Percutaneous treatment of high-risk pulmonary embolism: bridging the evidence gap

This editorial refers to ‘Outcomes with catheter-directed thrombolysis vs. catheter-directed embolectomy among patients with high-risk pulmonary embolism: a nationwide analysis’, by R. Sedhom et al. https://doi.org/10.1093/ehjacc/zuad004. We are in a period of rapid evolution in the management of pulmonary embolism (PE). However, morbidity and mortality related to PE remain high, and significant gaps persist in our understanding of the comparative effectiveness and safety of the diverse approaches available, spanning medical, percutaneous, and surgical strategies.1,2 Current PE guidelines recommend the use of systemic thrombolysis in patients with high-risk PE.3 However, more than half of patients with high-risk PE fail to receive systemic thrombolysis due to increased risk of bleeding or other factors.4 This has led to a growing interest in alternative therapies, including catheter-directed interventions which are recommended as an alternative to surgical embolectomy (Class IIa; level of evidence C) in patients for whom systemic thrombolysis has failed or is contraindicated.3 In addition, catheter-directed interventions are recommended as an alternative to systemic thrombolysis in other patients with PE who have worsening haemodynamics despite anticoagulation (Class IIa; level of evidence C).3 One approach to enhancing the appropriate utilization of these interventions has been the establishment of multidisciplinary pulmonary embolism response teams (PERT).5 Major society recommendations reflect the current state of evidence, indicating a lack of high-quality comparative effectiveness studies. Despite the limited evidence base, there has been a rapid adoption of percutaneous interventions for high-risk PE,2 with the most commonly used catheter-directed interventions being catheter-directed thrombolysis and catheter-directed embolectomy.6 Catheter-directed thrombolysis involves either a traditional dedicated catheter that infuses thrombolytics locally in the affected pulmonary artery or a more novel catheter that has a locally embedded high-frequency ultrasound emitter intended to mechanically disrupt the clot (EkoSonic endovascular system—Boston Scientific, Marlborough, MA).1 Catheter-directed thrombolysis allows for a reduction in the total dose of thrombolytics that would be otherwise given systemically. Catheter-directed embolectomy, conversely, relies on aspiration through large bore catheters to remove thrombus from central arteries. The two major FDA-approved embolectomy catheters are the FlowTriever (Inari Medical, Irvine, CA) and Indigo (Penumbra Inc., Almeda CA) systems.1,2,6 To date, there are no adequately powered randomized controlled trials comparing any of the diverse potential therapies for PE. This leaves physicians and multidisciplinary teams with little guidance when choosing from a wide array of options that include anticoagulation alone, systemic thrombolysis, surgical embolectomy, and/or catheter-directed interventions. There is an important need for comparative data to guide physicians in managing these high-risk patients. In this issue of the European Heart Journal Acute Cardiovascular Care, Sedhom et al. evaluate outcomes after catheter-directed thrombolysis vs. catheter-directed embolectomy in high-risk pulmonary embolism using a retrospective cohort from an administrative data set of sampled patients hospitalized between 2016 and 2019. The goal of the study was to describe clinical outcomes following treatment with either of these percutaneous approaches. High-risk pulmonary embolism was defined as PE with cardiogenic shock, mechanical ventilation, mechanical circulatory support (MCS), or vasoactive agents. Patients were identified using International Classification of Diseases, Tenth Revision, Clinical Modification (ICD 10-CM) codes. The primary outcome of the study was all-cause in-hospital mortality, while secondary outcomes included intracranial haemorrhage, non-intracranial haemorrhage, length of stay, cost, and 30-day readmission. A propensity score matching approach was used for covariate adjustment, and they used falsification endpoints to assess for residual confounding. The authors made several observations among this cohort of 3216 real plus imputed patients hospitalized for high-risk PE in which percutaneous treatment was employed. First, the most common percutaneous approach was catheter-directed thrombolysis (N = 1864 [58%]). Second, the overall use of catheter-directed interventions increased during the study period with an increase in the number of hospitals using either approach over time. Third, patients who received catheter-directed interventions were more likely to have high-risk features such as saddle embolus, cor pulmonale, or cardiogenic shock compared to those managed alternatively. Fourth, patients who underwent percutaneous intervention for high-risk PE had a high all-cause in-hospital mortality. Finally, in a propensity score–matched analysis, there were no significant differences in the primary and secondary outcomes between catheter-directed thrombolysis and catheter-directed embolectomy. These findings describe current experience with management of high-risk PE and are largely consistent with prior studies. Similar to findings in this report, Stein et al. found that despite advances in the management of patients presenting with high-risk PE in the USA, the mortality rate remains high (∼50%).4 This experience resembles that seen in Germany as well, with most deaths occurring early in the admission.7 Despite this, only a minority of these patients receive systemic thrombolytics or other non–catheter-based advanced therapies.4 Sedhom et al. add to this evidence by highlighting the growing utilization of catheter-based interventions. However, it is unclear how that increased utilization has impacted the outcomes of patients with high-risk PE. Current studies evaluating catheter-based interventions have been limited to haemodynamic outcomes. Therefore, the report from Sandhom et al. provides additional granular data on the clinical outcomes of patients with high-risk PE undergoing catheter-based interventions. There are several limitations to evaluating the comparative effectiveness of these approaches in this observational study. First, there is no way to account fully for selection bias, and patients treated with catheter-directed embolectomy were more likely to have a saddle PE and receive systemic thrombolysis, MCS, and/or inferior vena cava filter. While the authors attempted to adjust for these differences with propensity matching, this administrative database does not completely capture illness severity (e.g. lactic acid level, right heart catheterization indices, frailty). Second, the use of falsification endpoints to assess for residual confounding may be useful in certain clinical scenarios, but given the high mortality associated with high-risk PE, any falsification endpoint in this clinical situation is likely to be affected by the competing risk of death, making the results difficult to interpret. Third, the use of certain ICD-10-CM codes has been validated using clinical trials, but not for PE or in this cohort, and therefore, it is unclear what the performance or accuracy of these codes are in identifying this cohort, procedures, and non-mortality outcomes.8 Finally, the high mortality rate in this group raises concerns for immortal time bias, whereby patients that live long enough are more likely to get a certain treatment. This was not accounted for in this analysis and makes any interpretation of the comparative effectiveness of these approaches limited. The findings in this report offer us a view of the current practices and outcomes following the use of these novel treatments that will inform future studies. However, the limitations of this study highlight an important issue in the field of PE interventions, which is the absence of adequately powered randomized controlled trials. Table 1 summarizes the current landscape of evidence for advanced therapies in PE. Two studies currently underway evaluating are the HI-PEITHO trial (NCT04790370) and the PEERLESS trial (NCT05111613). HI-PEITHO is randomizing patients with intermediate-risk PE to either ultrasound-facilitated, catheter-directed thrombolysis or anticoagulation alone, while the PEERLESS trial is randomizing patients with intermediate or high-risk PE to either catheter-directed thrombolysis (with any commercially available device) or catheter-directed embolectomy with the FlowTriever system. Both trials are expected to complete enrolment in 2024 and will no doubt fill an important evidence gap that will help inform patients, clinicians, and regulators. Current landscape of evidence for catheter-based therapies in pulmonary embolism ULTIMA RCT (NCT01166997) CANARY RCT (NCT05172115) HI-PEITHO RCT (NCT04790370) FLARE (NCT02692586)a FLASH (NCT03761173)a EXTRACT-PE (NCT03218566).a PEERLESS RCT (NCT05111613) CANARY RCT (NCT05172115) SEATTLE II (NCT01513759)a FLASH (NCT03761173)a FLAME (NCT04795167)a ULTIMA RCT (NCT01166997) CANARY RCT (NCT05172115) HI-PEITHO RCT (NCT04790370) FLARE (NCT02692586)a FLASH (NCT03761173)a EXTRACT-PE (NCT03218566).a PEERLESS RCT (NCT05111613) CANARY RCT (NCT05172115) SEATTLE II (NCT01513759)a FLASH (NCT03761173)a FLAME (NCT04795167)a Colours: white indicates no evidence, blue indicates available evidence (single-arm studies or haemodynamic outcomes), and bold indicates studies that are currently underway. Completed studies listed have only shown haemodynamic improvements and have not evaluated patient-centred outcomes (e.g. mortality, readmission). RCT, randomized control trial. Single-arm studies. Current landscape of evidence for catheter-based therapies in pulmonary embolism ULTIMA RCT (NCT01166997) CANARY RCT (NCT05172115) HI-PEITHO RCT (NCT04790370) FLARE (NCT02692586)a FLASH (NCT03761173)a EXTRACT-PE (NCT03218566).a PEERLESS RCT (NCT05111613) CANARY RCT (NCT05172115) SEATTLE II (NCT01513759)a FLASH (NCT03761173)a FLAME (NCT04795167)a ULTIMA RCT (NCT01166997) CANARY RCT (NCT05172115) HI-PEITHO RCT (NCT04790370) FLARE (NCT02692586)a FLASH (NCT03761173)a EXTRACT-PE (NCT03218566).a PEERLESS RCT (NCT05111613) CANARY RCT (NCT05172115) SEATTLE II (NCT01513759)a FLASH (NCT03761173)a FLAME (NCT04795167)a Colours: white indicates no evidence, blue indicates available evidence (single-arm studies or haemodynamic outcomes), and bold indicates studies that are currently underway. Completed studies listed have only shown haemodynamic improvements and have not evaluated patient-centred outcomes (e.g. mortality, readmission). RCT, randomized control trial. Single-arm studies. In conclusion, PE continues to be an important cause of morbidity and mortality despite advances in management and increased utilization of catheter-based interventions. On-going randomized trials will inform us about the comparative effectiveness and safety of these approaches. Continued innovation and evidence generation are needed to improve short- and long-term outcomes for these high-risk patients. B.A.B. received grant support through Brigham and Women’s Hospital, Pfizer, Ionis, AstraZeneca, and Abbott Vascular and consulting/personal fees from Abiomed, SpectraWAVE, Endovascular Engineering, CSI, Philips, Abbott Vascular, Servier, Daiichi Sankyo, Janssen, and Quark Pharmaceuticals.