Identification and Basic Management of Bleeding Disorders in Adults

Rebecca Kruse-Jarres, MD; Tammuella C. Singleton, MD; Cindy A. Leissinger, MD


J Am Board Fam Med. 2014;27(4):549-564. 

In This Article

Diagnostic Considerations in an Adult With a Potential Bleeding Disorder

Abnormal Bleeding Symptoms

Bleeding is a common symptom and does not always indicate an underlying bleeding disorder. Symptoms such as gum bleeding, epistaxis, menorrhagia, petechiae, and bruising are especially common; in one study they were reported by anywhere from 22% to 85% of men and women without bleeding disorders.[2] Identification of pathologic bleeding may, therefore, prove challenging. Clinically significant mucocutaneous bleeding is defined as any of the following: spontaneous or provoked bleeding from 2 or more distinct mucocutaneous sites; bleeding from a single site warranting blood transfusions; or bleeding from a single site on 3 or more separate occasions.[3,4] Bleeding scoring systems have shown promise in retrospectively predicting bleeding phenotype in type 1 VWD[5] and prospectively excluding mild bleeding disorders in patients presenting with bleeding symptoms or abnormal coagulation study results,[6] but they require further investigation and validation for broader clinical use.

A thorough history and physical examination often provides clues as to whether bleeding is pathologic and may even point to potential underlying causes. Historical factors to explore are outlined in Table 1 .[7–11] During physical examination, the skin and mucous membranes should be inspected for stigmata of bleeding (eg, bruising, petechiae) and other findings suggestive of potential underlying causes of bleeding (eg, jaundice, telangiectasia). The presence of hepatomegaly, splenomegaly, or joint hypermobility may suggest potential diagnoses associated with bleeding. Skin or conjunctival pallor, tachycardia, or a cardiac flow murmur may indicate associated anemia. Historical and physical findings may suggest an abnormality of either primary hemostasis, which culminates in the formation of a platelet plug; secondary hemostasis, in which fibrin is formed via the coagulation "cascade"; or fibrinolysis, the normal breakdown of clots. A simplified schematic of the coagulation cascade and the corresponding laboratory assays for each pathway are provided in Figure 1. A more thorough review of the complex process of hemostasis and its various components is beyond the scope of this article but can be found elsewhere.[12–14]

Figure 1.

12,29 Simplified schematic of the coagulation "cascade." The coagulation cascade consists of three pathways: the intrinsic pathway, the extrinsic pathway, and the final common pathway, culminating in the formation of fibrin. This model of coagulation oversimplifies the process of in vivo coagulation but is useful for the correlation of coagulation assay (ie, activated partial thromboplastin time [aPTT], prothrombin time [PT]) abnormalities with specific pathways and, hence, coagulation factors. FI, fibrinogen; FII, factor II; FIX, factor IX; FV, factor V; FVII, factor VII; FVIII, factor VIII; FX, factor X; FXI, factor XI; FXII, factor II; HMWK, high-molecular-weight kininogen; PK, prekallikrein; TF, tissue factor.

Excessive bruising, epistaxis, bleeding after dental extraction, and menorrhagia are symptoms suggestive of quantitative or qualitative platelet disorders.[8] Patients with platelet abnormalities may also experience excessive bleeding after hemostatic challenges.[15] The presence of petechiae in particular suggests a platelet defect. Quantitative platelet abnormalities in adults are most often acquired; autoimmune (ie, idiopathic thrombocytopenic purpura [ITP]) and drug-induced thrombocytopenia account for the vast majority of cases of isolated thrombocytopenia.[16] Certain acquired thrombocytopenic conditions (eg, disseminated intravascular coagulation [DIC] and thrombotic thrombocytopenic purpura [TTP]/hemolytic uremic syndrome) may present with bleeding in acutely ill patients ( Table 2 ) but may also be the underlying reason for asymptomatic thrombocytopenia in an ambulatory patient. Congenital (or inherited) thrombocytopenias are usually diagnosed during childhood, but diagnosis could be delayed into adulthood, especially in individuals who do not regularly obtain health care. Numerous inherited thrombocytopenias exist; individual conditions may be identified based on platelet size, coexisting physical or laboratory abnormalities, and the presence of defective platelet function as well as an abnormal platelet count.[8,17,18] In most inherited forms, thrombocytopenia is mild: bleeding occurs only occasionally or after hemostatic challenge.[17] A family history of thrombocytopenia (including thrombocytopenia erroneously attributed to other causes such as ITP[16]) may suggest an inherited thrombocytopenia. Conversely, documentation of previously normal platelet counts would exclude congenital thrombocytopenia.[19]

Qualitative platelet disorders presenting in adults can be caused by medication (eg, aspirin and nonsteroidal anti-inflammatory drugs [NSAIDs]), uremia, cirrhosis, and myeloproliferative disorders.[3,20] Several inherited disorders of platelet function exist as well and are classified in various ways (by defective platelet function [eg, adhesion or aggregation] or platelet component [eg, receptors, granules, or membrane phospholipids]).[8,15] Although severe inherited disorders of platelet receptors (eg, Glanzmann thrombasthenia and Bernard-Soulier syndrome) and some platelet granule disorders typically present earlier in life, the majority of inherited platelet function disorders present during adulthood, often after hemostatic challenge.[8,21]

VWD may also be considered a disorder of platelet function, given the role of von Willebrand factor (VWF) in platelet adhesion and aggregation.[22] Consequently, VWD tends to present with clinical signs and symptoms similar to those of platelet disorders. In patients with a relevant history (eg, significant mucocutaneous bleeding, family history), initial hematologic laboratory evaluation should include specific testing for VWD (VWF antigen, VWF ristocetin cofactor activity, and factor VIII activity assays); other screening tests such as activated partial thromboplastin time (aPTT) and bleeding time or a platelet function analyzer (PFA-100; Siemens Health care Diagnostics Inc., Tarrytown, NY) may miss VWD.[23] Routine hemostatic screening laboratories may likewise miss other platelet function defects. Consequently, in patients with mucocutaneous bleeding who do not have thrombocytopenia or VWD, platelet aggregometry should be considered as an initial test for assessing platelet function,[3] in addition to evaluation of a peripheral blood smear for abnormalities in platelet morphology that are specific to certain conditions (eg, gray platelet syndrome).

In contrast to the superficial bleeding associated with platelet defects, coagulation factor defects result in delayed, deep bleeding, for example, into muscles or joints, as well as deep soft-tissue and mucocutaneous bleeding. Patients with milder congenital deficiencies or those with certain specific congenital deficiencies (eg, factor XI deficiency) are more likely to bleed after hemostatic challenge.

Menorrhagia is a common bleeding symptom in women, both those with and without bleeding disorders. Menorrhagia is the most common bleeding symptom in women with inherited bleeding disorders,[24] particularly menorrhagia that begins at menarche and persists into adulthood.[25] When menorrhagia begins after the age of 20 years, acquired bleeding diatheses should be considered, as should nonhematologic causes such as uterine pathology (eg, fibroids), hypothyroidism, and, in women older than 40 years of age, perimenopausal anovulation.[25] Consensus recommendations for the evaluation of acute menorrhagia have been published.[25] They include assessments of the patient's menstrual, bleeding, medication, and family histories; speculum and pelvic examinations with subsequent Papanicolaou test and endometrial biopsy, as appropriate, based on such factors as patient age and feasibility of performing these interventions through heavy menstrual bleeding; and (preferably intravaginal) ultrasound.[25] Recommended initial laboratory testing includes a complete blood count, pregnancy test, prothrombin time (PT), aPTT, fibrinogen, and, if feasible, VWF levels.[25] Additional studies may include tests of liver or platelet function or specific factor levels, as clinically indicated. Samples may also be drawn for storage for future testing,[25] especially if administration of transfusional therapies (eg, fresh frozen plasma [FFP]) is anticipated.

Hematologic Laboratory Abnormalities

Once a significant bleeding history is identified, an initial laboratory evaluation is generally undertaken to determine the underlying cause. Alternatively, an adult with an undiagnosed bleeding disorder may present with abnormal hematologic laboratory studies obtained as part of an evaluation for surgery or for some other reason. Increased sensitivity of the reagents used in coagulation assays, most notably PT and aPTT, has led to an increased incidence in abnormal results for these tests.[26] Screening coagulation laboratory studies have a low yield overall in the absence of any symptoms or family history of an underlying bleeding disorder.[27] Even among patients with a high pretest probability of having a bleeding disorder, only a minority of abnormalities correspond with a clinically significant bleeding diathesis.[28] Failure to identify the subset of individuals whose abnormal coagulation studies signify an as yet undiagnosed bleeding disorder, however, may have serious, if not grave, consequences, particularly if an invasive intervention is planned.

Abnormal Coagulation Assays. Prolongation of the aPTT or PT may indicate an acquired or congenital clotting factor deficiency or an inhibitor of one or more coagulation factors. Potential inhibitors include medication (namely anticoagulants), antibodies directed against specific coagulation factors, and nonspecific inhibitors (eg, lupus anticoagulants). A mixing study can be used to differentiate a deficiency from an inhibitor. In a mixing study, equal volumes of normal and patient plasma are combined, and then the coagulation study is repeated. In cases of coagulation factor deficiency, the presence of normal plasma replaces the missing factor(s), thereby normalizing the abnormal coagulation study. In contrast, when an inhibitor is present the abnormality persists after the addition of normal plasma. In some cases, a prolonged incubation period after mixing is necessary for accurate interpretation of results; therefore, it is imperative that the mixing study be incubated at 37°C for 2 hours. Readily identifiable causes of coagulation study abnormalities (eg, anticoagulant medication, systemic diseases such as liver disease, or artifactual prolongation [eg, caused by sample "contamination" with heparin]) should ideally be excluded before proceeding to a mixing study.[29]

The potential coagulation factors involved and, therefore, possible diagnoses can be narrowed down based on which coagulation study is abnormal[29–32] (Figures 1 and 2). Isolated prolongation of the aPTT indicates an abnormality of the intrinsic pathway (ie, of prekallikrein, high-molecular-weight kininogen, factor VIII [FVIII], factor IX, factor XI, or factor XII [FXII]).[29,30] An isolated prolonged aPTT can also indicate a lupus anticoagulant. An isolated prolonged PT indicates an abnormality of the extrinsic pathway (ie, factor VII [FVII]).[29,30] On occasion, congenital deficiencies of the final common pathway factors II (FII, also known as prothrombin), V (FV), and X (FX) and fibrinogen present with an isolated prolonged PT and a normal aPTT.[29] Not all conditions that prolong aPTT or PT are associated with a bleeding phenotype (Figure 2). For example, lupus anticoagulant is more likely to be associated with thrombosis than with bleeding, except in rare cases of associated antiprothrombin antibodies, which lead to bleeding symptoms and a prolonged PT in addition to prolonged aPTT.[32] Deficiencies of the contact-activating factors (FXII, prekallikrein, and high-molecular-weight kininogen) are rare and do not cause easy bleeding; however, they are associated with markedly prolonged aPTTs. When asked to evaluate an asymptomatic patient who has a markedly prolonged aPTT, testing for lupus anticoagulant should be considered first when the mixing study fails to correct and for FXII deficiency when the aPTT corrects in the mixing study.

Figure 2.

29–32 Differential diagnosis for abnormalities of aPTT and PT. Once the coagulation laboratory study abnormality has been identified, the differential diagnosis may be further narrowed down based on the specific coagulation study abnormalities (activated partial thromboplastin time [aPTT], prothrombin time [PT], or both); the presence or absence of bleeding symptoms; and the results of the mixing study. Note that prolonged incubation may be required for accurate mixing study results. *PT may also be prolonged by heparin (at high doses) or direct thrombin inhibitor (DTIs). aPTT may also be prolonged in FII and FX deficiencies. aPTT may also be prolonged in the setting of advanced liver disease or vitamin K deficiency. §Applies to 10% of lupus anticoagulants. DIC, disseminated intravascular coagulation; FII, factor II; FIX, factor IX; FV, factor V; FVII, factor VII; FVIII, factor VIII; FX, factor X; FXI, factor XI; FXII, factor XII; HMWK, high-molecular-weight kininogen; VWD, von Willebrand disease.

Prolongation of both aPTT and PT isolates the abnormality to the final common pathway, consisting of FV, FX, FII, and fibrinogen.[29,30] Congenital or acquired deficiencies of any of these factors may present with a prolonged PT and aPTT. Acquired deficiencies of single coagulation factors may occur in the setting of systemic diseases such as amyloidosis (FX) and myeloproliferative disease (FV) and must be differentiated from congenital deficiencies.[29] Deficiencies of multiple factors from both the intrinsic and extrinsic pathways or from all 3 pathways may also simultaneously prolong aPTT and PT.[29] Multiple factor deficiencies may occur as a result of severe liver disease, supratherapeutic warfarin doses resulting in deficiency of vitamin K–dependent factors, or consumptive coagulopathy (ie, DIC, which generally occurs in the setting of systemic illness and is therefore unlikely to present solely with asymptomatic coagulation study abnormalities). Potential inhibitors that may present with prolonged aPTT and PT include heparin, direct thrombin inhibitors, potent lupus anticoagulants, and other nonspecific inhibitors such as those associated with lymphoproliferative disorders or monoclonal protein disorders. Bleeding symptoms are generally a feature of all conditions that simultaneously prolong PT and aPTT.

Vitamin K deficiency and liver disease both may result in prolongation of PT or, in more advanced stages, of both PT and aPTT. The vitamin K–dependent coagulation factors (II, VII, IX, and X) may become depleted because of malabsorption, prolonged antibiotic use, or warfarin therapy.[29] Liver disease is distinguished from vitamin K deficiency by a deficiency of FV in addition to the vitamin K–dependent factors.[29] Liver disease is often quite advanced by the time abnormalities in coagulation laboratory studies are present; therefore, patients with liver disease are unlikely to present solely with asymptomatic coagulation laboratory abnormalities and often have concurrent physical signs (eg, jaundice, hepatomegaly) or other laboratory abnormalities indicative of impaired hepatic function (eg, thrombocytopenia, hypoalbuminemia, transaminitis).

Acquired coagulation factor inhibitors (or autoantibodies), most commonly directed against FVIII (a condition referred to as acquired hemophilia), deserve special mention because they may be associated with serious bleeding in adults with no history of bleeding. Acquired hemophilia is a rare condition (incidence of 1 to 4 per million per year[33]) that predominately affects older adults. In the largest collection of affected patients to date (n = 501), the median age at diagnosis was 74 years; however, younger women in particular may be affected as well because of an association with pregnancy.[34] In approximately half of cases, a coexisting underlying condition such as pregnancy is identified,[35] some of which (eg, cancer, autoimmune disease) are characterized by immune dysregulation. Acquired hemophilia should be suspected in an adult with new- or recent-onset bleeding who has no personal or family history of bleeding and presents with an isolated prolonged aPTT that does not correct in a mixing study.[35] Acquired hemophilia was associated with an especially high mortality—up to 41% in untreated patients[36–38] and 6% to 8% among effectively treated patients,[39,40]—mostly because of rebleeding. Because acquired hemophilia requires specialized treatment, prompt diagnosis is important, particularly when an invasive procedure is necessary.[38]

Thrombocytopenia. Thrombocytopenia is defined as a platelet count below the lower limit of the normal range (ie, <150,000/μL in most laboratories). In asymptomatic patients, artifactual thrombocytopenia as a result of platelet clumping may first be excluded by examining the peripheral blood smear.[41] Thrombocytopenia occurs because of impaired production,[16,17,19,20,41–43] destruction or consumption,[16,41,42,44–46] or sequestration[41,42] of platelets ( Table 2 ). Immune thrombocytopenias, either idiopathic (primary) or secondary to an autoimmune disease, may present with asymptomatic, isolated thrombocytopenia.[47] Other potential causes in ambulatory patients include medication; infectious agents such as Epstein-Barr, human immunodeficiency, or hepatitis C virus; or primary marrow failure.[41] Conversely, certain thrombocytopenic conditions may be excluded based on the lack of specific predisposing factors or acute illness (eg, shiga toxin–induced hemolytic uremic syndrome or DIC). Bleeding propensity in thrombocytopenic conditions usually depends on the platelet count. Bleeding is generally mild and limited to easy bruising when platelet counts exceed 20,000/μL.[41] The risk for spontaneous bleeding increases only after the platelet count decreases to <10,000/μL,[41] except in ITP, in which the increased presence of young, hyperfunctional platelets may preserve hemostasis even at platelet counts below this level.[48]

Abnormal Platelet Function. Historically, bleeding time was used as a screening test for qualitative platelet abnormalities, including in patients undergoing invasive procedures, particularly those who were recently exposed to medication that might alter platelet function (eg, aspirin or NSAIDs). This test also was used to screen for certain bleeding disorders, including VWD.[23,29] However, bleeding time has been shown to be relatively insensitive and poorly reproducible.[23,49,50] The test has not been shown to predict excessive surgical bleeding (particularly without other history suggesting a bleeding disorder) or to reliably identify aspirin- or NSAID-induced platelet dysfunction.[49] In addition, some patients with VWD have a normal bleeding time.[23] Therefore, this test is no longer routinely recommended as part of an initial hemostatic laboratory evaluation, especially to screen for VWD.[23] For assessing platelet function, the PFA-100 has supplanted the bleeding time in many laboratories.[51] In this technique, a sample of whole blood is passed through an aperture in a membrane coated with collagen and either epinephrine or adenosine diphosphate, and the amount of time it takes for the membrane to become occluded (ie, the closure time) is measured.[51] A closure time exceeding 300 seconds is considered prolonged.[52] Although the PFA-100 has been shown to have a relatively high sensitivity for detecting moderate and severe VWD, aspirin- and NSAID-related platelet dysfunction, and severe platelet function disorders, closure times may be normal in milder VWD and platelet function disorders, including relatively common platelet storage pool deficiencies,[51–55] thereby limiting the usefulness of the PFA-100 as a screening tool for all platelet function disorders.[54] If a patient is suspected of having a significant platelet function disorder, consultation with a hematologist is suggested to obtain more extensive platelet aggregation and release studies in a specialized laboratory setting.

Normal Initial Hematologic Laboratory Studies. Of note, several bleeding disorders are associated with normal routine initial hematologic laboratory studies (ie, platelet count, PT, aPTT, and PFA-100) ( Table 3 ). In some cases, the sensitivity of these studies for detecting certain conditions, such as some types of VWD or milder factor deficiencies, may be limited. In other cases, neither fibrin generation nor platelet function is impaired; therefore, coagulation studies and quantitative and qualitative platelet test results are normal. Alternative screening tests are indicated when these bleeding disorders are suspected based on clinical grounds[13,23,56–60] ( Table 3 ).