Risk Factors Associated with Fatal Dengue Hemorrhagic Fever in Adults: A Case Control Study

Abstract

Background. Dengue is endemic in most parts of the tropics with a significant mortality of 1-5%. Although individual case reports and case series have been published, large-scale case controls studies are few. The objective of this study was to find clinical and laboratory predictors of mortality in dengue. Methods. Hospital case record based case control study was performed. Results. Twenty fatalities with 80 controls were analyzed. Clinical parameters of postural dizziness (OR 3.2; 95% CI 1.1-8.9), bleeding (OR 31.9; 95% CI 6.08-167.34), presence of plasma leakage (OR 64.6; 95% CI 7.45-560.5), abdominal tenderness (OR 2.24; 95% CI 0.79-6.38), and signs of cardiorespiratory instability at admission increased the risk of dying from dengue. Altered consciousness was exclusively seen in 20% of cases. Laboratory parameters of elevated CRP (OR 1.652; 95% CI 1.28-2.14), AST, or ALT > 500 IU/L (OR 52.5; 95% CI 12.52-220.1) and acute kidney injury (AKI) (OR 103.5; 95% CI 13.26-807.78) during hospital stay increased the odds of dying. Need for assisted ventilation and multiorgan dysfunction (MOD) were exclusively seen in the cases. Multivariate logistic regression revealed bleeding at admission, AKI, and elevated hepatic transaminase >500 IU/L to be independent predictors of mortality. Conclusions. This case control study revealed that mortality from dengue could be predicted using clinical parameters at admission and low cost routine laboratory investigations.