Pulmonary embolism—an update on thrombolytic therapy

Pulmonary embolism is a commonly encountered disorder, usually precipitated by deep venous thrombosis, and is associated with significant morbidity and mortality. It accounts for 10% of all deaths in hospital, and is a major contributing factor in a further 10%.1 It can result in pulmonary hypertension and right ventricular dysfunction, and its mortality rate of approximately 14%,2 has changed little over the last 30 years.1We review the treatment of pulmonary embolism, with particular reference to thrombolytic therapy. The role of surgical or catheter‐based embolectomy will not be discussed. The initial diagnosis of pulmonary embolus is not always straightforward. In one report, pulmonary embolus was not clinically suspected in 70% of patients in whom it was subsequently considered to be the major cause of death.3 Conversely, another autopsy series found that 63% of patients thought to have had a pulmonary embolus had no such evidence.4 The ‘classical’ presentation is pleuritic chest pain of acute onset associated with shortness of breath and perhaps haemoptysis (peripheral pulmonary infarction). A preceding history of deep venous thrombosis may be present. The patient is usually tachycardic and tachypnoeic and may be hypoxic and hypotensive. However, patients may present with isolated dyspnoea, circulatory collapse (in particular cases with central pulmonary arterial obstruction) or pleuritic chest pain alone. Over 70% of fatal and non‐fatal pulmonary emboli have proximal deep‐vein thrombosis.5 One or more predisposing factors are present in 80–90% of patients with pulmonary embolus.6 The most common are immobilization for more than 1 week, a history of previous thromboembolism, recent trauma, and surgery (particularly of the lower limbs). Other risk factors include long‐distance air travel, inherited clotting disorders (e.g. factor V Leiden, antithrombin III deficiency, antiphospholipid antibodies)7 or the contraceptive pill in women who smoke or are diabetic. In surgical patients, the risk of thromboembolism increases with age, length of general anaesthetic, site of surgery (especially abdomen and lower limbs) and the presence of advanced cancer and previous thromboembolism.6 In non‐surgical patients, there are three major risk factors, namely cardio‐respiratory disorders, lower‐limb immobility and malignancy (especially of the uterus, pancreas, breast and stomach).5 The haemodynamic response to pulmonary embolism depends upon a number of factors, including the size of the embolus, the presence of co‐existent cardiopulmonary disease and the neurohumoral responses produced. An acute increase in right ventricular afterload occurs when approximately 25% of total pulmonary blood flow is acutely obstructed.8 Right ventricular systolic pressure continues to rise as the degree of obstruction increases (>50% acute obstruction of pulmonary arterial circulation), usually up to a maximum mean pulmonary artery pressure of 30 mmHg.8 If afterload continues to increase, the right ventricle begins to fail, with a rise in right atrial pressure, and clinical shock will ensue once forward cardiac output can no longer be sustained. The presence of systemic hypotension and thus right heart functional reserve, is the major determinant of acute survival.9 In contrast, long‐term survival depends to a large degree on co‐existent cardiopulmonary disease and other underlying disease processes.10 An echocardiographic assessment 6 weeks post‐acute event can accurately identify patients who are likely to have persistent pulmonary hypertension/right ventricular dysfunction in the longer term.11 General supportive measures should be given including analgesia, although opiates should be avoided in patients with hypotension due to their vasodilatory effects. Hypoxia should be treated with high‐percentage inspired oxygen, and hypotension treated with colloids, aiming to ensure a central venous pressure of 15–20 mmHg. In the majority of patients with pulmonary emboli, the standard therapy remains anticoagulation, initially with an intravenous bolus of 5000 IU of unfractionated heparin followed by an infusion to maintain the APTR at 1.5–2.5 times the control value until adequate replacement by oral warfarin, aiming for a target INR of 2–3 for 6–12 weeks.7 Barritt and Jordan first reported on the mortality benefit of anticoagulation in pulmonary embolism. They observed no deaths in 54 patients treated with anticoagulants (i.v. heparin over 36 h and nicoumalone for 14 days) while there were five deaths in 19 control patients who did not receive anticoagulants.12 Recent reports suggest that low‐molecular‐weight heparin may be as effective as unfractionated heparin in non‐life‐threatening pulmonary embolism.13,14 The rationale for anticoagulation is to provide prophylaxis against further thromboembolic events while the body's own fibrinolytic system gradually lyses the embolus.8 In contrast, the thinking behind the use of thrombolytic therapy (followed by anticoagulation) for pulmonary embolism is that actively dissolving the formed clot, more rapidly restores cardiopulmonary function to normal values. Chronic pulmonary hypertension may also be reduced, and the source of embolus in the venous system removed or reduced.9 Thrombolysis would thus be particularly beneficial in massive pulmonary embolism associated with haemodynamic compromise (Tables 1 and 2). The only established alternative treatment for this group of patients is pulmonary embolectomy, a procedure probably no more successful than thrombolysis15 and only available in a limited number of hospitals. More recently, transvenous catheter extraction or fragmentation of emboli has been tried, but these methods need further evaluation.7 In patients with massive pulmonary embolism, anticoagulation therapy alone may fail to resolve the pulmonary artery clot completely in up to 75% of patients at 1–4 weeks and in 50% at 4 months.16,17 Early uncontrolled studies of thrombolytic therapy found that streptokinase or urokinase delivered either peripherally or centrally via a pulmonary artery catheter produced angiographic and clinical improvement in patients with severe pulmonary embolus.18–24 Indications for thrombolytic therapy in pulmonary embolism Indications for thrombolytic therapy in pulmonary embolism Contra‐indications to thrombolytic therapy Contra‐indications to thrombolytic therapy Several small clinical studies have compared the efficacy of thrombolytic therapy and anticoagulants in pulmonary embolism. The early studies consistently demonstrated more rapid anatomical and/or physiological improvement in thrombolysis‐treated patients, whether or not there was haemodynamic compromise, but all were too small to assess the effects on mortality.25–30 One of the larger early studies, Phase 1 of the Urokinase Pulmonary Embolism Trial (UPET), compared peripheral administration of heparin alone with a 12‐h infusion of urokinase in 160 patients with angiographically‐proven pulmonary embolism.25 At 24 h, the urokinase‐treated group had significantly greater haemodynamic and anatomical improvements than the heparin‐treated patients, but by day 7 no difference between the two groups was noted on lung scans. During the 2 weeks following therapy, there was a trend towards fewer recurrent pulmonary emboli in the urokinase‐treated patients (17% vs. 23%). However, there was an increased incidence of severe bleeding (fall in haematocrit of ≥10% and/or the need for a blood transfusion of ≥2 units) in the urokinase group (27% vs. 14%). Similar small studies of either centrally or peripherally administered streptokinase showed clinical and angiographic benefit when compared to heparin administration alone.26,27,31 In a study using recombinant human tissue‐type plasminogen activator (rtPA), 101 haemodynamically stable patients with pulmonary embolus were randomized: 46 received rt‐PA (100 mg over 2 h) followed by heparin, and 55 received heparin alone.32 There was significant improvement in right ventricular wall motion at 24 h in the rt‐PA‐treated patients (39% vs. 17%). Pulmonary perfusion also improved by 14.6% compared to 1.5%. No clinical episodes of recurrence occurred in the rt‐PA group, whereas five of the heparin group had recurrent episodes within 14 days.32 More recently, a study based on a multicentre registry reported a significant reduction in in‐hospital mortality and recurrence rates of pulmonary embolus in patients receiving primary thrombolysis compared with those initially treated with heparin alone.33 Konstantinides et al. reviewed 719 consecutive haemodynamically stable patients with major pulmonary emboli. The inclusion criteria of the registry included clinical, echocardiographic and cardiac catheter findings signifying right heart failure and/or pulmonary hypertension caused by pulmonary embolism: (i) arterial hypotension (systolic blood pressure <90 mmHg or pressure drop of ≥40 mmHg for >15 min); (ii) cardiogenic shock (presence of arterial hypotension with signs of organ hypoperfusion and hypoxia); (iii) circulatory collapse with need for cardiopulmonary resuscitation; (iv) echocardiographic findings indicating increased right ventricular afterload and/or pulmonary hypertension; and (v) precapillary pulmonary hypertension by right heart cardiac catheterization (mean pulmonary artery pressure >20 mmHg in the presence of normal pulmonary wedge pressure). Patients were included if they met at least one of the above criteria at presentation, together with a diagnostic pulmonary angiogram, positive lung scan or at least three of the following: (i) syncope; (ii) tachycardia (heart rate >100/min); (iii) tachypnoea (respiratory rate>24/min); (iv) arterial hypoxaemia (PaO2 <70 mmHg on room air); (v) ECG signs of right ventricular strain. Primary thrombolysis was administered to 169 patients (23.5%), while the remaining 550 patients (76.5%) initially received heparin alone. Overall 30‐day mortality was significantly lower in the thrombolysis group compared to the heparin‐alone group (4.7% vs. 11.1%). Patients presenting with hypotension or syncope particularly benefited from thrombolysis. Recurrent pulmonary embolism during the in‐hospital phase was also significantly reduced in those who received early thrombolytic treatment (7.7% vs. 18.7%). It would appear from current knowledge that thrombolysis has significant advantages over anticoagulation alone in the treatment of patients with major pulmonary embolism, whether haemodynamically stable or not. Experimental studies of venous thromboembolism suggested that tPA was more potent than urokinase or streptokinase, prompting its assessment in pulmonary embolism.34,35 The clinical use of tissue plasminogen activator was first reported for pulmonary embolism in 1985.31 Soon after, 36 haemodynamically stable patients with angiographically‐proven pulmonary embolism (segmental or more proximal pulmonary artery involved), presenting within 5 days of onset of symptoms, received 50 mg rt‐PA peripherally over 2 h.36 Pulmonary angiography was repeated immediately after completion of the infusion. If significant clot lysis was evident, routine measures including anticoagulation were continued, otherwise an additional 40 mg rt‐PA was administered over 4 h with repeat angiography. Overall, 34 patients had some evidence of clot lysis after treatment with rt‐PA. Mean pulmonary artery pressures decreased significantly from 22 mmHg to 18 mmHg. Only two of the treated patients had a major bleeding complication. The FDA approved use of rt‐PA for acute pulmonary embolism (100 mg given intravenously over 2 h) in 1990. Limited comparative data are available on different thrombolytic agents in the treatment of pulmonary embolism. The phase II UPET trial compared regimens of 12 h urokinase, 24 h urokinase, or 24 h streptokinase (250 000 U bolus and 100 000 U/h for 24 h) in 167 patients. No differences in mortality or angiographic resolution were noted between the three groups.37 In a later study, Goldhaber and colleagues randomized 90 haemodynamically stable patients to either 100 mg rt‐PA (over 2 h) or urokinase (3 million units over 2 h, with the initial 1 million units given as a bolus over 10 min).38 Both drugs were administered peripherally, with all recruited cases having angiographically‐proven pulmonary embolism. Angiographic improvement was seen at 2 h post‐therapy in the rt‐PA and urokinase groups (79% vs. 67%). Additional perfusion lung scans at 24 h again were similar in the two groups. The European Cooperative Study Group for Pulmonary Embolism assessed 63 patients who were randomized to either urokinase (4400 U/kg bolus, 4400 U/kg/h for 12 h) or rt‐PA (10 mg bolus, 90 mg over 2 h) followed by heparin.39 The decrease in total pulmonary resistance at 12 h was similar in the two groups (53% and 48%, respectively), and bleeding complications were also similar. Current guidelines on streptokinase in the treatment of massive pulmonary embolism state that it should be given as a loading dose of 250 000 IU over 20–30 min followed by an infusion of 100 000 IU/h intravenously for up to 24 h.40 However, a recent study suggests that a 2‐h infusion of streptokinase is a safe and effective alternative regimen. Meneveau and colleagues randomized 66 patients to 100 mg rt‐PA (over 2 h) or 1.5 million units of streptokinase (over 2 h). All recruited cases had angiographically‐proven pulmonary embolus and both drugs were administered peripherally. At 1 h there was a significant reduction in total pulmonary resistance in the rt‐PA group compared with the streptokinase group (33% vs. 19%), but by 2 h post‐initiation of therapy, similar haemodynamic efficacy was noted. No significant differences in either pulmonary vascular obstruction (at 36–48 h) or bleeding complication rates were observed. In only two of the streptokinase patients (out of 43) was infusion discontinued because of a drop in systolic blood pressure.41 To date, it would appear that there is no difference in improvement in pulmonary vascular obstruction or bleeding complication rates between different thrombolytic agents. Experimental work has shown that thrombolysis continues for some time after tPA is cleared from the circulation, and that it can be accelerated and increased, together with reduced bleeding, if tPA is administered over a short period.42 No large studies of bolus therapy have been published to date. Levine randomized 58 haemodynamically stable patients to receive either a 2 min infusion of rt‐PA (0.6 mg/kg) or saline placebo in addition to a standard heparin infusion regimen.43 In the actively‐treated group, the percentage of patients with >50% resolution in perfusion defects at 24 h was increased compared to that in the placebo‐treated group (34% vs. 12%). However, no difference was apparent in lung scans obtained at 7 days. No major bleeding complications were noted in the rt‐PA‐treated patients. In a prospective but open trial, 54 patients with massive pulmonary embolism received a 10‐min infusion of rt‐PA at a dose of 1 mg/kg. By 48 h and 10 days, there were absolute improvements in perfusion defects of 11% and 31%, respectively.44 One of the patients died of an intracranial haemorrhage. There are few published studies of bolus regimens using urokinase or streptokinase. Two small open studies that used boluses of 15 000 U/kg and 20 000 U/kg of urokinase observed both an angiographic improvement and a reduction in pulmonary artery pressure, with a low incidence of major bleeding complications.45,46 There are, however, no published studies to date comparing bolus and other administration regimens. Thrombolytic therapy can be administered centrally rather than peripherally, using catheters sited in the right ventricle or in the pulmonary circulation. In theory, this should allow higher local concentrations of thrombolytic agent at the site of the pulmonary embolus, and might reduce bleeding complications. Several of the early studies using streptokinase involved locally delivered therapy over several hours (up to 72 h).26–28,47,48 In a small study of 10 patients with massive pulmonary embolus, treatment comprised intrapulmonary thrombolytic therapy (urokinase or streptokinase), anticoagulation and venocaval filters.49 A rapid response was seen in all, with significant improvements in oxygen saturation, pulmonary artery pressures, cardiac output and blood pressure compared to 10 patients over the same time period who had received heparin alone. Low‐dose streptokinase (10 000 U/h for 15–30 h) delivered locally may produce rapid lysis.50 In a brief report of three patients with massive PE who were treated with centrally administered tPA, all responded with rapid improvement of pulmonary artery pressures, and marked reductions in chest pain and shortness of breath.51 Verstraete et al. compared peripheral intravenous administration with local pulmonary administration of rt‐PA in 34 patients. Interestingly, they found that local pulmonary delivery did not confer any advantage over peripheral administration, with similar rates of lysis, bleeding and induction of systemic lytic state.52 From the studies to date, thrombolytic therapy (whether it is streptokinase, urokinase or tPA) administered peripherally or centrally, appears to be effective in all age groups for the treatment of massive pulmonary embolism. Elderly patients (i.e. >70 years) gain similar benefits to younger patients without an increased incidence of major bleeding complications when streptokinase is administered in a regimen comprising 250 000 U over 15 min and an infusion of 100 000 U/h for 12 h.53 Clinicians need to be diligent in diagnosing pulmonary embolism. At present it is both over‐ and under‐diagnosed in clinical practice. Clinical diagnosis can be improved if risk factor profiles are considered, since the majority of patients with pulmonary embolism have predisposing factors such as immobilization. Any underlying disease process should also be considered in the patient's management. The mainstay of diagnosis remains ventilation/perfusion lung scanning, but with suspected massive pulmonary embolism, urgent transthoracic echocardiography or spiral CT scanning should be used. In the majority of patients with pulmonary emboli, treatment will continue to comprise intravenous unfractionated heparin or subcutaneous low‐molecular‐weight heparin followed by oral anticoagulation for 6–12 weeks.40 Thrombolytic therapy should be considered in all patients with massive pulmonary embolism where there is evidence of acute pulmonary hypertension, right ventricular dysfunction and systemic hypotension. Although all randomized studies to date have been too small to address the issue of mortality benefit, there is objective evidence of haemodynamic benefit over that achieved by anticoagulation alone. Registry data also suggests reduced mortality, and decreased early recurrence of pulmonary embolism.33 All age groups and post‐operative patients also seem to benefit. Bolus and front‐loaded regimens (i.e. administered over ≤2 h) are simpler to use and appear to be as effective as longer‐duration regimens. At this stage, no major difference in efficacy between the different thrombolytic agents has been found. Streptokinase, however, should be avoided in those individuals who have received it previously for whatever indication, because of the likely presence of neutralization antibodies.54 Address correspondence to Dr R.S. More, Department of Cardiology, St Mary's Hospital, Milton Road, Portsmouth PO3 6AD