Evaluation of the Patient With Suspected Bleeding Disorders

BLEEDING HISTORY EVALUATION

When evaluating a bleeding history, it is important to recognize that the presenting problems are influenced by the following factors:

• 1.

  The nature and severity of the defect, and the presence of single or multiple risk factors for bleeding

• 2.

  Whether the bleeding problem is congenital or acquired

• 3.

  Antecedent exposure to hemostatic challenges (e.g., surgery, dental extraction, menses, and childbirth) and the risk for bleeding with each of these challenges

• 4.

  The presence of other medical problems (e.g., renal, hepatic, or thyroid disease), including anemia

• 5.

  Variability in the bleeding symptoms experienced by individuals without bleeding disorders (e.g., nosebleeds, bruising) and by individuals with known bleeding disorders, even within families with the same defect

• 6.

  Local factors (e.g., sun-damage to the skin, vascular lesions, diverticular disease, or cancerous lesions in the gastrointestinal [GI] tract) and the possibility of nonaccidental trauma

• 7.

  Treatments that increase the risk for bleeding (e.g., antiplatelet agents, such as aspirin and nonsteroidal antiinflammatory drug [NSAID] used for pain control, anticoagulants)

• 8.

  Whether treatments were used to prevent or control bleeding

• 9.

  Whether treatments prescribed for other reasons may have reduced bleeding (e.g., reduced menstrual bleeding while on oral contraceptives to prevent pregnancy)

A detailed history of bleeding episodes, including assessment of the family history, is critical for an accurate assessment of whether a bleeding diathesis is present, as well as the nature of such a bleeding tendency. Questions should therefore be aimed at determining the likelihood of a bleeding disorder being present as well as the type of the putative bleeding diathesis (is this a disorder of primary or secondary hemostasis?) and its inheritance pattern. The history should include an orderly description of bleeding during infancy and childhood, including umbilical stump bleeding (characteristic of factor XIII deficiency [see Chapter 135 ]), bleeding with circumcision (characteristically seen in boys with severe hemophilia A or B [see Chapter 134 ]), bleeding with loss of deciduous teeth, and bleeding with childhood trauma and surgeries (see Chapters 148 and 155 ). Bleeding with dental procedures, including wisdom tooth removal, should be explored. Questions such as “Did you have to go back for stitches? Did you awaken with a pillow covered with blood? How many days of bleeding did you experience after your dental extractions?” are more specific than “Did you bleed with tooth removal?” Patients with milder bleeding disorders may only bleed with procedures involving mucosal surfaces, due to the high levels of fibrinolytic activity at these sites. Epistaxis may be a presenting symptom of von Willebrand disease (VWD) or hereditary hemorrhagic telangiectasia (HHT) and is especially notable if it does not stop with pressure and requires either cautery or a visit to the emergency department. Other bleeding episodes, whether spontaneous or provoked, should be elucidated. Bleeding into muscles and joints is characteristic of moderate or severe disorders of humoral clotting factors and some fibrinolytic defects (e.g., Quebec platelet disorder ), whereas mucosal bleeding is typical of disorders of primary hemostasis. Easy bruising is a complaint voiced by many persons without underlying bleeding disorders, but certain historical features are worth noting. The new onset of bruising can herald a new thrombocytopenic disorder such as immune thrombocytopenia (ITP) (see Chapter 129 ) or acute leukemia (see Chapters 60 , 63 , 66 , 67 and 68 ), or it can point to acquired hemophilia (see Chapter 134 ). Bruising which only occurs over the hands and forearms in older individuals suggests the presence of senile purpura. Some cases of bruising can be explained by fragile skin caused by sun exposure, cigarette smoking, or corticosteroid use. Other cases of bruising can be provoked by use of serotonin reuptake inhibitors or herbal supplements. Lastly, there are rare syndromic bruising disorders such as the Gardner-Diamond syndrome, mostly seen in young women with underlying psychiatric disorders, with a characteristic pattern and appearance to the bruises.

Each individual surgical and major dental procedure undergone by the patient should be explored in depth. The details of bleeding, including timing (immediate or delayed), the need for transfusion, comments by the surgeon concerning the characteristics of the bleeding (e.g., did it start on the day of the surgery/procedure or days later?), any known anatomic sources of bleeding, etc. can illuminate the bleeding diathesis. Immediate bleeding may be more characteristic of a disorder of primary hemostasis, whereas delayed bleeding is seen more in patients with deficiencies in humoral clotting factors and fibrinolytic defects. Bleeding in patients with an underlying primary hemostatic defect is typically described as “diffuse oozing,” without the readily identifiable bleeding source seen with a surgical mishap, such as a severed vessel. Patients with primary hemostatic defects often report that their bleeding occurred on the day of surgery, rather than days later, which would suggest a coagulation factor deficiency or a fibrinolytic disorder. If a woman has bled with some procedures but not others, she should be asked if she was on estrogen-containing medications during the procedures in which she had good hemostasis, because estrogen can increase levels of von Willebrand factor (VWF), leading to normalization of hemostasis. Bleeding with some but not all procedures also suggests a mild rather than a severe disorder.

Women should be carefully questioned about their menstrual history. Duration and severity of flow are more important than the presence or severity of cramping. Although HMB is medically defined as loss of more than 80 ml of blood per menstrual cycle, few women are capable of determining this with any degree of precision. Pad or tampon use is imprecise as well because the number of sanitary products used may vary with the degree of fastidiousness of each patient. To that end, pictorial assessments of blood loss (depicting pads or tampons with varying degrees of saturation) have been devised, with scores given for numbers of products used and their saturation. An underlying bleeding disorder is found in up to one-third of adolescents and adults who present with HMB. Historical features correlated with a higher likelihood of an underlying bleeding disorder being found include “flooding” and “gushing” accidents, passage of clots larger than 1 cm, duration longer than 7 days, heavy flow greater than 3 days, and the development of iron deficiency. While bleeding during pregnancy is less common in women with VWD, platelet function disorders, and most other bleeding disorders (exceptions include fibrinogen disorders, plasminogen activator inhibitor-1 [PAI-1] deficiency, and factor XIII deficiency), postpartum hemorrhage is more commonly seen compared with the general population. Postpartum hemorrhage usually occurs 24 to 48 hours after delivery, and in women with bleeding problems, postpartum bleeding (lochia) can continue beyond the normal 6 weeks.

A family history of bleeding should be carefully sought out, with consideration of whether consanguinity is culturally accepted by the family. The family history evaluation may require several visits to fully document, as familial memories are probed. A family history of bleeding with surgical procedures, bleeding requiring transfusions, and HMB leading to hysterectomy at a young age should be queried. However, a negative family history does not rule out a congenital bleeding disorder. Approximately one-third of all cases of hemophilia A arise from spontaneous mutations. Many of the rare coagulation disorders, including deficiency of factors II, V, VII, and X (see Chapter 135 ), Glanzmann thrombasthenia (see Chapter 124 )), and VWD type 2 N (see Chapter 133 ), among others, are inherited in an autosomal recessive fashion, and other family members may be entirely asymptomatic or bleeding problems may be limited to siblings. (See Case 1: Illustration of a Mild, Inherited Bleeding Problem .)

Certain medications and herbal and dietary supplements increase the risk of bleeding. The use of these agents may precipitate a hemorrhage in those with milder bleeding disorders. The use of aspirin and nonsteroidal antiinflammatory agents impairs primary hemostasis (see Chapter 128 ), and their use should be avoided prior to surgery or prior to evaluation of the hemostatic system. Their inclusion in over-the-counter products seems ubiquitous, and careful attention to cold and flu remedy use is warranted. In the Southeast United States, aspirin-containing remedies such as Goody powders or BC powders are not viewed as medications, and their use must be specifically queried.

Without treatment, severe bleeding disorders typically cause abnormal bleeding with all major hemostatic challenges. In contrast, bleeding may not occur with every challenge in persons with milder defects. Age influences the bleeding history by increasing the likelihood of exposures to hemostatic challenges and the development of sequelae such as arthropathy in patients with severe hemophilia and Quebec platelet disorder. A severe unexplained bleed with surgery may be considered suspicious of a bleeding problem. However, if the person reports that prior challenges did not result in bleeding, this narrows down the possibilities to a mild inherited bleeding disorder, an acquired bleeding disorder, an iatrogenic condition (e.g., bleeding while on anticoagulant therapy or a technical problem during surgery that caused bleeding), or a problem that is not related to a congenital or acquired bleeding disorder (e.g., a postpartum hemorrhage after a caesarean section from uterine atony or retained products of conception).

A bleeding disorder assessment needs to consider both familial and personal bleeding symptoms (see Case 2: Illustration of The Importance of Assessing Both Personal and Familial Bleeding Problems ). However, a familial tendency toward bleeding may “pseudonormalize” bleeding symptoms and make recognition of abnormal bleeding patterns difficult. (See Case 3: Illustration of Delayed Recognition of Abnormal Bleeding .)

The timing of bleeding with challenges is evaluated to determine whether the bleeding problem reflects a common bleeding disorder, such as VWD, a platelet function disorder, an undefined mucocutaneous bleeding problem, or a rarer cause, such as a defect or deficiency in a coagulation factor or fibrinolytic protein (see Case 4: Illustration of the Importance of Assessing Bleeding Problems Over Time ). Although patients will often know whether bleeding began on the day of a challenge or if it started days later, their recall of timing details is often better when procedures were done without general anesthesia (e.g., dental extractions, biopsies).

A newborn or child presenting with spontaneous intracranial hemorrhage or a large cephalohematoma should be investigated for severe underlying bleeding disorders, such as thrombocytopenia, hemophilia, factor XIII deficiency, other coagulation factor deficiencies, or a severe defect in platelets or VWF (see Chapter 148 ). Trauma-related subdural or intracranial hemorrhage can also be a manifestation of a severe bleeding disorder. In adults, ischemic strokes are more frequent than hemorrhagic strokes (see Chapter 144 ), although hemorrhagic strokes appear to predominate in persons with some types of bleeding disorders (e.g., Quebec platelet disorder ), and they can also reflect complications of antithrombotic drug treatment.

Excessive or prolonged bleeding after childbirth or pregnancy loss can be problematic for some women with bleeding problems. In addition, severe fibrinogen disorders, PAI-1 deficiency, and factor XIII deficiency compromise carrying a pregnancy to term and need to be excluded if the patient has unexplained pregnancy losses that are associated with hemorrhagic placental abruption (see Case 5: Evaluation of an Isolated Symptom—Recurrent Pregnancy Loss With Bleeding ).

Bleeding Score

Objective frameworks for the evaluation of bleeding symptoms have been developed to codify clinical impressions into a numeric score. These bleeding assessment tools (BATs) were initially developed as research tools to quantify bleeding symptoms and severities. Several BATs and their subsequent versions have been refined. They have been studied and validated in specific populations, including those being evaluated for VWD, platelet disorders, symptoms of HMB, and other bleeding syptoms.

Several other tools have been developed with refinements of the original “Vicenza bleeding score” with similar formats: in a grid format, specific bleeding symptoms (e.g., epistaxis, dental hemorrhage) are scored in severity from trivial to severe, and then summed to a final bleeding score. Unfortunately, despite these refinements, the shared weakness of these tools is poor sensitivity (as low as 40%), which has not been greatly improved. Whether these tools are appropriate for use in the clinical setting is still debated, but because of their high negative predictive value, they are probably best for identifying patients who do not require further testing. In addition, the bleeding scores may be useful in predicting future bleeding: in a large, prospective study of VWD, a higher bleeding score is the strongest predictor of bleeding, even over other laboratory and clinical assessments. Whether the bleeding score’s predictive power holds true in other mild bleeding disorders is not yet known, although some studies have not shown the scores to be predictive of future bleeding. Finally, in response to the variability of bleeding scores utilized in clinical studies that make comparing these studies challenging, the International Society on Thrombosis and Haemostasis (ISTH) has proposed a new consensus BAT with the goal of expanding the understanding of bleeding rates and phenotypes more precisely in congenital bleeding disorders. Despite the ease of use and the good negative predictive value of various BATs, the judgement of an experienced clinician remains supreme. BATs have also been used to estimate bleeding risks as likelihood or odds ratios (relative to unaffected relatives or the general population) for experiencing different kinds of bleeding symptoms (e.g., bleeding with surgery, menses longer than 7 days, joint bleeds) and also risks for needing certain types of treatments, such as blood transfusion.

The Physical Examination

A careful physical examination may provide useful clues to the etiology of the patient’s bleeding ( Table 126.1 ). Examining the skin may reveal petechiae, indicating thrombocytopenia, or the characteristic ecchymoses, scarring, and lax skin seen with senile purpura. Patients with scurvy have characteristic perifollicular hemorrhages and “corkscrew” hairs. Telangiectasias around the lips or on the fingertips may signal the presence of HHT. Some patients with vascular Ehlers-Danlos syndrome exhibit characteristic “cigarette paper” skin, which is thin, crepey, and translucent. Bruises should be examined for their number, pattern, and age—if they are all the same color and lividity, they may have all occurred simultaneously. It is important to determine whether the pattern of distribution is indicative of self-infliction, as can be seen sometimes in some patients with Munchausen syndrome. Oculocutaneous albinism is associated with several platelet disorders, including the Hermansky-Pudlak syndrome and the Chédiak-Higashi syndrome (see Chapter 124 ). Splenomegaly can be associated with thrombocytopenia and may indicate underlying cirrhosis. Other stigmata of liver disease, such as spider angiomata, gynecomastia, asterixis, and jaundice, also suggest that the patient may have an associated liver coagulopathy. Joint hypermobility and skin hyperelasticity may be found in hypermobile Ehlers-Danlos syndrome, although not all patients with this disorder manifest the skin findings. A harsh systolic murmur may indicate severe aortic stenosis, which can cause an acquired type 2 VWD, with associated GI bleeding from arteriovenous malformations (see Chapter 133 ). An enlarged tongue, carpal tunnel syndrome, and periorbital purpura may point to amyloidosis (see Chapter 93 ), which can be associated with an acquired deficiency of factor X and many other clotting proteins, including factor V, VWF, α ₂ -antiplasmin, and PAI-1. See Case 6 for an illustration of physical examination leading to clues to the cause of worsening bleeding.