Original research

Thrombocytopenia in
neonates with polycythemia:
incidence, risk factors and
clinical outcome
Expert Review of Hematology Downloaded from informahealthcare.com by 77.160.46.41 on 12/30/14

Expert Rev. Hematol. Early online, 1–7 (2014)

Roos D Vlug1, Background: Polycythemia occurs in 1 to 5% of neonates and is associated with complications,
Enrico Lopriore*,1, including an increased risk of thrombocytopenia. Objective: To evaluate incidence, risk factors,
Marleen Janssen2, management and outcome of thrombocytopenia in neonates with polycythemia. Study design:
All neonates with polycythemia admitted to our neonatal intensive care unit between
Johanna M
2006 and 2013 were included in this retrospective study. We evaluated the incidence of
Middeldorp2, thrombocytopenia (platelet count <150  109/l) and severe thrombocytopenia (platelet count
Mirjam EA Rath1 and <50  109/l) and the correlation between platelet counts and hematocrit values. Results:
Vivianne EHJ The incidence of thrombocytopenia and severe thrombocytopenia was 51 (71/140) and 9%
Smits-Wintjens1
For personal use only.

1
Division of Neonatology, Department
of Pediatrics, Leiden University Medical
Center, Leiden, The Netherlands
2
Division of Fetal Medicine, Department
of Obstetrics, Leiden University Medical
Center, Leiden, The Netherlands
*Author for correspondence:
Tel.: +31 071 526 2909
Fax: +31 071 524 8198
e.lopriore@lumc.nl
(13/140), respectively. Platelet count was negatively correlated with hematocrit (spearman
correlation coefficient 0.233, p = 0.007). After multiple regression analysis, we found an
independent association between thrombocytopenia and being small for gestational age (OR:
10.0; 95%; CI: 1.2–81.7; p = 0.031). Conclusion: Thrombocytopenia occurs in 51% of
neonates with polycythemia and is independently associated with growth restriction. Increased
hematocrit is associated with decreased platelet count.
KEYWORDS: growth restriction . hyperviscosity . neonates . Polycythemia . thrombocytopenia
Neonatal polycythemia, defined as a venous of platelet precursor cells due to increased
hematocrit value of >65%, is a common prob- erythropoietin production [4]. The association
lem and occurs in 1–5% of all live births [1–3]. between neonatal polycythemia and thrombo-
The etiology of neonatal polycythemia includes cytopenia has only been studied in one small
intrauterine fetal hypoxia, placental transfusion study [4]. Data on the incidence and severity of
(late cord clamping) and chromosomal disor- thrombocytopenia in neonates with polycythe-
ders. Increased hematocrit leads to increased mia are scarce, and little is known about risk
blood viscosity and may result in impaired end- factors, neonatal management and outcome.
organ perfusion [2,3]. Polycythemia and hyper- The aim of this study is to estimate the inci-
viscosity can lead to a wide range of symptoms dence of and risk factors for thrombocytopenia
and/or complications, such as respiratory prob- in neonates with polycythemia and to study the
lems, feeding problems, lethargy, convulsions, correlation between severity of thrombocytope-
hypotonia, jitteriness, hypoglycemia, hyperbilir- nia and severity of polycythemia.
ubinemia, hypocalcemia, persistent pulmonary
hypertension of the newborn, cardiomegaly, Methods
necrotizing enterocolitis and cerebral infarc- All neonates with polycythemia born between
tions [3–7]. Another complication of polycythe- January 2006 and February 2013 at the Lei-
mia is thrombocytopenia (platelet count den University Medical Center were included
<150  109/l). The exact cause of thrombocy- in this retrospective observational study. Data
topenia in neonates with polycythemia is not were extracted from a patient database, medi-
well known. Thrombocytopenia could be due cal records and laboratory files. We recorded
to an increased platelet aggregation and adhe- the presence of polycythemia during the first
sion caused by hyperviscosity [4,5] or repression week after birth. Polycythemia was defined as

informahealthcare.com 10.1586/17474086.2015.997705  2014 Informa UK Ltd ISSN 1747-4086 1

 Original research Vlug, Lopriore, Janssen, Middeldorp, Rath & Smits-Wintjens
a venous hematocrit value >70 or >65% with clinical symp-
toms of polycythemia. Symptoms of polycythemia included
the need of respiratory support (mechanical ventilation, con-
tinuous positive airway pressure and supplementary oxygen),
hypocalcemia (<8 mg/dl), hypoglycemia (<46.8 mg/dl), hyper-
bilirubinemia, feeding problems, persistent pulmonary hyper-
tension of the newborn, jitteriness, necrotizing enterocolitis
and cerebral infarction. Hyperbilirubinemia was scored
according to the American Academy of Pediatrics [8]. Necro-
tizing enterocolitis was scored based on Bell’s staging crite-
ria [9]. We recorded the presence of thrombocytopenia during
the first week after birth. Thrombocytopenia was defined as a
platelet count <150  109/l and was classified as mild (101–
Expert Review of Hematology Downloaded from informahealthcare.com by 77.160.46.41 on 12/30/14
149  109/l), moderate (51–100  109/l), severe (21–50 
109/l) and very severe (£20  109/l). We excluded all neonates
without data on platelet count measurement during the first
week of life.
A concentrated platelet transfusion (single donor plasma-
reduced platelet apheresis concentrates) at a dose of 30  109/kg
was given if platelet count was <30  109/l in clinically stable
neonates; platelet count was <50  109/l in neonates with a
manifest bleeding, those undergoing a procedure with risk of
bleeding and in clinically unstable neonates with birth weight
<1500 g. In 2010, the transfusion trigger for nonbleeding and
For personal use only.
clinically stable neonates was lowered to 20  109/kg.
Partial exchange transfusion (PET) was given if hematocrit
value was >65% with clinical symptoms of polycythemia; if
hematocrit value was >70% without clinical symptoms. We
recorded the following hematological data: hemoglobin level,
erythrocytes, leucocytes, platelet counts and venous hematocrit
values at birth and in the first week of life. Erythroblastosis was
defined as erythroblast count >95th centile for gestational
age [10]. Leukocytopenia was defined as a white blood cell
count <5  109/l.
We recorded the following obstetric data: twin–twin transfu-
sion syndrome, twin anemia polycythemia sequence [11], mater-
nal hypertension (pre-existing or developed during pregnancy)
or preeclampsia, smoking, maternal diabetes (mellitus and grav-
idarum) and chromosomal disorders of the fetus/neonate. We
recorded the following neonatal data: gestational age at birth,
birth weight, small for gestational age (SGA), mode of delivery,
single or multiple, perinatal asphyxia, respiratory distress syn-
drome, sepsis and IVH (IVH) classified according to Volpe [12].
SGA was defined as birth weight <10th centile according to
Kloosterman [13]. Perinatal asphyxia was defined as a 5-min
Apgar score <5, heart rate decelerations on cardiotocogram
and/or arterial umbilical cord pH <7.0. Sepsis was defined as a
positive blood culture with clinical signs of infection.
The primary aim of this study was to investigate the inci-
dence and severity of and risk factors for thrombocytopenia in
neonates with polycythemia. Our secondary aim was to study
the correlation between platelet counts and hematocrit value.
We hypothesized that an increased hematocrit was associated
with a decreased platelet count. To study the severity of and
risk factors for thrombocytopenia, we compared neonates with
doi: 10.1586/17474086.2015.997705
(thrombocytopenia group) and without thrombocytopenia
(control group).
Statistical analyses
Data are reported as means and standard deviations (SD) or as
medians and interquartile range, as appropriate. The unpaired
Student’s t-test was used for parametric and Mann–Whitney
U test for nonparametric comparisons for continues variables.
Results of categorical variables were compared using Fisher’s
exact test. We calculated odds ratio (OR) and 95% CI by uni-
variate logistic regression to determine factors associated with
thrombocytopenia. Of all statistically different risk factors iden-
tified with univariate analysis, a multivariate logistic regression
analysis was performed to measure independent effects. The
Spearman rank correlation was used to study the relationship
between polycythemia and thrombocytopenia. A p value
<0.05 was considered to indicate statistical significance. Statisti-
cal analysis was performed using SPSS Statistics 20.0 (SPSS
Inc., Chicago, IL, USA).
Results
During the study period, a total number of 147 term and pre-
term neonates with polycythemia were admitted to our neona-
tal nursery. A flow chart of included neonates and information
on severity of thrombocytopenia is shown in FIGURE 1. We
excluded seven neonates due to lack of information on platelet
count during the first week after birth. The median gestational
age at birth was 36 weeks (interquartile range 27–42) with a
mean (SD) birth weight of 2486 g (875). Polycythemia was
diagnosed in 93% (130/140) of neonates in the first 72 h after
birth, of which the majority was diagnosed on day 1 (58%,
81/140). The remaining 7% was diagnosed between 72 h and
144 h. In 64% (89/140) of neonates at least one PET was nec-
essary to reduce hematocrit value, whereas 36% (51/150)
recovered spontaneously.
Incidence & severity of thrombocytopenia
Thrombocytopenia was detected in 51% (71/140) of neonates
and classified as mild in 26% (36/140), moderate in 16%
(22/140), severe in 7% (10/140) and very severe in 2% (3/140).
An overview of the baseline characteristics of all included neo-
nates with polycythemia with a breakdown in the two groups,
with and without thrombocytopenia is presented in TABLE 1. In
92% (65/71) of thrombocytopenic neonates, thrombocytopenia
resolved spontaneously without platelet transfusions. The lowest
mean platelet counts during the first week in the thrombocyto-
penia group and the control group were 93  109/l and 205 
109/l (p < 0.001), respectively. The lowest platelet count in the
thrombocytopenia group was 13  109/l.
Clinical findings
IVH was found in 4% (3/71) in the thrombocytopenia group com-
pared to 1% (1/69) in the control group without thrombocytope-
nia (p = 0.324). In the thrombocytopenia group, one polycythemic
neonate with mild thrombocytopenia had a grade 1 IVH, and two
Expert Rev. Hematol.
 Thrombocytopenia in neonatal polycythemia Original research

neonates had a grade 2 IVH (one with mild Neonates with

and the other with moderate thrombocyto- polycythemia
penia). The neonate with IVH in the con- n = 147
trol group without thrombocytopenia had a
grade 2 IVH. Detailed information on the Excluded:
clinical findings in the group with thrombo- no platelet counts
cytopenia and the control group is shown available (n = 7)
in TABLE 2. There was no significant difference
between the two groups in terms of mortal- Neonates with
ity. One neonate died in the thrombocyto- thrombocytopenia
penia group. This male neonate, born at n = 71 (51%)
35 weeks of gestation, died on the 7th day
of life due to respiratory failure, severe per- Mild Moderate Severe Very severe
Expert Review of Hematology Downloaded from informahealthcare.com by 77.160.46.41 on 12/30/14

sistent pulmonary hypertension of the new- thrombocytopenia thrombocytopenia thrombocytopenia thrombocytopenia

born and neonatal sepsis. n = 36 (26%) n = 22 (16%) n = 10 (7%) n = 3 (2%)

Hematologic outcome No platelet No platelet No platelet Platelet

An overview of the hematologic outcome transfusions transfusions transfusions transfusions
is shown in TABLE 3. Erythroblastosis was n = 36 (100%) n = 21 (95%) n = 8 (80%) n = 3 (100%)
detected in 34% (24/71) in the thrombo-
cytopenia group versus 6% (4/68) in the Platelet Platelet
control group (p < 0.001). Leukocytope- transfusions transfusions
nia was detected in 13% (9/71) of neo- n = 1 (5%) n = 2 (20%)
nates in the thrombocytopenia group
For personal use only.

versus 3% (2/69) in the control group Figure 1. Flow-chart showing the derivation of the study population.
(p = 0.032). Neonates with thrombocyto-
penia had higher venous hematocrit levels when compared with 13 (9/71) versus 1% (1/69), respectively, (OR: 9.9; 95%
the control group: at birth (hematocrit 71.7 vs 69.4%, respec- CI: 1.2–214; p = 0.010). Perinatal asphyxia occurred more fre-
tively, p = 0.013), as well as during the first week after birth quently in the thrombocytopenia group than in the control
(hematocrit 73.5 vs 71.4%, respectively, p = 0.008). Of the group, 6 (4/71) versus 0% (0/69) p = 0.045). Gestational age
thrombocytopenic neonates, 75% (53/71)
required at least one PET compared with
52% (36/69) of the control group Table 1. Baseline characteristics of included neonates with
(p = 0.006). A significant negative correla- polycythemia (n = 140).
tion was found between the lowest platelet
Thrombocytopenia Control p-value
count and venous hematocrit level
group, n = 71 group, n = 69
in polycythemic neonates (r =-0.233,
p = 0.007). A scatter plot of these data is Male, n (%) 36 (51%) 34 (49%) 0.50
shown in FIGURE 2. Cesarean delivery, n (%) 33 (46.5%) 14 (20%) 0.01
Gestational age at birth, weeks† 36 (27–41) 37 (27–42) 0.35
Risk factors for thrombocytopenia
‡
Detailed information on risk factors for Birth weight (g) 2214 (796) 2766 (868) <0.01
thrombocytopenia in neonates with poly- SGA (birth weight < 10th 9 (13%) 1 (1%) 0.01
cythemia is summarized in TABLE 4. Several centile), n (%)
risk factors were associated with throm-
Apgar at 5 min £ 5, n (%) 4 (6%) 0 (0%) 0.045
bocytopenia on univariate analysis,
including birth weight, SGA and perina- Dichorionic twins, n (%) 3 (4%) 4 (6%) 0.32
tal asphyxia. The mean (SD) birth weight Monochorionic twins, n (%) 21 (30%) 13 (19%) 0.10
in the thrombocytopenia group was sig- Twin-twin transfusion 15 (21%) 9 (13%) 0.16
nificantly lower compared to the control syndrome, n (%)
group, 2214 (800) and 2766 (870) g, Twin anemia polycythemia 9 (13%) 4 (6%) 0.21
respectively (OR: 1.1; 95% CI: 1.0–1.1; sequence, n (%)
p < 0.001). The incidence of SGA was †
Value given as median (range).
significantly higher in the thrombocyto- ‡
Value given as mean (SD).
penia group than in the control group, SGA: Small for gestational age.

informahealthcare.com doi: 10.1586/17474086.2015.997705

 Original research Vlug, Lopriore, Janssen, Middeldorp, Rath & Smits-Wintjens

Table 2. Clinical findings in neonates with and without thrombocytopenia.

Thrombocytopenia Control group, OR (95% CI) p-value
group, n = 71 (%) n = 69 (%)
Respiratory support, n (%) 34/71 (48) 22/69 (39) 2.0 (0.9–4.2) 0.053
Hypoglycemia, n (%) 33/71 (47) 23/69 (33) 1.7 (0.8–3.7) 0.112
Hyperbilirubinemia, n (%) 41/71 (58) 25/69 (36) 2.4 (1.2–5.0) 0.011
Hypocalcemia, n (%) 1/71 (1.4) 0/69 (0) nc 0.322
Feeding problems, n (%) 22/71 (31) 21/69 (30) 1.0 (0.5–2.2) 0.944
PPHN, n (%) 3/71 (4) 4/69 (6) 0.7 (0.1–4.0) 0.67
Jitteriness, n (%) 5/71 (7) 4/69 (6) 1.0 (0.2–4.9) 0.764
Expert Review of Hematology Downloaded from informahealthcare.com by 77.160.46.41 on 12/30/14

Necrotizing enterocolitis, n (%) 1/71 (1) 0/69 (0) nc 0.322

Intraventricular hemorrhage, n (%) 3/71 (4) 1/69(1) 3.0 (0.3–76.8) 0.324
Cerebral infarction, n (%) 1/71 (1) 0/69 (0) nc 0.322
Mortality, n (%) 1/71 (1) 1/69 (1) 1.0 (0.0–36.4) 0.984
Nc: Not calculated, PPHN: Persistent pulmonary hypertension of the newborn.

at birth, smoking, hypertension, sepsis and trisomy 21 were the 5–10% reported in the general neonatal intensive care unit
not associated with thrombocytopenia. On multivariate analy- population. In the majority of cases, neonatal thrombocytope-
sis, only SGA was independently associated with thrombocyto- nia was self-limiting and only a minority of neonates required
For personal use only.

penia (OR: 10.0; 95% CI: 1.2–81.7; p = 0.031). As birth
weight and SGA are closely related, birth weight was subse-
quently excluded from multivariate analysis.
Discussion
This study demonstrates that thrombocytopenia is common
among neonates with polycythemia, occurring in 51% com-
pared with 1–5% in the general population and 22–35% in
the neonatal intensive care unit population [14,15]. As shown in
this study, thrombocytopenia in polycythemic neonates is
mostly mild to moderate. Severe thrombocytopenia was present
in 9% of all neonates with polycythemia, which is similar to
treatment with platelet transfusions.
Several studies have been published on the relationship
between neonatal polycythemia and thrombocytopenia. The
reported incidence of thrombocytopenia in polycythemic neo-
nates ranges from 5.5 to 35% [4,7,16–18]. The incidence of
thrombocytopenia in this study (51%) is thus higher compared
to the previous reports. This difference could partly be because
of the high percentage of monochorionic twins in our study
population (24%), which are known to be at risk for develop-
ing twin–twin transfusion syndrome and/or twin anemia poly-
cythemia sequence. Both twin–twin transfusion syndrome and
twin anemia polycythemia sequence are known risk factor for

Table 3. Hematologic results factors in neonates with and without thrombocytopenia.

Thrombocytopenia Control group, OR (95% CI) p-value
group, n = 71 n = 69
First hemoglobin value after birth, g/dl† 22.4 (2.3) 21.6 (2.1) 1.3 (1.0–1.6) 0.059
†
First venous hematocrit after birth, % 71.7 (5.0) 69.4 (5.2) 1.1 (1.0–1.2) 0.013
†
Maximum venous hematocrit value, l/l 73.5 (4.5) 71.4 (4.5) 1.1 (1.0–1.2) 0.008
Partial exchange transfusion, n (%) 53/71 (75%) 36/69 (52%) 2.7 (1.3–5.9) 0.006
‡
Erythroblastosis, n (%) 24/71 (34%) 4/68 (6%) 8.2 (2.5–29.9) <0.001
§
Leukocytopenia, n (%) 9/71 (23%) 2/69 (3%) 4.9 (0.9–34.0) 0.032
†
First measured platelet count value, 10^9/l 128 (44) 214 (32) 0.9 (0.9–1.0) <0.001
†
Lowest platelet count in the first week, 10^9/l 93 (39) 205 (36) nc <0.001
†
Data presented in mean (SD).
‡
Corrected for gestational age.
§
Leukocytopenia was defined as a white blood cell count below 5  109/l.
nc: Not calculated.

doi: 10.1586/17474086.2015.997705 Expert Rev. Hematol.

 Thrombocytopenia in neonatal polycythemia Original research

developing polycythemia and/or throm- R2 Linear =

bocytopenia [11]. Our center is the 300 0.079
national referral center for complicated
monochorionic twin pregnancies and this
250
fact makes our findings from the
‘normal’ neonatal medical experience.
Care should thus be taken when compar-

Lowest platelet count

200
ing the reported incidence in the various
studies due to methodological differences
between the studies. 150
Little is known about the relationship
between thrombocytopenia and the sever-
ity of polycythemia in the literature. Pre- 100
Expert Review of Hematology Downloaded from informahealthcare.com by 77.160.46.41 on 12/30/14

vious studies suggest that neonates with

hematocrit levels above 70% have a higher
incidence of thrombocytopenia and possi- 50
bly require more specific therapy [4,16].
However, only one small study by
Acunas et al. compared 18 polycythemic 0
neonates with thrombocytopenia with
34 polycythemic neonates without throm- 55 60 65 70 75 80 85
bocytopenia. The authors found that Highest hematocrit value
polycythemic neonates with thrombocyto-
Figure 2. Negative correlation between lowest platelet count and the
penia were more likely to have severe clin-
For personal use only.

hematocrit value during the first week after birth.

ical symptoms [4]. This seems to be in
agreement with our findings, although the
definitions of (severe) clinical symptoms varied between higher incidence of PET correlated with higher hematocrit val-
the studies. ues and/or the higher incidence of symptoms in the
In our study, we found that the group with thrombocytope- thrombocytopenia group.
nia had a significantly higher venous hematocrit. In addition, The pathophysiological explanation for the development of
we also found a negative correlation between platelet counts thrombocytopenia in polycythemic infants remains unclear.
and hematocrit values. This correlation between thrombocyto- Low platelet counts in infants with polycythemia may occur
penia and polycythemia has previously been reported in a study because of impaired production secondary to tissue hypoxia
from Ekert et al. with 28 children between 3 weeks to 15 years (because of growth restriction or perinatal asphyxia), slow
of age with cyanotic congenital heart disease, in which the spleen blood flow because of hyperviscosity and decreased
severity of thrombocytopenia was proportional to the elevation plasma fraction with normal concentrations. In addition, the
of the hematocrit [19]. However, the relationship between sever- principal determinant of platelet production is thrombopoietin
ity of thrombocytopenia and hematocrit values has not been level. The production of thrombopoietin is largely constant
reported previously in polycythemic neonates. The only previ- and modulated by platelets by thrombopoietin binding to
ous study from Acunas et al. in which platelet counts and platelets’ thrombopoietin receptor. Because polycythemic
hematocrit values were compared in polycythemic neonates did infants have an increased total blood volume (along with an
not support these findings. The authors also do not report a increased hematocrit), as expected, the same platelet mass could
significant difference in hematocrit values between polycythe- be associated with thrombocytopenia in the study group with
mic neonates with and without thrombocytopenia [4]. In addi- polycythemia and with normal platelet count in controls. In
tion, the rate of PET in thrombocytopenic neonates was higher both these instances, the thrombopoietin levels will be identical
compared to the control group, respectively, 75 versus 52%. and platelet production the same. This basic physiological
This is probably because of the higher hematocrit levels we premise could then account for some of the reported differen-
found in the thrombocytopenia group. However, because PET ces in our study.
is controversial in neonates without symptoms of polycythemia Logistic regression analysis revealed that growth restriction is
another explanation could be the higher incidence of symptoms the only independent causative factor associated with thrombo-
of polycythemia in the thrombocytopenia group. The study cytopenia in polycythemic neonates. Several hypotheses have
done by Acunas et al. supports this finding although they did been postulated about the relationship between SGA and throm-
not find a difference in hematocrit value. For this reason, the bocytopenia focusing mostly on the link between intrauterine
authors suggest that thrombocytopenia could be a possible indi- growth restriction, placental disorders and chronic fetal hypoxia.
cator for PET instead of the classical criteria. In our study, the Chronic hypoxia may lead to increased erythropoietin

informahealthcare.com doi: 10.1586/17474086.2015.997705

 Original research Vlug, Lopriore, Janssen, Middeldorp, Rath & Smits-Wintjens

Table 4. Characteristics and risk factors in neonates with and without thrombocytopenia in the first week
after birth.
Thrombocytopenia Control Univariate p-value Multivariate p-value
group n = 71 group analysis analysis
n = 69 OR (95% CI) OR (95% CI)
Gestational age at birth, 36 (27–41) 37 (27–42) 1.0 (0.9–1.1) 0.355
weeks †
‡
Birth weight, (100 g) 22.14 (8) 27.66 (8.7) 1.1 (1.0–1.1) <0.001
SGA (birth weight < 10th 9/71 (13%) 1/69 (1%) 9.9 (1.2–214.0) 0.010 10.0(1.2–81.7) 0.031
centile), n (%)
Maternal hypertension/ 25/70 (36%) 19/69 (28%) 1.7 (0.8–3.8) 0.300
preeclampsia, n (%)
Expert Review of Hematology Downloaded from informahealthcare.com by 77.160.46.41 on 12/30/14

Smoking, n (%) 6/51(12%) 8/64 (13%) 0.9 (0.3–3.3) 0.905

Diabetes mellitus/ 1/66 (1.5%) 9/69 (13%) 0.1 (0.0–0.8) 0.011 0.1(0.02–1.0) 0.051
gravidarum
Trisomy 21, n (%) 3/71 (4%) 2/69 (3%) 1.5 (0.2–13.1) 0.672
Asphyxia, n (%) 4/71 (6%) 0/69 (0%) nc 0.045 nc nc
Sepsis, n (%) 3/71 (4%) 2/68 (3%) 1.5(0.2–12.9) 0.684
†
Value given as median (range).
‡
Value given as mean (SD).
nc: Not calculated.
For personal use only.

production and erythroblastosis, which may in turn impair meg-
akaryocytopoiesis and result in thrombocytopenia [20,21].
Acunas et al. compared potential causative factors for thrombo-
cytopenia in polycythemic neonates but found no difference in
birth weight or the presence of SGA. In contrast with our find-
ings, perinatal asphyxia in the study from Acunas et al. was iden-
tified as potential causative factor for thrombocytopenia [4].
Discordances in results may be because of the small number of
neonates with asphyxia included in our study (n = 4, 3%) com-
pared with a larger number of asphyxiated neonates (n = 21,
40%) reported in the study from Acunas.
To our knowledge, this is the first large study assessing the
correlation between polycythemia and thrombocytopenia.
Because of the retrospective design, the results of this study
should be interpreted with care. Prospective studies should be
conducted to confirm the correlation between the severity of
thrombocytopenia and the severity of polycythemia and to
study the risk factors and clinical outcomes of thrombocytope-
nia in polycythemic neonates.
In conclusion, the present study shows that neonates with poly-
cythemia are at increased risk (51%) of thrombocytopenia, and
the severity of thrombocytopenia is negatively correlated with the
severity of polycythemia. Although polycythemic neonates with
thrombocytopenia are more at risk to develop clinical symptoms,
thrombocytopenia was mostly mild and self-limiting and the risk
of IVH was not increased.
Financial & competing interests disclosure
The authors have no relevant affiliations or financial involvement with
any organization or entity with a financial interest in or financial conflict
with the subject matter or materials discussed in the manuscript. This
includes employment, consultancies, honoraria, stock ownership or options,
expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.

Key issues
. Neonatal polycythemia, defined as a venous hematocrit value of >65%, is a common problem and occurs in 1–5% of all live births.
. Neonatal polycythemia is associated with clinical and hematological complications, including an increased risk for thrombocytopenia.
. In this study with 140 polycythemic neonates, the incidence of thrombocytopenia (platelet count <150  109/l) and severe
thrombocytopenia (platelet count <50  109/l) was 51% (71/140) and 9% (13/140).
. Thrombocytopenia in polycythemic neonates is mostly mild and self-limiting and is not associated with an increased risk of
intraventricular hemorrhage.
. Platelet count was negatively correlated with hematocrit (spearman correlation coefficient -0.233, p = 0.007): that is, higher hematocrit
is correlated with a lower platelet count.
. Logistic regression analysis shows that growth restriction is the only independent causative factor associated with thrombocytopenia in
polycythemic neonates.

doi: 10.1586/17474086.2015.997705 Expert Rev. Hematol.

 Thrombocytopenia in neonatal polycythemia Original research

References
Papers of special note have been highlighted as:
. of interest
.. of considerable interest
1. Sankar MJ, Agarwal R, Deorari A, Paul VK.
Management of polycythemia in neonates.
Indian J Pediatr 2010;77:1117-21
2. Dempsey EM, Barrington K. Short and
long term outcomes following partial
exchange transfusion in the polycythaemic
newborn: a systematic review. Arch Dis
Child Fetal Neonatal Ed 2006;91:F2-6
3. Pappas A, Delaney-Black V. Differential
Expert Review of Hematology Downloaded from informahealthcare.com by 77.160.46.41 on 12/30/14
7. Wiswell TE, Cornish JD, Northam RS.
Neonatal polycythemia: frequency of clinical
manifestations and other associated findings.
Pediatrics 1986;78:26-30
8. American Academy of Pediatrics
Subcommittee on Hyperbilirubinemia.
Management of hyperbilirubinemia in the
newborn infant 35 or more weeks of
gestation. Pediatrics 2004;114:297-316
9. Bell MJ, Ternberg JL, Feigin RD, et al.
Neonatal necrotizing enterocolitis.
Therapeutic decisions based upon clinical
staging. Ann Surg 1978;187:1-7
10. Christensen RD, Henry E, Andres RL,
14. Roberts IA, Murray NA. Thrombocytopenia
in the newborn. Curr Opin Pediatr
2003;15:17-23
.. Review that summarizes the etiology,
incidence, management and outcome in
neonatal thrombocytopenia
15. Chakravorty S, Murray N, Roberts I.
Neonatal thrombocytopenia. Early Hum
Dev 2005;81:35-41
16. Katz J, Rodriguez E, Mandani G,
Branson HE. Normal coagulation findings,
thrombocytopenia, and peripheral
hemoconcentration in neonatal
polycythemia. J Pediatr 1982;101:99-102

diagnosis and management of polycythemia.

Pediatr Clin North Am 2004;51:1063-86
4. Acunas B, Celtik C, Vatansever U,
Karasalihoglu S. Thrombocytopenia:
11.
an important indicator for the application
of partial exchange transfusion in
polycythemic newborn infants? Pediatr Int
2000;42:343-7
. Study in twin anemia-polycythemia
5. Mimouni FB, Merlob P, Dollberg S,
Mandel D. Neonatal polycythaemia: critical
review and a consensus statement of the
Israeli Neonatology Association. Acta
For personal use only.
Bennett ST. Reference ranges for blood
17.
concentrations of nucleated red blood cells
in neonates. Neonatology 2011;99:289-94
Lopriore E, Slaghekke F, Oepkes D, et al.
Hematological characteristics in neonates
18.
with twin anemia-polycythemia sequence
(TAPS). Prenat Diagn 2010;30:251-5
19.
sequence showing the association between
polycythemia and thrombocytopenia in
recipient twins.
Morag I, Strauss T, Lubin D, et al.
Restrictive management of neonatal
polycythemia. Am J Perinatol 2011;28:
677-82
Krishnan L, Rahim A. Neonatal
polycythemia. Indian J Pediatr 1997;64:
541-6
Ekert H, Gilchrist GS, Stanton R,
Hammond D. Hemostasis in cyanotic
congenital heart disease. J Pediatr 1970;76:
221-30

Paediatr 2011;100:1290-6
6. Sarkar S, Rosenkrantz TS. Neonatal
polycythemia and hyperviscosity. Semin
Fetal Neonatal Med 2008;13:248-55
.. Review that summarizes the etiology,
incidence, management and outcome in
neonatal polycythemia
12. Volpe JJ. Intracranial hemorrhage: germinal
matrix-intraventricular hemorrhage of the
premature infant. in Volpe JJ, editor.
Neurology of the newborn. 4th edition.
Saunders; Philadelphia: 2001. pp. 428-93
13. Kloosterman GJ. [Intrauterine growth and
intrauterine growth curves]. Maandschr
Kindergeneeskd 1969;37:209-25
20. Wasiluk A, Mantur M, Kemona H, et al.
Thrombopoiesis in small for gestational age
newborns. Platelets 2009;20:520-4
21. Roberts I, Murray NA. Neonatal
thrombocytopenia. Semin Fetal Neonatal
Med 2008;13:256-64

informahealthcare.com doi: 10.1586/17474086.2015.997705