Ultrasound-assisted catheter-directed thrombolysis is increasingly popular for submassive PE. This involves placement of a catheter into one or both pulmonary arteries which infuses alteplase and also vibrates ultrasonically. The combination of localized alteplase and vibrational energy is intended to disrupt the clot. Unfortunately, the actual mechanism of action of this therapy remains unclear.
Ultrasound-assisted catheter-directed thrombolysis works
The only RCT investigating ultrasound-assisted catheter directed thrombolysis (CDT) was ULTIMA, which proved that it reduced pulmonary pressures, compared to heparin which did not. Subsequently eight uncontrolled studies have been published, which paint a reasonably consistent picture (table below)(1). Patients generally received about 25 mg of alteplase, usually via bilateral catheters in both pulmonary arteries. Although some bleeding complications occurred, there were no cases of intracranial or fatal hemorrhage (0/537 patients).
Mean pulmonary artery pressure (mPAP) decreased by an average of 9 mm. This is significant. For example, pulmonary hypertension is defined as mPAP > 25mm. A normal heart can generate a mPAP up to about 40 mm, beyond which it fails. Therefore, the difference between a normal mPAP (24 mm) and a mPAP causing cardiac arrest (~40mm) may be as little as ~15 mm.
The table below compares average hemodynamic improvements observed with ultrasound-assisted CDT versus improvements observed with systemic thrombolysis (2). Ultrasound-assisted CDT appears to be effective.
Why does ultrasound-assisted CDT work?
Although ultrasound-assisted CDT is a single intervention, it contains roughly three components:
-  ultrasonic energy
-  effect of local delivery of alteplase
-  systemic effect of ~25 mg alteplase (regardless of where it is infused)
The remainder of this post will explore evidence regarding these components.
[#1] Ultrasonic Energy
Ultrasonic energy augments thrombolysis in laboratory studies, but it doesn’t seem to work clinically. This may reflect limitations on the amount of energy which can be delivered without overheating the catheter.
Engelberger 2014 performed the only RCT available on ultrasound-assisted thrombolysis. These authors studied 48 patients undergoing catheter-based thrombolysis for DVT. All patients underwent placement of an ultrasound-assisted thrombolysis catheter and were treated with a slow infusion of 20 mg alteplase via this catheter. Patients were randomized as to whether or not the ultrasonic energy in the catheter was turned on. Reduction in clot size was identical regardless of whether ultrasonic energy was used (55% vs. 54%).
More recently, the PERFECT registry described 100 patients who underwent catheter-directed thrombolytic infusions for PE. Most patients (64%) were treated using CDT without ultrasound, with the remainder receiving ultrasound as well. There were no differences in the reduction in pulmonary pressures between the two groups:
Over the years, CDT without ultrasound has been attempted in a variety of diseases (e.g. MI, stroke). This has intuitive appeal, but little theoretical justification in PE.
Targeting pulmonary emboli is easier said than done
CDT makes the most sense for MI or CVA, where there is often a single thrombus causing a single culprit lesion. Alternatively, patients with severe PE typically have multiple lesions distributed throughout the pulmonary circulation. This raises a question of how it is possible to provide “directed” therapy to numerous target lesions.
~85% of patients in recent trials of CDT have had catheters placed into both pulmonary arteries to treat bilateral emboli. However, because the lungs receive 100% of the circulation, the average drug concentration in the lungs will be identical regardless of whether drug is infused peripherally or via bilateral pulmonary artery catheters.
CDT might improve therapy if placing the catheter next to a prominent clot increases the local concentration of alteplase. However, blood flow through an occluded pulmonary arterial system is fast and turbulent, so the alteplase doesn’t stay still. Studies in dogs have demonstrated that thrombolytic released adjacent to a clot is rapidly whisked away, sometimes winding up in the contralateral lung (Schmitz-Rode 1998).
Flow studies explain why thrombolytic agents administered via a catheter positioned adjacent to the embolus may have no more effect than systemically administered agents. –Schmitz-Rode 1998
Current techniques try to solve this problem by attempting to embed the catheter within the clot and infuse thrombolytic directly into the clot. However, these catheters have multiple side-ports, so thrombolytic may preferentially flow through unobstructed side-ports lying outside of the clot.
Catheter-directed thrombolysis ignores the significance of recirculating thrombolytic
CDT focuses on maximizing the first-pass effect of alteplase. This could make sense when performing CDT in a coronary artery, because the alteplase is diluted substantially prior to recirculation.
PE is different. For a patient receiving alteplase infusions into both pulmonary arteries, the average drug concentration in the lungs is equal to the subsequent systemic drug concentration. After flowing through the lungs, alteplase is sent to the systemic circulation and then it immediately returns to the lungs. All alteplase molecules remaining in circulation will keep circulating through the lungs over and over.
Re-circulating alteplase may play a greater role than the first-pass effect. The half-life of alteplase in the blood is ~4 minutes. Meanwhile, the time required for alteplase to pass through systemic circulation and return to the lungs may be roughly one minute (3). If alteplase re-circulates through the lungs every minute and levels fall with a half-life of 4 minutes, then the effect of recirculation may be estimated:
This is a crude estimate, but it illustrates how recirculating alteplase could have a dominant effect on thrombolysis. If recirculation were more important than the first-pass effect, then the site of infusion would be irrelevant.
[#2] CDT without ultrasound: Clinical evidence
This isn’t the first time there’s been excitement about CDT for PE. There was interest in CDT in the 1970s-1980s, with several uncontrolled studies showing it to be effective. Eventually, an RCT was performed in 1988 comparing CDT to an equal dose of alteplase via peripheral vein. Alteplase was equally effective regardless of where it was infused (Verstraete 1988). This study debunked CDT, which remained unpopular until the invention of ultrasound-assisted CDT. Newer catheters may be more effective than catheters used in the 1980s, but this has yet to be tested in an RCT.
Gaba 2014: Thrombolysis via a Swan-Ganz catheter
This study is strange but interesting. It describes 19 patients with submassive PE treated with alteplase infused at 0.5-1 mg/hr via a Swan-Ganz catheter. In 17/19 cases the pulmonary emboli were located bilaterally, but alteplase was infused unilaterally into one pulmonary artery. Patients underwent daily pulmonary angiography, with continuation of thrombolysis until pulmonary artery pressures and symptoms resolved. If angiography showed clearance of clot on the same side as the pulmonary artery catheter but residual clot on the contralateral side, then the catheter was re-positioned onto the other side.
Success was achieved in 18/19 cases, with complete or near complete clot dissolution (the remaining patient died from an intracranial hemorrhage (4)). Catheter repositioning to the contralateral lung was performed only in four cases. This study proves that infusing alteplase unilaterally can achieve resolution of bilateral pulmonary emboli, suggesting that CDT isn’t a precisely targeted therapy.
[#3] Systemic effect of ~25mg alteplase
Patients treated with ultrasound-assisted CDT usually receive ~25 mg alteplase (“quarter-dose”). This may have a greater systemic effect than is commonly appreciated. For example, four case reports describe success using slow infusions of peripheral 25 mg alteplase in PE patients at high risk of hemorrhage (5). Boone 2011 published a series of four patients who developed intra-operative pulmonary emboli which responded to tiny doses of alteplase (0.5-4 mg).
Aykan 2014 reported the effect of 25 mg alteplase administered peripherally over six hours to 27 patients with massive PE. This was a high-risk patient group, most of whom were over seventy years old. However the results were favorable, with no major hemorrhage or in-hospital mortality. Alteplase caused an impressive 23 mm drop in the pulmonary artery systolic pressure (from 57 +/- 8 mm to 34 +/- 3 mm)(6).
Additional data about the systemic efficacy of slow quarter-dose thrombolysis comes from patients with thrombosed prosthetic heart valves. Prosthetic valve thrombosis is similar to PE because it occurs in a central location, which receives 100% of the cardiac output. Slow peripheral infusion of 25 mg alteplase has been shown to be effective (Ozkan 2013).
Conclusion: Why does ultrasound-assisted CDT work?
Understanding why ultrasound-assisted CDT works seems like peeling away layers of an onion, with each layer bringing us closer to the truth. The outer-most layer is ultrasonic energy, which isn’t supported by the only RCT testing it. The next layer is CDT without ultrasound, which likewise was ineffective in the only RCT testing it. Its theoretical basis is also questionable. Ultimately, we are only left with the systemic effects of quarter-dose alteplase, which has proven efficacy and probably is the main mechanism behind all of these therapies.
Mechanism of action matters. For example, the PERFECT study concluded that ultrasonic energy was expensive and nonbeneficial, so CDT without ultrasound should be used. If CDT without ultrasound were also unnecessary, this would imply that similar benefits could be obtained with a slow peripheral infusion of quarter-dose alteplase. Administering alteplase peripherally could allow thrombolysis to be started sooner, achieved noninvasively, and be widely utilized in any hospital.
Stay tuned, this is the first of a two-part series on submassive PE.
- Choosing your poison: thrombolysis vs. heparin
- Resus: The crashing PE patient
- Other stuff
- References for this table: ULTIMA, SEATTLE II, PERFECT, Engelhardt 2011, Kennedy 2013, Dumantepe 2014, McCabe 2015, Engelberger 2015, Bagla 2015.
- It is difficult to make direct comparisons between different studies, because the effectiveness of any intervention will vary between populations (e.g. depending on the initial severity of pulmonary hypertension, time delay to therapy). Other variables limiting comparability include different time-points selected and different methods used to determine pulmonary artery pressures. References for this table: Meyer 1992, Meneveau 1997, Meneveau 1998, Tebbe 1999, Becattini 2010, Wang 2010, Sharifi 2013, Sharifi 2014
- Estimated based on ~4 liters blood volume divided by ~4 liters/minute cardiac output.
- The episode of intracranial hemorrhage occurred in a patient who had received relatively little alteplase, had normal fibrinogen levels (386-486 mg/dL), but had uncontrolled PTT values on a heparin infusion (with elevation of PTT > 200s). The authors attributed this intracranial hemorrhage to supra-therapeutic heparin infusion, rather than a consequence of alteplase infusion. This is consistent with my experience and interpretation of the literature, specifically that alteplase alone is fairly safe but the combination of alteplase with titrated heparin infusions can be dangerous.
- References are: Yildiz 2013, Biteker 2010, Sen 2014, and Aykan 2014.
- Unfortunately this study currently is only available in abstract format: Aykan AC et al. Low dose prolonged infusion of tissue type plasminogen activator therapy in massive pulmonary embolism. European Heart J 2014; 35(Suppl 1): 69.
- The PEITHO trial is often held up as an example of the bleeding risks of thrombolysis. However, it must be noted that patients in this trial were simultaneously treated with full-dose tenecteplase and loading boluses of heparin. This is probably inadvisable under any circumstance. The PEITHO trial is discussed further here.
Image credits: Opening image: Inception (Warner Brothers). White water rapids: en.wikipedia.org/wiki/Whitewater.
Latest posts by Josh Farkas (see all)
- Pulmcrit – Renoresuscitation, vasopressin, vepinephrine, and VANISH - August 15, 2016
- PulmCrit – Six reasons to avoid fluoroquinolones in the critically ill - August 1, 2016
- PulmCrit Wee – Pragmatic comparison of 33C vs. 36C after cardiac arrest - July 28, 2016