Pulmonary embolism (PE) is one of those diagnoses that finds its way into a lot of differential diagnoses. In the era of reasonably easy-but-occasionally-argue-with-radiologist to get CTPAs, the diagnosis has increased. An old study now, but in 2011 Wiener et al showed that the use of CTPA has increased the diagnosis but with no difference in mortality.
Unsurprisingly PE and, more broadly, venous thromboembolic diseases are a vast area of research and innovation. We all learn about Virchow’s triad at medical school, then we learn the probability and stratification scores, Wells, Geneva, PERC, PESI etc etc. Our language has changed over time from massive and sub-massive, to high risk and intermediate risk. Our treatments are advancing from systemic thrombolysis to catheter-directed thrombolysis and mechanical thrombectomy. We’ve even developed special teams to assess these patients (pulmonary embolism response teams; PERT).
In the UK we have NICE and BTS guidelines that provide a framework for our assessment and treatment of PE. Just to provide context and allow readers to compare and contrast, main features of the NICE guideline are:
Diagnosis: Physical exam, two-level Wells score, PERC score, age-adjusted D-Dimer, CTPA
Treatment: Outpatient (low-risk), systemic thrombolysis if haemodynamically unstable, consider catheter-directed thrombolysis in some patients, treatment dose low-molecular weight heparin for stable patients.
The 2026 AHA/ACC multi-society guideline is the first US guideline that covers the whole pathway from diagnosis to therapy. It’s a very long document, but very thorough. I particularly like the AHA guidelines in general for the clear way in which they present the strength of evidence for their recommendations. This post is a (my) summary and not intended to replace reading them for yourself! It will focus on what I think are some of the big game-changing features of how we might manage a patient with a pulmonary embolism.
What’s genuinely new?
- A–E Clinical Categories
- Greater emphasis on respiratory failure states
- Clearer outpatient endorsement
- Strong DOAC preference
- Formalisation of RV imaging reporting
- Explicit PERT recommendation
What’s reinforced?
- Pre-test probability first
- Smarter D-dimer use
- CTPA as first-line imaging
- LMWH > UFH
- Extended anticoagulation when appropriate
Let’s move on from pretending PE is just ‘massive’ or ‘not’
The guideline introduces Acute PE Clinical Categories A to E. The point is simple; blood pressure alone is an imprecise way to decide who is sick. You can be extremely unwell with a ‘normal’ BP if you’re compensating hard, hypoxic, tachycardic, and sliding into respiratory failure. Actionable recognition of pre-shock physiology is an important step. For those of you familiar with the management cardiogenic shock you will be aware of the SCAI classification of shock which similarly is graded A-E; perhaps this isn’t a coincidence that both guidelines share similar terminology. What this does do is provide a more explicit way of describing the systemic insults caused by a PE in a more nuanced way than massive/sub-massive or intermediate/high risk.
The guideline has a sensible approach to risk diagnostics. If you do one thing differently after reading this, let it be this: structure your pre-test probability and use your D-dimer intelligently. That is how you avoid turning CTPA into a reflex. The guideline suggests starting with clinical pre-test probability (Wells or revised Geneva are fine). If low risk, consider PERC to step away from the D-dimer treadmill and if low or intermediate probability, an age-adjusted D-dimer can safely rule out PE in many. The YEARS algorithm (and pregnancy-adapted YEARS) are reasonable ways to reduce imaging when used properly.
CTPA remains the default test when you need imaging. V/Q still has a role when CT is unsuitable, and V/Q SPECT is preferable to planar V/Q where it’s available. Echo is not a PE rule-in or rule-out test – but it is useful once PE is confirmed, because it tells you about the right ventricle. A positive CTPA (or high-probability V/Q) is sufficient for diagnosis and if PE is confirmed on CTPA, the report should include the numerical RV/LV ratio for risk stratification. Echo is ubiquitous and a lot of inpatients diagnosed with PE will get one. The report should include
- – RV/LV ratio
- – TAPSE
- – RV free wall motion
- – Estimated RVSP
- – Septal flattening
- – IVC collapse
The guideline is explicit about discharge. If you have the infrastructure (follow-up, anticoagulation access, and safety netting), you should not be filling beds with people in Categories A and B. This probably mirrors what we do in the UK anyway.
- Category A: discharge from the ED is reasonable and hospitalisation is not required.
- Category B: early discharge is generally recommended.
- Category C to E: hospitalise and match level of care to physiology and trajectory.
When you need parenteral anticoagulation, LMWH wins over UFH in most cases. If you can give a DOAC, do that rather than warfarin unless there is a reason not to.
- LMWH is preferred over UFH when parenteral anticoagulation is required.
- DOACs are preferred over VKAs for eligible patients.
- After a first PE without a major reversible risk factor (or with persistent risk), extended anticoagulation beyond the initial 3-6 months is recommended.
In critical care it is sometimes the case that additional caution is required when commencing any anticoagulation, in which case UFH may be preferred (high bleeding risk, considering thrombolysis, anticipating procedures, renal failure, potential ECMO)
- The severe end of PE care is messy because the evidence is messy. The guideline supports advanced therapies in the patients who are crashing (Category E1) and gives permission to consider them in selected Category D patients who look like they’re heading the wrong way. Previously the approach was ‘intermediate-high risk, observe closely’, now there is support for escalation before hypotension develops, in the right patient.
- Systemic thrombolysis, catheter-directed thrombolysis, thrombectomy, or surgical embolectomy may be reasonable in Category E1 and selected D1-2 patients.
- If you have a PE response team (PERT), use it. If you do not, this guideline is another reason to build one.
The post-PE clinic visit is where we find the people who are not bouncing back. The guideline recommends checking symptoms and function at follow-up visits for at least a year, with a low threshold to consider chronic thromboembolic pulmonary disease (CTEPD) in those with persistent dyspnoea.
Cheers!
Rich
Wiener R, Schwartz LM, Woloshin S. Time trends in pulmonary embolism in the united states: Evidence of overdiagnosis. Archives of Internal Medicine. 2011;171(9):831-837.
Creager MA, Barnes GD, Giri J, et al. 2026 AHA/ACC/ACCP/ACEP/CHEST/SCAI/SHM/SIR/SVM/SVN guideline for the evaluation and management of acute pulmonary embolism in adults. Circulation. 2026;153:e00-e00.
