Optimization of Left Ventricular Assist Device Support

LVAD

left ventricular assist device

RVF

right ventricular failure

The past 10 years have seen an increase in left ventricular assist device (LVAD) implantation, sicker patients at implant, and improved long-term LVAD survival.¹  This article provides an updated review of the fundamentals and limitations of LVAD optimization, particularly in the perioperative phase.

Notable areas for improvement in postoperative LVAD care are lack of a consensus definition for right ventricular failure (RVF), limited long-term options for persistent RVF, and the absence of randomized trials to guide perioperative therapies.

Right ventricular failure following LVAD placement is common, occurring in 10% to 40% of patients within 2 weeks after surgery, and confers substantial mortality and morbidity.¹  The 2014 Interagency Registry for Mechanically Assisted Circulatory Support definition of RVF, along with RVF severity, is provided in Table I.^2,3  Patients with severe or acute severe RVF exhibit significantly higher mortality.³  Although no validated predictive model for RVF exists, preoperative risk assessment and optimization remain valuable to gauge and minimize postoperative RVF. Postoperative RV management targets optimization of volume status and pulmonary vascular resistance; providing inotropic support; and minimization of tachyarrhythmias, hypotension, acidemia, and coagulopathy. Barring vasoplegia, bleeding, and substantial tricuspid regurgitation, a central venous pressure of 12 mm Hg or less should be targeted.⁴  Pulmonary vascular resistance reduction can be achieved by using inhaled nitric oxide or an inodilator, especially milrinone, which causes a concurrent reduction in systemic vascular resistance and should be used with caution in hypotensive patients. If feasible, early extubation should be pursued because negative intrathoracic pressure minimizes RV afterload and is associated with reduced RVF.⁴  Postoperative RVF necessitates gradual LVAD speed increases and a prolonged inotrope wean. If medically necessitated, early preemptive RVAD strategies are superior to rescue strategies in terms of mortality and end-organ preservation.^4,5 

Postoperative tachyarrhythmia management mirrors that in the non-LVAD heart failure population, with amiodarone often being the first-line antiarrhythmic agent. β-Blockers should be used with caution given their potential to worsen RV dysfunction. The presence of RVF imparts added susceptibility to hemodynamic compromise from atrial and ventricular tachyarrhythmias, and a rhythm-control strategy may be advisable, achieved medically or through cardioversion. Ventricular tachyarrhythmias should also trigger an investigation into possible preceding ventricular suction events, the detection of which should prompt reevaluation of pump speed and volume status. In addition, subcutaneous implantable cardioverter-defibrillators in particular should be interrogated postoperatively and reprogrammed accordingly to avoid inappropriate shocks that may result from intraoperative lead reorientation.⁶ 

Left ventricular assist device speed adjustment is another hallmark of postoperative care. The goal is to optimize forward flow and perfusion while simultaneously unloading the left ventricle and optimizing RV performance; the secondary goal is to promote at least intermittent pulsatility to reduce the long-term risk of aortic insufficiency and bleeding diathesis. The initial speed is set intraoperatively under transesophageal echocardiographic guidance to verify a midline interventricular septum and acceptable RV size and function. Thereafter, the speed is serially increased in parallel with inotrope weaning and with regular hemodynamic and echocardiographic assessments to ensure adequate systemic perfusion and RV function. Given that increases in speed both challenge RV function because of increased preload and facilitate RV function through decreased afterload, perhaps the most robust checkmark of tolerance of a new speed is subsequent reassessment of hemodynamics, including central venous pressure measurement, the ratio of central venous pressure to pulmonary capillary wedge pressure, and especially pulmonary arterial saturation. Echocardiographic evaluation of the position of the interventricular septum and right ventricle can further confirm the appropriateness of the speed adjustment. Besides RV function, volume shifts and vasoplegia also frequently compound the immediate postoperative hemodynamic milieu. Hence, the trajectory of speed increments often follows a nonlinear pace and also depends on the rate of postoperative RV recovery.

Common late LVAD complications include infection (31%), gastrointestinal bleeding (13%), and neurologic dysfunction (12%).⁷  The leading causes of death include withdrawal of support (19.4%), multiorgan failure (15.8%), heart failure (13.1%), and neurologic dysfunction (12.3%).⁷  Late RVF carries substantial morbidity and mortality. Regular outpatient echocardiographic assessments, with or without invasive hemodynamics, are an integral component of minimizing decompensated heart failure and are associated with reduced hospital readmission rates.⁸ Table II^9,10  summarizes common late LVAD complications.

In light of the growing presence of patients with LVADs in the health care system, there is no shortage of developments and innovations in this field. Further initiatives to create a streamlined definition and better prediction for postoperative RVF will facilitate preoperative patient selection and optimization. The recently developed EVAHEART2 LVAD (Evaheart, Inc) was engineered in part to target lower rates of postoperative RVF; its safety and efficacy are being evaluated in the ongoing COMPETENCE trial. Finally, BiVACOR (BiVACOR, Inc) is a total artificial heart device that is under development and that may fill the void in the need for durable biventricular support.