A cohort study to assess the new WHO Japanese encephalitis surveillance standards.

OBJECTIVE To assess the field-test version of the new WHO Japanese encephalitis (JE) surveillance standards. METHODS We applied the clinical case definition of acute encephalitis syndrome (AES), laboratory diagnostic criteria and case classifications to patients with suspected central nervous system (CNS) infections in southern Viet Nam. FINDINGS Of the 380 patients (149 children) recruited with suspected CNS infections, 296 (96 children) met the AES case definition. 54 children were infected with JE virus (JEV), of whom 35 (65%) had AES, giving a sensitivity of 65% (95% CI: 56-73) and specificity of 39% (95% CI: 30-48). Nine adults with JEV presented with AES. 19 JEV-infected children missed by surveillance included 10 with acute flaccid paralysis, two with flaccid hemiparesis and six with meningism only. Altering the case definition to include limb paralysis and meningism improved sensitivity to 89% (95% CI: 83-95), while reducing specificity to 23% (95% CI: 15-30). Six children that did not have AES on admission had reduced consciousness after admission. Cerebrospinal fluid (CSF) analysis diagnosed seven patients negative on serum analysis. Five patients with neurological manifestations of dengue infection had JEV antibodies in serum and would have been misdiagnosed had we not tested for dengue antibodies in parallel. CONCLUSION Children infected with JEV that presented with acute limb paralysis or neck stiffness only were missed by the surveillance standards, although some of them subsequently became encephalopathic. A footnote in the surveillance standards drawing attention to these presentations would be helpful. An acute CSF sample is more sensitive and specific than an acute serum sample.