Journal of the Neurological Sciences 244 (2006) 117 – 122
Neurological manifestations of dengue virus infection
U.K. Misra a,*, J. Kalita a, U.K. Syam a, T.N. Dhole b
Department of Neurology, Sanjay Gandhi PGIMS, Lucknow 226014, India
Microbiology, Sanjay Gandhi PGIMS, Lucknow 226014, India
Received 20 June 2005; received in revised form 15 November 2005; accepted 11 January 2006
Available online 9 March 2006
Aim:: Paucity of studies on neurological manifestations in dengue virus infection prompted this study. We aim to correlate clinical,
radiological and neurophysiological changes in dengue patients with neurological manifestations.
Method:: Consecutive IgM seropositive dengue patients admitted in neurology ward during 2003 – 2005 have been prospectively
evaluated. They were subjected to detailed clinical evaluation, blood counts, coagulation profile, serum chemistry including creatine kinase
(CK), cerebrospinal fluid (CSF), cranial CT and/or MRI, electroencephalogram (EEG), nerve conduction and needle electromyography
Results:: There were 17 patients, aged 5 to 56 years; 11 presented with encephalopathy and 6 with acute motor weakness. In the patients with
encephalopathy, seizures were present in 3, myoclonus in 1, CSF pleocytosis and EEG slowing in 8 each and globus pallidus and thoracic
spinal cord involvement on MRI in 1 patient each. In the pure motor weakness group, CK was elevated in 5 and EMG and muscle biopsy
were consistent with myositis in 1 patient each. The patients with pure motor weakness improved completely but in the encephalopathy group
3 died, 2 had partial, 1 poor and 5 complete recovery by 1 month.
Conclusion:: Dengue patients presenting with encephalopathy had more severe illness and worse outcome compared to acute pure motor
D 2006 Elsevier B.V. All rights reserved.
Keywords: Dengue; Encephalopathy; Myositis; MRI; EMG; EEG; Outcome
1. Introduction 450,000 cases are reported annually making it a leading
cause of morbidity .
Dengue is a flavi virus and is one of the most important Dengue has been regarded as a nonneurotropic virus .
human viral infections prevalent in southeast Asia, Africa, However there are reports describing neurological involve-
Australia and America where Aedes Aegypti, the mosquito ment in dengue virus infection [3 –6]. The neurological
vector of dengue virus, is present. More than 2.5 billion involvement in dengue virus infection includes encephalitis,
people live in tropical region and are at risk of dengue virus acute disseminated encephalomyelitis, transverse myelitis
infection. Moreover, the travelers to these areas are also at and Guillain Barre (GB) syndrome [5– 8]. The neurological
high risk of infection. About 100 million cases of dengue spectrum of dengue patients has been limited because of
fever and 250,000 cases of dengue hemorrhagic fever small number of case reports, paucity of imaging and
(DHF) occur every year all over the world. The incidence neurophysiological studies. These studies are often not
of DHF in Asia has risen steeply since 1980s and more than possible because the facilities for such investigations are not
available in the developing countries where dengue virus is
prevalent. In this communication, we report clinical,
radiological and neurophysiological findings in the patients
* Corresponding author. Fax: +91 522 2668017.
admitted in a neurology ward of a tertiary care teaching
E-mail addresses: email@example.com, firstname.lastname@example.org hospital, so as to learn about the neurological spectrum of
(U.K. Misra). dengue virus infection.
0022-510X/$ - see front matter D 2006 Elsevier B.V. All rights reserved.
118 U.K. Misra et al. / Journal of the Neurological Sciences 244 (2006) 117 – 122
2. Subjects and methods performed in CSF to diagnose Japanese encephalitis virus
coinfection. Patients with chronic liver or kidney failure
During 2003 – 2005, we managed 174 patients with were excluded.
neurological disorders possibly related to viral infections; The patients were subjected to detailed medical history
which included encephalitis in 101, transverse myelitis in and physical examination. Their age, area of residence,
12, Guillain Barre syndrome and polio like illness in 44 and symptoms of headache, body-ache, vomiting, fever, rash,
miscellaneous in 17 patients. During this period we hemorrhagic manifestations and focal neurological deficits
managed 24 patients with neurological syndrome associated were noted. Consciousness was assessed by Glasgow coma
with dengue virus infection. The diagnosis of dengue was scale (GCS) and mental status by mini-mental state
based on epidemiological (post monsoon period in an examination. Cranial nerve palsy was noted and limb
endemic area), clinical picture consistent with dengue weakness was graded on a 0 to V Medical Research
infection (fever, body-ache, hemorrhagic manifestations Council (MRC) scale. Muscle tone, reflex and sensations
and/or jaundice) and positive serum IgM ELISA for dengue were also recorded. Systemic manifestations such as
and exclusion of malaria and hepatitis. IgM ELISA was also lymphadenopathy, hepato-splenomegaly, jaundice, cardiac
Fp1 – A1
F3 – A1
C3 – A1
P3 – A1
O1 – A1
F7 – A1
T3 – A1
T5 – A1
Fp2 – A2
F4 – A2
C4 – A2
P4 – A2
O2 – A2
F8 – A2
T4 – A2
T6 – A2
Fz – A2
Cz – A2
TC=0.3 s 70 µV
HF=70 Hz 1 sec
Fig. 1. Electroencephalogram of a patient with dengue encephalopathy shows delta slowing.
U.K. Misra et al. / Journal of the Neurological Sciences 244 (2006) 117 – 122 119
abnormalities and respiratory findings were also recorded. ecchymoses were present in 2 and hematemesis or malena in
The laboratory tests included hemoglobin, blood counts, 2 patients. One patient had respiratory compromise neces-
hematocrit, platelet counts, blood sugar, blood urea nitro- sitating artificial ventilation. Liver and/or spleen were
gen, serum creatinine, bilirubin, trans-aminases, serum enlarged in 3 patients. Coagulopathy was present in 9 and
electrolytes, CK, prothrombin time and activated partial CSF revealed lymphocytic pleocytosis in 8 patients. EEG
thromboplastin time. Electrocardiogram, radiograph of chest was possible in 9 patients and 8 of them revealed theta to
in posteroanterior view, cranial CT scan and/or MRI were delta slowing (Fig. 1). Cranial CT was carried out in 5 and
carried out. Lumbar puncture was performed after ensuring was normal in all. Cranial MRI was done in 9 patients and
normal bleeding and coagulation functions. Cerebrospinal was normal in all except one who had T2 hyper-intensity in
fluid was analyzed for protein, sugar, cells, fungal and globus pallidus. This patient suffered from prolonged status
bacterial smear and culture. Electroencephalogram, nerve epilepticus (Fig. 2). Spinal MRI of the patient who had
conduction and concentric needle electromyography were paraparesis revealed T2 hyper-intensity of dorsal spinal
carried out in all the patients. The results of nerve cord. In this group, 3 patients died, 2 had partial, 1 poor and
conduction studies were compared with our laboratory’s 5 complete recovery by 1 month.
normal values . Patients were conservatively managed as
per WHO guidelines . In the patients with hemorrhagic 3.2. Acute pure motor weakness
diathesis, platelet concentrate and/or fresh frozen plasma
were administered as indicated. Outcome was defined at 1 Six patients had acute pure motor quadriparesis. They
month into complete (independent for activities of daily were referred as GB syndrome (3), dengue fever (2) and
living), partial (dependent for activities of daily living) and viral infection (1). The weakness occurred during 2 – 5 days
poor (bed ridden) . of fever and at the time of weakness fever ranged between
38 -C and 40 -C. Five had body-ache and 3 each had
headache and vomiting. Ecchymotic patches were noted in 3
3. Results patients and 1 had epistaxis and malena. Hypotension and
respiratory failure were not present in any patient. Liver was
There were 24 patients with dengue virus infection enlarged in 3 patients. All the patients had normal mental
presenting with different neurological manifestations. Seven status and cranial nerve examination; 4 patients had mild
patients were excluded because of associated Japanese upper and lower limb weakness (MRC grade III and IV) and
encephalitis, hepatic failure and intracerebral hemorrhage 2 had severe (grade II or below). The muscle tone was
in 2 each and coexistent malaria in 1 patient. Our results reduced in 3 patients. Biceps, triceps, knee and ankle
therefore are based on 17 patients. Their mean age was 34.6 reflexes were absent in 1 and reduced in 2 patients.
years, 5 were females. All the patients hailed from the state Serum CK was elevated in 5, and serum sodium and
of Uttar Pradesh, India; 12 were from urban and 5 from rural potassium were normal in all. The CSF did not reveal
areas and were admitted in post monsoon period. The
referral diagnosis was meningoencephalitis in 10, GB
syndrome in 3, dengue fever in 2, septicemia and viral
fever in 1 patient each. The patients were admitted with a
history of 3 –15 (median 7) days of illness. Fever was
present in all and ranged between 38 -C and 40 -C. The
neurological manifestations of the patients could be
categorized into 2 major groups:
1. Febrile encephalopathy (fever with altered sensorium)
2. Acute pure motor weakness (fever with muscle weakness
with or without myalgia)
3.1. Febrile encephalopathy group
Eleven patients presented with febrile encephalopathy
with headache in 6, vomiting in 7, generalized tonic clonic
seizures in 3 and myoclonus in 1. Three patients had 2
attacks of seizures, which were easily controlled by
phenytoin monotherapy. Four patients had behavioral
abnormality at the time of admission. The GCS score Fig. 2. Cranial MRI axial section in T2 sequence of a patient with dengue
ranged between 4 and 14 and 1 had decerebration. Three who suffered from status epilepticus showing hyper-intense lesion in
patients had quadriparesis and 1 paraplegia. Purpura and bilateral globus pallidus.
120 U.K. Misra et al. / Journal of the Neurological Sciences 244 (2006) 117 – 122
Table 1 and metabolic abnormalities. All the patients improved
Clinical features of neurological manifestations of dengue virus infection completely by the second week. The clinical and labora-
(encephalopathy group—serial no. 1 to 11; acute pure motor weakness—
serial no. 12 to 17) tory findings are presented in Tables 1 and 2.
Serial no. Age/sex GCS Seizure Weakness Reflex Outcome
1 6/M 4 GTCS Absent Hyper Complete
2 16/M 13 Absent Absent Normal Complete
3 17/M 12 Sec gen Q Hypo Partial
4 45/F 14 Myoclonus Absent Normal Partial In our study 11 patients had febrile encephalopathy and 6
5 50/M 13 GTCS Absent Normal Complete had acute pure motor weakness following dengue virus
6 30/F 12 Absent Paraplegia Hypo (LL) Poor infection. Encephalopathy in dengue virus infection may be
7 50/M 13 Absent Absent Hyper Death
due to cerebral edema, anoxia, hemorrhage, hyponatremia,
8 50/F 11 Absent Q Normal Complete
9 31/F 5 Absent Absent Hypo Death liver or kidney failure, microcapillary hemorrhage or release
10 68/M 10 Absent Absent Hyper Complete of toxic substances [5,6,13 – 15]. Seven dengue patients were
11 40/M 8 Absent Q Hyper Death excluded from our study because of associated intracerebral
12 32/M 15 Absent Q Normal Complete hemorrhage in 2, coinfection with Japanese encephalitis in 2
13 38M 15 Absent Q Hypo Complete
and malaria in 1 and severe hepatic dysfunction in 2 patients,
14 35M 15 Absent Q Hypo Complete
15 13/F 15 Absent Q Normal Complete which could result in altered sensorium (confounding
16 42/M 15 Absent Q Hypo Complete variables). In the remaining patients there was no significant
17 26/M 15 Absent Q Normal Complete hypotension, hyponatremia, liver and kidney impairment
GCS = Glasgow coma scale, M = male, F = female, Q = quadriparesis, severe enough to account for the alteration in consciousness.
GTCS = generalized tonic clonic seizure, Sec gen = partial motor seizure In metabolic and toxic encephalopathy CSF is normal. In our
with secondary generalization. patients with encephalopathy, 8 had lymphocytic pleocytosis
suggesting meningoencephalitis. Cranial CT and MRI scan
pleocytosis. Motor and sensory nerve conductions were were normal in all except in one who had globus pallidus
normal in all the patients. Concentric needle EMG involvement on MRI, which was possibly due to prolonged
performed 7 –10 days after the onset of illness was normal hypoxia following status epilepticus in this patient. Such
in 5. In 1 patient whose EMG was done on day 5, though lesions have been reported in neonatal asphyxia [16,17]. In
did not have any spontaneous activity but the MUPs were arbovirus encephalitis, however, high frequency of thalamic,
of short duration, low amplitude with 30– 40% polyphasia. basal ganglia and brainstem involvement has been reported
Muscle biopsy was done in the second week in one patient [18,19].
and revealed variation in fiber size, hyalinization and In a study from Vietnam 10 out of 21 patients had
myophagocytois which is suggestive of myositis. This encephalopathy, which was attributed to encephalitis in 9,
patient has been reported separately . In 2 patients, hepatic encephalopathy in 5 and other metabolic dysfunc-
wake EEG revealed theta slowing which has no other toxic tion in 4. CT scan however was carried out in 3 patients
Laboratory findings of dengue virus infection with neurological manifestations
Serial no. S bil SGPT S creat CK (U/L) Platelet EEG CSF [cell/mm3]/
(mg/dl) (U/L) (mg/dl) (thousand/mm3) [protein (mg/dl)]
1 1.5 106 0.1 – 33 Theta 350/74
2 0.2 138 1.7 235 317 Theta 50/60
3 0.3 88 1.3 71 246 Theta 20/42
4 0.6 70 1.6 114 400 Theta 5/56
5 0.5 117 1.0 – 140 Theta 40/49
6 0.4 95 0.8 48 252 Normal 0/40
7 0.5 115 5.2 326 16 Not done 25/49
8 1.8 270 0.7 590 45 Theta 2/34
9 0.4 20 0.9 700 144 Theta 25/24
10 0.5 131 1.0 80 67 Theta 20/128
11 3.9 144 5.9 140 115 Not done 30/74
12 0.2 231 1.1 550 97 Normal 0/29
13 0.4 135 1.3 592 120 Normal 0/25
14 0.5 70 1.2 160 62 Normal 0/36
15 0.4 168 0.6 346 13 Theta 5/42
16 2.9 202 1.6 3050 20 Theta 5/46
17 1.0 290 1.3 440 34 Normal Not done
S = serum, bil = bilirubin, creat = creatinine, CK = creatine phosphokinase, EEG = electroencephalogram, CSF = cerebrospinal fluid, SGPT = serum glutamate
U.K. Misra et al. / Journal of the Neurological Sciences 244 (2006) 117 – 122 121
only. The basis of diagnosing encephalitis in those patients more recently and prominently West Nile encephalitis.
was CSF pleocytosis, positive ELISA and PCR in CSF . Following dengue virus infection both GB syndrome and
In our study, CSF IgM ELISA was also positive in 3 out of 6 acute transverse myelitis have been reported [5,26]. We
patients with encephalopathy and none in pure motor have not found any patient with GB syndrome in spite of
weakness group. In a study on 27 dengue hemorrhagic detailed nerve conduction studies. One of our dengue
fever developing encephalopathy, 26 were comatose, 21 had patients presented with encephalomyelitis. Encephalopathy
convulsions and 1 hemiplegia. MRI done in 18 patients improved in 1 week but she remained paraplegic because of
revealed no abnormality in 4 patients, edema and scattered myelitis. Her MRI also showed T2 hyper-intensity in
focal lesion in 2, cerebral edema in 12 and cerebral thoracic spinal cord. It was interesting to note that in 2
hemorrhage in 1 . We however excluded 2 patients with patients with only motor weakness also had EEG slowing
dengue having intracerebral hemorrhage. High incidence which could be due to subclinical encephalopathy as has
(75%) of convulsion in dengue patients has been reported been reported in other viral infection .
from Vietnam , which is twice that seen in our study. The dengue patients with acute motor weakness im-
Similar high incidence of convulsions has also been proved completely within 2 weeks whereas those with
reported in JE from Vietnam . Presence of focal lesion encephalopathy only 45% improved completely by 1 month.
on MRI in the study of Cam et al.  may be responsible for 27% of the patients each either died or had poor or partial
high incidence of seizures and other underlying causes recovery suggesting a more severe illness in encephalopathy
should be therefore excluded. We did not observe such focal compared to myositis group. It seems that our patients
lesion in the MRI of our dengue patients. though present with 2 major syndromes: encephalopathy
Dengue is regarded as a nonneurotropic virus. There are and pure motor weakness, but some patients with enceph-
however recent reports on neurotropism or neuroinvasion of alopathy also had raised CK levels and some of the pure
dengue virus infection [5,6,21,22]. Dengue virus has been motor weakness patients had EEG slowing. This may
detected in the CSF in 5 out of 6 patients presenting with suggest a continuum of dengue virus infection. Pure motor
encephalitis confirming its neuroinvasion . Dengue weakness may represent a mild or early form of disease
virus antigens have also been demonstrated by immunohis- whereas encephalopathy may represent a severe form.
tochemistry in CNS biopsies from 5 fatal cases of dengue
infection associated with encephalopathy .
The syndrome of acute pure motor weakness in dengue
was quite characteristic and has not been comprehensively
We thank Prof. SK Shankar, Department of Neuropa-
evaluated in earlier reports. The clinical picture simulated
thology, NIHMANS, Bangalore for his valuable opinion on
GB syndrome and was the referral diagnosis in 3 out of the
biopsy. We also thank Mr. Rakesh Kumar Nigam for
6 patients. Presence of fever at the time of weakness, normal
nerve conduction studies and absence of albuminocytolog-
ical dissociation excluded the possibility of GB syndrome.
Muscle weakness in our patients was attributed to myositis References
because of reduced or normal reflexes, raised CK, myo-
pathic EMG and suggestive histopathology in another  Burke DS, Monath TP. Flaviviruses. In: Knipe DM, Howley PM,
patient. In our study EMG revealed myopathic changes Griffin DE, Lamb RA, Martin MA, Roizman B, Status SE, editors.
Fields virology. Philadelphia’ Lippincott Williams and Wilkins; 2001.
only in one patient in whom EMG was carried out on 5th p. 1043 – 126.
day; in the remaining patients with normal EMG findings,  Brinton MA. Flaviviruses. In: McKendall RR, Stroop WG,
EMG was performed after 7 days. The delay in EMG was editors. Hand book of neurovirology. New York’ Marcel Dekker;
due to associated thrombocytopenia, coagulopathy and 1994. p. 379 – 89.
working diagnosis being GB syndrome in which an early  Kho LK, Sumarmo, Wulur H, Jahja EC, Gubler DJ. Dengue
hemorrhagic fever accompanied by encephalopathy in Jakarta.
EMG is not considered useful. In influenza myositis, EMG Southeast Asian J Trop Med Public Health 1981;12:83 – 6.
has been reported to be normal in nearly half the patients  Nimmannitya S, Thisyakorn U, Hemsrichart V. Dengue haemorrhagic
. Dengue myositis has been reported in 2 patients and fever with unusual manifestations. Southeast Asian J Trop Med Public
was diagnosed on the basis of raised CK . The Health 1987;18:398 – 406.
 Solomon T, Dung NM, Vaughn DW, Kneen R, Thao LT, Raeng-
possibility of bleeding into the muscle resulting in pain
sakulrach B, et al. Neurological manifestations of dengue infection.
and paralysis is unlikely in our patients because the Lancet 2000;355:1053 – 9.
weakness involved all four limbs without any focal  Cam BV, Fonsmark L, Hue NB, Phoung NT, Poulsen A, Heegaard ED.
swelling. Moreover muscle histology did not reveal any Prospective case controlled study of encephalopathy in children with
evidence of bleeding. In all flavi virus infections involving dengue hemorrhagic fever. Am J Trop Med Hyg 2001;65:848 – 51.
the nervous system, a non-negligible proportion has a lower  Hendarto SK, Hadinegoro SR. Dengue encephalopathy. Acta Paediatr
Jpn 1992;34:350 – 9.
motor neuron syndrome with their other manifestations.  Hommel D, Talarmin A, Deubel V, Reynes JM, Drouet MT, Sarthou
This is true for Japanese encephalitis, St. Louis encephalitis, JL, et al. Dengue encephalitis in French Guiana. Res Virol 1998;149:
Murray Valley encephalitis, Powassan encephalitis and 235 – 8.
122 U.K. Misra et al. / Journal of the Neurological Sciences 244 (2006) 117 – 122
 Misra UK, Kalita J. Cinical neurophysiology. India’ Elsevier; 2004.  Deresiewicz RL, Thaler SJ, Shu L, Zamani AA. Clinical and
 World Health Organization. Dengue hemorrhagic fever: diagnosis, neurographic manifestations of Eastern Equine Encephalitis. N Engl
treatment, prevention and control. Geneva’ World Health Organiza- J Med 1997;336:1867 – 74.
tion; 1997.  Solomon T, Dung NM, Kneen R, Thao le TT, Gainsborough M,
 Misra UK, Kalita J, Kumar S, Mathur A. Radiological and Nisalak A, et al. Seizures and raised intracranial pressure in
neurophysiological changes in Japanese encephalitis. J Neurol Neuro- Vietnamese patients with Japanese encephalitis. Brain 2002;
surg Psychiatry 1994;57:1484 – 7. 125:1084 – 93.
 Kalita J, Misra UK, Mahadevan A, Shankar SK. Acute pure motor  Chardboonchart V, Khemarunmanus M, Yoksan S, Bhamarapravati
quadriplegia: is it dengue myositis? Electromyogr Clin Neurophysiol N. Dengue virus serotypic identification using suckling mouse and
2005;45(6):357 – 61. Western blot technique. Southeast Asian J Trop Med Public Health
 Lum LC, Lam SK, Choy YS, George R, Harun F. Dengue 1990;21:614 – 20.
encephalitis: a true entity? Am J Trop Med Hyg 1996;54:256 – 9.  Despres P, Frenkiel MP, Ceccaldi PE, Duarte Dos Santos C, Deubel V.
 Miagostovich MP, Ramos RG, Nicol AF, Nogueira RM, Cuzzi-Maya Apoptosis in the mouse central nervous system in response to
T, Oliveira AV, et al. Retrospective study on dengue fatal cases. Clin infection with mouse neurovirulent dengue viruses. J Virol
Neuropathol 1997;16:204 – 8. 1998;72:823 – 9.
 Chaturvedi UC, Dhawan R, Khanna M, Mathur A. Break-down of the  Lum LC, Lam SK, Choy YS, George R, Harun F. Dengue
blood – brain barrier during dengue virus infection of mice. J Gen Virol encephalitis: a true entity? Am J Trop Med Hyg 1996;54:256 – 9.
1991;72:859 – 66.  Hays AP, Gamboa ET. Acute viral myositis. In: Engel AG, Banker
 Volt T, Lemberg P. Damage of thalamus and basal ganglia in BQ, editors. Myology. New York’ Mc Graw Hill; 1986. p. 1439 – 66.
asphyxiated full term neonates. Neuropediatrics 1987;18:176 – 81.  Beauvais P, Quinet BE, Richardt JN. Dengue apoptosis of two cases.
 Kaufman SP, Miller SP, Ferriero DM, Glidden DH, Barkovich AJ, Arch Fr Pediatr 1993;50:905 – 7.
Partridge JC. Encephalopathy as a predictor of magnetic resonance  Esack A, Teelucksingh N. The Guillain Barre syndrome following
imaging abnormalities in asphyxiated newborn. Pediatr Neurol dengue fever. West Indian Med J 1999;48:36 – 7.
2003;28:342 – 6.  Westmoreland BF. EEG in cerebral inflammatory processes. In:
 Kalita J, Misra UK. Comparison of CT scan and MRI findings in the Niedermeyer E, da Silva FL, editors. Electroencephalography. Urban
diagnosis of Japanese encephalitis. J Neurol Sci 2000;174:3 – 8. and Schwarzenberg Baltimore; 1987 p. 259 – 74.