Prevalence and clinical differentiation of dengue fever in children in northern India.

STUDY: Although an increasing trend in outbreaks of dengue infection is seen in the northern Indian plains, the importance of dengue infection as a cause of acute undifferentiated febrile illness (AUFI) round the year is not known and the validity of clinical signs and simple laboratory parameters in differentiating dengue from other causes of AUFI has been sparsely reported.

OBJECTIVES: To study the prevalence of dengue infection as a cause of AUFI seen round the year and validity of clinical and simple laboratory features for its diagnosis.

METHODS: Consecutive children between 6 months and 12 years of age presenting to outpatients on 3 predecided weekdays with complaints of fever of 15 days or less duration and having no localizing signs of infection were enrolled over a 1-year period. Blood counts, liver function tests and ELISA test for dengue IgM were performed besides other investigations. Those testing positive for IgM were considered "probable dengue" (PD) while those with negative IgM tested after 5 days of illness were considered "non-dengue" (ND). Clinico-laboratory features were compared between PD and ND. A randomly selected subsample of IgM +ves was tested for dengue genome by real time PCR assay.

RESULTS: Of 298 children enrolled over 1 year, 56 (18.8%) tested positive for dengue IgM and 132 were ND. Comparing PD and ND, age, duration of illness at presentation, rash, bleeding manifestations, vomiting, platelet count, liver transaminases, serum proteins, albumen and bilirubin were significant features on univariate analysis. On logistic regression younger age, rash and higher serum alanine transaminase (sALT) levels were the only significant independent predictors for PD. Taking cutoff of age as 60 months or less and sALT > 40 units, one or more of these features were seen in 50/56 PD cases (sensitivity 89.3%). All of these were present in only 1 of 132 ND cases (specificity 99.2%). Randomly selected 44 of the 56 IgM +ve patients were subjected to PCR assay, of which 15 were positive.

CONCLUSIONS: Dengue transmission occurs round the year in this region. The cause of AUFI was definitely dengue in 15/298, i.e., 5% cases and likely to be dengue in another 13.8% cases. In children presenting to outpatients here with AUFI, young age, rash, and raised sALT are significant independent pointers to dengue. A combination of clinical and laboratory features including liver enzymes could be used to achieve high sensitivity and specificity. These results should be validated in a separate data set.