Pulmonary Embolism
An obstruction of the pulmonary arteries caused by a blood clot (embolus) or other material carried to the pulmonary vasculature by the circulatory system.
In the USA, pulmonary emboli and its primary cause, deep vein thrombosis, are estimated to lead to 110,000 hospitalizations annually in patients > 65 years. Annual incidence rates per 1000 persons aged 65 to 69 are 1.3 and 1.8 for pulmonary emboli and deep vein thrombi, respectively. Both rates increase with age.
Because the symptoms and signs are nonspecific, pulmonary embolism may be overdiagnosed or underdiagnosed, especially in the elderly. Patients with cardiac and respiratory disorders are especially at risk of misdiagnosis.
Etiology
Although a blood clot is the most common cause of pulmonary embolism, air, fat, bone marrow, foreign bodies, arthroplasty cement, and tumor cells also can obstruct the pulmonary vessels.
Bed rest and inactivity pose the greatest risk for developing deep vein thrombosis. Certain medical conditions common among the elderly (eg, trauma to leg vessels, obesity, heart failure, malignancy, hip fracture, myeloproliferative disorders) predispose them to venous thrombosis, as do smoking, estrogen use, tamoxifen therapy, the presence of a femoral venous catheter, surgery, and immobility. Risk factors for venous thrombosis are vessel wall injury, stasis, and conditions that increase the tendency of the blood to clot, including rare deficiencies of antithrombin III, protein C, and protein S as well as disseminated intravascular coagulation, polycythemia vera, or the presence of a lupus anticoagulant or antiphospholipid antibodies. Aging is also associated with increased coagulation and products of fibrinolysis, resulting in an overall prethrombotic state.
About 90% of blood clots that cause pulmonary embolism originate in the legs. The risk that a clot will embolize and lodge in the lungs is greater if the clot is in the popliteal or iliofemoral vein (about 50%) than if it is confined to the calf veins (< 5%). Less common sites of thrombosis that may lead to pulmonary embolism are the right atrium, the right ventricle, and the pelvic, renal, hepatic, subclavian, and jugular veins.
Symptoms and Signs
In elderly patients, the most common symptoms are tachypnea (respiratory rate > 16 breaths/minute), shortness of breath, chest pain that may be pleuritic, anxiety, leg pain or swelling, hemoptysis, and syncope. Patients who have small thromboemboli may be asymptomatic or have atypical symptoms. Nonspecific symptoms suggestive of pulmonary emboli in the elderly include persistent low-grade fever, change in mental status, or a clinical picture that mimics airway infection.
Patients with pulmonary embolism usually present with one of the following symptom patterns: (1) diagnostically confusing syndromes (confusion, unexplained fever, wheezing, resistant heart failure, unexplained arrhythmias); (2) transient shortness of breath and tachypnea; (3) pulmonary infarction (pleuritic pain, cough, hemoptysis, pleural effusion, pulmonary infiltrate); (4) right-sided heart failure along with shortness of breath and tachypnea secondary to pulmonary embolism; or (5) cardiovascular collapse with hypotension and syncope. Fewer than 20% of elderly patients have the classic triad of dyspnea, chest pain, and hemoptysis. If tachypnea is absent, pulmonary embolism is unlikely.
The most common physical findings are tachypnea, tachycardia, fever, leg edema or tenderness, cyanosis, and a pleural friction rub. Although most elderly patients with pulmonary embolism have deep vein thrombosis as the initial source of the embolus, only 33% have clinical signs of leg thrombosis--eg, leg swelling, tenderness, increased warmth, and Homans' sign.
About 33% of elderly patients with pulmonary embolism have pleural effusions, which are usually unilateral. About 67% of these effusions are bloody (red blood cell count > 100,000/mL) and must be distinguished from cancer and trauma. Patients with pulmonary embolism and a bloody pleural effusion generally have a pulmonary infiltrate on chest x-ray that suggests hemorrhagic consolidation of the lung parenchyma. The infiltrate usually resolves over several days. About 10% of patients with pulmonary emboli, especially those with severe heart failure, develop pulmonary infarction. The remainder of nonbloody effusions due to pulmonary embolism are exudates with elevated white blood cell counts (up to 75,000/mL) that mimic infected pleural effusions.
Syncope, a systolic blood pressure < 100 mm Hg, or a markedly decreased systolic blood pressure in a hypertensive patient suggests the possibility of a massive pulmonary embolism or, in a patient with marginal cardiopulmonary function, a significant embolus. Hypotension is ominous, because decreased aortic diastolic pressure may significantly reduce coronary artery blood flow to the overworked right ventricle, establishing a vicious circle.
A patient who is hypotensive because of pulmonary embolism has elevated right atrial and ventricular pressures (as measured by a pulmonary arterial catheter). Thus, a normal right atrial or ventricular pressure in a patient with hypotension argues against pulmonary embolism as the cause. An echocardiogram can help distinguish pulmonary embolism from right ventricular pressure overload, dissection of the aorta, pericardial tamponade, and myocardial infarction.
Diagnosis
The most important consideration for determining the extent of testing is the clinical assessment of pretest probability. The clinical pretest probability of pulmonary embolism places patients into low-, moderate-, or high-probability groups. This grouping is combined with the results of ventilation-perfusion scans or of spiral chest CT scans (see Laboratory Findings, below) to determine whether further testing is needed.
Patients with a normal ventilation-perfusion scan do not have pulmonary emboli, although about 1% have deep vein thrombosis, as determined by ultrasonography. Patients with a high-probability ventilation-perfusion scan and either a moderate or high pretest probability (about 10% of patients) almost certainly have pulmonary emboli and need treatment. Patients with a high-probability ventilation-perfusion scan but a low pretest clinical probability (about 2% of patients) need to undergo bilateral venous ultrasonography. If the results are positive (about 20% of patients), the patient should receive treatment; if they are negative (about 80% of patients), pulmonary angiography is usually necessary.
If patients with indeterminate ventilation-perfusion scans have a high pretest clinical probability (about 3% of patients), they should undergo venous ultrasonography. If the ultrasonography results are positive (about 30% of patients), anticoagulation therapy is indicated. If results are negative (about 70% of patients), pulmonary angiography is indicated.
Patients with indeterminate ventilation-perfusion scans who have either a low or moderate pretest clinical probability should undergo ultrasonography. If the results are positive, treatment is indicated. If the initial ultrasonography results are negative and there is adequate cardiopulmonary reserve, then anticoagulation therapy is unnecessary. However, serial ultrasonography should be performed on day 3 and possibly on day 7. During a 3-month follow-up, these patients have a rate of venous thromboembolic events of about 1%, which is similar to the rate in patients with normal ventilation-perfusion scans and normal initial ultrasonograms.
Pulmonary angiography should be strongly considered if there is great discordance between clinical suspicion and the ventilation-perfusion scan. In general, the greater the risk of not treating the patient for pulmonary embolism or the greater the risk of therapy, the greater the need for definitive angiographic diagnosis. Of patients with a low pretest clinical probability, a high-probability ventilation-perfusion scan, and a negative ultrasonogram, 33% have an embolus. Of patients with a high pretest probability, an indeterminate ventilation-perfusion scan, and a normal venous ultrasonogram, 50% have an embolus, as detected by pulmonary angiography.
Laboratory Findings
A chest x-ray, an ECG, and arterial blood gas values should be obtained. If pulmonary embolism is still considered likely, the next step is usually to obtain a ventilation-perfusion lung scan. If the lung scan is likely to be indeterminate (because of underlying lung disease), spiral chest CT scans may be useful. Finding deep vein thrombosis with ultrasonography indicates the need for anticoagulation and usually eliminates the need for further testing for pulmonary emboli. The gold standard for diagnosing pulmonary embolism is pulmonary angiography.
Lung scan: Ventilation-perfusion scans are usually the first test done to confirm pulmonary embolism. They are interpreted as normal, high probability, or indeterminate probability. A normal scan shows no perfusion defect and excludes pulmonary embolism. A high-probability scan shows one or more segmental or greater perfusion defects with normal ventilation or two or more large subsegmental perfusion defects (> 75% of a segment) with normal ventilation. High-probability scans indicate a 90% probability of pulmonary embolism. Many scans are, however, indeterminate and thus neither confirm nor rule out pulmonary embolism.
Pulmonary angiography: This test is not needed in everyone. It is invasive, and the dye load can cause severe renal complications. It should be used only when its outcome will definitively change treatment. In patients with renal failure, gadolinium-enhanced magnetic resonance pulmonary angiography is a better alternative because it determines the anatomy without the risk of contrast nephrotoxicity.
Two findings are pathognomonic: a constant intraluminal filling defect and a sharp cutoff of a vessel > 2.5 mm in diameter. A single small embolus may be missed, but multiple emboli are rarely missed. Because most patients with pulmonary embolism have many emboli, the incidence of false-negative pulmonary angiograms is low.
Spiral chest CT scan: Spiral chest CT scans are excellent for detecting pulmonary emboli in the central pulmonary arteries (sensitivity, 94%; specificity, 94%; positive predictive value, 93%; negative predictive value, 95%). Spiral chest CT scans, however, do not detect emboli in subsegmental vessels (which occur in about 6 to 16% of patients). In patients with known cardiopulmonary abnormalities or abnormal chest x-rays, spiral chest CT scans may be preferred to ventilation-perfusion lung scans.
Chest x-rays: Results of chest x-rays may be normal or may show nonspecific abnormalities, eg, atelectasis, an elevated hemidiaphragm, pleural effusion, or an infiltrate. An enlarged pulmonary artery on one side, hyperlucency of one lung because of reduced pulmonary vascular markings, or a pleural-based pyramidal infiltrate that points toward the hilus (Hampton's hump) is uncommon. A chest x-ray cannot establish or exclude a diagnosis of pulmonary embolism but can help diagnose other conditions with similar symptoms (eg, pneumothorax, pneumonia, rib fracture, heart failure).
ECG: ECG findings are usually nonspecific; 33% of patients with pulmonary embolism have a normal ECG. The most common abnormal findings are sinus tachycardia and nonspecific ST-segment and T-wave changes. Uncommon changes that strongly suggest pulmonary embolism indicate strain on the right side of the heart; these changes include T-wave inversion in precordial leads V1 through V4, transient right bundle branch block, new right or left deviation of the QRS axis, sudden onset of atrial fibrillation or other atrial arrhythmia, and ECG signs of right ventricular hypertrophy or right atrial enlargement. The S1Q3T3 pattern (prominence of S wave in lead I, Q wave in lead III, and T-wave inversion in lead III) also suggests pulmonary embolism. This pattern of right-sided heart strain is usually accompanied by T-wave inversion in the precordial leads.
Arterial blood gas studies: Pulmonary embolism often results in arterial hypoxemia because a low ventilation/perfusion ratio develops secondary to airway closure and bronchoconstriction in lung segments adjacent to the emboli. Intrapulmonary shunting of blood and a reduced mixed venous oxygen tension also contribute. Rarely, right-to-left shunting of blood occurs through a patent foramen ovale due to right atrial hypertension from massive pulmonary embolism.
An elevated alveolar-arterial gradient is nonspecific in the elderly population and is of minimal use in the diagnosis of acute pulmonary embolism. Conversely, a normal alveolar-arterial oxygen gradient does not exclude the diagnosis. However, a sudden decrease in partial pressure of arterial O2 (Pao2) that cannot be easily explained by another diagnosis is significant. If tachypnea is present, arterial blood gas values typically show a decrease in partial pressure of arterial CO2 (Paco2).
d-Dimer: Levels of d-dimer, a fibrin-specific product, are increased in patients with acute thrombosis. About 60% of patients < 50 who are suspected of having a pulmonary embolus have an abnormal d-dimer result. In contrast, 92% of patients > 70 have abnormal d-dimer levels, probably due to comorbid conditions. Therefore, if d-dimer test results are negative, deep vein thrombosis or pulmonary embolism is unlikely, but positive test results are not useful in patients > 70.
Venous ultrasonography and venography: For patients with an indeterminate ventilation-perfusion scan, serial ultrasonography performed at 3, 7, and 14 days may help exclude extension of a calf vein clot.
Digital subtraction angiography, MRI, and fiberoptic angioscopy: These tests are under study and are not yet routinely used. Digital subtraction angiography and MRI are less invasive and use less dye than pulmonary angiography.
Prognosis
The mortality rate for hospitalized patients > 65 with pulmonary embolism is 21%. If pulmonary embolism is the primary diagnosis, the mortality rate is 13%; if it is a secondary diagnosis, the rate is 31%. Thus, many diseases and medical conditions--including heart failure, chronic obstructive pulmonary disease, cancer, myocardial infarction, stroke, and hip fracture--greatly increase the risk of death among hospitalized patients > 65 with pulmonary embolism. Prognosis is poorest for patients with severe underlying cardiac or pulmonary disease.
Pulmonary embolism is believed to recur in 5 to 10% of patients despite ongoing heparin therapy. The likelihood of recurrent emboli is greatest in patients who have massive pulmonary embolization or for whom anticoagulant therapy is inadequate. If recurrence develops in the first few days of heparin or thrombolytic therapy, this treatment is usually continued.
In patients > 65 with a pulmonary embolus, the recurrence rate in the first year is 8%, and the 1-year mortality rate is 39% (21% inpatient mortality and an additional 18% mortality during the first year). Elderly patients with deep vein thrombosis but without pulmonary emboli have a 21% mortality rate in the first year. Recurrent pulmonary embolism leading to chronic pulmonary hypertension and cor pulmonale is uncommon.
Right ventricular hypokinesis, as identified by echocardiography, is present in about 60% of elderly patients with pulmonary emboli and a normal systemic arterial pressure. These patients have a twofold to threefold increase in mortality at 2 weeks, 3 months, and 1 year compared with patients with normal right ventricular function. The high incidence of right ventricular dysfunction in elderly patients may contribute to the high mortality rate.
Treatment
Supportive therapy includes providing supplemental O2 to achieve a Pao2 of 60 to 70 mm Hg, providing adequate intravascular fluid to maintain cardiac output, monitoring the patient for evidence of bleeding due to anticoagulant therapy, and avoiding drugs that adversely affect platelet function (eg, aspirin, other cyclooxygenase blockers).
If the patient is hypotensive, treatment includes volume expanders, thrombolytics, and an infusion of norepinephrine to increase aortic diastolic pressure and maintain coronary artery blood flow. In patients for whom thrombolysis is contraindicated or unsuccessful, transvenous catheter embolectomy to fragment the clot should be considered. Rarely, immediate surgery to remove a large clot from a major vessel may be attempted, but the survival rate among elderly patients is low.
Heparin prevents clot formation and extension. Because the risk of death from pulmonary embolism is greatest in the first few hours of development of a clot and because diagnostic test results often are not available for 8 to 12 hours, heparin should be given to patients with a moderate to high clinical probability of pulmonary embolism or deep vein thrombosis until all diagnostic results are available. Low-molecular-weight heparin (LMWH) is preferred to unfractionated heparin. LMWH can be given subcutaneously once or twice a day, and laboratory monitoring may not be necessary.
Long-term anticoagulation is begun in the hospital with heparin and is continued after discharge, usually with warfarin.
Thrombolytic (fibrinolytic) therapy should be considered for patients with deep vein thrombosis involving the iliofemoral system. It is also useful for patients with massive pulmonary embolism who have significant pulmonary hypertension, obstruction of multiple segments of the pulmonary circulation, right ventricular dysfunction, or systemic hypotension.
Clot lysis may reduce the incidence of recurrent thrombi and postphlebitic syndrome and returns pulmonary arterial pressure to normal more quickly than does heparin therapy. Thrombolytic therapy also relieves strain on the right side of the heart more quickly than heparin does. However, thrombolytic therapy does not improve survival. Risks include hemorrhage, including an approximate 1 to 2% risk of intracranial bleeding. The hemodynamic response and the rate of bleeding are similar across age groups.
Contraindications to thrombolytic therapy include eye or central nervous system surgery within the preceding 2 weeks, intracranial neoplasms or vascular abnormalities, stroke within the preceding 2 months, active bleeding, severe hypertension, and allergy to thrombolytic agents. Age is not a consideration.
Interruption of the inferior vena cava--usually with a Greenfield filter--may be required in patients who have a contraindication to anticoagulation; who do not respond to anticoagulant therapy, as demonstrated by recurrent emboli; who have pulmonary emboli from septic thrombophlebitis; or who have massive embolization from a clot in the legs.
Endarterectomy may be helpful in patients who have chronic pulmonary hypertension due to a clot occluding the main or lobar pulmonary arteries.
Prophylaxis decreases the incidence of fatal pulmonary emboli by two thirds in hospitalized patients at risk of developing venous clots. LMWH (eg, enoxaparin 40 mg sc once daily) is as effective and safe as prophylaxis with subcutaneous heparin (5000 IU sc bid or tid) and may reduce drug-induced adverse effects. Postoperative prophylaxis with LMWH (eg, enoxaparin 30 mg sc q 12 h for up to 14 days) also dramatically reduces the incidence of venous thrombosis after knee or hip replacement. For total hip replacement, some investigators find that 4 to 6 weeks of LMWH postoperatively may be more effective.
End-of-Life Issues
In some clinical situations, a decision is made not to attempt prevention, diagnosis, or treatment of deep vein thrombosis or pulmonary embolism. Such a decision should be based on a thorough understanding of the patient's wishes and quality of life. Palliative care at the end of life is discussed in Ch. 13. |
