In the CICM Part II, questions about PE are frequent. Usually, they ask about the acute management of a haemodynamically unstable patient. The examiners seem to have a particular interest in thrombolysis. PE is represented in the following past paper questions:

• Question 16 from the first  paper of 2015 (role of thrombolysis)
• Question 2 from the first  paper of 2014 (diagnosis of PE)
• Question 5 from the second paper of 2012 (role of thrombolysis)
• Question 21 from the first  paper of 2006 (acute management of massive PE)
• Question 2c from the first paper of 2000 (acute management of massive PE)
• Question 2c from the first  paper of 2004 (acute management of massive PE)

Question 13.2 from the second paper of 2013 asked about the predisposing factors for PE, but has otherwise somehow ended up in the haematology section (probably because the first half of it was about heparin resistance).

The best references for answering such SAQs are probably the 2014 ESC guidelines and 2012 ACCP Guidelines. More recent evidence is discussed in review articles such as Barco & Konstantinides (2018), but this high-quality stuff is kept under lock and key by the greedy fat cats at Springer-Verlag. The freegan reader is instead directed to meta-analysis or review publications such as Jimenez et al (2018) or Sista et al (2017). The latter is a particularly good summary of the existing guidelines and would be easy to recommend as the best last-minute revision resource for the exam candidate.

Classification of pulmonary embolism

Classification of PE occurs on clinical grounds and is necessary to determine the course of management. There are two main systems: the American way (AHA guidelines), and the European way (ESC guidelines). The AHA classify their PEs according to "massiveness", i.e. massive v.s sub-massive and "low risk", on the basis of haemodynamic stability TTE findings and cardiac enzymes. 

• Massive PE: is defined as "a pulmonary embolism of sufficient size to cause systemic arterial hypotension". This requires immediate thrombolysis or embolectomy. According to Barco et al (2018), these people have a 30-day mortality in the order of 20-40% (i.e. similar to ARDS and severe sepsis).
• Sub-massive PE: no shock, but RV dysfunction or myocardial necrosis (i.e. elevated troponin); evidence for thrombolysis is not as clear cut (but as of 2014 many trials have come out in support of thrombolysis for this group; see below). Mortality risk is around 3%
• "Low risk": all other PEs; for anticoagulation only.

This system has the advantage that the massive PE patients are easily identified (i.e. they look like they are about to die). The ESC instead use a scoring system which incorporates multiple variables (totally different to the AHA) in order to risk-stratify PE patients. Their scales are the Pulmonary Embolism Severity Index (PESI) or the simplified sPESI score. The table below which summarises these scales comes from Sista et al (2017); in turn, they got their data from Carrier et al (2009) and Jimenez et al (2010). 

None of these scales incorporates anything about clot burden into their risk stratification. For example, the term "saddle" is often used to describe pulmonary emboli which straddle the bifurcation of the pulmonary trunk. According to Sista et al, having one of these probably has some influence on mortality, irrespective of the clinical features.

Risk factors for pulmonary embolism

There is a fine article discussing the predisposing factors for PE.

They can be combined into a big complicated table.

Risk Factors for Pulmonary Embolism

Inherited risk factors Antithrombin III deficiency Protein C deficiency Protein S deficiency Factor V Leiden mutation Hyperhomocysteinaemia

Acute risk factors Surgery Trauma Pregnancy Burns CVC, Swan-Ganz Spinal injury with paralysis Immobility

Chronic risk factors Smoking Obesity Oral contraception Hormone replacement therapy Malignancy Heart failure Lupus anticoagulant Increasing age

Investigations for pulmonary embolism

In Question 2 from the first  paper of 2014, the college wanted their candidates to discuss TTE, CTPA, troponin and D-dimers in the diagnosis of PE. The examiners complained that "candidates did not answer the question as asked", which is an odd comment, given that the question was worded vaguely ("Briefly outline the role of each of the following"). It was not a "compare and contrast" question, which might lead one to conclude that many candidates ended up comparing and contrasting the diagnostic tests in a tabulated manner (that's what we robotically do when posed with a question like this). However, with that said, the college answer to this question discusses the advantages and disadvantages of these investigations. So, here is a "compare and contrast" sort of table, which incorporates both the model college answer and the 2014 ESC guidelines.

Investigations for Pulmonary Embolism

Test	Rationale and advantages	Limitations and disadvantages
History and clinical examination	Cheap and rapidly available Screening for predisposing factors and associated features is effective in building pre-test probability (Well's rule or the Geneva score)	Clinical features of PE (eg. chest pain, dyspnoea, tachycardia) are highly non-specific.
ECG features of RV strain	These are - inverted T waves in V1-V4 - S1Q3T3 pattern - RBBB	Only present in very severe cases Frequently, sinus tachycardia is the only ECG feature
TTE	Rapid bedside test No contrast or radiation exposure May reveal RV dysfunction, which would be an indication for thrombolysis	Interpreter-dependent accuracy Only 30-40% with PE have suggestive changes; thus a negative result does not exclude PE
CTPA	Rapid and easily available May reveal alternative chest pathology New scanners are better at excluding PE The college mentions PIOPED II (2006): sensitivity of 83% specificity of 96%	Both contrast and radiation exposure Risk of transport to and from the scanner What if you find a small sub-segmental PE? What clinical significance does this finding have?
Troponin	New high-sensitivity tests are good at excluding RV injury Troponin is elevated in 30 – 50% with moderate/large PE. A raised troponin in PE is associated with a poorer prognosis.	Poor specificity: will be elevated in a number of situations apart from PE.
D-dimer	Good sensitivity: a negative D-dimer excludes PE. Good way of excluding PE in well patientsd with a low pre-test probability Three-month thromboembolic risk was less than 1% in patients left untreated on the basis of a negative test result.	Poor specificity: will be elevated in a number of situations apart from PE. Critically ill populations invariably have an elevated D-dimer. NNT (number needed to test) is 3 in the ED, but over 10 in other scenarios
Lung scintigraphy (V/Q scan)	IV injection of Tc99m-labelled macroaggregated albumin particles are used in the perfusion scan Xe133 gas is then used as a ventilation scan. Total radiation exposure (1.1mSv) is much lower than in a CTPA  it is safe to withhold anticoagulant  therapy in patients with a normal perfusion scan.	Not available everywhere Will not detect small PEs Not available for urgent pre-thrombolysis confirmation of PE

Thrombolysis for massive pulmonary embolism

This has come up in Question 5 from the second paper of 2012 and Question 16 from the first paper of 2015, but generally speaking, it comes up in all the PE questions. It therefore requires some time spent upon it. The most important references for this are the 2014 ESC guidelines and 2012 ACCP Guidelines.

First, a few words about the contraindications:

Contraindications for Thrombolysis in PE

Absolute

Relative

any prior intracranial hemorrhage, known structural intracranial cerebrovascular disease (eg, arteriovenous malformation), known malignant intracranial neoplasm, ischemic stroke within 3 months, suspected aortic dissection, active bleeding or bleeding diathesis, recent surgery encroaching on the spinal canal or brain, and recent significant closed-head or facial trauma with radiographic evidence of bony fracture or brain injury.

age >75 years; current use of anticoagulation; pregnancy; noncompressible vascular punctures; traumatic or prolonged cardiopulmonary resuscitation (>10 minutes); recent internal bleeding (within 2 to 4 weeks); history of chronic, severe, and poorly controlled hypertension; severe uncontrolled hypertension on presentation (systolic blood pressure >180 mm Hg or diastolic blood pressure >110 mm Hg); dementia; remote (>3 months) ischemic stroke; and major surgery within 3 weeks.

This is straight from PulmCrit, who got it straight from the AHA guidelines (Jaff et al, 2011)

Rationale for thrombolysis

PE, in general, is managed with anticoagulation alone. In addition to this, clot burden may be decreased by either systemic thrombolysis, catheter-directed thrombolysis, clot fragmentation or surgical embolectomy. The rationale for this is the improvement of pulmonary blood flow, and thus improved haemodynamic performance of the systemic circulation. A long-term benefit of thrombolysis is the prevention of severe pulmonary hypertension which inevitably develops in the wake of large-scale pulmonary emboli.

Thus, thrombolysis may:

• Decrease clot burden
• Improve systemic haemodynamics by improving LV filling
• Prevent further RV injury
• Prevent progression to pulmonary hypertension

Use of thrombolysis in PE:

The 2014 ESC guidelines make the following statements:

• Greatest benefit is seen when it is given within 48 hours; however some benefit is still seen even when it is given as late as 6-14 days after the acute event.
• Most patients respond within 36 hours
• The haemodynamic benefits of thrombolysis are confined to the first few days after the treatment  (at one week after treatment the benefits are no longer apparent)
• The dose of alteplase is 0.9mg/kg:
  • 10% of the dose over 1 minute;
  • 90% of the dose over the subsequent hour
  • Maximum dose is 90mg.

Evidence for benefits of systemic thrombolysis in massive PE

As the college has mentioned, the evidence for this practice is dodgy. The first randomised trial for this thrombolysis in massive PE comes from 1995, and was performed in a group of only 8 patients. All 4 patients receiving thrombolysis had survived without pulmonary hypertension at 2 years follow-up; whereas all the heparin-treated patients had died.

More evidence was gradually accumulated in support of this technique. Ten years later, thrombolysis is seen as a mandatory first-line step in the treatment of massive PE. It appears to reduce mortality by 55% according to a 2004 meta-analysis (including data from 748 patients). This meta-analysis (Wan et al, 2004) contained a small sub-group (n = 154) of haemodynamically unstable PE patients, and within this small subgroup thrombolytic therapy reduced the risk of their composite endpoint of death and recurrent PE (from 19% to 9%). On the basis of this, all current guidelines seem to recommend thrombolysis for haemodynamically unstable massive PE.

Evidence of risk of thrombolysis in massive PE

Why not thrombolyse everybody? Well:

• In the a 2004 meta-analysis, it was found that in comparison to heparin alone thrombolysis doubles the risk of major bleeding from 12% to 22% (which makes some sort of perverse sense, as it tends to halve PE-associated mortality).
• Real-world registry data suggests that this risk of bleeding is probably underestimated by clinical trial data, given how spotlessly perfect their patient selection is (whereas at the coalface, in the ED and ICU, a fair few patients are retrospectively, posthumously, discovered to have had some contraindication to thrombolysis).
• In short, the risk of death from bleeding is significant and should be presented to the patient and family as a real possibility.

Evidence for benefits in the sub-massive group:

• Some trials were published around 2013-2014, raising suspicion that people with submassive PE should also be offered thrombolysis.
• PEITHO trial (2014) - multi-centre RCT, 1005 patients randomised
  • Received either heparin alone or thrombolysis plus heparin. The college quoted it in Question 16 from the first paper of 2015 (presumably - they referred to it as "a recent NEJM article").
  • Inclusion criteria were RV dysfunction and a raised troponin.
  • Haemodynamics improved, but with increased risk of major haemorrhage (6.3% vs 1.2%) and stroke (2.4% vs 0.2%).
  • 7-day mortality was essentially unaffected (1.2% vs 1.8%) in spite of this. 
• TOPCOAT trial (2014) - multi-centre RCT, 83 patients randomised
  • Patients treated with tenecteplase had fewer adverse outcomes, better functional capacity, and greater quality of life at 3 months
• MOPETT trial (2013) - single-centre RCT, 121 patients randomised
  • Question 16 from the first paper of 2015 also mentions that "lower dose (“safe dose”) thrombolysis has been investigated" in the group of patients with a large clot burden. Judging by the evidence they quote, they were talking about the MOPETT trial. 
  • Primary endpoints were pulmonary hypertension and the composite end point of pulmonary hypertension and recurrent PE at 28 months.
  • Mortality was a secondary endpoint.
  • Bottom line:
    • Much less pulmonary hypertension was observed at 28 months
      (16% vs 57%)
    • Lower mortality with thrombolysis (1.6% vs 10%) - but the study was not powered to detect this; in fact the numbers were 1 and 6 patients respectively, and so statistical significance was not reached.

Evidence for risks from thrombolysis in the sub-massive group

• In the submassive group, the risk was perhaps greater than benefit.  Question 16 from the first paper of 2015 mentions some sort of "recent meta-analysis" looking at the use of thrombolysis in haemodynamically stable patients. They probably meant this paper by Riera-Mestre et al (2014).
• The authors found that there was a very slight improvement in mortality; NNT to avoid one death was 125 patients.
• In contrast, the numbers needed to harm (NNH) for a major bleed were 27, and for an intracranial haemorrhage were 91.

Evidence surrounding thrombolysis in cardiac arrest

• The Australian Resuscitation Council guidelines ( Guideline 11.10, 2010) offers some lukewarm Class B (expert opinion based) recommendations, which are as follows:
  • Consider thrombolysis if PE is suspected as the cause of arrest
  • If you gave the alteplase, continue CPR for 30 minutes.
  • How much of it? A recent case report (Gabrilovich, 2013) mentions the use of a dose as small as 10mg, which was still effective in relieving the mechanical obstruction.
• These recommendations are made on the basis of small case series, and the guidelines statement has not been updated since 2010.
• In 2015, AHA's "Special Circumstances" guidelines statetement threw caution to the wind by recommending thrombolysis (Class IIA), waving aside objections about bleeding risk: "standard contraindications to thrombolysis may be superseded by the need for potentially lifesaving intervention".

Surgical embolectomy as an alternative

Surgical embolectomy is a possibility, but good outcomes are only seen when a strong and organised purpose-built team is looking after the process, rather than some ad-hoc on-call cardiothoracic surgeon. Furthermore, the patients need to be carefully selected, and the sort of patient most in need of embolectomy are also the patients least likely to be selected for surgery (i.e. they are in florid cardiogenic shock, or worse yet they failed thrombolysis and are now full of alteplase). Apparently, in this enlightened age, the rate of survival after surgical embolectomy is 85% at 1 month. 

Conclusion

For the purposes of answering questions in the final CICM exam, thrombolysis is an important and potentially life-saving step in the management of massive PE. If the patient does not meet criteria for thrombolysis, urgent percutaneous or open surgical embolectomy should be considered. The trainee who wishes to regurgitate this material may be well served by a visual mnemonic aid such as this "PE algorithm for Weill Cornell Medical College PE Advanced Care Team" from Sista et al 92017), which is based on the AHA risk stratification algorithm (Jaff et al, 2011).