INTRODUCTION — Pediatricians will be confronted with an infant, child, or adolescent either with overt bruising or bleeding or with a history of increased bleeding sometime during their busy clinical practice. This topic review will discuss the approach to such patients, including selection of the most appropriate laboratory tests to arrive at a working diagnosis. An approach to the adult with bleeding symptoms and neonatal thrombocytopenia are discussed separately. (See "Approach to the adult patient with a bleeding diathesis" and "Neonatal thrombocytopenia".)
BLEEDING HISTORY — Clinical evaluation of a bleeding patient begins with taking a careful history, taking into account the child's age, sex, clinical presentation, past history, and family history. While a bleeding history is being elicited, attention should be directed to the type of bleeding present (table 1):
While assessing a patient, one has to keep in mind that not all bleeding episodes suggest a generalized bleeding disorder. For example, epistaxis may be caused by rhinitis, trauma, superficial vessels, or dry air (see "Epidemiology and etiology of epistaxis in children"), and abnormal post-surgical bleeding (eg, tonsillectomy) may be caused by surgical trauma rather than a generalized bleeding disorder. However, clinicians should be alert to the possibility that bruising or bleeding judged to be abnormal (eg, due to frequency, duration, or severity of episodes, or lack of explanation for symptoms or physical findings) may be caused by a bleeding disorder or by nonaccidental injury (NAI). Furthermore, NAI and bleeding disorders are not mutually exclusive. Therefore, the history should include complete details as to the type of bleeding, location, degree of symptoms, nature of provoking injuries, and whether such injuries are consistent with the child’s development and level of activity [1]. (See "Physical abuse in children: Diagnostic evaluation and management".)
Age and sex of the patient — An inherited bleeding disorder should be strongly considered when the onset of bleeding manifestations occurs in infancy or early childhood and is associated with a positive family history. However, a negative family history does not exclude an inherited coagulation disorder, as up to one-third of patients with hemophilia have a negative family history [2]. (See "Genetics of the hemophilias", section on 'Genetic transmission'.)
The following are examples of typical presentations:
Family history — The family history is helpful in formulating a possible diagnosis of an inherited disorder of coagulation. The presence of bleeding manifestations only in male siblings and maternal uncles is suggestive of X-linked recessive inheritance, such as that seen in hemophilia A or B.
In contrast, in autosomal dominant traits such as hereditary hemorrhagic telangiectasia (Osler-Weber-Rendu disease), an accurate pedigree will show affected individuals of both sexes for several generations [11]. In autosomal recessive disorders, such as severe forms of the rarer coagulation factor deficiencies (eg, factor VII or factor XI deficiency), the family history may be negative; consanguinity increases the probability of such disorders [12]. (See "Hereditary hemorrhagic telangiectasia (Osler-Weber-Rendu syndrome)" and "Rare (recessively inherited) coagulation disorders".)
Medication — A thorough history of medication use, including herbal medicines, is mandatory. In particular, specific information should be sought about the ingestion of aspirin, aspirin-containing over-the-counter medications, and other non-steroidal anti-inflammatory drugs such as ibuprofen or naproxen. Such drugs impair platelet function and may exacerbate an underlying coagulation disorder. (See "Clinical presentation and diagnosis of von Willebrand disease", section on 'Clinical presentation'.) In addition, use of such drugs within one to two weeks of platelet function testing may cause abnormalities, which may lead to further expensive and unnecessary studies.
RELEVANT LABORATORY TESTS — As in adults with bleeding symptoms, certain laboratory tests of hemostasis are used as initial screening tests, whereas other more specific tests are performed at a later time in order to narrow the possibilities or make a definitive diagnosis. It should be emphasized that no current single test reliably screens the overall process of hemostasis (table 2 and algorithm 1), particularly for abnormalities of platelet function or the fibrinolytic pathway.
The usual initial screening tests include quantitation of platelets, examination of the peripheral blood smear, prothrombin time/international normalized ratio (PT/INR), activated partial thromboplastin time (aPTT), and at our center, the fibrinogen activity level. Normal values may vary with age and among different laboratories (table 3). In particular, normal prothrombin and activated partial thromboplastin times should be based upon an individual clinical laboratory's reference ranges [13]; published ranges should not be used to conclusively ascertain whether an individual result is normal or abnormal.
In an otherwise well child who presents with mucocutaneous bleeding, a complete blood count, platelet count, and examination of the peripheral blood smear are the most informative initial tests. Disorders that might be present in such a child depend upon whether the platelet count is normal or low, and, for the latter, whether pancytopenia (ie, thrombocytopenia, anemia, and neutropenia) is present.
Proper collection of the blood sample is essential for interpreting the results of clotting tests. Blood for coagulation tests should not be drawn from an existing heparinized indwelling line. A cleanly drawn venipuncture sample without air bubbles or tissue fluid contamination is the most appropriate sample for coagulation tests. Coagulation tests are performed on blood anticoagulated with a solution of sodium citrate in a ratio of nine parts of blood to one part of citrate. When the hematocrit is high (eg, newborns and children with cyanotic cardiac disease), the amount of citrate must be adjusted (reduced) to provide the proper ratio [14]. (See "Clinical use of coagulation tests", section on 'Obtaining the sample'.)
Platelet counting and the peripheral smear — The platelet count is performed most commonly using automated cell counters. (See "Automated hematology instrumentation".)
Platelets may also be counted directly on the blood smear. This permits assessment of platelet size, which helps to narrow the diagnostic possibilities in a patient with thrombocytopenia. (See 'Thrombocytopenia' below.)
Pseudothrombocytopenia — Examination of the peripheral blood smear is essential in patients with low platelet counts in order to exclude the presence of pseudothrombocytopenia caused by platelet aggregation after using EDTA as an in vitro anticoagulant (picture 1) [15]. Platelet aggregation causes falsely low platelet counts by the automated cell counter, but the platelet clumps are obvious on examination of the smear. Alternative anticoagulants (eg, trisodium citrate or heparin) may circumvent in vitro EDTA-associated platelet aggregation [16]. (See "Clinical manifestations and evaluation of thrombocytopenia in children", section on 'Verification of thrombocytopenia'.)
Prothrombin time (PT) — The production of fibrin via the extrinsic pathway and the final common pathway requires tissue thromboplastin (tissue factor), factor VII, factors X, V, prothrombin (factor II), and fibrinogen. The functioning of these pathways is measured by the PT (figure 1). This test bypasses the intrinsic pathway and uses "complete" thromboplastins (ie, tissue factor) capable of activating the extrinsic pathway.
The PT is sensitive to alterations in the vitamin K-dependent coagulation factors, especially factors II, VII, and X, and is used to monitor treatment with vitamin K antagonists. (See "Clinical use of coagulation tests", section on 'Prothrombin time' and "Therapeutic use of warfarin", section on 'Mechanism of action' and "Therapeutic use of warfarin", section on 'Laboratory monitoring'.)
Activated partial thromboplastin time (aPTT) — The aPTT measures the intrinsic and common pathways of coagulation (figure 1). It is called "partial" because clotting is initiated in vitro with agents that are only partial thromboplastins (ie, they are incapable of activating the extrinsic pathway). This aPTT is routinely used to evaluate intrinsic coagulation and the degree of heparin anticoagulation. (See "Clinical use of coagulation tests", section on 'Activated partial thromboplastin time'.)
The aPTT is sensitive to deficiencies of factors XII, XI, IX, and VIII and to inhibitors such as heparin (figure 1). It is less sensitive than the PT to deficiencies within the common pathway (eg, factors X, V, prothrombin, and fibrinogen) and is unaffected by alterations in factors VII and XIII. Although high levels of a single factor (eg, factor VIII) can shorten the aPTT, whether an association exists between a shortened aPTT and a hypercoagulable state remains controversial. (See "Clinical use of coagulation tests", section on 'Shortened clotting times'.)
Fibrinogen — Functional fibrinogen concentration is most commonly measured using a sensitized modification of the thrombin time (discussed in the next section), whereas structural (antigenic) levels are measured by immunologic assays. Immunologic and functional assays of fibrinogen may be discordant in patients with an inherited dysfibrinogenemia. (See "Disorders of fibrinogen".)
Thrombin time and reptilase time — The thrombin time (TT) is prolonged in the presence of heparin or hypofibrinogenemia (as in disseminated intravascular coagulation). (See "Clinical use of coagulation tests", section on 'Thrombin time'.)
Simultaneous measurement of TT and reptilase time (RT) is useful to assess the possibility of heparin contamination, which prolongs the former but not the latter. These tests measure conversion of fibrinogen to fibrin monomers and the formation of the initial clot by thrombin and reptilase, respectively (figure 1). Reptilase, a thrombin-like enzyme obtained from snake venom, differs from thrombin by generating fibrinopeptide A but not fibrinopeptide B from fibrinogen and by resisting inhibition by heparin via antithrombin III. Fibrin strand cross-linking, mediated by factor XIII, is not measured in these assays (figure 1). (See 'Clot solubility in urea' below.)
Tests for specific factor deficiencies and inhibitors — An abnormally prolonged PT or aPTT can be due to the absence or reduced concentration of a coagulation factor or the presence of an inhibitor to one of the coagulation factors:
Deficiencies of specific factors may be determined by assessing the PT or aPTT in mixtures of patient plasma with commercially available plasma deficient in known factors. Factor levels can be assessed functionally by comparing test results with standard curves generated by mixtures of serially diluted normal plasma and factor-deficient plasma. Immunologic assays also can be used to measure factor levels. Immunologic and functional assays should give equivalent results when a factor deficiency is present (generally referred to as “Type 1 deficiency”). Reduction in a functional assay with a normal immunologic assay suggests the presence of a functionally abnormal factor (“Type 2 deficiency”).
Antiphospholipid antibodies — In addition to factor inhibitors, certain antiphospholipid antibodies (lupus anticoagulants) also can result in a prolonged aPTT that is not correctable by the addition of normal plasma. The effect of these antibodies on the aPTT can be partially overcome by adding excess platelet phospholipid (particularly a hexagonal phase phospholipid) or by assessing the diluted Russell viper venom time [18]. (See "Clinical use of coagulation tests", section on 'Antiphospholipid antibodies'.)
In otherwise healthy children, the finding of a lupus anticoagulant is of no clinical consequence. However, when a lupus anticoagulant is present with other clinicopathologic features (eg, arthritis, serositis, renal abnormalities), an increased thrombotic risk may exist. (See "Clinical manifestations of the antiphospholipid syndrome".)
Clot solubility in urea — The initial clot, held together by noncovalent bonds, is soluble in urea. Subsequent transglutamination within the clot by activated factor XIIIa covalently crosslinks overlapping fibrin strands, which then are resistant to solubilization by urea (figure 1). The ability of urea to solubilize the mature clot reflects a severe deficiency of factor XIII [10]. However, if factor XIII deficiency is seriously suspected, the clot solubility may miss the diagnosis because the test is sensitive only at very low levels (Factor XIII 1 to 3 percent). Therefore, specific quantitative assays are recommended.
Tests for fibrinolysis — Fibrin- and fibrinogen degradation products (FDPs) are protein fragments resulting from the action of plasmin on fibrin or fibrinogen, respectively (figure 2). Elevated levels are seen in states of fibrinolysis such as disseminated intravascular coagulation (DIC). Specific measurement of the concentration of fibrin D-dimers, which are degradation products of cross-linked fibrins, is possible [19]. (See "Overview of hemostasis", section on 'Clot elimination and fibrinolysis' and "Disseminated intravascular coagulation in infants and children".)
Bleeding time and PFA-100 — The bleeding time (BT), which is a measure of platelet interaction with the vessel wall, is not performed routinely as a screening test in children (and adults) because of difficulty in administering the test in a standardized fashion. Moreover, a normal test does not predict the safety of surgical procedures [20]. We do not recommend it as a screening test prior to surgery. (See "Approach to the adult patient with a bleeding diathesis", section on 'Bleeding time' and "Preoperative assessment of hemostasis", section on 'Bleeding time'.)
PFA-100 is a platelet function analyzer designed to measure platelet-related primary hemostasis. It has been used as a screening test for investigation of various inherited and acquired disorders of primary hemostasis [21,22], but remains an inadequate tool for predicting risk of bleeding in the clinical setting [23,24]. We do not recommend PFA-100 as an adequate screening test for either von Willebrand disease or platelet function disorders due to insufficient sensitivity [25,26]. (See "Platelet function testing", section on 'The platelet function analyzer'.)
DIAGNOSTIC APPROACH — Routine screening in a child with a bleeding disorder includes a complete blood count, platelet morphology, and coagulation testing with the PT, aPTT, and fibrinogen level. This initial laboratory testing should allow the clinician to more narrowly define the diagnostic possibilities in the child with a bleeding disorder (table 2 and algorithm 1).
If all three coagulation tests are normal, the child may have an abnormality outside of the intrinsic, extrinsic, and common coagulation pathways. (See 'Normal initial coagulation tests' below.)
ABNORMAL INITIAL COAGULATION TESTS
Normal PT and prolonged aPTT — An isolated prolongation of aPTT is characteristic of intrinsic pathway coagulation factor (factors VIII, IX, XI, and XII), high molecular weight kininogen, or prekallikrein deficiency (algorithm 1). The aPTT also may be prolonged with an acquired inhibitor, such as the lupus anticoagulant. Because von Willebrand factor (vWF) protects factor VIII from proteolysis, decreased plasma vWF or a mutation in the factor VIII binding site in type 2N von Willebrand disease can also lead to decreased plasma factor VIII concentrations and a prolonged aPTT. (See "Clinical presentation and diagnosis of von Willebrand disease", section on 'Laboratory testing'.)
The following conditions are characterized by prolonged aPTT:
Prolonged PT and normal aPTT — An isolated prolongation of the PT is characteristic of inherited or acquired factor VII deficiency (figure 1 and algorithm 1). Inherited factor VII deficiency displays phenotypic and molecular heterogeneity, whereas acquired factor VII inhibitors are very rare occurrences during childhood [29]. (See "Acquired inhibitors of coagulation", section on 'Factor VII inhibitors' and "Rare (recessively inherited) coagulation disorders".)
Prolonged PT and aPTT
Well child — Prolongation of both PT and aPTT in a bleeding child who is otherwise well indicates an inherited disorder within the common pathway or an acquired disorder involving multiple pathways (figure 1 and algorithm 1).
Inherited deficiencies yielding this laboratory result include deficiency of factor X, V, II (prothrombin) or fibrinogen; these deficiencies are rare. (See "Rare (recessively inherited) coagulation disorders".)
Inherited disorders of fibrinogen (hypo- or afibrinogenemia) are autosomal recessive disorders, and bleeding associated with these disorders is treatable with cryoprecipitate or fibrinogen concentrates. Dysfibrinogenemia, an autosomal dominant disorder, may be associated with either bleeding or excessive clotting. (See "Disorders of fibrinogen".)
Sick child — In a sick child with prolongation of both PT and aPTT, disorders to consider are disseminated intravascular coagulation (DIC), fulminant sepsis with DIC, severe hepatocellular dysfunction, and severe vitamin K deficiency (algorithm 1). Because the production of factor V is independent of the status of vitamin K, the factor V level can be used to distinguish between vitamin K deficiency (in which factor V is normal) and liver disease or DIC (in which factor V is decreased).
Major vessel thrombosis, consumption coagulopathy in certain vascular lesions (see "Neonatal thrombocytopenia", section on 'Kasabach-Merritt syndrome'), and adult respiratory distress syndrome (ARDS) are other rare causes of prolonged PT and aPTT in a sick child.
An acquired inhibitor to prothrombin is seen rarely in a patient with systemic lupus erythematosus [30]. Similarly rare is an acquired inhibitor to factor X in amyloidosis [31]. An acquired inhibitor to factor V rarely has been reported in association with use of topical bovine thrombin, which may be used during vascular or cardiothoracic surgery in the form of "fibrin glue" [32]. Associated bleeding can at times be severe and unpredictable. (See "Acquired inhibitors of coagulation", section on 'Prothrombin inhibitors' and "Acquired inhibitors of coagulation", section on 'Factor X inhibitors' and "Acquired inhibitors of coagulation", section on 'Factor V inhibitors'.)
Accidental or intentional ingestion of warfarin or warfarin-containing rodenticides sufficient to cause bleeding usually results in a prolongation of the PT and aPTT because the vitamin K-dependent factors which are inhibited by warfarin are present in the extrinsic (factor VII), intrinsic (factor IX), and common pathways (factors II and X) (figure 1) [33]. Hemorrhage under such circumstances may be life-threatening and requires immediate treatment with combinations of intravenous vitamin K, fresh frozen plasma, and/or prothrombin complex concentrates (which contain all of the vitamin K-dependent coagulation factors). (See "Correcting excess anticoagulation after warfarin", section on 'Treatment'.)
Prolonged thrombin time — A prolonged TT may be due to hypofibrinogenemia, structurally abnormal fibrinogens (dysfibrinogenemias), or the presence of increased concentrations of fibrin split products (figure 1) [34]. Because heparin prolongs the TT but not the reptilase time (RT), the RT is useful for determining if heparin is the cause of a prolonged TT. (See 'Thrombin time and reptilase time' above and "Clinical use of coagulation tests", section on 'Thrombin time'.)
NORMAL INITIAL COAGULATION TESTS
Normal platelet count — In children with bleeding symptoms and an initial laboratory screen with a normal platelet count and initial coagulation screening tests, possible diagnoses include von Willebrand disease (vWD), factor XIII deficiency, platelet function disorder, vascular abnormality, and a fibrinolytic disorder.
Von Willebrand disease — vWD is the most common inherited bleeding disorder, with an estimated prevalence as high as 1 percent in some reports [25]. However, another report suggests substantially lower prevalence: Over 4500 children who presented to a children’s outpatient clinic were screened by asking “does your child have a problem with bleeding/bruising?” Approximately 5 percent of caretakers answered yes. With more detailed history, followed by specific testing for vWD, only 1 in 1000 was found to have a diagnosis of vWD [35].
There are three major types of vWD. Types 1 and 3 are quantitative deficiencies of vWF, whereas type 2 is a qualitative disorder. Laboratory tests for vWD include factor VIII assay, vWF activity (ristocetin cofactor assay) and vWF antigen. The platelet count may be low in some patients with type 2B von Willebrand disease. (See "Clinical presentation and diagnosis of von Willebrand disease", section on 'Laboratory testing'.)
Treatment of vWD is dependent on the correct typing and subtyping. (See "Treatment of von Willebrand disease".)
Factor XIII deficiency and other fibrinolytic disorders — Activated factor XIII is responsible for clot stabilization and cross linking of fibrin polymer (figure 1). Deficiency of this factor is an autosomal recessive disorder resulting in reduced clot stability and abnormal bleeding. One of the characteristic abnormalities of factor XIII deficiency is delayed separation of the umbilical cord and delayed bleeding from the umbilical stump. In the neonatal period, intracranial hemorrhage with little or no trauma and poor wound healing also are associated with the deficiency. If factor XIII deficiency is suspected, the quantitative assay should be performed [10]. (See 'Clot solubility in urea' above.)
Factor XIII has a very long half-life. A plasma-derived factor XIII concentrate (Fibrogammin) has completed extensive trials and is available for clinical use [10,36]. (See "Rare (recessively inherited) coagulation disorders", section on 'Factor XIII deficiency'.)
Deficiencies of alpha 2 antiplasmin and plasminogen activator inhibitor have also been associated with an increased bleeding tendency. (See "Thrombotic and hemorrhagic disorders due to abnormal fibrinolysis", section on 'Alpha-2-antiplasmin deficiency' and "Thrombotic and hemorrhagic disorders due to abnormal fibrinolysis", section on 'Plasminogen activator inhibitor'.)
Platelet function disorders — Studies to confirm the presence of qualitative disorders of platelet function include evaluation of platelet morphology on the peripheral blood smear, tests of platelet aggregation, and other tests of platelet function [37,38]. (See "Platelet function testing".)
Classic inherited disorders of platelet function are relatively rare and include:
Acquired causes of abnormal platelet function are much more common than inherited causes, and include use of aspirin and nonsteroidal antiinflammatory drugs, beta-lactam antibiotics, uremia, and the myeloproliferative and myelodysplastic syndromes (see "Congenital and acquired disorders of platelet function", section on 'Acquired platelet functional disorders'). As noted above, a history of ingestion of aspirin or aspirin-like products is a critical portion of the workup of any patient with a bleeding disorder. In addition, if platelet function tests are to be performed (eg, bleeding time, platelet aggregation), the subject must refrain from taking these products for at least one to two weeks before testing.
Vascular purpuras — Screening tests usually are normal in patients with bleeding disorders related to vascular abnormalities. They include:
Undetermined etiologies — Patients with a significant bleeding history for which no explanation exists are occasionally encountered. Physical abuse of the child should be considered in such cases. (See "Physical abuse in children: Diagnostic evaluation and management".) However, some disorders of hemostasis may escape detection with currently available methods.
Thrombocytopenia — Thrombocytopenia occurs in association with several of the disorders mentioned above. The appearance of the platelets on the peripheral smear helps to establish the diagnosis:
Pancytopenia — If the child has pancytopenia with or without organomegaly or lymphadenopathy, examination of the peripheral blood smear may reveal the presence of leukemic blasts, an observation that should be confirmed with a bone marrow examination. (See "Evaluation of the peripheral blood smear", section on 'Worrisome findings'.)
Another diagnosis to consider in a child with mucocutaneous bleeding and pancytopenia is aplastic anemia. Subjects with aplastic anemia present with varying combinations of symptomatic anemia, bleeding, and infection, depending upon the severity of the pancytopenia. Single or multiple skeletal anomalies may be present in children with the congenital forms of aplastic anemia. (See "Acquired aplastic anemia in children and young adults", section on 'Clinical presentation and diagnosis' and "Inherited aplastic anemia in children", section on 'Clinical presentation' and "Inherited aplastic anemia in children", section on 'Diagnosis'.)
SUMMARY AND RECOMMENDATIONS
ACKNOWLEDGMENT — The editorial staff at UpToDate, Inc. would like to acknowledge Dr. Indira Warrier, who contributed to an earlier version of this topic review.
All topics are updated as new information becomes available. Our peer review process typically takes one to six weeks depending on the issue.