Catheter-Directed Thrombolysis for Submassive Pulmonary Embolism

Abstract

Acute pulmonary embolism (PE) is a leading cause of morbidity and mortality in the United States. PE associated with right ventricular strain, termed submassive or intermediate-risk PE, is associated with an increased rate of clinical deterioration and short-term mortality. Trials have demonstrated systemic thrombolytics may improve patient outcomes, but they carry a risk of major hemorrhage. Catheter-directed thrombolysis (CDT) may offer similar efficacy to and a lower risk of catastrophic hemorrhage than systemic thrombolysis. Three prospective trials have evaluated CDT for submassive PE; ULTIMA, SEATTLE II, and PERFECT. These trials provide evidence that CDT may improve radiographic efficacy endpoints in submassive PE with acceptable rates of major hemorrhage. However, the lack of clinical endpoints, long-term follow-up, and adequate sample size limit their generalizability. Future trials should be adequately powered and controlled so that the short- and long-term effectiveness and safety of CDT can be definitively determined.

Objectives: Upon completion of this article, the reader will be able to identify the clinical presentation of submassive PE, and the role of catheter-directed therapies in the treatment of the disease process.

Accreditation: This activity has been planned and implemented in accordance with the accreditation requirements and policies of the Accreditation Council for Continuing Medical Education (ACCME) through the joint providership of Tufts University School of Medicine (TUSM) and Thieme Medical Publishers, New York. TUSM is accredited by the ACCME to provide continuing medical education for physicians.

Credit: Tufts University School of Medicine designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit ™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

Pulmonary embolism (PE) is one of the leading causes of cardiovascular morbidity and mortality in the United States. ^{1 2} The clinical presentation of acute PE is variable; some patients may present with minor symptoms, while others present with sudden death. ¹ Significant variability exists among hospitals in the management of acute PE. Several international guidelines have been developed, each with stratification algorithms which attempt to identify those patients at high risk of clinical deterioration and those who may need therapeutic escalation.

In 2011, the American Heart Association (AHA) released updated guidelines with clear definitions of “massive,” “submassive,” and “low-risk” PE. ³ While most submassive PE patients do very well with anticoagulation alone, some clinically deteriorate. In 2014, the European Society of Cardiology (ESC) Guidelines on Pulmonary Embolism attempted to further delineate this diverse group of patients. ⁴ The term “intermediate” PE was utilized instead of submassive, and this group was further divided into “intermediate-high risk” and “intermediate-low risk. This strategy is clinically useful, as it suggests that high-risk intermediate PE patients should be closely monitored and considered for therapeutic escalation, while low-risk intermediate PE patients are likely to clinically improve with anticoagulation alone.

Submassive Pulmonary Embolism Epidemiology and Natural History

Some sources cite ∼25 to 30% of all PE cases fall into the submassive category, while others state the frequency may be as high as 50%. ^{5 6} In 1990, Goldhaber et al established the International Cooperative Pulmonary Embolism Registry (ICOPER) composed of 2,454 patients with acute PE. After multivariate modeling, it was demonstrated that pulmonary emboli associated with right ventricular dysfunction was associated with a 90-day mortality rate twice as high as those without signs of right heart strain; up to 19%. ⁷ It should be mentioned that since this registry data, several randomized control trials have demonstrated significantly lower 90-day mortality rates for submassive PE (2–3%). ⁸

Systemic Thrombolysis in Patients with Submassive Pulmonary Embolism

Hemodynamically stable patients with normal right heart function and no myocardial ischemia have an excellent short-term prognosis with conservative management. Patients with massive physiology have mortality rates as high as 65%, for which aggressive therapy is the standard of care. The ideal treatment strategy for submassive PE patients, who fall in between these two extremes, remains much less clear. Identifying which patients do well with anticoagulation alone and which require more aggressive therapies is a puzzle that has yet to be solved.

Guidelines suggest that patients trending toward “massive PE physiology” with a low risk of bleeding may be considered for thrombolysis to prevent further deterioration. ^{3 4} Factors which may suggest a patient is trending toward massive PE, impending RV failure, and cardiogenic shock include persistent hypoxia, tachycardia or tachypnea, as well as evidence of multiorgan failure including elevated lactic acid and progressive hepatic or renal dysfunction. While not performed routinely, monitoring with serial echocardiography may also provide insight into trending RV kinetics and impending RV failure. ⁹

For patients with submassive PE, there is more equipoise surrounding thrombolytic therapy. Over the past few decades, more than 15 randomized trials have attempted to find a mortality benefit to systemic thrombolysis. The largest of these studies, the PEITHO trial, randomized over 1,000 patients to tenecteplase and therapeutic heparin or therapeutic heparin alone. The investigators found no statistically significant mortality benefit of the thrombolytics, though clinical deterioration occurred more frequently in the heparin-only group. There was also a significant increase in major bleeding events as well as a rate of intracranial hemorrhage of greater than 2% in the group treated with tenecteplase. ¹⁰ A recent meta-analysis did show a slight but statistically significant mortality reduction in submassive PE patients who received thrombolytics compared with those who did not. ¹¹

Rationale of Catheter-Directed Thrombolysis for Submassive PE

With some studies demonstrating relatively low 90-day mortality for submassive PE patients treated with anticoagulation alone (2–3%), and clearly increased risk of bleeding with systemic thrombolytics, many clinicians are reluctant to embrace therapeutic escalation. ^{10 12} Catheter-directed thrombolysis (CDT), as an alternative method of fibrinolytic drug delivery, is still controversial. Some are concerned that procedural risks compound the inherent bleeding risk of the thrombolytic agents. ¹³ Others see CDT as an effective, minimally invasive, and safe treatment to prevent clinical deterioration and improve RV function. ⁹

The ideal candidate for CDT has central thrombus (main or lobar) and is stable enough to tolerate a prolonged infusion. ^{14 15} With systemic thrombolytics, altered hemodynamics may direct a substantial fraction of the therapeutic dose away from the thrombus and into the systemic circulation. ¹⁶ The unfavorable safety profile of systemic thrombolytic administration and the surgical morbidity of open embolectomy have made CDT an attractive alternative for the treatment of submassive PE. ¹⁷ The goal of CDT is to achieve local delivery of thrombolytics with less systemic drug delivery and a lower risk of hemorrhage. ¹⁸

Placement of a multi-sidehole catheter directly into the thrombus facilitates a high intra-clot thrombolytic drug concentration. ¹⁹ Several animal models have demonstrated that direct infusion of fibrinolytics also increases intrathrombus pressure, which enhances the effectiveness and speed of clot lysis. ¹⁹

With systemic thrombolysis, the standard dose of tissue plasminogen activator (tPA) commonly used to treat PE is 100 mg. ²⁰ At least two studies have evaluated the safety and efficacy of low-dose tPA (50 mg). The MOPETT trial demonstrated that low-dose tPA may be a safe and effective treatment for moderate PE. ^{20 21} However, these studies had several methodological limitations that question the generalizability and validity of this practice, including lack of central randomization, lack of clear inclusion criteria, atypical definitions, lack of an independent core laboratory, and conduct at a single center. CDT has the potential to achieve equally effective thrombolysis, with a much lower overall dose of the fibrinolytic agent. ²² With CDT, a standard infusion rate of 1 to 2 mg/hour, for a total dose of 15 to 30 mg of tPA, is typically used. ²³

The primary goals of treatment escalation with CDT are to safely and rapidly dissolve pulmonary artery thrombus and restore RV function. CDT is a less invasive option than surgical embolectomy and has lower periprocedural morbidity.

Techniques and Catheters for CDT

Access to the pulmonary arterial tree is usually gained via the common femoral or internal jugular veins. Many interventionalists prefer the right common femoral vein, as it offers a relatively straight course to the right heart via the inferior vena cava. ²⁴ Ultrasonographic guidance is recommended for venous access to avoid inadvertent arterial puncture.

Once venous access is obtained, angiography of the right and left main pulmonary arteries can be performed. Catheters commonly used to access the pulmonary tree are usually 5 to 7 Fr and come in two basic varieties: pigtail and balloon tipped ( Fig. 1 ). ²⁴ The Grollman catheter is a popularly used option for accessing the pulmonary outflow tract. It features a gentle curve with a 90-degree reversed secondary curve, 3 cm from the pigtail tip, which allows the operator to traverse the contours and valves of the right heart. ²⁵

Fig. 1.

Examples of catheter options for pulmonary artery access. From left to right: Nyman, Grollman, standard straight pigtail flush catheter, balloon tipped occlusion catheter with distal side holes. (Reprinted with permission from Biam. ²⁴ )

At this point, pressure measurements of the pulmonary arteries are obtained to record baseline hemodynamics and determine contrast injection rate. ²⁶ A test injection is usually performed to help determine the injection rate. In a majority of circumstances, a rate of 20 to 25 mL/second is appropriate. In patients with severe pulmonary hypertension or tenuous right heart hemodynamics, the injection rate should be reduced by ∼10 to 15 mL/second. ²⁶ Though not routinely performed, right atrial and ventricular pressures may also help in the assessment of right circulation hemodynamics. ²⁷

The location of the pulmonary thrombus is known in most cases from the baseline CT angiogram ( Fig. 2 ). ²⁷ Next, a wire (e.g., exchange length Rosen or Bentson) traverses the thrombus, over which thrombus extraction devices or multi-sidehole catheters are advanced. Though any length sheath may be used, many interventionalists prefer a long sheath whose tip terminates in the main pulmonary artery. This longer sheath serves two purposes: it prevents unwanted motion or retraction of the catheter during the infusion and allows serial pulmonary artery pressure measurements. ²⁷ Two catheters are typically placed because pulmonary emboli are usually bilateral.

Fig. 2.

Case background information: a 40-year-old man with history of systemic lupus erythematosus and morbid obesity who underwent laparoscopic sleeve gastrectomy 4 weeks prior to presentation of acute shortness of breath. ( a ) Coronal slice of initial CT pulmonary angiogram demonstrating large filling defect in the right main pulmonary artery (arrow). ( b ) Axial slice from initial CT pulmonary angiogram demonstrating dilated right ventricle and straightening of the interventricular septum. ( c ) Intraprocedure fluoroscopic image demonstrating bilateral pulmonary arterial catheters, with catheter tips in the right and left pulmonary arteries (arrows). ( d ) Coronal slice of postprocedure CT pulmonary angiogram demonstrating near-complete resolution of the pulmonary embolism (arrow). ( e ). Axial slice of postprocedure CT pulmonary angiogram demonstrating resolution of right ventricular dilatation and convexity of the interventricular septum.

Once the catheter is embedded in the thrombus, a bolus of 2 to 6 mg of thrombolytic may be given prior to initiation of the infusion. ²⁸ The standard infusion rate is 1 to 2 mg/hour for a single catheter or 0.5 to 1 mg/hour if two catheters are in place. The infusion is done in typically 12 to 24 hours, for a total thrombolytic dose of 15 to 30 mg. ^{14 15} During the thrombolytic infusion, a subtherapeutic infusion of heparin is concurrently administered, with a partial thromboplastin time goal of ∼40 to 60 seconds. ²⁸ Notably, there is still significant center-to-center variability in dosing and treatment protocols.

Devices for Catheter-Directed Therapy

Catheter-directed therapy offers an arsenal of endovascular techniques for the removal and lysis of pulmonary arterial thrombi. ²⁹ In massive PE, the rotating pigtail catheter technique is a simple, cost-effective, and highly successful method for mechanical thrombus fragmentation. ³⁰ Numerous other mechanical techniques, such as balloon angioplasty, are also commonly used, but there is a scarcity of data on the true safety profile of these practices. ²³ There are also many catheter-based devices which are currently available for thrombus aspiration and extraction; however, these are usually reserved for massive PE. ^{30 31 32} These devices are a useful treatment option in patients who have absolute contraindications to thrombolytics. They also may help reduce the dose of thrombolytics in patients with relative contraindications, or can be used to rapidly reduce thrombus burden in patients with tenuous right heart kinetics. Safety and efficacy data of these devices are extremely limited, especially in the setting of submassive PE. Though clot fragmentation and mechanical thrombolytic techniques may be highly effective at achieving rapid thrombolysis, clot dislocation and distal embolization are well-documented potential complications to which the operator must remain cognizant. ^{33 34} Trials supporting the safety and clinical effectiveness of these devices in the setting of submassive PE are needed to justify routine adoption.

Ultrasound-assisted thrombolysis (USAT) catheters are a variant of the standard multi-sidehole catheter. Though the exact mechanism has not been elucidated, high-frequency ultrasound waves are thought to improve penetration of the thrombolytic agent and also accelerate enzymatic thrombolysis. ^{29 35} The catheter contains microinfusion pores for enhanced drug delivery as well as multiple ultrasound transducers along its length that deliver low-intensity, high-frequency ultrasound pulses. ²⁶ Data supporting its use include two prospective trials focused on CDT for severe PE: ULTIMA and SEATTLE II. In 2014, the EKOS system gained approval by the United States Food and Drug Administration for use in the pulmonary circulation.

It remains unclear whether USAT provides greater efficacy compared with standard CDT catheters. ³⁶ Lin et al conducted a methodologically limited retrospective study comparing the EKOS system to standard catheters in massive PE patients; the study demonstrated improved treatment outcomes and shorter time to thrombolysis in the USAT group. ³⁷ In 2014, Engelberger et al randomized 48 patients with acute deep vein thrombosis to treatment with USAT or standard CDT. In all patients, the thrombus was treated with the EKOS system, but ultrasound component was not switched on in the standard CDT group. They found no difference in thrombus burden in the patients treated with the ultrasound component activated compared with those who had it turned off. ³⁸ A subgroup analysis of the PERFECT prospective registry demonstrated no significant difference in pretreatment versus posttreatment pulmonary artery pressures between USAT catheters and standard catheters. ³⁹ Thus, the ideal catheter option and treatment protocol for CDT remains unclear. ^{39 40}

Current Literature on CDT

Compared with systemic thrombolysis, CDT has been insufficiently studied. The best evidence for CDT comes from three prospective trials: ULTIMA (a randomized controlled trial), SEATTLE II (a single arm cohort study), and PERFECT (a single-arm multicenter registry).

In 2014, the ULTIMA trial randomized 59 patients to receive treatment with ultrasound-assisted CDT plus heparin or heparin alone. The primary endpoint was reduction in right ventricle to left ventricle diameter (RV/LV) ratio by echocardiography. The investigators found a significant difference between the RV/LV ratio at 24 hours and baseline only in the group treated with CDT. ¹⁴ At 90 days, there was no statistically significant difference in the RV/LV ratio between the two groups, though RV dysfunction was more common in the heparin group. The clinical outcomes assessed included death, major/minor hemorrhage, hemodynamic decompensation, and recurrent venous thromboembolism (VTE). At the end of 90-day follow-up, there was one reported death (in the heparin-only group), no major bleeds, and no episodes of hemodynamic decompensation or recurrent VTE.

Perhaps the greatest limitation of the ULTIMA trial is that the primary endpoint (RV/LV ratio reduction) was not a clinical one. The trial was not powered to evaluate clinically relevant endpoints or the risk of bleeding. It is encouraging that no major bleeding complications and no serious adverse events related to the intervention were encountered, but strong conclusions about CDT's safety cannot be made based on these results. Of note, the trial excluded 84% of the screened patients, which potentially decreases the generalizability of the study, though the exclusions did approximate real-world practice.

The SEATTLE II study was a prospective, single-arm, multicentered trial designed to evaluate the safety and efficacy of ultrasound-assisted CDT. ¹⁵ The study enrolled 150 patients, 119 of whom were classified as submassive and 31 as massive. The primary safety endpoint was major bleeding within 72 hours; the primary efficacy endpoint was reduction in the RV/LV ratio at 48 hours by CT angiography. Hemorrhages were classified using the GUSTO (Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries) bleeding criteria. There were 17 major bleeds (1 GUSTO severe and 16 GUSTO moderate) and no intracranial hemorrhages. The one GUSTO severe bleed was a groin hematoma that caused transient hypotension and required vasopressor support. At 48 hours, there was a statistically significant reduction in RV/LV ratio and pulmonary artery systolic pressure compared with pre-CDT measurements.

Thirty-five patients did not have follow-up CT performed within the 48-hour time frame. Patients with complicated postprocedural events for whom imaging needed to be delayed, or those who had an excellent clinical response and were deemed to not need follow-up imaging, were potentially excluded. ¹⁵ Overall, the greatest limitation of SEATTLE II was the lack of a control/comparator group. Consequently, the safety and efficacy of CDT compared with systemic thrombolysis or anticoagulation cannot be ascertained.

The PERFECT study by Kuo et al was a prospective, multicentered registry, which enrolled 101 patients—73 with submassive PE and 28 with massive PE. The study did not require a specific procedural approach; each patient received immediate treatment with catheter-directed mechanical/pharmacomechanical thrombectomy and/or CDT. The PERFECT registry's primary endpoint was clinical success, defined as meeting all three of the following criteria: stabilization of hemodynamics, improvement in pulmonary hypertension and/or right heart strain, and survival to hospital discharge. ³⁹ Clinical success was achieved in 71 of 73 (97.3%) submassive PE patients. There were neither major procedure-related complications nor episodes of major bleeding or intracranial hemorrhage. There were 13 episodes of minor bleeding, most of which were related to the venous access site; all were self-limited without the need for blood transfusions.

Catheter-directed therapy was the initial treatment in 98 of 101 patients. The remaining three patients first received systemic tPA and were subsequently rescued with CDT after clinical deterioration necessitated further intervention. There were two patients in the PERFECT registry for whom CDT failed. It is postulated that irreversible pulmonary hypertension/RV failure or a component of acute on chronic PE may have been present. Acute on chronic thrombus poses a particularly difficult challenge, because these patients may not have a significant reduction in their pressures or thrombus burden following CDT. Identifying which acute on chronic PE patients require urgent thrombolysis is an important question for future prospective trials to answer. Similar to the SEATTLE II trial, the major limitation of the PERFECT registry is its lack of a control group. Thus, no comparative conclusions can be drawn.

These three studies provide some evidence that CDT has the potential to be relatively safe and efficacious in the treatment of submassive PE. All three studies found the decrease in pulmonary artery pressure after CDT to be statistically significant ( Fig. 3 ). However, each trial lacked key methodological features to allow CDT to be definitively accepted and routinely offered as a therapeutic option. The studies' limitations include low power (all three), the use of nonclinical endpoints (ULTIMA and SEATTLE II), and lack of comparator groups (SEATTLE II and PERFECT). ⁴¹ An important limitation of all three trials is the lack of long-term follow-up; none had clinical follow-up longer than 90 days postprocedure.

Fig. 3.

Results from PERFECT, SEATTLE II, and ULTIMA trials demonstrating similar mean systolic pulmonary artery pressures, at baseline and after intervention with CDT. (Reprinted with permission from Kuo et al. ³⁹ )

Future Directions

Determining which submassive PE patients should be considered for escalation of therapy is an ongoing research topic. ⁸ The ESC system of stratifying intermediate PE patients into high- and low-risk categories is a partially validated attempt to acknowledge differences between submassive PE patients, but more data are needed before this model becomes widely accepted. Clot location or overall clot burden could play a role in risk stratification, but neither the ESC nor the AHA guidelines account for these factors.

The three prospective trials on CDT for submassive PE provide an initial framework for future trials. These investigations should focus on clinically oriented short- and long-term outcomes that are valued by patients and providers. Long-term outcomes may include quality of life, dyspnea, exercise tolerance (e.g., 6-minute walk test), RV function, and clot burden. These studies should also evaluate the incidence of chronic thromboembolic pulmonary hypertension after CDT in patients with submassive PE. ⁴¹

A multidisciplinary consensus panel concluded that a triple-arm trial comparing systemic thrombolysis, CDT, and anticoagulation would be impractical given the number of patients and time necessary to complete the trial. ⁴¹ The panel concluded that a randomized controlled trial directly comparing CDT plus anticoagulation to anticoagulation alone in submassive PE patients was the highest research priority. ⁴¹ To impact the current body of literature, this type of trial would need to rigorously assess short- and long-term endpoints and bleeding rates. Repetition of a short-term trial, with only enough patients to demonstrate a difference in a physiologic parameter, would not influence current practice. ⁸

Future research is also needed to define and standardize the CDT protocol in submassive PE patients. The ideal and maximum thrombolytic dose, as well as catheter-directed techniques and device superiority, needs to be further delineated. ^{23 26} The use of ultrasound-assisted catheters versus standard catheters is still heavily debated. ³⁹ CDT has great potential to improve outcomes for patients with submassive PE, but proven clinical effectiveness and safety (as determined by well-designed, robust, and appropriately powered trials) are necessary for it to become the standard treatment for these patients.