Platelet transfusion in dengue fever: a randomized controlled trial

Abstract

Background: Immune mediated destruction of platelets can possibly lead to poor response to platelet transfusion in dengue fever.Weconducted this trial to study effects of platelet transfusion in dengue-related thrombocytopenia. Methods: A single center, randomized, non blind trial was conducted at Jinnah Hospital Lahore between August and October 2011. Adults of the age ≥14 years with dengue fever or dengue hemorrhagic fever and platelet counts <30,000/mm3 were eligible.Patients were randomized to treatment and control group. Treatment group received single donor platelets. Responsiveness to platelet transfusion was assessed by 1 hr post-transfusion platelet increments (PPI) and corrected count increments (CCI). Patients with PPI ≥10,000/mm3 and/or CCI≥5,000/mm3 1 hr post-transfusion were considered responders. Primary outcome measurewas platelet count increments at 24 and 72hrs. Secondary outcome measure was duration of bleeding. Results: Eighty seven (87) patientswere enrolled and43 (48.2%) received platelet transfusion. Mean PPI and CCI at 1 hr posttransfusion in treatment group were 18,800/mm3 (range -7,000 to 77,000) and 7,000/mm3 (range -1,050 to 27,260) respectively. 22 (53.6%) patients in treatment groupwere non-responders.Mean platelet incrementswere higher in treatment group as compared to control group at 24hrs (34,780/mm3 vs 4,280/mm3, p<0.001) and 72hrs (75,430/mm3 vs 32,840/mm3, p=0.001). At 24hrs, responders showed significantly higher increments (mean 53,310/mm3) than non-responders (mean 18,770/mm3) and controls (mean 4,280/mm3) (p=0.004, p <0.001 respectively). Significant difference was also found between non-responders and controls (p <0.001). Similarly, 72hrs post-transfusion platelet increments were significantly higher in responders (mean 103,440/mm3) as compared to non-responders (mean 51,430/mm3) and control group (mean 32,840/mm3) (p=0.001, p <0.001 respectively). However, no significant differences were found between nonresponders andcontrol group (p=0.104). Three severeanaphylactic reactions including two deaths occurred in treatment group. Both deathswere innon-responders.No significantdifferencewas found in duration of bleeding between treatment and control group. Conclusion: Although half of the patients show significant improvement in platelet count following platelet transfusion, no beneficial effect occurs on duration of bleeding. Platelet transfusion in dengue fever is associated with high incidence of severe complications including death and should not be routinely done. The risk of complications is higher in non-responders.