Methods
Study Design and Patients
We performed a retrospective cohort study of patients treated by a standardized USAT protocol for PE at the Bern University Hospital in Switzerland between April 2010 and January 2013. Patients were eligible for USAT if they had high- or intermediate-risk PE and evidence of embolus located in at least one main or lower lobe pulmonary artery as assessed by contrast-enhanced chest CT or conventional pulmonary angiography. During the study period, 67 PE patients (46 at intermediate risk and 21 at high risk) were eligible of whom 7 intermediate-risk patients were managed medically and 1 by surgical embolectomy, and 6 high-risk patients underwent surgical embolectomy and 1 received systemic thrombolysis. Overall, 38 intermediate-risk patients and 14 high-risk patients were treated by USAT. Patients signed informed consent for retrospective data collection.
High-risk PE was defined according to the American Heart Association as PE with sustained hypotension (systolic blood pressure <90 mmHg for at least 15 min or requiring inotropic support, not due to any cause other than PE), pulselessness, or persistent profound bradycardia (heart rate <40 b.p.m. with signs or symptoms of shock). Intermediate-risk PE was defined as PE with preserved systemic systolic blood pressure (≥90 mmHg) and evidence of right ventricular dysfunction on imaging. A right-to-left ventricular end-diastolic diameter ratio (RV/LV ratio) of >0.9 on echocardiography or multi-detector contrast-enhanced CT was used to confirm the presence of right ventricular dysfunction.
Medical records were reviewed and clinical data including symptoms, vital signs, comorbidities, risk factors for venous thrombo-embolism, and common laboratory values were recorded. Symptom duration was classified as acute, subacute, and acute-on-chronic. Acute PE refers to PE for which suggestive symptoms have been present for less than 14 days, subacute PE as symptom duration of 14–28 days, and acute-on-chronic PE refers to PE in patients with a previous history of symptoms suggestive of PE and acute exacerbation of symptoms within the last 14 days. Multi-detector contrast-enhanced CT images were used to determine the RV/LV ratio and the modified Miller score at baseline. This scoring system is based on the number of segmental pulmonary arteries involved. Non-occlusive thrombus in a segmental artery is assigned a score of 1, occlusive thrombus a score of 2. Thrombus in more proximal arteries receives a score equal to the number of segmental arteries supplied, with a maximal score of 36 points if the main pulmonary trunk is completely occluded. Transthoracic echocardiography was performed at baseline and after the completion of USAT at 24 h. Recorded echocardiographic loops were analysed by an experienced cardiologist (A.M.) for signs of right ventricular dysfunction, and graded as normal RV function, or mildly, moderately, or severely reduced RV function. Right and left ventricular dimensions were obtained from an end-diastolic apical four-chamber image by measuring the ventricular endocardial borders at the sub-annular plane located 1 cm above the annular plan and perpendicular to the interventricular septal axis.
Standardized Procedure of Ultrasound-assisted Catheter-directed Thrombolysis
All patients were treated with USAT using 12-cm treatment zone EkoSonic MACH4 Endovascular Systems (EKOS Corporation; Bothell, WA, USA). The EkoSonic Endovascular System consists of three main components: an Intelligent Drug Delivery Catheter (IDDC); a removable MicroSonic Device (MSD) containing multiple small ultrasound transducers distributed along the 12-cm long treatment zone; and the EkoSonic control unit. The MSD is placed through the central lumen of the IDDC to deliver high-frequency (2.2 GHz) and low-energy (0.5 Watt per transducer) ultrasound. The EkoSonic control unit provides power to the system and the user interface for operator control. It continuously adjusts the administered ultrasound energy according to the temperature at the treatment zone measured by the thermocouples within the IDDC. The EkoSonic Endovascular System was cleared by the US FDA in 2008 for the infusion of solutions into the pulmonary arteries and is C.E. certified for intravascular applications.
All patients received an intravenous bolus of unfractionated heparin of 80 units per kilogram body weight at initial presentation. The insertion of the catheter system was performed at the cardiac catheterization laboratory with continuous haemodynamic and electrocardiographic monitoring. Venous access was obtained at the right common femoral vein using a 6-French introducer sheath for patients who were scheduled for unilateral EkoSonic device placement or a 10-French double-lumen introducer sheath for those who were scheduled for bilateral EkoSonic device insertion. In five patients with concomitant ilio-femoral deep vein thrombosis, the contralateral common femoral vein was used for venous access. Invasive pressure tracings and a blood sample for the mixed venous oxygen saturation were obtained from the main pulmonary artery. Systemic arterial oxygen saturation was recorded through transcutaneous oxygen saturation measurement. A 0.035-inch hydrophilic guidewire (Terumo Corporation, Tokyo, Japan) and a standard 5-French multipurpose angiographic catheter were used to cross the embolic occlusion. To minimize the risk of pulmonary artery perforation, only the main and lower lobe pulmonary arteries were considered for catheter insertion. With the guidewire tip in a safe position within a large lower lobe segmental branch, the angiographic catheter was exchanged for the EkoSonic IDDC. Finally, the guidewire was removed and the MSD containing the ultrasound transducers was inserted into the IDDC. Bilateral device placement was performed in case of embolus located in both main and proximal lower lobe pulmonary arteries.
Per discretion of the interventional physician, a bolus of recombinant tissue plasminogen activator (rt-PA) of up to 5 mg per device was allowed but discouraged for patients with intermediate-risk PE. A continuous infusion of rt-PA at 1 mg/h and of saline coolant at 35 mL/h per catheter, and intravascular ultrasound delivery were then initiated. After catheter placement, patients were transferred to the intermediate or intensive care unit for continuous monitoring. After 5 h of treatment, the infusion rate of rt-PA was reduced to 0.5 mg/h per catheter for the remaining 10 h. The maximum suggested rt-PA dose was 20 mg for patients who had not received a bolus and 30 mg for those with bolus (10 mg bolus, 20 mg infusion). During USAT, intravenous unfractionated heparin was continuously administered with repetitive dose adjustments every 6 h to achieve and maintain an activated partial thromboplastin time ratio of 1.5–2.5.
At 15 h, the rt-PA infusion and ultrasound delivery were stopped. Thereafter, the EkoSonic devices were removed without fluoroscopic guidance in the intermediate or intensive care unit. After the removal of the MSD, invasive pressure tracings were recorded while slowly pulling back the EkoSonic IDDC. Once a typical pressure tracing of the main pulmonary artery trunk was obtained, a blood sample for the follow-up mixed venous oxygen saturation was taken from the IDDC. Follow-up systemic arterial oxygen saturation was obtained as described above. Finally, the EkoSonic IDDC and the introducer sheath were removed, and the puncture site manually compressed until local haemostasis was achieved.
Endpoints and Definitions
The primary endpoints of this study were the change of pulmonary artery pressure and cardiac index from the time of catheter placement to catheter removal. Cardiac index was derived from dividing cardiac output (CO) by body surface area (BSA). Cardiac output was calculated with the following formula according to the Fick principle: CO = VO2/[13.4 × Hb × (SaO2 – SvO2)]; where VO2 is the estimated oxygen uptake [calculated by: BSA × (161 − age × 0.54) for men; and BSA × (147.5 − age × 0.47) for women]; Hb the haemoglobin level in (g/dL), and SaO2 and SvO2 the oxygen saturation of arterial and venous blood, respectively. Body surface area was calculated according to the DuBois formula [BSA = 0.007184 × (height (cm) × weight (kg)].
Secondary endpoints were treatment-related success rates and complications. Technical success was defined as successful insertion of the EkoSonic device and the initiation of the rt-PA infusion with simultaneous high-frequency/low-energy ultrasound energy delivery. Stabilization of haemodynamics was defined as: cardiogenic shock or hypotension resolved, vasopressor dose reduced or discontinued, or signs of RV failure on echocardiography improved within 24 h of the end of the procedure. Bleeding complications were classified according to the International Society on Thrombosis and Haemostasis, where major bleedings are either (i) fatal bleeding, (ii) symptomatic bleeding in a critical area or organ, such as intracranial, intraspinal, intraocular, retroperitoneal, intraarticular or pericardial, or intramuscular with compartment syndrome, and/or (iii) bleeding causing a drop in haemoglobin level of ≥2 g/dL, or leading to transfusion of two or more units of whole blood or red cells.Minor bleedings are less severe bleedings not included in the definition of major bleedings. After 3 months, the patient or their treating physician were contacted by telephone to assess intermediate clinical outcomes.
Statistical Analysis
Data are presented as means ± standard deviations or absolute numbers and percentages for continuous and categorical variables, respectively. Where appropriate, categorical outcomes are presented as percentage with 95% confidence intervals (95% CI). Data were stratified with regard to intermediate- and high-risk PE. P-values for differences between the groups were calculated from unpaired t-tests or Wilcoxon rank test where appropriate for continuous variables, and χ tests for categorical variables. Parameters assessed before and after US-assisted CDT were compared using paired t-tests. All statistical analyses were performed using STATA version 9.1 (Stata Corp., College Station, TX, USA).