Keywords
anemia - thrombocytopenia - erythrocytes - platelets - transfusion - neonates
Anemia and thrombocytopenia occur frequently in neonates admitted to neonatal intensive
care units (NICUs), particularly in (very) preterm neonates. Management of these disorders
has traditionally been based on the administration of blood products, either red blood
cells (RBCs) or platelets to correct anemia and thrombocytopenia. Almost all very
preterm neonates admitted to a NICU require at least one RBC transfusion during admission
due to anemia,[1] whereas approximately 10% of very preterm neonates receive platelet transfusions
during their NICU stay (unpublished data from our center).
Although blood transfusions in the developed world are considered to be safe, long-term
effects of adult donor blood products on the health and development of preterm neonates
are not well known. In fact, the actual short-term benefits of transfusions in preterm
neonates are not well known either. To date, transfusion guidelines are not based
on solid scientific data but largely on expert opinion and dogmas. This lack of evidence
and scientific data resulted in a lack of international consensus on transfusion indications
and optimal thresholds. Consequently, transfusion practices between and within individual
countries vary greatly. Nevertheless, an increasing number of studies are gradually
shedding more light on this topic and report important new information on the safety
and effectiveness of transfusions in neonates.
This review aims to provide an overview of the available evidence and current transfusion
practices in neonates with anemia and thrombocytopenia and shortly discuss future
developments.
Neonatal Anemia and Red Blood Cell Transfusions
Neonatal Anemia and Red Blood Cell Transfusions
Anemia in neonates can be physiological, but can also be caused by nonphysiological
factors such perinatal or peripartum complications, clinical conditions such as sepsis
and cardiorespiratory disease, and hemolytic disease of the fetus and newborn. However,
one of the most important causes of anemia in very preterm neonates is iatrogenic
blood loss due to frequent laboratory testing. Estimates of iatrogenic blood loss
due to laboratory testing in the first month of life in preterm neonates admitted
to a NICU may amount up to one-third of the total blood volume.
The main treatment modality for neonatal anemia is still based on the administration
of RBC transfusions, but there is no international consensus regarding optimal hemoglobin
thresholds for RBC transfusions in preterm neonates. Postulated hemoglobin thresholds
for transfusions vary greatly between and within countries. In addition, the decision
to transfuse is sometimes made based on clinical judgment of the caregiver, irrespective
of national or local guidelines. Various national guidelines have been published in
the past decade and are summarized in [Table 1].
Table 1
Overview of international guidelines and thresholds for RBC transfusions and platelet
transfusions
|
British Committee for Standards in Haematology (2016)
|
Australian National Blood Authority (2016)
|
Canadian Blood Services (2017)
|
Dutch Guidelines Quality council (2019)
|
|
Neonatal anemia
|
|
Postnatal week
|
Respiratory support
|
No respiratory support
|
Respiratory support
|
No respiratory support
|
Respiratory support
|
No respiratory support
|
Respiratory support
|
No respiratory support
|
|
Week 1
|
10–12 g/dL
|
10 g/dL
|
11–13 g/dL
|
10–12 g/dL
|
11.5 g/dL
|
10 g/dL
|
11.5 g/dL
|
10 g/dL
|
|
Week 2
|
9.5–10 g/dL
|
7.5 g/dL
|
10–12.5 g/dL
|
8.5–11 g/dL
|
10 g/dL
|
8.5 g/dL
|
10 g/dL
|
8.5 g/dL
|
|
Week ≥ 3
|
8.5–10 g/dL
|
7.5 g/dL
|
8.5–11 g/dL
|
7–10 g/dL
|
8.5 g/dL
|
7.5 g/dL
|
8.5 g/dL
|
7.5 g/dL
|
|
Neonatal thrombocytopenia
|
|
Prophylactic in stable infant
|
25 × 109/L
|
10–20 × 109/L
|
20 × 109/L
|
25 × 109/L
|
|
Bleeding or invasive procedure
|
50 × 109/L
|
50 × 109/L
|
50 × 109/L
|
50 × 109/L
|
To date, only a few randomized controlled trials (RCTs) compared different transfusion
hemoglobin thresholds in very preterm neonates in regard to safety and effectiveness.
As shown in the latest Cochrane review,[2] no significant short-term nor long-term differences were found between liberal and
restrictive RBC transfusion groups. However, these trials may have underpowered to
detect a difference in the effect of transfusions on long-term neurodevelopmental
outcome.
Two larger ongoing trials in this field may shed further light in the near future
on the safety and effectiveness of liberal versus restrictive thresholds for RBC transfusions.
The TOP trial (Transfusion of Prematures, NCT01702805) in the United States has included
more than 1,800 preterm infants. A similar study, the ETTNO trial (Effects of Transfusion
Thresholds on Neurocognitive Outcome of Extremely Low Birth Weight Infants, NCT01393496),
was performed in Germany and included more than 900 preterm neonates. In both studies,
the primary outcome is death or significant neurodevelopment impairment in surviving
infants determined at 2 years of age. The results of these two studies, including
the long-term psychomotor assessment, are expected in the course of the coming year
and will certainly increase our current knowledge and help improve international transfusion
guidelines for RBC in neonates with anemia.
Although transfusions of RBC products are generally considered to be safe, short-term
or long-term complications might be under recognized in the neonatal population, especially
in (very) preterm neonates. Various studies in the adult population have identified
several potential complications associated with transfusions, including transfusion-related
lung injury and transfusion-associated circulatory overload. Whether these potential
transfusion risks can also occur in the neonatal population remains unclear. Recently,
several studies have suggested associations between RBC transfusions and intraventricular
hemorrhage, necrotizing enterocolitis (the so-called TANEC, “transfusion-associated
necrotizing enterocolitis”), retinopathy of prematurity, and iron overload.[3]
[4]
[5] The pathophysiological explanations for these associations and the evidence of a
possible causal relationship between transfusion and neonatal morbidities remain speculative.
Neonatologists should be aware of the current lack of evidence for optimal transfusion
guidelines and the increasing reports on potential side effects. Until new evidence
is gained, RBC transfusions in neonates with anemia should be used with caution. And,
since prevention is often better than cure, preventing neonatal anemia should be a
major aim.
Fortunately, prevention of neonatal anemia in preterm neonates can partly be achieved
by implementing inexpensive measures such as (1) delayed cord clamping and (2) minimization
of iatrogenic blood loss due to frequent laboratory testing. Both interventions may
result in significant reduction of the need for transfusions in preterm neonates.
Delaying cord clamping by at least 30 to 60 seconds after birth allows for a prolonged
placental transfusion and is now recommended in preterm and full-term neonates, as
stated in a guideline by the World Health Organization.[6] In preterm neonates, it is associated with higher hemoglobin levels after birth,
fewer RBC transfusions, better circulatory stability, and possibly a lower risk for
intraventricular hemorrhage and necrotizing enterocolitis compared with early cord
clamping.[7] Reducing iatrogenic blood loss can be achieved first by simply avoiding or minimizing
often unnecessary laboratory testing. This can also be achieved by using microtechnique
laboratory procedures which allow for smaller sampling volumes, the development of
noninvasive monitoring methods, and consequent use of fetal blood from the placenta
for baseline laboratory blood tests.[8]
[9]
[10] Other supportive measures such as the administration of erythropoietin can also
reduce the risk of anemia of prematurity and further investigations in the neonatal
population are needed.
Neonatal Thrombocytopenia and Platelet Transfusions
Neonatal Thrombocytopenia and Platelet Transfusions
Thrombocytopenia in neonates is defined as platelet count <150 × 109/L and classified as early or late onset. Early-onset thrombocytopenia, which can
be detected at birth or within 3 days after birth, is often caused by prenatal factors
such as severe intrauterine growth restriction, maternal disease (preeclampsia), perinatal
asphyxia, or fetal/neonatal alloimmune thrombocytopenia. Late-onset thrombocytopenia
occurs after the first 3 days after birth and is then most commonly caused by bacterial
sepsis, necrotizing enterocolitis, or thrombotic events associated with the use of
central lines.
In analogy to neonatal anemia and RBC transfusions, the mainstay of treatment in neonatal
thrombocytopenia is based on the administration of platelet transfusions. Similarly,
the management of thrombocytopenia in neonates is mainly based on dogmas and beliefs
rather than on solid scientific data. As a result, there is no international consensus
on the optimal indications and thresholds for platelet transfusions in neonates, leading
to wide worldwide variation in transfusion protocols. Several national guidelines
on platelet transfusion thresholds are summarized in [Table 1].
Platelet transfusions in preterm neonates are usually given “prophylactically,” i.e.,
indicated by a predefined platelet threshold, and only in a minority of cases “therapeutically”
in case of an active major bleed. The aim of prophylactic platelet transfusions in
preterm neonates is to prevent major bleedings such as intraventricular hemorrhages,
lung bleeding, or gastrointestinal bleeding. However, the benefit of these prophylactic
transfusions in nonbleeding neonates is unclear and controversial.[11] There is no clear correlation between platelet cut-off values and individual bleeding
risk.
A prospective observational study (Planet-1 study) showed that 81% of platelet transfusions
in neonates are given prophylactically.[12] Of the 194 neonates in this study with platelet counts <60 × 109/L, 73% were reported
to have minor hemorrhages, but only 9% developed severe hemorrhage. Of the 58 neonates
with the most severe thrombocytopenia, with a platelet count <20 × 109/L, only 9% developed major hemorrhage. Until just recently, only one randomized trial
was performed in preterm thrombocytopenic neonates comparing higher (<150 × 109/L) versus a lower (<50 × 109/L) platelet count threshold for prophylactic transfusion. The primary outcome in
this RCT by Andrew et al was the incidence of intracranial hemorrhage, which did not
significantly differ between the two treatment groups (26 vs. 28%, p = 0.73).[13] In a recently published international randomized trial (Planet-2 study), 660 preterm
neonates with severe thrombocytopenia were randomized to receive platelet transfusions
at platelet count thresholds of 50 × 109/L (liberal transfusion group) or 25 × 109/L
(restrictive transfusion group).[14] Mortality rate of major bleeding within 28 days of randomization was significantly
higher in the <50 × 109/L group compared with the <25 × 109/L group, respectively 26% (85/324) versus 19% (61/329) (odds ratio [OR]: 1.57; 95%
confidence interval [CI]: 1.06–2.32; p = 0.02). This study suggests that platelet transfusions in preterm neonates may cause
more harm than previously thought. Restrictive platelet transfusion guidelines may
thus be preferable compared with liberal transfusion guidelines. Surprisingly, the
incidence of bronchopulmonary dysplasia (BPD), a secondary outcome in the Planet-2
study, was also significantly higher in the liberal transfusion group than in the
restrictive group. The incidence of survival with BPD was 63% (169/269) in the liberal
transfusion group versus 54% (153/281) in the restrictive group (OR: 1.54; 95% CI:
1.03–2.30). This important finding suggests that platelet products may have proinflammatory
consequences and lead to organ damage. Although this trial clearly shows that guidelines
should now be adapted toward a lower transfusion threshold (platelet counts < 25 × 109/L), more research is needed to determine whether an even lower threshold should be
used to guide platelet transfusion. Ideally, the indication for prophylactic platelet
transfusions in preterm neonates should be based on an individual assessment of bleeding
risk (incorporating various relevant clinical risk factors) rather than on an absolute
platelet count level only. A reliable algorithm with a good predictive score quantifying
the absolute risk of bleeding is already available for adult patients, and it could,
in the future, prove to be useful in one of the most vulnerable group of patients
with a high risk of bleeding: very preterm neonates.
Conclusion
To optimize transfusion guidelines, the results of ongoing RCTs need to be carefully
evaluated and further research is needed in specific neonatal subpopulations to ascertain
appropriate thresholds. The greatest advances in transfusion practice are preventive
and supportive measures. Further reduction of iatrogenic phlebotomy, global implementation
of delayed cord clamping, and timely recognition and treatment of underlying conditions
such as infection may prevent a significant number of RBC transfusions. Until then,
blood product transfusions in neonates should be used with caution and neonatologists
should be aware that the evidence concerning the risks and benefits of transfusions
in neonates is fairly limited. Further investigations and large well-designed trials
in preterm neonates are urgently needed.
