Management of pulmonary
embolism
John A Strange David Pilcher
Abstract
Pulmonary embolism (PE) is a common condition with significant mortal-ity and morbidmortal-ity. Its occurrence frequently triggers referral to critical care services. Patients within critical care environments are also at elevated risk of developing venous thrombo-embolism and PE. This highlights the need for critical care clinicians to be confident in their approach to the patient with PE. Furthermore, the co-morbid conditions in this patient group may present additional challenges both in diagnosis (e.g. safe ac-cess to radiology) and management (e.g. relative contraindication to anti-coagulation/thrombolysis in trauma or intracranial haemorrhage). This brief review summarizes the contemporary evidence base regarding both diagnosis and treatment strategies and draws upon this to suggest a simple algorithm for investigation, risk stratification and management, particularly tailored to patients within a critical care setting.
Keywords Anticoagulation; computed tomography pulmonary angio-gram (CTPA); embolectomy; IVC filter; massive pulmonary embolism; pulmonary embolism; submassive pulmonary embolism; thrombolysis; venous thrombo-embolism
Royal College of Anaesthetists CPD matrix: 2C01, 2C03, 2C04, 1B00, 2C00
Definitions
The term pulmonary embolism (PE) encompasses the movement of abnormal material to the pulmonary arteries and through the pulmonary vasculature such that it obstructs blood flow; exam-ples include embolism of air/gas, fat and thrombus. The most common cause of PE is the migration of thrombus from veins (or right heart) to the pulmonary arterial tree. Other forms of PE are beyond the scope of this article.
Diagnostic considerations
PE is a commonly considered but relatively infrequently diagnosed condition in hospitalized patients. This is unsur-prising when one considers that the clinical presentation of PE varies from breathlessness in isolation to sudden death, mak-ing clinical assessments insensitive and highly unspecific (Table 1, signs, symptoms and differential diagnosis of PE).
Consideration of risk factors contributing to the development of venous thrombo-embolism (VTE) and PE (Table 2, Virch-ow’s triad, primary and secondary hypercoagulable states) may improve diagnostic rates, but a missed diagnosis, or the inappropriate application of treatment both carry considerable risks. The use of biomarkers and choice of imaging modalities can be guided by clinical decision rule (CDR) systems of which the most widely reported are the Well’s score and the Geneva score. These aim to stratify risk and focus resources on those most likely to benefit. The evolution of these tests and scoring systems has resulted in various approaches to investigating and treating possible PE; a suggested scheme geared more particularly to the critical care environment is outlined in Figure 1.
Investigation/severity assessment Bedside investigations
An arterial blood gas analysis (ABG) demonstrating hypoxia (with widened alveolarearterial oxygen gradient e Aea gradient) and hypocapnia with a concomitant increase in end-tidal CO2 gradient is suggestive of PE but lacks specificity,
equally a normal blood gas does not exclude PE. A normal ECG is found in one-third of cases, other findings include sinus tachy-cardia, T-wave inversion, right bundle branch block, p-pulmo-nale and other features suggestive of right ventricular strain. The classically described deep S wave in lead I, with a Q-wave and inverted T-wave in lead III (so called S1Q3T3) is rare and sug-gests more significant disease. The ECG is also important in screening for differential diagnoses. The chest X-ray may help to exclude common differentials such as pneumothorax, pneu-monia or pleural effusion. Identifying more specific abnormal-ities such as oligaemia and abnormal pulmonary artery contours is generally the preserve of radiologists.
Biomarkers
D-dimers are cross-linked fibrin degradation products. Serum levels are elevated in VTE and therefore PE. They have poor specificity and poor positive predictive value for PE. The most useful D-dimer result is a negative, which makes the diagnosis of
Learning objectives
After reading this article you should be able to:
C describe the disease entity of venous thrombo-embolism/pul-monary embolism (VTE/PE) and outline risk factors for its development, recognizing the varied spectrum of presentation in PE
C outline an appropriate diagnostic strategy for the evaluation of possible PE, risk stratifying according to clinical presentation and investigations
C understand the various treatment options for PE and how they should be utilized in light of the risk stratification and diag-nostic findings
John A StrangeMB BCh FFARCSI DIBICMis a Senior Registrar in Intensive Care Medicine at The Alfred Hospital, Melbourne, Australia. Conflicts of interest: none declared.
David PilcherMB BS FRACP MRCP FCICMis a Consultant Specialist in
Intensive Care Medicine at The Alfred Hospital, Melbourne, Australia. Conflicts of interest: none declared.
PE unlikely, although a high D-dimer concentration is an inde-pendent predictive factor associated with mortality.1
Measurements of troponin, brain natriuretic peptide (BNP) or NT-terminal pro-BNP (NT-pro-BNP) although not useful in diagnosing PE may stratify risk and determine prognosis in confirmed PE.2 Raised troponin predicts haemodynamic insta-bility in non-massive PE and increased risk of death regardless of PE size. In proven PE, low levels of BNP and NT-pro-BNP correlate with good outcomes,2 the latter is likely a superior predictor of outcome than troponin.3
Imaging
There is no ideal imaging modality in PE, studies show that confidence in any result can be improved by first assessing the pre-test probability of there being a PE. Unfortunately these studies are not representative of the critical care population, where a majority of patients have high pre-test probability of PE, in each scoring system.
Computed tomography
Computed tomography pulmonary angiography (CTPA) scan-ning, especially the multi-detector scanner (MD-CTPA), has now largely replaced lung ventilationeperfusion (V/Q) scanning as the cost-effective and reliable imaging procedure of choice in patients with suspected PE.4The CTPA scan has the advantage of greater diagnostic accuracy, being readily available at most hospitals, more rapid image-acquisition time, and the possibility of making an alternative diagnosis (Figure 2). High-resolution images to the level of segmental and in some cases sub-segmental pulmonary arteries can be obtained in a short time-period (often a single breath-hold). When compared to conven-tional angiography it appears reliable, with excellent sensitivity, specificity and accuracy.5The CTPA scan can also be used to assess the severity of PE. An increased right ventricular/left ventricular (RV/LV) ratio6and clot in the proximal branches of the pulmonary artery correlate with the clinical severity of PE. It is therefore recommended that the CTPA scan should be the principal imaging test for patients with high and moderate probability of PE.
Although inconclusive CTPA scans occur in around 10%, a negative CTPA result means that withholding anticoagulant therapy is safe. An emerging problem of CTPA scanning, however, is the increased detection (around 10%) of small peripheral emboli in subsegmental pulmonary arteries due to better visualization of these arteries. The clinical significance of these findings in critically ill patients is unknown, however these are usually unlikely to lead to a bad outcome if left untreated.
Ventilation/perfusion scans
Lung ventilation/perfusion scanning demonstrates regional abnormalities in the distribution of inhaled radioactive gas, and injected radioactive contrast agent respectively. Matched or mismatched defects are interpreted, and reported as low, intermediate or high probability for PE. This technique is still widely and effectively employed where CTPA is unavailable or contraindicated (such as intravenous contrast allergy) but is limited by the large proportion of patients with intermediate or low probability results e leaving clinical uncertainty as to who to treat (as many as 40% of these patients will have had a PE).
Echocardiography
Echocardiograms have poor negative predictive value (up to 50% of clots missed) but can show pathognomonic patterns for PE, and may identify clot in the right ventricle or proximal pulmo-nary arteries (generally only on trans-oesophageal studies). It is of greatest utility in the most severe cases, where haemodynamic instability may prevent safe transport to CT. In these patients trans-thoracic echocardiography (TTE) can be employed at the bedside to investigate the cause of haemodynamic instability, to exclude other diagnoses (tamponade, myocardial infarction, aortic dissection) and assess severity of known PE (the presence of RV dysfunction in any patient implies a more grave prog-nosis). In a selected group of patients, where there is a high index of clinical suspicion for PE, findings on TTE may be sufficiently compelling to allow the rapid institution of potentially life-saving treatments such as thrombolysis.
Clinical assessments aiding in the diagnosis of PE History
Previous DVT/PE
Family history of DVT/PE/sudden death History/family history of thrombophilia Secondary hypercoagulability (Table 2) Signs
Increasing risk of massive pulmonary embolism None
Tachycardia Moderate fever
RV dysfunction (raised JVP, parasternal heave, loud P2) Hypotension
Skin mottling Peripheral cyanosis Central cyanosis
Cardiovascular collapse/arrest
NB also important to examine for signs of DVT in limbs Differential diagnoses
Acute myocardial infarction Acute pulmonary oedema Asthma/exacerbations of COPD Pericardial tamponade Pleural effusion Fat embolism Pneumothorax Aortic dissection Rib fracture Anxiety Symptoms
Dyspnoea (most common) Pleuritic chest pain
Haemoptysis (late sign, lung infarction) Syncope
COPD, chronic obstructive pulmonary disease; DVT, deep vein thrombosis; JVP, jugular venous pressure; PE, pulmonary embolism; RV, right ventricular. Table 1
Disease spectrum
Massive pulmonary embolism
This is defined as PE causing sustained (>15 minutes) hypo-tension (systolic BP<90 mmHg) or a sustained significant drop in systolic blood pressure (>40 mmHg). Even when treated this condition has mortality exceeding 25% (65% if cardiopulmonary resuscitation is required). Acute RV failure is a very common feature. There may only be a brief window of opportunity to identify and address the condition. Patients remain at significant risk of death for several days after an event.
Submassive PE
Submassive PE typically describes other acute PEs, where pa-tients have evidence of RV dysfunction (best confirmed with echocardiography, but also possibly shown on CT) but have a normal blood pressure. This subgroup has up to four times the mortality risk and increased rates of recurrence, they may also go on to develop shock or RV thrombus.7Prevention of recurrence is a priority but therapies to remove clot such as thrombolysis may have a role in this group. All other patients with PE are haemodynamically stable and have normal RV function, the majority tend to follow an uneventful course (<2% mortality) unless further PE occurs.
Treatment
The major principles of management are aimed at either prevention of further embolization and thrombosis (anticoagulation and inferior vena caval filters), removal of established clot (thrombol-ysis and embolectomy) and concurrent haemodynamic support.
Anticoagulation
Anticoagulation decreases mortality in patients with PE. The risk of a major bleeding event secondary to anticoagulation is lower (<3%)8than the risk of death from undiagnosed PE (30%). This suggests that all confirmed cases and those with high clinical probability of PE should be anticoagulated e unless there is a compelling contraindication. Low-molecular-weight heparins (LMWH) are as effective and safe as unfractionated heparin and offer several advantages, including a longer half-life, increased bioavailability, a more predictable doseeresponse and fewer re-quirements for monitoring and dose adjustments. They should be readily used in the stable patient with PE. Unfractionated heparin offers rapid therapeutic dosing when weight based protocols are employed (Table 3e suggested dosing, complications and con-traindications of heparin therapy), and can be easily therapeuti-cally monitored and reversed with protamine if significant bleeding occurs. The predominant complication of both unfractionated heparin and LMWHs is bleeding. Both bleeding complications and heparin-induced thrombotic thrombocytopenia syndrome (HITTS) appear to be less common when LMWH is used.
Oral anticoagulants should be commenced when possible to allow LMWH or unfractionated heparin to cease. For warfarin, which has an initial procoagulant effect, this is when the inter-national normalized ratio (INR) is greater than 2.0. New oral anticoagulants, including rivaroxaban (competitively binds acti-vated factor X) and dabigatran (direct inhibitor of thrombin) have been developed. These drugs have more rapid onset of action and more predictable anticoagulant effects potentially allowing fixed dosing without routine laboratory coagulation Pathophysiology and risk factors for VTE
Virchow’s triadepathophysiology of intravenous clot formation Alterations in blood flow (stasis)
Vascular endothelial injury (vein wall damage)
Alterations in blood constituents (inherited and acquired hypercoagulable states)
Primary hypercoagulable states (inherited thrombophilias) Secondary (acquired) hypercoagulable states Factor V Leiden mutation (activated protein C resistance) found in>20% of
patients with confirmed VTE
Prothrombin gene mutation (probably at least as common as Factor V Leiden) Protein S deficiency
Protein C deficiency
The lupus anticoagulant (antiphospholipid antibody) Antithrombin III deficiency
Dysfibrinogenaemias (rare)
Inherited vena cava abnormalities (rare) Hyperhomocystinaemia (rare)
Immobility
Recent or current hospitalization Surgery within past 3 months Malignancy
Infection within past 3 months Pregnancy and puerperium
Trauma (especially with limb paralysis) Burns
Oestrogens (OCP/HRT), tamoxifen
Cardiovascular risk factors (smoking, obesity, hypertension, diabetes etc)
IV drug abuse, central venous devices Increasing age
Chronic renal failure or nephrotic syndrome Hyperviscosity syndromes (myeloma etc) Initially after commencement of warfarin without heparin/LMWH cover
HRT, hormone replacement therapy; LMWH, low-molecular-weight heparin; OCP. oral contraceptive pill; VTE, venous thrombo-embolism. Table 2
monitoring, however their place in critically ill patients remains to be established.
Inferior vena caval (IVC) filters
IVC filters are mechanical devices percutaneously inserted into the distal vena cava with the intention of physically entrapping clot as it embolizes from leg or pelvic veins (the most common sites of deep vein thrombosis (DVT)). They may be used to prevent further embolization or as primary prevention, although safety and efficacy have not been firmly established. An IVC filter is indicated for patients where anticoagulation is contraindicated and those who experience recurrent PE despite adequate anti-coagulation.9They may have a role in patients with massive or submassive PE who have undergone open surgical embolectomy or thrombolysis. Insertion is usually performed percutaneously in a radiology department, but it can be done at the bedside. They lower early recurrence but increase long-term DVT recur-rence rates. Newer retrievable designs may be more efficacious and safer if removed at the appropriate time (Table 4suggested role of IVC filters in PE).
Thrombolysis
There is a general consensus advocating the use of thrombolytics in massive PE with significant or ongoing haemodynamic insta-bility. Their use may result in dramatic improvements in hae-modynamics and oxygenation. However when compared to IV heparin therapy, clot resolution is similar after only a few days. Patient registry data suggest that mortality and recurrence of PE is lower when thrombolytics are used rather than heparin alone, however no head-to-head randomized trial has demonstrated a mortality difference. A meta-analysis of studies comparing thrombolysis with heparin showed a trend toward superiority with thrombolytics, this trend became a significant reduction in mortality when studies not including massive PE were excluded from the analysis. Thrombolysis of patients with submassive PE significantly reduces their chance of deteriorating to a degree that requires ICU admission.10
No study has shown a significant difference in the efficacy of different thrombolytic agentse a suggested protocol of two 10-unit doses of reteplase, separated by 30 minutes, is effective and simple to implement. There is no evidence that using a central INITIAL CLINICAL ASSESSMENT
Intermediate or high clinical probability of PE (using Clinical Decision Rule system)
Haemodynamically stable Haemodynamically unstable
INITIAL DIAGNOSTIC TEST (V/Q if CTPA contraindicated)Multi-detector CTPA scan Echocardiograph
SEVERITY ASSESSMENT FOR STRATIFICATION OF TREATMENT PE confirmed if thrombus in RV or PA PE likely if RV dysfunction If PE confirmed, stratify risk using
CTPA, blood tests and echocardiograph
RV/LV ratio proximal clot in main PA, Segmental or sub-segmental PE and CONFIRMATORY TESTS CTPA scan raised troponin, BNP or NT pro-BNP normal troponin, BNP or NT pro-BNP
Echocardiograph CTPA scan if safe Echocardiograph
to assess RV function
TREATMENT Anticoagulant therapy only Thrombolytic therapy / Embolectomy if RV dysfunctionAnticoagulant therapy
↑
BNP, brain natriuretic peptide; CTPA, computed tomography pulmonary angiography; LV, left ventricular; NT, NT-terminal; PA, pulmonary artery; PE, pulmonary embolism; RV, right ventricular.
venous or PA catheter for administering thrombolytics confers a treatment advantage or any reduction in bleeding complications. Delaying therapy in an unstable patient to gain central access cannot be justified and may result in arterial injury or pneumo-thorax (relative contraindications to thrombolysis).
Concerns around haemorrhagic complications are justified as major bleeding occurs in 10% of patients thrombolysed for PE (versus<3% with heparin infusion alone), though intracerebral haemorrhage is less common than might be feared (0.9%). Massive PE and submassive PE with RV dysfunction both carry significant mortality risk, employing thrombolytics in their
treatment can often be justified in the face of relative contrain-dication (such as recent surgery), the challenge is to balance an individual patient’s risks from acute PE with those from fibri-nolytic treatment itself.
Embolectomy
Mortality from surgical embolectomy for PE ranges from 25% to 50%. There remains a role for surgery but this is probably limited to patients with massive PE, treated within cardio-thoracic centres or where thrombolytic therapy is contra-indicated. Surgery has been advocated for removal of free floating RV thrombus.
Alternative approaches include, percutaneous embolectomy, catheter direct therapy with targeted thrombolysis or rheolysis (physical disruption of clot), or a combination of the above. These are emerging techniques without a large body of research to characterize their usefulness, mortality with these techniques remains above 20%.
Concurrent haemodynamic support
Shocked patients (massive PE) need urgent supportive care in parallel to definitive and preventative treatment. Insertion of a PA catheter may guide therapy with vasoactive agents and monitor response to thrombolysis but should not delay definitive treatment.
Initially volume loading with IV fluids may improve haemo-dynamics, however excessive intervention may overload an already stressed or injured right ventricle, leading to further decline and risking left ventricular failuree treatment should be titrated against clinical response.
Figure 2 Three images from a single computed tomography pulmonary angiography (CTPA) study performed with a high clinical suspicion of pulmonary embolism (PE). Image 1 demonstrates large PEs in the proximal right and inferior left pulmonary artery. Image 2 shows a significant concurrent pneu-mothorax. Image 3 demonstrates a right ventricular/left ventricular (RV/LV) ratio>1 signifying significant RV dysfunction. Together these images show the high utility of CTPA in diagnosis/exclusion of PE, diagnosis/exclusion of differential diagnoses, and in risk stratifying a patient so as to guide therapy.
Suggested dosing, heparin therapy
Heparin loading dose Initial maintenance infusion 80 units/kg (caution in
obesity/anasarca)
18 units/kg/hour
Six-hourly monitoring of activated partial thromboplastin time (APTT)e suggested dosing adjustments:
APTT Dose change (units/kg/hour)
Heparin re-bolus Repeat APTT
<35 þ4 80 units/kg At 6 hours
35e45 þ2 40 units/kg At 6 hours
46e70 No change None At 6 hours
71e90 2 None At 6 hours
>90 3 Stop infusion 1 hour At 6 hours Table 3
In massive PE blood supply to the RV can become compro-mised, clot causes increased pulmonary vascular resistance (increased PAPs) with right ventricular overload (increased
central venous pressure) leading to increased mean right ven-tricular pressure (RVPm) this is compounded by possible LVF
causing decreased mean arterial pressure (MAP) (right heart output¼ left heart output), the result is a drop in right ventric-ular coronary perfusion pressure (RVCPP ¼ MAP RVPm).
Values below 30 mmHg lead to significant cardiac ischaemia, worsening RV failure, shock and possibly death.
The pathogenesis outlined above suggests that therapy be aimed at increasing MAP (i.e. filling and pressor support) and at reducing RVPm (i.e. reducing PAPs/pulmonary vascular
resis-tance). The latter can be achieved with selective pulmonary va-sodilators (e.g. nitric oxide or inhaled prostacyclin) though these may result in systemic hypotension. Noradrenaline can coun-teract these concerns to a degree and is also the preferred ino-trope for its concomitant beneficiala- and b-adrenergic effects on MAP and cardiac output respectively. Caution should be exer-cised when using inotropes that have systemic vasodilatory ef-fects (such as milrinone or dobutamine), which may increase cardiac output without increasing MAP and therefore not significantly improve RVCPP.
Extracorporeal membrane oxygenation (ECMO) is an alter-native form of mechanical assistance, which may be available in specialized institutions. It should be considered for patients with PE who have had cardiopulmonary arrest or have very severe shock.
VTE prophylaxis
For inpatients (and a selected group of patients at home) pro-phylaxis of VTE is the most important aspect of PE management. Multiple studies have robustly demonstrated the need for, and effectiveness of prophylactic strategies in preventing DVT and PE across patient groups.9 Local protocols will vary but good
evidence exists for pharmacological prophylaxis with SC heparin, LMWHs or fondaparinux. IVC filters prevent recurrent PE over the long term, but at the cost of increased DVT risk. Mechanical approaches with graduated compression stockings and inter-mittent pneumatic compression devices are also options to be combined with anticoagulants or in their place if they are
contraindicated. A
REFERENCES
1 Grau E, Tenias JM, Soto MJ, et al. D-dimer levels correlate with mor-tality in patients with acute pulmonary embolism: findings from the RIETE registry. Crit Care Med 2007; 35: 1937e41.
2 Konstantinides S, Goldhaber SZ. Pulmonary embolism: risk assess-ment and manageassess-ment. Eur Heart J 2012; 33: 3014e22.
3 Maziere F, Birolleau S, Medimagh S, et al. Comparison of troponin I and N-terminal-pro B-type natriuretic peptide for risk stratification in patients with pulmonary embolism. Eur J Emerg Med 2007; 14: 207e11.
4 Anderson DR, Kahn SR, Rodger MA, et al. Computed tomographic pulmonary angiography vs ventilation-perfusion lung scanning in patients with suspected pulmonary embolism: a randomized controlled trial. J Am Med Assoc 2007; 298: 2743e53. 5 Winer-Muram HT, Rydberg J, Johnson MS, et al. Suspected acute
pulmonary embolism: evaluation with multi-detector row CT versus digital subtraction pulmonary arteriography. Radiology 2004; 233: 806e15.
6 Furlan A, Aghayev A, Chang CC, et al. Short-term mortality in acute pulmonary embolism: clot burden and signs of right heart dysfunction at CT pulmonary angiography. Radiology 2012; 265: 283e93.
7 Kucher N, Rossi E, De Rosa M, Goldhaber SZ. Prognostic role of echocardiography among patients with acute pulmonary embolism and a systolic arterial pressure of 90 mm Hg or higher. Arch Intern Med 2005; 165: 1777e81.
8 Schulman S, Beyth RJ, Kearon C, Levine MN. Hemorrhagic complica-tions of anticoagulant and thrombolytic treatment: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edition). Chest 2008; 133: 257Se98.
9 Kahn SR, Lim W, Dunn AS, et al. Prevention of VTE in nonsur-gical patients: antithrombotic therapy and prevention of throm-bosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012; 141(suppl 2): e195Se226S.
10 Konstantinides S, Geibel A, Heusel G, Heinrich F, Kasper W. Heparin plus alteplase compared with heparin alone in patients with submassive pulmonary embolism. N Engl J Med 2002; 347: 1143e50.
Suggested indications for insertion of inferior vena caval filter
Absolute indications
New or recurrent pulmonary embolism despite anticoagulation Contraindications to anticoagulation
Complications resulting from anticoagulation Other recommended indications
Following thrombolytic therapy Post-surgical embolectomy Extensive deep vein thrombosis Table 4
