Abstract

Part I: Thrombolysis for Deep Vein Thrombosis (DVT)
General Considerations
Iliofemoral DVT frequently leads to serious morbidity from the postthrombotic syndrome (PTS). Occlusion of the common femoral, external iliac, and common iliac veins obliterate the single venous outflow channel from the lower extremity. Spontaneous recanalization is rarely adequate to restore unobstructed venous drainage.
Observational studies have demonstrated unacceptably high postthrombotic morbidity, venous ulceration, and impaired quality of life (QoL) in patients treated with anticoagulation alone.1–3 A strategy for successful thrombus removal that avoids rethrombosis should reduce or eliminate PTS and potentially avoid recurrence.
Systemic Thrombolysis
A selected analysis from early randomized trials of systemic streptokinase administration demonstrated that venous valve function may be preserved in patients treated with lytic therapy compared to those treated with standard anticoagulation.4,5 An overview of results from 6 trials reported that systemic thrombolysis was 3.7 times more effective in producing some degree of lysis compared to heparin alone. 6 In a pooled analysis of 13 prospective studies, only 4% of the patients treated with heparin had successful or complete lysis compared to 45% of patients receiving systemic thrombolysis. 7 However, prolonged streptokinase infusions were often associated with allergic reactions and a hemorrhagic rate 3-fold higher than patients managed with heparin anticoagulation alone. 6
A randomized trial comparing recombinant tissue plasminogen activator (rt-PA) versus anticoagulation alone demonstrated that 58% of the patients receiving rt-PA achieved greater than 50% clot lysis compared to 0% in those receiving anticoagulation alone (P = .002) and that rt-PA-treated patients had a trend toward reduced PTS if lysis was successful (56% vs 25%, P = .07). 8 However, major bleeding was significantly higher in systemic thrombolysis compared to anticoagulation alone (P < .04).6,8,9
All trials of systemic thrombolytic therapy for acute DVT admitted patients with proximal DVT, not necessarily specifically those with iliofemoral DVT. Therefore, it is unknown whether patients with the most extensive venous thrombosis will improve or whether they face lower efficacy due to more extensive obliteration and greater thrombus burden as well as an increased risk of bleeding.
Catheter-Directed Thrombolysis
Catheter-directed thrombolysis (CDT) refers to infusion of a plasminogen activator directly into the thrombus using ultrasound-guided access to the deep venous system and fluoroscopic positioning of the catheter into the thrombus. Direct infusion of plasminogen activator into the thrombus takes advantage of accelerated thrombolysis by virtue of the plasminogen activator binding with fibrin-bound plasminogen. Avoiding systemic infusion has resulted in fewer major bleeding complications, and direct infusion of the lytic agent has been associated with improved efficacy.10–18
Successful CDT can be anticipated in 80% to 90% of the patients if treated within 14 days of symptom onset.10–19 Retrospective observations show that CDT results in improved QoL and that QoL is related to the magnitude of lytic success.19,20 The frequency and severity of PTS is directly related to the amount of residual thrombus at the completion of CDT. 21 A randomized trial involving 209 patients compared CDT followed by anticoagulation with anticoagulation alone (control group) for iliofemoral DVT. After 24 months of follow-up, the incidence of PTS was reduced from 55.6% in the control group to 41·1% in the CDT group (P = .047). The difference in PTS corresponded to an absolute risk reduction of 14.4% (95% confidence interval [CI] 0.2-27.9), and the number needed to treat was 7 (95% CI 4-502). At 6 months, iliofemoral patency was found in 65.9% of the patients in the CDT group and 47.4% in the control group (P = .012). Twenty bleeding complications related to CDT included 3 major and 5 clinically relevant bleeds. Randomized trials have demonstrated improved patency of the iliofemoral venous system10,11 and preserved venous valve function. 10
Bleeding complications in excess of 10% were observed in early studies, 15 but these have been reduced in more contemporary reports. The CaVenT investigators 22 reported that patients randomized to CDT plus anticoagulation had a major bleeding event rate of 3% versus 0% of patients randomized to anticoagulation alone. They used an rt-PA dose of 0.01 mg/kg per hour which was substantially lower than that used in earlier trials. This dose is consistent with most contemporary experiences, which often use a fixed dose of 1 to 2 mg of rt-PA per hour infused in 50 to 100 cm3 solution. Reduction in bleeding complications is likely multifactorial, including lower concentrations and overall dose of plasminogen activators, routine incorporation of ultrasound-guided vein puncture, and lower doses of heparin used during lytic infusion.
Pharmacomechanical Thrombolysis
Pharmacomechanical thrombolysis refers to percutaneous catheter-based techniques that integrate mechanical clot disruption in conjunction with intrathrombus infusion of a plasminogen activator.
Evidence does not exist to show that catheter-based mechanical thrombectomy alone, which includes aspiration, maceration, and/or fragmentation, has been effective for management of acute DVT.23–25 Clot manipulation in the absence of concomitant thrombolytic therapy has been associated with increased risk of symptomatic PE.23–25
Retrospective studies of pharmacomechanical techniques suggest that similar or improved efficacy can be achieved in shorter treatment times using reduced doses of plasminogen activator and reduced use of hospital and/or intensive care unit (ICU) length of stay without adversely affecting valve function.18,26–34 Several observational studies indicate that thrombus can be removed in some patients in a single procedure session,18,29–31 which reduces the need for hospitalization and eliminates the need to utilize ICU. Studies comparing postthrombotic morbidity in patients treated with CDT versus those treated with pharmacomechanical lysis are not available.
Recommendations
Systemic thrombolysis for patients with proximal DVT is not recommended due to low efficacy and increased risk of bleeding complications (level of evidence: high).
The CDT is recommended for patients with acute iliofemoral DVT (level of evidence: moderate). Patients with acute iliofemoral DVT at a center lacking expertise in CDT should be transferred to a center where expertise exists if indications for CDT are present.
Physicians puncturing deep veins should use ultrasound guidance for access (level of evidence: low).
In centers where expertise is available, pharmacomechanical thrombolysis is recommended as initial therapy for patients with iliofemoral DVT (level of evidence: low).
Pharmacomechanical thrombolysis is recommended in preference to CDT for iliofemoral DVT in centers where appropriate expertise is available (level of evidence: low).
Percutaneous mechanical thrombectomy alone (in the absence of thrombolytic therapy) is not recommended for the management of patients with acute DVT (level of evidence: low).
Patients treated with CDT or pharmacomechanical thrombolysis should receive the same intensity and duration of anticoagulation (level of evidence: low).
Part II: Thrombolysis for Pulmonary Embolism (PE)
General Considerations
Pulmonary embolism is a significant cause of mortality and can be associated with chronic thromboembolic pulmonary hypertension resulting in ongoing patient morbidity.35–37 Strategies to eliminate the acute pulmonary embolus are designed to improve survival and reduce long-standing morbidity of chronic thromboembolic pulmonary hypertension. 38
Outcomes are related to the severity of the PE. Short of sudden death, a number of factors have been used to identify patients at risk of poor outcomes, but although clinical features including age and comorbidities, influence the prognosis in acute PE34,39,40 and have been incorporated into clinical scores,41–44 they do not sufficiently predict the outcome in the absence of imaging or biomarkers. 45
Computed tomographic angiography
The burden of thrombus alone measured by quantitative assessment of a computed tomographic (CT) angiogram does not predict adverse outcomes. 46 However, CT scan measurement of right ventricular (RV) dilatation is associated with inhospital mortality, 47 30-day mortality, 48 and 3-month mortality. 49 A RV/left ventricular (LV) index of more than 0.9 is shown to be associated with adverse clinical outcomes.48,50 Ventricular septal deviation also predicts short-term mortality. 51 A meta-analysis of 2 studies involving 191 patients showed a pooled sensitivity of 65% (95% CI, 35%-85%) and specificity of 56% (95% CI, 39%-71%) for short-term mortality. 52
Echocardiography
Echocardiography can identify large pulmonary emboli obstructing the RV outflow to produce RV dysfunction. Parameters assessed include RV enlargement, septal deviation, tricuspid insufficiency, and increased pulmonary artery pressures. A systematic review of RV dysfunction defined by echocardiography involving 5 studies of 475 patients with stable PE revealed an odds ratio of 2.53 (95% CI, 1.17-5.50) for short-term mortality. 52 These studies showed a pooled sensitivity of 70% (95% CI, 46%-86%) and specificity of 57% (95% CI, 47%-66%) for short-term mortality. 52
Troponins
Troponin I and troponin T released from microinfarction of RV muscle are markers of myocardial injury. When elevated, they are associated with an adverse prognosis in patients with acute PE.53–58 A meta-analysis demonstrated that elevated troponin levels in patients with submassive PE were associated with a 19.7% mortality compared with 3.7% in patients with normal troponins (relative risk, 4.72; 95% CI, 3.45-6.47). 59
Natriuretic peptides
Natriuretic peptides that include brain natriuretic peptides (BNP) and N-terminal pro-BNP are released when the myocardium is placed on stretch and have been shown to predict adverse short-term outcomes in patients with acute PE. Literature reviews have demonstrated that mortality is increased 5- to 9.5-fold depending upon whether BNP or N-terminal pro-BNP was studied.60–62 A meta-analysis of 2 studies involving 170 patients showed a pooled sensitivity of 93% (95% CI, 14%-100%) and specificity of 59% (95% CI, 14%-92%) for short-term mortality. 52
Electrocardiography
There is a worse short-term prognosis if a PE is large enough to cause abnormalities in the conducting system that reveals right heart strain.63–73 These include sinus tachycardia, atrial arrhythmias, right bundle branch block, S1Q3T3 pattern, and ST-segment changes in V1 to V4.
Risk stratification for acute PE
The outcome for patients with acute PE depends on the hemodynamic compromise, the impact on the myocardium identified by RV dysfunction, myocardial damage, myocardial stretch, and cardiac electrical activity. Stratifying patients according to risk of morbidity and mortality is clinically helpful and is recommended in order to appropriately evaluate patients for treatment.
Massive PE is defined as acute PE causing sustained hypotension (systolic blood pressure less than 90 mm Hg for more than 15 minutes or requiring inotropic support), severe bradycardia (heart rate less than 40 bpm), or signs or symptoms of cardiogenic shock. In the MAPPET registry, inhospital mortality was 25% for patients presenting in cardiogenic shock and 65% for those requiring cardiopulmonary resuscitation compared to 8.1% in those who were hemodynamically stable. 74 Reports based on clinical predictors alone identify a systolic blood pressure less than 100 mm Hg as a predictor for an adverse outcome.43,44 In the ICOPER registry, the 90-day mortality rate for patients with acute PE and systolic blood pressure less than 90 mm Hg at presentation was 52.4% versus 14.7% in the remaining patients. 75
Submassive PE refers to the broad subset of patients who are defined as hemodynamically stable but with acute pulmonary emboli large enough to cause tachycardia, electrical disturbances on electrocardiography (EKG), RV dysfunction, or an increase in cardiac biomarkers.
Low-risk PE
Effect of Thrombolysis in Patients With PE
Most well-controlled randomized trials of thrombolysis for acute PE included a spectrum of patients with PE, many of whom would be well managed with anticoagulation alone. Many patients with low risk or submassive PE would not be expected to die so that judging success from mortality rates alone may underestimate the value of thrombolysis. Likewise, treating patients who may not benefit from lytic therapy will needlessly expose patients to an increased bleeding risk.
Randomized-controlled NIH-sponsored trials77,78 that compared lytic therapy versus heparin demonstrated more rapid and complete clearing of pulmonary emboli with lysis but with no reduction of mortality and an increased risk of bleeding. At 1-year follow-up, patients who underwent lytic therapy had better oxygen diffusing capacity and pulmonary capillary blood volume. 79 At 7-year follow-up, right heart catheterization demonstrated significantly reduced pulmonary artery pressures and pulmonary vascular resistance at rest and exercise. 80 This translated into significantly fewer patients who underwent lytic therapy having heart failure. The lytic group also had fewer recurrent DVTs and PEs as well as a reduced need for inferior vena cava filters.
A European randomized trial of thrombolytic therapy plus heparin versus heparin alone for submassive PE demonstrated improved results with primary lysis with significantly fewer patients requiring salvage lysis or aggressive clinical support. 81
A randomized study of patients with massive PE appeared to show a meaningful reduction in either recurrent PE or death, to 9.4% with thrombolytic therapy compared to 19.0% with anticoagulation alone (odds ratio 0.45, 95% CI, 0.22-0.90). 82 A small randomized study of massive PE was terminated by the Data and Safety Monitoring Committee because all 4 patients randomized to anticoagulation died whereas all 4 patients randomized to thrombolytic therapy survived. 83
Catheter-Based Interventions for PE
Direct mechanical intervention may be lifesaving for patients with massive or submassive PE who are deteriorating. Percutaneous catheter-based techniques can be performed as an alternative to systemic thrombolysis if there is a contraindication to systemic lysis or if surgical embolectomy is unavailable. Either catheter-based interventions or surgical embolectomy can be lifesaving if systemic thrombolysis has failed. 84
The early technique of aspiration thrombectomy with the Greenfield suction and embolectomy catheter (Boston Scientific, Natick, Massachusetts) is currently the only Food and Drug Administration-approved device, 85 but it has not been widely adopted because it is cumbersome and associated with many technical and physiologic difficulties. Advances in catheter-based technology has demonstrated that thrombus fragmentation can be performed with balloon catheters, pigtail catheters, impeller-based homogenization, rheolytic intervention, and ultrasound-accelerated thrombolysis.86–92
A systematic review of percutaneous therapy alone for patients with massive PE found an 81% success rate with mechanical therapy and 95% success rate when combined with infusion of a thrombolytic agent. 91 Since limited doses of plasminogen activators can be used safely without systemic effect and may substantially increase interventional success, it seems reasonable to incorporate both the pharmacologic and the mechanical advantage from catheter techniques for massive PE. The risk of pulmonary artery perforation increases when arteries smaller than 6 mm in diameter are treated. 93
Surgical Embolectomy
Operative embolectomy for acute massive PE remains a viable treatment option. It has been shown to be effective to rescue patients with failed systemic thrombolysis for massive PE. 84 Previous reports of operative mortality in the range of 25% to 30% have reduced enthusiasm for operative approaches, favoring alternative therapies. 91 However, recent contemporary series are associated with substantially improved outcomes.94,95 The contemporary procedure can often be performed without placing the patient on cardiopulmonary bypass and without aortic cross-clamping.
In light of the variety of techniques now available for patients with massive PE, it is advisable to develop a multidisciplinary team of surgeons, interventionalists, and physicians expert in thrombolytic therapy to design treatment algorithms for patients with potentially fatal PE.
Recommendations
All patients with PE should undergo risk stratification (level of evidence: high). Patients with massive PE should undergo thrombolytic therapy in the absence of risk factors for bleeding complications (level of evidence: high). Thrombolytic therapy should be considered in patients with submassive acute PE if they are not at high risk of bleeding complications (level of evidence: moderate). Thrombolytic therapy is not recommended for patients with low-risk PE (level of evidence: moderate). The same intensity and duration of anticoagulation should be offered to patients treated with thrombolytic therapy for PE (level of evidence: low). In patients with massive PE, catheter-based intervention or surgical embolectomy are reasonable alternatives (level of evidence: low). Catheter-based embolectomy or surgical embolectomy is recommended following unsuccessful thrombolysis for PE (level of evidence: low). Catheter-based intervention or operative surgical embolectomy can be considered for patients with submassive PE who are at increased risk of bleeding from systemic thrombolytic therapy (level of evidence: low).
Patients with acute PE who are at low risk are best treated with anticoagulation alone (level of evidence: moderate).
