Abstract
Objectives
A retrospective review of treatment of patients with massive or submassive pulmonary embolism (PE) using AngioJet rheolytic thrombectomy (ART) system with procedural modifications to improve on the previously reported outcomes.
Materials and Methods
Thirteen patients underwent emergent pulmonary artery thrombectomy for massive and submassive PE using ART with pharmacological and procedural modification, in comparison to prior reports. The modifications included the selective use of the Solent Omni AngioJet device in all subjects, distal contrast angiography via the AngioJet catheter before device activation, and limited short run times. Thrombolytic therapy was not used in any patient. Patients were monitored for short- and long-term outcomes. Long-term clinical follow-up and evaluation for persistent pulmonary hypertension with echocardiography was performed.
Results
The pharmacological and procedural modifications resulted in a favorable clinical response without any major complications and without any mortality. Procedure-related anemia (mean hemoglobin drop of 0.49 g/dl) was the only significant minor complication noted. There were no bleeding complications and no transfusion requirement. On a six-month follow-up, there was no mortality, and there were significant reductions in the pulmonary artery pressures.
Conclusion
Major and minor complications were reduced compared to prior reports using ART. A modified ART approach towards treatment of high-risk PE appears promising both in terms of efficacy and safety.
Keywords
Introduction
Pulmonary embolism (PE) is a common condition with variable outcomes. The subset of patients with massive or submassive PE demonstrates the highest mortality rates.1,2 There has been gradual acceptance of clot mitigation strategies for massive and submassive PE. 3 None of these therapies are optimal, and there are limitations to all available options. 4 One of the several options for clot management is the use of ART System, which currently has a warning for its use in this situation due to higher mortality and morbidity. A meta-analysis of ART for the treatment of PE has showed that this treatment is associated with a minor complication rate of 40% and a major complication rate of 28%. The complications noted are bradyarrythmias, heart block, hemoglobinuria, temporary renal insufficiency, hemoptysis, major hemorrhage, and procedure-related deaths. 5 We felt that these complications could be minimized or eliminated with procedural modifications. We report our experience with the AngioJet device, used with pharmacological and technical modifications in patients with massive or submassive PE.
Patients and methods
Our patient population consists of 13 consecutive patients with massive or submassive PE between February 2012 and January 2016. All patients were treated urgently with the ART system. Institution Review Board approval was obtained from DePaul Hospital, where all the procedures were performed, for the purposes of this study. The patients provided informed consent for publication of the images.
Technique
The technique for using AngioJet for pulmonary thrombectomy has been previously described.6,7 We performed our procedure with several modifications. Informed consent was obtained from all patients for the procedure. The procedure was performed in cardiac catheterization laboratory under conscious sedation unless the patient was intubated. Prior computerized tomography scans were reviewed to plan the procedure approach to identify vessels with largest thrombus burden as the treatment could then be focused on these areas. Heparin naïve patients were administered 70 units/kg intravenous heparin on arrival to catheterization laboratory. Activated clotting time was done for patients arriving to catheterization laboratory on intravenous heparin and additional heparin was administered as needed to obtain values above 200 s. Two patients, one with head injury and the other with neurosurgery within one week prior to presentation, were given heparin at a dose of 50 units/kg body weight. Thrombolytic therapy was not used in any patient. After site preparation, the right common femoral vein was accessed with a 6 French sheath. A 5 French angled pigtail catheter was advanced to the main pulmonary artery and then exchanged over a standard J-wire for 6 French 90 cm Ansel guiding sheath (Cook group, Bloomington, IN). Selective angiography was done via guiding catheter or via reintroduction of pigtail catheter to confirm proximal occlusion site. Next, a Glidewire (Terumo Medical, Somerset, NJ) was placed with aid of JR4 or multi-purpose catheter beyond thrombus. A Solent Omni AngioJet device (Boston Scientific, Marlborough, MA) was then advanced over the Glidewire in an inactivated state across the occlusion. An angiogram was done via the device to visualize size of the vessel distal to the occlusion. If vessel was noted to be at least 3 mm in diameter, the device was pulled back proximal to the clot and re-advanced through the clot after being activated. In the event that the wire was in a small vessel distally, the wire was withdrawn and repositioned or more commonly the device was activated only till the point where the vessel diameter was estimated to be at least 3 mm (Figure 1). Routinely, only one or two passes were made and angiography repeated to ensure clot extraction. Individual run times were monitored and kept at less than 20 s.
Initial pre-thrombectomy angiogram showing occluded right middle pulmonary artery (Plate 1). Post-thrombectomy angiography (Plate 2).
The goal of therapy was never to attempt complete clot extraction or achieve an angiographically desirable result but to decrease the clot burden to allow for improvement of oxygenation and hemodynamics (Figure 2).
Angiography demonstrating extensive thrombus in a large right upper pulmonary artery with compromised flow (a) in Plate 1. This was felt to be clinically significant and a decision was made to proceed with thrombectomy in this vessel. A guidewire could not be advanced via the AngioJet catheter into the distal vessel due to native vessel angulation. Therefore, an angled catheter (b) was advanced over a wire into the distal pulmonary artery and perform angiography to visualize the distal vessels (Plate 2). Plate 3 shows that the angled catheter has been exchanged for the Solent Omni AngioJet device (d). Angiography post-thrombectomy demonstrated decreased clot burden and improved flow in the right pulmonary artery. There is a filling defect (e) consistent with residual thrombus, but additional runs were not done to attempt to eradicate this clot as the flow was felt to be much improved and adequate. Device was activated only till point (c) on Plate 2.
Inferior Vena Cava (IVC) filters were implanted on an individualized basis at the end of the procedure and discussion regarding filter removal was undertaken at first follow-up visit.
Patient’s hemodynamics and oxygenation was continuously monitored during the procedure. We used the calculated saturation to FiO2 ratio to quantitatively assess oxygenation. This has been used as a surrogate for PaO2 to FiO2 ratio.
Pulmonary artery pressures were obtained during the procedure. Run times were documented. Close monitoring of hemodynamic and respiratory status was done post-procedure during hospitalization. Complete blood counts and basic metabolic panel were monitored post-procedure.
Follow-up
Repeat echocardiography was done three to six months post-procedure to assess the right ventricular size and pulmonary artery pressures. Clinical follow-up of at least six months was available in all patients. Statistical analysis was done with t test for all continuous variables.
Results
Patient population
Table 1 details the clinical characteristics of the 13 patients enrolled in this analysis. Mean age was 56.2 ± 15.9 years. There were 10 male and 3 female patients. Presentation was acute dyspnea in seven patients, syncope in three, hypotension in two, and cardiac arrest in one patient. Five patients were felt to have massive PE and eight had submassive presentation. The etiologies were recent surgery (five patients), hypercoagulability (two patients), cancer (two patients), immobility (two patient), and no identifiable cause in two patients. One patient was diagnosed to have methelene tetrahydrofolate reductase (MTFHR) gene and another had an unidentifiable reason on routine testing but had several family members and a personal history of thromboembolism at young age. Eleven patients had first PE episode and two had previous PE. One patient had a history of deep venous thrombosis (DVT).
Patient characteristics.
DIC: disseminated intravascular coagulation; DVT: deep venous thrombosis; PE: pulmonary embolism; RV: right ventricle; LV: left ventricle.
Four patients were on vasopressors, two on ventilators, and five on non-rebreather masks. All patients had right ventricle strain evidenced on computed tomography angiography with a mean right ventricle to left ventricle ratio of 1.43. All patients had a ratio of >0.9. All patients in addition had elevated cardiac Troponin I levels. Eight patients had saddle embolism, four had main stem thrombus, and nine had bilateral mid-lung embolism. One patient did not have central PE and only had bilateral PE. Ten patients were diagnosed to have DVT during hospitalization on venous Doppler imaging.
Hemodynamic data
There were substantial changes in all parameters in the right direction (Table 2). Patient’s heart rates, systolic blood pressures, and oxygenation improved significantly immediately post-procedure.
Hemodynamic data.
Sat/FiO2: saturation to FiO2 ratio.
Technical success
There was 100% success defined as absence of death or major complications. Major complications recorded were stroke, myocardial infarction, pulmonary hemorrhage, hemoptysis, and major bleeds requiring transfusions.
Complications
Transient bradycardia with was noted in five patients during device activation which was sinus bradycardia with heart rate less than 50 beats per minute in three patients and second degree Mobitz type 1 atrioventricular block in two additional patients. None of the patients required atropine or pacing and the arrhythmias self-terminated on cessation of device activation. The device was activated for a mean of 15.6 s with each run in 11 patients, and the data were not available for two patients.
We evaluated for complications with checking blood counts, platelet counts, and renal function before and 24 h after the procedure. There was an average drop in hemoglobin of 0.49 ± 0.77 g/dl (p<0.05) within 24 h of the procedure; however, there were no transfusions needed. Platelets count decreased from 227,076.9 ± 89.9/mcL to 221,230.8 ±73.1/mcL (p > 0.5, not significant) within 24 h of the procedure. Lastly, the renal function measured as creatinine level decreased from 1.09 ± 0.39 mg/dl to 0.99 ± 0.34 mg/dl (p < 0.05). Acute renal failure defined as a 25% increase in baseline creatinine or an absolute increase of >0.5 mg/dl was not noted in any patient.
Follow-up
The pulmonary artery pressures during procedure averaged 57/27 mm Hg with a mean of 37.1 ± 7.2 mm Hg. Echocardiography done three to six months post-procedure showed mean pulmonary artery systolic pressures had decreased to 28.75 ± 11.9 mm Hg in 12 patients (p < 0.005). One patient did not have detectable tricuspid regurgitation, and therefore pulmonary artery pressures could not be calculated. Right ventricular size and function were normal in all but one patient. There was no mortality at six months or the longest available follow-up.
Discussion
Clot mitigation strategies are now well accepted for high-risk patients with PE and appear to improve outcomes. The goal of treatment of massive and submassive PE is to improve short-term mortality and prevent long-term pulmonary hypertension. There is no ideal treatment available. Thrombolytic therapy is often used but up to two third of all treatment eligible patients do not qualify for thrombolytic therapy. 8 Patients who do qualify and receive thrombolytic therapy have a high risk of both local and systemic hemorrhagic complications.9,10 Surgical embolectomy is limited to highly experienced centers and despite improvements has a mortality rate of 19% in a newer pooled analysis. 11 There is no standard catheter-directed therapy (CDT) protocol and multiple techniques are used. Clot fragmentation using a rotating pigtail catheter can result in distal embolization and worsening pulmonary hypertension. 12 Rheolytic thrombectomy has traditionally been associated with a high complication rate. 5 Other devices are limited by requirement for surgical cut downs, availability in United States, and large bore access. Data regarding the success and safety of these other devices are lacking. 4 The Ekos system is the only approved thrombectomy system in United States but requires adjunctive thrombolytic therapy.
An ideal device should be easy to use, be effective in clot retrieval, work independently of thrombolytic therapy, and above all have low complication rates. CDT holds significant promise but still has hurdles, often requires adjunctive thrombolysis and there are high procedural complications. None of the currently available devices fulfills all the above criteria.
There are several prior reports using ART for treatment of high-risk patients with PE. Our goal was to improve efficacy and reduce both the hemorrhagic and non-hemorrhagic complications with the use of this device. The hemorrhagic complications found on prior reports were felt to be both due to thrombolytic therapy and instrumentation of pulmonary arteries. Thrombolytic therapy is frequently used systemically or as an adjunct to CDT. These agents have well-described major hemorrhagic complications.9,10 Besides, a majority of patients with massive or submassive PE are not candidates for thrombolytic therapy. 8 Nine of the 13 patients we studied had contraindications to thrombolysis; however, the procedure was performed without thrombolytic therapy in all reported patients. We feel that this approach definitely contributed to the absence of any hemorrhagic complication in our patient population.
We used the Solent Omni AngioJet (Boston Scientific, Marlborough, MA) device in all patients. Multiple previous studies have used the XVG or Xpeedior catheter, but there is no prior information on using the Solent Omni catheter. We feel there are distinct advantages in using this particular device for mechanical rheolytic thrombectomy. The Xpeedior device is an older generation device and no longer in production. The XVG is still in production but is significantly weaker than the Solent Omni device, being a 5 French system (Boston Scientific, Marlborough, MA). The Solent Omni device provides more effective clot retrieval. This proves particularly efficacious in the treatment of main stem or saddle embolus where a less potent device may not be able to achieve adequate clot retrieval due to larger vessel diameter or extensive thrombus burden. A more effective mechanical device can also obviate the need for thrombolytic therapy.
Secondly, it has a distal port, which can be used to provide contrast angiography distal to the catheter with the wire in place. Due to this factor, we can modify the procedure significantly to assess how far down to advance the device. We feel that hemorrhagic complications are more likely to occur when the device is activated in a small distal vessel due to the risk of vessel trauma, rupture, and pulmonary hemorrhage. In a standard approach, the activated device is advanced through the clot without any prior knowledge of distal vessel anatomy. There is uncertainty in terms of the size of the vessel distal to the thrombus. The operator can inject contrast to opacify the vessels distal to the occlusive thrombus to determine wire position and vessel diameter. This therefore identifies a distal point beyond which the device was not activated.
Lastly, the operator was also attentive of aggressive device manipulation and activation times. The goal was always felt to be clinical and hemodynamic improvement rather than complete clot extraction on angiography. We felt that repeated passes and prolonged run time would be associated with non-hemorrhagic complications: bradyarrhythmias, hemoglobinuria, and thrombocytopenia. Therefore, the run frequency and individual run times were kept at a minimum. Despite this, we obtained adequate clot removal, probably due to the higher suction power of this device. We have not found any distinct angiographic advantage to increased number of runs.
We obtained marked improvements in hemodynamic and respiratory status, with normalization of heart rates, blood pressure, and oxygenation. We were able to wean off vasopressors and respiratory support in all patients. The procedure was well tolerated with limited minor complications and no major complications. There were procedure-related bradyarrhythmias, which were self-limited and did not require specific treatment. We did experience a significant average drop in hemoglobin of 0.49 mg/dl. However, this was minor enough that none of the patients required blood transfusions. There was no significant procedure-related thrombocytopenia. We attribute this to keeping the activated device run times to a minimum. Lastly, unlike prior reports, we had no decline in renal function. Conversely, we found a reverse trend with an improvement in renal function, probably due to improved hemodynamics and subsequently improved renal perfusion.
Outliers
This device performed well in a high-risk patient subset with very low complication rates. One of the patients with both acute and chronic PE had poor thrombus extraction with this device. This was felt to be due to the presence of organized thrombus from prior PE. It is likely that the device does not perform well in chronic organized clot in the absence of adjunctive thrombolytic therapy. This patient demonstrated persistent pulmonary hypertension and right ventricular dilatation.
A second patient with both acute and chronic PE had angiographic improvement in thrombus burden with ART, but the pulmonary arterial pressures remained elevated. However, right ventricular size and function normalized. Persistently elevated pulmonary pressures were felt to be due to chronic thromboembolic pulmonary hypertension probably as a sequela of previous PE. Except for these two patients, all other patients demonstrated marked reductions in pulmonary artery pressures.
Limitations
There was no control group in our study, but there are obvious disadvantages to this procedure and technique. The procedure could be potentially less efficacious in the absence of thrombolytic therapy, but we feel that the Solent Omni device performs well in the setting of acute thrombus without the aid of thrombolysis. Local thrombolysis can be instituted during the procedure for a low-bleeding-risk patient if clot retrieval is felt to be inadequate. This is particularly feasible with the low heparin dose used in our study. However, the risk associated with thrombolytic therapy is not warranted for the majority of patients. Secondly, the procedure as described would be more time consuming and lead to a higher contrast volume and radiation exposure. These, we believe is unavoidable in the interest of the higher safety noted with the procedural modifications.
There are several other limitations to this report. This is a small sample size of heterogeneous group of patients, which may be a factor in influencing results. However, this report details all comers without patient selection. A second drawback is that one operator performed all the procedures, and there is some concern regarding reproducibility. Despite this, the techniques used are easy to apply for an operator who is familiar with the device and angiographic anatomy of the pulmonary circulation.
Conclusion
The procedure was modified to maximize efficacy and decrease complications. It appears that these modifications do result in an improved outcome for these high-risk patients. This report could potentially influence a renewed interest in this technology for treatment of high-risk patients with PE. If applied safely, the advantage of this treatment is its applicability to all patients, as it does not rely on the use of thrombolytic therapy. Besides, it provides a prompt clinical response in critically ill patients where thrombolytic therapy, locally or systemically administered, could take several hours to be effective.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.