Clinical criteria to diagnose dengue in its early stages.

Abstract

Introduction. Clinical differentiation of dengue from other diseases with similar symptoms is difficult. The case definition of the World Health Organization (WHO) has high sensitivity but its specificity is very low.
Objective. A diagnostic scale was formulated for early clinical diagnosis of dengue that provided greater accuracy than that of the WHO definition.
Materials And Methods. A cohort of 251 adults (> 12 years of age) with unspecific acute febrile syndrome was selected from clinics located in Bucaramanga, Colombia. They consisted of 125 cases of dengue (serologically and/or virologically confirmed) and 126 with other febrile diseases. Clinical manifestations encountered during the first four days of dengue disease were determined, along with the diverse diagnostic combinations that were presented.
Results. : The scale consisted of the following criteria: presence of rash, positive tourniquet test, absence of nasal discharge, arthralgias, absence of diarrhea (1 point for each finding), leukocyte count < 4,000/mm3 (3 points) and platelet count < 180.000/mm3 (2 points). In a receiver-operating-characteristic curve, the predictive area of 81.0% was significantly superior to the one produced with WHO criteria, (70.0%, p < 0.001). Febrile syndrome with at least a 3 point score obtained the following values: sensitivity = 95.2%; specificity = 27.8%; positive predictive value = 56.7%; negative predicative value = 85.4%. With a 6-point score, sensitivity = 70.4%; specificity = 78.6%; positive predictive value = 76.5%; negative predicative value = 72.8%. With at least 8 points: sensitivity = 42.4%; specificity = 96%; positive predictive value = 91.4%; negative predictive value = 62.7%. With 9 or 10 points, specificity and positive predictive value were of 100%.
Conclusion. The described scale proved useful for early clinical diagnosis of dengue, but requires validation for its application in endemic areas.