Using Clinical Profiles and Complete Blood Counts to Differentiate Causes of Acute Febrile Illness during the 2009–11 Outbreak of Typhoid and Chikungunya in a Dengue Endemic Area

Abstract

Abstract Background and aims After the 2009–11 outbreak of typhoid and chikungunya (CHIK) in Thailand, an effort was made to use complete blood counts and clinical profiles to differentiate these diseases to facilitate earlier specific treatment. Methods Patients aged 2–15 years having fever on first visit ≤3 days without localizing signs were enrolled retrospectively. Typhoid fever was confirmed by hemoculture, dengue by nonstructural protein-1 or polymerase chain reaction (PCR), and CHIK by PCR. Febrile children with negative results for these infections were classified as other acute febrile illness (AFI). Results Of the 264 cases, 56, 164, 25 and 19 had typhoid fever, dengue viral infection (DVI), CHIK and other AFI, respectively. Arthralgia had sensitivity, specificity, positive predictive value (PPV) and negative predictive value of 0.96, 0.97, 0.80 and 0.99, respectively, to differentiate CHIK from the others. After excluding CHIK by arthralgia, the PPV of the WHO 1997 and 2009 criteria for DVI increased from 0.65 and 0.73 to 0.95 and 0.84, respectively. Children with one of myalgia, headache or leukopenia had sensitivity of 0.84, specificity of 0.76 and PPV of 0.92 to differentiate DVI from typhoid and other AFIs. Patients with one of abdominal pain, diarrhea or body temperature >39.5°C were more likely to have typhoid fever than another AFI with PPV of 0.90. Conclusion Using this flow chart can help direct physicians to perform more specific tests to confirm the diagnosis and provide more specific treatment. Nevertheless, clinical follow-up is the most important tool in unknown causes of febrile illness.