Comparative Outcomes of Ultrasound-Assisted Thrombolysis and Standard Catheter-Directed Thrombolysis in the Treatment of Acute Pulmonary Embolism

Abstract

Objectives:

The objective of this study was to compare the outcomes of patients undergoing ultrasound-accelerated thrombolysis (USAT) and standard catheter-directed thrombolysis (CDT) for the treatment of acute pulmonary embolism (PE).

Methods:

The records of all patients in our institution having undergone CDT or USAT for massive or submassive PE from 2009 to 2014 were retrospectively reviewed. Standard statistical methods were used to compare characteristics and to assess for longitudinal change in outcomes.

Results:

Sixty-three patients, 27 CDT and 36 USAT, were treated for massive (12.7%) or submassive (87.3%) PE. Of which, 96.8% were treated for bilateral PE. Baseline patient characteristics did not differ between the 2 treatment groups. There was no difference in total dose of lytic administered (CDT: 23.2 ± 13.7 mg; USAT: 27.5 ± 12.9 mg; P = .2). Two patients in the CDT and 1 in the USAT groups required conversion to surgical thrombectomy (CDT: 7.4%; USAT: 2.8%; P = .6). Rates of major and minor bleeding complications (CDT: 11.0%; USAT: 13.9%; P = .8) did not differ significantly between the CDT and USAT groups. Estimated survival at 90 days was 92% for CDT and 93% for USAT and 82% at 1 year for both groups (P = .8). All echocardiographic parameters improved significantly from baseline to 1-year follow-up, but quantitative improvement did not differ between groups.

Conclusion:

This study suggests no statistical differences in clinical and hemodynamic outcomes or procedural complication rates between USAT and standard CDT for the treatment of acute PE. Prospective studies are needed to further evaluate comparative and cost-effectiveness of different interventions for acute massive and submassive PE.

Introduction

Acute pulmonary embolism (PE) carries a high morbidity and is the third leading cause of cardiovascular mortality in the United States, accounting for 5% to 10% of in-hospital deaths,¹ with a 90-day mortality rate as high as 15%.² Treatment of acute PE has traditionally been limited, with the majority of patients receiving systemic anticoagulation for stabilization and eventual dissolution of the thrombus. Systemic infusion of thrombolytic medications has played an increasing role in the management of patients deemed to be at higher risk, potentially decreasing mortality and hemodynamic compromise but at the expense of increased bleeding complications.^3–5 Surgical thrombectomy under cardiopulmonary bypass remains an option of last resort due to the associated morbidity and mortality.^6,7

The recent development of catheter-directed therapies such as catheter-directed thrombolysis (CDT), ultrasound-accelerated thrombolysis (USAT), and pharmacomechanical thrombectomy has introduced more tools for the treatment of acute PE.¹ Proponents of these techniques suggest that they may provide a similar therapeutic benefit as systemic thrombolysis, while decreasing the dose of thrombolytic required and potentially decreasing the risk of adverse bleeding events.^8–10

Catheter-directed thrombolysis requires placement of a multiside hole infusion catheter within the pulmonary arterial thrombus burden under angiographic guidance. Thrombolytic medications are slowly infused through the catheter, which is left in place for the duration of the treatment. Ultrasound-accelerated thrombolysis (Ekosonic device; EKOS Corp, Bothell, Washington) is a modification of this therapy utilizing a proprietary system of local high-frequency, low-power ultrasound to dissociate the fibrin matrix of the thrombus, allowing deeper penetration of lytic medication.¹¹ Such devices have been used in multiple vascular beds and come with a significant increase in incremental disposable cost. Although several observational series have been published regarding the ability of these therapies to decrease the PE clot burden and improve echocardiographic markers of right heart function compared to systemic thrombolysis, few comparative studies have been performed. The objectives of this study were to compare hemodynamic and clinical outcomes for CDT and USAT in the treatment of patients with acute PE.

Methods

This single-institution retrospective cohort study was approved and exempted from informed consent by the institutional quality improvement review committee.

Patients

Patients were identified through the use of the institutional electronic medical record. Patients who underwent CDT for the treatment of acute PE from 2009 to 2014 were included in the study. For diagnosis and classification of PE, all patients underwent diagnostic contrast-enhanced chest computed tomography (CT), echocardiography, and evaluation of cardiac bio-markers (troponin) and were subsequently classified as having high risk (massive) or intermediate risk (submassive) PE according to the current American Heart Association and European Society for Cardiology consensus guidelines.^7,12 High-risk PE was defined as the presence of sustained hypotension requiring vasopressor or inotropic support. Intermediate-risk PE was defined as the absence of hemodynamic instability but with right ventricular dysfunction on echocardiography or CT angiography, a positive troponin biomarker, or Simplified Pulmonary Embolism Severity Score¹³ greater than 1. Patients who did not meet the criteria for high- or intermediate-risk PE and patients who underwent treatment at an outside institution before transfer were excluded from the study.

Procedures

For CDT, the choice of either standard CDT or USAT and the technique used for each interventional procedure were based on individual physician preference; all interventionalists were proficient with either system. Catheter-directed thrombolysis was performed using Cragg-McNamara (Boston Scientific, Marlborough, Massachusetts) or Uni-Fuse (AngioDynamics, Latham, New York) multisidehole catheters, and USAT was performed using the EkoSonic system (EKOS Corp).

All procedures were performed in a standard angiography suite or hybrid operating room with a fixed-imaging fluoroscopic system. The decision for placement of a vena cava filter was determined preoperatively by the interventionalist and multidisciplinary treatment team. Venous access was obtained from either a femoral or jugular approach under ultrasound guidance. A vena cava filter was placed at this time at the discretion of the interventionalist for patients judged to have low cardiopulmonary reserve or with residual iliofemoral venous thrombus identified on duplex. Areas of heavy thrombus burden based on initial pulmonary angiography and CT findings were then identified, with subsequent catheter placement across the area of greatest thrombus burden. For patients with bilateral emboli, a second catheter was placed in a similar fashion in the contralateral pulmonary arterial tree.

A loading dose of 2 to 4 mg of alteplase (Activase; Genentech, South San Francisco, California) was administered at the interventionalist’s discretion; initiation of low-dose thrombolytic infusion was then begun at rates ranging from 0.5 to 1 mg per catheter per hour. Alteplase was the sole thrombolytic medication used for infusion. Concurrent ultrasound therapy using the EkoSonic system, if applicable, was initiated prior to leaving the interventional suite.

All patients were monitored in an intensive care unit (ICU) during the duration of infusion therapy. Fibrinogen levels and activated partial thromboplastin times were drawn, recorded, and assessed at set intervals. Regardless of treatment modality, patients were maintained on a systemic therapeutic unfractionated heparin infusion following institutional dosing protocol for the duration of thrombolytic therapy, then eventually transitioned to long-term anticoagulation therapy (warfarin, enoxaparin, or a factor Xa inhibitor) prior to discharge. Repeat angiography was performed at the discretion of the interventionalist.

Conditions for termination of the lysis procedure varied across procedure periods. From 2009 to 2011, termination of thrombolysis was based upon clinical improvement as well as repeat angiography demonstrating greater than 50% clot resolution; lytic infusion time and dose were tailored accordingly but were discontinued if no improvement was noted after 48 hours. From 2011 onward, termination was generally based on improved clinical and/or echocardiographic parameters assessed 12 to 24 hours after the initiation of thrombolysis and total dose of lytic administered rather than thrombus clearance on imaging. Parameters for termination included significant improvement in hemodynamics such as tachycardia and hypotension, reduction in supplemental oxygen or dyspnea, and improvement in right heart parameters on repeat echocardiogram. This was followed by removal of the infusion catheters at the bedside.

Intervention characteristics were also stratified into 2 groups based on the year the procedure was performed (2009–2011 and 2012–2014) to account for changes in CDT protocol and as USAT was not performed in our institution prior to 2012.

Outcome Measures

Primary outcomes were survival and periprocedural hemodynamic stabilization. Hemodynamic stabilization was defined as either the resolution of existing hemodynamic instability (sustained hypotension or need for vasopressor/inotropic support) or the absence of new onset of hemodynamic instability in previously stable patients subsequent to the initiation of catheter-directed therapy.

Adverse events included periprocedural death, major bleeding requiring transfusion or intervention, minor bleeding, and stroke. Other secondary outcomes included progression to surgical conversion, length of ICU stay, supplemental oxygen use at discharge, and recurrent PE.

All baseline, immediate postprocedure, and within-1-year follow-up echocardiograms were recorded and reanalyzed. Echocardiographic parameters included qualitative measures of right ventricular systolic function as determined by the interpreting cardiologist, right ventricular size, right ventricular to left ventricular end-diastolic size ratio (RV/LV ratio), tricuspid regurgitant jet velocity, and the estimated systolic pulmonary arterial pressure. Two independent readers measured and verified the RV/LV ratio for each echocardiogram.

Statistical Analysis

Statistical comparisons for baseline characteristics were performed using Fisher exact, 2-sample t test, Wilcoxon rank sum test, analysis of variance, and Kruskal-Wallis test depending on the nature of the data. Time-to-event data were analyzed using the Kaplan-Meier method and survival curves compared using the log-rank test. Paired t testing and change scores were used for repeated measures data. All statistical analytics were performed using Stata SE 13.1 (College Station, Texas).

Results

Patient Characteristics

Sixty-nine patients undergoing catheter-directed intervention for acute PE were identified; 6 did not undergo either CDT or USAT and were excluded. Sixty-three patients were included in the study—27 (43%) patients underwent CDT and 36 (57%) patients underwent USAT. There were no significant differences in baseline characteristics between the CDT and USAT subgroups. The 2 cohorts had a similar average age of around 60 years and were predominantly female (Table 1). Nearly all patients had bilateral PE and received treatment with 2 catheters. There were more high-risk PEs in the CDT group and more intermediate-risk PEs in the USAT group, but these differences were not significant. Other important baseline characteristics such as concurrent diagnosis of deep venous thrombosis (DVT), presence of a hypercoagulable state, presence of a malignancy, or history of PE did not differ between the groups.

Table 1.

Baseline Characteristics by Treatment Type.^a

	CDT (N = 27)	USAT (N = 36)	P
Age	57.07 ± 17.49	60.63 ± 12.78	.35
Male sex	10 (37.0)	17 (47.2)	.45
PE severity category			.065
High risk	6 (22.2)	2 (5.6)
Intermediate risk	21 (77.8)	34 (94.4)
Troponin elevation	21 (77.8)	29 (80.6)	.5
Bilateral PE	25 (92.5)	36 (100)	.17
Hypercoagulable state	3 (11.1)	3 (8.3)	.00
Recent surgery	9 (33.3)	6 (16.6)	.15
Malignancy	3 (11.1)	7 (19.4)	.49
Hormone therapy	2 (7.4)	3 (8.3)	1.00
Recent travel	4 (14.8)	8 (22.2)	.53
Concurrent DVT	12 (44.4)	20 (55.6)	.45
History of DVT	6 (22.2)	5 (13.8)	.51
History of PE	6 (22.2)	4 (11.1)	.30
Coronary artery disease	4 (14.8)	4 (11.1)	.72
Congestive heart failure	1 (3.7)	2 (5.6)	1.00
Pulmonary hypertension	1 (3.7)	2 (5.6)	1.00

Abbreviations: CDT, catheter-directed thrombolysis; DVT, deep venous thrombosis; PE, pulmonary embolism; USAT, ultrasound-accelerated thrombolysis.

^aAll values are n (%).

Procedural Details

The CDT subgroup had more angiography suite trips (2.0 ± 0.7 vs 1.4 ± 0.6; P < .001) and longer total lysis procedure time (23.7 ± 13.9 hours vs 15.4 ± 5.3 hours; P = .002), but the 2 groups had a similar total amount of tissue plasminogen activator infused (23.2 ± 13.7 mg vs 27.6 ± 12.9 mg; P = .6). Vena cava filter placement was more common in the CDT group than the USAT group (62%, n = 13 vs 29%, n = 10; P = .03).

Clinical Outcomes

Primary outcomes did not differ significantly between the 2 groups. Survival for CDT and USAT groups was 100% and 97% at 30 days, 92% and 93% at 90 days, and 82% in both groups at 1 year (all standard errors <10%, log-rank P = .8; Figure 1). There was only 1 periprocedural death (3%, P = 1.0); this was a patient with submassive PE in the USAT group who underwent the intervention without immediate complication but had sudden cardiac arrest and died just prior to anticipated discharge. All but 5 patients demonstrated hemodynamic stabilization—3 in the CDT group and 2 in the USAT group (11.1% vs 5.6%; P = .6). Three of these patients progressed to surgical thrombectomy via median sternotomy with cardiopulmonary bypass (CDT: N = 2, 7.4%; USAT: N = 1, 2.8%), whereas the remaining 2 recovered after continued infusion. Catheter-directed thrombolysis patients trended toward a longer average length of ICU stay compared to USAT (4.96 ± 5.61 days vs 2.83 ± 4.55 days; P = .1), although this trend did not persist after stratification by year of procedure performed (CDT vs USAT after 2012: 2.82 ± 1.40 days vs 2.91 ± 4.75 days; P = .9). Overall length of stay (LOS) did not vary between intervention types or eras of procedure (median LOS prior to 2012: 8 days [interquartile range, IQR: 6–13]; after 2012: 7 days [IQR: 4.5–8]; P = .14).

Figure 1.

Survival between the 2 treatment cohorts. All standard errors are <10% within the plot range.

There were no differences in rates of major or minor bleeding (Table 2), surgical conversion, supplemental oxygen use at discharge or follow-up, or recurrent PE. No bleeding complications required operative intervention. There were no strokes in either group.

Table 2.

Adverse Events by Treatment Type.^a

	CDT (N = 27)	USAT (N = 36)	P
Hemodynamic stabilization	24 (89.0)	34 (94.4)	.6
Perioperative death	0 (0.0)	1 (2.8)	1.00
Major bleeding	0 (0.0)	2 (5.6)	.5
Minor bleeding	3 (11.1)	3 (8.3)	1.00
Surgical conversion	2 (7.4)	1 (2.8)	.57

Abbreviations: CDT, catheter-directed thrombolysis; USAT, ultrasound-accelerated thrombolysis.

^aAll values are n (%).

Hemodynamic Outcomes

Echocardiographic parameters were measured at baseline and at 1 year. All parameters demonstrated significant improvement over this time period without significant differences between the 2 groups (Table 3). Echocardiograms performed immediately postprocedure were also recorded, but there were not enough examinations to conduct a reliable comparative statistical analysis.

Table 3.

Echocardiogram Parameters From Baseline to 1-Year Follow-Up.^a

	CDT		USAT		P
	Baseline	1-Year	Baseline	1-Year
RV/LV ratio	1.10 ± 0.06	0.86 ± 0.12	1.13 ± 0.19	0.77 ± 0.19	1.00
Tricuspid regurgitation jet velocity, m/s	3.25 ± 0.90	2.48 ± 0.48	3.34 ± 0.64	2.45 ± 1.04	.85
Systolic PA pressure, mm Hg	58.33 ± 24.96	36.42 ± 9.50	57.01 ± 17.10	37.98 ± 19.39	.35
RV dilation	10 (90)	2 (17)	15 (100)	6 (67)	.18
RV systolic dysfunction	10 (90)	0 (0)	11 (73)	1 (7)	1.00

Abbreviations: CDT, catheter-directed thrombolysis; LV, left ventricular; PA, pulmonary artery; RV, right ventricular; USAT, ultrasound-accelerated thrombolysis.

^aValues are mean ± standard deviation or n (%).

Discussion

The usage of catheter-directed techniques to treat acute high- or intermediate-risk PE has increased over the past several years in both common practice and in the published literature as practitioners attempt to find a therapy that provides the hemodynamic improvements and mortality benefits of systemic thrombolysis while reducing the risk of adverse bleeding events.^1,3,8

Several observational series have shown early efficacy of catheter-directed techniques for improving hemodynamic parameters measured using echocardiography.^8,10 The single-arm SEATTLE II¹⁴ and randomized ULTIMA⁹ trials showed improvement in cardiovascular hemodynamics with USAT, with ULTIMA demonstrating an immediate advantage of USAT over anticoagulation. Initial results from the PERFECT registry by Kuo and colleagues¹⁵ also showed global improvements in hemodynamics for catheter-directed lysis and a high degree of clinical success.

Despite the overall efficacy of CDT, limited comparative data exist for different interventional methods as many recent series report primarily on thrombolysis with USAT.⁸ An early series by Lin and colleagues¹⁶ of 33 patients with high-risk PE suggested potential benefit of USAT for angiographic clearance of thrombus burden with more bleeding events in the CDT group. Using data from the PERFECT registry, Kuo and colleagues also analyzed comparative outcomes of CDT and USAT but could not demonstrate any hemodynamic or clinical difference between the 2 therapies while noting the comparatively higher disposable cost of USAT catheters.¹⁵

Our retrospective analysis of 63 patients examines short-term clinical outcomes and midterm echocardiographic measures of hemodynamic compromise in patients undergoing CDT. Our results do not show a difference between USAT and CDT, with both modalities demonstrating similar rates of clinical outcomes such as survival and hemodynamic stabilization with similar procedure length and lytic dose in the time-adjusted cohorts. Midterm echocardiographic parameters were also similar between the 2 groups.

Recent studies have moved away from relying on angiographically recognized thrombus burden as a biomarker of risk,^7,9 instead focusing on clinical outcomes and hemodynamic measures of right heart function. This was also reflected in the change in practice at our institution between 2011 and 2012, as CDT supplanted systemic thrombolysis as the therapy of choice for high- and intermediate-risk PE, and USAT was introduced. This shift from reliance on thrombus burden clearance to clinical and hemodynamic parameters for termination of thrombolysis partially explains the higher lysis duration time and mean number of procedures in the CDT group prior to 2012. During this early period, angiographic clot resolution was used to decide on termination of lysis and required repeat angiography and prolongation of lytic infusion. The overall management of patients with acute PE in both interventional and noninterventional settings has also become more standardized at our institution with the formation of a multidisciplinary PE response team composed of a lead attending pulmonologist, an interventionalist (vascular surgeon and/or cardiologist), and a cardiac surgeon.

The expected benefit of USAT has been dependent on the device’s ability to increase penetration of lytic into thrombus using high-frequency, low-power ultrasound, due to its reversible effects on fibrin dissociation.^9,11 This benefit has been shown to result in faster thrombus clearance in selected vascular beds in some studies, such as the recently published DUET study comparing USAT and CDT in arterial occlusions.¹⁷ However, contrary evidence exists in the venous circulation, with the recent BERNUTIFUL trial demonstrating no difference in time to thrombus clearance in lower extremity DVT.¹⁸

Our results and other published data on the use of USAT for venous thromboembolism indicate that further prospective studies are needed to determine the comparative and cost-effectiveness of these 2 therapies in order to justify the higher disposable cost of USAT. In addition, the claim for increased effectiveness with USAT needs to be further verified in patients with acute PE, in order to avoid any false sense of security from the promise of rapid clot clearance in potentially unstable patients who might otherwise benefit from other interventions.

Limitations

This study is limited by the retrospective nature of data collection and analysis. Despite being among the larger of the observational catheter-directed PE cohorts, the sample size remains statistically small, limiting the ability to draw inference from the data. The cohorts also reflect a changing procedural protocol within our institution, as intervention characteristics and conditions for lysis termination and ICU or hospital discharge become more standardized with time. Another limitation to the study is the absence of consistent postprocedure testing demonstrating immediate echocardiographic improvement, around which some studies have been focused. Finally, functional and quality of life measures were not collected predischarge or at follow-up, limiting the ability to determine the effectiveness for patient-centered outcomes not relating to mortality. However, despite the stated limitations, the cohort groups were similar and had a similar treatment course, allowing us to make our comparative analysis.

Conclusion

Catheter-directed interventions for acute high- or intermediate-risk PE demonstrate efficacy in clinical outcomes. However, the comparison of clinical and hemodynamic outcomes for standard CDT and USAT does not demonstrate significant differences to justify the preferential use of USAT and merits future prospective study of the comparative and cost-effectiveness of these 2 treatment modalities.