The initial steps to evaluating bleeding in older adults, like many assessments in medical practice, begin with a careful history and physical examination. Identifying and integrating key components of the patient’s history and physical examination with laboratory data often allows providers to more precisely diagnose the underlying etiology of a bleeding diathesis. Patients should be asked about the initial onset and site of bleeding, prior bleeding episodes, a history of iron deficiency anemia or transfusions, significant bleeding following minor procedures (eg, tooth extraction), and a history of heavy or long menses (in women).
The presence of other family members with similar bleeding tendencies, particularly in the presence of bleeding that began after birth or during childhood, may suggest an inherited bleeding diathesis. A thorough medication history, including the use of anticoagulants, herbal agents, and over-the-counter (OTC) products, should be obtained. Patients may need to be asked specifically about the use of aspirin, nonsteroidal anti-inflammatories, or other OTC antiplatelet agents. To assess for the possibility of dietary vitamin K deficiency, providers should inquire as the patient’s typical diet and whether there have been any changes recently to the diet.
The physical examination may identify petechiae (small capillary hemorrhages most common on the legs, feet, and ankles), small ecchymotic lesions, epistaxal bleeding, or gingival bleeding if the disorder is primary one of platelet number/function or blood vessels. In contrast, older patients with an inherited or acquired coagulation disorder typically present with large, palpable ecchymoses, soft tissue hematomas, hemarthrosis, and extensive postprocedural bleeding.
Laboratory Evaluation of Coagulation Pathways in Older Adult Patients
Laboratory evaluation includes routine tests available through most laboratories as well as more specialized tests that may require a reference laboratory and the expertise of a hematologist for interpretation. General screening tests include a platelet count, hemoglobin and/or hematocrit, prothrombin time (PT), INR, and aPTT. The PT assay adds calcium and thromboplastin (tissue factor) as a platelet substitute to measure the function of the extrinsic and final common pathways (see Figure 106-2). As such, the PT detects deficiencies in the vitamin K–dependent coagulation factors: factors II, VII, IX, and X. If a deficiency in factor VII is suspected, the PT is the most sensitive test available. The aPTT measures components of the intrinsic and common pathway and can detect deficiencies of—or inhibitors to—factors II, V, VIII, IX, X, XI, and XII (see Figure 106-2). The aPTT can also be prolonged in patients with abnormalities in fibrinogen levels. Other routine testing (eg, liver or renal function panels, peripheral smears, disseminated intravascular coagulation panel, etc) may be indicated depending on the patient’s clinical presentation, history, and physical examination.
Causes of a prolonged PT with a normal aPTT
In older patients suspected of having a coagulopathy, examination of any abnormalities in the PT and/or aPTT can help guide clinicians as to the differential diagnostic and appropriate management steps (Table 106-5). Causes of a prolonged PT with a normal aPTT include warfarin or other VKA use, vitamin K deficiency, systemic heparin exposure, factor VII deficiency (inherited or acquired), early liver disease, and in some early cases of DIC.
TABLE 106-5PATTERNS AND EVALUATION OF A PROLONGED PROTHROMBIN TIME (PT) AND/OR ACTIVATED PARTIAL THROMBOPLASTIN TIME (aPTT)* |Favorite Table|Download (.pdf) TABLE 106-5 PATTERNS AND EVALUATION OF A PROLONGED PROTHROMBIN TIME (PT) AND/OR ACTIVATED PARTIAL THROMBOPLASTIN TIME (aPTT)*
|PT RESULT ||aPTT RESULT ||POTENTIAL ETIOLOGY |
|Prolonged ||Normal || |
Intended or surreptitious VKA use
Dietary vitamin K deficiency (mild)
Factor VII deficiencya
Early liver disease
Heparin exposure (high doses)
|Normal ||Prolonged || |
Circulating lupus anticoagulant
Antiphospholipid antibody syndrome
Factor VIII, IX, XI, XII deficiencya
von Willebrand diseasea
|Prolonged ||Prolonged || |
Inherited disorder of common pathway
Advanced liver disease
Systemic fibrinolytic therapy
Dietary vitamin K deficiency (severe)
Combined heparinoid and VKA exposure
Deficiency or inhibitor of factor II, V, Xa
Deficiency or inhibitor to fibrinogena
Causes of a normal PT with a prolonged aPTT
Causes of a normal PT with a prolonged aPTT include anticoagulant use (eg, heparin), deficiencies in factor VIII (hemophilia A, von Willebrand disease), factor IX (hemophilia B), and factor XI. Patients with hemophilia A and B, which are the most common causes, usually present earlier in life with recurrent soft tissue and joint bleeding. Factor XI deficiency has a more variable bleeding history, often following surgical procedures, and has been characteristically described in Ashkenazi Jews but is also reported in other ethnic groups. An isolated prolongation in the aPTT may also be due to the presence of a circulating acquired inhibitor. These disorders include antiphospholipid antibody syndrome, although this disorder causes thrombosis instead of bleeding and usually presents earlier in life. In addition, acquired inhibitors to factor VIII may be seen in older adults with solid tumors (adenocarcinoma being most common among these), hematologic malignancies (chronic lymphocytic leukemia being most common among these), rheumatoid arthritis, systemic lupus erythematosus, and drug reactions (penicillin, phenytoin, interferon, fludarabine, and others). The incidence of an acquired factor VIII inhibitor is rare, reported to be approximately 1.5 cases per million per year in unselected patients from the United Kingdom and Wales. In up to 50% of patients, there may be no underlying, identifiable disorder. Older adults presenting with the abrupt onset of a large spontaneous hematoma or ecchymoses without a history of trauma or prior known bleeding diathesis should prompt consideration by clinicians of an acquired factor VIII inhibitor secondary to an underlying disorder, as described above. Often, this bleeding may be severe with one survey reporting a mortality rate of 22% in these patients.
Causes of a prolonged PT and a prolonged aPTT
Finally, patients with prolongation of both the PT and aPTT should be suspected of having either an inherited disorder of the common pathway or an acquired disorder that affects the intrinsic, extrinsic, and/or common pathways in combination. In older adults, the most common causes of a prolonged PT and aPTT include anticoagulation with a VKA or heparinoid or simultaneous treatment with a VKA and heparin, such as may occur during the treatment of acute venous thromboembolism. Other possible causes include vitamin K deficiency, more advanced liver disease or DIC, and administration of systemic fibrinolytic agents. Additional diagnostic measures in this setting may include measurements of fibrinogen and fibrinogen degradation products.
Abnormal Platelet Number in Older Adult Patients
The normal half-life of circulating platelets in healthy humans is 8 to 10 days. Any process that disrupts this, leading to reductions in platelet number, may increase the risk of, or cause, bleeding in older adults. Quantitative disorders of platelet number in older adults are numerous and some of the more common of these are outlined in Table 106-6. Broadly, the etiology of thrombocytopenia in older adults can be typically thought of decreased platelet production, increased platelet destruction or consumption, increased platelet sequestration, dilutional effects, and spurious thrombocytopenia.
TABLE 106-6COMMON CAUSES OF THROMBOCYTOPENIA IN OLDER ADULTS |Favorite Table|Download (.pdf) TABLE 106-6 COMMON CAUSES OF THROMBOCYTOPENIA IN OLDER ADULTS
ETIOLOGY OF THROMBOCYTOPENIA
Decreased Platelet Production
Nutrient deficiencies (B12 or Folate)
Bone marrow failure (e.g. aplastic anemia, PNH)
Bone marrow infiltration (e.g. tumor, TB, Sarcoidosis)
Prescription or over-the-counter medications
Chronic alcohol abusea
Acute or chronic infectionb
Increased Platelet Destruction
Immune thrombocytopenic purpura
Acute or chronic infectionb
Systemic lupus erythematosus
Disseminated intravascular coagulation
Mechanical prosthetic valve
Thrombotic thrombocytopenic purpura
Antiphospholipid antibody syndrome
Chronic alcohol abusea
Pulmonary embolism or hypertension
Spurious (e.g. pseudothrombocytopenia)
Dilutional (e.g. in setting of trauma with massive transfusion)
Decreased production of platelets is most often encountered in patients with disorders that impair bone marrow thrombopoiesis. These disorders commonly include nutrient deficiencies, myelodysplastic syndromes, and acute or chronic infectious syndromes. These are particularly common in older adults. For example, vitamin B12 and folate deficiency can be found in up to 24% and 10% of older adults, respectively. Similarly the median age at diagnosis for patients with myelodysplastic syndromes exceeds 65 in most studies, with an onset of disease earlier than 50 being very uncommon. Finally, older adults are at markedly increased risk of infection from bacterial, viral, and fungal causes—many of which may result in bone marrow suppression and associated thrombocytopenia. Many of these disorders are covered in more detail in other chapters within this book. Increased destruction or consumption can be a result of accelerated, antibody-mediated clearance from medications, ingested substances, or an autoimmune process.
Primary Immune Thrombocytopenia
Primary ITP (previously also referred to as idiopathic thrombocytopenic purpura or immune thrombocytopenic purpura) is characterized by persistent thrombocytopenia (platelet count < 150 k/μL) as a result of antibodies (mostly IgG and less commonly IgM and IgA) produced against platelet membrane glycoprotein GPIIb-IIIa. Antibody-coated platelets are then rapidly cleared and destroyed by the reticuloendothelial system via recognition by Fc-receptors on macrophages and dendritic cells. These antibodies may also result in defective platelet production. ITP is a commonly encountered cause of thrombocytopenia in adults. The prevalence of ITP, which rises with age, varies with different populations, ranging between 9 and 23 cases per 100,000. In patients over the age of 65, there is an equal incidence between men and women.
Many patients (20%–30%) will be asymptomatic and the thrombocytopenia will be noted incidentally. Recent guidelines recommend that a platelet count of less than 100 k/μL should be the threshold for a diagnosis of ITP. Further classifications can include newly diagnosed (< 3 months duration), persistent (3–12 months duration), and chronic (> 12 months duration). For patients with symptoms, the most common presentation is bleeding in the skin or mucous membranes, ranging from the rare severe hemorrhage to the more commonly encountered mild petechial or purpuric lesions on the extremities and epistaxis. Symptom onset is usually insidious but occasionally can be abrupt.
ITP is a diagnosis largely made by exclusion. In ITP, the peripheral blood smear is usually normal except for the identification of low numbers of platelets with the occasional large-appearing platelets. However, platelet morphology is normal without identifiable defects in intracellular platelet granules or uniformly small or large platelets. While the evaluation for ITP in older adults may include a thorough history and physical, complete blood cell counts, reticulocyte counts, peripheral blood smear, and a direct antiglobulin test, a low platelet count may be the only identifiable abnormality. Often, diagnostic tests are usually helpful only in excluding other potential etiologies of the thrombocytopenia. Bone marrow biopsies may be undertaken in select patients but are typically not routinely required.
Secondary Immune Thrombocytopenia
Thrombocytopenia due to other underlying immune-mediated disease processes are common in older adults. For example, medications can also induce thrombocytopenia, particularly in patients with recurrent acute thrombocytopenia not otherwise explained by other processes. While the timing, severity, and duration of drug-induced thrombocytopenia can vary substantially, in many patients thrombocytopenia will be first evident within 1 week of initiation of the causative drug (Figure 106-3). Common drugs associated with thrombocytopenia are shown in Table 106-7 and an exhaustive list of drugs linked to low platelet counts can be accessed online (http://www.ouhsc.edu/platelets). Providers may also consider reporting cases of drug-induced thrombocytopenia to the US FDA (or similar organizational bodies in other countries) Adverse Event Reporting System at www.fda.gov/medwatch.
Categorization of drug-induced thrombocytopenia by time of onset of the thrombocytopenia. (Adapted from Kenney B, Stack G. Drug-induced thrombocytopenia. Arch Pathol Lab Med. 2009;133(2):309–314. With permission from Archives of Pathology & Laboratory Medicine. Copyright 2009 College of American Pathologists.)
TABLE 106-7COMMON MEDICATIONS IMPLICATED AS CAUSES OF DRUG-INDUCED THROMBOCYTOPENIA |Favorite Table|Download (.pdf) TABLE 106-7 COMMON MEDICATIONS IMPLICATED AS CAUSES OF DRUG-INDUCED THROMBOCYTOPENIA
|Abciximab ||Hydrochlorothiazide |
|Acetaminophen ||Ibuprofen |
|Cilastatin/imipenem ||Inamrinone |
|Clopidogrel ||Interferon-γ |
|Cyclosporine ||Linezolid |
|Diazepam ||Oxaliplatin |
|Dactinomycin/actinomycin ||Phenytoin |
|Diclofenac ||Piperacillin |
|Digoxin ||Quinine |
|Dipyridamole ||Ranitidine |
|Eptifibatide ||Rifampin |
|Famotidine ||Tirofiban |
|Fludarabine ||TNF-α |
|Furosemide ||Trimethoprim/sulfamethoxazole |
|Heparinoids ||Valproic acid |
|Hepatitis B vaccine ||Vancomycin |
In hospitalized patients, heparin is one of the most common causes of thrombocytopenia and in some cases may result in marked increases in the risk of both arterial and venous thrombosis (eg, HIT). The diagnosis of HIT can be challenging and providers are encouraged to use validated clinical prediction rules in combination with the results of serologic testing and/or functional platelet assays.
Acute and chronic infections, including hepatitis C, HIV, influenza, and Helicobacter pylori may also cause a secondary immune thrombocytopenia, although precise mechanisms and pathways remain incompletely understood. In some of these infectious syndromes, thrombocytopenia can also result from ineffective thrombopoiesis as a result of bone marrow infiltration and/or megakaryocyte infection. Finally, secondary ITP can also be a complication of rheumatologic and autoimmune disorders, including systemic lupus erythematosus (SLE), antiphospholipid antibody syndrome, IgA deficiency, and common variable immunodeficiency (CVID).
In addition to immune-mediated thrombocytopenias, low platelet counts in older adults can be the result of other, nonimmune causes. One of the most commonly encountered causes is spurious thrombocytopenia, also referred to as pseudothrombocytopenia, being identified in about 0.1% of all blood samples in adults. Spurious thrombocytopenia is due to exogenous platelet clumping in the presence of ethylenediaminetetraacetic acid (EDTA), which is found within some vacutainer tubes. This can be identified by examining the blood film for platelet clumping and repeating a platelet count using blood drawn into an alternative anticoagulant such as citrate buffer. In addition, enhanced platelet consumption can occur in processes such as DIC and thrombotic thrombocytopenia purpura and hemolytic uremic syndromes (TTP-HUS). Dilutional thrombocytopenia may be seen in trauma patients who have undergone massive fluid resuscitation protocols that include large number of packed RBC transfusions.
Abnormal Platelet Function in Older Adult Patients
Bleeding in older adults may stem not from absolute reductions in platelet count but rather from inhibition of normal platelet functions (eg, activation, adhesion, and aggregation) that impair hemostasis. Traditional antiplatelet agents, including aspirin, nonsteroidal anti-inflammatories, clopidogrel, ticagrelor, dipyridamole, and cilostazol may all increase the risk of bleeding in older adults. The relative risk increase for bleeding varies among different agents and between different patients. A number of other medications commonly prescribed to older adults may also inhibit platelet activation or aggregation. These include nitrates, calcium channel blockers, and selective serotonin reuptake inhibitors (SSRIs). Many of the mechanisms underlying these drugs’ effects on platelet function remain incompletely understood.
Acquired abnormalities of platelet function in older adults may be due to acute or chronic disease processes. For example, liver disease may cause abnormalities in platelet aggregation even in the absence of absolute thrombocytopenia. Similarly, acute and chronic kidney disease has been associated with various pathophysiologic effects on platelet function. For example, patients with renal dysfunction have abnormal platelet aggregation responses to adenosine diphosphate (ADP) and epinephrine as well as impaired platelet adhesion. While the underlying etiologies of this platelet dysfunction is likely multifactorial, kidney disease does appear to cause defects in platelet glycoprotein IIb/IIIa, an integral membrane protein mediating both platelet aggregation and adhesion. Interestingly, recent evidence also suggests that uremia is associated with differential changes in the platelet proteome that may contribute the bleeding seen in older adults with kidney disease.
Anticoagulant Use in Older Adults
Anticoagulants, including VKAs, TSOACs, heparinoids, and others, increase the risk of bleeding. Additional information on these various anticoagulants and their use in specific disease settings (eg, atrial fibrillation, stroke) is beyond the scope of this chapter but can be found in other chapters in this book.
While bleeding rates may vary substantially based on the population, disease, and anticoagulants studied, in general 2% to 3% of patients on anticoagulants will have a major bleeding complication each year and another 14% experience minor bleeding. Compared to younger subjects, older adults have an increased risk of anticoagulant-associated bleeding. This relative risk increase depends to some degree among the patient group being studied, the anticoagulant class and intensity prescribed, and the presence or absence of comorbidities such as kidney disease, uncontrolled hypertension, a history of bleeding, liver disease, concomitant antiplatelet agent or nonsteroidal anti-inflammatory drug (NSAID) use, and others. While the magnitude of the risk varies, in general the risk of bleeding increases approximately 40% for every 10-year incremental age increase.
This increased risk may present a challenging clinical dilemma for practitioners as older patients, while having an increased bleeding risk, also have a significantly higher risk of thromboembolic disease, including venous thromboembolism, cardiovascular disease, and stroke, compared to younger subjects. As a result, providers may be more hesitant to anticoagulate older patients, even in the absence of contraindications, due to the perception that the risks of treatment outweigh the benefits. This hesitation may be reflected in utilization studies where 50% or fewer older adults with atrial fibrillation are prescribed OACs for stroke prevention. In addition to age being cited as a deterrent to OAC prescription, other reasons include concerns with compliance, monitoring, falls, cognitive impairment, decreased drug clearance, and polypharmacy. A recent systematic review of 30 studies reinforced that these and other patient factors remain significant obstacles to OAC use. Each advancing decade of life is associated with a 14% reduction in OAC use (especially VKAs) and, additionally, patients may frequently refuse anticoagulation due to their fears of bleeding. As one example, approximately one-third of atrial fibrillation patients enrolled in the AVERROES and ACTIVE A studies declined to take a VKA partly as a result of their perceptions of bleeding risk.
Compared with VKAs, TSOACs such as dabigatran, rivaroxaban, apixaban, and edoxaban, offer some safety and efficacy advantages. A growing body of literature also demonstrates that TSOACs generally have a favorable profile in older adults as well, although there are some differences between the various TSOACs. The reader is referred to several recent reviews for more information on this topic.
While bleeding risk increases with age, TE risk rises as well. In many older adults, there is an overall net clinical benefit with appropriate OAC prescription. This has been most rigorously examined in older adults with atrial fibrillation. Typically, the “net clinical benefit” is defined as the annual rate of ischemic strokes prevented by warfarin minus the annual rate of intracranial hemorrhage (ICH) caused by warfarin, the latter multiplied by a factor of 1.5 to reflect the greater clinical severity of ICH than ischemic stroke (and the greater impact of OAC on TE risk compared to ICH risk). The net clinical benefit of OAC increases progressively with age and patients 85 years and older may have the greatest benefit. Reflective of these and other emerging data, many guidelines, recommendations, and clinical practice patterns in OAC prescription and management are shifting away from a focus on identifying high-risk patients who should be treated with OAC and toward identification of truly low-risk patients in whom OAC may safely be deferred.
Management of Bleeding Risk in Older Patients Prescribed OACs
As noted above, recent changes in antithrombotic management emphasize a careful bleeding risk assessment in patients considered for OAC. To assist providers in this assessment, various bleeding risk assessment tools have been published. These tools have been validated with VKAs but not yet with the TSOACs. A recent review concluded that the HAS-BLED score (Tables 106-8 and 106-9) provides the best risk estimation for major bleeding risk, at least in atrial fibrillation (AF) patients treated with OAC. Nevertheless, the performance of the HAS-BLED was only modest (c-statistic 0.58–0.80) and remains to be validated with the TSOACs. These bleeding risk assessment tools are also limited in that the majority of predicted bleeds are extracranial with low mortality rates and disability. As a result, tools such as the HAS-BLED should not be used to exclude patients from OAC, but rather to identify and eliminate modifiable bleeding risk factors (Table 106-10).
TABLE 106-8THE HAS-BLED BLEEDING RISK SCORE |Favorite Table|Download (.pdf) TABLE 106-8 THE HAS-BLED BLEEDING RISK SCORE
| ||RISK FACTORS ||SCORE |
|H ||Hypertensiona ||1 |
|A ||Abnormal liverb or kidney functionc ||1 or 2 |
|S ||Stroke ||1 |
|B ||Bleedingd ||1 |
|L ||Labile INRe ||1 |
|E ||Elderlyf ||1 |
|D ||Drugsg or alcoholh ||1 or 2 |
TABLE 106-9THE NUMBER OF ESTIMATED BLEEDING EVENTS PER 100 PATIENT-YEARS BASED ON THE HAS-BLED RISK CATEGORY |Favorite Table|Download (.pdf) TABLE 106-9 THE NUMBER OF ESTIMATED BLEEDING EVENTS PER 100 PATIENT-YEARS BASED ON THE HAS-BLED RISK CATEGORY
|SCORE ||RISK CATEGORY ||BLEEDS/100 PATIENT-YEARS |
|0–1 ||Low ||1.2–2.8 |
|2 ||Moderate ||3.6–5.4 |
|≥ 3 ||High ||6.0–9.5 |
TABLE 106-10MODIFIABLE BLEEDING RISK FACTORS IN OLDER ADULTS PRESCRIBED ANTICOAGULANTS AND POTENTIAL STRATEGIES TO REDUCE THE RISK |Favorite Table|Download (.pdf) TABLE 106-10 MODIFIABLE BLEEDING RISK FACTORS IN OLDER ADULTS PRESCRIBED ANTICOAGULANTS AND POTENTIAL STRATEGIES TO REDUCE THE RISK
|RISK FACTOR ||POSSIBLE INTERVENTION(S) |
|Concomitant NSAID use ||Replace NSAID with acetaminophen or selective COX-2 inhibitor |
|Concomitant use of an antiplatelet agent ||Eliminate antiplatelet agent if not indicated |
|Highly variable time in therapeutic range (TTR)a ||Consider referral to anticoagulation clinic, if available, or switch to TSOAC |
|Frequent falls or high risk of falls ||Consider evaluation by Faint & Fall Clinic or geriatric specialist |
|Uncontrolled hypertension ||Control blood pressure to appropriate goal |
|History of gastrointestinal bleeding ||Consider proton pump inhibitor (PPI) or H2 antagonist |
|Poor adherence or compliance ||Education; referral to anticoagulation clinic if available |
In addition, the real and/or perceived risk of traumatic ICH from falls in older adults may be viewed as a contraindication to OAC use. While fall risk assessment in older adults is an important aspect of geriatric care, studies of AF patients on OAC who have a high fall risk have not consistently demonstrate an increased risk of traumatic ICH. An older modeling study suggested that AF patients would have to fall more than 295 times annually before the risk of traumatic ICH would exceed the risk of ischemic stroke. However, in older patients who do fall, the average number of falls they sustain each year is fewer than two and many recommend that high fall risk by itself should not prevent OAC use if the benefits otherwise outweigh the risks. For clinicians with significant concerns about individual patients, a referral to a faint and fall clinic or a comprehensive fall risk assessment should be considered.
Other strategies to reduce the risk of bleeding in older adults prescribed OACs include a thorough medication reconciliation to identify concomitantly prescribed medications that might interact with an OAC or increase the risk of bleeding. In addition, patients, family members, and caregivers should receive comprehensive education on OACs as insufficient education may increase the risk of bleeding. Referral to a Thrombosis or Coumadin Clinic for education and monitoring may improve compliance and the therapeutic time in range (if on a VKA) while also preventing avoidable medication interactions. These specialty anticoagulation clinics may also have a valuable role in following patients on TSOACs by providing regular patient follow-up, counseling, education, and laboratory monitoring (eg, for early identification of renal dysfunction where TSOAC dose adjustments may be needed).