Tweet
he Lancet 1999; 354:1253-1256DOI:10.1016/S0140-6736(99)04379-2
Early ReportOutbreak of Nipah-virus infection among abattoir workers in Singapore
Dr Nicholas I PatonMD
a 
, Yee Sin LeoMRCP a, Sherif R ZakiMD b, Alexander P AuchusMD c, Kim En LeeMRCP d, Ai Ee LingFRCPA e, Suok Kai ChewMSc f, Brenda AngMMed a, Pierre E RollinMD b, T UmapathiMRCP d, Ivy SngFRCPA e, Cheng Chuan LeeMRCP a, Erle LimMRCP c and T.G KsiazekDVM b
See Commentary
Summary
Introduction
Methods
Results
Discussion
References
Summary
BackgroundIn March 1999, an outbreak of encephalitis and pneumonia occurred in workers at an abattoir in Singapore. We describe the clinical presentation and the results of investigations in these patients.
MethodsClinical and laboratory data were collected by systemic review of the case records. Serum and cerebrospinal fluid (CSF) samples were tested for IgM antibodies to Nipah virus with an IgM capture ELISA. Reverse-transcriptase PCR was done on the CSF and tissue samples from one patient who died.
FindingsEleven patients were confirmed to have acute Npah-virus infection based on raised IgM in serum. Mpah virus was identified by reverse-transcriptase PCR in the CSF and tissue of the patient who died. The patients were all men, with a median age of 44 years. The commonest presenting symptoms were fever, headache, and drowsiness. Eght patients presented with signs of encephalitis (decreased level of consciousness or focal neurological signs). Three patients presented with atypical pneumonia, but one later developed hallucinations and had evidence of encephalitis on CSF examination Abnormal laboratory findings included a bw lymphocyte count (nine patients), low platelet count, low serum sodium, and high aspartate aminostransferase concentration (each observed in five patients). The CSF protein was high in eight patients and white-blood-cell count was high in seven Chest radiography showed mild interstitial shadowing in eight patients. Magnetic resonance imaging (MRI) showed focal areas of increased signal intensity in the cortical white marker in all eight patients who were scanned. The nine patients with encephalitis received empirical treatment with intravenous aciclovir and eight survived.
InterpretationInfection with Nipah virus caused an encephalitis illness with characteristic focal areas of increased intensity seen on MRI Lung involvement was also common, and the disease may present as an atypical pneumonia.
Back to top
Introduction
On March 16, 1999, a 47-year-old abattoir worker presented to Singapore General Hospital with fever, headache, and confusion. His level of consciousness deteriorated rapidly over the next 12 h, and he died from what appeared to be a viral encephalitis. Within 24 h of this patient's admission, a second abattoir worker presented to the same hospital with a similar illness. That patient's family mentioned that his brother (also an abattoir worker) was currently in another hospital with severe headache and pneumonia. Enquiries showed that a fourth abattoir worker with similar symptoms had been admitted to a third hospital in Singapore. The Ministry of Health was notified.
On March 19, 1999, the abattoirs in Singapore were closed down and a screening programme began for all abattoir staff. Over the next week, more than 500 abattoir workers were screened at the Communicable Disease Centre for febrile illness with symptoms and signs of respiratory or neurological disease. Patients with such features were admitted to hospital for further observation. In addition, all hospitals in Singapore instituted prospective surveillance for potential cases of illness among abattoir workers. 35 possible cases of infection were identified by these two approaches.
A similar illness had been observed among pig-farmers in Malaysia for several months before our cases in Singapore.1 Although that illness was initially thought to be Japanese encephalitis, virological studies done at the Centers for Disease Control and Prevention (CDC), USA, indicated that a new Hendra-like paramyxovirus, recently named Nipah virus, was associated with some cases.1,2 We report the investigation of and findings in the abattoir workers affected in the Singapore outbreak.
Back to top
Methods
One team of infectious-disease physicians, in conjunction with the Ministry of Health and other agencies, managed or comanaged all the cases detected in Singapore. In addition to prospective observation and data collection, the case records of these patients were reviewed retrospectively so that additional data could be obtained on laboratory tests and clinical progress.
Serum and cerebrospinal-fluid (CSF) samples were tested at the CDC for IgM antibodies to Nipah virus with an IgM capture ELISA based on Hendra virus antigens, and for IgG antibodies with an indirect IgG ELISA. These assays do not cross-react with other paramyxoviruses known to cause disease in human beings. The diagnosis was made on IgM in single samples with cut-off values previously established for similar IgM capture assays used in the diagnosis of haemorrhagic fever. Reverse-transcriptase PCR was done on CSF samples and on tissue samples from the patient who died after detection of Nipah virus. Tissue taken at necropsy from the patient who died was tested by immunohistochemistry with hyperimmune mouse serum against Hendra virus.
Back to top
Results
Patients
11 patients of 35 abattoir workers with symptoms of the illness admitted to hospital had IgM antibodies to Nipah virus. This total included all four index cases. Only one of these patients had IgG detected in serum. The median age was 44 years (range 24?66). All patients were men; ten were Chinese, and one Indian. All worked in the same abattoir but had different jobs: four were directly involved in slaughtering of pigs; two collected pig blood for sale at a market; one removed internal organs after slaughter; and the remaining four herded pigs around the abattoir before slaughter. Investigations revealed that this abattoir had imported pigs from a farm in Malaysia that had been affected by Nipah virus. The onset of the illness was between March 9 and March 19 for all patients. Nine of the patients had received vaccine against Japanese encephalitis (inactivated mouse-brain-derived virus) because of concern about the possible outbreak of that illness in Malaysia; eight had received two doses and one a single dose.
Case report
The first recognised case of encephalitis was a 47-year-old man who presented to the hospital with a 1-week history of headache, and a 2-day history of fever, cough productive of sputum, shortness of breath, drowsiness, and confusion. He had a temperature of 37?6?C and a pulse rate of 82 beats per min, and bilateral crackles were detected on auscultation of the chest. The patient was obtunded and disoriented. Deep-tendon reflexes were absent and there were bilateral Babinski's signs.
Laboratory investigations showed the following abnormalities: a low lymphocyte count (5?9% of a total white-blood-cell count of 4?4?109/L), low platelet count (87?109/L), low serum sodium (127 mmol/L), and high serum glucose (12?2 mmol/L). Lumbar puncture showed an opening pressure of 12 cm of water and clear CSF. The findings in CSF were: protein 0?4 g/L, glucose 7?9 mmol/L, and a white-blood-cell count of 5?106/L (lymphocytes).
The results of gram staining, indian-ink stain, and stains for acid-fast bacilli were negative as were those of latex agglutination tests for streptococci, neisseria, and Escherichia coli, and a cryptococcal-antigen test. CSF-antibody tests for herpes simplex, measles, mumps, Japanese encephalitis, dengue (completion-fixation tests), and Nipah viruses (both IgM capture and indirect IgG ELISA tests) were negative.
The chest radiograph showed slight cardiomegaly and distension of peripheral vessels, but no consolidation. Magnetic resonance imaging (MRI) revealed several hyperintensities in the frontal subcortical white matter on the right. An electroencephalogram showed severe diffuse encephalopathy with periodic lateralising epileptiform discharges over both temporal regions.
Empirical therapy with intravenous aciclovir and ceftriaxone was started on the day of admission to hospital. Over the next 12 h the patient had two generalised tonic-clonic seizures and became comatose with only a flexion response to pain. Reflexes were absent in arms and legs. He died 48 h later after cardiac arrest.
Necropsy showed systemic vasculitis with endothelial damage and syncytial cell formation. Necrotic areas with little inflammatory reaction were diffusely scattered in the central nervous system. Immunohistochemical assays with a polyclonal hyperimmune serum against Hendra virus gave positive results, and showed extensive viral antigen primarily in endothelial cells and parenchymal cells within the central nervous system. Smaller amounts of viral antigen were also seen in other tissues examined. The virus was isolated and detected by reverse-transcriptase PCR in the tissues and was antigenically and genetically related to Hendra virus, and identical to the virus (now known as Nipah virus) that caused viral encephalitis in Malaysia.1,2
Clinical presentation of the illness
The clinical characteristics of the 11 patients with serologically confirmed Nipah virus infection are summarised in the table. The commonest symptoms on presentation were fever (ten), headache (eight), drowsiness (five), nausea (five), confusion (four), vomiting (four), cough (four), giddiness (three), and myalgia (three). The commonest clinical findings at presentation were fever (>37?5?C) in seven patients and decreased level of consciousness in five. Five patients were tachycardic, but none was hypotensive or tachypnoeic. Two had bilateral lung crackles on auscultation. Abdominal examination was normal in all cases. Neurological signs were evident in several patients on admission: one patient had focal brain-stem involvement (unilateral ptosis, reduced palatal movements, dysphonia, and dysarthria), two had lower motor-neuron signs (diminished or absent reflexes), one had bilateral Babinski's signs, and cerebellar signs were noted in three patients (nystagmus in two, dysmetria in two, and unsteady gait in one). There was no evidence of autonomic dysfunction. The total white-blood-cell count was lower than normal (below 4?109/L) in three patients, the proportion of lymphocytes was low (below 20%) in nine, the platelet count was low (less than 140?109/L) in five, serum sodium was low (less than 135 mmol/L) in five, and the concentration of aspartate aminotransferase was slightly high (greater than 36 IU/L) in five. The haemoglobin concentration and indicators of renal and hepatic function were normal.
Table 1. Summary of presenting symptoms, signs, and investigations
Lumbar puncture was done for ten of the 11 patients. The amount of protein was slightly raised in five patients and substantially raised (about ten times the upper limit of normal) in three patients. The CSF white-cell-blood count was slightly raised in four patients and substantially raised in three. Lymphocytes predominated in all CSF samples except in the sample with the highest total count, in which 70% of the cells were polymorphs. Microscopy and culture, complement-fixation tests, and serological tests for bacterial and viral pathogens were negative for all CSF samples. Antibodies to Nipah virus were not detected in any of the CSF samples tested and reverse-transcriptase PCR detected Nipah virus in only one CSF sample from the patient who died.
The chest radiograph was abnormal at presentation or during the course of the illness in eight cases. Most patients had only slight abnormalities, and focal or more diffuse interstitial shadowing was the usual pattern. Computed tomography (CT) scans of the head were normal in all cases. MRI scans were done for eight patients, and all scans showed one or more focal areas of increased signal intensity, mainly in the white matter (figure). Electroencephalography done for five patients showed abnormality in all: four patients showed intermittent rhythmic slow wave activity without evidence of seizure activity and one patient had periodic lateralising epileptiform discharges over both temporal regions.
Click to enlarge image
Figure. MRI showing hyperintense lesions in deep cortical white matterMRI is by fluid attenuation inversion recovery sequences.
Two patients presented with atypical pneumonia with fever and infiltrates on chest radiography. Sputum cultures could not be obtained. Serological tests for mycoplasma, chlamydia, and Q fever were negative, but one patient had high titre of antibodies to legionella (1/1024) in samples taken during the acute illness and during convalescence; these findings suggest recent but not acute legionella infection. This patient had a prominent headache that persisted for several weeks into the illness, and the MRI scan showed that he had similar changes to the other patients, thus corroborating a diagnosis of acute Nipah-virus infection. The respiratory symptoms resolved within 1 week in these two patients, but the link between this response and empirical treatment with clarithromycin is unclear. A third patient who presented with pneumonia (blood cultures and serological tests for causes of atypical pneumonia were all negative) later had a transient episode of visual hallucinations (seeing pigs). Lumbar puncture showed an increased number of white-blood-cells in the CSF, and the diagnosis was revised to that of encephalitis. This patient was treated empirically with intravenous aciclovir and ceftriaxone, as were the eight with neurological features at presentation. Neurological recovery took from a few days to several weeks, although several patients initially deteriorated and four required mechanical ventilation. Only one patient died, although four had persistent neurological deficits (cognitive impairment and cerebellar signs) on discharge from hospital.
Back to top
Discussion
We have described the clinical features of a new infectious disease caused by Nipah virus, a previously unknown Hendra-like paramyxovirus. The illness most commonly presents as encephalitis with fever and altered level of consciousness, with or without focal neurological signs. Cerebellar signs were also prominent in several of our patients. In many patients there was associated pulmonary involvement, and the infection can present as an atypical pneumonia with fever, cough, headache, and opacities on chest radiography, but without the neurological features of encephalitis.
One characteristic feature of the disease was the presence of one or more focal high intensity lesions on MRI scans. These lesions were located in the white matter in most cases although in one case they were present in the superficial cortex. The lesions are non-specific indicators of a focal inflammatory process and would be consistent with demyelination or vasculitis. However, the finding of a focal vasculitis at necropsy supports the latter as the likely cause. Similar findings have been described in one of the reported cases of infection with the Hendra virus3 although in that patient the changes were located in the cortical grey matter. These MRI features are helpful in distinguishing infections by Nipah virus from those by Japanese encephalitis, because the latter characteristically leads to bilateral abnormalities in the thalamus, commonly with associated haemorrhage.4
We made the diagnosis of Nipah-virus infection in our patients by the detection of IgM antibodies in serum. Only one of the patients had IgG antibodies in the blood; this finding confirms that these patients were investigated in the acute stage of the disease. Early sampling may partly explain why we did not find antibodies to Nipah virus in the CSF samples. In addition, if the encephalitis is a result of secondary seeding after initial infection of another organ such as the lung, CSF-antibody production might occur at a late stage in this disease.
The disease seems to have occurred in Malaysia several months before this outbreak in Singapore.1,2 Most of the cases there were men working on pig farms. This association led to the natural suspicion that the Japanese-encephalitis virus was the cause. However, investigations in Malaysia and at the CDC have identified a new paramyxovirus, similar to Hendra virus.1,2 Hendra, named after the suburb of Brisbane, Australia where the first case was observed, has caused only three previously identified human infections.5,6,7 All three patients had close contact with diseased horses, and the virus probably spread by respiratory secretions from the infected horses. Of these patients, one presented with a severe and fatal pneumonia, one had a mild and non-specific febrile illness, and one presented initially with meningitis, which resolved, only to relapse after 1 year as a fatal encephalitis. Thus, there are some similarities between the clinical manifestations of the Hendra virus and the cases of infection with Nipah virus that we have described. However, further analysis of the virus circulating in Malaysia and Singapore has shown that there is about a 10% difference in the aminoacid sequence between the two viruses. This study therefore described a disease caused by a new and distinct human viral pathogen.
All 11 patients, with the exception of two patients who presented with an atypical pneumonia, were treated empirically with intravenous aciclovir; apart from the first identified case, all survived. However, several patients deteriorated before making a recovery, so whether the antiviral therapy influenced the natural course of the disease is unclear. Ribavirin has shown some activity against other paramyxovirus infections such as respiratory syncytial virus8 and therefore may have some use in treatment of this new infection. However, because high-dose intravenous ribavirin has toxic effects, it should be used with caution. The drug of choice for treatment of this infection remains unclear.
There are two abattoirs in Singapore, both of which import pigs from Malaysia. All 11 patients with confirmed infection with Nipah virus were employed at the same abattoir. Subsequent investigations have shown that this abattoir had imported pigs from a farm in Malaysia affected by the virus. Every pig is routinely examined before and after death in the abattoirs of Singapore, but no usual illness was detected around this time (S B Chua, personal communication). The mode of transmission of Nipah virus is currently unknown, although it is likely to require close contact with pigs. Our patients had various different jobs in the abattoir but all had direct involvement with the animals. No secondary cases in the family or contacts of abattoir workers have been identified to date, and the outbreak in Singapore was effectively contained when the importation of pigs was suspended and the abattoirs closed. Epidemiological case-control studies are underway at our centre and elsewhere to address issues of transmission. Our patients will be monitored closely to determine the long-term sequelae, if any, of this disease.
Back to top
Contributors
A E Ling, P Rollin, and T Ksiazek did the virological studies. S Zaki and I Sng did the analysis and identification of Nipah virus in the patients who died. N Paton, Y S Leo, A Auchus, K E Lee, B Ang, T Umapathi, C C Lee, and E Lim contributed to the clinical management of the patients and control of the outbreak. S K Chew coordinated the control of the outbreak at the Ministry of H ealth. N Paton wrote the first draft of the paper, and all investigators reviewed and contributed to the final draft.
Acknowledgments
We thank Kwong Ming Fock for permission to describe the cases that presented at Changi General Hospital; Meng Cheong Wong, Asok Kurup, Mun San Lam, Sin Yew Wong, and Chai Beng Tan for their assistance in managing the inpatients at the various hospitals; the nurses and doctors of the intensive-care units and isolation wards who provided care to these patients; Edmund Monteiro and the staff at the Communicable Disease Centre who ran the screening programme; Sin Bin Chua and the officers working in the Primary Production Department, the Ministry of the Environment, and the Ministry of Health, Singapore who contributed to the control of the outbreak.
Back to top
<!--start tail=-->References
1. Centers for Disease Control. Outbreak of Hendra-like virus?Malaysia and Singapore, 1998?1999. MMWR Morb Mortal Wkly Rep 1999; 48: 265-269. MEDLINE
2. Centers for Disease Control. Update: outbreak of Nipah virus?Malaysia and Singapore, 1999. MMWR Morb Mortal Wkly Rep 1999; 48: 335-337. MEDLINE
3. O'Sullivan JD, Allworth AM, Paterson DL, et al. Fatal encephalitis due to novel paramyxovirus transmitted from horses. Lancet 1997; 349: 93-95. Abstract | Full Text | Full-Text PDF (110 KB) | MEDLINE | CrossRef
4. Kumar S, Misra UK, Kalita J, Salwani V, Gupta RK, Gujral R. MRI in Japanese encephalitis. Neuroradiology 1997; 39: 180-184. MEDLINE | CrossRef
5. Selvey LA, Wells RM, McCormack JG, et al. Infection of humans and horses by a newly described morbillivirus. Med J Aust 1995; 162: 45.
6. Murray K, Selleck P, Hooper P, et al. A morbillivirus that caused fatal disease in horses and humans. Science 1995; 268: 94-97. MEDLINE
7. Paterson DL, Murray K, McCormack JG. Zoonotic disease in Australia caused by a novel member of the paramyxoviridae. Clin Infect Dis 1998; 27: 112-118. MEDLINE
8. Hall CB, McBride JT, Walsh EE, et al. Aerosolised ribavirin treatment of infants with respiratory syncytial virus infection; a randomised double-blind study. N Engl J Med 1983; 308: 1443-1447. MEDLINE
Back to top
<!--end tail-->Affiliations
a. Communicable Disease Centre, Tan Tock Seng Hospital, Singapore
b. Centers for Disease Control and Prevention, Atlanta, USA
c. Department of Neurology, Singapore General Hospital
d. National Neuroscience Institute, Singapore
e. Department of Pathology, Singapore General Hospital
f. M inistry of Health, Singapore
Correspondence to
r Nicholas Paton, Department of Infectious Diseases, Communicable Disease Centre, Tan Tock Seng Hospital, Moulmein Road, Singapore 308433
Early ReportOutbreak of Nipah-virus infection among abattoir workers in Singapore
Dr Nicholas I PatonMD
a , Yee Sin LeoMRCP a, Sherif R ZakiMD b, Alexander P AuchusMD c, Kim En LeeMRCP d, Ai Ee LingFRCPA e, Suok Kai ChewMSc f, Brenda AngMMed a, Pierre E RollinMD b, T UmapathiMRCP d, Ivy SngFRCPA e, Cheng Chuan LeeMRCP a, Erle LimMRCP c and T.G KsiazekDVM bSee Commentary
Summary
Introduction
Methods
Results
Discussion
References
Summary
BackgroundIn March 1999, an outbreak of encephalitis and pneumonia occurred in workers at an abattoir in Singapore. We describe the clinical presentation and the results of investigations in these patients.
MethodsClinical and laboratory data were collected by systemic review of the case records. Serum and cerebrospinal fluid (CSF) samples were tested for IgM antibodies to Nipah virus with an IgM capture ELISA. Reverse-transcriptase PCR was done on the CSF and tissue samples from one patient who died.
FindingsEleven patients were confirmed to have acute Npah-virus infection based on raised IgM in serum. Mpah virus was identified by reverse-transcriptase PCR in the CSF and tissue of the patient who died. The patients were all men, with a median age of 44 years. The commonest presenting symptoms were fever, headache, and drowsiness. Eght patients presented with signs of encephalitis (decreased level of consciousness or focal neurological signs). Three patients presented with atypical pneumonia, but one later developed hallucinations and had evidence of encephalitis on CSF examination Abnormal laboratory findings included a bw lymphocyte count (nine patients), low platelet count, low serum sodium, and high aspartate aminostransferase concentration (each observed in five patients). The CSF protein was high in eight patients and white-blood-cell count was high in seven Chest radiography showed mild interstitial shadowing in eight patients. Magnetic resonance imaging (MRI) showed focal areas of increased signal intensity in the cortical white marker in all eight patients who were scanned. The nine patients with encephalitis received empirical treatment with intravenous aciclovir and eight survived.
InterpretationInfection with Nipah virus caused an encephalitis illness with characteristic focal areas of increased intensity seen on MRI Lung involvement was also common, and the disease may present as an atypical pneumonia.
Back to top
Introduction
On March 16, 1999, a 47-year-old abattoir worker presented to Singapore General Hospital with fever, headache, and confusion. His level of consciousness deteriorated rapidly over the next 12 h, and he died from what appeared to be a viral encephalitis. Within 24 h of this patient's admission, a second abattoir worker presented to the same hospital with a similar illness. That patient's family mentioned that his brother (also an abattoir worker) was currently in another hospital with severe headache and pneumonia. Enquiries showed that a fourth abattoir worker with similar symptoms had been admitted to a third hospital in Singapore. The Ministry of Health was notified.
On March 19, 1999, the abattoirs in Singapore were closed down and a screening programme began for all abattoir staff. Over the next week, more than 500 abattoir workers were screened at the Communicable Disease Centre for febrile illness with symptoms and signs of respiratory or neurological disease. Patients with such features were admitted to hospital for further observation. In addition, all hospitals in Singapore instituted prospective surveillance for potential cases of illness among abattoir workers. 35 possible cases of infection were identified by these two approaches.
A similar illness had been observed among pig-farmers in Malaysia for several months before our cases in Singapore.1 Although that illness was initially thought to be Japanese encephalitis, virological studies done at the Centers for Disease Control and Prevention (CDC), USA, indicated that a new Hendra-like paramyxovirus, recently named Nipah virus, was associated with some cases.1,2 We report the investigation of and findings in the abattoir workers affected in the Singapore outbreak.
Back to top
Methods
One team of infectious-disease physicians, in conjunction with the Ministry of Health and other agencies, managed or comanaged all the cases detected in Singapore. In addition to prospective observation and data collection, the case records of these patients were reviewed retrospectively so that additional data could be obtained on laboratory tests and clinical progress.
Serum and cerebrospinal-fluid (CSF) samples were tested at the CDC for IgM antibodies to Nipah virus with an IgM capture ELISA based on Hendra virus antigens, and for IgG antibodies with an indirect IgG ELISA. These assays do not cross-react with other paramyxoviruses known to cause disease in human beings. The diagnosis was made on IgM in single samples with cut-off values previously established for similar IgM capture assays used in the diagnosis of haemorrhagic fever. Reverse-transcriptase PCR was done on CSF samples and on tissue samples from the patient who died after detection of Nipah virus. Tissue taken at necropsy from the patient who died was tested by immunohistochemistry with hyperimmune mouse serum against Hendra virus.
Back to top
Results
Patients
11 patients of 35 abattoir workers with symptoms of the illness admitted to hospital had IgM antibodies to Nipah virus. This total included all four index cases. Only one of these patients had IgG detected in serum. The median age was 44 years (range 24?66). All patients were men; ten were Chinese, and one Indian. All worked in the same abattoir but had different jobs: four were directly involved in slaughtering of pigs; two collected pig blood for sale at a market; one removed internal organs after slaughter; and the remaining four herded pigs around the abattoir before slaughter. Investigations revealed that this abattoir had imported pigs from a farm in Malaysia that had been affected by Nipah virus. The onset of the illness was between March 9 and March 19 for all patients. Nine of the patients had received vaccine against Japanese encephalitis (inactivated mouse-brain-derived virus) because of concern about the possible outbreak of that illness in Malaysia; eight had received two doses and one a single dose.
Case report
The first recognised case of encephalitis was a 47-year-old man who presented to the hospital with a 1-week history of headache, and a 2-day history of fever, cough productive of sputum, shortness of breath, drowsiness, and confusion. He had a temperature of 37?6?C and a pulse rate of 82 beats per min, and bilateral crackles were detected on auscultation of the chest. The patient was obtunded and disoriented. Deep-tendon reflexes were absent and there were bilateral Babinski's signs.
Laboratory investigations showed the following abnormalities: a low lymphocyte count (5?9% of a total white-blood-cell count of 4?4?109/L), low platelet count (87?109/L), low serum sodium (127 mmol/L), and high serum glucose (12?2 mmol/L). Lumbar puncture showed an opening pressure of 12 cm of water and clear CSF. The findings in CSF were: protein 0?4 g/L, glucose 7?9 mmol/L, and a white-blood-cell count of 5?106/L (lymphocytes).
The results of gram staining, indian-ink stain, and stains for acid-fast bacilli were negative as were those of latex agglutination tests for streptococci, neisseria, and Escherichia coli, and a cryptococcal-antigen test. CSF-antibody tests for herpes simplex, measles, mumps, Japanese encephalitis, dengue (completion-fixation tests), and Nipah viruses (both IgM capture and indirect IgG ELISA tests) were negative.
The chest radiograph showed slight cardiomegaly and distension of peripheral vessels, but no consolidation. Magnetic resonance imaging (MRI) revealed several hyperintensities in the frontal subcortical white matter on the right. An electroencephalogram showed severe diffuse encephalopathy with periodic lateralising epileptiform discharges over both temporal regions.
Empirical therapy with intravenous aciclovir and ceftriaxone was started on the day of admission to hospital. Over the next 12 h the patient had two generalised tonic-clonic seizures and became comatose with only a flexion response to pain. Reflexes were absent in arms and legs. He died 48 h later after cardiac arrest.
Necropsy showed systemic vasculitis with endothelial damage and syncytial cell formation. Necrotic areas with little inflammatory reaction were diffusely scattered in the central nervous system. Immunohistochemical assays with a polyclonal hyperimmune serum against Hendra virus gave positive results, and showed extensive viral antigen primarily in endothelial cells and parenchymal cells within the central nervous system. Smaller amounts of viral antigen were also seen in other tissues examined. The virus was isolated and detected by reverse-transcriptase PCR in the tissues and was antigenically and genetically related to Hendra virus, and identical to the virus (now known as Nipah virus) that caused viral encephalitis in Malaysia.1,2
Clinical presentation of the illness
The clinical characteristics of the 11 patients with serologically confirmed Nipah virus infection are summarised in the table. The commonest symptoms on presentation were fever (ten), headache (eight), drowsiness (five), nausea (five), confusion (four), vomiting (four), cough (four), giddiness (three), and myalgia (three). The commonest clinical findings at presentation were fever (>37?5?C) in seven patients and decreased level of consciousness in five. Five patients were tachycardic, but none was hypotensive or tachypnoeic. Two had bilateral lung crackles on auscultation. Abdominal examination was normal in all cases. Neurological signs were evident in several patients on admission: one patient had focal brain-stem involvement (unilateral ptosis, reduced palatal movements, dysphonia, and dysarthria), two had lower motor-neuron signs (diminished or absent reflexes), one had bilateral Babinski's signs, and cerebellar signs were noted in three patients (nystagmus in two, dysmetria in two, and unsteady gait in one). There was no evidence of autonomic dysfunction. The total white-blood-cell count was lower than normal (below 4?109/L) in three patients, the proportion of lymphocytes was low (below 20%) in nine, the platelet count was low (less than 140?109/L) in five, serum sodium was low (less than 135 mmol/L) in five, and the concentration of aspartate aminotransferase was slightly high (greater than 36 IU/L) in five. The haemoglobin concentration and indicators of renal and hepatic function were normal.
Click to view table
Table 1. Summary of presenting symptoms, signs, and investigations
Lumbar puncture was done for ten of the 11 patients. The amount of protein was slightly raised in five patients and substantially raised (about ten times the upper limit of normal) in three patients. The CSF white-cell-blood count was slightly raised in four patients and substantially raised in three. Lymphocytes predominated in all CSF samples except in the sample with the highest total count, in which 70% of the cells were polymorphs. Microscopy and culture, complement-fixation tests, and serological tests for bacterial and viral pathogens were negative for all CSF samples. Antibodies to Nipah virus were not detected in any of the CSF samples tested and reverse-transcriptase PCR detected Nipah virus in only one CSF sample from the patient who died.
The chest radiograph was abnormal at presentation or during the course of the illness in eight cases. Most patients had only slight abnormalities, and focal or more diffuse interstitial shadowing was the usual pattern. Computed tomography (CT) scans of the head were normal in all cases. MRI scans were done for eight patients, and all scans showed one or more focal areas of increased signal intensity, mainly in the white matter (figure). Electroencephalography done for five patients showed abnormality in all: four patients showed intermittent rhythmic slow wave activity without evidence of seizure activity and one patient had periodic lateralising epileptiform discharges over both temporal regions.
Click to enlarge image
Figure. MRI showing hyperintense lesions in deep cortical white matterMRI is by fluid attenuation inversion recovery sequences.
Two patients presented with atypical pneumonia with fever and infiltrates on chest radiography. Sputum cultures could not be obtained. Serological tests for mycoplasma, chlamydia, and Q fever were negative, but one patient had high titre of antibodies to legionella (1/1024) in samples taken during the acute illness and during convalescence; these findings suggest recent but not acute legionella infection. This patient had a prominent headache that persisted for several weeks into the illness, and the MRI scan showed that he had similar changes to the other patients, thus corroborating a diagnosis of acute Nipah-virus infection. The respiratory symptoms resolved within 1 week in these two patients, but the link between this response and empirical treatment with clarithromycin is unclear. A third patient who presented with pneumonia (blood cultures and serological tests for causes of atypical pneumonia were all negative) later had a transient episode of visual hallucinations (seeing pigs). Lumbar puncture showed an increased number of white-blood-cells in the CSF, and the diagnosis was revised to that of encephalitis. This patient was treated empirically with intravenous aciclovir and ceftriaxone, as were the eight with neurological features at presentation. Neurological recovery took from a few days to several weeks, although several patients initially deteriorated and four required mechanical ventilation. Only one patient died, although four had persistent neurological deficits (cognitive impairment and cerebellar signs) on discharge from hospital.
Back to top
Discussion
We have described the clinical features of a new infectious disease caused by Nipah virus, a previously unknown Hendra-like paramyxovirus. The illness most commonly presents as encephalitis with fever and altered level of consciousness, with or without focal neurological signs. Cerebellar signs were also prominent in several of our patients. In many patients there was associated pulmonary involvement, and the infection can present as an atypical pneumonia with fever, cough, headache, and opacities on chest radiography, but without the neurological features of encephalitis.
One characteristic feature of the disease was the presence of one or more focal high intensity lesions on MRI scans. These lesions were located in the white matter in most cases although in one case they were present in the superficial cortex. The lesions are non-specific indicators of a focal inflammatory process and would be consistent with demyelination or vasculitis. However, the finding of a focal vasculitis at necropsy supports the latter as the likely cause. Similar findings have been described in one of the reported cases of infection with the Hendra virus3 although in that patient the changes were located in the cortical grey matter. These MRI features are helpful in distinguishing infections by Nipah virus from those by Japanese encephalitis, because the latter characteristically leads to bilateral abnormalities in the thalamus, commonly with associated haemorrhage.4
We made the diagnosis of Nipah-virus infection in our patients by the detection of IgM antibodies in serum. Only one of the patients had IgG antibodies in the blood; this finding confirms that these patients were investigated in the acute stage of the disease. Early sampling may partly explain why we did not find antibodies to Nipah virus in the CSF samples. In addition, if the encephalitis is a result of secondary seeding after initial infection of another organ such as the lung, CSF-antibody production might occur at a late stage in this disease.
The disease seems to have occurred in Malaysia several months before this outbreak in Singapore.1,2 Most of the cases there were men working on pig farms. This association led to the natural suspicion that the Japanese-encephalitis virus was the cause. However, investigations in Malaysia and at the CDC have identified a new paramyxovirus, similar to Hendra virus.1,2 Hendra, named after the suburb of Brisbane, Australia where the first case was observed, has caused only three previously identified human infections.5,6,7 All three patients had close contact with diseased horses, and the virus probably spread by respiratory secretions from the infected horses. Of these patients, one presented with a severe and fatal pneumonia, one had a mild and non-specific febrile illness, and one presented initially with meningitis, which resolved, only to relapse after 1 year as a fatal encephalitis. Thus, there are some similarities between the clinical manifestations of the Hendra virus and the cases of infection with Nipah virus that we have described. However, further analysis of the virus circulating in Malaysia and Singapore has shown that there is about a 10% difference in the aminoacid sequence between the two viruses. This study therefore described a disease caused by a new and distinct human viral pathogen.
All 11 patients, with the exception of two patients who presented with an atypical pneumonia, were treated empirically with intravenous aciclovir; apart from the first identified case, all survived. However, several patients deteriorated before making a recovery, so whether the antiviral therapy influenced the natural course of the disease is unclear. Ribavirin has shown some activity against other paramyxovirus infections such as respiratory syncytial virus8 and therefore may have some use in treatment of this new infection. However, because high-dose intravenous ribavirin has toxic effects, it should be used with caution. The drug of choice for treatment of this infection remains unclear.
There are two abattoirs in Singapore, both of which import pigs from Malaysia. All 11 patients with confirmed infection with Nipah virus were employed at the same abattoir. Subsequent investigations have shown that this abattoir had imported pigs from a farm in Malaysia affected by the virus. Every pig is routinely examined before and after death in the abattoirs of Singapore, but no usual illness was detected around this time (S B Chua, personal communication). The mode of transmission of Nipah virus is currently unknown, although it is likely to require close contact with pigs. Our patients had various different jobs in the abattoir but all had direct involvement with the animals. No secondary cases in the family or contacts of abattoir workers have been identified to date, and the outbreak in Singapore was effectively contained when the importation of pigs was suspended and the abattoirs closed. Epidemiological case-control studies are underway at our centre and elsewhere to address issues of transmission. Our patients will be monitored closely to determine the long-term sequelae, if any, of this disease.
Back to top
Contributors
A E Ling, P Rollin, and T Ksiazek did the virological studies. S Zaki and I Sng did the analysis and identification of Nipah virus in the patients who died. N Paton, Y S Leo, A Auchus, K E Lee, B Ang, T Umapathi, C C Lee, and E Lim contributed to the clinical management of the patients and control of the outbreak. S K Chew coordinated the control of the outbreak at the Ministry of H ealth. N Paton wrote the first draft of the paper, and all investigators reviewed and contributed to the final draft.
Acknowledgments
We thank Kwong Ming Fock for permission to describe the cases that presented at Changi General Hospital; Meng Cheong Wong, Asok Kurup, Mun San Lam, Sin Yew Wong, and Chai Beng Tan for their assistance in managing the inpatients at the various hospitals; the nurses and doctors of the intensive-care units and isolation wards who provided care to these patients; Edmund Monteiro and the staff at the Communicable Disease Centre who ran the screening programme; Sin Bin Chua and the officers working in the Primary Production Department, the Ministry of the Environment, and the Ministry of Health, Singapore who contributed to the control of the outbreak.
Back to top
<!--start tail=-->References
1. Centers for Disease Control. Outbreak of Hendra-like virus?Malaysia and Singapore, 1998?1999. MMWR Morb Mortal Wkly Rep 1999; 48: 265-269. MEDLINE
2. Centers for Disease Control. Update: outbreak of Nipah virus?Malaysia and Singapore, 1999. MMWR Morb Mortal Wkly Rep 1999; 48: 335-337. MEDLINE
3. O'Sullivan JD, Allworth AM, Paterson DL, et al. Fatal encephalitis due to novel paramyxovirus transmitted from horses. Lancet 1997; 349: 93-95. Abstract | Full Text | Full-Text PDF (110 KB) | MEDLINE | CrossRef
4. Kumar S, Misra UK, Kalita J, Salwani V, Gupta RK, Gujral R. MRI in Japanese encephalitis. Neuroradiology 1997; 39: 180-184. MEDLINE | CrossRef
5. Selvey LA, Wells RM, McCormack JG, et al. Infection of humans and horses by a newly described morbillivirus. Med J Aust 1995; 162: 45.
6. Murray K, Selleck P, Hooper P, et al. A morbillivirus that caused fatal disease in horses and humans. Science 1995; 268: 94-97. MEDLINE
7. Paterson DL, Murray K, McCormack JG. Zoonotic disease in Australia caused by a novel member of the paramyxoviridae. Clin Infect Dis 1998; 27: 112-118. MEDLINE
8. Hall CB, McBride JT, Walsh EE, et al. Aerosolised ribavirin treatment of infants with respiratory syncytial virus infection; a randomised double-blind study. N Engl J Med 1983; 308: 1443-1447. MEDLINE
Back to top
<!--end tail-->Affiliations
a. Communicable Disease Centre, Tan Tock Seng Hospital, Singapore
b. Centers for Disease Control and Prevention, Atlanta, USA
c. Department of Neurology, Singapore General Hospital
d. National Neuroscience Institute, Singapore
e. Department of Pathology, Singapore General Hospital
f. M inistry of Health, Singapore
Correspondence tor Nicholas Paton, Department of Infectious Diseases, Communicable Disease Centre, Tan Tock Seng Hospital, Moulmein Road, Singapore 308433
Comment