POST-LVAD right ventricular failure

Abstract

Patient selection and implantation timing are key determinants of success for therapy with a left ventricular assist device (LVAD). End-stage heart failure patients with stable disease on inotropic treatment are the best candidates, whereas patients with cardiogenic shock are considered too ill for LVAD support and should be receive temporary mechanical circulatory support devices to optimize their condition before LVAD implantation. However, if implantation is delayed, outcomes may worsen due to secondary organ damage caused by prolonged end-stage heart failure, with the potential for right heart failure to develop and lead to death. Most patients with advanced left ventricular failure assessed for LVAD implantation also have some degree of right ventricular dysfunction. Though LVADs are effective for treating left ventricular failure, they do not intrinsically treat, and in some instances may worsen, right ventricular failure (RVF). Indeed post-LVAD RVF is a major complication of device implantation and significantly increases postoperative morbidity and mortality. The etiology is often multifactorial, including pre- existing right ventricular dysfunction, leftward shifting of the interventricular septum, excessive volume overload, and suboptimal pulmonary afterload reduction. Different echocardiographic, hemodynamic, and biologic markers may help to the prevention, early diagnosis, and effective treatment of post- LVAD RVF. Specifically, post-LVAD RVF results in poor filling of the left ventricle and poor LVAD output that often necessitate additional right ventricular support with inotropes and pulmonary vasodilators, or rarely, a right-sided mechanical device. Additional treatments that can improve right ventricular function after LVAD implant include annuloplasty to reduce the severity of tricuspid regurgitation, aggressive diuresis to reduce volume overload, and treatment to maintain aortic valve patency in every cycle, to lower excessive left ventricular loading.