Trans R Soc Trop Med Hyg 2017; 111: 433–439

doi:10.1093/trstmh/trx079 Advance Access publication 29 January 2018

Prophylactic and therapeutic interventions for bleeding in dengue:

a systematic review

REVIEW
Senaka Rajapaksea,*, Nipun Lakshitha de Silvab, Praveen Weeratungaa, Chaturaka Rodrigoa,c
and Sumadhya Deepika Fernandod

a
Department of Clinical Medicine, Faculty of Medicine, University of Colombo, Colombo 00800, Sri Lanka; bDepartment of Clinical

Downloaded from https://academic.oup.com/trstmh/article/111/10/433/4828073 by guest on 20 January 2021

Medicine, Faculty of Medicine, General Sir John Kotelawala Defence University, Kandawala Estate, Ratmalana, Sri Lanka; cDepartment
of Pathology, School of Medical Sciences, University of New South Wales, Sydney, NSW, Australia; dDepartment of Parasitology, Faculty
of Medicine, University of Colombo, Colombo, Sri Lanka

*Corresponding author: Email: senaka@med.cmb.ac.lk

Received 24 September 2017; revised 7 December 2017; editorial decision 12 December 2017; accepted 19 December 2017

The global incidence of dengue has increased sevenfold between 1990 and 2013. Despite a low case fatality
rate (<1%), during epidemics, due to the large number of people affected, overall mortality rates can be signifi-
cant. The risk of clinically significant bleeding in dengue is unpredictable and often contributes to an adverse
outcome. This systematic review focuses on the evidence for prophylactic and therapeutic interventions for
bleeding in dengue infection. PubMed, CINAHL, Cochrane Library, Embase and Google Scholar were searched for
randomized, quasi-randomized and non-randomized, prospective or retrospective studies that had a control
group alongside an intervention aimed at stopping or preventing bleeding in dengue infection. Eleven studies
that included 1904 patients in 12 study arms were eligible. These assessed the role of platelet transfusion [two
randomized controlled trials (RCTs) and three non-randomized studies], plasma transfusion (one RCT), recombin-
ant activated factor VII (one RCT), anti-D globulin (two RCTs), immunoglobulin (one RCT) and interleukin 11 (one
RCT) as prevention or treatment for bleeding. Due to significant heterogeneity in study design and outcome
reporting, a meta-analysis was not performed. Currently there is no evidence that any of the above interventions
would have a beneficial effect in preventing or treating clinically significant bleeding in dengue.

Keywords: Bleeding, Dengue, Platelets, Systematic review

Introduction to denote severe disease, plasma leakage and not haemorrhage is

Dengue is a flavivirus infection transmitted by Aedes mosquitoes.
It is a disease with global implications and results in considerable
morbidity and mortality. Transmission occurs in at least 128
countries and almost 4 billion people are at risk.1 The total num-
ber of recorded cases of dengue has increased exponentially,
from less than a thousand cases per year globally in the 1950s
to an annual incidence of 58.4 million cases in 2013. South and
South-east Asia account for a major proportion of the global bur-
den of dengue, with an estimated mortality of 1.29 and 8.49 per
million population, respectively, in 2013.2
Dengue encompasses a clinical spectrum of disease, from sim-
ple dengue fever, to severe disease characterized by thrombocyto-
paenia, plasma leakage, bleeding and multi-organ dysfunction.
The most striking manifestation of dengue infection is the plasma
leakage syndrome, which, when severe, leads to shock. In extreme
cases, multi-organ dysfunction syndrome occurs, leading to death.
Although the term ‘dengue haemorrhagic fever’ is commonly used
the hallmark of severe dengue.
Nonetheless, haemorrhage does occur. Bleeding manifestations
in dengue range from minor cutaneous and mucosal bleeding to
deep visceral bleeding, including intracranial, pulmonary, gastro-
intestinal and intraperitoneal haemorrhage. Bleeding manifesta-
tions in dengue are seen in 20–60% of hospitalized patients in
different case series.3–5 Although severe bleeding is generally con-
sidered an unusual manifestation of dengue,4 bleeding is respon-
sible for a significant proportion of fatal outcomes. For example, in
a nationwide retrospective death review in Malaysia from 2013 to
2014, severe bleeding was responsible for 29.7% of deaths.6
The exact pathophysiological mechanism behind bleeding in
patients with dengue infection has not been clearly elucidated.
Several haemostatic abnormalities have been described in patients
with dengue, which in combination might predispose to bleeding.
Severe thrombocytopaenia, prolonged prothrombin time (PT), pro-
longed activated partial thromboplastin time (aPTT), prolonged
thrombin time (TT), reduction of certain coagulation factors,

© The Author(s) 2018. Published by Oxford University Press on behalf of Royal Society of Tropical Medicine and Hygiene. All rights reserved.
For permissions, please e-mail: journals.permissions@oup.com.

433
S. Rajapakse et al.

increased tissue factor level, alterations in the fibrinolytic system
and reduced protein C and protein S levels are some of the abnor-
malities reported in the literature.7–9 It is also recognized that
most of these haemostatic abnormalities spontaneously resolve
without any clinically significant bleeding.7,10 On the other hand,
major bleeding manifestations in dengue do not necessarily correl-
ate with coagulation abnormalities.11 In particular, the degree of
thrombocytopaenia does not closely relate to the risk and severity
of bleeding in dengue.12,13 Thus coagulation parameters may not
be a useful guide for determining the risk of bleeding in dengue.
Furthermore, there is no consensus in the guidelines as to whether,
marker for prevention of bleeding. Of the included studies, ran-
domized controlled trials (RCTs) were examined qualitatively for
risks of bias based on the Cochrane risk of bias assessment
tool.18 Quality scores for individual studies were not calculated,
as it is not perceived by all as an objective measure of risk of
bias.19 Meta-analysis could not be performed due to study het-
erogeneity, and publication bias was not assessed due to the
small number of available studies for each comparison.
Results

Downloaded from https://academic.oup.com/trstmh/article/111/10/433/4828073 by guest on 20 January 2021

or when, correction of thrombocytopaenia or deranged coagula-
tion parameters should be attempted.14,15 Since thrombocyto- The initial search yielded 676 results. After screening of abstracts,
paenia is the most striking abnormality seen in dengue patients, 20 potential papers were identified. After excluding uncontrolled
platelet transfusion has been widely used, both prophylactically as studies, case reports, small case series and reviews, nine papers
well therapeutically, in managing dengue patients in the past, remained. Another two papers were identified by cross-referencing.
based on the assumption that thrombocytopaenia is a significant The PRISMA flow diagram for study selection is shown in Figure 1.
contributor towards haemorrhage.16,17 This practice has declined Of the 11 studies, 5 studies on platelet transfusion (2 RCTs and 3
over the years. Various other therapeutic agents aimed at prevent- non-randomized studies), 1 RCT on plasma transfusion, 1 RCT on
ing or reducing bleeding have been studied, including recombinant recombinant activated factor VII, 2 RCTs on intravenous anti-D
activated factor VII (rFVIIa), anti-D globulin and intravenous globulin, 1 RCT on IVIg and 1 RCT on interleukin 11 (IL-11) were
immunoglobulin (IVIg), among others. At present, the guidelines selected for further evaluation. The reviewers’ assessment of risk of
for the management of dengue do not recommend the use of any bias is summarized in Figure 2.
therapeutic agent for prevention or treatment of bleeding.14,15
We conducted a systematic review to determine the efficacy
and safety of prophylactic and therapeutic measures used to Platelet and plasma transfusion
manage bleeding in dengue infection. RCTs
The first RCT on platelet transfusion in dengue was a single-
centre non-blinded study conducted in 2011 in Pakistan.20 The
Materials and methods primary outcome measure considered by the investigators was
an increase in platelet count; a reduction in clinical bleeding
This review includes randomized, quasi-randomized or non-
was the secondary outcome. Adult patients with dengue and a
randomized, prospective or retrospective studies that had a con-
trol group looking at interventions for bleeding manifestations in
dengue. We searched PubMed, Embase, CINAHL, Cochrane Library
and Google Scholar using the search terms ‘bleeding’ OR ‘hemor-
rhage’ OR ‘haemorrhage’ AND ‘dengue’ in the title or abstract
fields with no restriction on date limits. References provided in full
papers were also used to identify additional papers for review. Only
articles published in English were included. The last date of the
search was 3 July 2017. The review is registered with PROSPERO
(CRD42017074419).
Two reviewers (NLdS and PW) independently screened the
abstracts and selected articles to assess the full text articles.
Eligible studies were finalized by consensus among all authors.
Data from individual studies were extracted using a customized
data extraction pro forma designed by the authors. The data
items extracted from each eligible study included the location
of the trial, participant demographics, patient characteristics of
the test and control groups, severity of disease, details of inter-
vention and outcome measures. The primary outcome measure Figure 1. PRISMA diagram of the systematic review. We searched
we considered was documented episodes of bleeding following PubMed, Embase, CINAHL, Cochrane Library and Google Scholar using
the search terms ‘bleeding’ OR ‘hemorrhage’ OR ‘haemorrhage’ AND
the intervention. Secondary outcome measures considered were
‘dengue’ in the title or abstract fields with no restriction on date limits.
adverse events and mortality. Some studies also report the This yielded 676 results. After screening of abstracts, 20 potential papers
platelet count and need for platelet transfusions as an indica- were identified. After excluding uncontrolled studies, case reports, small
tion of treatment efficacy. Although there is currently no evi- case series and reviews, nine papers remained. References provided in
dence to link platelet count to bleeding risk in dengue, for the full papers were also used to identify additional papers for review.
sake of completeness we included studies where an outcome Another two papers were identified by this method. A total of 11 studies
measure was improvement in platelets, as a possible surrogate were reviewed for this article.

434
Transactions of the Royal Society of Tropical Medicine and Hygiene
Figure 2. Risk of bias graph. Eight RCTs are included in the risk of bias
summary diagram. Only two studies have a clear description of methods
of randomization so that selection bias is effectively minimized.
Although some studies have a high risk of performance and detection
bias due to a lack of blinding or a poor mechanism of blinding, outcome
assessment was not affected in most of the studies. Except for two
studies, all the other studies either had no loss of data or used ITT ana-
lysis. As a result, attrition bias is minimized.
platelet count <30 000/μl without bleeding manifestations or
with mild bleeding were included. In this study, the method of
randomization and measures for allocation concealment were
not specified and therefore the risk of selection bias is unclear.
The study was non-blinded and therefore was at risk of having
performance and detection bias; however, since the outcome
measures were objective, these factors may not have had a sig-
nificant effect on the final results. Nineteen patients (46.4%)
were considered responders, based on the platelet increase 1 h
after transfusion (post-transfusion platelet increment >10 000/μl
or corrected count increment >50 000/μl). Although platelet
transfusion increased platelet counts, this did not result in a
significant difference in progression to severe bleeding, time
needed for cessation of bleeding or a reduction in new-onset
bleeding; notably, there was only one patient (in the treatment
group), thus conclusions on the effects of bleeding cannot be
clearly made from this study. Platelet transfusion–related compli-
cations were experienced by three (7%) patients.
The second study, a multicentre RCT on prophylactic platelet
transfusion in dengue, recruited 372 adult patients with dengue
and platelet counts <20 000/μl, without significant bleeding,
between 2010 and 2014 in Malaysia and Singapore.21 Patients
with comorbidities such as chronic liver and kidney disease,
active peptic ulcer disease, anticoagulant use, haematological
diseases and critically ill patients were excluded from the study.
Downloaded from https://academic.oup.com/trstmh/article/111/10/433/4828073 by guest on 20 January 2021
The primary endpoint was clinical bleeding, excluding petechiae.
The randomization method described in this study was strong,
resulting in a low risk of selection bias. The outcome measures
were objective, hence the effect of lack of blinding is likely to
have been low. The analysis was based on intention to treat
(ITT) and the risk of attrition bias was low. Platelet transfusion
was not shown to be superior to supportive care in preventing
clinical bleeding and was associated with 13 adverse events,
compared with 2 adverse events in the control group (relative
risk 6.26 [95% confidence interval 1.43 to 27.34]). The mean
time to recovery of platelet count >50 000/μl was not signifi-
cantly different between the two groups. Notably, in this cohort
of patients, plasma leakage occurred in only 13% (47/369) of
patients, which is lower than the figures reported in other hospi-
talized cohorts.22,23 The generalizability of these results to
patients with severe dengue with the potential towards a higher
incidence of significant bleeding is questionable.
The effects of fresh frozen plasma (FFP) on platelet counts in
dengue were evaluated in a RCT in 2008 in Sri Lanka.24 The trial
randomized 109 adult dengue patients with platelet counts of
10 000–40 000/μl to receive 3 units of FFP or normal saline as
placebo. The primary outcome measures were an increase in
platelet counts at 12, 24 and 48 h post-transfusion. The ran-
domization method was sound, with a low risk selection bias.
Although the study was described as a blinded study, measures
taken towards blinding appear inadequate, resulting in a high
risk of performance bias. Nonetheless, like in the other studies
mentioned above, the objective nature of the outcome mea-
sures minimizes the impact of this. Statistical analysis was done
using ITT analysis and bias due to incomplete data was mini-
mized. Platelet counts increased in patients receiving FFP at 12
h, with 31 400/μl (SD 37 000) in the treatment group and 18
000/μl (SD 26 500) in the control group (p=0.04). However, this
increase was not sustained at 24 and 48 h. Notably, bleeding
was not an outcome measure in this study, and the fact that
platelet counts improved may not correlate with prevention of
bleeding. One patient in the intervention group developed an
anaphylactic reaction to FFP. A detailed breakdown of other
adverse events was not provided.
Non-randomized comparative studies
A retrospective study from Malaysia compared 106 paediatric
patients presenting between 1991 and 2000 with complicated
dengue fever, including dengue shock syndrome, who received
platelets or FFP transfusion (n=53) or no intervention (n=53).25
Patients with clinically significant bleeding (evidenced by low/
normal haematocrit) were excluded. While no reduction in
435
S. Rajapakse et al.
bleeding tendency was seen, transfusion of platelets or FFP
increased the risk of pulmonary oedema from 6.5% (95% confi-
dence interval [CI] 2.9 to 29.2) to 30% (95% CI 19.9 to 42.5)
(p=0.006) and prolonged hospital stay from 5 d (95% CI 3 to
17.3) to 7 d (95% CI 4 to 17) (p<0.001).
A single-centre retrospective analysis of patients with dengue
in Singapore in 2004 included 256 patients whose platelet count
was <20 000/mm3 who had no clinical bleeding.12 There was no
benefit seen in patients receiving platelet transfusions in terms of
clinical bleeding, increment in platelet counts, time to recovery of
platelet counts to >50 000/μl or length of hospital stay. One
patient in the group given platelet transfusions died of major
gastrointestinal bleeding and two patients among those who did
not receive platelets developed minor bleeding. The decision to
transfuse platelets was made based on the treating clinicians’
judgement, without any predetermined guidelines or protocols.
Another retrospective observational study in Singapore evalu-
ated the clinical outcome of prophylactic platelet transfusion in
patients with dengue and platelet counts <20 000/μl.26 There
were 788 eligible patients for analysis and 486 received prophy-
lactic transfusions. Outcome variables were compared with 302
patients who did not receive platelet transfusions. Recovery of
thrombocytopaenia was higher in those who received transfu-
sions. The increment in platelet counts was 8000/μl (95%
CI −6000 to 43 000) in the transfused group compared with
5000/μl (95% CI −6000 to 31 000) in the non-transfused group
(p<0.0001). The median length of hospital stay was 6 d (95% CI
4–8) in transfused group and 5 d (95% CI 5–7) in the non-
transfused group (p<0.0001). Eighty-nine patients (18.3%) in the
transfused group and 28 patients (9.3%) of those not transfused
developed mucosal bleeding (p=0.001). There were many base-
line differences in the study groups that might have resulted in
bias; there were higher numbers of patients with severe dengue,
severe bleeding or plasma leakage in the group receiving platelet
transfusion; liver enzymes were higher and white cell counts and
platelet counts were lower in this group. Since transfusions were
given based on the treating clinicians judgement, it is possible
that the threshold for transfusion was lower in patients who
were more sick. To correct for this, the investigators performed a
propensity score matching analysis that showed there was no
difference in clinical bleeding, mucosal bleeding or internal bleed-
ing between the two groups after adjustment for variables that
may have influenced the treating clinicians’ decision to transfuse
platelets.
rFVIIa
RCTs on rFVIIa
The effect of rFVIIa was studied in an RCT of children with den-
gue haemorrhagic fever grade II–III with active bleeding in
Thailand and the Phillipines.27 The primary outcome was
response of bleeding. If bleeding completely stopped, treatment
was considered effective, if partially controlled it was considered
partially effective and if it remained the same or worsened,
treatment was considered ineffective. The method of random-
ization is not clearly mentioned in this study. The study was
blinded and performance bias was prevented by providing an
identical-looking placebo. However, there is the possibility of
436
detection bias since authors have not clearly described the
measures taken to blind the investigators who assessed clinical
outcomes. At 2 h after administration, the control of bleeding
improved in the treatment group. An effective response to
bleeding was seen at 2 h in 12/16 (75%) patients in the inter-
vention group and 4/9 (44.4%) in the control group, but subse-
quently no difference was seen. Platelet transfusions were given
more often in the control group (33.3%) compared with the
treatment group (6.3%). Red cell and plasma transfusions were
not different between the two groups. The study had a sample
size of 25 and had minor violations in protocol in relation to the
Downloaded from https://academic.oup.com/trstmh/article/111/10/433/4828073 by guest on 20 January 2021
timing of repeat administrations of rFVIIa. Overall, this study
shows no evidence of benefit with rFVIIa for bleeding in dengue.
Non-randomized comparative studies
There were no observational studies of rFVIIa with a control
group.
Intravenous anti-D globulin
RCTs on anti-D globulin
In an RCT published from the Philippines in 2007 that studied the
response of thrombocytopaenia to anti-D globulin injection in
patients with dengue haemorrhagic fever, beneficial effects were
shown in terms of improvement of thrombocytopaenia.28 Adult
and paediatric patients with dengue haemorrhagic fever were
randomized to receive either anti-D globulin or placebo and
platelet counts were monitored. The outcome measure was an
increase in platelet count. The method of randomization was not
specified in this paper. Double-blinding was done effectively, pre-
venting performance and detection bias. All recruited patients
were included in the analysis. Of 47 patients recruited, only 24
had platelet counts <50 000/μl, and in this group a greater
increase in platelet count was seen in those receiving anti-D
globulin (n=12); this difference was not seen in those with plate-
let counts >50 000/μl. There were two deaths, one each from the
treatment and placebo groups.
Another RCT from India conducted in 2010 showed that
administration of anti-D globulin resulted in an increase in plate-
let count as well as a reduction in bleeding.29 Patients with den-
gue and platelet counts <20 000/μl were enrolled into the
study. Fifteen patients with no bleeding or mild bleeding were
included in each group. In the intervention group, anti-D globu-
lin was administered in addition to standard supportive care
and platelet transfusion. Random sequence generation was
done using a computer-based system, however, it is not clear
whether they used precautions to maintain allocation conceal-
ment. The study was not blinded. All included patients were
analysed, resulting in low attrition bias. Platelet transfusion
requirements were lower in the intervention group (187 vs
342 ml; p=0.01). However, the indication or justification for
platelet transfusion is not clear in the methodology and we do
not consider this as a valid outcome measure based on the
results in the previous section. Grade I bleeding, which was pre-
sent in five patients in the intervention group, reduced to zero at
48 h, whereas in the control group the reduction was from six to
five (p=0.014). Similarly, grade II bleeding was significantly
Transactions of the Royal Society of Tropical Medicine and Hygiene
reduced at 24 h (five to zero and seven to four in the interven-
tion and control groups, respectively; p=0.032). At other time
intervals considered, differences were not significant. There were
no deaths, and no serious complications were reported. No
patients required red cell transfusion.
Non-randomized comparative studies
There were no observational studies with a control group evalu-
ating the role of anti-D globulin in dengue.
IVIg
RCTs on IVIg
There is only one RCT on the effect of IVIg on thrombocyto-
paenia in dengue infection.30 This study from the Philippines in
2007 included 31 patients with platelet counts of 20 000–80
000/μl without clinical evidence of bleeding who were rando-
mized to receive IVIg or placebo. The outcome measure was an
increase in platelet count and no clinical bleeding. It is unclear
at which point the randomization was done, and there were
more patients with dengue haemorrhagic fever in the control
group. This was not a blinded study. Whether there was a loss of
data prior to analysis is unclear. No difference in platelet counts
were seen in the two groups.
Non-randomized comparative studies
There are no observational studies on the use of IVIg in dengue.
Tranexamic acid
Tranexamic acid is an anti-fibrinolytic agent that preserves and
stabilizes fibrin’s matrix structure and thereby helps to control
bleeding. Its use is recommended in some guidelines14 to miti-
gate menstrual bleeding and gastrointestinal bleeding. There
are no controlled studies on its use in dengue.
Recombinant human IL-11
RCTs on IL-11
A RCT from Pakistan enrolling 40 adult patients evaluated the
effect of recombinant human IL-11 in patients with dengue
with platelet counts of 10 000–30 000/μl without clinical bleed-
ing.31 The method of randomization in the study is unclear.
Blinding was done effectively, thus the risks of performance and
detection bias are low. All included patients were analysed,
without any attrition bias. Comparisons of platelet counts in the
placebo and intervention groups at 12 h (20 650±3963.1/μl vs
26 150±7513.32; p=0.007), 24 h (21 400±7576.1/μl vs 30 550
±8009.7; p=0.001), 36 h (25 700±9073.4/μl vs 35 700±8442.4;
p=0.001) and 48 h (30 000±10 125.53/μl vs 41 050±8629.66;
p=0.001) showed that the increases in platelet counts at 24, 36
and 48 h were greater in the treatment group, although no dif-
ference in clinical bleeding was seen.
Non-randomized comparative studies
There were no observational studies on the role of recombinant
human IL-11 in dengue.
Discussion
There is no trial evidence, or evidence from observational studies,
to support the use of platelet transfusion or plasma transfusion to
prevent clinically significant bleeding in patients with dengue infec-
tion and thrombocytopaenia. Data from observational studies are
Downloaded from https://academic.oup.com/trstmh/article/111/10/433/4828073 by guest on 20 January 2021
fraught with problems, however, because of the lack of clear
guidelines regarding the indications for platelet transfusion. In the
past, platelet transfusions were given freely, based on clinical judg-
ment, at various thresholds.17 There is no evidence of benefit in
using platelet transfusions and hence most current guidelines do
not recommend the use of platelet transfusion in dengue,14 and
indeed, in the light of emerging evidence demonstrating the lack
of benefit, the practice of routinely transfusing platelets when
counts drop below a certain threshold has decreased over the
years. Conversely, studies of a restrictive strategy for platelet trans-
fusion in dengue with thrombocytopaenia have shown that such
restrictive strategies have no adverse effect on outcome.17,32
Notably, the two RCTs on platelet transfusion20,21 were on
patients with severe thrombocytopaenia but less severe disease.
Thrombocytopaenia does not correlate with bleeding risk,12,13 nor
is there evidence to show that it correlates with the severity of
dengue—in particular, the occurrence of plasma leakage. There is
definite evidence of increased risk of adverse effects, particularly
allergic reactions, with the use of platelet transfusion. The evidence
of benefit in terms of platelet count with transfusion of FFP is lim-
ited and does not translate into a reduction of clinical bleeding.
The pathophysiological relationship between thrombocyto-
paenia and bleeding in dengue is not clearly understood.
Thrombocytopaenia in dengue is due to several causes and its
pathogenesis is complex. Bone marrow suppression, evidenced
by hypocellularity and attenuation of megakaryocyte maturation,33
together with peripheral destruction of platelets through activation
of the complement cascade, antibody-mediated lysis and phago-
cytosis34,35 occur. Platelet dysfunction is also known to take
place.34 Bleeding in dengue, on the other hand, is multifactorial,
and platelets may play only a small part in its pathogenesis, and
associated coagulopathy and vascular damage are important
contributors.36 This may explain why the correlation between
platelet count and bleeding is poor.
The fact remains that there is little trial evidence on the role of
platelet transfusion in patients with thrombocytopaenia and mod-
erate to severe bleeding. There is clearly a need for further study on
this. Currently the guidelines recommend administration of fresh
packed cells or fresh whole blood in patients who have moderate
to severe bleeding in dengue.14,15 There are currently no published
studies evaluating the place of blood transfusion in this situation.
Administration of rFVIIa may result in short-term improve-
ment in control of bleeding in patients with dengue and active
bleeding but shows no overall benefit. Its high cost limits its
use, estimated at around US$4241 per additional quality
adjusted life year.37 It may be of benefit in patients with severe
bleeding, as a short-term measure. Intravenous anti-D globulin
has shown some evidence of benefit in increasing platelet
437
S. Rajapakse et al.
counts in patients with dengue, although the impact on clinical
bleeding is unclear. There is no evidence of benefit with the use
of IVIg with regards to improvement in platelet count or reduc-
tion in bleeding, and there is no justification to use this thera-
peutic agent in the treatment of dengue infection.38 The single
randomized study on IL-11 showed no effect on clinical bleed-
ing and its use cannot be recommended. There is no trial evi-
dence to support or refute the use of tranexamic acid.
It is noteworthy that almost all studies looked at prevention
or minimizing bleeding; there were no studies that examined
the place of any of these measures in the management of
severe haemorrhage. In the absence of evidence, it is likely that
clinicians will continue to use platelets, FFP and blood, as well
as other measures such as tranexamic acid and rFVIIa, in the
setting of severe haemorrhage.
Conclusions
(1) Prophylactic platelet transfusion should not be routinely pre-
scribed in patients with dengue with no bleeding based on
low platelet count.
(2) Therapeutic platelet transfusion should not be routinely pre-
scribed in patients with dengue with thrombocytopaenia
and mild bleeding.
(3) There is inadequate evidence to support or refute the use of
platelet transfusion in patients with severe bleeding in dengue.
(4) There is a need for further, well-designed RCTs to evaluate
the role of platelets and plasma transfusion in patients in
both the prevention of bleeding and in the setting of clinic-
ally significant bleeding in dengue infection.
(5) There is currently insufficient evidence regarding the role of
rFVIIa, anti-D globulin, Ig or tranexamic acid in the preven-
tion or treatment of bleeding in dengue infection and there
is a place for further research on these therapeutic agents.
Authors’ contribution: SR and PW conceived the study. NLdS and PW
did the literature search and selected the abstracts to access full arti-
cles. All authors read the full articles and synthesized the evidence. NLdS
wrote the initial manuscript. SR, DF and CR critically revised the manu-
script. All authors read and approved the final manuscript. SR is the
guarantor of the paper.
Funding: None.
Conflicts of interest: None declared.
Ethical approval: Not required.
References
1 Brady OJ, Gething PW, Bhatt S, et al. Refining the global spatial limits
of dengue virus transmission by evidence-based consensus. PLoS
Negl Trop Dis 2012;6(8):e1760.
438
2 Stanaway JD, Shepard DS, Undurraga EA, et al. The global burden of
dengue: an analysis from the Global Burden of Disease Study 2013.
Lancet Infect Dis 2016;16(6):712–23.
3 Laul A, Laul P, Merugumala V, et al. Clinical profiles of dengue infec-
tion during an outbreak in northern India. J Trop Med 2016;2016:
5917934.
4 Thomas L, Brouste Y, Najioullah F, et al. Prospective and descriptive
study of adult dengue cases in an emergency department, in
Martinique. Med Mal Infect 2010;40(8):480–9.
5 Rai S, Chakravarti A, Matlani M, et al. Clinico-laboratory findings of
patients during dengue outbreak from a tertiary care hospital in
Downloaded from https://academic.oup.com/trstmh/article/111/10/433/4828073 by guest on 20 January 2021
Delhi. Trop Doct 2008;38(3):175–7.
6 Woon YL, Hor CP, Hussin N, et al. A two-year review on epidemiology
and clinical characteristics of dengue deaths in Malaysia,
2013–2014. PLoS Negl Trop Dis 2016;10(5):e0004575.
7 Chuansumrit A, Chaiyaratana W. Hemostatic derangement in den-
gue hemorrhagic fever. Thromb Res 2014;133(1):10–6.
8 Sosothikul D, Thisyakorn U, Thisyakorn C. Hemostatic studies in den-
gue patients. Southeast Asian J Trop Med Public Health 2015;46
(Suppl 1):43–5.
9 Chuang YC, Lin YS, Liu CC, et al. Factors contributing to the disturb-
ance of coagulation and fibrinolysis in dengue virus infection.
J Formos Med Assoc 2013;112(1):12–7.
10 Wills BA, Oragui EE, Stephens AC, et al. Coagulation abnormalities in
dengue hemorrhagic fever: serial investigations in 167 Vietnamese
children with Dengue shock syndrome. Clin Infect Dis 2002;35(3):
277–85.
11 Krishnamurti C, Kalayanarooj S, Cutting MA, et al. Mechanisms of
hemorrhage in dengue without circulatory collapse. Am J Trop Med
Hyg 2001;65(6):840–7.
12 Lye DC, Lee VJ, Sun Y, et al. Lack of efficacy of prophylactic platelet
transfusion for severe thrombocytopenia in adults with acute
uncomplicated dengue infection. Clin Infect Dis 2009;48(9):1262–5.
13 Lum LC, Goh AY, Chan PW, et al. Risk factors for hemorrhage in
severe dengue infections. J Pediatr 2002;140(5):629–31.
14 Gunawardane NWA, Dissanayake U, Sellahewa K, et al. Guidelines on
management of dengue fever and dengue hemorrhagic fever in
adults. Colombo, Sri Lanka: Ministry of Health, 2010.
15 World Health Organization. Dengue: guidelines for diagnosis, treat-
ment, prevention and control. Geneva: World Health Organization,
2009.
16 Kurukularatne C, Dimatatac F, Teo DL, et al. When less is more: can
we abandon prophylactic platelet transfusion in dengue fever? Ann
Acad Med Singapore 2011;40(12):539–45.
17 Thomas L, Kaidomar S, Kerob-Bauchet B, et al. Prospective observa-
tional study of low thresholds for platelet transfusion in adult den-
gue patients. Transfusion 2009;49(7):1400–11.
18 Higgins JP, Altman DG, Gotzsche PC, et al. The Cochrane
Collaboration’s tool for assessing risk of bias in randomised trials.
BMJ 2011;343:d5928.
19 Greenland S. Quality scores are useless and potentially misleading:
reply to ‘Re: A critical look at some popular analytic methods’. Am J
Epidemiol 1994;140(3):300–301.
20 Khan Assir MZ, Kamran U, Ahmad HI, et al. Effectiveness of platelet
transfusion in dengue fever: a randomized controlled trial. Transfus
Med Hemother 2013;40(5):362–8.
21 Lye DC, Archuleta S, Syed-Omar SF, et al. Prophylactic platelet trans-
fusion plus supportive care versus supportive care alone in adults
with dengue and thrombocytopenia: a multicentre, open-label, ran-
domised, superiority trial. Lancet 2017;389(10079):1611–18.
Transactions of the Royal Society of Tropical Medicine and Hygiene
22 Srikiatkhachorn A, Krautrachue A, Ratanaprakarn W, et al. Natural history
of plasma leakage in dengue hemorrhagic fever: a serial ultrasono-
graphic study. Pediatr Infect Dis J 2007;26(4):283–90; discussion 291–2.
23 Balasubramanian S, Janakiraman L, Kumar SS, et al. A reappraisal of
the criteria to diagnose plasma leakage in dengue hemorrhagic
fever. Indian Pediatr 2006;43(4):334–9.
24 Sellahewa KH, Samaraweera N, Thusita KP, et al. Is fresh frozen plas-
ma effective for thrombocytopenia in adults with dengue fever? A
prospective randomised double blind controlled study. Ceylon Med J
2008;53(2):36–40.
25 Lum LC, Abdel-Latif Mel A, Goh AY, et al. Preventive transfusion in den-
gue shock syndrome—is it necessary? J Pediatr 2003;143(5):682–4.
26 Lee TH, Wong JG, Leo YS, et al. Potential harm of prophylactic plate-
let transfusion in adult dengue patients. PLoS Negl Trop Dis 2016;10
(3):e0004576.
27 Chuansumrit A, Wangruangsatid S, Lektrakul Y, et al. Control of
bleeding in children with dengue hemorrhagic fever using recombin-
ant activated factor VII: a randomized, double-blind, placebo-
controlled study. Blood Coagul Fibrinolysis 2005;16(8):549–55.
28 De Castro RAC, De Castro J-AA, Barez MYC, et al. Thrombocytopenia
associated with dengue hemorrhagic fever responds to intravenous
administration of anti-D (RH0-D) immune globulin. Am J Trop Med
Hyg 2007;76(4):737–42.
29 Pannu AK, Bhalla A, Singhal M, et al. Safety and efficacy of a single
®
dose of anti-D (WinRho ) in severe thrombocytopenia secondary to
dengue virus infection. Indian J Crit Care Med 2017;21(2):80–84.
30 Dimaano EM, Saito M, Honda S, et al. Lack of efficacy of high-dose
intravenous immunoglobulin treatment of severe thrombocytopenia
in patients with secondary dengue virus infection. Am J Trop Med
Hyg 2007;77(6):1135–38.
31 Suliman MI, Qayum I, Saeed F. Randomized clinical trial of human
interleukin-11 in dengue fever-associated thrombocytopenia. J Coll
Physicians Surg Pak 2014;24(3):164–8.
32 Sharma A, Charles K, Chadee D, et al. Dengue hemorrhagic fever in
Trinidad and Tobago: a case for a conservative approach to platelet
transfusion. Am J Trop Med Hyg 2012;86(3):531–5.
33 Srichaikul T, Nimmannitya S. Haematology in dengue and dengue
haemorrhagic fever. Best Pract Res Clin Haematol 2000;13(2):
261–76.
Downloaded from https://academic.oup.com/trstmh/article/111/10/433/4828073 by guest on 20 January 2021
34 Srichaikul T, Nimmannitya S, Sripaisarn T, et al. Platelet function dur-
ing the acute phase of dengue hemorrhagic fever. Southeast Asian J
Trop Med Public Health 1989;20(1):19–25.
35 Honda S, Saito M, Dimaano EM, et al. Increased phagocytosis of pla-
telets from patients with secondary dengue virus infection by human
macrophages. Am J Trop Med Hyg 2009;80(5):841–5.
36 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.
37 Naing C, Poovorawan Y, Mak JW, et al. Cost-utility analysis of an
adjunctive recombinant activated factor VIIa for on-demand treat-
ment of bleeding episodes in dengue haemorrhagic fever. Blood
Coagul Fibrinolysis 2015;26(4):403–7.
38 Koizumi N, Gamage CD, Muto M, et al. Serological and genetic ana-
lysis of leptospirosis in patients with acute febrile illness in Kandy, Sri
Lanka. Jpn J Infect Dis 2009;62(6):474–5.
439