A systematic review and meta-analysis of predictive accuracy of clinically relevant biomarkers of severe dengue.

BACKGROUND

OBJECTIVES

Even though commercially available assays can diagnose dengue fever at the point of care, no prognostic biomarkers accurately predict the possibility of severe dengue. Predictive biomarkers may lower dengue-associated morbidity and mortality by timely management. A systematic review and meta-analysis were performed to analyze the predictive value of clinically relevant biomarkers of severe dengue.

METHODS

The study protocol was designed based on updated PRISMA guidelines. An electronic search of PubMed/MEDLINE, Scopus and Embase was carried out using specific search terms for articles analysing clinically significant biomarkers of dengue across all age groups between January 2000 and June 2025. The meta-analysis component was modified to synthesize the pooled area under the curve (AUC) of ROC (Receiver Operating Characteristic) with a 95% confidence interval (CI).

RESULTS

Eleven cross sectional studies reporting AUC of ROC curve with 95% confidence interval, of nine biomarkers in serum samples from Thailand, India, Mexico, and Sri Lanka were qualified for the quantitative synthesis. The overall pooled AUCROC curve value of selected biomarkers was 0.86 (95% CI 0.81-0.90). The pooled AUCROC curve value of serum ferritin was 0.88 (95% CI 0.82-0.95).

INTERPRETATION

CONCLUSION

This meta-analysis concludes that elevated serum ferritin levels within the initial 4-5 days of fever among dengue patients predict progression to severe dengue.