Meta-analysis of the association between significant factors and the risk of DSS.

Pooled ORs with corresponding 95% CIs of the published results were calculated where more than one study had investigated the marker.
*Factor was presented as both dichotomous (frequency of higher values) and continuous (higher value) variables.
#Factor was presented as a dichotomous (frequency of higher values) variable.
†Factor was presented as a continuous variable.
aadjusted odds ratio calculated after the addition of potential missing studies using the trim and fill method of Duvall and Tweedie.
Hemoconcentration was defined as an increase of hematocrit and presented as both dichotomous (frequency of higher values) and continuous (higher value) variables.
Neurological signs (any signs) were defined as patients had any signs of convulsion, decreased consciousness, drowsiness, and lethargy.
A particular dengue serotype infection was defined as a dichotomous variable versus infection with another DENV serotype (e.g., DENV-2 vs. non-DENV-2). Only studies investigated all four strains were included for the analysis.