A 45 year old male patient underwent LVAD implantation for INTERMACS level 1 ischemic cardiomyopathy. An HVAD system was implanted utilizing a minimally invasive approach with bilateral minithoracotomy on extracorporeal membrane oxygenation support. The patient was discharged from the hospital after 3 weeks. On postoperative day 740, the patient presented with dyspnea and low flow. On admission, there were no signs of hemolysis, but TTE revealed a poorly unloaded dilated left ventricle and a peak velocity over the outflow of 356 cm/sec. Before CT scan, the patient decompensated and had to be intubated and inotropic support was initiated. The CT scan revealed a stenosis of the outflow graft. The patient underwent subsequent outflow graft stenting. The procedure was performed as described in the first case. Left ventricular assist device flow immediately normalized. The logfiles findings were similar to the first case; however, the decrease in power consumption and flow rate developed after a much longer period (1 month; Figure 2). The patient recovered and was discharged. Interestingly, he developed a classical in-pump thrombus with hemolysis and increased power uptake in the following month as well as 1 year later. In both instances, thrombolytic therapy was successful. During a follow-up period of 1.5 years, the patient showed no signs of recurrent outflow graft stenosis.
A 73 year old male patient underwent implantation of a HeartWare HVAD for ischemic heart failure. Minimally invasive LVAD implantation was performed via a left thoracotomy approach and the outflow graft toward the left subclavian artery. The patient was discharged 4 weeks postoperatively. In the course, he developed a severe driveline infection originating from a driveline-associated large bowel erosion. Several procedures including major abdominal surgery were necessary. Two years later, the patient presented with dyspnea and low flow. Transthoracic echocardiography revealed a full, unloaded left ventricle at a peak velocity over the outflow graft of 748.5 cm/sec. A CT scan revealed a stenosis of the outflow graft close to the anastomosis with the subclavian artery (Figure 3). In contrast to the previous two cases, thrombus formation in the outflow graft was suspected in this case. Therefore, initial treatment was thrombolytic therapy, which however was unsuccessful. Subsequently, a SMART stent (12 × 40 mm) was implanted by the interventional radiologist supported by balloon protection. Embolic protection was performed by balloon blockade centrally to anastomosis during the intervention. The VAD flow normalized immediately, and the postinterventional course was uneventful. The logfiles showed a steady decrease of power consumption and flow until an almost total occlusion (Figure 2). The patient was discharged 3 days after the intervention. The patient had no episode of restenosis or pump thrombosis, but finally died from septicemia and multiorgan failure 3 months later.
Obstruction of the LVAD outflow graft is a rare but potentially lethal complication. Differentiation of outflow graft stenosis, pump thrombosis, and inflow obstruction of the VAD can be challenging.
Although hemolysis parameters are obviously misleading in such cases, the logfiles can be a useful diagnostic tool. Although true pump thrombosis results in increased power consumption and consecutively increased calculated flow rates, obstructions of the inflow and outflow lead to decreased power consumption and consecutively decreased calculated flow rates. However, stenosis of the outflow graft shows a rather slow and steady decrease in power consumption over days or even weeks until an almost total occlusion of the pump is reached. In contrast to this, occlusion of the inflow shows an immediate drop in power consumption without any previous signs (Figure 2).
For diagnosis of outflow obstruction, the following diagnostic algorithm is recommended: in each patient in whom pump thrombosis is suspected, an outflow graft obstruction is recommend.
In all patients with LVAD with new onset of dyspnea and changes in pump flow, blood samples were analyzed for free hemoglobin and LDH levels, transthoracic echocardiography was performed, and the logfiles were generated.
If no significant abnormalities were observed in hemolysis parameters, but a full left ventricle in the echo and a slow decrease in power consumption in the logfiles, outflow graft obstruction is suspected. For final verification of the diagnosis, CT scan was performed before angiography together with consecutive outflow graft stenting.
Pump replacement has been traditionally undertaken to address this issue. But in cases of outflow graft obstruction, the complication rate is high because the whole system of the outflow graft has to be replaced.5 Medical thrombolytic therapy has shown increased success in cases of pump thrombosis; in our limited experience, outflow graft stenosis seems not to be a good entity for this treatment.6 In outflow obstruction cases, hemolysis parameters are low, and therefore, a real thrombus formation seems unlikely. The morphological picture in the CT scans looks more like an obstruction than a thrombosis. Meticulous placement and length adjustment of the outflow graft during LVAD implantation is crucial to avoid kinking of the prosthesis, leading to flow problems immediately after or even during implantation.
In all of our three cases, outflow graft stenosis was experienced after a rather long time on the device; therefore, it seems not to be related to a poor primary alignment of the graft. Development of stenosis might be related to changes of the geometry of the mediastinum because of remodeling of the right ventricle. One could speculate that one or the other surgical access might be more prone to outflow obstruction, but actually regarding the outflow graft in all three cases, different approaches were used: case 1: full sternotomy; case 2: bilateral minithoracotomy; and case 3: left-sided minithoracotomy with outflow graft toward subclavian artery. If the outflow graft obstruction would be related to a technical problem during the implantation, one would expect that the problem occurs immediately after the implantation. In all of our cases, the outflow obstruction developed more than 2 years after the implantation.
In case 2, we used a bilateral minithoracotomy approach: in such cases, we usually pass an umbilical tape from one incision toward the other, which serve as guiding structure for the graft later on. The alignment of the graft is observed under direct vision simultaneously from both thoracotomies, similarly correct length adjustment and alignment can easily be achieved. This is because of the attachment of the graft toward the subclavian artery that is more prone to this kind of complication as it has to pass the chest wall; hence, the graft is passed through a ring-enforced gortex graft here, to avoid any kind of kinking. On the other hand, there can be a size mismatch between graft and subclavian artery, but until now, there is no evidence that this is an issue. Another explanation is tissue formation between the graft and a goretex graft, which is used to protect the outflow grafts in case of reentry. For this reason, the use of goretex grafts to protect the outflow graft was stopped. Only if the outflow is attached to the subclavian artery, protection of the graft is mandatory because it has to pass the chest wall.
Regarding postinterventional platelet inhibition, clopidogrel was included for 6 months in all three patients to their running anticoagulative regimens.
In all the three cases, outflow graft stenting could be performed without any complication. Special attention to neurological protection is necessary. Until now, no neurological complications of the procedure were reported; however, cerebral embolization is a potential risk. Although this article only describes three cases, the results seem promising and suggest that interventional stent placement should be considered in this rare complication of LVAD therapy.
1. Lahpor J, Khaghani A, Hetzer R, et al. European results with a continuous-flow ventricular assist device for advanced heart-failure patients. Eur J Cardiothorac Surg 2010.37: 357–361.
2. Hanke JS, ElSherbini A, Rojas SV, Avsar M, Shrestha M, Schmitto JD. Aortic outflow graft stenting in patient with left ventricular assist device
outflow graft thrombosis. Artif Organs 2016.40: 414–416.
3. Pham DT, Kapur NK, Dermody M, Halin N. Stenting of an outflow graft obstruction after implantation of a continuous-flow, axial-flow left ventricular assist device
. J Thorac Cardiovasc Surg 2015.150: e11–e12.
4. Oezpeker C, Zittermann A, Ensminger S, et al. Systemic thrombolysis versus device exchange for pump thrombosis management: A single center experience. ASAIO J 2016.62: 246–251.
5. Stulak JM, Dunlay SM, Sharma S, et al. Treatment of device thrombus in the HeartWare HVAD: Success and outcomes depend significantly on the initial treatment strategy. J Heart Lung Transplant 2015.34: 1535–1541.
6. Webber BT, Panos AL, Rodriguez-Blanco YF. Intravenous thrombolytic therapy for patients with ventricular assist device thrombosis: An attempt to avoid reoperation. Ann Card Anaesth 2016.19: 192–196.
Keywords:Copyright © 2018 by the American Society for Artificial Internal Organs
left ventricular assist device; outflow stenosis; stentgraft