British Journal of Anaesthesia, 122 (1): 19e31 (2019)

doi: 10.1016/j.bja.2018.09.010
Advance Access Publication Date: 25 October 2018
Review Article

CLINICAL PRACTICE

Perioperative thrombocytopenia: evidence,

evaluation, and emerging therapies
A. Nagrebetsky1,*,#, H. Al-Samkari2,#, N. M. Davis1, D. J. Kuter2 and
J. P. Wiener-Kronish1
1
Department of Anesthesia Critical Care and Pain Medicine, Massachusetts General Hospital, Harvard
Medical School, Boston, MA, USA and 2Division of Hematology, Massachusetts General Hospital, Harvard
Medical School, Boston, MA, USA

*Corresponding author. E-mail: anagrebetsky@mgh.harvard.edu

#
These authors contributed equally to the work.

Abstract
Thrombocytopenia is a common perioperative clinical problem. While global haemostasis is influenced by many patient-
and procedure-related factors, the contribution of thrombocytopenia to bleeding risk is difficult to predict, as platelet
count does not linearly correlate with likelihood of bleeding. Thus, the widely used definition of thrombocytopenia and
grading of its severity have limited clinical utility. We present a summary and analysis of the current recommendations
for invasive procedures in thrombocytopenic patients, although the platelet count at which any given procedure may
safely proceed is unknown. The benefits and risks of preoperative platelet transfusions should be assessed on a patient-
by-patient basis, and alternatives to platelet transfusion should be considered. In non-emergent surgeries or in post-
operative thrombocytopenic patients, haematology consultation should be considered to guide diagnostics and man-
agement. We present a pragmatic approach to the evaluation of perioperative thrombocytopenia.

Keywords: surgical procedures; operative; thrombopoietin; romiplostim; eltrombopag; platelets; desmopressin

antifibrinolytic agents in the acute setting and throm-

Editor’s key points
 Most ‘threshold’ recommendations for preprocedural
platelet transfusions are based on low quality evidence or
opinion. Existing data suggest no linear relationship be-
tween platelet count and risk of spontaneous bleeding.
 Given the risks associated with platelet transfusion,
alternatives should be considered: desmopressin and
bopoietin receptor agonists for select groups in the
elective setting.
 Thrombopoietin receptor agonists effectively in-
crease platelet count in thrombocytopenic patients,
but are currently off-label except in patients with
thrombocytopenia as a result of chronic liver
disease.

Editorial decision: 2 September 2018; Accepted: 2 September 2018

© 2018 British Journal of Anaesthesia. Published by Elsevier Ltd. All rights reserved.
For Permissions, please email: permissions@elsevier.com

19
 20 - Nagrebetsky et al.
Thrombocytopenia is typically defined as a platelet count
below 150109 L 1 and is common in the perioperative setting,
occurring in approximately 5e10% of patients.1,2 While hae-
mostatic function depends on multiple factors, platelet num-
ber and function are key elements. In this review, we address
the role of platelet count and platelet function in the periop-
erative care of patients undergoing non-cardiac surgery or
procedures and offer an approach to the clinical management
of the thrombocytopenic surgical patient. The safety of
regional anaesthesia in patients with abnormal haemostasis is
an extensive topic that falls outside the scope of this review.
This article is intended to serve as a pragmatic review for
use in daily clinical practice. The manuscript is structured in a
way that may help a practicing clinician in charge of a patient
with thrombocytopenia to quickly find the most important
information relating to the current recommendations and
underlying evidence. We did not carry out a systematic liter-
ature review. A systematic review would decrease the proba-
bility of excluding relevant publications, but would also result
in a larger and more detailed manuscript that could be difficult
to use as a quick clinical reference.
Platelet count and clinical outcomes
A normal platelet count is within the range of 150e450109
L 1, subject to inter-laboratory variation.3 Without physiologic
or pathophysiologic perturbations, the baseline platelet count
is relatively stable over the lifetime of an individual.4 Thus, a
substantial change in platelet count is suggestive of an alter-
ation in normal physiology, such as pregnancy or an under-
lying pathological process.
Data from a large observational study in asymptomatic non-
hospitalised surgical patients without indications for coagu-
lation testing suggest that there may be a relationship between
preoperative thrombocytopenia and 30-day mortality.1 In a
multivariable analysis, a platelet count of 101e150109 L 1 was
associated with a 30% higher likelihood of death [adjusted odds
ratio 1.31; 95% confidence interval (CI) 1.11e1.56, P0.01], and a
platelet count of 100109 L 1 nearly doubled the likelihood of
death (adjusted odds ratio 1.93; 95% CI 1.43e2.61, P0.001).
Furthermore, patients with a platelet count of 100109 L 1
were more likely to develop sepsis, renal complications, and
adverse pulmonary outcomes, though the authors did not
report on the relationship between platelet count and bleeding.
The multiple adverse outcomes in thrombocytopenic pa-
tients may not simply be related to the role of platelets in
haemostasis. A normal platelet count is 15e40 times higher
than is necessary to achieve haemostasis.5 While this may be a
reflection of the functional reserve commonly observed with
biological systems, it could suggest that other functions of
platelets require higher platelet counts.6 Indeed, platelets are
involved not only in haemostasis, but also in inflammatory
and immune responses and wound healing.7 In pathologic
states, platelets may promote excess inflammation and be
associated with organ damage, such as acute kidney injury
and acute lung injury.8 The recent finding that more than half
of the body’s normal platelet production occurs in the lungs
may offer some additional support for these arguments.9
Thrombocytopenia and the risk of bleeding
Thrombocytopenia has been subcategorised for surgical pur-
poses into mild (100e149109 L 1), moderate (50e99109 L 1),
and severe (<50109 L 1).10 The clinical utility of such
subcategorisation is limited as the relationship between
platelet count and bleeding risk is not linear and depends on
platelet function and other patient-specific variables (Fig. 1).
Data from large studies suggest that the risk of spontaneous
bleeding is difficult to predict until platelet count decreases to
extremely low values, below approximately 10109 L 1.11
Because high-quality data quantifying the bleeding risk in
thrombocytopenic surgical patients does not exist, we
explored data from other medical populations. Most studies
assessing the risk of spontaneous bleeding by platelet count
were performed in patients with haematologic malignancies,
in whom anaemia, chemotherapy, and radiation may further
modify the risk of bleeding. In these patients, platelet counts
above 10109 L 1 are sufficient to allow for normal thrombin
generation12 and therefore prevent spontaneous bleeding.
This is supported by data demonstrating that platelet trans-
fusion to a threshold higher than 10109 L 1, the threshold for
prophylactic transfusion recommended by AABB (formerly,
American Association of Blood Banks) guidelines, does not
affect the incidence of spontaneous bleeding (Fig. 1).13,14
In patients with platelet counts of 10109 L 1, the rate of
spontaneous bleeding is high, but the rate of clinically-relevant
major bleeding is not. In a recent study of more than 1200 cancer
patients with platelet counts of 10109 L 1 or lower for at least 5
days, 70% experienced at least one episode of bleeding classified
as WHO Grade 2 or higher, but only 2% had Grade 4 bleeding and
only one died of hemorrhage.15 The WHO bleeding scale broadly
classifies bleeding into Grade 0 (no bleeding), Grade 1 (minimal
bleeding, such as petechiae), Grade 2 (mild blood loss, such as
visible blood in stool or urine), Grade 3 (gross blood loss, such as
profuse GI bleeding or minor intracerebral haemorrhage), and
Grade 4 (debilitating, life-threatening bleeding). Compared with
patients with a platelet count of 81109 L 1, patients with a
platelet count of 5109 L 1 had a higher risk of bleeding (odds
ratio 3.1; 95% CI 2.0e4.8). In this study, no distinct pattern of
decreasing bleeding risk with increasing platelet count was
seen in the range from 6 to 80109 L 1.
In addition to platelet count, many other factors affect the
likelihood of bleeding. For example the risk of bleeding varies
with age; in a study of 117 patients with immune thrombo-
cytopenia (ITP), the risk of major bleeding was almost 30 times
higher in patients older than 60 yr in comparison with patients
younger than 40 yr.16 The risk of bleeding also depends on the
aetiology of thrombocytopenia: ITP patients bleed less
commonly than do other patients at similarly low platelet
counts, probably because of the increased size and function of
their platelets. Perioperative administration of fluids, blood
products, and medications may further attenuate the hae-
mostatic effectiveness of a given platelet count.
Thrombocytopenia and procedures
There is extremely limited data to inform perioperative man-
agement of patients with thrombocytopenia. Figure 1 sum-
marises acceptable platelet counts and indications for platelet
transfusions in thrombocytopenic patients undergoing inva-
sive procedures. Most such ‘threshold’ recommendations are
based on low quality evidence, expert opinion, or practice
review.17e19 Nearly all studies examining prophylactic platelet
transfusions in preoperative thrombocytopenic patients are
retrospective and observational. Nonetheless, we have
attempted to review those studies that might be of clinical
utility in the perioperative management of patients with
thrombocytopenia.
 Perioperative thrombocytopenia - 21

Fig 1. Clinically significant platelet count thresholds/targets and indications for platelet transfusion in patients undergoing elective surgery
or invasive procedures. *As graded in the source document. AABB (formerly, American Association of Blood Banks); ASGE, American
Society for Gastrointestinal Endoscopy; BCSH, British Committee for Standards in Haematology; JPAC, Joint United Kingdom Blood
Transfusion and Tissue Transplantation Services Professional Advisory Committee.

A large systematic review of 17 randomised controlled trials Noncardiac surgery

and 55 observational studies examined the role of prophylactic
In a single-centre retrospective cohort study of 13 978 non-
platelet transfusions in thrombocytopenic patients, finding
cardiac surgical patients, 860 patients had a platelet count of
reduced rates of clinically significant bleeding, but no mortality
100109 L 1 and 71 received platelet transfusion before
benefit, in patients with haematologic malignancies receiving
operation (none received platelets during surgery).21 Preoper-
prophylactic platelet transfusions for a platelet threshold of 10
ative platelet transfusion did not improve patient outcomes
or 20109 L 1.20 In a subset of patients undergoing central
or decrease perioperative red blood cell (RBC) requirements.
venous catheter placement, bleeding rates were similar for
Patients receiving preoperative platelet transfusion were more
those receiving prophylactic platelet transfusion and those
likely to have higher ASA physical status, lower baseline
who did not, although all patients with platelet counts below
haemoglobin concentrations and platelet counts, higher rates
50109 L 1 received platelet transfusion.20
 22 - Nagrebetsky et al.
of postoperative ICU admission, and a longer hospital stay.
Additionally, they were more likely to receive red cell
transfusion.21
In one study of 246 patients with end-stage liver disease,
platelet count was not associated with higher blood product
transfusion requirements.22 It is worth noting that certain
guidelines relating to liver instrumentation in cirrhotic patients
suggest specific platelet thresholds17 (such as UK guidelines
recommending >50109 L 1 for a liver biopsy); the findings
from this study do not support such recommendations.
In a subset of surgical patients, thrombocytopenia may be a
component of a pathophysiologic cascade associated with the
underlying disease. In such patients, the severity of thrombo-
cytopenia may reflect the severity of the disease. For example,
in patients with solid tumours of the liver23 and brain,24 pre-
operative thrombocytopenia was associated with postoperative
mortality. In recipients of liver transplants, the degree of
thrombocytopenia after surgery, a probable marker of intra-
operative pathophysiologic perturbations, was also associated
with surgical complications, graft dysfunction, and sepsis.25
Interventional radiology procedures
A large single-centre retrospective study examined the utility of
prophylactic platelet transfusions in patients undergoing
interventional radiology procedures.26 In this study of 18 204
patients, 2060 patients had a platelet count of 100109 L 1
before needle placement. Some 203 of the 2060 thrombocyto-
penic patients received platelets. Prophylactic platelet trans-
fusions did not reduce bleeding or improve clinical outcomes
when utilised for patients with counts >50109 L 1. In a sensi-
tivity analysis in patients with platelet counts 50109 L 1,
prophylactic platelet transfusions did not reduce the frequency
of RBC transfusion. Of note, patients who were transfused in
this study were more likely to have a haematologic malignancy,
lower baseline platelet count, increased incidence of emer-
gency procedures, and a higher rate of general anaesthesia.
Acute abdominal complications in thrombocytopenic
patients with haematological malignancies
Acute abdominal complications are common in patients with
malignancies, particularly after chemotherapy.27 Surgical
problems that are more likely in cancer patients include hae-
morrhage, peritonitis, cholecystitis, intestinal obstruction,
appendicitis, and splenic infarction.28 A retrospective review
of 58 haematologic malignancy patients undergoing emer-
gency abdominal surgery concluded that intraoperative
thrombocytopenia was not associated with prognosis; the
frequency of surgical bleeding was not reported.27
Thrombocytopenia and the airway
It is likely that severe thrombocytopenia increases the risk of
bleeding with airway manipulation, though supporting evi-
dence is lacking. In one study of 166 patients with median
duration of thrombocytopenia (30109 L 1) of 18 days, there
was only one non-epistaxis bleeding episode involving the
airway, a ‘throat and mouth haemorrhage’.29
Limited data is available on the safety of flexible bron-
choscopy in thrombocytopenic patients. In a retrospective
study of 150 thrombocytopenic patients who underwent
bronchoscopy, only one patient had bleeding that required
continuous suctioning; it resolved spontaneously.30 Subgroup
analysis of 35 patients (platelet count 30109 L 1) demon-
strated no episodes of clinically significant bleeding. Some of
the patients received prophylactic platelet transfusions, but in
many of these patients, platelet counts did not increase
beyond 50109 L 1. In another retrospective study of 37
mechanically-ventilated ICU patients with a mean platelet
count 27109 L 1, 12 patients had tracheobronchial bleeding
that did not require RBC transfusion.31 In a prospective study
of 66 flexible bronchoscopies in 47 patients with a median
platelet count of 37109 L 1 undergoing haematopoietic stem
cell transplantation, five patients had bleeding complica-
tions.32 Severe epistaxis occurred in one patient with a platelet
count of 14109 L 1, minimal epistaxis was documented in
three other patients with platelet counts between 14 and
55109 L 1, and a patient with a platelet count of 120109 L 1
developed haemoptysis. Each study concluded that flexible
bronchoscopy is safe in most patients with thrombocytopenia.
In line with the existing data, the British Thoracic Society
Guidelines suggest that bronchoscopy with lavage can be
safely performed at platelet counts >20109 L 1.33
Qualitative platelet dysfunction and
procedures
Acquired platelet dysfunction
Acquired qualitative platelet defects secondary to uraemia,
cardiopulmonary bypass, and medications are common in
surgical patients. Typical offending medications include
aspirin, the non-aspirin NSAIDs, P2Y12 adenosine diphosphate
receptor antagonists (ticlopidine, clopidogrel, prasugrel, tica-
grelor, and cangrelor), dipyridamole, and glycoprotein IIb/IIIa
antagonists (abciximab, eptifibatide, and tirofiban). Selective
serotonin reuptake inhibitors are thought to reduce platelet
function, but the clinical impact is less clear. Antiplatelet
agents are routinely held before operative management
whenever possible in the elective surgical setting, but patients
often require emergency surgery in the setting of bleeding
provoked or exacerbated by an antiplatelet medication. More
recently, a randomised trial to evaluate the effect of platelet
transfusions for patients who had suffered acute spontaneous
primary intracerebral haemorrhage while on antiplatelet
agents found that platelet transfusion was inferior to standard
of care; there were more serious adverse events (e.g. enlarge-
ment of intracerebral haemorrhage or infections) in the pa-
tients who received platelet transfusions (42% compared with
29% receiving standard care).34 Importantly, patients in this
trial did not undergo invasive neurosurgical management of
the haemorrhage. This study concluded that platelet trans-
fusion may be deleterious for this patient population, chal-
lenging the long-accepted notion that patients on aspirin
presenting with cerebral haemorrhage benefit from platelet
transfusion.34
Patients who require ophthalmology procedures while on
antiplatelet agents are a noteworthy subgroup in clinical
practice. A comprehensive guide to perioperative manage-
ment of such patients has been published previously.35 The
treatment of uremic platelet dysfunction is discussed below in
the section discussing alternatives to platelet transfusion.
Inherited platelet disorders
Inherited platelet disorders may result in qualitative platelet
dysfunction, varying degrees of thrombocytopenia, or both.

The clinical spectrum of bleeding is highly variable in this
diverse group of disorders. A retrospective analysis of bleeding
and transfusion with invasive procedures in a heterogenous
group of 423 patients with inherited platelet disorders found
excessive bleeding risk in those with Bernard-Soulier syn-
drome, autosomal variant Glanzmann thrombasthenia, and
Hermansky-Pudlak syndrome.36 Bleeding was twice as com-
mon in inherited platelet function disorders (24.8% of pro-
cedures) compared with inherited thrombocytopenias (13.4%
of procedures). The latter were associated with infrequent
surgical bleeding until the platelet count was below 68109
L 1.36
Evaluation of thrombocytopenia in the
perioperative patient
A complete evaluation of thrombocytopenia in any patient
begins by obtaining a thorough haematologic history,
including prior platelet count trends and analysis of any prior
thrombocytopenic episodes. Also critical is an exacting
medication history, including times of drug administration. In
many cases, patients with mild thrombocytopenia of known
aetiology may proceed to surgery without additional periop-
erative workup or haematology consultation. In cases of un-
clear aetiology, severe thrombocytopenia, or unexpected
bleeding, haematology consultation is warranted to guide
further diagnosis and treatment. The following discussion
focuses on the causes of unexplained thrombocytopenia in the
postoperative patient, in whom a normal platelet count was
documented before surgery and in whom no known prior
thrombocytopenic diagnosis exists.
Artifactual thrombocytopenia
The first step in the diagnostic workup of the thrombocyto-
penic postoperative patient is assessment for artifactual
thrombocytopenia (‘pseudothrombocytopenia’). The complete
blood count is drawn in a tube containing the anticoagulant
ethylenediaminetetraacetic acid (EDTA). Approximately 0.1%
of individuals have clinically insignificant antibodies that
result in platelet clumping in the presence of EDTA (Fig. 2a) or
platelet satellitism around leukocytes (Fig. 2b) resulting in a
falsely low platelet count.37,38 Because the clumping of
Fig 2. Artifactual thrombocytopenia. (a) Platelet clumping occurring secondary to pre-existing antibodies that result in platelet clumping in
the presence of ethylenediaminetetraacetic acid; and (b) Platelet satellitism, with a rosette of platelets coating the surface of a normal
neutrophil.
Perioperative thrombocytopenia - 23
pseudothrombocytopenia occurs usually only in the setting of
EDTA, performing an automated platelet count from patient
blood in a citrated or heparinised tube instead of an EDTA tube
can confirm the diagnosis and allow for an accurate estimate
of the true platelet count.
Non-artifactual thrombocytopenia in the
postoperative patient
Having excluded artifactual causes, thrombocytopenia may
then be classified according to its basic pathophysiologic
process (i.e. decreased platelet production, increased platelet
destruction, platelet sequestration or dilution, or some com-
bination of these processes). Table 1 details the most likely
aetiologies of acute thrombocytopenia in the postoperative
setting according to underlying pathophysiology. Examination
of the peripheral blood film is often suggestive or diagnostic
for a number of these disorders. The most relevant findings in
the perioperative patient include large platelets (also quanti-
fied by an elevated mean platelet volume, suggestive of
consumptive thrombocytopenia) and fragmented RBCs (also
called schistocytes, suggestive of thrombotic microangiopathy
or destruction of red cells and platelets by endovascular de-
vices). If the peripheral film is revealing of a likely aetiology,
evaluation should be tailored to the likely differential
diagnosis.
If platelet transfusion has occurred, a post-transfusion
platelet count should always be obtained within 30 min of
the completion of the first transfusion episode to allow
calculation of the corrected count increment (CCI). Failure of
the CCI to increase by the expected magnitude can provide
valuable diagnostic information, such as the presence of an-
tiplatelet antibodies indicative of an acute immunologic
destructive process or platelet alloimmunisation.39
Certain aetiologies of thrombocytopenia are suggested by
history alone. Administration of large quantities of i. v. fluids
during operation may result in a dilutional thrombocyto-
penia, but with conservation of overall platelet mass and no
significant clinical consequence; in this situation, propor-
tional reductions in haemoglobin, haematocrit, and the white
blood cell count are typically seen along with the platelet
decrease. Massive red cell transfusion may have the same
result.40 Severe thrombocytopenia (often <10109 L 1) in the
 24 - Nagrebetsky et al.
Table 1 Typical aetiologies of thrombocytopenia in the postoperative patient by pathophysiologic mechanism
Decreased platelet production Increased platelet destruction
Drug-induced marrow Immune thrombocytopenia
suppression Drug-induced immune platelet destruction
Infection Heparin-induced thrombocytopenia
Liver disease (thrombopoietin Post-transfusion purpura
deficiency) Sepsis
Cardiopulmonary bypass
Extracorporeal membrane oxygenation (ECMO)
Continuous venovenous haemodialysis
Microangiopathy (e.g. thrombotic microangiopathy, disseminated
intravascular coagulation)
Neonatal alloimmune thrombocytopenia (neonates only)
days after blood product (typically platelet) transfusion may
herald post-transfusion purpura secondary to prior sensiti-
sation to certain platelet glycoproteins (typically HPA-1a)
foreign to the patient (either via prior transfusion or
pregnancy).
Infection or sepsis may lead to precipitous decreases in the
platelet count over a short period of time, which may be
because of disseminated intravascular coagulation (DIC) or to
the secretion of neuraminidase by certain bacteria; neur-
aminidase then cleaves sialic acids from platelets and leads to
rapid platelet destruction by the Ashwell-Morrel receptor in
the liver.41 Acute, decompensated DIC is not uncommon in
critically ill patients. The diagnosis of DIC is made in the setting
of oozing from an i. v. line or venepuncture sites, mucocuta-
neous and possibly deeper tissue bleeding, thrombocytopenia,
hypofibrinogenemia, and elevated prothrombin and partial
thromboplastin times. D-dimer measurement is elevated in
these patients, although this test may be of little utility in the
postoperative patient who would be expected to have an
elevated D-dimer at baseline.
Additionally, thrombocytopenia may be attributable to the
medications (vide infra) administered to treat infection and it is
oftentimes impossible to discern whether thrombocytopenia
is as a result of infection or to the antibiotics.
Figure 3 illustrates an approach to thrombocytopenia in the
postoperative patient. Of note, bone marrow examination is not
mentioned in this algorithm; it is not a routine part of this
evaluation and is reserved for rare circumstances in which
acute bone marrow failure is a major consideration, such as
drug-induced aplastic anaemia. While an in-depth discussion of
each major aetiology of postoperative thrombocytopenia is
beyond the scope of this review, drug-induced thrombocyto-
penia and heparin-induced thrombocytopenia will be discussed
in further detail owing to their considerable perioperative rele-
vance and management challenges.
Drug-induced thrombocytopenia
Drugs may cause thrombocytopenia by a variety of different
pathophysiologic mechanisms, including immune-mediated
destruction (e.g. vancomycin), bone marrow suppression (e.g.
linezolid), precipitation of thrombotic microangiopathy (e.g.
cyclosporine), and induction of platelet activation (e.g. hepa-
rin). The most common drugs resulting in thrombocytopenia
in the postoperative setting are heparin agents and antibiotics.
For specific queries, an online database listing all relevant
Platelet sequestration
or dilution
Significant i.v. fluid
administration
Massive red blood cell
transfusion
Splenomegaly
drug-induced thrombocytopenia publications for any given
drug is available.42
The drugs most relevant to the perioperative setting with the
greatest evidence for association with thrombocytopenia after
exposure include glycoprotein IIb/IIIa inhibitors, amiodarone,
ampicillin, haloperidol, ibuprofen, naproxen, piperacillin, ra-
nitidine, trimethoprim-sulfamethoxazole, and vancomycin.43
Each agent only rarely causes thrombocytopenia, so analysis
of timing of drug initiation vs thrombocytopenia onset is critical
to establish the link. For most drugs, the incidence of throm-
bocytopenia after exposure is not known and data associating
the agent with thrombocytopenia are limited to published case
reports and case series.
The severity of thrombocytopenia caused by any drug is
highly variable.44 If the patient’s platelet count is high enough
that postoperative bleeding is not a significant concern, the
count can be monitored over the course of drug treatment
assuming no concerns for a more serious aetiology (e.g. drug-
induced thrombotic microangiopathy). When the platelet
count decreases to the point that bleeding is a concern, drug(s)
considered to be the most likely offenders should be dis-
continued, if possible. The diagnosis may sometimes be
confirmed clinically if the platelet count recovers after drug
discontinuation.
Most drugs result in thrombocytopenia via an immune-
mediated mechanism (although the rate of immune destruc-
tion, and therefore the nadir platelet count, varies widely);
therefore, diagnostic testing in the form of an assay for drug-
dependent antiplatelet antibodies may be performed, but re-
sults typically are not known in time to guide urgent clinical
decision-making. This laboratory confirmation may be espe-
cially helpful to determine the specific cause when multiple
drugs are suspected. Importantly, a positive test is helpful in
confirmation of the specific drug responsible, but negative
testing does not exclude a drug as a cause of
thrombocytopenia.45
In cases in which the suspected drug is essential and
required for an extended duration, platelet transfusion can be
attempted, but may be ineffective in the setting of brisk
immune-mediated platelet destruction. Unlike in primary ITP,
corticosteroids have not demonstrated efficacy in drug-
induced ITP.46 Therefore, an individualised approach to
management is warranted. Intravenous immunoglobulin may
work well, but if it is ineffective, use of a thrombopoietin (TPO)
receptor agonist is a reasonable consideration. Although an
off-label use for TPO receptor agonists, these agents are
 Perioperative thrombocytopenia - 25

Fig 3. Approach to the patient with postoperative thrombocytopenia. AIHA, autoimmune haemolytic anaemia; DIC, disseminated intra-
vascular coagulation; DTI, direct thrombin inhibitor; ECMO, extracorporeal membrane oxygenation; HELLP, haemolysis-elevated liver
enzymes-low platelets syndrome; HIPA, heparin-induced platelet activation; HIT, heparin-induced thrombocytopenia; ITP, immune
thrombocytopenia; IVF, i.v. fluid; PF4, platelet factor 4; Plt, platelet; PTP, post-transfusion purpura; RBC, red blood cell; SRA, serotonin
release assay; TCP, thrombocytopenia; TMA, thrombotic microangiopathy.

approved by the US Food and Drug Administration (FDA) for
chronic ITP and have also demonstrated efficacy in manage-
ment of a different subset of drug-induced thrombocytopenia
(chemotherapy-induced thrombocytopenia).47,48
Heparin-induced thrombocytopenia
Heparin-induced thrombocytopenia (HIT) occurs in up to 2.5%
of inpatients receiving heparin and is most common in
surgical patients (especially those subject to major trauma or
major surgery).49e51 HIT results from immunologic recognition
of heparin-platelet factor 4 (PF4) complexes typically 5e10
days after heparin initiation, resulting in platelet activation,
thrombocytopenia, and high thrombotic risk.52,53 Incidence of
venous thromboembolism is 17e55%, incidence of arterial
thrombosis is 3e10%, and mortality is 5e10% in patients with
HIT.49,54 Asymptomatic HIT must be distinguished clinically
from the modest platelet count decreases that may occur in
 26 - Nagrebetsky et al.
the first several days after beginning heparin, a clinically
insignificant entity known as heparin-associated thrombocy-
topenia (HAT). The distinction between these two entities is
especially relevant to those patients undergoing cardiopul-
monary bypass, in whom nearly half develop HAT, but only
1e2% develop HIT.55
The first step in diagnosis is application of the 4Ts scoring
system (Table 2).56 Patients with intermediate or higher
probability of HIT by the 4Ts score should undergo assessment
for anti-heparin-PF4 antibodies [typically using an IgG-specific
enzyme-linked immunosorbent assay (ELISA)], receive four-
limb Doppler ultrasonography to screen for occult venous
thrombosis,57 and heparin should be replaced with an alter-
native agent (typically the direct thrombin inhibitors arga-
troban or bivalirudin, although fondaparinux can be used off-
label).49,58 If ELISA testing is suggestive of HIT, a functional
assay (serotonin release assay or heparin-induced platelet
aggregation) assessing for platelet activation in the presence
of heparin is confirmatory.59 If confirmed, all heparin-
containing products are to be avoided except under specific,
time-limited, controlled clinical circumstances in which no
other suitable options exist (such as cardiopulmonary bypass).
After withdrawal of heparin, the platelet count will grad-
ually normalise over the following 3e4 days; once 100109
L 1, i. v. direct thrombin inhibitor therapy should be transi-
tioned to an oral anticoagulant or s. c. fondaparinux.58 The
greatest clinical experience is with warfarin, though a direct
oral anticoagulant is reasonable.57 The duration of the pro-
thrombotic state in a patient with HIT is not entirely clear.
For those without thrombosis, guidelines suggest anti-
coagulation should continue for a minimum of 4 weeks.49
Platelet function testing
Platelet function assessment is performed to identify quali-
tative platelet defects (e.g. inherited platelet function disor-
ders, assessment of pharmacologic antiplatelet agents). While
thrombocytopenia is much more common in the perioperative
setting than qualitative defects, a brief discussion of the most
common platelet function testing follows, as it may be rele-
vant in select cases.
A platelet function test of historical interest is the bleeding
time, largely abandoned because of poor reproducibility and
Table 2 4Ts score for diagnosis of heparin-induced thrombocytopenia. Scoring is as follows: 3 points, low probability of HIT; 4e5
points, intermediate probability of HIT; 6e8 points, high probability of HIT.56 Plt, platelet
Parameter 2 Points
Thrombocytopenia Plt count decrease >50% from
baseline with Plt nadir
20109 L 1
Timing of platelet Onset 5e10 days after heparin
count decrease exposure OR onset 1 day if prior
heparin exposure within the past
30 days
Thrombosis or New proved thrombosis, skin
other sequelae necrosis, or acute systemic
reaction after i.v. heparin bolus
Other causes for None apparent
thrombocytopenia
poor sensitivity to non-severe platelet function defects.60 A
more modern screening test for qualitative platelet abnor-
malities is platelet function analyser (PFA) testing, a specific
instrument that reports the length of time before a platelet
plug forms to stop flow through an aperture coated with
platelet agonists (closure time or CT). While convenient, PFA
testing is sensitive to haematocrit and platelet count and
insensitive to mild platelet defects, limiting its value both
generally and in the perioperative setting.60
A more useful platelet function assay is light transmission
platelet aggregometry. In this laboratory test, platelet-rich
plasma is placed in cuvettes between a light source and
photocell, with various platelet agonists (such as adenosine
diphosphate, collagen, epinephrine, and arachidonic acid)
added, resulting in platelet activation and aggregation. This
increases light transmission through the cuvette over time
which is read by the photomultiplier cell and captured as a
tracing for each agonist. Platelet response to each agonist al-
lows for diagnosis of specific platelet defects; for example,
aggregation in response to arachidonic acid is diminished in
patients on aspirin. While light transmission platelet aggreg-
ometry is considered the gold standard, its major limitation is
its susceptibility to numerous pre-analytical variables that can
impact test results (such as agitation of the tubes before
testing, as in a hospital’s pneumatic tube system). It cannot be
performed accurately at platelet counts under 100109 L 1.
Moreover, it is poorly standardised, has a long turnaround
time, and requires numerous tubes of blood to be drawn.61
Assessment of viscoelastic properties of blood in vitro
(thromboelastography) may also provide useful information
on platelet function. However, this technique evaluates the
combined effects of many components of haemostasis. Thus,
it is not a platelet function test in a strict sense and is outside
of the scope of this review.
Alternatives to platelet transfusion in the
thrombocytopenic surgical patient
While platelet transfusion is the only currently available
therapy capable of achieving a rapid, generally predictable
increase in platelet count, platelet transfusion has several
risks. In addition to the relatively common complications of
blood product transfusion (volume overload and febrile non-
1 Point 0 Points
Plt count decrease 30e50% from Plt count decrease <30% or
baseline OR Plt nadir 10e19109 L 1
nadir <10109 L 1
Consistent with onset 5e10 days after Onset <4 days without
heparin exposure but not entirely recent heparin exposure
clear OR onset after 10 days from
heparin exposure OR onset 1 day
with prior heparin exposure within
the past 30e100 days
Progressive or recurrent thrombosis, None
erythematous skin lesions, or
suspected, unproved thrombosis
Possible Definite

haemolytic transfusion reactions), less common but impor-
tant risks associated with platelet transfusion include
transfusion-associated acute lung injury, allergic reactions,
clinically-relevant immunomodulation, post-transfusion pur-
pura, infectious risk, and alloimmunisation with subsequent
ineffectiveness of platelet transfusion. Because platelets
cannot be frozen, they have the highest risk of bacterial sepsis
of any blood product.62 The immunomodulatory effects of
blood product transfusion are particularly salient in the sur-
gical and critical care populations, for whom the transient
immunosuppression has been demonstrated to increase
infection risk63 and even increase the risk of cancer relapse
after oncologic surgery.64
Given these risks and the relative scarcity of platelets as a
resource, proper alternatives to platelet transfusion should
always be considered in the thrombocytopenic surgical pa-
tient. Non-specific haemostatic agents such as desmopressin,
the antifibrinolytic agents (epsilon aminocaproic acid and
tranexamic acid), or procoagulant bypass agents [recombinant
factor VIIa and activated prothrombin complex concentrates
(aPCC)] are a consideration, and the TPO receptor agonists
(romiplostim, eltrombopag, avatrombopag and lusu-
trombopag) can raise the platelet count pharmacologically.
Raising the haematocrit may also improve haemostasis in the
patient who is both thrombocytopenic and anaemic.
Desmopressin
Desmopressin (1-deamino-8-D-arginine vasopressin, brand
name DDAVP) is a rapid-acting vasopressin analogue that acts
principally by inducing the release of large von Willebrand
factor multimers from Weibel-Palade bodies in the vascular
endothelium.65 Desmopressin can be given by i. v., s. c., or
intranasal routes and is commonly used in patients with mild
haemophilia A and milder forms of von Willebrand disease to
minimise bleeding in the perioperative setting. Desmopressin
is also commonly used to promote haemostasis in patients
with uremic platelet dysfunction based on improvements
demonstrated in platelet function testing,66,67 although there
have been no large, randomised studies demonstrating the
magnitude of this benefit. It is potentially attractive from a
mechanistic perspective for use as a non-specific haemostatic
agent in thrombocytopenic patients, as increased circulating
von Willebrand factor could theoretically improve platelet
adhesion to the endothelial surface. One small randomised
trial of cirrhotic thrombocytopenic patients undergoing dental
extraction demonstrated no difference between desmopressin
and platelet transfusion in reducing bleeding events, but only
one bleeding event occurred in the trial population, rendering a
comparison of efficacy impossible.68 Despite the lack of evi-
dence, desmopressin can be considered a non-specific hae-
mostatic agent in thrombocytopenic surgical patients who are
uremic or refractory to platelet transfusions. This agent is well-
tolerated, with hyponatremia being the most common adverse
effect. This necessitates plasma sodium monitoring in all pa-
tients and administration no more frequently than three times
daily.69 Tachyphylaxis develops rapidly after the initial dose as
a result of depletion of von Willebrand factor from endothelial
cells,65 so desmopressin is ideal for a short duration of use.
Antifibrinolytic agents
The antifibrinolytic agents epsilon aminocaproic acid and tra-
nexamic acid are synthetic analogues of the amino acid lysine
Perioperative thrombocytopenia - 27
and function as non-specific haemostatic agents via competitive
inhibition of plasminogen, blocking its binding to fibrin strands
and reducing clot lysis. These agents are used in a wide variety of
clinical settings mostly outside the thrombocytopenic setting to
promote haemostasis, especially in patients with mucocuta-
neous bleeding or perioperative bleeding. Efficacy has been
demonstrated in non-thrombocytopenic patients undergoing
open cardiac surgery,70,71 orthopaedic surgery,72 and extracor-
poreal membrane oxygenation,73 among numerous others.
Considerably less evidence is available in thrombocytopenic
patients, however. Efficacy was shown in a small ITP cohort,74
but results were equivocal in patients with bone marrow fail-
ure.75 It is reasonable to consider a trial of antifibrinolytic therapy
in a thrombocytopenic patient refractory to platelet transfusion
or for whom platelet transfusion is not acceptable (i.e. a Jeho-
vah’s Witness).
Procoagulant bypass agents
The procoagulant bypass agent recombinant factor VIIa
(rFVIIa, eptacog alfa) and activated prothrombin complex
concentrates (aPCC) are widely used to achieve haemostasis in
patients with haemophilia and factor inhibitors. The largest
study utilising these agents in the thrombocytopenic setting
was a randomised trial of 100 thrombocytopenic hematopoi-
etic stem cell transplant patients with moderate or severe
bleeding; rFVIIa administration was not effective in improving
the patient’s bleeding score, but did result in increased inci-
dence of venous thromboembolism.76 A Cochrane analysis of
29 randomised controlled trials of rFVIIa use for the preven-
tion and treatment of bleeding outside of the haemophilia
setting more generally (including studies of patients with and
without thrombocytopenia) demonstrated decreased red cell
transfusion requirements and decreased blood loss, but no
evidence of mortality benefit and an increase in arterial
thromboembolic events with the use of rFVIIa.77 Given the lack
of evidence demonstrating efficacy, the considerable financial
cost, and the concern for promotion of thrombosis, rFVIIa and
aPCC should not be considered as alternatives to platelet
transfusion.
Thrombopoietin receptor agonists
The TPO receptor agonists romiplostim, eltrombopag, ava-
trombopag, and lusutrombopag act on bone marrow mega-
karyocytes and megakaryocyte progenitors to increase
platelet production. Because of the normal kinetics of
thrombopoiesis, the platelet count begins to increase
approximately 5e7 days after administration of a TPO re-
ceptor agonist and peaks 10e14 days later (so none of these
agents are capable of achieving a rapid platelet increase).
Given these time constraints, use of these agents for the
management of perioperative thrombocytopenia has been
examined in multiple studies. A retrospective cohort study of
47 thrombocytopenic patients receiving romiplostim to in-
crease platelet count for surgery (median baseline platelet
count 47109 L 1) showed that romiplostim was highly effi-
cacious for this purpose, with median platelet count on the
day of surgery of 164109 L 1 and all but one patient under-
going surgery on the originally scheduled date.78 Median time
to platelet count 100109 L 1 was 13 days. Most patients in
this cohort underwent major surgery (e.g. open cardiac sur-
gery, total joint arthroplasty, abdominopelvic surgery, or
neurological surgery).
 28 - Nagrebetsky et al.
Because thrombocytopenia of chronic liver disease is due in
large part to TPO deficiency, use of TPO receptor agonists is an
attractive prospect in this patient population. In a large,
randomised, controlled trial of 292 patients with chronic liver
disease and thrombocytopenia who required a minor elective
invasive procedure, eltrombopag or placebo was administered
for 14 days before a planned elective invasive procedure.
Eltrombopag was effective in reducing platelet transfusions
and bleeding, but the study was terminated early because of
increased incidence of portal venous thrombosis in the
eltrombopag arm. Similarly, more recent studies of novel non-
peptide TPO receptor agonists (lusutrombopag and ava-
trombopag) for perioperative thrombocytopenia in chronic
liver disease patients have demonstrated similar efficacy out-
comes without increased rates of portal vein thrombosis.79,80
Avatrombopag was approved by the US FDA in May 2018 for
treatment of perioperative thrombocytopenia in patients with
chronic liver disease, based on the results on two randomised
controlled trials enrolling a total of 435 patients demonstrating
increased platelet counts (approximately 30e50109 L 1 higher
on average) treated with 5 days of avatrombopag vs placebo
before minor invasive procedures.80 There was no increase in
thromboembolism in patients treated with avatrombopag.
Subsequently, lusutrombopag was approved by the US FDA in
July 2018 for treatment of perioperative thrombocytopenia in
patients with chronic liver disease based on the results of two
randomised controlled trials enrolling a total of 312 patients
demonstrating similar improvements in platelet count after 7
days of preoperative lusutrombopag.79
Raising the haematocrit
Haemostasis is improved in renal failure patients by
increasing the haematocrit through transfusion or erythro-
poietin treatment.81e84 The proposed mechanism is that the
higher RBC mass (which flows in the centre of the blood vessel)
‘pushes’ platelets closer to the endothelial cell wall, thereby
increasing the haemostatic effect. No definitive studies have
documented such an increased haemostatic effect in throm-
bocytopenic patients who are not uremic.
Conclusions
The bleeding risk in thrombocytopenic patients undergoing
surgical procedures is difficult to predict, and at any given
platelet count, congenital or acquired platelet dysfunction
may render available platelets less effective. Thus, the platelet
count at which a given procedure carries an acceptable risk of
bleeding is unique in each patient. Recommendations for
procedure-specific ‘safe’ platelet counts often lack a robust
evidence base and may reflect anecdotal experiences of those
performing the procedure. The true magnitude of benefit of
prophylactic platelet transfusion remains unclear and the
need for platelet transfusion should be assessed on a patient-
by-patient basis. In non-emergent surgeries or postoperative
thrombocytopenic patients, haematology consultation may be
helpful in diagnosing the aetiology of thrombocytopenia,
selecting appropriate therapy, and determining the utility of
platelet transfusion or its alternatives.
Authors’ contributions
Designed the review, collected data, wrote the manuscript,
created tables and figures: A.N., H.A.
Revised the review: A.N., H.A., N.D.
Wrote a section of the manuscript: N.D., J.W.K.
Critically revised the review: D.K., J.W.K.
Declarations of interest
H.A. is a consultant for Agios. D.J.K. is a consultant for Amgen,
Genzyme, Pfizer, Agios, Novartis, Syntimmune, Argenx, Fuji-
film, Rigel, Momenta, Principia, Zafgen, Seattle Genetics, Up-
to-Date, Merck, Alnylam, and Bioveratif and receives
research funding from Bristol Myers Squibb, Amgen, Syn-
timmune, Principia, Rigel, Agios, Alnylam, and Bioveratif.
Acknowledgements
We thank our colleagues for providing photomicrographs for
use in this review: A. Sohani of the Department of Pathology,
Massachusetts General Hospital, Boston, MA, USA (Fig. 2a) and
Y. Li of the Department of Pathology, Advocate Christ Medical
Center, Oak Lawn, IL, USA (Fig. 2b).
References
1. Glance LG, Blumberg N, Eaton MP, et al. Preoperative
thrombocytopenia and postoperative outcomes after
noncardiac surgery. Anesthesiology 2014; 120: 62e75
2. Qureshi AI, Zafar A, Miran MS, Jani VB. Carotid endarter-
ectomy in patients with thrombocytopenia: analysis of
the national surgical quality improvement program reg-
istry. Oper Neurosurg (Hagerstown) 2017; 13: 150e6
3. Buckley MF, James JW, Brown DE, et al. A novel approach
to the assessment of variations in the human platelet
count. Thromb Haemost 2000; 83: 480e4
4. Brecher G, Cronkite EP. Morphology and enumeration of
human blood platelets. J Appl Physiol 1950; 3: 365e77
5. Slichter SJ. Relationship between platelet count and
bleeding risk in thrombocytopenic patients. Transfus Med
Rev 2004; 18: 153e67
6. Lam FW, Vijayan KV, Rumbaut RE. Platelets and their in-
teractions with other immune cells. Compr Physiol 2015; 5:
1265e80
7. Li Z, Yang F, Dunn S, Gross AK, Smyth SS. Platelets as
immune mediators: their role in host defense responses
and sepsis. Thromb Res 2011; 127: 184e8
8. de Stoppelaar SF, van ’t Veer C, van der Poll T. The role of
platelets in sepsis. Thromb Haemost 2014; 112: 666e77
9. Lefrancais E, Ortiz-Munoz G, Caudrillier A, et al. The lung
is a site of platelet biogenesis and a reservoir for haema-
topoietic progenitors. Nature 2017; 544: 105e9
10. Williamson DR, Albert M, Heels-Ansdell D, et al. Throm-
bocytopenia in critically ill patients receiving thrombo-
prophylaxis: frequency, risk factors, and outcomes. Chest
2013; 144: 1207e15
11. Josephson CD, Granger S, Assmann SF, et al. Bleeding
risks are higher in children versus adults given prophy-
lactic platelet transfusions for treatment-induced hyper-
proliferative thrombocytopenia. Blood 2012; 120: 748e60
12. Chantarangkul V, Clerici M, Bressi C, Giesen PL, Tripodi A.
Thrombin generation assessed as endogenous thrombin
potential in patients with hyper- or hypo-coagulability.
Haematologica 2003; 88: 547e54

13. Slichter SJ, Kaufman RM, Assmann SF, et al. Dose of pro-
phylactic platelet transfusions and prevention of hemor-
rhage. N Engl J Med 2010; 362: 600e13
14. Kaufman RM, Djulbegovic B, Gernsheimer T, et al. Platelet
transfusion: a clinical practice guideline from the AABB.
Ann Intern Med 2015; 162: 205e13
15. Uhl L, Assmann SF, Hamza TH, Harrison RW,
Gernsheimer T, Slichter SJ. Laboratory predictors of
bleeding and the effect of platelet and RBC transfusions
on bleeding outcomes in the PLADO trial. Blood 2017; 130:
1247e58
16. Cortelazzo S, Finazzi G, Buelli M, Molteni A, Viero P,
Barbui T. High risk of severe bleeding in aged patients
with chronic idiopathic thrombocytopenic purpura. Blood
1991; 77: 31e3
17. Norfolk D, editor. Handbook of transfusion medicine. 5th Edn.
United Kingdom: Blood Services, TSO; 2013
18. British Committee for Standards in Haematology, Blood
Transfusion Task Force. Guidelines for the use of platelet
transfusions. Br J Haematol 2003; 122: 10e23
19. Van Os EC, Kamath PS, Gostout CJ, Heit JA. Gastroenter-
ological procedures among patients with disorders of
hemostasis: evaluation and management recommenda-
tions. Gastrointest Endosc 1999; 50: 536e43
20. Kumar A, Mhaskar R, Grossman BJ, et al. Platelet trans-
fusion: a systematic review of the clinical evidence.
Transfusion 2015; 55(1116): 27. quiz 1115
21. Warner MA, Jia Q, Clifford L, et al. Preoperative platelet
transfusions and perioperative red blood cell re-
quirements in patients with thrombocytopenia undergo-
ing noncardiac surgery. Transfusion 2016; 56: 682e90
22. Thakrar SV, Mallett SV. Thrombocytopenia in cirrhosis:
impact of fibrinogen on bleeding risk. World J Hepatol 2017;
9: 318e25
23. Zhang Z, Zhang Y, Wang W, et al. Thrombocytopenia and
the outcomes of hepatectomy for hepatocellular carci-
noma: a meta-analysis. J Surg Res 2017; 210: 99e107
24. Dasenbrock HH, Devine CA, Liu KX, et al. Thrombocyto-
penia and craniotomy for tumor: a national surgical
quality improvement program analysis. Cancer 2016; 122:
1708e17
25. Pamecha V, Mahansaria SS, Kumar S, et al. Association of
thrombocytopenia with outcome following adult living
donor liver transplantation. Transpl Int 2016; 29: 1126e35
26. Warner MA, Woodrum D, Hanson A, Schroeder DR,
Wilson G, Kor DJ. Preprocedural platelet transfusion for
patients with thrombocytopenia undergoing interven-
tional radiology procedures is not associated with reduced
bleeding complications. Transfusion 2017; 57: 890e8
27. Mokart D, Penalver M, Chow-Chine L, et al. Surgical
treatment of acute abdominal complications in hematol-
ogy patients: outcomes and prognostic factors. Leuk Lym-
phoma 2017; 58: 2395e402
28. Hohenberger P, Buchheidt D. Surgical interventions in
patients with hematologic malignancies. Crit Rev Oncol
Hematol 2005; 55: 83e91
29. Diedrich B, Remberger M, Shanwell A, Svahn BM,
Ringden O. A prospective randomized trial of a prophy-
lactic platelet transfusion trigger of 10 x 10(9) per L versus
30 x 10(9) per L in allogeneic hematopoietic progenitor cell
transplant recipients. Transfusion 2005; 45: 1064e72
30. Nandagopal L, Veeraputhiran M, Jain T, Soubani AO,
Schiffer CA. Bronchoscopy can be done safely in patients
with thrombocytopenia. Transfusion 2016; 56: 344e8
Perioperative thrombocytopenia - 29
31. Kim YH, Suh GY, Kim MH, et al. Safety and usefulness of
bronchoscopy in ventilator-dependent patients with se-
vere thrombocytopenia. Anaesth Intensive Care 2008; 36:
411e7
32. Weiss SM, Hert RC, Gianola FJ, Clark JG, Crawford SW.
Complications of fiberoptic bronchoscopy in thrombocy-
topenic patients. Chest 1993; 104: 1025e8
33. Du Rand IA, Blaikley J, Booton R, et al. British thoracic
society guideline for diagnostic flexible bronchoscopy in
adults: Accredited by NICE. Thorax 2013; 68: i1e44
34. Baharoglu MI, Cordonnier C, Al-Shahi Salman R, et al.
Platelet transfusion versus standard care after acute
stroke due to spontaneous cerebral haemorrhage associ-
ated with antiplatelet therapy (PATCH): a randomised,
open-label, phase 3 trial. Lancet 2016; 387: 2605e13
35. Kong KL, Khan J. Ophthalmic patients on antithrombotic
drugs: a review and guide to perioperative management.
Br J Ophthalmol 2015; 99: 1025e30
36. Orsini S, Noris P, Bury L, et al. Bleeding risk of surgery and
its prevention in patients with inherited platelet disor-
ders. Haematologica 2017; 102: 1192e203
37. Pegels JG, Bruynes EC, Engelfriet CP, von dem Borne AE.
Pseudothrombocytopenia: an immunologic study on
platelet antibodies dependent on ethylene diamine tetra-
acetate. Blood 1982; 59: 157e61
38. Bobba RK, Doll DC. Platelet satellitism as a cause of
spurious thrombocytopenia. Blood 2012; 119: 4100
39. Kuter DJ. General aspects of thrombocytopenia, platelet
transfusions, and thrombopoietic growth factors. In:
Kitchens C, Konkle B, Kessler C, editors. Consultative he-
mostasis and thrombosis. Philadelphia: Elsevier Saunders;
2013. p. 103e16
40. Diab YA, Wong EC, Luban NL. Massive transfusion in
children and neonates. Br J Haematol 2013; 161: 15e26
41. Li Y, Fu J, Ling Y, et al. Sialylation on O-glycans protects
platelets from clearance by liver kupffer cells. Proc Natl
Acad Sci U S A 2017; 114: 8360e5
42. Platelets on the web. Hematology-oncology section. Depart-
ment of Medicine, Department of Biostatistics & Epide-
miology, College of Public Health University of Oklahoma
Health Sciences Center. Available from: https://www.
ouhsc.edu/platelets/ditp.html. [Accessed 1 August 2018].
43. Reese JA, Li X, Hauben M, et al. Identifying drugs that
cause acute thrombocytopenia: an analysis using 3
distinct methods. Blood 2010; 116: 2127e33
44. Aster RH, Curtis BR, McFarland JG, Bougie DW. Drug-
induced immune thrombocytopenia: pathogenesis,
diagnosis, and management. J Thromb Haemost 2009; 7:
911e8
45. Curtis BR, McFarland JG, Wu GG, Visentin GP, Aster RH.
Antibodies in sulfonamide-induced immune thrombocy-
topenia recognize calcium-dependent epitopes on the
glycoprotein IIb/IIIa complex. Blood 1994; 84: 176e83
46. Pedersen-Bjergaard U, Andersen M, Hansen PB. Drug-
induced thrombocytopenia: clinical data on 309 cases and
the effect of corticosteroid therapy. Eur J Clin Pharmacol
1997; 52: 183e9
47. Al-Samkari H, Marshall AL, Goodarzi K, Kuter DJ. The use
of romiplostim in treating chemotherapy-induced
thrombocytopenia in patients with solid tumors. Haema-
tologica 2018; 103: e169e72
48. Soff GA, Miao Y, Devlin SM, et al. Romiplostim for
chemotherapy-induced thrombocytopenia (CIT). Results
of a phase 2 trial. Blood 2017; 130: 289
 30 - Nagrebetsky et al.
49. Linkins LA, Dans AL, Moores LK, et al. Treatment and
prevention of heparin-induced thrombocytopenia:
antithrombotic therapy and prevention of thrombosis, 9th
ed: American college of chest physicians evidence-based
clinical practice guidelines. Chest 2012; 141(2 Suppl):
e495Se530S
50. Greinacher A, Alban S, Omer-Adam MA, Weitschies W,
Warkentin TE. Heparin-induced thrombocytopenia: a
stoichiometry-based model to explain the differing im-
munogenicities of unfractionated heparin, low-
molecular-weight heparin, and fondaparinux in different
clinical settings. Thromb Res 2008; 122: 211e20
51. Lubenow N, Hinz P, Thomaschewski S, et al. The severity
of trauma determines the immune response to PF4/hep-
arin and the frequency of heparin-induced thrombocyto-
penia. Blood 2010; 115: 1797e803
52. Arepally GM. Heparin-induced thrombocytopenia. Blood
2017; 129: 2864e72
53. Visentin GP, Ford SE, Scott JP, Aster RH. Antibodies from
patients with heparin-induced thrombocytopenia/throm-
bosis are specific for platelet factor 4 complexed with
heparin or bound to endothelial cells. J Clin Invest 1994; 93:
81e8
54. Kuter DJ, Konkle BA, Hamza TH, et al. Clinical outcomes in
a cohort of patients with heparin-induced thrombocyto-
penia. Am J Hematol 2017; 92: 730e8
55. Selleng S, Malowsky B, Strobel U, et al. Early-onset and
persisting thrombocytopenia in post-cardiac surgery pa-
tients is rarely due to heparin-induced thrombocytopenia,
even when antibody tests are positive. J Thromb Haemost
2010; 8: 30e6
56. Pouplard C, Gueret P, Fouassier M, et al. Prospective
evaluation of the ’4Ts’ score and particle gel immuno-
assay specific to heparin/PF4 for the diagnosis of heparin-
induced thrombocytopenia. J Thromb Haemost 2007; 5:
1373e9
57. Cuker A. Management of the multiple phases of heparin-
induced thrombocytopenia. Thromb Haemost 2016; 116:
835e42
58. Schindewolf M, Steindl J, Beyer-Westendorf J, et al. Use of
fondaparinux off-label or approved anticoagulants for
management of heparin-induced thrombocytopenia. J Am
Coll Cardiol 2017; 70: 2636e48
59. Linkins LA, Warkentin TE. The approach to heparin-
induced thrombocytopenia. Semin Respir Crit Care Med
2008; 29: 66e74
60. Harrison P, Mackie I, Mumford A, et al. Guidelines for the
laboratory investigation of heritable disorders of platelet
function. Br J Haematol 2011; 155: 30e44
61. Cattaneo M, Cerletti C, Harrison P, et al. Recommenda-
tions for the standardization of light transmission
aggregometry: a consensus of the working party from the
platelet physiology subcommittee of SSC/ISTH. J Thromb
Haemost 2013; 11: 1183e9
62. Corash L. Bacterial contamination of platelet components:
potential solutions to prevent transfusion-related sepsis.
Expert Rev Hematol 2011; 4: 509e25
63. Alter HJ, Klein HG. The hazards of blood transfusion in
historical perspective. Blood 2008; 112: 2617e26
64. Wang T, Luo L, Huang H, et al. Perioperative blood
transfusion is associated with worse clinical outcomes
in resected lung cancer. Ann Thorac Surg 2014; 97:
1827e37
65. Zeigler ZR, Megaludis A, Fraley DS. Desmopressin (d-
DAVP) effects on platelet rheology and von Willebrand
factor activities in uremia. Am J Hematol 1992; 39: 90e5
66. Kim JH, Baek CH, Min JY, Kim JS, Kim SB, Kim H. Desmo-
pressin improves platelet function in uremic patients
taking antiplatelet agents who require emergent invasive
procedures. Ann Hematol 2015; 94: 1457e61
67. Escolar G, Cases A, Monteagudo J, et al. Uremic plasma
after infusion of desmopressin (DDAVP) improves the
interaction of normal platelets with vessel sub-
endothelium. J Lab Clin Med 1989; 114: 36e42
68. Stanca CM, Montazem AH, Lawal A, Zhang JX, Schiano TD.
Intranasal desmopressin versus blood transfusion in
cirrhotic patients with coagulopathy undergoing dental
extraction: a randomized controlled trial. J Oral Maxillofac
Surg 2010; 68: 138e43
69. Das P, Carcao M, Hitzler J. DDAVP-induced hyponatremia
in young children. J Pediatr Hematol Oncol 2005; 27: 330e2
70. Daily PO, Lamphere JA, Dembitsky WP, Adamson RM,
Dans NF. Effect of prophylactic epsilon-aminocaproic
acid on blood loss and transfusion requirements in pa-
tients undergoing first-time coronary artery bypass
grafting. A randomized, prospective, double-blind study.
J Thorac Cardiovasc Surg 1994; 108: 99e106. discussion
106e8
71. Lu J, Meng H, Meng Z, et al. Epsilon aminocaproic acid
reduces blood transfusion and improves the coagula-
tion test after pediatric open-heart surgery: a meta-
analysis of 5 clinical trials. Int J Clin Exp Pathol 2015; 8:
7978e87
72. Zufferey P, Merquiol F, Laporte S, et al. Do antifibrinolytics
reduce allogeneic blood transfusion in orthopedic sur-
gery? Anesthesiology 2006; 105: 1034e46
73. Buckley LF, Reardon DP, Camp PC, et al. Aminocaproic
acid for the management of bleeding in patients on
extracorporeal membrane oxygenation: four adult case
reports and a review of the literature. Heart Lung 2016; 45:
232e6
74. Bartholomew JR, Salgia R, Bell WR. Control of bleeding in
patients with immune and nonimmune thrombocyto-
penia with aminocaproic acid. Arch Intern Med 1989; 149:
1959e61
75. Estcourt LJ, Desborough M, Brunskill SJ, et al. Anti-
fibrinolytics (lysine analogues) for the prevention of
bleeding in people with haematological disorders.
Cochrane Database Syst Rev 2016; 3: CD009733
76. Pihusch M, Bacigalupo A, Szer J, et al. Recombinant acti-
vated factor VII in treatment of bleeding complications
following hematopoietic stem cell transplantation.
J Thromb Haemost 2005; 3: 1935e44
77. Simpson E, Lin Y, Stanworth S, Birchall J, Doree C, Hyde C.
Recombinant factor VIIa for the prevention and treatment
of bleeding in patients without haemophilia. Cochrane
Database Syst Rev 2012; 3: CD005011
78. Al-Samkari H, Marshall AL, Goodarzi K, Kuter DJ. Romi-
plostim for the management of perioperative thrombo-
cytopenia. Br J Haematol 2018; 182: 106e13
79. Afdhal N, Duggal A, Ochiai T, et al. Platelet response to
lusutrombopag, a thrombopoietin receptor agonist, in
patients with chronic liver disease and thrombocytopenia
undergoing non-emergency invasive procedures: results
form a phase 3 randomized, double-blind, placebo-
controlled study. Blood 2017; 130: 291

80. Terrault N, Chen YC, Izumi N, et al. Avatrombopag before
procedures reduces need for platelet transfusion in pa-
tients with chronic liver disease and thrombocytopenia.
Gastroenterology 2018; 155: 705e18
81. el-Shahawy MA, Francis R, Akmal M, Massry SG. Recom-
binant human erythropoietin shortens the bleeding time
and corrects the abnormal platelet aggregation in hemo-
dialysis patients. Clin Nephrol 1994; 41: 308e13
82. Gordge MP, Leaker B, Patel A, Oviasu E, Cameron JS,
Neild GH. Recombinant human erythropoietin shortens
Perioperative thrombocytopenia - 31
the uraemic bleeding time without causing intravas-
cular haemostatic activation. Thromb Res 1990; 57:
171e82
83. Moia M, Mannucci PM, Vizzotto L, Casati S, Cattaneo M,
Ponticelli C. Improvement in the haemostatic defect of
uraemia after treatment with recombinant human
erythropoietin. Lancet 1987; 2: 1227e9
84. Vigano G, Benigni A, Mendogni D, Mingardi G, Mecca G,
Remuzzi G. Recombinant human erythropoietin to correct
uremic bleeding. Am J Kidney Dis 1991; 18: 44e9
Handling editor: J.G. Hardman