-seven (62%) patients were male; however, the proportion of male patients varied from 46% to 74% during the various outbreaks. Fever, required as part of the probable case definition, was universally (100%) present, followed by altered mental status (90%), headache (73%), severe weakness (67%), cough (62%), difficult breathing (69%), diarrhea (29%), and seizures (23%) (table 1). The clinical presentation started with mild-to-severe fever, followed by altered mental status (median time after onset of illness, 4 days; range, 0 10 days) or cough with respiratory difficulty (median time after onset of illness, 4 days; range, 0 13 days). Compared with the other outbreaks, during the Faridpur outbreak, respiratory difficulty was more common (75%) and seizures were less common (3%). The case-fatality rate was high (73%; range, 69% 75%). Most patients who died did so shortly after onset of illness (mean duration SD from onset of illness to death, 7 4.6 days; median, 6 days; range, 2 36 days). In the Faridpur outbreak, during which most persons became infected through person-to-person transmission [19], a short, specific period of exposure could be determined for 11 patients who were exposed to only 1 infected patient for defined periods. Seven patients had a single close exposure (touched or were in the same room with an infected patient for at least 10 min) Clinical Presentation of Nipah Virus CID 2008:46 (1 April) 981

over the course of 1 day, and 4 had close exposures with a patient with Nipah virus infection during 2 consecutive days. On the basis of the exposure histories of these 11 patients, the median incubation period was 9 days (range, 6 11 days). During all 4 outbreaks, a total of 64 serum specimens were obtained from 51 patients. Two specimens were obtained from 13 patients; 23 patients died before a second sample could be obtained, and the remaining patients had only 1 specimen obtained after illness. We used antibody test results from all patients to assess the relationship between onset of symptoms and seroconversion (table 2). During the first 5 days of illness, 6 (66%) of 9 patients were IgM positive, and only 2 were IgG positive. By 2 weeks after symptom onset, all patients were IgM positive, and after 2 weeks, all patients were still IgG positive. Between 2 and 3 months after symptom onset, IgM antibody levels began to decrease, and after 2 years, no survivors were IgM positive, but 100% remained IgG positive. In total, 5 patients who were Nipah antibody negative when the first specimens were tested had detectable IgM antibodies in specimens obtained at least 2 weeks after the onset of illness. CSF specimens from 6 patients with confirmed Nipah virus infection were analyzed; 3 (50%) had a WBC count!5 cells/ mm 3. CSF glucose levels were within normal limits, and CSF protein levels were elevated in all patients except 1, who had normal levels. There was no bacterial growth yielded by any of the CSF cultures (table 3). Patients who died were significantly older (median age, 28 years; range, 2 60 years) than those who survived (median age, 15 years; range, 4 50 years; P p.018 ). Patients who died were more likely than survivors to have altered mental status (OR, 12.4; 95% CI, 2.1 128.3), difficulty breathing (OR, 6.0; 95% CI, 2.0 18.5), documented temperature 137.8 C (OR, 11.2; 95% CI, 1.4 96.0), and abnormal (diminished or extensor) planter reflexes (undefined OR; P p.004) (table 4). All surviving patients had laboratory-confirmed cases by definition, compared with 22 (33%) of 67 patients who died ( P!.001). Chest radiographs were performed for 5 patients from the Faridpur outbreak; all had a history of cough and difficulty breathing, and radiographic findings for all 5 were consistent with acute respiratory distress syndrome (figure 1). In patients with chest radiograph confirmed acute respiratory distress syndrome, respiratory symptoms (cough and difficulty breathing) started within 2 5 days after the onset of fever. All 5 patients with acute respiratory distress syndrome died. Only 1 patient who had severe respiratory difficulty received mechanical ventilation; a chest radiograph was not performed for this patient, and the patient died on day 17 of his illness. DISCUSSION There are important differences between the clinical descriptions of Nipah virus illness in Bangladesh and those from Malaysia. Cough or breathing difficulties were common during all of the first 4 outbreaks in Bangladesh and were particularly pronounced during the Faridpur outbreak. In the Faridpur outbreak, additional patients had respiratory symptoms similar to those experienced by the patients with documented acute respiratory distress syndrome, but chest radiographs were not performed because of the lack of facilities at treatment centers. In Malaysia and Singapore, respiratory symptoms were reported less frequently. Only 14% of Malaysian patients had a history of nonproductive cough [8], and 3 (27%) of 11 patients in Singapore had atypical pneumonia, along with abnormal chest radiographic findings [3, 4, 6, 23]. Respiratory symptoms were also common in patients during the outbreak in Siliguri, India in 2001, providing further evidence of a clinical presentation of Nipah virus infection in the southern Asian region that was different from the clinical presentation described in Malaysia [9]. Person-to-person transmission was documented during 1 outbreak in Faridpur [19] and was suggested in the Meherpur outbreak in Bangladesh [11]. Health care workers and hospital visitors were infected after exposure to hospitalized patients during the outbreak of Nipah virus infection in Siliguri, India, also suggesting person-to-person transmission [9]. The high prevalence of respiratory symptoms, especially cough, may have contributed to the transmissibility of Nipah virus infection from person to person that was observed in Bangladesh and Siliguri but not in Malaysia [5, 24, 25]. Coinfections with other respiratory pathogens have not been documented and were not systematically investigated. Genetic differences in Nipah virus strains between those isolated in Malaysia and those isolated in Bangladesh could explain the differences in clinical presentation observed, specifically the importance of respiratory symptoms in Bangladeshi patients [13, 26]. Because Nipah virus has been isolated from respiratory secretions of humans [27] and commonly causes pulmonary symptoms in other animals [28 30], it is plausible that some strains cause pulmonary symptoms in humans. Comparing sequence differences between patients with and without severe respiratory manifestations would help to evaluate this hypothesis. The case-fatality rate was higher in Bangladesh and Siliguri than in Malaysia. We found that clinical features, including respiratory symptoms, fever, and absence of plantar reflexes, were associated with death. In Bangladesh, the higher casefatality rate could be related to suboptimal health care. Diagnostic and supportive care facilities were limited in the areas affected by the outbreaks in Bangladesh. Other factors contributing to a higher case-fatality rate in Bangladesh may include any or a combination of the following: a more virulent Nipah virus genotype, exposure to secretions and/or excretions of patients with extremely high viral loads, and suppressed 982 CID 2008:46 (1 April) Hossain et al.

immunity, perhaps because of malnutrition (commonly seen in Bangladesh). WBC counts in CSF specimens were not elevated in 3 of 6 patients with Nipah virus infection, a finding similar to that for the Malaysian patients [8]. These findings suggest that normal WBC count and normal chemical parameters in CSF do not rule out Nipah virus infection in patients with encephalitis. There are several limitations to our study. The context of an investigation of an outbreak of Nipah virus infection in Bangladesh is challenging because of the quick progression from symptom onset to death, coupled with reluctance of sick patients to seek care and inadequate diagnostic infrastructure in health care facilities. For these reasons, many infected patients in Bangladesh were not hospitalized, and clinical evaluations were not performed. Even when hospital records were available, they were often incomplete. These challenges also produce delays in identification and investigation of outbreaks, which contributes to incomplete recall or recall bias by patients and/or relatives. This might have resulted in underreporting of some symptoms, especially when the recall time was long, as was the case for the Meherpur outbreak. In addition, because of delays in outbreak detection, the confirmed case definitions from the first and second outbreaks in Meherpur and Naogaon were based on the presence of IgG (rather than IgM) antibody to Nipah virus antigen. A case definition based on the presence of IgG antibody is reasonable, even if suboptimal, because all patients from these 2 outbreaks experienced encephalitis during a very specific period and in a specific geographic location. Although our case definition was based on the presence of IgG antibody, 3 of 4 patients with laboratory-confirmed cases who lived in Naogaon also had IgM antibodies, providing further evidence that presence of IgG antibodies is a reasonable indicator of past infection. In addition, there has been no evidence of another outbreak with similar impact in the area in the recent or more remote past; thus, previous exposure was unlikely. A third limitation is that our series may be biased toward more-severe cases, considering our reliance on hospital surveillance for case detection in many areas. However, door-todoor visits in areas where outbreaks occurred limited the effect of this bias. Asymptomatic infection was reported in 8% of patients with laboratory-confirmed cases in Malaysia [25]; however, there is no evidence of asymptomatic Nipah virus infection in Bangladesh during outbreaks, although cases of mild illness were identified. The findings of our investigation suggest that Nipah virus infections occurred in all age groups and that fever, altered mental status, cough, and respiratory symptoms were the most common symptoms among those infected in Bangladesh. Severe neurologic manifestations are consistently the most substantial and severe components of Nipah virus infection. Severe chronic sequelae occur in many survivors, adding to the significant public health burden of Nipah virus associated disease [31]. Future priorities should include ongoing surveillance and investigation of outbreaks of Nipah virus infection, identification and evaluation of strategies to prevent Nipah virus transmission, and improvement of clinical management of cases in resource-poor settings. Acknowledgments We thank the Centers for Disease Control and Prevention; World Health Organization; Institute of Epidemiology Disease Control and Research; Ministry of Health and Family Welfare, Bangladesh; Dr. Abu Taher Azad, Dr. Sultana Monira Hossain, and Dr. A. K. M. Saifuddin Ekram, for assistance in data collection; the study participants and their relatives; Mr. Milton Quiah, for administrative support; the Director and staff of the Dhaka, Faridpur, and Rajshahi Medical College Hospitals; and Dr. M. A. Salam, for his critical review of the manuscript. Financial support. Centers for Disease Control and Prevention, World Health Organization, and Ministry of Health and Family Welfare, Bangladesh. Potential conflicts of interest. All authors: no conflicts. References 1. Harcourt BH, Tamin A, Halpin K, et al. Molecular characterization of the polymerase gene and genomic termini of Nipah virus. Virology 2001; 287:192 201. 2. Update: outbreak of Nipah virus Malaysia and Singapore, 1999. MMWR Morb Mortal Wkly Rep 1999; 48:335 7. 3. Chew MH, Arguin PM, Shay DK, et al. Risk factors for Nipah virus infection among abattoir workers in Singapore. J Infect Dis 2000; 181: 1760 3. 4. Chua KB. A novel approach for collecting samples from fruit bats for isolation of infectious agents. Microbes Infect 2003; 5:487 90. 5. Parashar UD, Sunn LM, Ong F, et al. Case-control study of risk factors for human infection with a new zoonotic paramyxovirus, Nipah virus, during a 1998 1999 outbreak of severe encephalitis in Malaysia. J Infect Dis 2000; 181:1755 9. 6. Chan KP, Rollin PE, Ksiazek TG, et al. A survey of Nipah virus infection among various risk groups in Singapore. Epidemiol Infect 2002; 128: 93 8. 7. Chua KB, Bellini WJ, Rota PA, et al. Nipah virus: a recently emergent deadly paramyxovirus. Science 2000; 288:1432 5. 8. Goh KJ, Tan CT, Chew NK, et al. Clinical features of Nipah virus encephalitis among pig farmers in Malaysia. N Engl J Med 2000; 342: 1229 35. 9. Harit AK, Ichhpujani RL, Gupta S, et al. Nipah/Hendra virus outbreak in Siliguri, West Bengal, India in 2001. Indian J Med Res 2006; 123: 553 60. 10. Gurley ES, Montgomery JM, Hossain MJ, et al. Risk of nosocomial transmission of nipah virus in a Bangladesh hospital. Infect Control Hosp Epidemiol 2007; 28:740 2. 11. Hsu VP, Hossain MJ, Parashar UD, et al. Nipah virus encephalitis reemergence, Bangladesh. Emerg Infect Dis 2004; 10:2082 7. 12. Luby SP, Rahman M, Hossain MJ, et al. Foodborne transmission of Nipah virus, Bangladesh. Emerg Infect Dis 2006; 12:1888 94. 13. AbuBakar S, Chang LY, Ali AR, Sharifah SH, Yusoff K, Zamrod Z. Isolation and molecular identification of Nipah virus from pigs. Emerg Infect Dis 2004; 10:2228 30. 14. Bellini WJ, Harcourt BH, Bowden N, Rota PA. Nipah virus: an emergent paramyxovirus causing severe encephalitis in humans. J Neurovirol 2005; 11:481 7. 15. World Health Organization. Nipah virus. Wkly Epidemiol Rec 2002; 77:297 9. Clinical Presentation of Nipah Virus CID 2008:46 (1 April) 983

16. Nipah virus outbreak(s) in Bangladesh, January-April 2004. Wkly Epidemiol Rec 2004; 79:168 71. 17. World Health Organization. Nipah-like virus, Bangladesh. Wkly Epidemiol Rec 2004; 79:93. 18. ICDDR,B. Nipah encephalitis outbreak over wide area of western Bangladesh, 2004. Health and Science Bulletin 2004; 2:7 11. 19. Gurley ES, Montgomery JM, Hossain MJ, et al. Person-to-person transmission of Nipah virus in a Bangladeshi community. Emerg Infect Dis 2007; 13:1031 7 20. Quddus R, Alam S, Majumdar MA, et al. A report of 4 patients with Nipah encephalitis from Rajbari district, Bangladesh in the January 2004 outbreak. Neurol Asia 2004; 9:33 7. 21. Daniels P, Ksiazek T, Eaton BT. Laboratory diagnosis of Nipah and Hendra virus infections. Microbes Infect 2001; 3:289 95. 22. Chan YP, Chua KB, Koh CL, Lim ME, Lam SK. Complete nucleotide sequences of Nipah virus isolates from Malaysia. J Gen Virol 2001; 82: 2151 5. 23. Lee KE, Umapathi T, Tan CB, et al. The neurological manifestations of Nipah virus encephalitis, a novel paramyxovirus. Ann Neurol 1999; 46:428 32. 24. Mounts AW, Kaur H, Parashar UD, et al. A cohort study of health care workers to assess nosocomial transmissibility of Nipah virus, Malaysia, 1999. J Infect Dis 2001; 183:810 3. 25. Tan KS, Tan CT, Goh KJ. Epidemiological aspects of Nipah virus infection. Neurol J Southeast Asia 1999; 4:77 81. 26. Harcourt BH, Lowe L, Tamin A, et al. Genetic characterization of Nipah virus, Bangladesh, 2004. Emerg Infect Dis 2005; 11:1594 7. 27. Chua KB, Lam SK, Goh KJ, et al. The presence of Nipah virus in respiratory secretions and urine of patients during an outbreak of Nipah virus encephalitis in Malaysia. J Infect 2001; 42:40 3. 28. Murray K, Selleck P, Hooper P, et al. A morbillivirus that caused fatal disease in horses and humans. Science 1995; 268:94 7. 29. Selvey LA, Wells RM, McCormack JG, et al. Infection of humans and horses by a newly described morbillivirus. Med J Aust 1995; 162:642 5. 30. Mohd Nor MN, Gan CH, Ong BL. Nipah virus infection of pigs in peninsular Malaysia. Rev Sci Tech 2000; 19:160 5. 31. Sejvar JJ, Hossain J, Saha SK, et al. Long-term neurological and functional outcome in Nipah virus infection. Ann Neurol 2007; 62:235 42. 984 CID 2008:46 (1 April) Hossain et al.