Red blood cells (RBC) are nonnucleated biconcave discs whose major cellular component is hemoglobin (Hb), an oxygen transport protein. Erythropoietin, a hematopoietic growth factor produced in the kidney, is the major regulator of red cell production. A normal red cell life span is approximately 120 days. Normal developmental factors influence Hb values in children including age, gender, and sexual maturity necessitating the need for age-appropriate reference values.

Hb production begins early in gestation and undergoes a series of transitions. Hb is formed by two pairs of identical subunits called globin chains. Fetal erythropoiesis consists of an orderly evolution through a series of different Hbs. All forms of Hb are made up of a combination of two α-like globin proteins (α or ξ) and two β-like globin proteins (β, δ, γ, or ε). In the embryo, the predominant Hbs include Gower 1 (ξ₂ε₂), Gower 2 (α₂ε₂), and Portland (ξ₂γ₂). In the fetus, there is a transition to fetal Hb (α₂γ₂). After birth, there is a final switch to adult Hb (α₂β₂).

At birth, neonates have a mean Hb of 15.9 g/dL (±1.86) and an elevated mean corpuscular volume (MCV) of 110 fl (±5) (1). After birth, red cell production quickly decreases likely secondary to the abrupt increase in oxygen concentration. The half-life of neonatal red cells is shorter (average 23.3 days) than in adults and is even shorter in premature infants (average 16.6 days) (2). The Hb naturally decreases over the first 2 to 3 months of life (physiologic nadir) and then slowly increases in the fourth to sixth months of life. Throughout childhood, there is a mild increase in the mean Hb. In males during puberty, as the Tanner stage increases, the Hb increases. There is no relationship between Hb and Tanner stage in females.

White blood cells (WBC) play an integral role in the immune system. There are five types of WBC including neutrophils, eosinophils, basophils, lymphocytes, and macrophages/monocytes. Individual WBC percentages have limited clinical utility; instead, absolute counts should always be considered. The WBC reference range will vary with age. In general, newborns have a higher total WBC (mean 18.1 k/μL) that will quickly decrease over the first week of life. Lymphocyte predominance is seen from 2 weeks to approximately 5 years of age, and then neutrophils become predominant.

Platelets are small anucleated cell particles that are made in the bone marrow via fragmentation of megakaryocytes; production is mediated via thrombopoietin. Platelets circulate for approximately 7 to 10 days and are subsequently removed via the reticuloendothelial system. By 18 weeks of gestation, the plasma platelet concentration reaches the adult range of 150 to 450 k/μL. At birth, neonates have the same platelet count range and volume as adults. Data regarding platelet function in neonates demonstrate hyporeactivity to some agonists and hyperreactivity to others (1). Platelets play an integral role in hemostasis.

Hemostasis refers to the coordinated process that stops bleeding at the site of vascular injury through the formation of an impermeable platelet and fibrin plug. Hemostasis is achieved through the following three main mechanisms:

Vascular constriction decreases blood flow at the site of injury. Platelets adhere to the exposed subendothelium, through von Willebrand factor (vWF) tethering, forming an initial platelet plug. Simultaneously, coagulation is initiated by the exposure of tissue factor. Tissue factor binds and activates FVII that activates the coagulation cascade, resulting in a small thrombin burst. This small thrombin burst stimulates further platelet activation and the activation of coagulation on the platelet surface. On this increased platelet surface, a large amount of thrombin is formed that is sufficient to convert fibrinogen to fibrin
leading to stable clot formation. To limit clot formation at the site of injury, activated procoagulant proteins are inhibited by anticoagulant proteins (protein C, S, antithrombin, thrombomodulin, heparin cofactor II, and tissue factor pathway inhibitor). Clot degradation is initiated by the fibrinolytic system (plasminogen, tissue plasminogen activator [tPA], and urokinase plasminogen activator). This is a finely balanced system, and a derangement at any level can result in a tendency for bleeding or a prothrombotic state.

In the neonate, these hemostatic processes are in place but in different concentrations than adults. In normal postnatal development, many values normalize by 6 months of age, although changes can still be seen throughout childhood (3,4). Understanding the difference in neonatal values is imperative when interpreting coagulation studies to ensure the correct diagnosis of either a bleeding or clotting disorder. It also has direct implications for the use of specific hemostatic interventions in a neonate (i.e., unfractionated or low–molecular-weight heparin [LMWH] therapy).

Coagulation proteins do not cross the placenta and are independently synthesized by the fetus; most are present by 10 weeks of gestation and gradually increase with gestational age (5,6). In a neonate, the following procoagulant proteins are decreased including the vitamin K–dependent factors (II, VII, IX, and X) and the contact pathway factors (XI, XII, prekallikrein, and high–molecular-weight kininogen) (4,6,7). This results in a prolonged prothrombin time (PT) and partial thromboplastin time (PTT) when compared to normal adult values. Conversely, neonates have increased plasma vWF levels and elevated levels of circulating ultra-large von Willebrand multimers (8,9). Similar to the procoagulant factors, the inhibitors of coagulation are also decreased. The anticoagulant proteins including protein C, S, antithrombin, heparin cofactor II, and tissue factor pathway inhibitor are decreased, resulting in a slower rate of thrombin inhibition (4,6,7,10). The fibrinolytic system is also depressed secondary to a unique neonatal glycoform of plasminogen that is inefficiently converted to plasmin (11). Neonates will have markedly elevated D-dimer values at birth lasting up to 3 days (7,10).

Adolescents and children with congenital and acquired heart disease are at increased risk for hematologic abnormalities including red cell anomalies, bleeding, and thrombosis. The following sections discuss individual hematologic disorders describing the effects on the normal heart and in addition paying attention to particular concerns regarding the child and adolescent with congenital and acquired heart disease.

Two genetic disorders, Barth syndrome and 22q11.2 deletion syndrome, may have both heart disease and associated white cell disorders. Barth syndrome is an X-linked recessive mitochondrial disorder that is characterized by a dilated cardiomyopathy (or sometimes hypertrophic) with endocardial fibroelastosis, skeletal myopathy, growth retardation, neutropenia, and organic aciduria (59,60). It is secondary to mutations in the TAZ gene that is responsible for remodeling the mitochondrial phospholipid: cardiolipin (61). The etiology of neutropenia in Barth syndrome is unclear at this time. Treatment can include granulocyte colony-stimulating factor to prevent severe neutropenia.

Chromosome 22q11.2 deletion syndrome encompasses a heterogeneous group of disorders. The most common phenotypic features include conotruncal cardiac malformations, immunologic dysfunction, developmental delay, and palate deformities (62). Immunologic dysfunction is the result of thymic aplasia or hypoplasia resulting in a variable T-cell deficiency with a resultant increase in infections and autoimmune disease (63). Decreased regulatory T cells may play a role in the increased incidence of autoimmune disorders (63,64). Autoimmune disease can include acquired cytopenias (65,66).

Hemostatic derangements complicated with bleeding are very common in pediatric patients with heart disease. The holy grail of hemostatic testing would be one test that is rapidly available and adequately reports on all components of hemostasis. Unfortunately, this test does not exist, leaving clinicians with a series of coagulation tests that reflect various aspects of hemostasis.

Hemostatic testing: The PT and PTT serve as screening tests for procoagulant factor deficiencies. The PT and PTT do not provide a global picture of hemostasis. Proper specimen handling is imperative for coagulation assays. Heparin contamination of a specimen is the most common reason for a prolonged PTT in a hospitalized patient. Heparinase can be used to remove heparin from a specimen. Coagulation assays are also sensitive to the plasma-to-citrate ratio in the specimen tube. Underfilling of the specimen tube will falsely prolong the coagulation assays. Patients with significant polycythemia (hematocrit >55%) can also have an altered plasma-to-citrate ratio and will have falsely prolonged coagulation assays. A special specimen tube with a decreased amount of citrate should be used for patients with significant polycythemia. This is of particular concern when using the prothrombin time/international normalized ratio (PT/INR) to adjust warfarin dosing in cyanotic/polycythemic patients. Factor deficiencies that can cause an isolated prolonged PTT include either a deficiency in factor VIII, IX, XI, XII, prekallikrein, or high–molecular-weight kininogen. An isolated prolonged PT is secondary to factor VII deficiency. Prolongation in both the PT and PTT can be seen in the setting of factor II, V, X, or fibrinogen deficiency.

For monitoring unfractionated heparin (uFH) therapy, the PTT is commonly used as a surrogate marker of heparin activity. For therapeutic heparin, the PTT is titrated to 1.5 to 3 times control. There are limitations to the PTT because it is not a direct measure of heparin and it is affected by variables other than heparin including an elevated factor VIII, fibrinogen, or a Lupus anticoagulant. For example, when the FVIII level is >250%, the PTT is significantly shortened and will no longer reflect the heparin effect. (Factor VIII is an acute phase reactant and can be elevated in inflammatory states.) For a direct measure of heparin activity against factor X, the uFH anti-Xa level can be used with a recommended therapeutic range of 0.35 to 0.7 U/mL (note this is different from the LMWH anti–Xa-recommended therapeutic range of 0.5 to 1 U/mL).

The activated clotting time (ACT) is used to monitor higher heparin doses given to patients undergoing cardiac surgery with cardiopulmonary bypass (CPB), cardiac catheterization, or extracorporeal membrane oxygenation (ECMO). The ACT is a whole blood clotting time that is simple to perform with a rapid turnaround time. It is more sensitive to a wider range of heparin doses than the PTT. The limitations to this assay are that it is imprecise, proper specimen handing and timing are imperative, it is affected by the same variables that can prolong the PTT assay, and it is affected by the platelet count.

Fibrinogen is the final procoagulant protein in the coagulation pathway. It is converted by thrombin into fibrin. Fibrin is then cross linked by factor XIII to make a stable clot. In the setting of a bleeding patient, it is imperative that the fibrinogen level is maximized to ensure adequate hemostasis. The thrombin time measures the conversion of fibrinogen to fibrin and is affected by quantitative or qualitative abnormalities of fibrinogen, the presence of thrombin inhibitors, and fibrinogen degradation products. Heparin contamination of the specimen can prolong the thrombin time.

For an assessment of primary hemostasis, a platelet count indicates the number of platelets that are available but does not provide any information regarding platelet function. At this time, there is limited availability to rapidly assess platelet function. The bleeding time assesses platelet and capillary function but this technique is patient- and operator-dependent and has fallen out of favor (67). It is insensitive to mild platelet defects and does not consistently predict a bleeding tendency (68). The platelet function analyzer (PFA-100) is an in vitro screen for platelet function (68). The PFA-100 is prolonged in the setting of anemia or thrombocytopenia and is insensitive to mild platelet aggregation defects; it has also fallen out of favor as a screening test.

Thromboelastography (TEG) is a whole blood point-of-care assay that provides a real-time graph of clot formation and stability. Values measured include R (time necessary for initial clot formation), K (clot kinetics), α (rate of clot formation), maximum amplitude (maximum strength of the clot), and A-60 to maximum amplitude ratio (fibrinolysis). TEG has been utilized in pediatric CPB and in patients with LV assist devices (69,70,71).

Platelet mapping assay is a modification of the standard TEG and is used clinically to assess the effectiveness of antiplatelet medications. This assay utilizes whole blood that is heparinized to suppress thrombin generation, and then known platelet agonists are added to the sample to assess platelet activation. This assay is currently utilized most commonly in pediatric patients with an LV assist device to monitor antiplatelet therapy.

Postoperative bleeding is a common complication in pediatric patients undergoing CPB and is associated with increased morbidity and mortality (94,95). CPB is a nonphysiologic state and the passage of blood through the artificial circuit results in the activation of the coagulation, fibrinolytic, and inflammatory systems. The coagulation system is activated through the contact pathway and ultimately results in the formation of fibrin. With fibrin deposition within the CPB circuit, there is platelet adherence and activation. Prematurely activated platelets are no longer available for hemostasis.

Post-CPB bleeding is more common in pediatric patients than in adults. Clinical factors associated with postoperative bleeding include younger age (<1 year), lower weight (<8 kg), cyanosis, and prolonged circuit and deep hypothermia times (96,97,98). One major contributing factor is the degree of hemodilution that occurs in pediatric patients with the institution of cardiac bypass. This is secondary to the large discrepancy between a pediatric patient’s blood volume and the circuit volume. An adult patient experiences approximately 25% hemodilution with the institution of bypass, while a pediatric patient can experience up to 60% of hemodilution (99). This dilution is further exacerbated by developmental hemostasis in neonates who, at baseline, already have lower procoagulant and anticoagulant proteins (4).

To prevent postoperative CPB bleeding, multiple approaches have been utilized including adequate anticoagulation during CPB, modified ultrafiltration, adequate reversal of anticoagulation post-CPB, transfusion of blood components, and antifibrinolytics (99). High-dose heparin is used during CPB to prevent thrombosis of the circuit. Heparin exerts its anticoagulant effect through the binding of antithrombin. Neonates developmentally have lower antithrombin levels that can contribute to a variable heparin effect. There are also significant issues with monitoring heparin therapy during CPB in pediatric patients. It has been shown in children that the ACT values do not correlate with heparin levels. Specifically, the ACT overestimates the heparin effect because it is prolonged by other variables including hemodilution and hypothermia (99,100,101). Adequate reversal of heparin with protamine is dependent on knowing the correct heparin concentration. Too little protamine means heparin is still circulating or excessive unbound protamine has anticoagulant properties (102,103). Hyperfibrinolysis is also an important mechanism that leads to post-CPB bleeding (104,105). All three antifibrinolytic therapies (aprotinin, tranexamic acid, and aminocaproic acid) have been shown to decrease post-CPB hemorrhage (106). Aprotinin was removed from the US market in 2008 because of its association with renal failure and death in adult recipients (106,107). Despite the use of these therapies, hemorrhage following CPB remains a frequent occurrence.

Although some bleeding from indwelling mediastinal drains is expected after cardiac surgery, the rate of bleeding should decrease as each postoperative hour goes by. Excessive bleeding is a clinical concern that warrants immediate attention and constant vigilance. In the immediate postoperative period, bleeding of <5 mL/kg/h is often associated with minor abnormalities in coagulation status. Red cell transfusion may be necessary to correct a postoperative anemia but blood component administration is rarely necessary. Bleeding 5 to 10 mL/kg/h should prompt notification of the cardiothoracic surgeon and continued evaluation of the patient at the bedside. Any abnormalities in coagulation parameters should be corrected. Blood loss must be balanced by RBC transfusion. The patient must be closely monitored for persistence or an increase in the rate of bleeding that may signal the presence of a surgical bleeding site or may be the result of loss of coagulation factors secondary to the ongoing hemorrhage. Bleeding of >10 mL/kg/h that persists or increases will likely result in hemodynamic compromise if not abated. The cardiothoracic surgeon should decide whether reexploration is needed to exclude a bleeding site or to remove thrombus that may be perpetuating further bleeding. The cardiac intensivist must exclude concomitant hemopericardium and/or hemothorax (often herald by tachycardia, hypotension, desaturation, and/or tamponade physiology) while keeping up with the blood loss (RBC transfusion) and coagulopathy (blood component replacement). Even after the decision has been made to reoperate, the bedside medical team must continue to be vigilant with RBC, volume, and blood component replacement as indicated by hemodynamic parameters and laboratory data.

Recombinant factor VIIa (rFVIIa, eptacog alfa [activated], NovoSeven, Novo Nordisk Pharmaceuticals Inc., Princeton, NJ) is currently utilized as an off-label hemostatic agent in the management of postoperative CPB bleeding. It is an analog of the naturally occurring procoagulant factor VII and induces hemostasis by binding to exposed tissue factor at the site of endothelial injury and stimulating thrombin formation (108,109). Current studies indicate
that children and adults with cardiovascular disease account for a large proportion of off-label rFVIIa use (110,111).

There are concerns regarding the thrombotic potential of rFVIIa and this concern is further heightened in pediatric patients with CHD because of a known increased incidence of thrombosis (112). Current evidence to support off-label use of rFVIIa in pediatric CPB is limited. Only one pediatric randomized clinical trial of rFVIIa in CPB has been conducted and it was found that there was no benefit to the prophylactic use of rFVIIa among 82 infants undergoing surgery for correction of CHD (113). The primary end point, time to chest closure, was actually prolonged in the treatment group. No difference was noted in the secondary end points of surgical blood loss or use of blood products (113). A recent systematic review regarding the off-label use of rFVIIa in pediatric cardiac surgery reviewed 29 studies that included a total of 169 patients (114). The authors concluded that there is no supporting evidence to use rFVIIa prior to cardiac surgery to prevent bleeding. Randomized clinical trials should be performed that utilize appropriate clinical outcomes to address the use of rFVIIa to treat bleeding after cardiac surgery. Therapeutic options for postoperative bleeding are listed in Table 75.2.

Patients on ECMO or who have ventricular assist devices (VADs) are at increased risk for both bleeding and thrombotic complications. Similar to CPB, both processes have the passage of blood through an artificial circuit resulting in activation of coagulation. Anticoagulation remains the mainstay to prevent thrombotic outcomes. In ECMO, uFH is commonly used for anticoagulation. In patients with a VAD, one type of anticoagulation (uFH, LMWH, or warfarin) is commonly combined with antiplatelet therapy although anticoagulation management will vary by specific device manufacturer recommendations.

Thrombosis has long been recognized as a clinical problem to those caring for adolescents and children with congenital and acquired heart disease. Much work has focused on the diagnosis, treatment, and prevention of thrombosis in Kawasaki disease (115,116,117,118,119,120,121,122,123,124,125) and to a lesser extent on the thrombotic complications associated with cardiac catheterization (126,127,128,129,130,131,132,133,134,135,136) and cardiomyopathies (137,138,139,140,141).

The prevention and management of thromboses related to prosthetic valves (142), arrhythmias (143,144), and pulmonary hypertension (145,146,147,148,149,150) in children has largely been extrapolated from the adult literature. For the past decade, the single-ventricle population has been identified as a particularly high-risk population for thrombosis and their potentially devastating sequelae (151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167).