Dengue fever is a significant mosquito-borne viral disease in the tropics and subtropics, and remains a seasonal challenge to emergency and in-patient care. The disease spectrum is from simple febrile illness to dengue with warning signs, plasma leakage, severe bleeding and shock. Dengue incidence has increased significantly, due to urbanization, travel, vector adaptation and climate-sensitive transmission dynamics, as revealed by global mapping studies and clinical reviews [1],[2]. The haematological profile of dengue is unique. Leukopenia, hemoconcentration and thrombocytopenia are common findings and are often helpful in aiding clinical diagnosis and monitoring. Thrombocytopenia can be caused by marrow suppression, immune-mediated platelet destruction, peripheral consumption, and activation of endothelial cells and direct interaction between the virus and megakaryocyte-platelet pathways [3],[4]. Platelet count is a routine parameter monitored in dengue but its interpretation is complex. A very low platelet count is a concern for bleeding and severe disease, but bleeding is also affected by vascular permeability, coagulopathy, liver dysfunction, platelet activation and host immune response. Guidelines therefore discourage giving platelets as a prophylactic in otherwise stable patients, based on platelet counts alone, and stress careful clinical monitoring [5],[6]. Platelet indices, immature platelet fraction and platelet kinetics have been used in previous studies to assess recovery or severity. But, there are still many hospitals that rely on absolute platelet count as the primary lab parameter for admission, referral or transfusion. The present study evaluated the correlation between platelet count and severity of dengue in confirmed cases and determined a clinically useful threshold for severe dengue [7].

This is a prospective observational study in the medicine department of tertiary care hospital during one dengue season. After obtaining written informed consent, consecutive patients aged 15 years and older with acute febrile illness and positive NS1 antigen and/or dengue IgM serology were enrolled. Patients with chronic liver disease, chronic kidney disease, pregnancy, known platelet disorder, hematological malignancy, recent chemotherapy, and mixed infections like malaria, leptospirosis or enteric fever were excluded. The final sample was 160 patients with dengue. Clinical classification was based on presence of fever, abdominal pain, persistent vomiting, mucosal bleeding, lethargy, hepatomegaly, fluid accumulation, shock or severe organ involvement as dengue without warning signs, dengue with warning signs or severe dengue. The demographic data, day of illness, clinical symptoms, vital signs, bleeding manifestation, fluid therapy, transfusion requirement, length of stay and outcome were recorded. CBC was done every 24 hours until recovery or until the patient was discharged. Hematocrit rise was determined from baseline or lowest stable hematocrit. Liver enzymes and renal function were measured as indicated. The data were analyzed statistically using SPSS version 26. Continuous variables were compared using ANOVA or Kruskal-Wallis test depending on distribution. Chi-square test was used to compare categorical variables. Spearman correlation was used to determine the correlation between platelet count and severity grade. The threshold of platelet count in severe dengue was identified using ROC curve analysis. A p-value <0.05 was considered significant

Among 160 patients, 72 (45.0%) had dengue without warning signs, 61 (38.1%) had dengue with warning signs and 27 (16.9%) had severe dengue. The mean age was 31.6 ± 12.8 years, and 94 patients (58.8%) were male. Most patients presented between day 4 and day 6 of illness. Severe dengue patients had a higher frequency of abdominal pain, persistent vomiting, mucosal bleeding and hypotension. Table 1. Clinical profile according to dengue severity category. Variable Without warning signs (n=72) With warning signs (n=61) Severe dengue (n=27) p-value Age (years) 30.4 ± 11.6 32.1 ± 13.4 33.6 ± 13.7 0.518 Male sex, n (%) 40 (55.6) 36 (59.0) 18 (66.7) 0.593 Day of illness at admission 4.8 ± 1.2 5.1 ± 1.3 5.4 ± 1.1 0.084 Abdominal pain, n (%) 12 (16.7) 29 (47.5) 19 (70.4) <0.001 Mucosal bleeding, n (%) 6 (8.3) 13 (21.3) 14 (51.9) <0.001 Hypotension, n (%) 0 (0.0) 5 (8.2) 13 (48.1) <0.001 Mean stay (days) 3.1 ± 1.0 4.4 ± 1.5 6.8 ± 2.3 <0.001 Table 2. Hematological parameters across severity groups. Parameter Without warning signs With warning signs Severe dengue p-value Lowest platelet count (×10^3/µL) 102.4 ± 44.6 68.1 ± 31.2 33.7 ± 16.8 <0.001 Platelet nadir day 5.2 ± 1.0 5.6 ± 1.1 6.1 ± 1.3 0.002 WBC count (×10^3/µL) 3.8 ± 1.4 3.2 ± 1.2 2.9 ± 1.0 0.006 Peak hematocrit (%) 39.6 ± 4.8 43.1 ± 5.2 47.8 ± 6.0 <0.001 AST (U/L) 72 ± 44 124 ± 81 218 ± 136 <0.001 ALT (U/L) 58 ± 36 92 ± 64 169 ± 104 <0.001 Table 3. Diagnostic performance of platelet thresholds for severe dengue. Platelet threshold Sensitivity (%) Specificity (%) AUC p-value ≤100 ×10^3/µL 96.3 43.6 0.70 <0.001 ≤75 ×10^3/µL 88.9 61.7 0.75 <0.001 ≤50 ×10^3/µL 81.5 74.4 0.82 <0.001 ≤25 ×10^3/µL 37.0 93.2 0.65 0.004 Spearman analysis showed an inverse correlation between platelet count and severity grade (r=-0.64, p<0.001). Platelet count also correlated inversely with hematocrit rise (r=-0.41, p<0.001), AST level (r=-0.36, p<0.001) and length of hospital stay (r=-0.45, p<0.001). Twenty-one patients received platelet transfusion. Transfusion was given for active bleeding in 15 patients and for very low platelet count with hemodynamic instability in 6 patients. No mortality occurred, but 9 severe dengue patients required intensive monitoring

This study revealed that there was a significant negative correlation between platelet count and severity of dengue. Severe dengue patients had the lowest platelet nadir, delayed platelet recovery, higher liver enzyme elevation and hemoconcentration. The results are in line with the biological knowledge that thrombocytopenia is due to decreased production and increased destruction and peripheral consumption in the critical phase of dengue [8],[9]. The optimal combination of sensitivity and specificity for severe dengue in our study was obtained with platelet count <50 ×10^3/µL. The specificity, however, was not high enough to allow platelet count to be used as a single predictor. Patients with very low counts had no bleeding, while others with higher counts had warning signs of plasma leakage. This agrees with the idea of using platelet count in conjunction with trends in the hematocrit, hemodynamics and clinical warning signs [10],[11]. The frequency of bleeding did increase significantly with decreasing platelet count, but bleeding was not solely dependent on platelet count. Previous research has demonstrated that platelet dysfunction, endothelial injury, activation of complement, coagulation abnormalities and hepatic involvement are factors that increase the risk of bleeding. Thus, giving routine prophylactic platelet transfusion to stable dengue patients without bleeding may not yield any benefit and may pose unnecessary risks to patients [12],[13]. There is a correlation between thrombocytopenia and increased AST/ALT, indicating that liver involvement is associated with more severe systemic disease. Platelet count, hematocrit, leukocyte count, transaminases and hypotension have been combined into similar scoring systems to enhance the prediction of complications. These are more clinically useful than platelet counts alone, particularly in the busy hospital setting during outbreaks [14],[15]. Platelet indices and immature platelet fraction may give further information concerning marrow recovery and platelet turnover. Recent studies have reported that MPV, PDW and IPF are dynamic markers during dengue and can predict recovery before the absolute platelet count increases. These measures may be valuable where advanced analyzers are available [16]. Single season recruitment and absence of viral serotype testing were limitations of the study. However, repeated platelet monitoring and the use of a standardized severity classification further support the results. Further multicenter evaluations with fluid leakage markers and viral load should be performed to develop more accurate severity prediction models [17]. The nadir of platelets was also significant. Most patients had the lowest counts on days 5-6 of illness, when plasma leakage and warning signs are most likely to occur. If the patient presents early, a single platelet count may thus be misleading. A series of trends is safer than a single measurement, especially if clinical symptoms are changing. The results justify a risk-stratified strategy that depends on the platelet count, hematocrit trend, oral intake, abdominal symptoms, bleeding, comorbidities and the possibility to follow up. Moderate thrombocytopenia (100,000 to 150,000 platelets) with stable vital signs may be treated conservatively with close monitoring, and patients with declining platelets and increasing hematocrit should be monitored more closely. The importance of public health should also be highlighted. In times of outbreaks, hospitals may receive an overwhelming number of referrals based solely on platelet count. Using platelet thresholds along with warning signs in a standardized triage protocol can decrease unnecessary platelet transfusion and target resources to the high-risk patient and help counsel patients about platelet recovery [17]. Patients should be counseled that platelet recovery may start after defervescence, and may be delayed from subjective improvement. Isolated platelet values can cause repeat testing or transfusion requests due to anxiety. Clarifying the anticipated course, alerting to warning signs that need immediate review, and explaining the difference between thrombocytopenia and active bleeding can help to enhance adherence to follow-up and minimize panic-driven admissions. This cohort is free of mortality, which may be due to timely recognition, supportive care and referral of high-risk patients. However, the need for intensive monitoring in severe dengue indicates that low platelet count with shock, bleeding or organ involvement is clinically significant. Laboratory markers and bedside parameters should be included in future hospital protocols for greater uniformity of care [17].

The platelet count decreased significantly as the severity of the disease increased and there was strong inverse correlation with warning signs, hemoconcentration and hospital stay. A platelet count of ≤50 ×10^3/µL was a good indicator for severe dengue, but not a substitute for clinical evaluation. The best way to monitor dengue patients is by combining the serial platelet trend with the hematocrit and warning signs

1. Bhatt S, Gething PW, Brady OJ, et al. The global distribution and burden of dengue. Nature. 2013;496(7446):504-507. doi:10.1038/nature12060. PMID:23563266. 2. Simmons CP, Farrar JJ, Nguyen vV, Wills B. Dengue. N Engl J Med. 2012;366(15):1423-1432. doi:10.1056/NEJMra1110265. PMID:22494122. 3. Guzman MG, Gubler DJ, Izquierdo A, Martinez E, Halstead SB. Dengue infection. Nat Rev Dis Primers. 2016;2:16055. doi:10.1038/nrdp.2016.55. PMID:27534439. 4. Martina BEE, Koraka P, Osterhaus ADME. Dengue virus pathogenesis: an integrated view. Clin Microbiol Rev. 2009;22(4):564-581. doi:10.1128/CMR.00035-09. PMID:19822889. 5. Srichaikul T, Nimmannitya S. Haematology in dengue and dengue haemorrhagic fever. Baillieres Best Pract Res Clin Haematol. 2000;13(2):261-276. doi:10.1053/beha.2000.0073. PMID:10942625. 6. de Azeredo EL, Monteiro RQ, de Oliveira Pinto LM. Thrombocytopenia in dengue: interrelationship between virus and the imbalance between coagulation and fibrinolysis and inflammatory mediators. Mediators Inflamm. 2015;2015:313842. doi:10.1155/2015/313842. PMID:25999666. 7. Khazali AS, et al. Thrombocytopenia in dengue infection: mechanisms and a clinical perspective. Rev Med Virol. 2024. PMID:39397710. 8. Hottz ED, Bozza FA, Bozza PT. Platelets in immune response to virus and immunopathology of dengue. Front Med (Lausanne). 2018;5:121. doi:10.3389/fmed.2018.00121. PMID:29740515. 9. Lye DC, Lee VJ, Sun Y, Leo YS. Lack of efficacy of prophylactic platelet transfusion for severe thrombocytopenia in adults with acute uncomplicated dengue infection. Clin Infect Dis. 2009;48(9):1262-1265. doi:10.1086/597773. PMID:19323630. 10. Malavige GN, Fernando S, Fernando DJ, Seneviratne SL. Dengue viral infections. Postgrad Med J. 2004;80(948):588-601. doi:10.1136/pgmj.2004.019638. PMID:15466994. 11. Khatri S, Sabeena S, Arunkumar G, Mathew M. Utility of platelet parameters in serologically proven dengue cases with thrombocytopenia. Indian J Hematol Blood Transfus. 2018;34(4):703-706. doi:10.1007/s12288-018-0924-2. 12. Lee TH, Wong JGX, Leo YS, et al. Potential harm of prophylactic platelet transfusion in adult dengue patients. PLoS Negl Trop Dis. 2016;10(3):e0004576. doi:10.1371/journal.pntd.0004576. PMID:26986619. 13. World Health Organization. Dengue: guidelines for diagnosis, treatment, prevention and control. Geneva: WHO; 2009. PMID:23762963. 14. Bhaskar M, et al. Predictive scoring system for risk of complications in dengue fever. J Family Med Prim Care. 2022. PMID:35860477. 15. Logia P, et al. Predictors of clinically significant bleeding in adult dengue patients. Cureus. 2023. PMCID:PMC10701565. 16. Sontakke RA, Aglave NR, Dua H. Correlation of platelet parameters with the severity of thrombocytopenia in dengue fever in children aged less than 18 years. Cureus. 2024;16(3):e56829. doi:10.7759/cureus.56829. PMID:38654784. 17. Looi KW, Matsui Y, Kono M, et al. Evaluation of immature platelet fraction as a marker of dengue fever progression. Int J Infect Dis. 2021;110:187-194. doi:10.1016/j.ijid.2021.07.048.