Pulmonary Embolism in 2026: A Practical Approach to Diagnosis, Risk Stratification and Management

From clinical suspicion and D-dimer testing to the new AHA/ACC PE categories, anticoagulation, reperfusion therapy and follow-up

Pulmonary embolism (PE) ranges from an incidental finding in an asymptomatic patient to catastrophic obstructive shock and cardiac arrest. The 2026 guideline introduces a more granular approach to this spectrum through AHA/ACC Acute Pulmonary Embolism Clinical Categories A–E, integrating clinical severity, respiratory compromise, haemodynamics, cardiac biomarkers and right ventricular (RV) function.

The practical message is simple: diagnosing PE is only the first step. Once PE is confirmed, the clinician must rapidly determine how sick the patient is, where they should be treated, and whether anticoagulation alone is sufficient or advanced therapy should be considered.

When should acute pulmonary embolism be suspected?

PE should be considered when compatible symptoms or signs occur in an appropriate clinical context. The most common symptoms are dyspnoea and pleuritic chest pain. Haemoptysis, syncope and shock occur less frequently.

The examination should specifically look for evidence of haemodynamic or respiratory compromise and concurrent deep-vein thrombosis. Tachycardia or hypotension may indicate haemodynamic compromise, while tachypnoea and hypoxaemia suggest significant impairment of gas exchange.

Risk assessment should include factors such as recent surgery, hospitalisation, immobility, pregnancy, oestrogen exposure, trauma, cancer, inflammatory disease and inherited or acquired thrombophilia.

The guideline recommends a targeted history and comprehensive physical examination to establish the clinical pre-test probability before ordering diagnostic imaging.

How should suspected PE be investigated?

The diagnostic strategy begins with clinical probability, not with CT pulmonary angiography (CTPA).

Practical diagnostic algorithm

A positive CTPA confirms PE. A high-probability V/Q scan is also sufficient to establish the diagnosis.

What is the role of PERC?

The Pulmonary Embolism Rule-Out Criteria can be used in patients with a very low clinical probability of PE, approximately <15%.

All of the following need to be absent:

When the clinical probability is very low and all PERC criteria are negative, no further investigation for PE is required.

How should D-dimer be used in 2026?

D-dimer should primarily be used in patients with low or intermediate clinical probability, rather than as an indiscriminate screening test.

The guideline supports two important approaches.

Age-adjusted D-dimer

In patients with low or intermediate clinical probability, an age-adjusted D-dimer below the threshold can exclude PE without imaging.

For fibrinogen-equivalent-unit assays:

Age-adjusted threshold = age × 10 μg/L

This strategy reduces unnecessary pulmonary imaging while maintaining a low diagnostic failure rate.

YEARS

YEARS assesses three clinical features:

• Clinical signs of DVT
• Haemoptysis
• PE considered the most likely diagnosis

The guideline uses:

A D-dimer below the applicable threshold can exclude PE without imaging. Pregnancy-adapted YEARS may also be considered in pregnant adults.

What is the preferred imaging test?

CTPA is the preferred imaging modality when imaging is required. It provides direct visualisation of pulmonary arterial thrombus, may identify alternative diagnoses and also provides information about RV size.

V/Q imaging remains important when CTPA cannot be performed. When available, V/Q SPECT is preferred to planar V/Q scanning.

A common mistake is to use echocardiography as a diagnostic test for PE. The guideline specifically states that echocardiography should not be used to confirm or refute the diagnosis.

Once PE is diagnosed, what should happen next?

This is where the 2026 guideline introduces its most important conceptual change.

Every patient should undergo outcomes risk stratification and be assigned an AHA/ACC Acute PE Clinical Category.

The system ranges from:

A → incidental/asymptomatic disease

through

B and C → symptomatic PE of increasing severity

to

D → incipient cardiopulmonary failure

and

E → established cardiopulmonary failure.

Importantly, the category is determined by the most severe clinical, laboratory or imaging finding, and it can change as the patient's condition evolves.

What are the new AHA/ACC Acute PE Clinical Categories?

A respiratory modifier (R) can be added when clinically important respiratory compromise accompanies the category. In Category C this includes significant hypoxaemia, tachypnoea or supplemental oxygen requirement; more severe oxygen or ventilatory requirements characterise the respiratory modifiers in Categories D and E.

Clinical severity thresholds

The guideline considers a low clinical severity score to include:

PESI ≤85, sPESI = 0 or Bova ≤4

while elevated clinical severity includes:

PESI >85, sPESI ≥1 or Bova >4.

Which tests matter for PE risk stratification?

Risk assessment should not depend on clot size alone.

The new framework combines:

Clinical severity → haemodynamics → respiratory status → biomarkers → RV function

Clinical scores

For Categories A and B, the guideline recommends Hestia, PESI and/or sPESI to identify patients at low short-term risk.

For haemodynamically stable Category C or D PE, a validated PE-specific score can help identify patients at increased risk of adverse outcomes. NEWS or NEWS2 may also be used as alternatives in selected patients.

Biomarkers

In Category C PE, at least one cardiac biomarker should be measured:

Troponin or BNP

In Categories C–E, lactate should also be measured to assist short-term risk stratification. Elevated lactate may reveal clinically important hypoperfusion even before frank hypotension develops.

RV function

RV size and function should be assessed using CTPA and/or echocardiography when relevant to risk stratification.

On CTPA, the guideline recommends reporting a numerical RV/LV ratio rather than a subjective assessment. When echocardiography is performed, assessment should include parameters such as RV/LV ratio, RV diameter, TAPSE, estimated RV systolic pressure, septal motion and other indices of RV dysfunction.

What is “normotensive shock” and why does it matter?

A normal systolic blood pressure does not necessarily mean that the patient is haemodynamically safe.

The guideline highlights normotensive shock, particularly in Category D2.

Evidence of hypoperfusion may include:

Lactate >2 mmol/L, acute kidney injury, reduced urine output, cardiac index ≤2.2 L/min/m² or other evidence of impaired organ perfusion.

In Category C3, a MAP <80 mmHg may also identify patients at greater risk of deterioration.

This is an important practical shift: do not wait for persistent hypotension before recognising a deteriorating PE.

Which patients can be managed as outpatients?

Selected patients in Categories A and B can be managed without prolonged hospital admission.

The guideline recommends using a validated decision tool such as:

Hestia, PESI or sPESI.

For outpatient treatment to be appropriate, the patient must also have:

immediate access to anticoagulant therapy and rapid, reliable expert follow-up.

Clinical judgement, social circumstances, treatment access and patient preference remain important even when a numerical risk score suggests low risk.

The Hestia criteria are particularly practical because they ask whether there are reasons why home treatment would be unsafe, including haemodynamic instability, need for reperfusion therapy, bleeding risk, prolonged oxygen requirement, severe pain, severe renal or hepatic dysfunction, pregnancy or social/medical barriers to outpatient care.

What is the cornerstone of PE treatment?

Anticoagulation

Unless there is an absolute contraindication, anticoagulation should be initiated in patients with acute PE to reduce recurrent VTE and death.

For patients eligible for oral treatment, DOACs are recommended over vitamin K antagonists, unless contraindicated.

When initial parenteral anticoagulation is required in Categories C1–E1, the guideline recommends LMWH over unfractionated heparin (UFH) because of lower recurrent VTE and major bleeding.

If PE is strongly suspected at Category C2 severity or above, imaging is delayed and bleeding risk is low, therapeutic anticoagulation while awaiting imaging may be beneficial.

How are oral anticoagulants initiated?

The guideline describes the following treatment approaches:

The guideline itself advises clinicians to confirm drug doses, contraindications and interactions using current product information.

How should treatment differ according to the AHA/ACC category?

A useful way to translate the guideline into practice is:

Confirmed acute PE

The guideline recommends multidisciplinary PERT assessment for Categories C–E to improve acute care delivery and facilitate appropriate escalation.

When should systemic thrombolysis be considered?

Systemic thrombolysis is not a routine treatment for haemodynamically stable PE.

The balance between preventing cardiovascular collapse and causing major bleeding becomes progressively more favourable as PE severity increases.

The guideline notes that systemic thrombolysis in Category E can reduce mortality and recurrent PE, but bleeding risk remains central to treatment selection.

What about catheter-directed thrombolysis, mechanical thrombectomy and surgical embolectomy?

The new guideline does not recommend routinely escalating all patients with RV strain or positive biomarkers to an invasive procedure.

Instead, intervention becomes progressively more appropriate as clinical instability develops.

For mechanical thrombectomy specifically, the guideline considers it reasonable in Category E1 and potentially appropriate in D1–D2, while its benefit in C2–C3 remains uncertain.

Surgical embolectomy is reasonable in Category E1 and may be considered in selected D1–D2 patients. It is not recommended over anticoagulation alone in A–C3, and in Category E2 without mechanical circulatory support it is not recommended over other advanced therapies for preventing short-term mortality.

How should shock and RV failure be supported?

Patients with PE-related shock require more than thrombus-directed therapy.

For Categories D2–E2 with cardiogenic shock, vasopressors and/or inotropes are recommended to maintain cardiac output and systemic perfusion.

The supportive text identifies norepinephrine as the generally preferred vasopressor. If cardiac output remains inadequate, an inotrope such as dobutamine may be required in selected patients.

Volume loading should be cautious. Patients with RV failure may be preload dependent, but excessive fluid can further dilate the RV and worsen haemodynamics. Small fluid boluses may be considered when reduced preload is suspected rather than indiscriminate large-volume resuscitation.

How should hypoxaemia and respiratory failure be managed?

Moderate-to-severe hypoxaemia may be treated with heated high-flow nasal cannula, which the guideline considers preferable to standard nasal cannula oxygen for improving oxygenation.

Intubation deserves particular caution.

In PE with RV dysfunction, sedation, loss of sympathetic tone and positive-pressure ventilation can reduce RV preload and worsen RV afterload, precipitating sudden haemodynamic collapse.

Therefore:

Deep sedation and mechanical ventilation should be avoided unless clinically necessary.

When intubation cannot be avoided in a Category C–E patient, vasopressors, inotropes and—where appropriate and available—VA-ECMO should be readily available as rescue haemodynamic support.

When should VA-ECMO be considered?

For Category E2 PE with refractory cardiogenic shock, VA-ECMO is a reasonable option when appropriate resources and expertise are available.

It can stabilise systemic perfusion and oxygenation while allowing RV recovery or providing a bridge to further treatment.

Patients on VA-ECMO should generally continue parenteral systemic anticoagulation in the absence of significant bleeding. The benefit of routinely adding thrombectomy or catheter-directed thrombolysis once VA-ECMO has been established remains uncertain.

When is an IVC filter indicated?

An IVC filter is not an alternative to anticoagulation when anticoagulation can be given safely.

The clearest indication is acute PE when anticoagulation cannot be administered.

When a filter is needed:

Retrievable filters are preferred to permanent filters, and removal should be attempted as soon as PE risk has fallen sufficiently and anticoagulation becomes feasible.

Routine IVC filter placement in a patient already receiving therapeutic anticoagulation should not be performed.

Filter placement may also be considered in selected patients with recurrent PE despite optimal therapeutic anticoagulation.

How long should anticoagulation continue?

The guideline defines the initial treatment phase as 3–6 months. The decision after this period depends primarily on recurrence risk rather than simply whether the patient “feels better”.

Examples of major reversible factors include substantial surgery, prolonged hospitalisation with bed confinement, lower-limb fracture and Caesarean section.

Persistent risk factors include active cancer, autoimmune disease, inflammatory bowel disease and chronic immobility.

The decision to continue extended anticoagulation should therefore repeatedly balance recurrent VTE risk against bleeding risk, rather than being treated as a one-time decision.

What follow-up is recommended after acute PE?

Management does not end when the patient leaves hospital.

The guideline recommends:

Persistent dyspnoea should not automatically be attributed to deconditioning. Between one-third and one-half of patients with symptomatic PE report ongoing dyspnoea or activity limitation during follow-up, and chronic thromboembolic pulmonary disease should be considered when symptoms persist.

What is the practical 2026 approach to acute PE?

A clinician-focused pathway

What are the key messages for clinical practice?

The 2026 approach moves beyond simply labelling PE as “massive”, “submassive” or “low risk”. The central question after confirming PE is where the patient sits on the spectrum from Category A to E and whether that status is changing.

Low-risk patients can increasingly be treated outside hospital. In contrast, a normotensive patient with rising lactate, RV dysfunction and cardiac biomarker elevation may be considerably less stable than their blood pressure suggests.

Anticoagulation remains the foundation of treatment. DOACs are preferred for eligible patients receiving oral therapy, while LMWH is preferred when initial parenteral treatment is required in most higher-risk presentations. Advanced reperfusion therapies should be targeted to patients whose clinical severity justifies their bleeding and procedural risks—not simply because a large clot is visible on CT.

Finally, PE care should extend beyond the acute admission. Anticoagulation duration, recurrence risk, bleeding risk and persistent dyspnoea should all be actively reassessed after discharge.

Source

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: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2026;153:e977–e1051. doi:10.1161/CIR.0000000000001415.