Temperature as a Predictor of Neonatal Sepsis

Abstract

In 2016, approximately 6 million children died worldwide. A total of 1 million of these children died within the first 24 h of their birth. Majority of these deaths are due to infectious diseases like sepsis. Blood culture is used to diagnose sepsis. But these methods cannot be used in the areas where medical facilities do not exist. Therefore, we require simple ways to predict sepsis. In this paper, we calculated the prediction capability of temperature to predict sepsis. The sensitivity of temperature minimum, temperature maximum and temperature variation were 54.96%, 46.56% and 64.12%, respectively, whereas specificity were 79.09%, 75.45% and 76.82%, respectively. In conclusion, a decent accuracy was achieved using temperature as a sole predictor of sepsis. We speculate that better results can be obtained if we study temperature patterns as predictors.