Here we describe, for the first time, the use of the FlowTriever system with large-bore percutaneous thrombus aspiration for intermediate-high and high-risk PE in the Netherlands. Using the FlowTriever system, the technical success rate was 100%, 30-day all-cause mortality was 29% and major bleeding was recorded in 24% of patients, of which 1 bleeding complication seemed to be directly related to the FlowTriever procedure. Furthermore, a significant reduction was seen in mPAP and RVEDD.

Treatment of massive PE using thrombolysis remains a challenge in daily practice due to the high mortality and bleeding rates, particularly in high-risk patient subgroups [6]. In current guidelines, ST is recommended in high-risk PE in order to reduce RV failure and mortality [4]. Unfortunately, evidence supporting these recommendations is slim [13, 14]. Importantly, ST is known to be associated with a concomitant high risk of bleeding and may therefore not be suitable in a select group of patients. A meta-analysis reported on 12 studies comparing ST with anticoagulation [7]. Mortality in high-risk and intermediate-risk PE patients receiving ST was 6.9 and 1.0%, respectively. Major bleeding in all patients treated with ST was 9.9% (1.7% fatal or intracranial haemorrhage). The largest trial in this meta-analysis (PEITHO) included 1005 intermediate-risk patients, randomised to ST plus heparin versus placebo plus heparin [14]. All-cause mortality in the ST group was 1.2% and major bleeding 11.5%. The study protocol, however, dictated the exclusion of patients with a significant bleeding risk or any other condition which the investigators felt would cause harm.

Studies including intermediate-risk patients treated with the FlowTriever system initially showed low mortality and bleeding rates. In 2019 the FLARE study enrolled 106 intermediate-risk patients treated with the FlowTriever system of which only two were concomitantly treated with ST [15]. One death was reported and one major bleeding. Similar results were reported by Wible et al. in their study on the use of the FlowTriever system in 46 intermediate-high and intermediate-low-risk patients [11]. In-hospital mortality was 0% and major bleeding was seen in 4.6%. In a small case series with 8 patients (1 high risk, 7 intermediate risk) a mortality rate of 13% was seen, without any bleeding complications [10]. More recently, Luedemann et al. reported on the use of the FlowTriever system in a cohort of intermediate-high and high-risk patients, with a 19% all-cause 30-day mortality and 19% major bleeding [16]. A study by Kucher et al. enrolled 15 high-risk PE patients with a contraindication for ST and reported one death [17]. Eight of these patients received VA-ECMO. Recently, the FLAME trial enrolled 53 high-risk patients [18]. More than one-fifth of these patients presented after circulatory arrest. Major bleeding was seen in 11% and mortality was as low as 1.9%, although notably comatose patients and patients who had been resuscitated for more than 30 min were excluded. Lastly, the ongoing FLASH registry investigators reported on their findings in 800 intermediate-risk and high-risk (8%) patients [19]. In 32% ST was contraindicated and ST had failed in 10%. All-cause 30-day mortality was low (0.8%) and major bleeding at 48 h was 1.4%.

Mortality rates were lower in these studies compared with our results. However, this can be attributed to the significant proportion of patients in our study presenting with high-risk PE (67%) and circulatory arrest (48%). Circulatory arrest due to PE has a detrimental outcome with a high mortality rate of up to 90% [20] and in our study 4 out of 6 deceased patients presented with arrest.

Major bleeding rates in our study were relatively high. Again, this is most likely a reflection of our indiscriminate patient selection where 8 patients (38%) with ST failure and 13 patients (62%) with a major bleeding risk factor were included. Four out of five patients with a major bleeding had received ST prior to the FlowTriever procedure. Importantly, only one major bleed seemed to be related to the FlowTriever procedure (cardiac tamponade).

In our centre, the FlowTriever procedure is performed by a dedicated interventional cardiology team that is proficient in large-bore access and has experience in RV (outflow tract)/pulmonary artery interventions. Furthermore, the on-site availability of (rescue) VA-ECMO offers additional salvage options in patients presenting with circulatory arrest or obstructive shock; the use of VA-ECMO can reduce mortality rates to 7–35% based on previous literature [17, 20,21,22]. VA-ECMO is initiated in the Cath Lab, using an ultrasound and fluoroscopy assisted protocol, which allows swift concomitant FlowTriever use with minimal delay. In this setting, mechanical thrombectomy can result in a rapid return of circulation due to clot removal, thereby reducing the duration of RV failure and hypoperfusion.

Our study has several limitations. The relatively modest sample size, single-arm design and heterogeneity of the cohort pose challenges in drawing definitive conclusions. Furthermore, the availability of comprehensive haemodynamic, echocardiographic and procedural data varied among patients and was limited, especially in those patients presenting in the most severe clinical condition, i.e. circulatory arrest. Consequently, it is important to recognise that the positive effect on mPAP and RV dimensions may be overestimated and conclusions must be drawn with caution. Thirdly, haemodynamic parameters such as heart rate and blood pressure may have been influenced by inotropics and/or vasopressin and may therefore not correctly reflect the true clinical condition. Finally, two patients were lost to follow-up as they were transferred back to the referring centre within a few days. Procedure-related complications were not seen; however, bleeding complications within 30 days may have been missed. Survival status was available for all patients.

Future questions to be addressed include whether mechanical thrombectomy alone is superior to ST when testing for mortality safety outcomes, and overall cost. Furthermore, which patients are ideal candidates for this procedure is yet to be determined. Studies that are currently ongoing could shed light on this matter. The PEERLESS study will compare mechanical thrombectomy with catheter-directed thrombolysis [23]. Simultaneously, the PEERLESS II trial is a randomised controlled trial currently recruiting (NCT06055920) and will compare mechanical thrombectomy with anticoagulation versus anticoagulation alone in intermediate-risk patients. Furthermore, the investigators of the ongoing FLASH registry, a large multicentre study on the use of mechanical thrombectomy in intermediate-high and high-risk PE, will continue to report on their findings in the future. Finally, the TORPEDO-NL trial is currently recruiting and will randomise high-risk PE patients to mechanical thrombectomy or ST.