The diagnosis of acute symptomatic PE should be considered in all patients who report new onset dyspnea, worsening of their usual dyspnea, chest pain, syncope or hypotension with no alternative explanation, particularly when the basic complementary tests (chest X-ray, electrocardiogram and arterial blood gases) rule out other differential diagnoses. Table 2 lists some of the usual radiological and electrocardiographic findings in PE. Standardized testing using clinical prediction rules classifies patients into categories with different PE prevalence (approximately 10% for low probability, 25% for intermediate probability and >60% for high probability) and facilitates the interpretation of other diagnostic tests. The Wells and Geneva scores (Table 3) have been the most extensively validated.

• •

  The use of adequately validated clinical scales (Wells or Geneva) is recommended as a first step in the diagnostic approach to the hemodynamically stable patient with suspected PE.

D-dimer is a fibrin degradation product present in the thrombus, generated when the clot undergoes proteolysis by plasmin. It is a highly sensitive test, but has low specificity, because the finding of high levels may also be associated with other clinical situations such as advanced age, infection, cancer, pregnancy or hospital admission.

In normotensive patients with a low or intermediate probability of PE, a negative high sensitivity (≥95%) D-dimer (<500ng/ml) excludes the diagnosis of PE. In patients who do not receive anticoagulant therapy, the incidence of VTE in the following 3 months is 0.14% (95% confidence interval [CI], 0.05–0.41).5 Low or moderate sensitivity methods for determination of D-dimer (<95%) (Table 4) only exclude the disease in the group of patients with low clinical probability (or with PE unlikely according to the dichotomized Wells score).

• •

  It is recommended that the sensitivity of the D-dimer method used in each setting is known.

• •

  A negative high sensitivity D-dimer excludes PE in patients with low or intermediate clinical probability.

• •

  A negative moderate or low sensitivity D-dimer excludes PE in patients with low clinical probability (or PE unlikely).

• •

  It is recommended not to measure D-dimer in patients with a high probability of PE.

The most commonly used imaging tests are multidetector chest computed tomography angiography (angio-CT), ventilation/perfusion (V/Q) lung scan and venous ultrasound of the LE, with or without Doppler (DU).

The most commonly used prognostic factors in normotensive patients with PE are summarized in Table 5. In general, these tools assess the patient's clinical condition and general health status, right ventricular dysfunction, the thrombotic burden and myocardial damage.

The most useful tools for classifying low risk patients with acute symptomatic PE are the prognostic clinical scores. The combination of a set of variables that takes into account the age, comorbidity and cardiovascular repercussion of PE reliably identifies a subgroup of patients (approximately 30%) with a less than 2% risk of early mortality. The Pulmonary Embolism Severity Index (PESI) and simplified PESI (sPESI) clinical scores (Table 6) have been extensively validated as excellent tools for the identification of these low risk patients.14,15 The sPESI score is easier to use than the original, while retaining its prognostic ability.

The value of combining the clinical scores and some biochemical markers (particularly brain natriuretic peptide [BNP] or high sensitivity troponin [hsTnT]) or imaging tests (transthoracic echocardiogram or DU) for the identification of this group of low risk patients has not been completely clarified.

• •

  It is recommended to use well validated prognostic clinical scores (PESI or simplified PESI) as a first step for the identification of patients with low-risk PE.

The most widely used method in clinical practice for evaluating right ventricular function is transthoracic echocardiography. Various quantitative parameters (right ventricular end-diastolic diameter in the parasternal long axis view, ventricular end-diastolic diameter ratio in the apical 4C projection, tricuspid regurgitation peak velocity, diameter of the inferior vena cava, pulmonary artery acceleration time or TAPSE [Tricuspid Annular Plane Systolic Excursion]) have been used to estimate the degree of ventricular dysfunction. Its use for the identification of patients with intermediate-risk PE is limited by its operator-dependence, cost and lack of continuous availability in many centers. Furthermore, there is no echocardiographic pattern of right ventricular dysfunction that is sufficiently reliable as to justify, on its own, the use of fibrinolytic treatment.16

The images generated by angio-CT not only enable the diagnosis of PE to be confirmed or discarded, but they can also assess the extension of the arterial obstruction and the presence or not of right ventricular dilatation.17,18 Although ventricular size volumetric reconstruction studies may offer a more accurate prognostic assessment, more recent studies use the ratio between the ventricular diameters as a parameter of ventricular dysfunction (with different cut-off points). Its isolated use for identifying patients with intermediate-risk PE who could benefit from fibrinolytic treatment is not recommended.

Various meta-analyses have demonstrated the prognostic value of myocardial ischemia (identified by elevated troponin I or T levels) in hemodynamically stable patients with PE, although the power of the association is not sufficiently robust as to justify therapeutic escalation.19,20 Cardiomyocyte stress causes release of natriuretic peptides into the circulation (BNP and NT-pro-BNP).21,22 Several studies and meta-analyses suggest that they are useful for identifying patients with intermediate-risk PE. However, the sensitivity of these cardiac biomarkers for death due to PE is insufficient to establish the indication for thrombolytic treatment.

An association has been shown between the persistence of thrombotic material in the lower extremity deep vein system and short- and mid-term mortality of patients with PE.23 The presence of concomitant DVT, together with other imaging tests and cardiac biomarkers, may serve to identify a patient profile with a particularly high risk of complications associated with the PE itself.

• •

  A combination of prognostic tests (which identify right ventricular dysfunction, myocardial ischemia, cardiomyocyte stress or thrombotic burden) is suggested for identifying patients with intermediate risk PE.

For decades, unfractionated heparin (UFH) has been the drug of choice for the treatment of PE. It exerts its anticoagulant action by binding to antithrombin and potentiating its effect in the inactivation of a series of activated coagulation factors, mainly thrombin (IIa). It is usually administered intravenously (iv) as a continuous infusion, but it is also safe and effective using the subcutaneous route (sc). It requires monitoring using the activated partial thromboplastin time (APTT), which should be 1.5–2.5 times the control value, so an initial infusion rate of 18U/kg/h is used. As well as continuous infusion, an 80U/kg bolus is usually administered to achieve a more rapid anticoagulant effect. Patients who are treated with lower doses and who do not reach a therapeutic APTT in the first few days of treatment have an increased risk of recurrent VTE. It is not known if there is a relationship between sub-therapeutic APTT levels and recurrence in patients who are treated initially with the recommended doses. UFH is currently reserved for patients in whom the use of fibrinolytic treatment is considered (intermediate- or high-risk PE), and for patients at a high risk of bleeding who are to receive anticoagulant therapy. For patients with severe renal failure (creatinine clearance <30ml/min), some indirect evidence suggests that low molecular weight heparins (LMWH) (at the doses recommended in the Summary of Product Characteristics) could be safer and more effective than UFH.24

LMWH are prepared from the fractionation of UFH by chemical or enzymatic methods. They have potent antiXa action (greater than that of UFH). Most studies suggest that LMWH and UFH are equivalent in terms of efficacy (recurrent VTE) and safety (major bleeding) for treatment of the acute phase of VTE. In a meta-analysis that included 1951 patients from 12 studies comparing LMWH with UFH, LMWH were associated with fewer recurrent VTE (odds ratio [OR] 0.63; 95% CI, 0.33–1.18) and fewer major bleeding events (OR 0.67; 95% CI, 0.36–1.27), with no differences in mortality (OR 1.20; 95% CI, 0.59–2.45).25

Fondaparinux is a synthetic pentasaccharide that selectively inhibits factor Xa without inactivating thrombin. It is administered subcutaneously once daily at weight-adjusted doses and does not require monitoring. The Matisse investigators evaluated its efficacy and safety in the treatment of acute DVT and PE.26 Compared with UFH, they did not observe any differences in the rate of recurrent thromboembolic events (1.3% vs 1.7% in the acute phase), major bleeds (1.3% vs 1.1%) or mortality in the first 3 months of follow-up. A potential advantage of this drug is that it is not associated with heparin-induced thrombocytopenia (HIT).

• •

  The use of LMWH or fondaparinux instead of UFH is suggested in hemodynamically stable patients with acute PE.

• •

  The use of LMWH at doses adjusted to the UFH is suggested in patients with acute PE and severe renal failure.

Studies demonstrating that heparin treatment for 5 days instead of 10 is a safe, effective practice also showed that oral anticoagulants could be started on the first day of anticoagulation, without losing efficacy. Two clinical trials with warfarin have shown that: (1) loading doses (which may cause hemorrhages) should be avoided without this incurring a delay in reaching therapeutic INR levels; (2) commencing vitamin K antagonists (VKA) at lower doses avoids excessive falls in protein C levels, which would theoretically induce a state of hypercoagulability.

• •

  In hemodynamically stable patients with PE, it is recommended that parenteral anticoagulation be maintained for at least 5 days, and until the INR is >2.0 for 24h.

New oral anticoagulants have recently been introduced that act differently to the VKA, among which are dabigatran (direct thrombin inhibitor), rivaroxaban and apixaban (factor Xa inhibitors). Rivaroxaban is a direct selective factor Xa inhibitor. In the joint analysis of two clinical trials (for DVT and PE), rivaroxaban was associated with an efficacy similar to that of standard treatment (hazard ratio [HR] 0.87; 95% CI, 0.66–1.19) while major bleeding was reduced to half (HR 0.54, 95% CI, 0.37–0.79).27,28 Its oral administration, at a dose of 15mg twice daily for 3 weeks followed by 20mg once daily, could provide a simple, single-drug approach to the acute and long-term treatment of PE.

Apixaban has been evaluated in a clinical trial for the treatment of patients with DVT or PE (34%) for the first 6 months after a thrombotic event.29 Compared with standard treatment, apixaban showed similar efficacy (relative risk [RR] 0.84; 95% CI, 0.60–1.18) and a statistically significant reduction in major bleeding (RR 0.31; IC 95%, 0.17–0.55) and clinically relevant non-major bleeding (RR 0.48; 95% CI, 0.38–0.60). It was administered orally, at a dose of 10mg twice daily for the first 7 days followed by 5mg twice daily.

The efficacy and safety of dabigatran has not been evaluated during the first 10 days (on average) of treatment of acute symptomatic PE.

• •

  Treatment with rivaroxaban as monotherapy is suggested in hemodynamically stable patients with PE.

The results of a clinical trial and some cohort studies suggest that, compared with hospitalization, outpatient treatment in low risk patients is equally effective and safe in terms of recurrent VTE, bleeding events and mortality.30 Outpatient treatment can be considered in patients with PE who meet the following requirements: (1) clinically stable with good cardiopulmonary reserve, and a low risk validated clinical score (e.g. PESI or sPESI), (2) good social support with rapid access to medical care and (3) expected treatment compliance.

• •

  In patients with low-risk PE and adequate home conditions, early discharge is suggested instead of standard discharge (more than 5 days admission).

Several meta-analyses have shown that early mobilization is a safe practice in patients with proximal DVT.31 The evidence is less consistent for patients with symptomatic PE. The presence of concomitant DVT worsens the prognosis in patients with PE, particularly when associated with right ventricular dysfunction and myocardial ischemia.32

• •

  Early mobilization is suggested in patients with low-risk PE.

• •

  Rest is suggested for the first few days of treatment in patients with intermediate-risk PE.

Thrombolytic treatment accelerates clot lysis and hemodynamic improvement occurs more rapidly than with UFH treatment, although there are no differences in the residual thrombosis after 5–7 days. Analysis of a subgroup of patients in the Urokinase Pulmonary Embolism Trial (UPET), which compared urokinase followed by UFH or UFH alone, showed that fibrinolytic treatment reduces mortality in patients with shock secondary to massive PE.33 Based on these findings, and in the absence of a high risk of bleeding, thrombolytic treatment is indicated in patients with acute symptomatic PE and hemodynamic instability (defined as cardiogenic shock or sustained systolic blood pressure <90mmHg, not due to hypovolemia, sepsis or cardiac arrhythmias).34 Hemodynamic instability is rare, but up to half of patients with PE without hemodynamic instability have clinical and echocardiographic signs of right ventricular dysfunction (intermediate-risk PE). Although not routinely recommended,35 and pending publication of the results of the Pulmonary Embolism Thrombolysis Study (PEITHO),36 the decision to use thrombolytic treatment in a very selected group of patients will depend on their risk of bleeding and the severity of the clinical symptoms.

Thrombolytic treatment administered for 2h is safer and more effective than 12–24-h regimens. Streptokinase and recombinant tissue plasminogen activator (rt-PA) are equally effective with this short administration regimen. It should be administered via a peripheral route. Administration through a central line is not more effective and increases the risk of bleeding at the venous access insertion site.

• •

  The administration of systemic fibrinolytic treatment is recommended in patients with PE and cardiogenic shock.

• •

  The administration of systemic fibrinolytic treatment is suggested in patients with PE and hypotension without a high risk of bleeding.

• •

  It is recommended not to administer fibrinolytic treatment in most hemodynamically stable patients with PE.

• •

  In hemodynamically stable patients with intermediate-risk PE and a low risk of bleeding, particularly in the under 75s, it is suggested that the administration of fibrinolytic treatment be assessed.

Vena caval filters are indicated in PE patients with a contraindication for anticoagulation therapy. Decousus et al. published a clinical trial evaluating the usefulness of vena caval filters, as a complement to conventional anticoagulation, in patients with DVT and high-risk PE.37 The filters reduced the frequency of PE during the first 12 days of treatment, and a tendency toward a lower number of fatal episodes was detected. After 2 years, there were no differences between the two groups in terms of mortality or recurrent VTE, due to an increase in the frequency of DVT in the group treated with filters. These findings indirectly support the use of vena caval filters in patients in whom anticoagulation is contraindicated in the acute phase. A recoverable filter should be inserted when possible, and should be removed as soon as anticoagulation can be commenced.

• •

  Placement of an inferior vena caval filter is recommended in hemodynamically stable patients with PE and a contraindication for anticoagulation.

In experienced centers, mechanical thrombus fragmentation is performed in patients with high-risk PE and a contraindication for the use of fibrinolytics. Pulmonary embolectomy is another method of treatment for high-risk PE. It is indicated in cases of right-sided heart thrombi, high risk of paradoxical arterial embolism or in patients with high-risk PE in whom fibrinolysis has not been effective.38 The results improve if patients undergo the procedure before developing cardiogenic shock.

• •

  In patients with high-risk PE and who (i) have a contraindication for fibrinolysis, (ii) fibrinolysis has failed, or (iii) the shock is likely to cause death of the patient before the fibrinolysis is effective, the use of interventionist catheterization techniques or pulmonary embolectomy is suggested if the necessary resources and expertise are available.

Patients with acute symptomatic PE should receive supplementary oxygen to obtain saturations higher than 92%. Oxygen therapy, especially in patients with right-sided heart overload, acts as a vasodilator and may contribute to the decrease in pressure in the pulmonary arteries.

Pleuritic chest pain is a common symptom in patients with PE. It can be relieved on most occasions by administering non-steroidal anti-inflammatory agents within 24–48h. The administration of these drugs does not increase the risk of bleeding in acute PE.

The administration of fluids (<500ml) may be beneficial for increasing the cardiac output in patients with PE, low output and sustained systemic pressures. Dopamine or dobutamine can be used in patients with low output and sustained systemic pressures. For hypotensive patients with PE, adrenaline combines the beneficial effects of noradrenaline and dobutamine.

A meta-analysis of studies that performed chest angio-CT for suspected PE found that the incidence of subsegmental PE was 4.7% (95% CI, 2.5–7.6) and 9.4% (95% CI, 5.5–14.2) in patients undergoing single- and multidetector CT. respectively.39 There were no differences between the two groups in the incidence of thrombotic events during the first 3 months of follow-up when the patients did not receive anticoagulant therapy based on a negative angio-CT.

Incidental PE (unsuspected) is detected in approximately 2% of patients (most with cancer) in whom a chest CT is performed for reasons other than suspected PE. Some indirect evidence suggests that incidental PE worsens the prognosis in these patients.40 The recommendation for indicating anticoagulant therapy is more consistent when the incidental PE is associated with concomitant PE, the PE is lobar or in the main arteries and the risk of bleeding is not high.

• •

  Anticoagulant therapy is suggested in patients with incidental PE in a segmental, lobar or main location.

The presence of right-sided heart thrombi, particularly when they are mobile, significantly worsens the prognosis in patients with PE. Some indirect evidence suggests that fibrinolysis or surgical embolectomy is more effective than conventional anticoagulant therapy in these patients.

The appropriate duration of anticoagulant treatment requires a balance between the risk of recurrent VTE and the risk of bleeding complications.

The risk of recurrent VTE depends on (1) the efficacy of treatment of the acute episode, (2) a minimum duration of long-term treatment, (3) the possibility of the patient having an intrinsic risk factor for suffering a new VTE episode.

In a meta-analysis that included 2925 patients who had suffered a first episode of VTE not secondary to cancer, and who had received treatment of different durations, the risk of recurrence increased significantly for treatment durations less than 3 months (HR 1.52; 95% CI, 1.14–2.02).41 There were no differences in the risk of recurrence for treatment durations of 3 months compared to durations of 6 or more months (HR 1.19; 95% CI, 0.86–1.85). Based on this evidence, it is agreed that long-term treatment of patients with PE should have a minimum duration of 3 months.

In the aforementioned meta-analysis,41 the risk of recurrence was significantly lower for events caused by a temporary risk factor than for unprovoked events (HR 0.55; 95% CI, 0.41–0.74). In a systematic review42 that included 15 clinical trials, the risk of recurrent VTE during follow-up for patients with unprovoked VTE was 2.3–2.5 times higher than that of patients with provoked VTE, 7.9–10.6 times higher than patients with VTE provoked by a surgical risk factor, and 1.4–1.8 times higher than that of patients with VTE provoked by a non-surgical risk factor.

Although the literature is not consistent, for the purpose of decision-making on treatment duration, these guidelines stratify PE as provoked by a major risk factor, provoked by a minor risk factor, unprovoked or secondary to cancer (Fig. 4).

Fig. 4.

Risk of recurrence according to the pulmonary embolism precipitating factor.

(0.06MB).

These guidelines only make recommendations or suggestions for drugs approved in Spain for this indication.

In a review of 4 studies in which serial imaging tests were carried out, the percentage of patients with residual PE was 87% at 8 days, 68% at 6 weeks, 65% at 3 months, 57% at 6 months and 52% at 11 months.58 However, in a study of 673 patients from a single center who were followed-up for 3 months, the rate of symptomatic recurrent VTE was only 3%.59 Therefore, there is disagreement between the percentage of patients with residual thrombosis and the percentage of patients with recurrent VTE.

The incidence of symptomatic chronic thromboembolic pulmonary hypertension (CTEPH) in patients who have had an episode of PE varies according to the series and follow-up period, and has reached up to 3.8%.60 Some characteristics of the initial PE episode (age >70 years, young age, female gender, pulmonary systolic pressure >50mmHg, massive or sub-massive PE, recurrent or idiopathic PE) increase the risk of developing CTEPH.61

• •

  In patients with a history of PE, it is recommended not to use thoracic imaging tests to evaluate the persistence of residual thrombosis or reperfusion of the initial defects.

• •

  In patients with a history of PE and signs or symptoms suggestive of CTEPH, it is recommended to perform a follow-up transthoracic echocardiogram.

VTE is associated with an occult neoplasm in approximately 10% of patients. The Trousseau study62 compared the utility of extensive screening for cancer by thoracoabdominal CT and mammography (in women) with limited screening. The results did not show any differences in the incidence of cancer or mortality between the two patient groups. Extensive screening significantly increased the healthcare costs due to the additional diagnostic process associated with false positive results.

• •

  In patients with an episode of unprovoked PE, it is suggested not to perform specific tests searching for neoplasia if there are no clinical symptoms or basic complementary examinations that suggest the presence of this disease.