Attached
Background: Dengue causes considerable morbidity and mortality in Sri Lanka. Immune mediated and cytokine related factors
contribute to its evolution from an asymptotic infection to severe
forms of dengue. Previous studies have analysed the association of
individual cytokines with clinical disease severity. In contrast, we
have viewed this evolution to severe dengue as the behaviour of
a complex dynamic system. We therefore analysed the combined
effect of multiple cytokines that interact dynamically with each
other in order to generate a mathematical model to predict the
occurrence of severe dengue. We expect this to have predictive
value in detecting severe cases and improve outcomes.
Methods & Materials: We analysed data on 11 adult patients
with dengue fever (DF) and 25 patients with dengue haemorrhagic
fever (DHF) recruited from the Colombo South Teaching Hospital,
Sri Lanka. Platelet activating factor (PAF), sphingosine 1- phosphatase (S1P), IL1, TNF and IL10 were used as the cytokine
parameters for the model. Hierarchical clustering was used to
detect factors that correlated with each other. Their interactions
were mapped using Fuzzy Logic mechanisms with the combination
of Hamacher and OWA operators.
Results: Clustering indicated that S1P and IL1 levels were associated with each other. Since, PAF, IL-10 and TNF- have shown to
associate with severe dengue, they were combined together by allocating these cytokines a higher prominence in the model. Operator
value below 0.3 in the overall model correctly predicted development of DHF with 76.6% accuracy. A region of ambiguity was
detected in the model for the value range 0.35 to 0.55. However,
in six instances patients with DHF indicated operator values above
0.6 and in four instances, patients with DF showed operator values
below 0.35. The accuracy of this model in predicting severe dengue
was 76.19% at 96 hours from the onset of illness, 75% at 108 hours
and 74.07% at 120 hours.
Conclusion: The results show a robust mathematical model that
explains the evolution of dengue infection to its serious forms. This
model should be further improved by including additional parameters and be validated on other data sets
