Prophylactic platelet transfusion may not reduce the risk of clinical bleeding in adults with dengue and thrombocytopenia

EBM Focus - Volume 12, Issue 12

Read the full EBM Focus and earn CME credit.

Reference - Lancet 2017 Mar 7 early online (level 2 [mid-level] evidence)

  • Prophylactic platelet transfusion for patients with dengue infection and thrombocytopenia is not uncommon despite a lack of recommendations advocating its use and a lack of evidence assessing its efficacy.
  • A recent open-label trial randomized 372 adults (75% men) with dengue and ≤ 20,000 platelets/mcL to prophylactic platelet transfusion plus supportive care vs. supportive care alone.
  • Rates of clinical bleeding between the two groups were similar (21% vs. 26%, with a 95% CI for difference -15.08% to +5.34%), as were rates of severe dengue (5% vs. 4%) and other outcomes. However, a potential benefit of transfusion cannot be excluded, and these results may not apply to children, women, or adults with very low platelet levels.

Dengue, a mosquito-transmitted virus, is common in tropical areas of southeast Asia and the Americas and has increased in prevalence over the past decades in those areas as well as non-endemic areas including the United States and Europe (Nature 2013 Apr 25;496(7446):504, Nat Rev Dis Primers 2016 Aug 18;2:16055). Symptoms of infection are usually mild, including self-limited fever, but can progress to a severe manifestation with impaired hemostasis and organ failure; supportive care is the mainstay of treatment (Nat Rev Dis Primers 2016 Aug 18;2:16055). Transient thrombocytopenia is common, but prophylactic platelet transfusion has not been recommended due to a dearth of evidence assessing its efficacy (WHO Handbook for the Clinical Management of Dengue 2012). Nevertheless, a survey of about 300 physicians who have managed patients with dengue revealed that almost 22% would consider transfusion for platelet counts ≤ 20,000 platelets/mcL in the absence of bleeding (PLoS Negl Trop Dis 2012;6(6):e1716).

To directly assess the efficacy of prophylactic platelet transfusion, a recent open-label trial randomized 372 adults (mean age 45 years, 75% men) in Singapore and Malaysia with dengue and thrombocytopenia (≤ 20,000 platelets/mcL) to prophylactic platelet transfusion plus supportive care vs. supportive care alone. Transfusion consisted of 4 units of pooled platelets daily for each day with ≤ 20,000 platelets/mcL. Patients had dengue for a mean of 5 days at the time of randomization, and warning signs for risk of severe dengue were present in 95%. A history of mild clinical bleeding (mostly transient gingival or nose bleed) was reported in 13% of patients allocated to transfusion and 19% allocated to supportive care alone (statistical comparison not reported). Patients were excluded for persistent mild bleeding, any severe bleeding, pregnancy, breastfeeding, recent peptic ulcer or anticoagulant use, and liver, kidney, autoimmune, and hematological conditions. The primary outcome was clinical bleeding excluding petechiae within 7 days or by hospital discharge (whichever occurred earlier), and patients were also assessed at 21 days post-randomization.

Clinical bleeding by day 7 or hospital discharge occurred in 21% of patients with platelet transfusion added to supportive care vs. 26% with supportive care alone (not significant, 95% CI for difference -15.08% to +5.34%). Most of the bleeding consisted of gingival bleeding, epistaxis, and hematoma. Similar percentages were observed in 77 patients with baseline platelet count ≤ 10,000 platelets/mcL. Rates of bleeding by day 21 between the two groups were also similar, with clinical bleeding in 22% vs. 27% and severe bleeding in 2% vs. 4%. The two groups also had similar rates of disease progression, with severe dengue in 5% vs. 4%, plasma leakage (pleural effusion, ascites, or ≥ 20% increase in serum hematocrit) in 11% vs. 10%, and dengue hemorrhagic fever or dengue shock syndrome in 1% vs. 1%. In 188 patients who had transfusion (taking into account 11 patients who crossed over from their allocated groups), adverse events possibly related to transfusion occurred in 4.8%, including 1 patient each with anaphylaxis, transfusion-related acute lung injury, and fluid overload. No patients died or required packed red blood cell or whole blood transfusion.

This trial showed that the addition of prophylactic platelet transfusion to supportive care may not reduce the risk of clinical bleeding in adults with dengue and ≤ 20,000 platelets/mcL. However, the wide confidence interval in risk difference cannot exclude a 15% difference in favor of transfusion, and fewer patients in the transfusion group had a history of mild clinical bleeding. Generalizability to important patient groups is limited. For example, the trial excluded children, and only 25% of the patients were women, who might have an increased risk of bleeding (Southeast Asian J Trop Med Public Health 2012 Jul;43(4):890). Also, it is unclear if there is a platelet count threshold below which patients may benefit from platelet transfusion. The subgroup analysis of 77 patients with ≤ 10,000 platelets/mcL suggests that any such threshold is likely lower than 10,000, but this supposition is tempered by the low number of patients analyzed. Despite these limitations, this trial offers the best evidence to date in support of previous studies - mostly observational or small - showing a lack of benefit, and some harm, with prophylactic platelet transfusion for dengue with thrombocytopenia.

See the Dengue topic in DynaMed Plus for more information. DynaMed users click here.