KEY POINTS
- 1.
Platelets are anuclear cellular fragments that are released from megakaryocytes and are involved in primary hemostasis.
- 2.
The normal platelet counts in newborn infants have been traditionally defined as 150 to 450 × 10 9 /L. These counts decline during the early neonatal period but then begin to rise toward the end of the first week.
- 3.
Platelet production involves the production of thrombopoietic factors such as thrombopoietin (TPO); expansion and differentiation of megakaryocyte progenitors; and production and release of platelets.
- 4.
The safe levels of blood platelet counts in neonates are still unclear. In neonates, spontaneous bleeding from thrombocytopenia does not occur when platelet counts are >100 × 10 9 /L. The risk may not be much higher even at 50 × 10 9 /L and may increase only <20 × 10 9 /L.
- 5.
Thrombocytopenia is seen frequently in neonates and can occur due to a large variety of causes. Some causes are potentially more dangerous than others. These patients need to be carefully evaluated.
- 6.
The best protocols for clinical management and platelet transfusions are a subject of ongoing debate.
Platelets are anuclear cellular fragments that are released from megakaryocytes and are involved in primary hemostasis. The megakaryocyte progenitor cells differentiate under the stimulus of thrombopoietin (TPO), and once mature, generate and release platelets into the bloodstream. Circulating platelets have a half-life of 7 to 10 days. During primary hemostasis, the activated platelets change shape and aggregate with damaged red cells and leukocytes to seal the damaged capillary walls or completely plug the leaking vessels. In secondary hemostasis, these clots get strengthened with fibrin and other products released from the coagulation cascade.
Platelet Counts in Newborn Infants
The normal platelet counts in newborns and infants have been traditionally defined as 150 to 450 × 10 9 /L. Platelet counts decline over the first few days after birth but then begin to rise toward the end of the first week ( Fig. 45.1 ). However, these definitions may need to be redefined with increasing survival of premature infants. Wiedmeier et al. studied the platelet counts in 47,000 infants delivered between 22 and 42 weeks’ gestation and showed that the mean platelet count at birth was >200 × 10 6 /μL even in the most preterm infants. However, platelet counts in the 100 to 149 × 10 9 /L range were not uncommon. The 5th percentile was 104 × 10 6 /μL for those born at <32 weeks’ gestation and 123 × 10 6 /μL for late-preterm and term neonates. The platelet counts at birth increased with advancing gestational age, by approximately 2 × 10 6 /μL for each week of gestation.
The true prevalence of thrombocytopenia in asymptomatic newborns is unknown. Thrombocytopenia in neonates (as in adults) has been traditionally defined as a platelet count <150 × 10 9 /L, and classified as mild (100–150 × 10 9 /L), moderate (50–99 × 10 9 /L), and severe (<50 × 10 9 /L). Based on these definitions, large studies in unselected populations estimated an overall incidence of neonatal thrombocytopenia of 0.7% to 0.9%. Another study of 5632 unselected newborns found platelets below 150 × 10 3 /μL in approximately 1% of all neonates ; about one-third were term infants born after uncomplicated pregnancies and deliveries with no known maternal factors or unusual physical findings. However, the incidence was much higher in infants admitted to the neonatal intensive care unit (NICU), ranging from 18% to 35%. The incidence of thrombocytopenia is inversely correlated to the gestational age so that the most immature neonates are the most frequently affected: platelet counts less than 150 × 10 9 /L were found at least once during the hospital stay in 70% of extremely low birth weight infants.
Thrombocytosis is classified as mild (platelet counts 500–700 × 10 6 /μL), moderate (700–900 × 10 6 /μL), severe (900–1000 × 10 6 /μL), and extreme (>1000 × 10 6 /μL). Primary thrombocytosis, a myeloproliferative disorder, is caused by monoclonal or polyclonal abnormalities of hematopoietic cells or by abnormalities in TPO biology. It is extremely rare in children and the frequency may be about 1 in 10 million. Secondary or reactive thrombocytosis is not uncommon in young infants. In a recent study, the 95th percentile upper reference range was 750 × 10 6 /μL over the first 90 days of life. The most common causes of reactive thrombocytosis in neonates and children are infections, tissue damage (surgeries, trauma, burns), and anemia (frequently iron deficiency). Reactive thrombocytosis has also been described in association with medications such as corticosteroids, maternal exposure to methadone or psychopharmaceutical drugs, and metabolic diseases, myopathies, or neurofibromatosis.
Fetal and Neonatal Platelet Production
Platelet production involves four steps: (1) production of thrombopoietic factors such as TPO, (2) expansion of megakaryocyte progenitors, (3) differentiation of megakaryocytes through a unique endomitotic process, and (4) production and release of platelets. TPO stimulates expansion of hematopoietic stem cells and downstream progenitor cells and promotes megakaryocyte differentiation and platelet production. It is produced mainly in the liver but also in the kidney, smooth muscle, and marrow cells. Neonates have higher plasma TPO concentrations than adults. Stem cell factor, interleukin (IL)-3, IL-11, IL-6, and erythropoietin also stimulate megakaryopoiesis and thrombopoiesis.
Megakaryocyte progenitors include the burst-forming unit–megakaryocytes (BFU-MKs) and the more mature colony-forming unit–megakaryocytes (CFU-MKs). BFU-MKs produce large multifocal colonies containing ≥50 megakaryocytes. The CFU-MKs generate smaller (3–50 cells/colony) unifocal colonies. Megakaryocytes show typical morphologic features during endoreduplication, with large cells containing polyploid nuclei. Unlike megakaryocyte progenitors, mature megakaryocytes do not generate colonies but differentiate from small mononuclear cells to large polyploid cells. Fetal/neonatal and adult megakaryocytes show important differences in morphology and biology and in the modal ploidy (the number of sets of complete chromosomes; Table 45.1 ). Fetal/neonatal megakaryocytes are smaller in size, have lower ploidy levels, produce fewer platelets, but proliferate at higher rates. These low-ploidy but mature megakaryocytes can rapidly populate the rapidly expanding bone marrow space and blood volume in the fetus/neonate to maintain normal platelet counts. However, the reserve capacity for platelet production is lower, and neonates rapidly become thrombocytopenic during stress.
| Parameter | Fetal/Neonatal Megakaryocytes | Adult Megakaryocytes |
|---|---|---|
| Size | Smaller | Larger |
| Polyploidy | Less polyploid (2–4N) | More polyploid (up to 16N) |
| Proliferation | Hyperproliferative in ex vivo culture | Less proliferative in ex vivo culture |
| Maturation | Express maturation markers | Express maturation markers |
| Proplatelet formation | Form fewer proplatelets | Form more proplatelets |
The sequence of events in platelet release from megakaryocytes is not well known. Mature megakaryocytes may migrate to a perivascular site and extend a process through the endothelium, giving rise to proplatelets, which then release platelets ( Fig. 45.2 ). An alternate mechanism may be the release of platelets in the lungs due to shear forces.
On a stained blood smear, platelets appear as dark purple spots, about 20% the diameter of red blood cells. The smear is used to examine platelets for size, shape, qualitative number, and clumping. Normal, resting platelets are disc-shaped but develop numerous long pseudopodia upon activation. These processes are important for the formation of hemostatic plugs ( Fig. 45.3 ). Platelets contain four major types of granules, namely α-granules, dense bodies, lysosomes, and the more recently described T-granules. The constituents of each are summarized in Fig. 45.4 .
Platelet Function and Primary Hemostasis
In neonates, spontaneous bleeding from thrombocytopenia does not occur when platelet counts are >100 × 10 9 /L. The risk may not be much higher even at 50 × 10 9 /L and may increase only <20 × 10 9 /L. The risk of bleeding in newborn infants may be related more to trauma sustained during the birthing process than to the platelet counts. The most feared bleeding complication is intracranial hemorrhage (ICH), due to the associated risk of adverse neurologic outcomes and mortality.
Neonatal platelets seem to be less responsive than those from adults to most agonists in terms of adhesion, aggregation, and activation. This hyporeactivity is more pronounced in preterm infants. These studies have been performed with a variety of stimulants including adenosine diphosphate (ADP), epinephrine, collagen, thrombin, and thromboxane analogs ( Fig. 45.5 ). The postulated mechanisms of this hyporeactivity include (1) fewer α 2 -adrenergic receptors, (2) impaired calcium mobilization after exposure to collagen, (3) decreased thromboxane-induced signaling, and (4) lower expression of protease-activated receptor 1.
Despite these differences in responsiveness to canonical agonists, healthy full-term neonates have similar/enhanced primary hemostasis compared with adults. Bleeding times in healthy term neonates are shorter than those in adults. Similarly, in in-vitro platelet function studies using Platelet Function Assay (PFA)–100, which measures the closure time taken to occlude a small aperture, the time is shorter in term neonates than samples from older children or adults. This enhanced platelet/vessel wall interaction in neonates may be related to their higher hematocrits, higher mean corpuscular volumes, and higher concentrations of von Willebrand factor, all of which compensate for the hyporeactivity of neonatal platelets. Platelets from preterm infants seem less reactive than those from full-term infants, leading to longer bleeding times. However, even these bleeding times were near or within the normal range for adults.
Evaluation of Neonatal Thrombocytopenia
When evaluating a thrombocytopenic neonate, the most frequent causes are listed by premature and full-term neonates by the postnatal age ( Table 45.2 ). Infection/sepsis should always be considered (regardless of the time of presentation and the infant’s appearance), because any delay in diagnosis and treatment can have life-threatening consequences.
| Premature | Full Term | |||
|---|---|---|---|---|
| Early Onset (<72 Hours) | 7–14 Days | >14 Days | Early Onset (<72 Hours) | Late Onset (>7 Days) |
|
|
|
|
|
The differential diagnosis for thrombocytopenia has been classically divided into disorders of decreased platelet production versus increased platelet consumption. However, emerging information shows that most disorders might actually be a mixture of the two mechanisms; causes that were traditionally identified to involve peripheral destruction of the platelets may actually involve the bone marrow progenitors in many cases and may also show an element of decreased megakaryopoiesis and thrombopoiesis in the bone marrow.
Any report of platelet counts below 150 × 10 3 /μL should be confirmed with a repeat test, preferably with a blood sample obtained by venipuncture and then evaluated ( Fig. 45.6 ). A thorough physical examination should be done to identify any cutaneous or oral petechiae or purpura. In ill infants, sepsis and disseminated intravascular coagulation are important causes of thrombocytopenia. Premature very low birth weight infants and those with gram-negative infections are also very likely to have low platelet counts. TORCH (toxoplasmosis, other agents, rubella, cytomegalovirus (CMV), or herpes simplex) infections may present with a whole range, mild to severe, of multisystem illness. In these infants, thrombocytopenia may be caused by bone marrow suppression and/or peripheral destruction. Signs of TORCH infections such as microcephaly, hepatosplenomegaly, or cutaneous “blueberry muffin” rash should be noted ( Fig. 45.7 ). Necrotizing enterocolitis (NEC) can also cause thrombocytopenia very early in the clinical course of the disease. Critically ill infants with multisystem organ failure from any cause can develop thrombocytopenia due to destruction of platelets in various end organs such as the lung, even if they do not show laboratory evidence of diffuse intravascular coagulation. Finally, infants on extracorporeal support for oxygenation or renal replacement therapy can develop thrombocytopenia because of platelet consumption upon contact with the foreign membranes. Thrombocytopenia could appear as the initial presenting sign of sepsis, TORCH infection, or other serious condition while the baby still appears to be otherwise well.
Early Onset of Thrombocytopenia inWell-Appearing Infants
In an otherwise healthy-appearing infant, placental insufficiency may be the most likely cause of thrombocytopenia. These infants usually develop only mild to moderate thrombocytopenia (50–150 × 10 3 /μL) that resolves spontaneously within 7 to 10 days after birth. This diagnosis should be considered in small-for-gestational-age infants with a history of intrauterine growth restriction or maternal hypertension, diabetes, or preeclampsia.
Well-appearing infants can show extremely low platelet counts due to immune-mediated neonatal alloimmune thrombocytopenia (NAIT) or autoimmune platelet destruction, in which maternal antibodies passed to the newborn in-utero lead to destruction of the baby’s platelets. Of these, NAIT produces the most pronounced thrombocytopenia, with platelets typically <50 × 10 9 /L. It occurs when the fetus inherits a paternal platelet antigen not carried by the mother; this antigen then becomes a target for maternal antibodies ( Fig. 45.8 ). Maternal platelets are not targeted and remain within normal range. NAIT affects an estimated 1 in 800 to 1000 live births. The true incidence may be higher, because milder cases might go undetected and the severe cases lead to intrauterine death.
Unlike Rh-incompatibility, NAIT frequently causes disease in a woman’s first pregnancy. The severe thrombocytopenia caused by NAIT carries a significant risk of potential morbidity and mortality. Intrauterine death or ICH may occur as early as at 14 to 16 weeks’ gestation, resulting in a relatively high incidence of intrauterine ICH (>10%). Approximately 10% to 30% of newborns with NAIT will develop ICH, with about half already having developed it in utero; neurologic sequelae and death will occur in 20% and 10% of affected neonates, respectively.
Immune thrombocytopenia occurs because of the passive transfer of antibodies from the maternal to the fetal circulation. There are two distinct types of immune-mediated thrombocytopenia: (1) NAIT and (2) autoimmune thrombocytopenia. In NAIT, the antibody is produced in the mother against a specific human platelet antigen (HPA) present in the fetus but absent in the mother. The antigen is inherited from the father of the fetus. The anti-HPA antibody produced in the maternal serum crosses the placenta and reaches the fetal circulation, leading to platelet destruction and apoptosis of megakaryocyte progenitors with decreased platelet production. The antigens responsible for NAIT are results of single-nucleotide polymorphisms in genes encoding any the major glycoproteins located on the platelet surface, particularly glycoprotein (GP)IIb/IIIa. The platelet antigens are named using an HPA nomenclature; antigens are numbered chronologically, according to the date of their initial report. The biallelic antigens were given an alphabetic designation of “a” or “b” in the order of their frequency (higher frequency for “a”). Sixteen HPA antigens have been identified so far. The frequency of each varies within ethnic groups: in the White population, antibodies to HPA-1a are the major cause of NAIT, followed by HPA-5a and, less frequently, HPA-9b, HPA-3a and HPA-3b, and HPA-15. Antibodies to HPA-4b are the predominant cause of NAIT in the Japanese population.
The diagnosis of NAIT should be considered in infants with platelet counts below 50 × 10 3 /μL. An infant suspected to have NAIT should be examined with a head ultrasound for ICH and followed up for any progression. In addition, the combination of severe neonatal thrombocytopenia with a parenchymal (rather than intraventricular) ICH is highly suggestive of NAIT.
If blood cannot be collected from the parents in a timely fashion, neonatal serum may be screened for the antiplatelet antibodies. However, low antibody concentrations in the neonate, coupled with binding of the antibodies to the infant’s platelets, can result in false-negative results. It is still unclear if there is any correlation between the affinity of the antibodies and the severity of disease.
If there is clinical suspicion for NAIT, testing for antiplatelet antibodies should be performed on the infant and/or the mother. The definitive diagnosis of NAIT involves two steps: (1) genotyping studies to identify the HPA carried by the neonate but not by the mother and (2) identification of corresponding maternal anti-HPA antibodies in the newborn by enzyme-linked immunosorbent assay. If positive, both parents as well as the neonate can be genotyped for the five most frequently identified HPA types involved with NAIT. These tests can help in reproductive planning for these families, because if the fetus carries the incompatible antigen, there is a 90% likelihood of recurrence in subsequent pregnancies. Unfortunately, >80% of neonatal thrombocytopenia believed clinically to be caused by NAIT lacks demonstrable HPA incompatibility, and the information on the patterns of HPA incompatibility in non-White populations is limited.
Thrombocytopenia secondary to NAIT resolves gradually because the causative maternal antibodies may take 8 to 12 weeks to be cleared. , These infants require close monitoring for hemorrhages during this period and also for developmental milestones for at least 18 months. Families of children affected by NAIT should be encouraged to receive reproductive counseling and future obstetric care at specialty centers. ,
Autoimmune thrombocytopenia typically causes mild to moderate thrombocytopenia, due to maternal autoantibodies that target both maternal and fetal platelets. Maternal platelet counts are expected to be low, but because the mother may carry several different types of antiplatelet immunoglobulins that may not all cross the placental barrier, the severity of her own and the neonate’s thrombocytopenia may not correlate with each other. Hence, infants born to mothers with idiopathic thrombocytopenic purpura, systemic lupus erythematosus, or other autoimmune disorders should be screened for platelet counts at birth, regardless of maternal platelet count at delivery. Platelet levels eventually normalize at 2 to 8 weeks after birth as maternal autoantibodies get cleared from the baby’s circulation.
In <1% of all instances, thrombocytopenia may occur as part of a genetic syndrome. These diagnoses are suspected more often in well-appearing term infants, although preterm or sicker infants are not at lower risk of these conditions. A localized skin lesion, discoloration, or palpated mass may represent a hemangioma of Kasabach-Merritt syndrome, which can consume platelets, and may also be informative. Fanconi anemia and thrombocytopenia-absent-radii syndrome show upper extremity abnormalities. Chromosomal abnormalities such as trisomies and Turner syndrome can be also be associated with thrombocytopenia. Other genetic disorders, such as congenital amegakaryocytic thrombocytopenia (CAMT) with proximal radio-ulnar synostosis (ATRUS), can also present with thrombocytopenia. If there is marked thrombocytopenia, a blood smear should be examined for abnormalities in platelet morphology, which may not be detected by automated platelet counting. Some genetic causes show abnormal platelet size, such as the small platelets in Wiskott-Aldrich syndrome (WAS) and X-linked thrombocytopenias. Patients with conditions such as Bernard-Soulier syndrome or Jacobsen syndrome have large platelets ( Fig. 45.9 ).
