Venous Thromboembolism and Pulmonary Hypertensive Diseases

From WiserWiki

[edit] Venous Thromboembolism and Pulmonary Hypertensive Diseases

Randolph J. Lipchik

Kenneth W. Presberg

[edit] VENOUS THROMBOEMBOLISM

Venous thromboembolism (VTE) remains a common and serious problem that can lead to premature mortality and long-term morbidity. Physicians in almost all areas of patient care will encounter patients at risk for this disease. VTE often occurs in patients with significant underlying medical problems.^[1] In recent studies the overall 1-year mortality rate for patients who developed VTE was 20% to 40%. Cardiac disease, chronic lung disease, and malignancy accounted for the majority of the deaths.^[2] The diagnosis of pulmonary embolism (PE) can be difficult because the signs and symptoms may be similar to the patient's chronic condition.

[edit] Epidemiology

More than 5 million cases of deep venous thrombosis (DVT) occur annually in the United States. Of these patients, approximately 500,000 will have a clinically apparent PE, about 10% of whom will die. The Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) study found that 2.5% of patients who survive to hospitalization will die of PE and that 8% will experience a recurrence in the following months. A small percentage of survivors will fail to resolve the intravascular thrombi and will develop chronic pulmonary hypertension (Fig. 80-1). Current theory stipulates that PE is a complication of venous thrombosis, which is preventable and amenable to early diagnosis and treatment. Despite these advances, the overall incidence of VTE has not decreased, partly because of the increasing age of the U.S. population. Most patients enrolled in the PIOPED study were older than 60. The VTE incidence of 1 in 1000 at age 65 increases to approximately 3 in 1000 at age 85. Unfortunately, many cases still arise because of a failure to provide proven preventive therapy for patients at risk.

Figure 80-1 Estimated incidence of venous thromboembolism and associated complications in the United States.

Approximately 90% of clinically significant PEs arise from the deep veins of the lower extremities; however, not all venous thrombi of the lower extremities or other venous systems pose such a serious risk. Thrombi that remain confined to the calf veins do not cause significant PE. Thrombi may also originate in the pelvic veins, but emboli are less hazardous from these smaller veins. Axillary and subclavian vein thrombosis is more often seen, with central venous catheterization a major risk factor. Axillary thrombus may also arise spontaneously in a young adult from exercise, usually in the setting of thoracic outlet venous compression (Paget-Schroetter syndrome). Clinically significant PE occurs less often from thrombi involving these sites, but all DVT complications of the lower extremities have been described for thrombosis in these other locations.

Data from natural history studies of VTE indicate that the vast majority of patients who die from PE do so within the first few hours of the event, probably from recurrent embolism.^[3] Almost all fatal recurrences are within the first week of presentation. Therefore overall morbidity and mortality can be significantly reduced only by prevention of DVT, and in-hospital mortality can be decreased by early initiation of effective therapy to treat VTE and prevent recurrent PE.

[edit] Pathophysiology

[edit] Risk Factors.

Conditions that promote DVT were first described by Virchow and include vascular intimal injury, blood stasis, and hypercoagulability. Patients can be divided according to the degree of risk and their management directed accordingly (Box 80-1). In the low-risk group, DVT occurs in less than 0.5% of patients. Medium-risk patients are older, with a 2% to 10% incidence of DVT. DVT may occur in 20% of high-risk patients. In many patients, significant risks converge to promote DVT development.

A subset of patients may be predisposed to thrombotic events because of a primary or secondary imbalance in coagulation.^[4] These conditions are generally referred to as hypercoagulable states (Box 80-2). These disorders are uncommon, and diagnostic screening is best directed toward those who are young, have a family history of thrombosis, had unexplained recurrent events, or had abnormal resolution despite adequate anticoagulation. Recently discovered genetic disorders associated with an increased risk of thrombosis include the factor V Leiden mutation, which imparts a resistance to activated protein C, and the prothrombin gene mutation (A²⁰²¹⁰ allele). Some of these genetic disorders are limited to certain ethnic groups, which may partly explain the variability in the incidence of VTE among different populations. Certain disorders, such as the antiphospholipid syndrome, may lead to recurrent devastating complications. Other individuals with inherited thrombophilias may not experience significant complications; a cohort of asymptomatic antithrombin III–deficient patients had a low prevalence of thrombotic events. Anticoagulation is not necessary until patients are at risk or develop a complication. Long-term primary prophylactic anticoagulation has not been beneficial for any of these disorders.

Trousseau first described the well-known association of malignancy with VTE in 1865. A peculiar form of migrating superficial venous thrombosis may antedate the signs and symptoms of cancer by years. A new DVT in an otherwise healthy individual may also be an early clue to the presence of an occult malignancy; however, exhaustive diagnostic tests for an occult malignancy have no proven benefit. Directed examinations of the lung, gastrointestinal tract, breasts, and reproductive organs (the most common sites) are reasonable.

[edit] Complications.

Death from this disease is the result of a PE arising from DVT. PE has multiple respiratory and hemodynamic consequences. Hyperventilation is almost universal and relatively proportional to the degree of vascular obstruction. The exact mechanisms are unknown, but lung mechanoreceptors may play an important role. The embolism leads to vascular obstruction and decreased perfusion, which increases alveolar dead space and ventilatory requirements. The obstruction may be relieved in hours by normal fibrinolysis, and dead space may return to normal shortly after the event. The many causes of hypoxemia include a decrease in mixed-venous oxygen pressure (Po₂) if cardiac output is decreased, ventilation/perfusion ( / ) inequality, and occasionally an increase in right-to-left shunt through a patent foramen ovale. Hyperventilation tends to increase arterial Po₂ (Pao₂) toward normal, limiting the diagnostic utility of Pao₂ for PE. The alveolar-arterial (A-a) gradient, however, remains elevated in almost all patients. Severe hypoxemia in a patient without underlying lung disease signals a massive embolism.

Atelectasis appears after 24 hours of obstruction and partly results from depletion of surfactant secondary to an interruption of nutrient supply to the type II alveolar cells of the lung. Infarction is uncommon because of the dual arterial blood supply of the lung. Hemoptysis and other signs attributed to infarction occur 24 to 48 hours after the acute event but may be the first signs of PE. The hemodynamic consequences of PE may also be life threatening. If the embolism causes more than 50% of the vascular bed to be acutely obstructed, right ventricular afterload will increase precipitously, and right ventricular failure may ensue. The normal right ventricle can sustain adequate blood flow only up to a mean pulmonary artery pressure (PAP) of about 40 mm Hg or a systolic pressure of about 65 mm Hg. Compounds that lead to pulmonary vasoconstriction (e.g., serotonin, thromboxane A₂) are released by platelet aggregation and contribute to PAP elevation. The patient's ability to tolerate these insults largely depends on the underlying cardiopulmonary status, so the cardiopulmonary manifestations of PE can vary greatly among patients. Fortunately, the thrombus begins to resolve within hours after treatment, and restoration of 25% of the luminal diameter allows sufficient flow to normalize a perfusion scan. Much of the obstruction is relieved within days, and resolution is maximal at approximately 8 weeks.

Chronic thromboembolic pulmonary hypertension involving major vessels results from recurrent submassive or massive emboli that do not resolve.^[5] Many of these patients do not have a documented history of PE and did not receive prior treatment. Most do not have recognized hypercoagulable states, but an intrinsic defect in fibrinolysis is strongly suspected; lupus anticoagulant has been found in 10% of these patients. When major vessel thromboemboli do not sufficiently resolve over weeks, the right ventricle adapts over time to the increased afterload, hypertrophies, and may sustain systemic or suprasystemic pressures. These chronic adaptations are usually documented by an electrocardiogram (ECG) or echocardiogram. Patient survival has been correlated to the mean PAP; 5-year survival is significantly decreased when the mean PAP is greater than 35 mm Hg. Patients with a mean PAP greater than 55 mm Hg have a very poor prognosis and are at risk of sudden death. Progressive right ventricular dysfunction is signaled by further increases in right atrial pressure and right-sided cardiac volumes. Hypoxemia in these patients does not correlate with the degree of pulmonary hypertension. Some may be candidates for surgical thromboendarterectomy, whereas others have very limited therapeutic options.

[edit] Patient Evaluation

[edit] History.

The diagnosis of PE is one of the most difficult diagnoses to make on clinical grounds because the clinical signs, symptoms, and basic laboratory studies are neither sensitive nor specific, and the only definitive diagnostic test, pulmonary angiography, is invasive and expensive. The history may strengthen clinical suspicion of PE and with objective tests may allow the physician to rule in or rule out the diagnosis with reasonable certainty. Although PE can occur silently, the sudden onset of dyspnea is usually the most common symptom reported. The multicenter Urokinase Pulmonary Embolism Trial (UPET) identified the most common symptoms associated with angiographically proven PE (Table 80-1). Dyspnea, pleuritic chest pain, and cough were most often seen. The classic triad of dyspnea, pleuritic chest pain, and hemoptysis occurred in only 28% of cases, but two of the three symptoms were present in 65%. Dyspnea, chest pain of any quality, and a sense of apprehension were present in 44%; two of these three symptoms were reported in 81% of patients. Syncope, reflecting inadequate systemic perfusion from obstruction of the pulmonary vasculature, was seen in 13% of patients overall, usually with massive emboli.

Table 80-1 Incidence of Symptoms and Signs Associated with Acute Pulmonary Embolism

✢Urokinase Pulmonary Embolism Trial, 327 patients. Modified from Bell WR, Simon TL, DeMets DL: The clinical features of submassive and massive pulmonary emboli, Am J Med 62:355, 1977.

†Prospective Investigation of Pulmonary Embolism Diagnosis, 117 patients. Modified from Stein PD et al: Clinical, laboratory, roentgenographic, and electrocardiographic findings in patients with acute pulmonary embolism and no preexisting cardiac or pulmonary disease, Chest 100:598, 1991.

In the more recent PIOPED study the subset of patients without prior cardiac or pulmonary disease was examined to identify symptoms that could be attributed solely to the pulmonary embolism (Table 80-1). Dyspnea was present in 73% of cases, pleuritic chest pain in 66%, and cough in 37%. Hemoptysis occurred in only 13% of patients. The lack of specificity of these symptoms was confirmed, since their prevalence was not significantly different in patients proven not to have PE. Since at least 90% of pulmonary emboli arise from the deep veins of the lower extremities, symptoms of DVT should also be elicited. With thrombosis of the iliac, femoral, or popliteal veins, unilateral leg swelling may be the only symptom. In calf vein thrombosis, pain is the most common symptom, but because of multiple veins and collaterals, an isolated clot may be asymptomatic. In UPET, symptoms of lower extremity DVT were present in only 21% of patients. In a small percentage, PE may arise from thrombosis of upper extremity veins, usually in association with a longstanding central line or in cases of trauma. Upper extremity thrombosis is often asymptomatic, but pain and swelling are the most common symptoms, when present. Although nonspecific, the symptoms discussed, in conjunction with the conditions associated with increased risk for venous thrombosis, are important diagnostic clues in the investigation of thromboembolic disease.

[edit] Physical Examination.

In patients with PE the examination can be surprisingly unremarkable. Evidence of acute pulmonary hypertension (right ventricular heave, right-sided S₃, large jugular a wave, increased P₂) is seen infrequently and in the setting of massive embolism. Table 80-1 lists the most common signs of PE. In PIOPED, for patients with no prior cardiac or pulmonary disease, tachypnea was present in 70%, tachycardia in 30%, and rales in 51%. DVT was clinically evident in only 11%. In UPET these signs were present in 92%, 44%, and 58%, respectively. Phlebitis occurred in 32% of patients. The greater incidence of signs in UPET may have resulted from selection criteria. UPET patients were included only if the embolus involved at least one segmental artery, whereas PIOPED patients could have smaller emboli. All these findings are not very specific, however, since patients without PE have a similar incidence of the same signs. The physical findings for lower extremity DVT are also deceptive. Two thirds of DVTs are clinically silent, and in patients with leg swelling, tenderness, or a positive Homans' sign, only half will actually have DVT. Therefore the physical findings alone are not particularly helpful but can be additional diagnostic clues.

[edit] Laboratory Studies

[edit] General Tests.

Again, routine laboratory studies are not particularly specific. Classically, arterial blood gases (ABGs) demonstrate hypoxemia and hypocapnia, the latter resulting from hyperventilation. These findings are supportive evidence but are also found in a variety of other settings. In patients without prior cardiac or pulmonary disease, Pao₂ was the same for patients with and without PE, and 26% of those with angiogram-proven PE had a normal Pao₂ (greater than 80 mm Hg). ECG abnormalities are common with PE and usually nonspecific. In UPET, ST-segment and T-wave abnormalities occurred in 64% of patients. Evidence of acute cor pulmonale (S₁Q₃T₃, right bundle branch block, p pulmonale, or right axis deviation) was less common, but one or more of these occurred in 25% of cases. Sinus tachycardia was present in 43%. Other rhythm disturbances (premature ventricular or atrial beats, atrial fibrillation) occurred in 11% of patients, with atrial fibrillation accounting for 3%. These abnormalities persisted for 5 to 6 days. The ECG was normal in only 13% of patients.

The chest radiograph most often demonstrates findings of atelectasis, small pleural effusion, infiltrates, or elevated hemidiaphragm but may be normal. Other features, such as a pleural-based density (Hampton's hump) and regional loss of vascularity with proximal vascular fullness or cutoff (Westermark's sign), have been associated with PE. Overall, however, the chest film has poor sensitivity and specificity for PE. Its main value is as an adjunct to the / scan and its ability to detect an alternate cause for the patient's symptoms.

[edit] Specific Tests.

Pulmonary angiography is the definitive study for the diagnosis of PE. A catheter is introduced, most often through the femoral vein, and advanced to the main pulmonary artery of interest. PAP is measured and then contrast injected while films from multiple views are taken. Subselective injection of smaller arteries can be done if necessary. A thrombus must be clearly outlined to make the diagnosis of PE. Although definitive, angiography is not suitable as a routine test because it is invasive, expensive, not available at all centers, and has potential complications, such as anaphylactoid reactions to intravenous (IV) contrast and worsening of renal dysfunction, particularly in the diabetic patient. Pulmonary hypertension was once thought to be a contraindication to angiography, but in large centers with experience the procedure can be done safely. This modality is reserved for patients whose diagnosis of PE is uncertain after clinical evaluation and noninvasive imaging studies.

Perfusion lung scanning has been in use for 30 years and is a sensitive but nonspecific method for evaluating pulmonary perfusion. Macroaggregated albumin labeled with technetium 99m is injected intravenously, and anterior, posterior, lateral, and oblique views of the chest are taken. The distribution of particles in the lung reflects the distribution of blood flow. Localized defects can occur in areas of lung consolidation or collapse, with pulmonary vasoconstriction caused by local alveolar hypoxia and vascular obstruction. The chest radiograph is critical in evaluating lung scan perfusion defects. Abnormal perfusion without an abnormality on chest film is much more specific for pulmonary vascular obstruction (embolism) than abnormal perfusion that corresponds to an area of parenchymal consolidation. Ventilation scanning is often performed after the perfusion scan to increase its specificity. Xenon¹³³ is inhaled for several minutes to fill all areas of the lung. The patient then breathes ambient air, and the washout of the isotope is studied. With normal ventilation the lungs clear rapidly and symmetrically. Areas of retained radioactivity indicate abnormal ventilation. Areas of normal ventilation with abnormal perfusion (mismatch) are very suggestive of PE.

The PIOPED trial was unique in that it combined the clinical estimate of the likelihood of PE with the findings of / scanning in patients with angiographically proven PE, thus strengthening the diagnostic utility of lung scanning (Table 80-2).^[6] With high clinical suspicion the probability of PE is as high as 40% when the / scan is read as low probability; however, when the scan is near normal or normal, the chance of PE is very low regardless of the level of clinical suspicion. However, the diagnosis of PE can still be difficult. Approximately 67% of PIOPED patients fell into nondiagnostic categories, and further testing was required. In patients with chronic obstructive pulmonary disease (COPD), up to 90% of scans are nondiagnostic.

Table 80-2 Likelihood of Pulmonary Embolism (PE) Based on Clinical Estimates and Ventilation/Perfusion Scans

Because most pulmonary emboli arise from the lower extremity, techniques for the detection of DVT are important modalities to consider in a PE workup. Contrast venography is the definitive test for detection of DVT but is invasive and requires large amounts of IV contrast material. Noninvasive modalities include ¹²⁵I-fibrinogen scanning, impedance plethysmography (IPG), and venous compression ultrasound. ¹²⁵I-fibrinogen incorporated into freshly forming thrombus was a sensitive method for detecting thrombus formation in calf, popliteal, and distal thigh veins. This test required active thrombus formation so was of little value in the immediate diagnosis of DVT; because of the risk of transmissible infection, it is no longer available for clinical use. IPG assesses venous drainage from the thigh after partial release of an occlusive cuff. Using a mild electric current, the impedance of the thigh is measured after application of an occlusive cuff and should decrease as venous blood is allowed to drain proximally. Failure of impedance to fall suggests obstruction to venous outflow due to a thrombus. Calf thrombi do not usually affect venous outflow so are not detected reliably by IPG, and extrinsic compression of thigh vein(s) or elevated central venous pressure (congestive heart failure) may give false-positive results. Ultrasound examination of the proximal veins of the lower extremity is a sensitive and specific test for DVT. If the vessel is adequately visualized and cannot be compressed, the diagnosis of DVT is confirmed. Visualization of thrombus is unreliable, however, and ultrasound is somewhat operator dependent. The choice of IPG or ultrasound is usually based on local availability. More recent studies have shown that both IPG and ultrasound are much less sensitive than initially reported.

Helical (or spiral) computed tomography (CT) technology has greatly enhanced the diagnostic capability of CT scans. During continuous x-ray exposure the patient moves through the scanner without stopping, often during a single breath-hold. This results in a helix-shaped volume of data over a larger distance. Narrow slices and overlapping reconstruction result in vastly improved spatial resolution and imaging of small vessels. Therefore a relatively small bolus of IV contrast can opacify the pulmonary vasculature, allowing imaging of intravascular thrombi. Multiple studies have documented sensitivities and specificities for PE ranging from 79% to 95% and 93% to 97%, respectively. Helical CT often can provide an alternative cause for the patient's symptoms, which the / scan cannot. The major concern with helical CT is the small, peripheral emboli that are not easily seen. In recent studies, patients who received no anticoagulation after either a normal helical CT scan or a normal to low-probability / scan had similar risks of recurrent PE and mortality. This suggests a good outcome after a negative helical CT scan but requires further study. Whether the CT scan should replace the / scan or substitute for standard pulmonary angiography is the subject of ongoing debate. We favor the use of helical CT for chronically ill and hospitalized patients because they are more likely to have abnormal chest radiographs that result in nondiagnostic / scans. For most outpatients with normal chest films, the / scan should remain the initial imaging study.

The plasma concentration of d-dimer, a fibrin degradation product, is often elevated in patients with thrombosis, but a normal level may eliminate DVT and PE as diagnoses. Using a simple whole-blood d-dimer assay, available in Europe, Canada, and Australia, investigators have found that a normal test may help to exclude PE in patients with a low pretest probability of PE or a nondiagnostic / scan. This would obviate the need for further testing.

[edit] Diagnosis

A chest radiograph should be obtained, and if it does not reveal an alternate diagnosis or is normal, a / scan should follow. If the perfusion scan is normal, PE is essentially ruled out, and therapy or further workup is unnecessary. The lack of subsequent DVT or PE in this setting has been confirmed in several prospective studies. A high-probability / scan indicates an 85% to 90% probability of PE; if combined with a high clinical suspicion, however, the positive predictive value is as high as 96%. In this setting, treatment is initiated without further evaluation. Unfortunately, most patients do not have normal or high-probability scans and fall into more difficult categories. Of PIOPED patients, 73% had low-probability or intermediate-probability / scans, which, combined with different levels of clinical suspicion, indicate PE probabilities of 4% to 66% (see Table 80-2). This degree of uncertainty requires further diagnostic investigation. The first option is to perform pulmonary angiography, which is the definitive test but is invasive and not universally available. Helical CT scanning may be a more reasonable option. The alternative is to study the lower extremity for evidence of DVT. Contrast venography is an option, but for reasons previously mentioned, noninvasive studies have become more prevalent. Finding a proximal DVT with IPG or lower extremity ultrasound warrants therapy, making the diagnosis of PE unnecessary because the therapy is the same. With PE a proximal DVT is found in approximately half the cases.

With a nondiagnostic / scan and a negative noninvasive leg study, probability of PE may still be as high as 30% to 50%, and pulmonary angiography or helical CT is indicated to clarify the diagnosis. Fig. 80-2 shows a diagnostic strategy for the diagnosis of PE that applies equally well for females as well as males. In a separate analysis of female PIOPED patients, signs, symptoms, and risk factors were essentially the same as in male patients, except that oral contraceptive use (not postmenopausal estrogen use) in the setting of surgery was associated with a more frequent diagnosis of PE. In addition, the sensitivity of a high-probability / scan was diminished in this group, and angiography was needed more frequently to confirm the diagnosis of PE.

Figure 80-2 Diagnostic algorithm for venous thromboembolic disease. / , Ventilation/perfusion; CT, computed tomography; PE, pulmonary embolism; DVT, deep venous thrombosis. (From Lipchik RJ, Kuzo RS, Goodman LR: Clin Pulm Med 5:109, 1998.)

[edit] Differential Diagnosis.

Because the symptoms associated with pulmonary embolism are nonspecific, the initial evaluation of the patient with a chest radiograph and general laboratory studies usually identifies alternative diagnoses. For example, chest pain and dyspnea may indicate pneumococcal pneumonia, acute myocardial infarction, or pneumothorax. Fever, elevated white blood cell count, purulent sputum, and segmental infiltrate on the chest film occur with pneumonia. Signs of acute ischemia on the ECG with evidence of congestive heart failure should direct the focus toward ischemic heart disease. The chest radiograph alone should confirm the presence of a pneumothorax. Irritation of the diaphragm by abdominal processes such as pancreatitis or subdiaphragmatic abscess may also produce lower chest pain and a sense of dyspnea, but these can usually be distinguished from PE by physical examination. The failure to find a clear alternative diagnosis and the presence of predisposing factors for thromboembolic disease are the diagnostic clues that increase the suspicion for PE and lead to the next level of investigation.

[edit] Management

[edit] Prophylaxis for Deep Venous Thrombosis.

Key principles for the administration of prophylactic regimens include (1) a well-defined risk group, (2) a simple and easily implemented modality, and (3) a modality that is safe for widespread use. The most frequently used modalities have varying efficacy among the different risk groups^[7] (Table 80-3). Some, such as graded elastic stockings, should be limited to a few select patient groups and have little applicability to high-risk populations. The anticoagulant drugs, including heparin, warfarin, and low-molecular-weight heparin (LMWH), are effective in most high-risk groups. The LMWH enoxaparin is approved for DVT prophylaxis in patients undergoing hip or knee replacement surgery or general surgery. The different LMWHs have specific activities and safety profiles. They need to be individually studied in different clinical settings as they become available. They are not recommended for patients with lumbar puncture or spinal anesthesia and must be used with caution in patients with liver or kidney failure. For patients at high risk for VTE, a combination of intermittent compression devices and an anticoagulant should be strongly considered.

Table 80-3 Deep Venous Thrombosis (DVT) Prophylaxis

✢Reduction in hospital mortality.

[edit] Treatment for Venous Thromboembolism.

Heparin remains the initial treatment of choice for DVT and PE, unless specific contraindications exist, because it immediately inhibits thrombus formation. Heparin binds to antithrombin III, inducing a conformational change in its active center that allows antithrombin III to rapidly inactivate factors IIa (thrombin), Xa, and IXa so that the body's thrombolytic mechanisms can proceed unopposed. Although thrombus formation is arrested, heparin does not prevent embolization of established thrombus. The embolization risk decreases when the thrombus is either dissolved or organized; therefore thromboembolism within the first few days is not a failure of therapy. Heparin's efficacy increases with increasing doses, but at higher doses the risk of bleeding becomes significant. Therapy is usually adjusted to maintain the activated partial thromboplastin time (PTT) in a therapeutic range (1.5 to 2.5 times the control). A delay in achieving a therapeutic PTT or allowing the PTT to fall into a nontherapeutic range increases the risk of recurrent DVT from 7 to 15 times, whereas recurrent thrombus occurs in less than 5% of patients who are adequately anticoagulated. Heparin can be administered intravenously or by multiple daily subcutaneous injections. The IV route is most common and usually involves a bolus followed by a continuous infusion. The dose of heparin is adjusted according to the PTT drawn every 4 to 6 hours until a therapeutic range has been reached; thereafter the PTT is checked daily. Table 80-4 shows one nomogram for managing heparin therapy. Heparin is usually administered for 5 days, overlapping with oral anticoagulants.

Table 80-4 Weight-based Heparin Therapy for Venous Thromboembolism

Anticoagulation is continued for 3 to 6 months on an outpatient basis with coumadin, maintaining the prothrombin time (PT) at approximately 1.5 times control, or an international normalized ratio (INR) of 2.0 to 2.5. The INR, or ratio of patient PT/control PT^ISI (ISI, international sensitivity index for thromboplastin), allows the reporting of therapeutic range to be universally applicable. Coumadin can be started simultaneously with heparin so that after 5 days the INR should be in therapeutic range, allowing the discontinuation of heparin. Coumadin is usually initiated with daily 5-mg doses, with further adjustments made according to the INR. Coumadin should be continued for 3 to 6 months or indefinitely if a predisposing condition persists.

The use of oral anticoagulants alone for the initial treatment of proximal DVT results in three times the rates of thrombus extension and PE compared with IV heparin therapy. When properly diagnosed and treated, recurrence and mortality rates for PE are 8.3% and 2.5%, respectively. The major complications of these medications must also be considered. The major complication of heparin is hemorrhage, which usually occurs with coexistent disease or coagulopathy (e.g., uremia, unsuspected peptic ulcer disease, thrombocytopenia, concomitant aspirin use) but can occur with a therapeutic PTT. Reversible heparin-associated thrombocytopenia (HAT), mediated by a heparin-IgG immune complex, has been noted more frequently with bovine-derived than porcine-derived heparin. With porcine heparin the incidence of HAT is 2.4% for therapeutic heparin and 0.3% for prophylactic heparin. The usual time of onset is between 3 and 15 days from the onset of therapy. The incidence of arterial or venous thrombosis after HAT is 0.4%. Other heparin complications include osteoporosis, alopecia, skin necrosis, and hypoaldosteronism.

As with heparin, warfarin's major complication is hemorrhage. The rare but serious complication of skin necrosis can occur with initiation of therapy and may be caused by a rapid fall in protein C, a vitamin K–dependent inhibitor of coagulation factors Va and VIIIa. A decrease in protein C before a reduction in the other factors results in a transient hypercoagulable state with thrombosis of subcutaneous vessels. Treatment requires discontinuation of warfarin and administration of vitamin K.

The LMWHs approved for use in orthopedic DVT prophylaxis offer an alternative to standard unfractionated heparin in the treatment of DVT and PE. Enoxaparin is approved by the U.S. Food and Drug Administration for DVT therapy. Subcutaneous LMWH in fixed body-weight-adjusted doses may be a better alternative than standard IV heparin in treating lower extremity DVT. No IV administration, no PTT monitoring, and the ease of outpatient therapy are major potential advantages. Considerable cross-reactivity occurs between heparin and LMWH in the setting of HAT, but danaparoid, a heparinoid, has been used successfully in such patients.

Because the greatest risk for PE occurs in patients with proximal DVT, some question whether venous thrombi in the calf need to be treated at all. Studies show good outcomes without anticoagulation in some patients. Using serial IPG of the lower extremities in patients with clinically suspected DVT, patients were not anticoagulated unless they had evidence of proximal extension of thrombus (confirmed venographically). When the IPG remained normal, there were no recurrences of DVT or episodes of symptomatic PE during follow-up. In a similar trial of patients with clinically suspected PE, anticoagulation was not initiated as long as serial IPG studies remained negative for proximal thrombus. At 3 months' follow-up the rate of PE was as low as for patients with normal perfusion scans, supporting the belief that thrombi from the proximal veins of the lower extremities are the usual cause of significant pulmonary emboli.^[8] Whether serial noninvasive studies are practical, cost-effective, and safe remains to be seen.

[edit] Thrombolysis.

Tissue plasminogen activator, streptokinase, and urokinase are all approved for treatment of PE. The only consensus indication for thrombolytic therapy in acute PE is in the setting of submassive or massive embolism and hemodynamic compromise.^[3] Many physicians would administer this therapy only after initial resuscitative efforts have failed to restore adequate hemodynamics. Thrombolysis accelerates clot lysis and improves hemodynamics within hours. Peripheral IV administration may be equivalent to intrapulmonary artery infusion, allowing for lesser delays until treatment. At day 7, however, the effects of thrombolysis plus heparin are equivalent to heparin alone. Furthermore, thrombolysis does not improve mortality in patients with PE.

The use of thrombolytic agents in patients with DVT without PE is more controversial. Many physicians will consider use of these drugs in a patient with acute iliofemoral thrombosis or with vascular compromise (phlegmasia) to improve symptoms and prevent long-term disability. Some studies have shown earlier and more complete recovery of venous flow, but long-term venous competence and decreased morbidity have yet to be demonstrated. The benefits of this treatment in patients with DVT alone need to be balanced with the small but serious risks of major hemorrhage, including intracranial bleeding. The standard of care for proximal DVT remains heparinization followed by warfarin therapy.

[edit] Inferior Vena Cava Filters.

Inferior vena cava (IVC) ligation therapy has been used since the late 1800s. It caused severe venous stasis, however, and development of venous collateral flow may again lead to recurrent PE. Therefore alternative methods were needed to prevent PE when standard therapy had failed or was not feasible. For the last 25 years IVC filters have been in use, most widely the Greenfield filter, although newer designs have been implemented in the past 10 years. These filters can trap emboli as small as 2 mm in diameter and allow for a significant proportion of their volume to be obstructed while preserving a sufficient cross-sectional area for venous flow. Box 80-3 lists the recommended indications for and complications of IVC filters. None of the indications has been substantiated in controlled clinical trials, but the decreased rate of PE compared with untreated DVT supports their use when anticoagulation fails or cannot be administered. IVC filters have also been used successfully in trauma patients who cannot be anticoagulated. Complication rates of modern IVC filters are low.^[9]

[edit] Thromboembolic Pulmonary Hypertension.

Symptomatic patients with mean PAPs greater than 35 mm Hg should have a thorough evaluation to determine if they are candidates for surgical therapy.^[5] The surgical options include pulmonary thromboendarterectomy or lung transplantation. Thromboendarterectomy has a 6% to 15% perioperative mortality risk, but survivors recover sufficiently to return to work or other usual activities. Candidates for this surgery need to have proximal organized vascular thrombosis, as determined by angiography or helical CT scan. Pulmonary angioscopy is available at select centers and allows for direct visualization of thrombus. / scanning invariably shows a segmental or larger defect but usually underestimates the degree of obstruction. Those who are not candidates for endarterectomy can be considered for vasodilator therapy or lung transplantation. All patients should have supplemental oxygen to keep their oxygen saturation above 90% at rest, during sleep, and with exertion. Most physicians recommend lifelong anticoagulation and an IVC filter. Thrombolytic therapy has no proven value once this condition is established.

[edit] Thromboembolic Disease in Pregnancy.

Pregnancy presents unique problems in VTE and its management for a variety of reasons. During pregnancy, increases in clotting factors V, VII, VIII, IX, X, XII, and fibrinogen create a hypercoagulable state; however, this is usually balanced by an increase in baseline fibrinolytic activity. Other factors associated with an increased risk of PE are older maternal age, race (black vs. white), operative delivery, and prior thromboembolism. The diagnostic strategy is as discussed earlier. The total radiation dose to the fetus from a chest radiograph, / scan, and pulmonary angiogram (via the brachial vein) is approximately 0.05 rad, a low dose, particularly balanced against the potential mortality of a PE.

Heparin or an LMWH is the drug of choice during pregnancy for thromboembolic disease because their size precludes transit across the placenta. The daily requirement of IV heparin can be given subcutaneously in two or three daily doses, but monitoring is necessary to ensure that the PTT measured at the midpoint between doses is maintained at 1.5 to 2 times the control. The potential risk of osteoporosis with long-term heparin use becomes a problem in this setting; 20,000 U/day for more than 20 weeks is associated with an increased risk of bone demineralization, but even prophylactic heparin may result in vertebral compression fractions in up to 1.6% of pregnant women. The demineralization is reversible with discontinuation of heparin but may be a slow process. The incidence of osteopenia with LMWH appears to be less.

Warfarin is contraindicated during pregnancy because of well-known teratogenic effects. First-trimester exposure results in a characteristic set of findings, including nasal hypoplasia, depressed bridge of the nose, epiphyseal stippling, and a high rate of developmental impairment. Exposure in the second trimester is associated with central nervous system and ophthalmologic abnormalities. Overall, 13% of pregnancies exposed to warfarin result in abnormal infants. The use of thrombolytic agents is reserved for life-threatening situations and carries the risk of abruptio placentae and fetal death, although there are anecdotal reports of uncomplicated use of thrombolytic therapy during pregnancy.

[edit] PULMONARY HYPERTENSIVE DISEASES

Although flow through the pulmonary circulation usually equals that through the systemic circulation, PAPs are normally much lower than those in the systemic arteries. This difference in arterial pressure reflects the resistance of the pulmonary vessels being proportionately lower than that in their systemic counterparts. The pulmonary circulation has a high compliance because of nonmuscularized, distensible small vessels and capillaries. Therefore significant increases in pulmonary blood flow are accommodated with only slight increases in pressure. According to Poiseuille's equation (law) describing flow through a tube, resistance to flow is inversely proportional to the radius of the tube to the fourth power. Therefore, as vascular remodeling occurs and the pulmonary arterioles are narrowed, resistance may increase precipitously, leading to significant increases in PAP at normal blood flows. For obstructive vascular diseases such as PE, increases in PAP are usually not evident until more than 30% to 50% of the vascular bed has been lost. A diagnosis of pulmonary arterial hypertension is made when the mean PAP is greater than 18 mm Hg. Since extensive pulmonary vascular changes are often present before severe pulmonary hypertension results, many patients will develop symptoms and present at an advanced stage.

[edit] Pathophysiology

Box 80-4 lists disorders associated with pulmonary hypertension, presenting the chronic disorders according to the 1998 World Health Organization Classification of Pulmonary Hypertension. Although thromboembolic disease usually involves macroscopic clots that cause a heterogenous loss of vessels, resulting in abnormal perfusion scans or angiograms, multiple small emboli occasionally may result in a relatively uniform distribution of obstruction. In situ thrombosis is particularly common in patients with sickle cell disease and may also occur with the hypercoagulable states. The hypertensive effects of either the obstructive or restrictive lung diseases may be related to obliteration of pulmonary vessels along with the primary lung damage. Alternatively, hypoxia may contribute to the elevated PAP. Hypoxia is one of the most potent pulmonary vasoconstrictors, and administration of oxygen may acutely reverse some of the PAP elevation. Chronic hypoxemia, however, can lead to vascular remodeling, which may be only partially reversed with time. Chronic high blood flow through left-to-right shunts via atrial or ventricular septal defects may also result in irreversible pulmonary vascular remodeling. Similarly, chronically elevated left atrial pressure from mitral stenosis, atrial myxomas, or diseases that cause increased pulmonary venous resistance may also cause pulmonary hypertension. Collagen vascular disease may cause pulmonary hypertension secondary to interstitial lung disease or by primary vascular involvement. The prognosis is quite poor in patients who develop pulmonary vasculitis or other primary pulmonary vascular disease. Portal hypertension from hepatic or extrahepatic origin is associated with potentially severe pulmonary hypertension. These patients usually do not have hepatopulmonary syndrome, and therefore resting Pao₂ is usually normal.

Primary pulmonary hypertension (PPH) is a diagnosis of exclusion when no other cause can be found. PPH occurs more often among young adult women, but the pathogenesis remains obscure. A significant association exists between the use of fenfluramine-derived anorexic agents and PPH, and fenfluramine has been taken off the market in the United States.^[10] More patients have been exposed to these drugs, and an increased number of PPH cases is expected.

[edit] Patient Evaluation

Unfortunately, most patients with pulmonary hypertension will develop symptoms late in the course. Dyspnea with effort and eventually at rest is the most common complaint of patients with severe pulmonary hypertension. Any persistent complaint of effort intolerance should be taken seriously. Pulmonary hypertension may be associated with substernal chest pain indistinguishable from angina but may result from pulmonary artery (PA) distention rather than from myocardial ischemia. As pulmonary vascular changes progress, resistance increases and cardiac output decreases, leading to further limitation and occasionally syncopal episodes. Hemoptysis is uncommon. As pulmonary vascular resistance becomes excessive, the right side of the heart dilates, and right atrial and ventricular end-diastolic pressure rises. Systemic venous hypertension is required for adequate venous return, and peripheral edema becomes apparent. Other signs of pulmonary hypertension include jugular venous distention and v waves, a palpable PA impulse, and a precordial lift from right ventricular hypertrophy and distention. A loud pulmonic component of the second heart sound and a right-sided S₃ or S₄ may be heard. Holosystolic and diastolic murmurs may be present when tricuspid or pulmonary valvular regurgitation are present, respectively. Flow murmurs over the lung fields have been described in patients with chronic pulmonary hypertension from PE. As cardiac function declines, peripheral edema, cyanosis, and other signs of hypoperfusion may be seen. Raynaud's phenomenon is often seen in patients with collagen vascular disease or PPH.

[edit] Diagnosis

The general diagnostic approach to the patient with suspected pulmonary hypertension is shown in Fig. 80-3. Chest radiographs can provide information regarding any lung disorder and secondary vascular changes. The central pulmonary vessels are often dilated, and a right descending PA greater than 19 mm in diameter suggests pulmonary hypertension. A main PA greater than 30 mm on CT scan is also suggestive. Oligemia can be present in the more peripheral regions of the lung. Disorders that raise pulmonary venous pressure, such as pulmonary venoocclusive disease, may lead to pulmonary edema. Pao₂ at rest may be normal but tends to fall with exercise, and as carbon dioxide pressure decreases, the A-a gradient increases. Hypoxemia stimulates ventilation, and a compensated respiratory alkalosis is a common finding. The alveolar dead space may be relatively high because some ventilated regions of the lungs are underperfused, and it may fail to decrease normally with exercise. Diffusing capacity usually decreases as pulmonary vasculature is lost; however, decreases in diffusing capacity can be mild in patients with PPH or other diseases that primarily affect the pulmonary blood vessels. ECGs are abnormal in most patients. Right axis deviation, right ventricular hypertrophy with strain, and right atrial enlargement are common. Echocardiography is useful because it permits evaluation of the right ventricle and estimation of systolic PAP. Studies with solutions that generate microbubbles, which serve as a contrast agent, help identify the presence of right-to-left shunts. Perfusion scans, helical CT scans, and pulmonary angiography are essential for detecting thromboembolic disease. Cardiac catheterization is frequently needed to rule out congenital heart defects, to obtain accurate hemodynamic measurements, and to evaluate the efficacy of therapy.

Figure 80-3 Diagnostic algorithm for pulmonary hypertension. COPD, Chronic obstructive pulmonary disease; PPH, primary pulmonary hypertension; L → R, left to right; PAP, pulmonary artery pressure; RAP, right atrial pressure; RVEDP, right ventricular end-diastolic pressure. (From Rich S: Prog Cardiovasc Dis 31:205, 1988.)

[edit] Treatment

General treatment options for pulmonary hypertension are summarized in Box 80-5. When pulmonary hypertension is secondary to an underlying disease, treatment of the primary disorder is the treatment of choice. For example, if pulmonary hypertension is caused by hypoxemia, supplemental oxygen may result in lower PAPs, improved right-sided heart function, and enhanced survival. On the other hand, many patients are quite ill at presentation, and treatment options may be limited. Patients with PA hypertension are candidates for vasodilator therapy. Oral and IV vasodilator agents can cause systemic hypotension, which limits their use and requires hemodynamic monitoring in an intensive care unit when they are initiated. Inhaled nitric oxide is used as a screening agent and is selective in that it does not cause systemic hypotension; it can predict acute responses to oral and IV vasodilators. Patients experience symptomatic benefit from vasodilator therapy if their pulmonary vascular resistance decreases by 30% or more.

Recent studies have shown improved survival with titrated high-dose calcium channel blockers or continuous IV epoprostenol (prostacylin) therapy in patients with PPH.^[11] These patients uniformly have significant decreases in both PAP and pulmonary vascular resistance with therapy. Many of these patients who did not improve acutely with therapy did demonstrate improved hemodynamics with long-term IV epoprostenol. Similar benefits have been seen in some patients with scleroderma and other diseases.^[12] Epoprostenol therapy is limited to patients with American Heart Association class III or IV symptoms; it is expensive and requires comprehensive patient training. Warfarin also improves survival in patients with PPH independent of their response to other therapy. Some patients with endstage pulmonary hypertension are candidates for lung transplantation.

[edit] EVIDENCE-BASED MEDICINE

The primary source for this chapter was MEDLINE. Electronic searches dating back to 1994 were conducted up to July 1999, focusing on systematic reviews and randomized trials.

[edit] REFERENCES