The Ultrasound-Accelerated Thrombolysis of Pulmonary Embolism (ULTIMA) trial is the only randomized controlled trial to compare CDI and anticoagulation vs. anticoagulation alone for intermediate-risk PE. The trial’s conclusion was that CDI improves RV systolic function compared to anticoagulation alone at 24 h and 90 days [49]. The results are in agreement with our experience comparing CDI and anticoagulation alone for intermediate-risk PE; it showed improved early right ventricular function and shorter ICU length of stay at the expense of potentially higher major complication rates with no difference in 30-day mortality or decompensation rates between the two groups [51].

Few studies have compared CDI to systemic thrombolytics for intermediate- and high-risk PE. The recent National Inpatient Sample (NIS) study compared the two treatment modalities in a propensity-matched comparison with in-hospital mortality as the primary outcome. Both in-hospital mortality and intracranial hemorrhage rates were lower for the CDI group compared to the systemic thrombolysis group, but CDI had a higher cost of hospitalization [48]. The study (and the database) was limited by the absence of confounding variables such as PE type, vasopressor use, thrombolytic dose, and anticoagulation regimens. Our recent experience comparing those two groups (CDI and systemic thrombolysis) reveals equivalent clinical and echocardiographic (RV/LV ratio) 30-day outcomes at a potentially lower major bleeding and stroke rates. Randomized studies comparing the two treatment modalities have some ethical and methodological concerns given the data suggesting a lower complication rate for CDI and the large sample size needed to show a clinically significant mortality difference between the two groups, if any.

Several non-comparative studies including the Prospective, Single-arm, Multi-center Trial of EkoSonic Endovascular System and Activase for Treatment of Acute Pulmonary Embolism (SEATTLE II); the Pulmonary Embolism Response to Fragmentation, Embolectomy, and Catheter Thrombolysis (PERFECT) registry; and multiple case series have presented their favorable results with CDI use [19, 40, 41, 45, 47, 50–53]. CDI clinical success rates, defined as treatment completion without major bleeding, stroke, or other major treatment-related events, decompensation, or in-hospital death, have been consistently high across studies ranging between 85 and 100% (Table 30.1).

Yet, despite the data about the efficacy and relative safety of CDI, they should not be portrayed as risk-free procedures [59, 60]. Death, major bleeding, and procedural-related complications such as coronary sinus rupture, cardiac tamponade, tricuspid valve rupture, arrhythmias, and acute renal failure requiring hemodialysis have been reported [50, 54]. Apart from the randomized ULTIMA trial and the PERFECT registry reporting a major bleeding rate of 0%, real-life experience with CDI has revealed a major bleeding rate ranging between 3 and 10%, which still compares favorably against systemic lysis with a range between 0 and 33%; a recent meta-analysis reported a major bleeding rate of around 9% associated with systemic thrombolysis [19, 38, 48–50, 54]. Different major bleeding criteria have led to variabilities and discrepancies in the reporting of major bleeding rates [46, 60]. As a result, pooled analyses of CDI complication rates remain statistically and clinically heterogeneous due to differences in patient selection and study design. The true complication rate of these interventions is difficult to obtain. Using the Global Utilization of Streptokinase and TPA for Occluded Arteries (GUSTO) criteria, we performed a meta-analysis on bleeding outcomes for CDI and obtained a pooled major bleeding rate of 3.5% [60]. Our experience with CDI has revealed a failure rate (major adverse event or no improvement) of around 15% and a major bleeding rate of around 7% [51, 54, 60]. Predictors of CDI failure and complications included PE type, older age, and major contraindications to thrombolytics. While CDIs can be used in patients with relative contraindications to systemic thrombolytics, a major contraindication remains a prohibitive factor for the use of thrombolytics with CDI. However, the near 0% stroke rate of CDIs compared to the 2% stroke rate of systemic thrombolytics has been the major drive toward increased CDI [19, 48–50, 52].

Cost-effectiveness is another aspect to consider in the comparison between CDI and anticoagulation or systemic thrombolysis. Data from the NIS revealed higher hospitalization costs associated with CDI compared to systemic thrombolysis [48]. The added cost with CDI will need to be justified by superior short- and long-term clinical outcomes or a superior quality of life, if any.

Enrolling randomized clinical trials comparing standard vs. ultrasound-assisted CDI for submassive PE [SUNSET sPE] and studying the optimal duration and dose of thrombolytics for submassive PE [OPTALYSE PE] are currently under way to allow a better understanding and standardization of the various CDI techniques and protocols [61].

The most recent American College of Chest Physician (ACCP) and European Society of Cardiology guidelines recommend systemic thrombolysis over CDI for high-risk acute PE patients and selected intermediate-risk PE patients who fail to improve or deteriorate with anticoagulation. CDIs are suggested over systemic thrombolysis if the patient has a high risk of bleeding, and local expertise and resources are available [12, 22].

The contemporary use of CDI has expanded to include catheter interventions with or without thrombolytics. The latter may involve thrombus fragmentation and/or aspiration/suction thrombectomy techniques; both have been described and used in patients with major contraindications to thrombolysis. However, their safety and efficacy remain controversial [3, 50, 62]. Depending on patient factors, contraindications to thrombolytics, and physician preference, the appropriate CDI technique is selected (Table 30.2).

The usual technique irrespective of catheter use involves ultrasound-guided vein access through a transjugular or transfemoral approach. Two single-lumen sheaths (two access sites) or single dual-lumen sheaths are used for bilateral PEs. An inferior vena cava (IVC) filter is placed prior to pulmonary arteriograms, if deemed necessary. A standard J wire is guided through the right atrium toward the right ventricle and then to the main pulmonary artery. When large devices are planned to be used, care should be taken at this step to prevent tricuspid valve injury (if the wire goes through the chordae tendineae). To avoid this, a pigtail catheter can be used to cross the valve or an inflated Swan-Ganz catheter. Once the pigtail is within the main pulmonary artery, an arteriogram is done to locate the clot and initiate thrombolysis by directing the lytic catheters toward the clot or suction thrombectomy by introducing the suction system.

Standard CDI use involves a 5 or 10 cm non-Food and Drug Administration (FDA)-approved multiside hole infusion catheter placed unilaterally or bilaterally in the pulmonary arteries across the heaviest clot burden. While there is no standardized protocol, our protocol uses a 2–4 mg of on-table thrombolytic infusion followed by the initiation of thrombolytic at a 0.5–1 mg/h per catheter. The catheter side holes allow for the local infusion of thrombolytics into the surrounding thrombus (Fig. 30.2a). A modification of this standard catheter system is the ultrasound-assisted thrombolysis. The EkoSonic Endovascular System (EKOS Corporation, Bothell, WA, USA) is an example of a FDA-approved setup that combines the standard CDI catheter with an ultrasound-emitting core (Fig. 30.2b). Ultrasound transducers are placed along a wire that is introduced on the infusion catheters with the proposed benefit of faster thrombus resolution compared to the standard CDI technique. Ultrasound energy loosens the thrombus’ fibrin strands and allows for more contact of the lytic agent with the thrombus [63]. Despite the theoretical advantage, the clinical superiority of ultrasound-assisted CDI over standard CDI remains to be proven [52, 64].

The optimal dose of thrombolytic has not been established and ranges between 15 and 25 mg across studies, which is substantially lower than the 100 mg standard systemic dose. The indication to stop lysis has not been established; however, clinical parameters such as heart rate, blood pressure, and oxygen requirement frequently guide the management of PE patients and the decision on termination of lysis. Other adjuncts contributing to the decision of lysis termination include invasive cardiac monitoring or echocardiographic parameters such as the right ventricle to left ventricle ratio. The OPTALYSE PE trial is currently randomizing submassive PE patients to one of four treatment arms with thrombolytic doses ranging between 4 and 24 mg and infusion times ranging between 2 and 6 h to determine the optimal treatment strategy for submassive PE.

The rate of heparin infusion during lytic administration is a controversial issue. While some recommend minimal heparin (500 units/h) to prevent bleeding complications, our group has agreed on a low-dose heparin protocol (atrial fibrillation protocol) with a target activated partial thromboplastin time (aPTT) between 60 and 80 s, to prevent further clot formation.

Major contraindications to thrombolytics have contributed to the advancements seen with catheter interventions. Hemodynamically unstable PE patients with a major contraindication such as a recent stroke or surgery can be attended to using several adjuncts to CDI. Rotating pigtail and balloon embolectomy catheters have been used as thrombus fragmentation CDI techniques [65]. At present, mechanical thrombus fragmentation is combined with aspiration/suction thrombectomy to prevent distal embolization of clot fragments, provide rapid clot resolution, and avoid the use of thrombolytics. Suction thrombectomy devices break down into two groups, small- and large-bore suction thrombectomy catheters, none of which are FDA approved for acute PE.

Practically any catheter attached to a large syringe can serve as a thrombectomy system. The current small-bore suction thrombectomy catheters being marketed include the Aspire (Control Medical Technology, Park City, UT) and Pronto XL (Vascular Solutions, Minneapolis, MN) catheters. The 14F Pronto XL catheter employs a 60 mL lockable syringe through which manual aspiration is performed. The catheter is rotated under suction which is controlled and modified using a roller clamp near the syringe [66] (Fig. 30.3a). The Aspire handheld mechanical aspirator can be connected to any catheter to allow forceful aspiration of the clot (Fig. 30.3b).

Large-bore suction devices such as the Vortex AngioVac System (AngioDynamics, Latham, NY) have recently entered the market and involve en bloc removal of emboli. The 18F suction device makes use of an extracorporeal veno-venous bypass circuit which drains, filters, and reinfuses the blood (cleared from clot) for up to 6 h. This FDA-approved technique (though not for PE) has been successful in small case series, but more evidence is needed [67]. The primary drawback is the rigidity of the catheter, which makes positioning and advancing the catheter in and beyond the pulmonary artery quite challenging (Fig. 30.4a). Other aspiration devices currently being assessed include the Indigo System CAT8 aspiration catheter (Penumbra Inc., Alameda, CA) and the FlowTriever (Inari Medical, Irvine, CA) (Table 30.2). The 20F FlowTriever device uses three expanding spiral wires to envelop portions of the clot, after which retraction and aspiration are simultaneously applied to capture most of the clot within the spiral wires [68] (Fig. 30.4b). The FlowTriever pulmonary embolectomy clinical study (FLARE) trial will help define the successes and failures of this device. The 8F Indigo catheter uses a powerful vacuum aspiration mechanism along with a separator wire that continuously breaks down the clot for constant aspiration. Its major drawback is aspirating a large volume of blood in the process (Fig. 30.4c). Both Indigo and FlowTriever catheters are relatively new; apart from a case report describing the FlowTriever mechanisms, the literature lacks evidence supporting either catheter and data about the safety of these devices is still absent [68].

The AngioJet Rheolytic Thrombectomy System (Boston Scientific, Marlborough, MA) is a pharmacomechanical system implementing a high-velocity jet to remove intravascular thrombus. The catheters have been used in the past for massive PEs; however, the FDA has recently issued a black box warning regarding their use after a series of adverse events and deaths [69, 70]. Therefore, it is currently best if this device is avoided.

The evidence from suction thrombectomy devices (small and large bore) comes from case series and reports. We are limited in drawing solid conclusions about the efficacy and safety of thrombectomy techniques; however, these devices remain the only option for high-risk PE patients with a high risk of bleeding. Recent guidelines suggest the use of suction thrombectomy devices in massive PE patients with high bleeding risks if appropriate expertise and resources are available [22].

Surgical pulmonary thrombectomy is reserved for patients with absolute contraindications or those who have failed systemic thrombolysis. But given the advancements in CDI, these patients are now being treated using aspiration/suction thrombectomy devices. In contemporary practice, surgical pulmonary thrombectomy is viewed as a last resort option given the significant morbidity and mortality associated with the procedure [71].

Implementing a multidisciplinary approach in the management of acute PE is gradually becoming the standard. At our institution, a pulmonary embolism response team (PERT) represents this approach and has a unified algorithm for the treatment of acute PE (Fig. 30.5). PERT members include pulmonary, critical care physicians, cardiologists, and vascular and cardiothoracic surgeons. The most important factor in determining management is hemodynamic stability and secondarily the bleeding risk. Hemodynamically unstable patients require immediate revascularization, and systemic thrombolysis is their most common treatment strategy, unless the bleeding risk is high at which we consider a catheter intervention. Hemodynamically stable patients are risk stratified into intermediate- or low-risk PE based on PESI score, laboratory markers, and imaging findings. Low-risk PE patients receive anticoagulation alone. Intermediate-risk PE patients are further stratified into intermediate-low- and intermediate-high-risk PE which in turn determines the treatment strategy. It is always prudent to assess the risk-benefit ratio of our interventions or treatments by assessing the bleeding risk. Given the lack of randomized trials comparing CDI techniques with anticoagulation and/or systemic thrombolysis, patient selection and multidisciplinary communication are of utmost importance to achieve better outcomes.