Abstract
INTRODUCTION: A prospective observational study was conducted to identify simple characteristics that can be used to predict the development of dengue hemorrhagic fever (DHF) in resource-limited settings. METHODS: All patients admitted to a tertiary care hospital in Sri Lanka with confirmed dengue infection, based on the nonstructural protein 1 antigen test during the 2017 dengue outbreak, were included. Sociodemographic, clinical, and laboratory investigation data collected during the first 10 days of fever were statistically compared between DHF and non-DHF patients using Student's t-test, odds ratio (OR), and confidence interval (CI). RESULTS: Of the 490 dengue patients monitored, 244 (49.8%) developed DHF. A significantly high incidence of diabetes mellitus (OR: 2.23; 95% CI: 1.09-4.54), respiratory tract allergies (OR: 1.95; 95% CI: 1.24-3.05), recurrent upper respiratory tract infections (OR: 3.58; 95% CI: 2.29-5.59), past history of dengue (OR: 2.56; 95% CI: 1.04-6.28), urban living (OR: 1.52; 95% CI: 0.84-2.78), active lifestyle (OR: 11.44; 95% CI: 7.50-17.46), along with a rapid reduction in daily platelet (~34%) and white blood cell (WBC) (~25%) counts, and rise of serum alanine transaminase (ALT) (~67%) during the initial few days of fever were detected in DHF cases. CONCLUSIONS: Comorbidities, living conditions, and physical activity levels appear to be associated with the development of DHF. A reduction of over 25% in platelet and WBC counts and a rise of more than 50% in ALT compared to the previous day's values may help in the early detection of DHF.