Role of Transcatheter Treatment in Intermediate- to High-risk Pulmonary Embolism

Venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), is associated with a major global burden of disease, with an annual incidence ranging from 0.75 to 2.69 per 1,000 people.¹ The major risk factors for VTE include recent hospitalisation, recent surgery, cancer and immobilisation, in addition to specific genetic conditions.² There is strong evidence of higher VTE incidence in the elderly, probably reflecting a higher prevalence of comorbidities and risk factors in this population.¹ As a result of this, the incidence of VTE in the general population is increasing with the rising average age and, consequently, it represents an important and growing public health concern.² Regarding mortality, VTE remains the leading cause of in-hospital death and the third leading cause of cardiovascular death after MI and stroke.³

According to Barco et al., PE alone is responsible for 86,930 deaths (0.46%) of 18,726,382 total deaths reported, with a large heterogeneity across the 213 countries included in the survey, partially explainable by socioeconomic characteristics, management practices or prevalence of risk factors.⁴ PE presentation is extremely variable, ranging from asymptomatic to sudden death, with a wide spectrum of haemodynamic conditions.⁵ Therefore, correct risk stratification is crucial for adequate management and prompt identification of patients requiring a more aggressive treatment or a rapid treatment escalation.^5,6

High-risk PE (HR PE) is defined by the presence of haemodynamic instability and is associated with a 30-day mortality rate of 30%.⁵ In these patients, who account for 5% of hospitalised patients with PE, the most recent European Society of Cardiology (ESC) guidelines recommend urgent systemic thrombolytic therapy.⁶

Intermediate–high-risk PE (IHR PE) is characterised by echocardiographic or CT signs of right ventricular dysfunction (RVD) and elevated blood markers of myocardial injury in the absence of cardiogenic shock.⁶ The treatment strategy is based primarily on anticoagulation therapy, with a 30-day mortality rate ranging from 6.6% to 12.6%.⁷

It has been reported that despite being initially stable, 5% of these patients will experience haemodynamic deterioration in the first few hours after presentation.⁸ In this setting, systemic thrombolysis can be an important therapeutic resource but at the cost of a significant risk of intracranial or other life-threatening bleeding.^8,9 Moreover, it has been estimated that, owing to a high perceived bleeding risk, at least half of haemodynamically unstable patients do not receive systemic thrombolysis.¹⁰

Catheter-directed therapy (CDT), both catheter-directed thrombolysis and mechanical thrombectomy, has emerged in recent years as an alternative or complementary strategy for the treatment of PE, increasing interest in the scientific community. According to the 2019 ESC guidelines, CDT should be considered in high-risk (HR) patients with contraindications to or failure of systemic thrombolysis and in IHR patients presenting signs of haemodynamic deterioration, together with rescue thrombolytic therapy or surgical embolectomy.⁶

Nevertheless, a recent position paper from the ESC underlines that treatment of IHR PE should focus on preventing haemodynamic instability rather than intervening once shock has manifested.¹⁰ Therefore, conservative management with anticoagulants alone would not be sufficient, especially in patients presenting with signs of impending decompensation.⁸

In this context, CDT could reduce the thrombotic burden without significantly increasing the risk of bleeding.³

In addition to the high early mortality risk, PE is associated with increased long-term mortality (compared to general population) and long-term sequelae, such as post-PE syndrome and chronic thromboembolic pulmonary hypertension.^6,7,11 It is unclear if early reperfusion therapies, including interventional approaches and thrombolysis, have an impact on long-term clinical symptoms, functional limitation or persistent/new onset of pulmonary hypertension.^5,6,12,13

This review aims to discuss the critical aspects of the stratification and management of IHR PE patients and the potential role of CDT, summarising current evidence and future perspectives.

Risk Stratification

The most widely adopted acute PE classification is the one proposed in the 2019 ESC guidelines (Table 1). This differs slightly from that previously suggested by the American Heart Association (AHA), which provided for a division into three groups: massive, submassive and low-risk PE.^14.

According to ESC guidelines, patients presenting with acute PE should be stratified into three risk categories – high, intermediate and low – based on haemodynamic and clinical presentation, presence of RVD and elevated troponin levels; each risk group correlates with a different early (in-hospital or 30-day) mortality rate.⁶

The main differences between the two classifications have been discussed in a more recent AHA document.⁵ For simplicity, it has been proposed that the ESC’s high-risk and low-risk categories should correspond, respectively, to the AHA’s massive and low-risk groups, while the ESC’s intermediate-risk (IR) category should include all patients classified as submassive in earlier AHA scientific statements.⁵

HR PE is characterised by haemodynamic instability, defined as a systolic blood pressure (BP) <90 mmHg.⁶ The identification of HR PE represents the very early stage of the PE stratification process, as the primary objective is to not delay reperfusion therapy (usually with systemic thrombolysis) in all the patients with the highest mortality risk. For this reason, in this phase, laboratory tests and even angio-CT, if not readily available, are not mandatory.⁶ Bedside transthoracic echocardiography, instead, plays a key role in the differential diagnosis with other causes of shock because the absence of echocardiographic signs of right ventricular (RV) overload or dysfunction practically excludes PE as the cause of haemodynamic instability.⁶ On the other hand, the presence of clear signs of RV overload in patients with a high clinical suspicion of PE, with no other evident causes of shock, is sufficient to justify urgent fibrinolytic administration if CT is not immediately feasible.⁶

After excluding the presence of haemodynamic instability, the evaluation of additional clinical, laboratory and imaging parameters, along with patient’s comorbidities and medical history, is needed for further stratification.

Stable patients with signs of RVD and/or a positive troponin marker and Pulmonary Embolism Severity Index (PESI) class III–V or simplified PESi (sPESI) ≥1 are considered as IR. This risk class includes two further subgroups, depending on: the presence of both RVD and elevated troponin levels (IHR); or only one or neither of them (intermediate–low).⁶

Notably, patients with signs of RVD and/or a positive troponin marker must be considered as IR, even if they are PESI class I–II or sPESI=0.⁶ The low-risk category includes all patients presenting with acute PE who do not meet the criteria for HR and IR PE (Figure 1).

Risk Stratification and Treatment Decisions in Intermediate–High-risk Patients

The IHR subgroup is at the high end of the severity spectrum in haemodynamic stable patients: it includes a population characterised by a relatively high early mortality rate (6.6–12.6%) and elevated risk of haemodynamic decompensation.^7,8

The randomised controlled PEITHO trial showed that 5.0% of IHR patients, initially treated with anticoagulation alone, progressed toward haemodynamic collapse or death with a mean interval from randomisation of 1.79 ± 1.60 days.⁸ In the single-arm PEITHO 2 trial, these events occurred less frequently (1%).¹⁵

As a consequence, according to ESC guidelines, IHR patients should be closely monitored in the first 2–3 days to ensure they remain stable.⁶ During this period, risk stratification should be continuously repeated, and factors associated with a greater risk of haemodynamic decompensation should be investigated.

Major clinical, imaging and laboratory indicators of severity of pulmonary embolism in normotensive patients are summarised in Figure 2.

Clinical Parameters of Severity

The PESI is the most validated score; it combines PE-related severity factors and the patient’s baseline status to assess overall 30-day mortality and other adverse medical outcomes.^6,16,17 It evaluates 11 items and stratifies patients into five severity classes, from an inpatient risk of death and complications of <1% in class I to 10–24.5% in class V.¹⁶ All the 11 items are patient characteristics that can be easily assessed with physical examination and through taking a medical history, with no need for laboratory tests or radiological examination. A simplified version of PESI (sPESI), with only six variables, was validated in 2010, and has demonstrated similar results in terms of sensitivity and specifity.¹⁷

Both PESI and sPESI have good sensitivity but scarce specificity, with an area under the curve (AUC) of 0.78 (95% CI [0.77–0.80]) to predict mortality; this means that about 20% of the patients are not correctly classified by the score.^16,17

As confirmed by subsequent studies, PESI and sPESI estimate the risk of death from any cause within 30 days and work well in clinical practice to help identify patients with low-risk PE who may be candidates for early hospital discharge or home treatment.^5,18,19

In contrast, the AHA and ESC PE risk classifications are intended to distinguish high-risk patients who may benefit from more intensive monitoring and treatment, and are based on 30-day mortality risk, with an emphasis on PE-related deaths.⁵

Notably, PESI does not consider RV dysfunction and elevated serum troponin levels, which have been shown to be independent risk factors for early mortality.²⁰

Apart from PESI and sPESI, other relevant clinical indicators of PE severity in haemodynamic stable patients have been found.^8,10 Heart rate >100 BPM, BP 90–100 mmHg, respiratory rate >20 breaths per minute, oxygen saturation <90%, chronic heart failure and active neoplasm are associated with an increased risk of early PE-related death and haemodynamic deterioration.¹⁰

Evaluation of these patients’ characteristics and presenting symptoms is useful in the decision-making and treatment strategy for IHR patients. However, no outcome studies have shown that patients with any combination of risk factors for an early adverse outcome may clinically benefit from upfront reperfusion therapy.¹⁰

Right Ventricular Dysfunction

RV failure due to acute pressure overload is considered the primary cause of death in severe PE.⁶ Therefore, detecting signs of RVD through echocardiography or CT pulmonary angiography (CTPA) plays a lead role in confirming clinical suspicion of PE and in risk stratification.

The echocardiographic definition of RVD varies across studies and, because of the right ventricle’s morphology and irregular shape, it cannot be reliably assessed using a single parameter.^6,21 The most common parameters include RV hypokinesis, McConnell’s sign, RV end-diastolic diameter, pulmonary artery pressure (PAP), tricuspid annular plane systolic excursion (TAPSE) and the right ventricle/left ventricle (RV/LV) diameter ratio.^6,21–24 Among these, an RV/LV diameter ratio of ≥1.0 and TAPSE <16 mm are the findings most frequently associated with an unfavourable prognosis.^21,22 However, the data available in the literature are often conflicting.

In a retrospective, single-centre study of 1,416 patients with PE, an echocardiographic RV/LV diameter ratio of ≥0.9 was found to be an independent predictive factor of hospital mortality (OR 2.66; p=0.01) with a sensitivity of 72% and specificity of 58%.²³

Similarly, in a prospective study of 411 patients with IR and low-risk PE, initially treated with anticoagulation alone, TAPSE was identified as a reliable clinical predictor of 30-day PE-related mortality and/or rescue thrombolysis due to haemodynamic decompensation (AUC 0.91; 95% CI [0.856–0.935]; p=0.0001).²²

A subsequent systematic review of haemodynamically stable patients found that, for each unit increase in the RV/LV diameter ratio, the odds of all-cause mortality increased by more than 2.5-fold (OR 2.79; 95% CI [1.92–4.04]); additionally, for every 1 mm decrease in TAPSE, the odds of combined adverse events increased by 1.3-fold (OR 1.31; 95% CI [1.03–1.67]).²⁴

A more recent meta-analysis confirmed that an increased RV/LV diameter ratio and abnormal TAPSE were associated with a higher risk of short-term death in all patients with PE.²¹ However, in a sensitivity analysis of haemodynamically stable patients, neither TAPSE nor an increased RV/LV diameter ratio showed a significant correlation with death.²¹ In the same study, RV hypokinesis correlated with an increased risk of short-term adverse outcome with significant heterogeneity (RR 1.60; 95% CI 1.14–2.25), while an augmented RV diameter was not associated with an increased risk of death or adverse outcome (standardised mean difference 0.58; 95% CI [0.05–1.21]).²¹

In patients with suspected PE, the presence of thrombus in transit and patent foramen ovale should be investigated, as they are both associated with increased mortality.⁶

Thrombus in transit is a rare condition in the context of PE (prevalence of 8.1% in all PE cases), and is associated with obstructive shock (48.9%) and a high overall mortality rate (20.4%).²⁵ There are no specific indications in the guidelines addressing its management, but favourable survival odds are observed with thrombolytic therapy and surgical thrombectomy.²⁵

CTPA is nowadays a widely available diagnostic tool and is often the first imaging modality pursued in suspected PE, especially in haemodynamically stable patients. RV dysfunction is typically evaluated in CTPA by measuring the RV end-diastolic diameter or the RV/LV diameter ratio in the transverse or four-chamber view; additionally, qualitative indicators of RV dysfunction (i.e. leftward bowing of the interventricular septum or reflux in inferior vena cava) can be observed.⁶

The prognostic value of an RV/LV diameter ratio ≥0.9 was assessed in a prospective multicentre cohort study involving 457 patients. Specifically, an RV/LV diameter ratio of ≥0.9 (measured on the axial plane in non-ECG-gated images) was an independent predictor of an adverse in-hospital outcome, both in the overall population with PE (HR 3.5; 95% CI [1.6–7.7]) and in haemodynamically stable patients (HR 3.8; 95% CI [1.3–10.9]).²⁶

A following large meta-analysis confirmed that, in patients with acute PE, an RV/LV diameter ratio of ≥1.0 was associated with a 2.5-fold risk of all-cause mortality (OR 2.5; 95% CI [1.8–3.5]) and adverse outcome (OR 2.3; 95% CI [1.6–3.4]) and a fivefold risk of pulmonary embolism-related mortality (OR 5.0; 95% CI [2.7–9.2]).²⁷

In current guidelines, an RV/LV diameter ratio of ≥1.0 (instead of 0.9) is proposed as the cut-off that better correlates with a poor prognosis.⁶ However, in a recent study of 609 consecutive PE patients, the combination of an axial RV/LV diameter ratio of ≥1.5 and reflux of contrast medium in the inferior vena cava has been suggested as an optimised definition of RV dysfunction, as it was associated with the best prognostic performance for predicting adverse outcomes in both unselected (OR 3.7; 95% CI [2.0–6.6]) and normotensive patients (OR 2.8; 95% CI [1.1–6.7]).²⁸

In conclusion, the evaluation of RVD can be performed both by echocardiography and by CT and is of paramount importance in the risk stratification of PE; however, a global and multiparametric evaluation is recommended, which must be integrated with laboratory tests, clinical data and patient characteristics (Supplementary Table 1).

Serum Biomarkers

Troponins T and I are serum biomarkers indicative of myocardial injury, traditionally used for diagnosing MI and rapidly available in the urgent setting. Elevated troponin levels are found in 30–60% of patients with PE but can also occur in conditions such as acute pericarditis, myocarditis, severe heart failure, sepsis and acute renal failure.⁶

In patients with PE, elevated levels of troponin (both I and T) are linked to higher short-term mortality, increased risk of PE-related death and adverse outcomes.^29,30 This association holds true even in haemodynamically stable patients and remains consistent both for conventional troponin (cTn) and for high-sensitivity troponin (hsTn) assays.^29,31

In one study, hsTn assay using a cut-off value of 14 pg/ml was associated with a prognostic sensitivity and a negative predictive value superior to cTn assay and was the only assay that demonstrated a correlation with long-term survival.³¹ Conversely, a recent cohort study of 834 patients with haemodynamically stable PE observed that using hsTn, compared with cTn more frequently detected elevated values but failed to predict the 30-day risk of a complicated course, suggesting that use of hsTn may result in overestimation of the risk in patients with stable PE.³²

According to the ESC guidelines, cardiac troponin I or T elevation are defined as concentrations above the normal limits, and thresholds depend on the assay used.⁶ However, age-adjusted hsTnT cut-off values (≥14 pg/ml for patients aged <75 years and ≥45 pg/ml for those ≥75 years) may further improve the negative predictive value of this biomarker.^6,33

In haemodynamically stable PE, troponin markers, due to their high negative-predictive value, can be considered in clinical practice, together with clinical severity parameters and imaging findings of RVD, as an instrument to identify patients with a low risk of poor prognosis.^34,35

B-type natriuretic peptide (BNP) is synthesised as an inactive prohormone (pro-BNP) that is split into the active hormone BNP and the inactive N-terminal fragment (NT–pro-BNP). It is released by ventricular cardiomyocytes in response to stretch and, therefore, represents a serum biomarker of right ventricle overload.^36,37

The prognostic role of BNP and NT-proBNP has been established in various meta-analyses, as higher levels of these markers have been associated with increased rates of early mortality, early adverse events, and PE-related mortality.^30,36,37

Elevated BNP and NT-proBNP have limited specificity and a positive predictive value in predicting early mortality for normotensive patients with PE. However, low BNP or NT-proBNP levels can reliably exclude the likelihood of an adverse early clinical outcome, demonstrating high sensitivity and a strong negative predictive value.⁶ The ESC suggests a cut-off of ≥600 ng/l for NT-proBNP.⁶

In a recent study, the predictive value of BNP was estimated in proportion to the upper normal limit (UNL) and a cut-off of 3.5 times the UNL was associated with increased rates of both all-cause mortality and PE-related mortality.³⁸ This approach aims to minimise measurement discrepancies between different assays across laboratories.

Although BNP is excluded from the main criteria in current guidelines, it plays a fundamental role in decision-making for reperfusion therapy, particularly for IHR PE. BNP is included in the Composite Pulmonary Embolism Shock (CPES) score, developed to identify patients with normotensive shock, who are known to be at an increased risk of early mortality among those who are haemodynamically stable.^39,40

Lactate is a marker of an imbalance between tissue oxygen supply and demand. Elevated arterial plasma levels of ≥2 mmol/l predict PE-related prognosis both in unselected and in initially normotensive PE patients.^6,41–44 Plasma lactate levels >2 mmol/l are associated with increased in-hospital mortality, PE-related mortality, all-cause death and the composite endpoint of all-cause death and clinical deterioration, independently of hypotension or RVD at presentation.^41,42

Selectively in normotensive patients, a large multicentre prospective cohort study showed that raised lactate is an independent predictor of PE-related mortality or clinical deterioration; in addition, the combination of RV dysfunction, elevated troponin and increased lactate predicted a 6.6-fold increase in the risk of short-term PE-related adverse events.⁴³

Normotensive Shock

The ESC classification does not reliably predict clinical deterioration in IHR patients, which may ultimately lead to haemodynamic instability and death, usually due to progressive RV dysfunction and cardiogenic shock.^45,46

The FLASH registry is an observational registry studying patients with acute PE treated with catheter-based mechanical thrombectomy using the FlowTriever (Inari) system.³⁹ An interim analysis of the first 384 patients, where an invasive assessment of cardiac index was conducted before and after thrombectomy, revealed that one-third of IR haemodynamically stable patients were actually in a state of normotensive shock.

Normotensive shock is defined by a reduced cardiac output (≤2.2 l/min/m²) alongside a systolic BP of ≥90 mmHg, maintained by compensatory mechanisms.³⁹ Notably, 17.4% of patients with normotensive shock had a sPESI score of 0.³⁹

The FLASH authors developed a 6-point CPES score comprising elevated cardiac troponins, elevated natriuretic peptides, RVD, saddle PE, concomitant DVT and tachycardia.³⁹ A score of 6 was a significant predictor of normotensive shock (OR 5.84; 95% CI [2.00–17.04]), with the prevalence of normotensive shock increasing in tandem with higher scores: a score of 0 corresponded to a 0% prevalence of normotensive shock, while a score of 6 showed a prevalence of 58.3% (OR 5.8; 95% CI 2.0–17.0]).³⁹

In a subsequent retrospective study, CPES score predicted death (aHR 1.76; 95% CI [1.04–2.96]; p=0.033), resuscitated cardiac arrest (aHR 1.99; 95% CI [1.17–3.38]; p=0.011), and haemodynamic decompensation (aHR 1.96; 95% CI [1.34–2.89]; p=0.001).⁴⁰

Another study confirmed the good prediction performance of the CPES score, highlighting that with a threshold of 4 points for a positive test, instead of 3 points, the shock score classified a much lower proportion of patients as positive, but showed a significantly greater ability to predict a complicated course (death from any cause, haemodynamic collapse or recurrent PE).⁴⁷ Results were similar in the analyses restricted to the IR subgroup.⁴⁷

These findings hold significant potential for managing IHR PE, as patients with normotensive shock may be the most likely to benefit from more aggressive interventions, such as thrombolysis and catheter-directed reperfusion. However, large randomised controlled trials demonstrating short-term clinical outcome benefits of CDT over anticoagulation alone in IHR PE are still lacking, and the cardiogenic shock prognosis score requires validation in larger studies.

RISA-PE is a recently proposed classification system that adapts the five Society for Cardiovascular Angiography and Interventions shock stages to address RV failure caused by acute PE:

• A: right ventricular dysfunction and troponin elevation;
• B: A+serum lactate >2 mmol/l or shock index ≥1;
• C: persistent hypotension;
• D: obstructive shock;
• E: cardiac arrest.

It was tested retrospectively in a cohort of 334 consecutive IHR and HR PE patients, assessed by local pulmonary embolism response teams (PERTs) and treated with CDT, and it has been proposed to enhance acute PE risk stratification and patient selection for CDT. In-hospital all-cause mortality increased progressively with higher RISA-PE stages (1.2%, 6.4%, 19.0%, 25.6% and 57.7% for stages A, B, C, D and E, respectively; p-value for linear trend <0.001), independently from other variables associated with increased mortality, such as respiratory insufficiency and the presence of bilateral central PE. Moreover, the RISA-PE classification demonstrated higher predictive accuracy for in-hospital all-cause mortality compared to the risk stratification suggested in ESC guidelines. However, further evaluation in larger studies is necessary before these stages can be integrated into the clinical decision-making process.⁴⁸

Indications for Catheter-directed Therapies

Presently, CDT is not the first-line treatment in either HR or IR patients with PE.

As mentioned above, in the 2019 ESC guidelines, CDT is recommended for high-risk patients with a contraindication to or failure of systemic thrombolysis, as well as for IHR patients showing signs of haemodynamic deterioration despite adequate anticoagulation therapy (i.e. failure of anticoagulation).⁶ In a position paper, the ESC clarified the definition of treatment failure of thrombolytic and anticoagulation therapy and proposed a flow chart for treatment-decision strategies (Figure 3).¹⁰

The core of this diagram is the presence of a PERT, a panel of specialists from different disciplines (cardiology, pulmonology, haematology, vascular medicine, anaesthesiology/intensive care, cardiothoracic surgery and radiology) that coordinates and expedites the process of decision-making, identifying the risk factors and favouring a patient-tailored approach.⁶ Given the complexity of PE, the creation of PERTs is encouraged to standardise the treatment and improve clinical outcomes.⁴⁹

The importance of PERTs has also been emphasised in two position papers recently published by the Italian Society of Interventional Cardiology and the Interventional Cardiology Working Group of the Italian Society of Cardiology.^50,51 These documents highlight the importance of a multidisciplinary approach, recommending the establishment of PERTs to assess patients for advanced reperfusion therapies, especially when standard treatments are contraindicated or ineffective. Integrated in-hospital pathways and regional PERT networks are recommended to ensure equitable access to advanced therapies and a standardised PE management. Additionally, the authors call for larger randomised clinical trials to evaluate the efficacy of transcatheter therapies compared to current standards, aiming to improve patient outcomes through evidence-based guidelines and promote further research and education in this field.^50,51 The growing importance of CDT as part of the therapeutic armamentarium is highlighted as bridging the gap between evolving standards of care and current clinical practice.

Regarding the indication for CDT, as suggested by ESC, in initially stable patients, the development of overt cardiorespiratory instability after starting anticoagulant treatment is a clear treatment failure and warrants an immediate escalation of emergency treatment with systemic thrombolysis, if this is not contraindicated.^6,10

Sometimes, haemodynamic deterioration is more subtle, indicated by a progressive increase in heart or respiratory rate, a progressive decrease in systemic BP or oxygen saturation, or by worsening signs of organ hypoperfusion (decrease in urinary output, increase in lactate levels).¹⁰ In these cases, the PERT should discuss rescue reperfusion therapy.¹⁰

However, treatment failure can also be defined as a lack of improvement, indicated by no progress in clinical, imaging or laboratory parameters after 24–48 hours of therapy. If, following this period, a patient still meets the criteria for the IHR class (TAPSE <16 mm and RV/LV diameter ratio ≥1), reperfusion therapies (systemic thrombolysis or CTD) should be considered, even in the absence of clear signs of haemodynamic deterioration.¹⁰

The decision to consider a patient for reperfusion therapy should depend not only on the patient’s baseline conditions and overall bleeding risk but also on local expertise, availability of CTD and logistical factors.¹⁰ Consequently, a multidisciplinary approach is essential, and a PERT should always be involved in patients’ evaluation.

Additionally, an understanding of clot timing (whether it is fresh and fibrin-rich or chronic and fibrotic) helps in selecting the right patient candidates for CTD and in choosing the most appropriate device for the intervention, significantly improving procedural success and patient outcomes. In this context, recognising acute-on-chronic scenarios or acute PE on underlying chronic pulmonary hypertension is crucial.

Catheter-directed Therapy Technologies

CDT techniques can be broadly categorised, based on thrombus-removal mechanisms, into catheter-directed thrombolysis, catheter-directed mechanical thrombectomy and pharmaco-mechanical thrombectomy (Supplementary Table 2). The aim of these therapies is to remove the thrombus from the pulmonary arterial bed, reduce RV overload and, consequently, recover RV function.

Treatment should be guided by the improvement in patient’s clinical parameters (BP, oxygen saturation, heart and respiratory rate, etc.) and by the reduction of pulmonary artery invasive pressure and RV dysfunction. Thus, complete thrombus removal is typically unnecessary to achieve a favourable outcome.¹⁰

Local thrombolysis devices enable the direct infusion of thrombolytics into the thrombus through multi-hole standard catheters (such as pigtail), dedicated catheters (e.g. Uni-Fuse [AngioDynamics] or Cragg-McNamara [Medtronic]), or ultrasound-emitting catheters (such as the EkoSonic or EKOS system [Boston Scientific]).

Specifically, ultrasound-accelerated catheter-directed thrombolysis (USAT) combines thrombolytics infusion with ultrasound emission, which is intended to accelerate clot dissolution by loosening and thinning fibrin strands, thereby exposing more drug receptor sites.

USAT with EKOS is the CDT with the largest scientific evidence. Safety and efficacy of the device were documented in the ULTIMA, SEATTLE II and OPTALYSE PE trials, as well as in a prospective registry (KNOCOUT PE) with a wider population.^52–54 Overall, USAT was demonstrated to be effective in reducing the RV/LV diameter ratio with a low risk of major bleeding (1.6%).^49–54

The SUNSET sPE Trial compared standard catheter-directed thrombolysis (SCDT) with USAT in IR patients, demonstrating a similar reduction in thrombotic burden, apparently with a better RV/LV diameter ratio reduction in the SCDT group and a major bleeding rate of 5% in the USAT group.⁵⁵ In the PEITHO1 trial, the major bleeding rate was 11.5%.⁸

A subsequent meta-analysis revealed no significant differences between USAT and SCDT in terms of intensive care unit (ICU) or hospital length of stay, though SCDT showed a slightly greater reduction in RV/LV diameter ratio (−0.16; p=0.003).⁵⁶ Additionally, an observational analysis of a registry of 39,430 patients found no difference in in-hospital mortality between the two groups.⁵⁶ Both studies also indicated similar rates of bleeding events across treatments.^56,57

The major advantage of SCDT and USAT is the reduced dose of thrombolytics compared to systemic thrombolysis, theoretically lowering the bleeding risk. The total dose and administration protocols may vary between centres, with treatment duration extending for up to 24 hours.¹⁰ This type of approach should be considered in patients without anticoagulation or thrombolysis contraindication if they are haemodynamically stable.

Mechanical thrombectomy primarily involves thrombus aspiration devices. Thrombus aspiration is achieved by applying suction, either manually or with a dedicated system, through large-bore catheters (20 Fr or larger) or medium-bore catheters (8–16 Fr).¹⁰

The FlowTriever system includes a set of three telescopic aspiration catheters (16 Fr, 20 Fr and 24 Fr), a 60 ml aspiration syringe and a catheter with a laser-cut open cell element for clot mechanical disruption (replacing the previous four catheters with nitinol disks). The recent introduction of the FlowSaver filter allows aspirated blood to be reinfused into the patient.

The prospective, multicentre, single-arm FLARE study assessed the safety and the efficacy of the device in 104 IHR patients. At 48 hours follow-up, a mean reduction rate of 0.38 in the RV/LV diameter ratio was observed (p<0.0001) and the mean PAP was reduced by 2.0 mmHg (p=0.001).⁵⁸ Six major adverse events were reported in four patients, with no device-related deaths. The major bleeding rate was 0.9%.⁵⁸

FLASH is an ongoing, prospective registry designed to investigate the second-generation FlowTriever.⁵⁸ The results from the first 800 patients (7.9% HR PE and 76.7% IHR PE) found a significant reduction in the RV/LV diameter ratio (p<0.0001) and in PAP (−23.0%; p<0.0001) as well as demonstrating a 0.3 l/min/m² mean increase in cardiac index (18.9%; p<0.0001) in patients with depressed baseline values. The major adverse event rate was 1.8% at 48 hours, with no device-related deaths.⁵⁹

The Indigo mechanical thrombectomy system (Penumbra) includes 7 Fr (replacing the first generation 8 Fr catheter), 12 Fr and 16 Fr aspiration catheters, a pump providing suction and a separator wire (for 7 Fr and 12 Fr only). The new-generation catheters are endowed with a computer-aided thrombectomy technology that aims to reduce blood loss.

The first generation of 8 Fr Indigo catheters was investigated in the EXTRACT-PE trial, which included 119 patients with IR PE. This observed a RV/LV diameter ratio reduction at 48 hours of 0.43 (p<0.0001), with a major adverse events rate of 1.7% and one device-related death.⁶⁰

The next-generation Indigo devices are being evaluated in the observational, single-arm STRIKE-PE study. Initial results from the first 150 patients (94.7% IR PE and 5.3% HR PE), who were treated with the 12 Fr catheter, have been published.⁶¹ The mean RV/LV diameter ratio showed a 25.7% reduction (p<0.001) and both systolic and median PAP decreased significantly (8.9 mmHg and 5.8 mmHg, respectively; p<0.001). Four patients (2.7%) experienced a combination of major adverse events within 48 hours, including major bleeding, clinical deterioration, pulmonary vascular injury and cardiac injury linked to the device. No device-related deaths or cardiac injuries were reported within the first 48 hours.

At 90-day follow-up, patients demonstrated statistically significant improvements in Borg dyspnoea scores and quality of life measures, and New York Heart Association class distribution returned to pre-index PE levels.

A recent meta-analysis compared mechanical thrombectomy using the Indigo and FlowTriever devices with ultrasound-assisted thrombolysis using the EKOS device.⁶²

Technical success was similar in the two groups (99.6 versus 99.4%).⁵⁹ Thrombectomy was associated with a longer mean procedure time, lower mean blood loss, shorter mean ICU stay and shorter mean overall hospital stay.⁶² EKOS performed better in reducing thrombotic burden (measured with the Miller Index) and PAP.⁶² The authors concluded that both procedures were effective in PE treatment, with comparable results.⁶²

Conversely, data from REAL-PE, a large retrospective observational study comparing FlowTriever and EKOS, indicated a higher bleeding risk associated with the FlowTriever group, with no observed differences in median length of stay, all-cause 30-day readmission or in-hospital mortality.⁶³ Notably, the study did not include ESC stratification of patients or details regarding previous or concurrent thrombolytic therapy.

The recently published PEERLESS trial is a pivotal randomised controlled study comparing large-bore mechanical thrombectomy (LBMT) using the FlowTriever system with catheter-directed thrombolysis in patients presenting with intermediate-risk PE.

Conducted across 57 sites in the US, Germany and Switzerland, the trial enrolled 550 patients with acute IR PE. Participants were randomised in a 1:1 ratio to receive either LBMT or catheter-directed thrombolysis. The primary endpoint was a hierarchical composite assessed at hospital discharge or within 7 days after the procedure, encompassing all-cause mortality, intracranial haemorrhage, major bleeding, clinical deterioration and/or escalation to bailout, and intensive care unit admission and length of stay.

Results demonstrated a significant advantage for LBMT over catheter-directed thrombolysis, with a win ratio of 5.01 (95% CI [3.68–6.97]; p<0.001). This outcome was primarily driven by lower rates of clinical deterioration and/or bailout (1.8 versus 5.4%; p=0.04) and less post-procedural ICU use, including fewer admissions (41.6% versus 98.6%) and shorter stays exceeding 24 hours (19.3% versus 64.5%). No significant differences were observed between the two groups regarding mortality, intracranial haemorrhage or major bleeding.

Secondary endpoints favoured LBMT, with patients exhibiting lower respiratory rates at 24 hours (18.3 ± 3.3 versus 20.1 ± 5.1; p<0.001) and fewer instances of moderate to severe dyspnoea, as measured by the modified Medical Research Council scale (13.5 versus 26.4%; p<0.001).

Additionally, LBMT was associated with shorter total hospital stays (4.5 ± 2.8 versus 5.3 ± 3.9 overnight stays; p=0.002) and fewer all-cause readmissions within 30 days (3.2% versus 7.9%; p=0.03). The 30-day mortality rates were comparable between the two groups (0.4% versus 0.8%; p=0.62).⁶⁴

More randomised trials are needed to establish if one device is clearly superior to another. In current clinical practice, the choice of the most appropriate CDT should be evaluated on a case-by-case basis, considering the patient’s clinical presentation and comorbidities.

Mechanical thrombectomy is preferable when thrombolytic or anticoagulant therapy is contraindicated, when systemic thrombolysis has failed or when there are signs of imminent haemodynamic deterioration, owing to its capacity for immediate thrombus removal without a need for thrombolytics. On the other hand, USAT is a relatively easy and safe procedure that has proven effective, although it requires a prolonged infusion time (Figure 4).

Expected Benefits of Catheter-directed Therapies

During a procedure, CDTs allow direct access to the thrombus, enabling aspiration, targeted delivery of thrombolytics or pharmaco-mechanical intervention.

From a technical perspective, when performing local thrombolysis, it is essential to stabilise the catheter as distally as possible to ensure the drug is delivered effectively along its entire course within the thrombus. Conversely, thrombus aspiration procedures involve greater technical complexities.

Thrombus aspiration offers the advantage of rapidly restoring pulmonary perfusion, which in turn improves haemodynamics in realtime. Although a clear definition of procedural success has not been established, prior studies suggest that a reduction of >7 mmHg in mean PAP may correlate with positive outcomes, along with a decrease in heart rate and an increase in BP and oxygen saturation. This haemodynamic improvement should lead to a marked reduction in RV strain and lower short-term mortality (<1% in observational registries).⁶⁰ Patients also often experience symptomatic relief, including a rapid resolution of dyspnoea and chest pain.⁵⁸

The long-term benefits of CDT are becoming increasingly evident as observational studies show favourable outcomes in terms of long-term mortality, functional status and quality of life.^65–67 The targeted nature of the therapy minimises the risk of residual clot burden, which can lead to post-PE syndrome characterised by chronic dyspnoea, exercise limitation and reduced quality of life.⁶⁸

In addition, by effectively treating the thromboembolic event early, CDT may reduce post-embolic pulmonary hypertension and the clinical impact of any recurrent thromboembolism, which is the main cause of death in the first month.¹¹

However, further studies are needed in this field to evaluate the clinical impact in everyday practice.

Conclusion

Acute PE poses significant clinical challenges, particularly for IHR patients, underscoring the need for accurate risk stratification and timely intervention. The ESC guidelines emphasise a three-tiered risk model – high, intermediate and low – aligned with mortality risk and management requirements.

Rapid identification of IHR patients, due to their elevated risk of haemodynamic deterioration, is crucial to guide them towards the correct therapy promptly. The PESI score and other clinical tools help to stratify these cases, while the addition of cardiac markers such as troponins and RVD parameters has further refined risk prediction.

Emerging tools, including CDT such as USAT and mechanical thrombectomy, provide targeted options for patients with contraindications to anticoagulation or for those unresponsive to it. While promising, these therapies require more robust validation.

The PERT, with its multidisciplinary approach, enables the comprehensive integration of diagnostic and therapeutic tools, including CDT, in managing patients with PE. Therefore, this team plays an increasingly pivotal role in addressing complex cases, facilitating tailored and adaptive treatment strategies that enhance outcomes in life-threatening PE scenarios.

Click here to view Supplementary Material.