DYNAMIC HEMATOLOGICAL CHANGES AS EARLY PREDICTORS OF SEVERE DENGUE: A COMPARATIVE STUDY IN ISLAMABAD DURING THE 2024–2025 OUTBREAKS

Abstract

Background: Dengue fever is a major and recurring public health challenge in Islamabad, Pakistan, with seasonal outbreaks causing significant morbidity and healthcare burden. Hematological parameters, including platelet count, haematocrit, and neutrophil-to-lymphocyte ratio (NLR), are central to dengue diagnosis and severity assessment, but most existing research relies on single-timepoint laboratory values that may fail to capture the dynamic changes preceding clinical deterioration. This study aimed to evaluate and compare hematological trends in laboratory-confirmed dengue patients admitted during the 2024 and 2025 outbreak seasons in Islamabad, to determine their predictive value for progression to severe dengue per WHO 2009 criteria, and to examine sex-based differences in these parameters across both outbreak years.

Methods: A retrospective observational comparative cohort design was employed. Data from laboratory-confirmed dengue patients (NS1, PCR, or IgM positive), aged ≥1 year, admitted between September and November of 2024 and 2025 at major hospitals and diagnostic centres in Islamabad were analysed (minimum calculated sample size = 300, derived using the Events Per Variable principle). Platelet count, haematocrit, NLR, haemoglobin, and white blood cell (WBC) count were extracted from hospital laboratory information systems and stratified by sex and outbreak year. Patients with pre-existing hematological disorders, co-infections, or incomplete laboratory records were excluded. Group means were compared using Welch's independent-samples t-test.

Results: A total of 309 laboratory-confirmed dengue patients (186 male, 123 female) were analysed. Male patients in 2024 had a mean platelet count of 145,702 cells/µL and haematocrit of 41.50%, with a mean NLR of 4.72; in 2025, male patients showed significantly higher mean platelets (174,630 cells/µL; p<0.05) and a numerically higher NLR (5.12). Female patients had lower mean haematocrit than males in both years (35.70% in 2024; 37.08% in 2025), and haematocrit and haemoglobin differed significantly between sexes in 2024 (p<0.001 for both). The 2025 cohort showed a modest improvement in platelet counts across both sexes compared with 2024, suggesting year-to-year variation in outbreak severity.

Conclusion: Dynamic hematological monitoring — particularly serial platelet counts, NLR trajectory, and haematocrit variability — offers prognostic information beyond single-timepoint assessment in dengue patients. Sex-based differences in haematocrit and platelet trends support the use of sex-stratified clinical thresholds, and year-to-year variation between the 2024 and 2025 Islamabad outbreaks supports the hypothesis that viral or immunological factors contribute to inter-outbreak differences in severity. A validated, locally derived hematology-based risk score represents a practical, cost-effective tool for early triage of high-risk dengue patients in resource-constrained settings.