Patients with left ventricular assist devices (LVADs) present unique challenges in thromboembolic management. Blood clotting is affected by a variety of factors, classically articulated by Virchow’s triad: blood chemistry, surface contact, and flow patterns.1 All the three factors are altered in LVAD patients. Blood chemistry is modified by anticoagulants and other influences. The foreign surfaces of the pump components induce platelet activation. Blood flow patterns are altered as blood is pumped out of the heart’s apex into the aorta, bypassing the aortic valve. LVAD support reduces left ventricular pressure, which can remain lower than aortic pressure throughout the cardiac cycle, even when the heart is contracting. Under these “series” flow conditions, all blood flow is exiting the heart through the LVAD, and the aortic valve remains closed.2 This abnormal intraventricular flow pattern can reduce or eliminate the vortex formation and washout that discourages thrombus formation in the normal heart. Once a thrombus forms, its presence can further alter the flow field and worsen the local flow stasis in a positive feedback loop.
We report the case of a 79-year-old male patient (weight 76 kg and height 175 cm), with a history of multiple infarctions, ischemic disease, and a triple coronary bypass surgery 14 years prior, who was diagnosed with heart failure and recommended for a LVAD implant. The patient had acquired an anterior left ventricular aneurysm with extensive myocardial calcification which had extruded at the junction of the proximal septal with more normal myocardium, in the left ventricular outflow tract (LVOT). The heart was akinetic in this region and had a cardiac output of 2.2 L/min. The patient was implanted with a HeartMate II continuous flow ventricular assist device (Thoratec, Pleasanton, CA). During the implant surgery, the calcified material was removed from the left ventricle (LV) wall, and the proximal septal perforation was repaired with felt pledgets. The LVAD was set to a pump speed of 9.4 krpm, and echo examination revealed that the aortic valve was not opening. The patient initially tolerated the surgery well and achieved hemostasis and was maintained in the hospital on aspirin (81 mg/day) and Coumadin (2 mg/day) in addition to other standard medical therapy. Figure 1 illustrates the time course of treatment in which the patient was relatively stable for several weeks. Five weeks after the LVAD implant, the patient showed signs of stroke. Transesophageal echocardiography revealed a pedunculated mass extending from the area of repair along the proximal septum that appeared to be a large thrombus. The Doppler flow images shown in Figure 2 illustrate the inflow through the mitral valve and outflow through the apex to the LVAD, with little or no blood circulation along the septum. The mass resided beneath the aortic valve in the altered flow gyre created by the LVAD flow pattern (arrows). Anticoagulant therapy was increased, but did not relieve the symptoms. A second surgery was performed in which additional calcific material was removed from the previous aneurysmal site as well as a thrombus on the proximal septum. The site was re-repaired, and a felt patch secured over the septal perforation. Transesophageal echocardiography showed a lack of thrombus in the repair area beneath the aortic valve immediately after the procedure. However, within a week, echocardiography demonstrated that the clot had reformed in the same location. The patient was scheduled for a LVAD replacement but died from a blood clot to the brain before surgery. An examination of his heart at autopsy confirmed the presence of a large intraventricular thrombus attached to the septal wall at the repair site in the LVOT (Figure 3), which was listed as the cause of death.
The patient described in this case report illustrates the positive feedback interaction of flow disturbances, foreign surfaces, and thrombus formation. Abnormal flow patterns such as stagnation areas have previously been correlated with thrombosis, especially in the presence of medical devices.3–5 Small thrombi can form and grow in areas of blood flow stagnation, which present a serious risk for stroke. Previous studies on the atrial appendage have found that slow flow and reduced ejection are strong predictors of clot formation.3 Slow filling and emptying velocities evidence decreased vortex formation in the atrium, which is an important flow structure in the normally functioning atrium for ensuring good washout during each cardiac cycle.4 The changes in intraventricular flow that accompany LVAD use can similarly lead to areas of stasis. Vulnerable areas in LVAD patients include the aortic outflow tract, in particular aortic cusps with stagnant flow, and the LV apex near the LVAD inflow cannula.6 These LVAD-related flow patterns have not been extensively studied until recently, through experimental studies in our mock loop. Particle image velocimetry (PIV) is an experimental flow measurement technique in which small fluorescent particles flowing with the fluid are illuminated with a laser and imaged with a high-speed camera. The distance moved by each particle is divided by the time between images to calculate the velocity.7 Particle image velocimetry imaging of flow through a transparent rubber ventricle supported by a HeartMate II LVAD was performed in our mock loop using clinically relevant hemodynamic conditions. Results from these experiments (Figure 4) show that systolic vortex behavior is altered with LVAD support, increasing flow stasis near the LVOT which creates an increased risk for thrombus formation. This problem is further exacerbated by the presence of a foreign surface, such as the repair felt. This risk has now been illustrated in the case report presented, indicating the need for continued attention to flow field analysis as a tool for predicting thrombus potential.
LVADs provide tremendous benefits for patients by reducing the symptoms of heart failure and improving the quality of life. However, the accompanying changes have some secondary consequences that remain problematic, such as thromboembolism. The LVAD changes the blood path through the heart, creating an abnormal intraventricular flow field that decreases washout, especially near the LVOT, because it is a site of flow stasis. The combined effect of foreign material, disrupted blood chemistry, and altered flow patterns in LVAD-supported hearts increases the risk of thrombus formation. This may be especially relevant in recipients of mechanical circulatory assist with continuous flow devices, both temporary and chronic, that run at continuous speeds and can create areas of threatening intraventricular flow architecture in the akinetic heart. This creates a unique liability for recipients having prosthetic material in the LVOT which contains thrombogenic material (felt) or is resistant to thrombus adherence (mechanical aortic valve made of pyrolytic carbon). One possible strategy would be to operate the pump in a pulsatile mode, which would introduce transient flow structures into the ventricle that would aid in blood mixing. However, at the present time, the continuous flow LVADs do not provide for explicit pulsatility control. Another suggestion is to use a more biocompatible material for myocardial wall repair, such as CorMatrix (CorMatrix Inc., Roswell, GA). Either of these steps would be expected to improve the positive feedback thrombus formation linked to altered intraventricular flow patterns similar to those predicted using experimental fluid mechanics.
The authors thank the support of equipment by Thoratec, Inc. and Medtronic, Inc. They also appreciate the assistance of the Mechanical Circulatory Support team at Sharp Memorial Hospital in San Diego, CA.
1. Hanson S, Ratner BDRatner BD, Hoffman AS, Schoen FJ, Lemons JE. Testing of blood-materials interactions. In: Biomaterials Science: An Introduction to Materials in Medicine. 1996 San Diego, California Academic Press
2. May-Newman K, Enriquez-Almaguer L, Posuwattanakul P, Dembitsky W. Biomechanics of the aortic valve in the continuous flow VAD-assisted heart. ASAIO J. 2010;56:301–308
3. Goswami KC, Yadav R, Bahl VK. Predictors of left atrial appendage clot: A transesophageal echocardiographic study of left atrial appendage function in patients with severe mitral stenosis. Indian Heart J. 2004;56:628–635
4. Fyrenius A, Wigström L, Ebbers T, Karlsson M, Engvall J, Bolger AF. Three dimensional flow in the human left atrium. Heart. 2001;86:448–455
5. Bluestein D. Research approaches for studying flow-induced thromboembolic complications in blood recirculating devices. Expert Rev Med Devices. 2004;1:65–80
6. Estep JD, Stainback RF, Little SH, Torre G, Zoghbi WA. The role of echocardiography and other imaging modalities in patients with left ventricular assist devices. JACC Cardiovasc Imaging. 2010;3:1049–1064
7. Willert CE, Gharib M. Digital particle image velocimetry. Experim Fluids. 1991;10:181–193