Meningitis determined by oligosymptomatic dengue virus type 3 infection: Report of a case

Introduction 

Dengue infection is caused by a flavivirus, and is recognized as one of the most important mosquito-borne human infections of the 21st century.1 The number of cases has increased enormously, arising from more than 100 tropical and subtropical countries of the world, with 2.5 billion people living in areas of risk.2 Dengue virus type 3 (DENV-3) was isolated in Brazil for the first time in December 2000. During the DENV-3 epidemic of 2002, the number of dengue hemorrhagic fever (DHF)/dengue shock syndrome (DSS) cases and deaths exceeded the number of these cases caused by other serotypes of dengue virus in other epidemics.3 This virus type is also more related to neurovirulence.4

Several neurological manifestations have been described each year, however the prevalence of neurological manifestations is unknown. Atypical neurological complications have been reported as acute disseminated encephalomyelitis,5 neuromyelitis optica,6 and Guillain–Barré syndrome, the last one with poor or without dengue infection signs before the neurological picture.7 Encephalitis is the most common neurological manifestation associated with dengue infection. Unlike other viral infections, meningitis determined by dengue infection is a rare complication.8

We describe a case of meningitis caused by an oligosymptomatic dengue infection that occurred during the dengue epidemic of 2007–2008, in the city of Rio de Janeiro, Brazil.

Case report 

A 24-year-old woman complained of a frontal, throbbing headache, refractory to analgesic drugs. One day after the onset of pain, she developed fever and vomiting. The symptoms remained for four days. There was no myalgia, joint pain, bleeding, diarrhea, cutaneous rash, or past history of dengue. A neurological examination was normal, except for the presence of nuchal rigidity +/4+. Hemogram/platelet count and biochemistry were normal. A brain computed tomography scan did not show any abnormalities. Cerebrospinal fluid (CSF) was collected by lumbar puncture on the fourth day of symptoms and showed 362×106 cells/l (66% of lymphocytes), glucose of 86mg/dl, and total protein of 144mg/dl. On analysis of the CSF, bacterial/fungal cultures and immunological testing for syphilis (VDRL) were negative, as well as human T-lymphotropic virus I (HTLV-I), herpes simplex virus (HSV), varicella zoster virus, cytomegalovirus (CMV), and HIV1/2 antibodies (by ELISA). During hospitalization the serum and CSF were dengue IgM and IgG antibody negative (ELISA, Panbio, Australia). PCR for dengue was positive in the serum, for virus type 3. This was performed using a reverse transcriptase-PCR (RT-PCR) following the procedure described by Kumaria and Chakravarti9 (Figure 1). PCR for HSV, CMV, and enterovirus were negative. After analgesic treatment and venous hydration, the patient's symptoms disappeared completely.

View Large Image Download to PowerPoint

Figure 1. Dengue type 3 detection by RT-PCR. Lane 1, molecular standard; lane 2, patient sample (dengue type 3); lane 3, positive dengue 3 control; lane 4, positive dengue 2 control; lane 5, negative control.

Discussion 

Our case occurred during the dengue epidemic of 2007–2008 and did not have the typical infection symptoms of dengue. There was no history of muscle pain, diarrhea, joint or abdominal pain, petechiae or purpura, thrombocytopenia, rash, or bleeding. The main symptom was the severe headache, which was refractory to analgesics and throbbing, determining hospitalization. According to the literature,10 headache is described as extremely severe in 51.8% of dengue cases and severe in 27.1%. However, CSF analysis is not used as a routine exam to differentiate a common dengue headache from meningitis. This fact makes dengue meningitis underestimated. Our case, besides the intense headache, also presented nuchal rigidity, suggesting a meningeal involvement.

To our knowledge, there have been no other descriptions of meningitis caused by dengue virus type 3 infection, since only dengue encephalitis has been reported. This is the first description of the disease in adults, confirmed by PCR. The hypothesis of meningismus was discarded, since the CSF sample showed active inflammation, characterized by pleocytosis and hyperproteinorachia. Although the virus was not found in the CSF, its presence in the serum confirms the diagnosis. Dengue PCR has a sensitivity of 98–99% and a specificity of 100% in serum during the first five days of symptoms.11 The pathogenesis of dengue in the central nervous system is still unknown. As a hypothesis, the virus would invade the central nervous system. The breakdown of the blood–brain barrier during dengue infection in mice, determining virus invasion, was demonstrated by Chaturvedi et al.12 However, the sensitivity and specificity of the PCR technique for the CSF have not been established. Hence, a low concentration of the virus in the CSF may not be detected. The negativity of dengue IgM antibody was expected because the blood and CSF samples were collected before the seventh day after the onset of symptoms. Dengue IgM antibody can appear, in half of the patients, within 2–3 days of defervescence.13

Other causes of viral meningitis, mainly that as a result of enteroviruses, should be investigated in the differential diagnosis. Enteroviruses are responsible for 85% of viral meningitis cases, being usually benign and associated with a broad range of diseases, from an undifferentiated febrile illness to distinct clinical syndromes (herpangina, hand-foot-mouth syndrome, and pleurodynia).14 Our case had non-specific clinical characteristics and the enterovirus diagnosis could only be discarded by the negative PCR test.

In conclusion, our report demonstrates that CSF analysis is necessary to differentiate headache due to systemic infection from that due to meningitis. In addition, meningitis can be the first manifestation of dengue infection. In endemic areas, dengue infection should be suggested and tested as a probable etiological agent of meningitis.

Conflict of interest: No conflict of interest to declare.