Management of Pulmonary Embolism: 2010 State-of-the-Art Update
The morbidity and mortality of venous thromboembolism remain underrecognized and underappreciated. Suspected pulmonary embolism should be risk stratified using a validated clinical risk prediction tool; intermediate to high clinical suspicion requires objective diagnostic testing to confirm or refute the diagnosis. Therapy with unfractionated heparin, low molecular weight heparin, or fondaparinux should be initiated while diagnostic testing is pursued. Conversion to vitamin K antagonists requires a minimum of 5 days’ overlap between the parenteral agent and the vitamin K antagonist. Anticoagulation should be continued for a minimum of 3 to 6 months. Longer or even indefinite therapy may be required with a persistent hypercoagulable state. In patients with cancer, low molecular weight heparin monotherapy for the initial 3 to 6 months is preferred. In stable patients with normal biomarkers and a normal echocardiogram, accelerated discharge and outpatient therapy may be considered. In patients with hemodynamic instability, systemic thrombolytic therapy, catheter-directed therapy, or surgical embolectomy may be considered. Cancer screening and/or thrombophilia testing should be pursued only if the findings will directly affect patient therapy or long-term care.
Venous thromboembolism (VTE), including deep venous thrombosis (DVT) and pulmonary embolism (PE), is a common cause of cardiovascular morbidity and mortality. Despite the recognition that 30% to 60% of patients with DVT have PE at the time of diagnosis, most studies dichotomize DVT and PE, with or without DVT, into separate clinical entities [1–3]. The annual incidence of VTE has been estimated at 300,000 to 600,000 cases [4,5]. The incidence is similar among races and between genders [6,7], with an age- and gender-adjusted annual incidence of approximately 1.17 per 1,000 individuals . Older patients are at increased risk [6–8], and although some studies have suggested the incidence of VTE in hospitalized adults is stable , others have suggested an increasing incidence . As the population ages and medical care becomes more complex, increased incidence may be expected.
The signs and symptoms of DVT and PE are notoriously nonspecific. All cases of VTE should be objectively documented using reliable diagnostic modalities. Patients may be asymptomatic; have nonspecific signs and symptoms, such as fever, tachycardia, pleuritic chest pain, or dyspnea; or present with cardiogenic shock or cardiac arrest. Upon presentation, patients with suspected PE should be risk stratified into low, intermediate, or high pretest probability for VTE using validated clinical models [9, 10]. Further diagnostic strategies using D dimer, CT angiography, or ventilation–perfusion imaging should be used to confirm or refute the diagnosis .
Absolute and relative contraindications to anticoagulation
Known sensitivity to the proposed therapy
Active peptic ulcer disease
Recent organ biopsy
Arterial trauma or puncture of a noncompressible arterial structure
Gastrointestinal or genitourinary bleeding within the past 10 days
Severe thrombocytopenia or severe anemia
Known bleeding diathesis
Surgery, major stroke, or major trauma within the past 2 weeks
History of intracranial, spinal, or ophthalmic bleeding
Severe uncontrolled hypertension
Severe liver disease
Patients with confirmed PE require hospitalization. Increased 3-month mortality is observed in patients older than 70 years and those with cancer, congestive heart failure, chronic obstructive pulmonary disease, systolic blood pressure less than 90 mm Hg, or right ventricular hypokinesis . The Pulmonary Embolism Severity Index (PESI) is a validated prediction rule that uses 11 clinical characteristics to stratify patients according to mortality risk associated with PE [14,15]. For patients in the lowest-risk classes (I and II), 30-day all-cause mortality was less than 4%, whereas for patients in the highest-risk class (V), mortality was greater than 20% . Elevated cardiac troponin and brain-type natriuretic peptide (BNP) levels may indicate right ventricular dysfunction. Normal troponin and BNP have a high negative predictive value for in-hospital death and may be used to stratify low- versus intermediate-risk patients . Outpatient therapy of PE has not been widely advocated; however, several authors have outlined options for accelerated discharge in clinically stable, low-risk patients without right-heart strain or injury [17,18]. Other authors have advocated outpatient management in patients in PESI risk class I or II who are at low risk for poor outcomes [15,19].
Risk factors for thrombosis
Factor V Leiden gene mutation
Prothrombin gene mutation
Protein C deficiency
Protein S deficiency
Factor VIII excessa
Plasminogen activator inhibitor excessa
Polycythemia rubra vera
Recent surgery, hospitalization, immobilization (eg, plaster cast)
Advanced age (>40 years)
Hormone replacement therapy
Selective estrogen receptor modulators
Acute inflammatory states
Infection, systemic inflammatory response syndrome, sepsis
Inflammatory bowel disease
Disseminated intravascular coagulation
Heparin-induced thrombocytopenia with or without thrombosis
Paroxysmal nocturnal hemoglobinuria
No trial or model has evaluated whether patients in the highest PESI class with an associated mortality greater than 20% benefit from intensive care unit (ICU) admission. In general, patients with systemic hypotension, demonstrated right ventricular dysfunction, elevated cardiac biomarkers, or significant supplemental oxygen requirements should be evaluated for ICU admission .
Volume resuscitation and vasopressor support may be required in patients with massive PE. Judicious volume support is appropriate; however, volume overload may increase right ventricular ischemia and decrease cardiac output and therefore should be avoided. Norepinephrine is the preferred vasopressor for persistent hypotension in massive PE .
Diet and lifestyle
Patients treated with vitamin K antagonists (VKAs) should consume consistent dietary vitamin K. Many medications and herbal products influence therapy . Patients should report all medication changes, over-the-counter drugs, herbal therapy, and vitamins.
Patients should be cautioned regarding an increased bleeding risk while receiving anticoagulant therapy (Table 1). Patients should avoid high-risk activities likely to cause trauma, bleeding, or extensive bruising. Assuring recognition of the potential bleeding risk by emergency medical service personnel through a medical alert necklace, bracelet, or wallet card may be lifesaving.
UFH, LMWH, or fondaparinux should be initiated while awaiting diagnostic testing. LMWH or fondaparinux may be preferable to UFH because of predictable bioavailability, unmonitored administration, and subcutaneous dosing [13••,25,26].
VKAs may be administered on the first day of treatment along with parenteral therapy. A minimum of 5 days’ overlap with a stable international normalized ratio (INR) greater than 2.0 on 2 consecutive days is required when converting from parenteral to oral therapy. Monitoring and dosing of the VKA should be adjusted to maintain a target INR of 2.5 with a range between 2 and 3 [13••].
In patients with cancer, LMWH monotherapy is preferred for the initial and long-term management of VTE [13••].
The rates of major hemorrhage in clinical trials using UFH, LMWH, or fondaparinux are 1% to 2% [25–27]. Major hemorrhage with VKAs occurred in 2% of patients within the first 3 months of treatment and in 1% per year thereafter .
Thrombolytic therapy should be considered in patients with massive PE (defined by hemodynamic compromise) unless there are major contraindications owing to bleeding risk [13••].
UFH, LMWH, and fondaparinux are used for initial therapy. These drugs bind to and activate antithrombin (AT) via a specific pentasaccharide sequence. Once bound, the natural anticoagulant activity of AT is increased by several hundred-fold.
Unfractionated heparin sodium
UFH is a mixture of glycosaminoglycans derived from bovine lung or porcine intestine. Approximately 30% of the molecules contain the pentasaccharide sequence that binds and activates AT. The UFH–AT complex accelerates the clearance of serine proteases, predominantly factor Xa and thrombin [29•].
Low molecular weight heparin
Enzymatic or chemical cleavage of UFH produces a mixture of low molecular weight glycosaminoglycans, also called LMWH. Most of the AT–LMWH complexes are too small to bind thrombin. Therefore, anti-Xa activity predominates. Compared with UFH, LMWH has superior bioavailability; fewer platelet, cellular, and protein interactions; a more predictable clearance; and a longer half-life that allows for once- or twice-daily subcutaneous (SC) administration [29•].
Fondaparinux is a synthetic pentasaccharide analogue that binds and activates AT. The AT-fondaparinux complex inhibits only factor Xa. It has excellent bioavailability (100%) and a long half-life (∼17 h), which allows once-daily dosing.
- Unfractionated heparin
Adjusted dose, activated partial thromboplastin time (aPTT)-monitored SC UFH therapy: 250 U/kg every 12 h with midinterval (6-hour) aPTT monitoring has been used historically [29•].
In one trial, unmonitored SC UFH given as a 333-U/kg initial dose followed by 250 U/kg every 12 h was noninferior to SC weight-based enoxaparin sodium in patients with DVT .
IV UFH and aPTT-adjusted SC heparin require anticoagulation monitoring. We recommend an aPTT range standardized to an anti-factor Xa assay with a target range of 0.3 to 0.7 U/mL [29•].
- Low molecular weight heparin
Enoxaparin sodium: 1 mg/kg SC every 12 h (US Food and Drug Administration [FDA] approved for outpatient treatment of DVT without PE); 1.5 mg/kg SC daily (FDA approved for inpatient treatment of DVT with or without PE).
Dalteparin sodium: 100 IU/kg SC every 12 h (not FDA approved); 200 IU/kg SC daily (FDA approved for monotherapy of symptomatic VTE in patients with cancer). After 4 weeks, the dosage may be decreased to 150 IU/kg SC daily.
Tinzaparin sodium: 175 IU/kg SC daily (FDA approved for treatment of DVT with or without PE).
Fondaparinux (FDA approved for the treatment of DVT and PE) is dosed by body weight: less than 50 kg, 5 mg SC daily; 50 to 100 kg, 7.5 mg SC daily, greater than 100 kg, 10 mg SC daily.
Absolute and relative contraindications and other considerations related to anticoagulation are listed in Table 1. Active heparin-induced thrombocytopenia (HIT) or a history of HIT is an absolute contraindication for administration, continuation, or initiation of UFH or LMWH. Parenteral anticoagulants must be used cautiously in the setting of pending or recent neuroaxial anesthesia and should be avoided in patients with recent (<48 h) traumatic epidural or spinal catheter placement.
Main drug interactions
Caution must be exercised when using drugs known to increase the risk of bleeding (ie, antiplatelet agents, aspirin, clopidogrel, ticlopidine, and nonsteroidal anti-inflammatory drugs) concomitantly with other anticoagulants. The use of glycoprotein IIb/IIIa inhibitors or thrombolytic agents should be individualized.
Main side effects
Bleeding, either overt or subclinical, is the main adverse side effect of UFH therapy. Patients should be monitored closely for visible bleeding, hypotension, decreasing hematocrit, back pain, leg pain, or thigh weakness, which may be associated with subclinical blood loss. Immune-mediated HIT occurs in 3% to 5% of patients receiving UFH and approximately 1% of those receiving LMWH. Thrombocytopenia with thrombosis, similar to HIT, has been documented with fondaparinux . Platelet counts should be monitored every other day during UFH therapy and weekly with LMWH and fondaparinux therapy. Injection site pain, irritation, hematoma, and ecchymosis are associated with SC administration. In addition, reversible elevations in liver transaminases are seen with all three agents. Osteopenia and osteoporosis may occur with prolonged administration of UFH and less commonly with LMWH.
The half-life of UFH is approximately 60 to 90 min. UFH may be preferred if there is an anticipated need to hold or stop therapy. UFH is eliminated by the reticuloendothelial system and is preferred over LMWH or fondaparinux in patients with severe renal disease (creatinine clearance [CrCl] <30 mL/min). All three drugs are considered pregnancy class B. Protamine sulfate offers a reliable antidote for over-anticoagulation with UFH. One milligram of protamine sulfate will neutralize 100 U of UFH. LMWH is approximately 60% reversed by protamine. The maximum dose of protamine is 50 mg during a single 10-minute prolonged IV infusion. Hypotension, bradycardia, dyspnea, flushing, and anaphylaxis may occur during administration.
There is no reliable antidote for fondaparinux. Given its long half-life, the drug may persist for 36 to 48 h after administration. Use in patients with increased bleeding risks is not advised.
Ease of administration and reliable pharmacokinetics make LMWH and fondaparinux particularly suitable for outpatient VTE treatment.
LMWH typically can be administered without anti-factor Xa monitoring; however, monitoring may be helpful in children, pregnant women, morbidly obese patients, and patients with significant renal or hepatic insufficiency. A chromogenic anti-factor Xa level of 0.6 to 1.0 U/mL has been recommended for enoxaparin sodium. The anticoagulant effect of fondaparinux can be measured by anti-factor Xa activity. However, monitoring has not been advocated and standardized assays are not readily available.
UFH is fairly inexpensive; however, LMWH and fondaparinux are considered cost-effective compared with UFH. Although these drugs are more costly, cost savings are realized in the form of outpatient therapy, decreased monitoring, and ease of administration and preparation.
Vitamin K antagonists
Warfarin is the only VKA available in the United States. VKAs inhibit the terminal γ-carboxylation of clotting factors II, VII, IX, and X and the natural anticoagulants protein C and protein S. The decarboxylated proteins are nonfunctional within the coagulation cascade [33•].
Most patients can begin therapy with 5 mg of warfarin daily. In hospitalized, debilitated, or elderly patients, doses of 2 to 2.5 mg may be warranted [33•]. In stable outpatients with a low bleeding risk, one may consider using a 10-mg dose . Genetic variants of the CYP2C9 and VKORC1 enzymes alter warfarin dosage requirements; however, routine evaluation for these mutations is not recommended at this time.
Absolute and relative contraindications and other considerations for anticoagulation are listed in Table 1. Warfarin is associated with a well-characterized embryopathy. Patients should not use warfarin during pregnancy or become pregnant while on therapy. Warfarin is considered pregnancy category X by the FDA.
Main drug interactions
Concomitant use of drugs known to increase the risk of bleeding must be done with caution. Drugs that augment or compete with the cytochrome P-450 enzyme complex, displace warfarin from albumin, delay or inhibit gastrointestinal absorption, or alter the natural flora of the gastrointestinal tract may affect VKA therapy. The list of drugs and herbal products that affect or are affected by warfarin is ever growing .
Main side effects
Similar to the parenteral anticoagulants, the main adverse event occurring with VKAs is bleeding, either overt or subclinical (see “Parenteral Agents”). Another common, sometimes distressing, side effect of warfarin is hair loss or thinning; however, this is reversible with drug discontinuation.
Monitoring the effect of warfarin in patients with an elevated baseline prothrombin time (PT; ie, with a lupus anticoagulant, liver dysfunction, or malnutrition) may be difficult and the results unreliable . In these settings, targeting individual vitamin K-dependent coagulation factors II and X to activity levels between 20% and 25% usually is recommended. The FDA considers generic warfarin and name brand drugs (Coumadin, Bristol-Myers Squibb, Princeton, NJ, and Jantoven, Upsher-Smith Laboratories, Minneapolis, MN) to be equivalent. However, patients should advise their practitioner if they are aware of a change of their dispensed drug. Increased monitoring may be warranted. The effect of warfarin can be readily reversed using fresh frozen plasma with concomitant administration of vitamin K.
Generic warfarin and even brand name drugs are fairly inexpensive. However, INR monitoring increases the cost of therapy. The INR should be monitored daily at first, then weekly. Once stable therapy is achieved, INR should be monitored at least every 4 weeks [33•]. Whole-blood fingerstick monitoring has increased the availability of point-of-care testing through anticoagulation clinics and even allows self-testing and self-monitoring .
Direct thrombin inhibitors
Lepirudin, a recombinant hirudin analogue; argatroban, a synthetic derivative of l-arginine; and bivalirudin, a 20-amino acid peptide, are FDA-approved direct thrombin inhibitors (DTIs). These drugs bind and inhibit thrombin; they are monitored using the aPTT, but all cause some prolongation of the PT.
(FDA approved for the treatment of HIT and associated thrombosis.) The recommended dosage is a 0.4-mg/kg IV bolus followed by a 0.15-mg/kg/hour IV infusion adjusted to maintain a therapeutic target aPTT 1.5 to 2.5 times the median laboratory control value. Reducing the initial bolus to 0.2 to 0.3 mg/kg provides therapeutic anticoagulation without exceeding the target aPTT range. The half-life is approximately 60 min [37,38].
(FDA approved for the prevention or treatment of thrombosis in HIT.) The recommended dosage is a 2-µg/kg/min IV infusion adjusted to maintain an aPTT 1.5 to 3 times the laboratory control. A dose reduction to 0.5 to 1 µg/kg/min should be considered in critically or severely ill patients. The half-life of argatroban is approximately 45 min; the aPTT should be measured every 2 to 4 h after initiation or dose adjustment [37,39].
(FDA approved for percutaneous coronary intervention [PCI] with or without HIT.) The recommended dosage during PCI is a 0.75-mg/kg IV bolus followed by a 1.75-mg/kg/hour infusion. A maintenance infusion of 0.2 mg/kg/hour may be used for up to 20 h following the procedure. One off-label study using bivalirudin in HIT was successful when the initial dose was adjusted based on renal function. The study investigators recommended an IV infusion rate of 0.15 mg/kg/hour in patients with a CrCl greater than 60 mL/min, 0.08 to 0.1 mg/kg/hour in those with a CrCl of 30 to 60 mL/min, and 0.03 to 0.05 mg/kg/hour in those with a CrCl less than 30 mL/min or receiving renal replacement therapy [40•]. The half-life of bivalirudin is approximately 23 min; the aPTT should be measured 2 h after starting therapy or adjusting the dosage. The dosage may be adjusted to maintain the aPTT at two to three times the baseline control.
Absolute and relative contraindications and other considerations for anticoagulation are included in Table 1. DTIs must be used cautiously, if at all, in the setting of pending or recent neuroaxial anesthesia and should be avoided in patients with recent traumatic epidural or spinal catheter placement. Lepirudin should be avoided in patients with a CrCl less than 30 mL/min. Dose adjustments may be required in patients with a CrCl of 30 to 50 mL/min. Argatroban should be avoided in patients with significant liver dysfunction. Bivalirudin is cleared only in part by the kidneys. Dose adjustment may be warranted in patients with renal insufficiency. All DTIs are considered pregnancy category B by the FDA.
Main drug interactions
See “Parenteral Agents.”
Main side effects
See “Parenteral Agents.” Anti-hirudin antibodies that may decrease drug clearance and prolong the aPTT have been demonstrated following exposure to lepirudin. Anaphylaxis has been documented with repeat lepirudin exposure.
All DTIs have some laboratory effect on the PT. Conversion to warfarin during DTI therapy requires careful monitoring and assessment. Argatroban has a more significant effect on the PT than bivalirudin or lepirudin. None of the DTIs have a specific antidote. Recombinant activated factor VII has been shown to increase thrombin formation and may be helpful for severe bleeding. With relatively short half-lives, most of these drugs are cleared fairly rapidly. Supportive care with blood products may be required.
DTI therapy is considerably more expensive than the other available parenteral agents. Routine use in VTE is not advocated. However, in HIT, DTIs are the treatment of choice.
The role of thrombolysis in PE is somewhat controversial. Studies clearly demonstrate improvement in hemodynamics and angiographic appearance of PE following thrombolytic administration. However, most trials have failed to demonstrate a meaningful reduction in recurrent events or death. A recent meta-analysis of five trials enrolling a total of 464 patients with hemodynamically stable, submassive PE failed to identify any benefit from tissue plasminogen activator administration compared with heparin in this patient population . Another meta-analysis of 11 trials of thrombolysis compared with anticoagulation alone demonstrated no benefit to thrombolysis in the pooled analysis. However, in five trials including patients with massive, hemodynamically unstable PE, a significant reduction in recurrent PE or death was identified: 9.4% with thrombolysis compared with 19% with anticoagulation alone (odds ratio, 0.45; 95% CI, 0.22–0.92) . Therefore, systemic thrombolysis has been advocated for some patients with massive PE and hemodynamic instability [13••].
Thrombolytics convert plasminogen to plasmin. Plasmin nonspecifically cleaves thrombus-bound fibrin, circulating fibrin, and fibrinogen. Therefore, active therapeutic thrombolysis frequently is complicated by a systemic coagulopathy.
Streptokinase, urokinase (not available in the United States), and alteplase (recombinant tissue plasminogen activator [rt-PA]) are FDA approved for treatment of acute, massive PE [43•]. Tenecteplase and reteplase, FDA approved for acute myocardial infarction (AMI), have been investigated in small trials of massive PE [44, 45]. Desmoteplase, not FDA approved, has been studied in stroke and recently in patients with massive PE .
rt-PA is produced from cell culture. Reteplase and tenecteplase are recombinant, engineered, deletion variants of rt-PA designed to have improved fibrin specificity and longer half-lives.
Loading dose of 250,000 U IV over 30 min, followed by 100,000 U/hour over 24 h.
Loading dose of 4,400 U/kg/hour IV over 10 min, followed by 4,400 U/kg/hour for 12 h.
The FDA-approved dosage is 100 mg IV delivered over 2 h by continuous peripheral infusion without the use of heparin. In Europe, rt-PA is administered using a 10-mg bolus, followed by a 90-mg continuous IV infusion with concomitant UFH .
Studied in massive PE using the approved AMI dosage of a 10-U IV bolus with an additional 10-U IV bolus 30 min later .
The recommended AMI dosing is a 30- to 50-mg body weight-adjusted bolus administered over 5 to 10 s .
Doses of 180 and 250 µg/kg IV demonstrated similar or even greater efficacy compared with alteplase, 100 mg, for massive PE .
Contraindications for thrombolysis include any increased risk for bleeding; active bleeding; known thrombolytic hypersensitivity; recent organ biopsy; arterial trauma or puncture of a noncompressible site; gastrointestinal or genitourinary bleeding within the previous 10 days; thrombocytopenia or severe anemia; known bleeding diathesis; a history of intracranial, spinal, or intraocular bleeding; cardiopulmonary resuscitation, major surgery, stroke, or major trauma within the previous 14 days; severe hypertension; severe liver disease; and pregnancy. In all patients, the consideration for thrombolysis must be weighed against the increased risk for bleeding.
Main drug interactions
See “Parenteral Agents.”
Main side effects
Hemorrhage, including intracranial (1–3%) or life-threatening hemorrhage, is the most worrisome side effect. Streptokinase may be antigenic; repeat exposure may result in anaphylactic reactions.
In the United States, concomitant heparin infusion during thrombolysis is not advocated. Following thrombolysis, the aPTT should be measured; if it is less than 2.5 times the control laboratory value, then heparin may be initiated. Some authors advocate maintaining fibrinogen greater than 100 mg/dL during therapy. For bleeding, supportive care with fresh frozen plasma and cryoprecipitate may be helpful.
Per unit dose, thrombolytics are expensive. Cost-effectiveness in lives saved has not been evaluated. Thrombolysis may be considered in hemodynamically unstable PE; however, routine use is not recommended.
Inferior vena cava filter placement
IVC filters may prevent massive PE when there is a contraindication to anticoagulation, a bleeding complication related to anticoagulation, or a recurrent event despite adequate anticoagulation. Relative indications for filter insertion include massive PE or limited cardiopulmonary reserve, in which a small recurrent PE may be poorly tolerated; large iliofemoral or IVC thrombosis, especially associated with thrombolysis; and planned pulmonary thromboendarterectomy.
IVC filters do not treat VTE. Their use should be considered prophylactic. If clinically feasible, patients should be treated with appropriate anticoagulation.
There are 12 FDA-approved IVC filters, six of which are approved for both permanent and optionally retrievable use. Placement access typically is via the internal jugular or femoral veins; however, the subclavian, axillary, and even brachial veins may be used. Each filter has a unique deployment device, and the introducer size varies among devices. Ideal positioning is in the infrarenal IVC. Imaging of the IVC is required before and after placement for accurate deployment. Suprarenal placement has been advocated in pregnant women, women of childbearing status, patients with distal IVC thrombosis, those with extensive ovarian vein thrombosis, or those with an infrarenal IVC greater than 40 mm.
There are few absolute contraindications to IVC filter placement. Allergy to the metal alloy may affect device choice. Relative contraindications include contrast dye allergy, significant renal insufficiency, extensive IVC thrombosis at the level of the renal veins requiring suprarenal filter placement, and HIT.
Procedural complications include insertion site bleeding or infection, inadvertent puncture of an adjacent artery or anatomic structure (eg, pneumothorax), and accidental deployment in the incorrect vessel. Device-related complications include maldeployment with tilting or failure to fully deploy, migration during or after deployment, component or device fracture, strut penetration, and guidewire entrapment. Thrombotic complications, including recurrent VTE and insertion site thrombosis, are well documented. One randomized trial compared long-term outcomes in anticoagulated patients with or without IVC filters. At 8 years, patients with an indwelling IVC filter had a significant decrease in PE (6.2% vs 15.1%; P = 0.008) and an increase in secondary DVT (35.7% vs 27.5%; P = 0.042) but no difference in postthrombotic syndrome (70.3% vs 51.0%; P = 0.83) compared with patients treated with anticoagulation alone . Insertion site thrombosis has been documented in up to 20% to 30% of patients. Recurrent PE has been documented in up to 10% of patients and may be fatal in up to 2% .
There is no specific indication for an optionally retrievable filter. If the anticoagulation risk is time limited, optionally retrievable IVC filters may be particularly attractive. This allows retrieval of the device when it no longer is required and avoids the risks of a long-term indwelling apparatus. The ongoing PREPIC 2 study is a multicenter, randomized trial designed to assess the efficacy and safety of the ALN optionally retrievable filter (ALN Implants Chirurgicaux, Ghisonaccia, France) implanted for 3 months in anticoagulated patients with symptomatic PE. Measured outcomes will include recurrent PE at 3 months and symptomatic VTE, filter complications, and death at 6 months .
No studies have compared the cost-effectiveness of IVC filters versus anticoagulation alone. Placement and retrieval essentially double the procedure-related costs.
Pulmonary artery catheter-directed therapy (CDT), including suction thrombectomy, rheolytic thrombectomy, thrombus maceration or fragmentation, and catheter-directed thrombolytic infusion, may be considered in experienced centers if anticoagulation alone has not provided sufficient clinical improvement and there are contraindications to systemic thrombolytic therapy.
CDT for massive PE has not been considered the standard of care. The published literature consists of case reports and small series using a variety of techniques [51•]. As such, there is no standard procedure for pulmonary artery CDT. Each procedure is operator dependent and requires considerable skill and expertise.
Operator experience should guide procedural attempts. Other than the inability to gain endovascular access and contrast dye allergy, there are few absolute contraindications to CDT. Anticoagulation should be maintained during and after the procedure; the inability to use systemic anticoagulation is a relative contraindication to CDT.
Although the quality of the published literature makes it difficult to assess actual CDT complication rates, they likely are similar to those of other endovascular procedures. One systematic review reported groin hematoma, transient bradyarrhythmia, transient heart block, hemoglobinuria, hemoptysis, contrast-induced renal insufficiency, embolus dislocation, pulmonary artery dissection, cerebral and noncerebral hemorrhage, cardiac tamponade, and procedure-related death [51•].
No interventional devices are approved for use in the pulmonary artery. There are no organized clinical trials assessing the benefit of CDT for PE. Wide adoption of these techniques for managing massive PE seems unlikely . CDT may be particularly useful if surgical embolectomy is not available or if the patient has contraindications to surgery.
No data are available regarding the cost or cost-effectiveness of CDT.
Pulmonary artery embolectomy may be indicated in patients with massive, hemodynamically unstable PE when there are contraindications to thrombolysis. The mortality associated with the procedure is acceptable compared with other interventions.
Pulmonary embolectomy is performed under cardiopulmonary bypass with or without cardioplegia and hypothermia. The main pulmonary artery trunks are opened, and manual thrombectomy is performed under direct visualization [53,54]. Intraoperative placement of an IVC filter is common; use of a right ventricular assist device may be required.
Pulmonary embolectomy should be considered in patients with massive, hemodynamically unstable PE in whom thrombolysis is contraindicated or after the administration of thrombolytics without hemodynamic improvement.
Late mortality following surgical embolectomy is related to underlying comorbidities, not the procedure itself. In one trial, there was a significant reduction in pulmonary artery pressures. Following the procedure, all patients were classified as New York Heart Association functional class I or II .
No data are available regarding the cost or cost-effectiveness of surgical embolectomy. The procedure may be considered lifesaving for patients with massive PE in extremis.
Several novel oral anticoagulants are undergoing clinical evaluation and may be available within the next few years. Appealing pharmacologic characteristics of these drugs include a rapid onset of action, a predictable linear pharmacokinetic profile, fixed dosing without routine therapeutic monitoring, and few drug interactions [55••,56]. These agents are discussed further in “New and Emerging Anticoagulant Therapies for Venous Thromboembolism,” also in this issue.
No potential conflicts of interest relevant to this article were reported.