Intermediate-Term Outcomes for Patients With Submassive Pulmonary Embolism Treated With Catheter-Directed Thrombolysis

Objective. We aimed to assess the intermediate-term outcomes for patients receiving catheter-directed thrombolysis (CDT) for submassive pulmonary embolism (PE). Background. Previous research has shown improvements in right ventricular (RV) function and dilation at 24 hours when CDT was used to treat submassive PE. Methods. Consecutive patients presenting with submassive PE treated with directed t-PA infusion at a single center were identified and included in this study. Outcomes included cardiovascular mortality, RV function by echocardiogram, 30-day readmission, and major bleeding. Results. The study population was 79 patients with submassive PE; 46% were men, with an average age of 58 years and an average pulmonary embolism severity index (PESI) score of 108. One patient died of cardiovascular causes during the index hospitalization. There were no additional deaths within 30 days of admission. The observed 30-day mortality rate was low compared with that predicted by PESI (1.3% vs 4.0%-11.4%). Fifty-two patients had follow-up echocardiography available for evaluation after CDT. Of these, 62% showed return to normal RV function and size, and 19% demonstrated mild residual RV dysfunction or dilation. Eight patients (10%) had a hospital readmission within 30 days of discharge, including 6 admissions due to cardiopulmonary complications or minor bleeding and 2 for non-cardiopulmonary or bleeding-related reasons. The observed readmission rate of 10% was similar to historic rates of 12.8%. Conclusions. Intermediate-term follow-up for CDT demonstrates high success rates with low adverse event rates. Further randomized data are needed to study the long-term benefits of CDT.

J INVASIVE CARDIOL 2021;33(12):E949-E953. Epub 2021 November 11.

Key words: pulmonary embolism severity score, right ventricular dysfunction, thrombolysis

Pulmonary embolism (PE) remains a major healthcare burden, representing the third leading cause of death for hospitalized patients in the United States. It accounts for 600,000 hospitalizations and up to 180,000 deaths annually.¹ Prompt detection and effective treatment of PE may reduce the likelihood and severity of complications, such as right ventricular (RV) failure and death.^2,3 Long-term sequelae include chronic thromboembolic pulmonary hypertension (CTEPH), post thrombotic syndrome, and residual heart failure. In the PEITHO trial, 2.7% of survivors of intermediate-risk PE had evidence of CTEPH at 6-month follow-up.⁴

Treatment is guided by risk stratification based on evidence of hemodynamic instability and RV dilation according to guidelines published by the European Society of Cardiology (ESC) and the American Heart Association (AHA).^5,6 Massive (AHA) or high-risk PE (ESC) leads to an elevated RV-to-left ventricular (LV) ratio and systemic hypotension. Patients with submassive (AHA) or intermediate-risk PE (ESC) show signs of RV dilation and/or elevated cardiac enzymes without evidence of hemodynamic instability. In low-risk patients, neither RV dysfunction nor hemodynamic instability is present. Risk stratification may additionally be informed by calculation of the pulmonary embolism severity index (PESI)^7,8 or simplified PESI,⁹ which assign points to clinical variables to predict mortality.

Submassive PE represents 40% of all PE cases, and the appropriate therapy for submassive PE remains to be determined.¹⁰ Current guidelines recommend that anticoagulation be used instead of thrombolysis for patients without hypotension or cardiac arrest, and systemic thrombolytic therapy is preferred to catheter-directed thrombolysis (CDT) for PE with hypotension.¹¹ Systemic fibrinolytic therapy has been shown to reduce cardiovascular collapse, but leads to intracranial hemorrhage in 3% of patients treated for PE.^12,13 CDT has been used with increasing frequency to treat patients with high-risk PE, and may hasten recovery and mitigate the risk of long-term sequelae.

CDT is hypothesized to offer enhanced penetration of the thrombus with reduced doses of thrombolytic agents compared with systemic thrombolysis. It may therefore reduce the rate of intracerebral hemorrhage and other major bleeding. Previous research has suggested favorable short-term outcomes for CDT, including recovery from pulmonary hypertension, 24-hour improvement in RV function, and reduced rates of intracranial hemorrhage compared with systemic thrombolytics,^14,15 but no direct comparison has been conducted. To date, little research has described outcomes after discharge of patients receiving CDT, and the long-term risks or benefits of this therapy remain inadequately understood. In this study, we investigate the intermediate- and long-term outcomes of patients treated with CDT.

Eighty-one consecutive patients with acute submassive PE treated with CDT at a single academic center (University of Pennsylvania-Presbyterian Medical Center, Philadelphia, Pennsylvania) between November 2014 and July 2019 were identified and screened, and 79 were enrolled in this retrospective study. All patients had computed tomography (CT) angiography-confirmed PE and had a thrombus meeting the definition of intermediate- to high-risk submassive PE, which required evidence of RV dilation and/or elevated biomarkers without hemodynamic instability. Hemodynamic instability is defined in this study and by the ESC as cardiac arrest or systolic blood pressure < 90 mm Hg or requiring vasopressor support with end-organ hypoperfusion or persistent hypotension with end-organ hypoperfusion. In this study, we also considered patients requiring extracorporeal membranous oxygenation (ECMO) to meet the definition of hemodynamic instability. RV enlargement was defined as RV-to-LV ratio >1 on CT or echocardiography. Elevated biomarkers were defined as elevated troponin I, troponin T, brain natriuretic peptide (BNP) and/or N-terminal pro-brain natriuretic peptide (NT-proBNP). The decision to institute CDT was recommended by the attending physician and was based on significant RV dysfunction, elevated biomarkers, and adverse findings on clinical assessment including persistent tachycardia and dyspnea with mild exertion, and/or severe, persistent hypoxemia (typically, an oxygen requirement of >5 L/min or need for high-flow nasal cannula). In general, patients with submassive PE who were selected for CDT were among those considered to be at elevated risk within the spectrum of submassive PE.

Patients were treated with CDT per institutional protocol. Individual informed consent for this retrospective review was waived by the institutional review board due to the nature of the study, but each patient gave informed consent for CDT. Venous access was achieved via the femoral or internal jugular veins. All patients received directed tissue plasminogen activator (t-PA) infusion at a rate of 1 mg/hour if treating 1 lung, or 0.5 mg/hour in each lung if treating 2 lungs. Thrombolysis was administered through a standard Cragg-McNamara infusion catheter (Vascular Solutions). Therapy was continued until hemodynamic or symptomatic improvement occurred, or the treating physician determined that further benefit was unlikely to be achieved.

Demographic and clinical data were collected by individual chart review. Variables included baseline demographics and risk factors for PE. All echocardiograms during and after index admission were analyzed for RV dilation and systolic dysfunction and dilation and pulmonary artery systolic pressure (PASP). When the PASP was estimated to be within a range of values, the upper limit was recorded for analysis. PESI was calculated as previously described using the earliest available vital signs measured at the time of the index admission.7,8 Calculation of PESI was limited to those patients in whom supplemental oxygen was administered before record of respiratory rate and oxygen saturation was conducted because of the dependence of the score on these values.

Outcomes were individually evaluated in all patients. They included 30-day all-cause and cardiovascular mortality, echocardiographic evidence of RV recovery, readmission, and major bleeding, which was defined as bleeding requiring transfusion, intra-articular, intraocular, intracranial hemorrhage, and echocardiographic evidence of pulmonary hypertension. Minor bleeding was defined as bleeding that came to medical attention that was neither intracranial nor requiring transfusion.

Eighty-seven patients presenting with submassive PE were screened for this study. Six patients with massive PE requiring vasopressors or ECMO were excluded. These are not submassive PEs and should not be discussed. Eighty-one patients presenting with submassive PE were screened for this study. One patient (1.2%) was excluded because CDT was followed by mechanical thrombectomy, and 1 patient who underwent unsuccessful CDT therapy at an outside hospital with unclear dosing regimen before transfer to the study center for additional CDT was excluded. A total of 79 patients were analyzed. Baseline patient characteristics are shown in Table 1. The median t-PA infusion time was 24 hours (range, 9.5-114 hours).

The high-risk nature of this population is demonstrated by the elevated PESI scores, RV dilation by preprocedure echocardiography, and elevated cardiac biomarkers. Thirty-six of 75 patients (48%) with sufficient vitals at initial presentation to calculate PESI presented with PESI >105, which is categorized as class IV or V. RV dilation was present in all patients with preprocedure echocardiography. In addition, 84% of the 75 patients with available data had elevated biomarkers, including elevated NT-proBNP or troponin (Table 2). Furthermore, relative contraindications to systemic thrombolysis were present in 12% and absolute contraindications were present in an additional 8%.¹⁶

One of 79 patients (1.3%) died of PE-related causes during the index hospitalization. There were no additional deaths within 30 days of admission. There were 2 additional mortalities within 1 year of admission, including 1 death from cardiovascular decompensation and 1 from lung cancer. Median time to follow-up was 597 days (range, 113-1730 days).

Eight patients (10%) were readmitted to a hospital within 30 days of discharge, including 6 admissions due to cardiopulmonary complications. These included 2 events of recurrent PE, 2 events of syncope, 1 event of hypotension and dyspnea of unclear origin, and 1 event of severe tachycardia necessitating in-hospital management. Two patients were readmitted for non-cardiopulmonary reasons, which included back pain and non-healing lower-extremity wounds.

Symptoms of dyspnea were available for assessment after treatment in the medical records of 73 patients. Fifty-two patients (71%) regained normal function with no dyspnea including with exertion (New York Heart Association [NYHA] class 1). Among the 19 patients with residual dyspnea, clinical notes were often insufficient to accurately classify the severity of the symptoms.

Fifty-two patients (65%) had follow-up transthoracic echocardiography (TTE) available for evaluation after CDT. The median follow-up time was 58 days (range, 1-1084 days). Thirty-two patients (62%) demonstrated full RV recovery upon follow-up imaging, defined as a clinical echocardiography classified to have absent RV dilation and normal RV systolic function. A further 12 patients (23%) demonstrated mild RV dilation and/or systolic dysfunction. Seven patients (13%) had residual RV systolic dysfunction or dilation classified as moderate, and the remaining 1 patient demonstrated severe RV dilation or systolic dysfunction at the time of their last echocardiography. Among patients with their last follow-up TTE before the median follow-up time, 58% had normal RV function. In patients who underwent follow-up echocardiography after the median follow-up time, 65% had normal RV function. These 2 rates of RV dysfunction were not significantly different (P=.58). There were no readmissions within 30 days or other mortality within 1 year in the patients with moderate or severe dilation or dysfunction.

Thirty-six patients (46%) had follow-up echocardiography that included estimation of PASP. Of these patients, 15 had PASP <30 mm Hg, 9 had a PASP between 30 and 40 mm Hg, and 12 had PASP ≥40 mm Hg.

There were 3 instances of bleeding; 2 were temporally related to thrombolysis. One patient developed a minor gastrointestinal bleed not requiring transfusion during the index admission, and 1 patient who had recently undergone orthoscopic partial meniscectomy developed hemarthrosis while receiving CDT. Additionally, there was 1 instance of menorrhagia occurring 2 days after discharge from index admission resulting in anemia and requiring transfusion.

In this study, we describe favorable intermediate-term outcomes of 79 patients treated with CDT. Aujesky et al previously found 30-day mortality rates of 4.0% and 11.4% for patients with PESI scores between 106 and 125;⁷ the mean in this population was 108. There was 1 death in the first 30 days (1.3%), which is lower than the rate predicted by PESI, and rates of readmission were comparable to historic controls.

Patients were selected for CDT instead of more conservative treatments because they were believed to be at high risk. This was demonstrated by high PESI scores and cardiac strain. Eighty-four percent of patients had elevated cardiac biomarkers, consistent with right heart strain. Forty-eight percent of patients presented with PESI categorized as class IV or V. Among submassive PE patients, these individuals are at the highest risk of death, with previous estimates of 30-day mortality ranging from 4%-25%.⁷ In a previous study of all patients presenting to a single center with submassive PE, only 18% of patients had a PESI categorized as class IV or V at the time of admission.¹⁷

Despite the elevated clinical risk of patients included in this analysis, rates of death were low compared with those predicted by clinical risk. There was 1 instance of death within 30 days of treatment. The observed all-cause, 30-day mortality rate (1.3%) was low compared with the rate predicted by PESI (4.0%-11.4%).⁷ Rates of readmission were comparable to historical controls. Ten percent of patients were readmitted within 30 days of discharge from index hospitalization, which is comparable to previously published rates of 12.8%.¹⁸

The patients studied here had a lower bleeding rate than previously reported for systemic thrombolysis. In a meta-analysis of patients treated with systemic t-PA for PE, 9% experience major bleeding,¹⁹ and the International Cooperative Pulmonary Embolism Registry (ICOPER) documented a 3% rate of intracranial hemorrhage.^12,13 Compared with systemic thrombolysis, CDT enables treatment of PE with a lower dose of t-PA, which may mitigate bleeding risk. Whereas doses of 100 mg of t-PA administered over 1-2 hours may be used for systemic thrombolysis, 20 patients receiving CDT in this study were treated with 1 mg/hour t-PA for a median of 25 hours. The risk of bleeding in patients receiving heparin monotherapy for therapeutic anticoagulation is highly dependent on individual risk factors, including age, gender, comorbidities, and exposure to non-steroidal anti-inflammatory drugs or anticoagulants, but has historically been documented to be approximately 9%.²¹ In this study, there were 3 bleeding events within 30 days of CDT (3.8%), and only 2 were temporally related to thrombolysis. Of these 2, only 1 was categorized as major bleeding. This involved 1 event of hemarthrosis following lytic therapy in a patient with recent knee surgery.

Absolute and relative contraindications to systemic thrombolysis prevent two-thirds of patients presenting with massive PE from receiving systemic t-PA,22 and CDT might be effective even in patients who are not candidates for systemic fibrinolysis. In this study, 20% of patients had contraindications to systemic t-PA, including 6 patients (8%) with absolute contraindications.¹⁶ These patients did not experience higher rates of complications despite other apparent contraindications to t-PA. One of the 16 patients with absolute or relative contraindications to tPA was readmitted for dyspnea within 30 days of discharge from index hospitalization, and there were no instances of death within 30 days in this population. While the ideal patient population for CDT remains to be identified, CDT may be a promising option for some patients who are not eligible for systemic thrombolysis.

A majority of patients (65%) regained normal RV systolic function and size after CDT treatment, and another group (23%) have only mild residual RV dysfunction or dilation. The prevalence of persistent RV dysfunction in our study is generally consistent with previous reports in other series that included patients treated with thrombolytics, although there is a wide range depending on the population studied.^23-25 RV dysfunction and dilation occur rapidly following PE, and prompt intervention to reduce clot burden may mitigate chronic sequelae.^2,3

Functional recovery is demonstrated by the high rate of NYHA class 1 symptoms at the time of follow-up. A majority of patients (71%) described no symptoms of dyspnea at the most recent visit available for assessment in their medical record. Further classification of the 29% of patients with residual shortness of breath into NYHA classes 2-4 was limited by the unavailability of detailed clinical data.

Assessment of pulmonary artery pressure was limited in this retrospective study both by physician-dependent reporting of echocardiographic assessment of PASP and the absence of follow-up invasive hemodynamic measurement. However, 9 of 36 patients (25%) with echocardiographic assessment of PASP had measurements exceeding 40 mm Hg, which is concerning for possible development of pulmonary hypertension. Previous research has reported rates of symptomatic CTEPH of 1.0% at 6 months after PE and 3.1% after 1 year.²⁶ While CDT appears to have been beneficial for this population, echocardiographic data suggest residual pulmonary hypertension in some patients, which may in some cases be associated with residual clot burden.

Study limitations. This study has several limitations. By nature of the retrospective design, we were limited to clinical data already present in patient records. Systematic, protocol-driven data collection was not possible. Measurement of clinical variables used to calculate the PESI was inconsistent, and in some cases, supplemental oxygen was initiated prior to measurement of vital signs. Echocardiographic evidence of RV function was determined by the physician reading the study, and we were unable to account for interindividual variability while analyzing echocardiograms. Follow-up echocardiography was conducted as requested by the patients’ healthcare providers, and its timing in relation to presentation and availability for analysis after discharge is variable among subjects. This imaging was absent in 27 patients. In 6 of 12 patients (50%) with mild residual RV dysfunction at the time of their last echocardiogram, the last study was conducted within 14 days of CDT. A further limitation is the absence of a control population. With limited data available, comparison of CDT to anticoagulation alone or systemic thrombolysis cannot be made, although the risks of bleeding and mortality appear to be low, and the cardiopulmonary and functional outcomes appear favorable.

The optimal therapy for submassive PE remains the subject of ongoing investigation. Low-risk patients may benefit from anticoagulation alone, but more aggressive therapy may improve outcomes in higher-risk patients. This study demonstrates that CDT is an effective treatment in patients with high-risk submassive PE. In this sample, we observed low mortality and bleeding, and limited residual RV dysfunction. These data therefore suggest that CDT effectively reduces sequelae of PE with low rates of complications, but large-scale, protocol-driven studies with defined endpoints are needed to better determine the safety and efficacy and to determine which patients receive the greatest benefit from this therapy compared with other interventions.

Safety and Outcome of Rheolytic Thrombectomy for the Treatment of Acute Massive Pulmonary Embolism

IAGS 2022 / 16th Biennial IAGS Meeting proceedings / 7.3 Contemporary Management of Intermediate- and High-Risk Pulmonary Embolism

Use of the STEMI Team for Treatment of Patients with Pulmonary Embolism: A Pilot Study

Preserved Left Atrial Function Following Left Atrial Appendage Closure for Stroke Prevention